Handbook of Behavior, Food and Nutrition
wwwwwwwwwwwwwww
Victor R. Preedy · Ronald Ross Watson Colin R. Martin Editors
Handbook of Behavior, Food and Nutrition
Editors Prof. Victor R. Preedy King’s College London Department Nutrition & Dietetics 150 Stamford St. London SE1 9NH UK Prof. Colin R. Martin University of the West of Scotland Ayr Campus KA8 0SR Ayr UK
Prof. Ronald Ross Watson University of Arizona Health Sciences Center Department of Health Promotion Sciences 1295 N. Martin Ave. Tucson, AZ 85724-5155 USA
ISBN 978-0-387-92270-6 e-ISBN 978-0-387-92271-3 DOI 10.1007/978-0-387-92271-3 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2011921927 © Springer Science+Business Media, LLC 2011 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)
This book is dedicated to Miss Caragh Brien, my wonderful daughter. Colin R. Martin
wwwwwwwwwwwwwww
Foreword
The Editors are to be commended for bringing together what is arguably the best and most comprehensive text book on the complex interrelationships between the brain, behaviour, food selection and choice. There are more than 200 chapters describing not only how behaviour affects our food selection but also how what we eat affects how we behave. Content covers genetics, sensory factors, endocrine and neuro-endocrine processes, neurology, behaviour, psychology, physiology, the act of eating, food choice, selection, preferences, appetite, pregnancy, human development, children and adolescents, ageing, anorexia nervosa, bulimia nervosa, obesity, nutrient excess and toxicity, alcoholism, quality of life, body image and much more. The authors have helpfully included chapters on changing eating behaviour and attitudes. The International Handbook of Behavior, Food and Nutrition can truly be said to be research based, theoretical, factual, scientific, academic and practical. The International Handbook of Behavior, Food and Nutrition is without doubt a quality text attractive to the wide range of practitioners and intelligent readers with an interest in these areas. These contributions are a testament not only to the skills of the authors, but the Editorial attributes of Professor Preedy, Watson and Martin. They have marshalled together a truly international team of experts. I especially like the structuring of the chapters. Each contribution has a mini-dictionary, “key facts” and “summary points” to facilitation the navigation across fields of interest. This is a cross-disciplinary book of tremendous importance to the health of individuals and society at large and deserves wide dissemination. The Editors and all the authors are to be congratulated. Prof Betty Kershaw DBE FRCN Emeritus Dean The School of Nursing and Midwifery The University of Sheffield United Kingdom
vii
wwwwwwwwwwwwwww
Preface
In this book the Editors aims to disseminate important data pertaining to the modulatory effects of foods and nutritional substances on behavior and neurological pathways and visa versa. This ranges from the neuroendocrine control of feeding to the effects of disease on the brain. The importance of this book pertains to the fact that food is an essential component of cultural heritage but the effects of perturbations in the food-cognitive axis can be quite profound. The complex inter-relationship between neuropsychological processing, diet and behavioural outcome is explored within the context of the most contemporary psychobiological research in the area. This comprehensive psychobiologically and pathology-themed text examines the broad spectrum of diet, behavioural and neuropsychological interactions from normative function to occurrences of severe and enduring psychopathology. The Editors have taken a scientific and objective stand and included chapters that scrutinize the relationships between the brain, behaviour, food and nutrition in a scientific and rational way. In very simple terms this books addresses limitations in other works that may individually look at the one-way-traffic of either food and behavior. This book examines via two-way-traffic at multiple levels. For example, it examines at both preclinical and clinical levels, genes and populations, and how (a) components in food will affect our sensory responses and (b) how our behavior and sensory responses affect what foods we eat, their pattern of consumption and so on. This book consists of over 200 chapters, and is conveniently divided into 5 main sections to represent the various specialty areas, namely: General, normative aspects and overviews Pathological and abnormal aspects Specific conditions and diseases Changing eating behaviour and attitudes Selective methods
The Editors recognize the difficulty in assigning chapters to specific sections. For example in order to describe normative features, abnormal aspects of diet and behavior may also be described. Chapters on food choice may have coverage on the developing brain, behavior and neuroendocrinology. Thus, some chapters can potentially be assigned to several sections. However, this is resolved with the excellent indexing systems that Springer is renowned for. The chapters are well illustrated with numerous tables and figures. This book represents a multidisciplinary “one-stop-shop” of information with suitable indexing of the various pathways and processes. The chapters are written by ix
x
national or international experts or specialists in their field. The Editors recognize that very often experts in one field may be novices in another. To bridge this knowledge-divide the authors have incorporated sections on “Applications to other areas health and disease”, “Key Facts or Features” and “Summary Points” This reference book is for nutritionists, dietitians, food scientists, behavioral scientists, psychologists, doctors, nurses, physiologists, health workers and practitioners, college and university teachers and lecturers, undergraduates and graduates.
Preface
Contents
Volume 1 Part I: General and Normative Aspects: Evolutionary and Genetic 1 Diet and Brain Evolution: Nutritional Implications of Large Human Brain Size.................................................................. William R. Leonard, J. Josh Snodgrass, and Marcia L. Robertson 2 Epigenetics, Phenotype, Diet, and Behavior........................................ Patrick O. McGowan, Michael J. Meaney, and Moshe Szyf 3 The Michigan State University Twin Registry (MSUTR): Genetic, Environmental, and Neurobiological Influences on Food and Diet-Related Behavior..................................................... Sarah E. Racine, Kristen M. Culbert, S. Alexandra Burt, and Kelly L. Klump
3
17
33
Part II: General and Normative Aspects: Sensory 4 Cerebral Activity to Visual Presentation of Food............................... Jyrki T. Kuikka
47
5 Infant Visual Acuity and Relationships with Diet and Nutrition...... Michelle P. Judge, Carol J. Lammi-Keefe, and Holiday Durham
59
6 Acquired Tastes: Establishing Food (Dis-)Likes by Flavour–Flavour Learning............................................................... Remco C. Havermans and Anita Jansen
73
7 Personality Traits in the Context of Sensory Preference: A Focus on Sweetness............................................................................ Paul Richardson and Anthony Saliba
85
xi
xii
8 Changes to Taste Perception in the Food Industry: Use of Cyclodextrins.............................................................................. Giani Andrea Linde, Antonio Laverde Jr, and Nelson Barros Colauto 9 Infantile Olfactory Learning................................................................. Katsumi Mizuno
Contents
99
119
10 Texture and Diet Related Behavior: A Focus on Satiation and Satiety.............................................................................................. 133 Annette Stafleu, Nicolien Zijlstra, Pleunie Hogenkamp, and Monica Mars 11 Sensory Education: French Perspectives............................................. Caroline Reverdy
143
Part III: General and Normative: Endocrine and Neuroendocrine 12 The Role of Cholecystokinin (CCK) in Eating Behavior................... Mihai Covasa and Timothy Swartz
161
13 The Role of Ghrelin in Eating Behavior.............................................. Mihai Covasa and Timothy Swartz
175
14 The Role of Glucagon-Like Peptide-1 (Glp-1) in Eating Behavior.................................................................................................. Mihai Covasa and Timothy Swartz
189
15 The Role of PYY in Eating Behavior and Diet.................................... Jennifer L. Scheid and Mary Jane De Souza
203
16 The Role of Enterostatin in Eating Behavior and Diet....................... Charlotte Erlanson-Albertsson
217
17 The Paradoxical Role of Glucose-Dependent Insulinotropic Polypeptide (GIP) in Diet and Eating Behaviour in Health and Disease............................................................................................. L.R. Ranganath and J. Pinkney 18 The Relationship between the IGF System, Nutrition, and Behavior........................................................................................... Moira S. Lewitt
241
259
19 Leptin and the CNS............................................................................... Jenni Harvey
271
20 Oxytocin and Appetite........................................................................... Céline Caquineau and Gareth Leng
289
Contents
xiii
21 The Role of Apolipoprotein APO A-IV in Eating Behavior and Diet................................................................................................... Giuseppe Derosa and Sibilla Anna Teresa Salvadeo 22 Brain Histamine Affects Eating and Drinking Behaviours................ Leonardo Munari and Maria Beatrice Passani 23 Ectopic Brain Peptides Posing as Adipokines: Fat as a Novel Site of kiss1 Expression.......................................................................... Russell Brown, Syed A. Imran, and Michael Wilkinson 24 Orexigenic Hypothalamic Peptides Behavior and Feeding................ Jon F. Davis, Derrick L. Choi, and Stephen C. Benoit
303
319
337
355
25 Circadian Neuroendocrine-Immune Aspects of Feeding Behavior: Lessons from Calorie-Restricted or High-Fat-Fed Rats..................... Ana I. Esquifino and Daniel P. Cardinali
371
26 Development of Regulation of Food Intake by the Gut and the Brain: Modeling in Animals.................................................... Takashi Higuchi and Chuma O. Okere
387
Part IV: General and Normative Aspects: Neurological 27 Dietary Proteins and Satiety-Related Neuronal Pathways in the Brain............................................................................................. Gilles Fromentin, Nicolas Darcel, Catherine Chaumontet, and Daniel Tomé 28 The Importance of Trace Elements for Neurological Function......... Joel G. Anderson and Keith M. Erikson 29 Cannabinoid Cb1 Receptor Antagonists/Inverse Agonists and Food-Seeking Behavior.................................................................. John D. Salamone, Kelly Sink, Kristen N. Segovia, Patrick A. Randall, Peter J. McLaughlin, V. Kiran Vemuri, and Alexandros Makriyannis 30 The Nutritional Neurotrophic Neoteny Theory: Evolutionary Interactions Among Diet, Brain, and Behavior................................... Nūn Sava-Siva Amen-Ra 31 Forebrain Activation by Postoral Nutritive Substances..................... Takashi Kondoh and Kunio Torii 32 Oral Administration of Phosphatide Precursors Enhances Learning and Memory by Promoting Synaptogenesis....................... Mehmet Cansev and Ismail H. Ulus
407
423
441
457
469
489
xiv
Contents
33 Gender Differences in Brain Activation by Food Stimulation........... Gene-Jack Wang, Nora D. Volkow, Frank Telang, Panayotis K. Thanos, and Joanna S. Fowler
505
34 Food Perception in Adults: Neuroimaging Findings.......................... Alexandra P.F. Key, Evonne J. Charboneau, and Ronald L. Cowan
515
Part V: General and Normative Aspects: Behavioral and Psychological 35 The Breastfed Infant’s Neurobehavioral Organization: Implications for Child Health and Cognitive Development..................................... 533 Sybil L. Hart, Shera C. Jackson, and L. Mallory Boylan 36 Meal Composition and Cognitive Function......................................... Louise Dye, Alexa Hoyland, Daniel Lamport, and Clare Lawton
547
37 Dietary Amino Acids and Mood........................................................... Reeta Rintamäki and Timo Partonen
565
38 Food Coloring, Sodium Benzoate Preservative, and D-serine: Implications for Behavior...................................................................... Kenji Hashimoto
577
39 Malonyl-CoA Signaling in the CNS: Hypothalamic Control of Feeding Behavior and Energy Expenditure...................... M. Daniel Lane and Seung Hun Cha
585
40 Emotional and Behavioral Aspects of Chocolate Eating.................... E.L. Gibson
601
41 Psychological and Physiological Consequences of Drinking Tea....... E.L. Gibson and J.A. Rycroft
621
42 Coffee and Its Effects on the Brain...................................................... Marcellino Monda, Giovanni Messina, Claudia Vicidomini, Andrea Viggiano, Domenico Tafuri, Teresa Iannaccone, Sergio Chieffi, and Bruno De Luca
637
43 Selective Attention as a Mediator Between Food Motivation and Disposition to Act............................................................................ Jaime A. Pineda and David S. Leland
645
44 The Pleasures and Memory of Food and Meals.................................. Paul Rozin and Dina Gohar
659
45 Explicit and Implicit Attitudes to Food............................................... Maria Czyzewska, Reiko Graham, and Natalie A. Ceballos
673
Contents
xv
46 Impact of Eating and Lifestyle Behaviors on Body Weight: Beyond Energy Value.............................................................. Vicky Drapeau, Marion Hetherington, and Angelo Tremblay 47 Food Neophobia and Sensation Seeking.............................................. Thomas R. Alley and Kathleen A. Potter
693
707
48 The Effects of Prior Beliefs and Learning on Consumers’ Acceptance of Genetically Modified Foods: Implications for Diet and Behavior............................................................................ 725 Wallace E. Huffman, Matthew Rousu, Jason F. Shogren, and Abebayehu Tegene 49 Food Cravings: A Central Construct in Food Intake Behavior, Weight Loss, and the Neurobiology of Appetitive Behavior.............. Corby K. Martin, F. Joseph McClernon, Anastasia Chellino, and John B. Correa 50 Non-sensory Factors Which Influence Choice Behavior of Foods That Have a Positive Effect on Health.................................. Gastón Ares
741
757
51 The Concept of Chronotype in Eating Behaviors............................... Christoph Randler
771
52 Feeding and Sleep Behavior.................................................................. Chin Moi Chow and Christopher Herrera
783
53 Healthy Choices? The Implications of Direct and Indirect Stimuli for Product Perception and Food Consumption.................... Vivianne H.M. Visschers and Thomas A. Brunner
797
Volume 2 Part VI: General and Normative Aspects: Physiological 54 The Physiological Relationships Between the Brainstem, Vagal Stimulation, and Feeding............................................................ Andreas Stengel and Yvette Taché 55 Anticipatory Physiological Regulation in Feeding Biology................ Michael L. Power and Jay Schulkin 56 Food Intake and Heart Rate Variability: Toward a Momentary Biopsychosocial Understanding of Eating Behavior........................... Hiroe Kikuchi, Kazuhiro Yoshiuchi, Ken Ohashi, Fumiyo Sato, Yoshiyuki Takimoto, Akira Akabayashi, and Yoshiharu Yamamoto
817
829
845
xvi
Contents
57 Branched-Chain Amino Acids and Central Fatigue: Implications for Diet and Behavior...................................................... Eva Blomstrand
865
58 The Menstrual Cycle: Psychological, Behavioral, Physiological, and Nutritional Factors......................................................................... Olga van den Akker
879
Part VII: General and Normative Aspects: Feeding and Eating 59 Feeding Behavior and Body Mass Index.............................................. Gian Franco Adami
891
60 Blood Glucose Patterns and the Control of Feeding Behavior: A New Framework for the Control of Meal Initiation....................... L. Arthur Campfield, Alexandra C. Smith, and Françoise J. Smith
911
61 Dynamics of Feeding Behavior: Role of Hypothalamic and Satiety Signals................................................................................. B.S. Zanutto and J.E.R. Staddon
929
62 Food Unpredictability and Foraging.................................................... Sarah E. Overington and Louis Lefebvre
941
63 The Impact of Family Meals on Diet and Food Behaviors................. Sarah J. Woodruff and Rhona M. Hanning
957
Part VIII: General and Normative Aspects: Food Choice, Selection and Preferences 64 Television and Food Choice................................................................... Emma J. Boyland and Jason C.G. Halford 65 The Construction of Eating Episodes, Food Scripts, and Food Routines................................................................................. Carole A. Bisogni, Margaret Jastran, and Christine E. Blake
973
987
66 Endogenous Opioids, Opioid Receptors, and Incentive Processes.... 1011 Mauricio R. Papini and Leonardo A. Ortega Part IX: General and Normative Aspects: Appetite 67 Central Regulation of Appetite and Satiety Behavior........................ 1023 Edward B. Lee and Rexford S. Ahima
Contents
xvii
68 Role of the Gastrointestinal Tract in Peptide Hormone Release and Appetite.............................................................................. 1035 Joan Khoo, Christopher K. Rayner, Christine Feinle-Bisset, and Gary Wittert 69 Manipulation of Diet to Alter Appetite................................................ 1051 Joanne A. Harrold and Jason C.G. Halford 70 Disinhibition, Appetite, and Weight Regulation in Adults................. 1069 Eleanor J. Bryant 71 The Effects of Exercise on Appetite Control....................................... 1087 Catia Martins, Denise Robertson, and Linda Morgan 72 Whey Protein and Satiety: Implications for Diet and Behavior........ 1107 Sylvia M.S. Chung Chun Lam and Paul J. Moughan Part X: General and Normative Aspects: Fatty Acids 73 Dietary n-3 Polyunsaturated Fatty Acids and Brain Lipid Fatty Acid Composition......................................................................... 1127 Gudrun V. Skuladottir 74 Mother–Child Long Chain Polyunsaturated Fatty Acid Relationships: Implications for Diet and Behavior............................. 1139 S.A. van Goor, D.A.J. Dijck-Brouwer, and F.A.J. Muskiet Part XI: Pathology and Abnormal Aspects: Genetic 75 Genetic Markers, Weight Reduction, and Behavioral Changes in Lifestyle............................................................................... 1159 Thomas Reinehr and Anke Hinney 76 The Role of Gene Polymorphisms in Susceptibility to Anorexia Nervosa and Bulimia Nervosa.......................................... 1175 Palmiero Monteleone and Mario Maj 77 Changes in Brain Gene Expression in Nutrient Deficiencies: An Example with Iron........................................................................... 1201 Erica L. Unger, Narasimha Hegde, and James R. Connor Part XII: Pathology and Abnormal Aspects: Sensory 78 When Taste Triggers Sociophobia........................................................ 1217 Matthieu J. Guitton
xviii
79 Why Do We Dislike So Many Foods? Understanding Food Aversions....................................................................................... 1231 Christina L. Scott and Ronald G. Downey 80 Influence of Cognitive Biases in Visual Evaluation of Food Amount in Patients Affected by Eating Disorders.............................. 1245 Piergiuseppe Vinai 81 Visual Processing, Food Cravings and Weight-Loss Dieters.............. 1261 Eva Kemps and Marika Tiggemann 82 Abnormal Physiologic Responses to Touch in Feeding Difficulties...... 1273 Donna Scarborough Part XIII: Pathology and Abnormal Aspects: Endocrine and Neuroendocrine 83 Gut Peptides and Enteral Feeding in Critically Ill Patients: Implications for Gastric Dysmotility and Appetite............................. 1285 N.Q. Nguyen and R.H. Holloway 84 The Brain and Leptin Resistance and Implications for Food-Related Disorders................................................................... 1301 Nina Eikelis and Gavin Lambert 85 The Role of Bile Acids in Gut-Hormone-Induced Weight Loss After Bariatric Surgery: Implications for Appetite Control and Diabetes........................................................ 1317 Rachel E. Roberts, Jamsaid Alaghband-Zadeh, and Carel W. Le Roux 86 Orlistat and the Influence on Appetite Signals.................................... 1331 Mark Ellrichmann Part XIV: Pathology and Abnormal Aspects: Neurological 87 Gastrointestinal Disorders in Neurologically Impaired Children................................................................................. 1353 Alja Gössler and Karel Krafka 88 Dysphagia: Neurological and Behavioral Aspects.............................. 1375 Dorianne Feldman and Marlís González-Fernández 89 Neuropsychological Aspects of Eating Disorders – A Focus on Diagnostic Criteria............................................................................ 1387 Jennie C. Ahrén 90 Lexical-gustatory Synesthesia and Food- and Diet-related Behavior.................................................................................................. 1397 Julia Simner
Contents
Contents
xix
Part XV: Pathology and Abnormal Aspects: Behavioral and Psychological 91 Rethinking the Eating Disorder Continuum: A Categorical Approach to Abnormal Eating............................................................. 1411 Jessie L. Miller and Tracy Vaillancourt 92 Cued Overeating.................................................................................... 1431 Anita Jansen, Remco C. Havermans, and Chantal Nederkoorn 93 Frontal Behavioral Symptoms in Prader-Willi Syndrome................. 1445 Kaeko Ogura, Toshikatsu Fujii, and Etsuro Mori 94 Culture to Culture: Fat-Phobia and Somatization............................. 1457 Samir Al-Adawi, Sanjay Jaju, Ibrahim Al-Zakwani, and Atsu S.S. Dorvlo 95 Examining the Relationship Between Binge Eating and Coping Strategies in Adolescents.................................................. 1475 Susana Sierra-Baigrie, Serafín Lemos-Giráldez, and Eduardo Fonseca-Pedrero 96 Nutritional Influences on Antisocial Behavior.................................... 1487 David Benton 97 Disordered Eating and Mental Workload........................................... 1501 Ethan E. Hull, Jennifer E. Phillips, and Dana L. Rofey 98 Psychiatric Comorbidity in Eating Disorders..................................... 1515 Tahany M. Gadalla 99 Cognitive Performance Deficits of Dieters........................................... 1525 Michael W. Green 100 Disturbed Growth in Early Life and Later Neurocognitive Development Associated with Psychiatric Disorders.......................... 1541 Shiro Suda and Nori Takei 101 Coronary Heart Disease, Diet and Neurocognitive Functioning....... 1555 Colin R. Martin, Mick P. Fleming, and David R. Thompson
Volume 3 Part XVI: Pathology and Abnormal Aspects: Physiological 102 Exercise, Appetite, and Energy Balance: The Interactions Between Energy Expenditure and Intake, and the Implications for Weight Management........................................................................ 1569 Stephen Whybrow, Neil King, and James Stubbs
xx
103 Head Injury: Metabolic, Nutritional, and Energy Considerations........................................................................................ 1585 Christine Charrueau, Béatrice Morio, and Christophe Moinard Part XVII: Pathology and Abnormal Aspects: Feeding and Eating 104 Behavioral Consequences of Force-feeding......................................... 1603 Malgorzata Starzomska and Marek Smulczyk 105 Psychological Stress, Diary Methods, and Eating Behavior.............. 1619 Daryl B. O’Connor, Fiona Jones, and Mark Conner 106 Parental Restriction and Their Children’s Food Choices and Intake............................................................................................... 1635 Harriëtte M. Snoek 107 The False Memory Diet: False Memories Alter Food Preferences.... 1645 Daniel M. Bernstein, Nicole L.M. Pernat, and Elizabeth F. Loftus 108 Eating Pattern and Bariatric Surgery.................................................. 1665 Gian Franco Adami Part XVIII: Pathology and Abnormal Aspects: Food Choice, Selection and Preferences 109 Food Addiction: Analysis With an Animal Model of Sugar Bingeing................................................................................... 1687 Nicole M. Avena, Miriam E. Bocarsly, and Bartley G. Hoebel 110 The Lingering Impact of Negative Food Experiences: Which World War II Veterans Won’t Eat Chinese Food?................. 1705 Brian Wansink, Koert van Ittersum, and Carolina Werle 111 Changes in Food Neophobia and Food Preferences During a Weight Reduction Session: Influence of Taste Acuity on the Individual Trajectory................................................................. 1715 Marie-Odile Monneuse, Claude Marcel Hladik, Bruno Simmen, and Patrick Pasquet Part XIX: Pathology and Abnormal Aspects: Appetite 112 Dyspepsia and Appetite Regulation..................................................... 1731 Takashi Akamizu 113 Overnutrition in Mothers and Appetite Regulators in Offspring..... 1745 Hui Chen and Margaret J. Morris
Contents
Contents
xxi
114 Sex Hormones and Appetite in Women: A Focus on Bulimia Nervosa.................................................................................................... 1759 Sabine Naessén and Angelica Lindén Hirschberg Part XX: Pathology and Abnormal Aspects: Lipids 115 Dietary n-3-Polyunsaturated Fatty Acid Deprivation and Cytokine Signaling Pathways in the Brain................................... 1771 Sophie Laye, Virginie F. Labrousse, and Veronique De Smedt-Peyrusse 116 Dietary Supplementation of Omega-3 Polyunsaturated Fatty Acids in Autism............................................................................ 1787 Pierluigi Politi, Hellas Cena, and Enzo Emanuele 117 Docosahexaenoic Acid and Cognitive Dysfunction............................. 1797 Michio Hashimoto, Hossain Md Shahdat, and Masanori Katakura Part XXI: Pathology and Abnormal Aspects: Miscellaneous Topics 118 Maternal Dietary Intake of N-Nitroso Compounds from Cured Meat and the Risk of Pediatric Brain Tumors.................................... 1817 Michael Huncharek 119 Behavioral Aspects of Nonalcoholic Fatty Liver Disease: Diet, Causes, and Treatment................................................................. 1833 Giulio Marchesini, Chiara Nuccitelli, Elena Centis, Silvia Di Domizio, Alessandro Suppini, Rebecca Marzocchi, and Riccardo Dalle Grave 120 Interactions Between Diet, Immune System and Brain Function in the Symptom Profile of Chronic Fatigue Syndrome....................... 1845 Yvonne Christley, Tim Duffy, and Colin R. Martin 121 The Relationship Between Nutrition and Neurocognitive Function in Schizophrenia.................................................................... 1859 Mick P. Fleming and Colin R. Martin 122 Neuronal Circuits and Neuroendocrine Responses Involved in Dehydration Induced by Water Restriction/Deprivation.............. 1873 Zheng-Hua Zhu, Bai-Ren Wang, James S. McTaggart, and Li-Ze Xiong 123 Feminine Norms and Disordered Eating............................................. 1897 Melinda A. Green, David Kugler, Ashley Stillman, Christopher Davids, Katherine Read, Kelly Siglin, and Amanda Jepson 124 The Dietary Antioxidants Alpha-Tocopherol and Alpha-Lipoic Acid and Their Synergy in Brain Disorders.................................................................................. 1911 Oscar Gonzalez-Perez
xxii
125 Obsessive Eating..................................................................................... 1923 Naoko Narita, Mami Tazoe, and Masaaki Narita 126 Eating Disorders and Suicide................................................................ 1939 Antonio Preti, Maria Valeria Camboni, and Paola Miotto Part XXII: Pregnancy 127 Neuroendocrine Mechanisms of Change in Food Intake During Pregnancy.................................................................................. 1969 Alison J. Douglas 128 Effect of Pregnancy and Consciousness Factors on Food-Related Behavior.................................................................................................. 1985 J.K. Chun, S.W. Lim, and W.I. Cho 129 Diet in the Aetiology and Management of Postpartum Depression: Knowing the Facts............................................................ 2009 Vassiliki Costarelli Part XXIII: Developmental, Children and Adolescents 130 Prenatal Diet and Stress Responsiveness............................................. 2023 Susanne R. de Rooij 131 Early Nutrition and Postnatal Brain Growth in the Preterm Infant............................................................................. 2041 Richard W.I. Cooke 132 Perinatal Undernutrition and Brain-Derived Neurotrophic Factor.............................................................................. 2055 Didier Vieau, Sylvain Mayeur, Marie-Amélie Lukaszewski, Fabien Delahaye, Isabelle Dutriez-Casteloot, Christine Laborie, Sylvie Deloof, Jean Lesage, and Christophe Breton 133 The Developing Brain and Dietary Omega-3 Fatty Acids.................. 2069 Sheila M. Innis 134 Dietary Choline for Brain Development.............................................. 2089 Amy R. Johnson and Steven H. Zeisel 135 Iodine and Brain Development............................................................. 2105 Pere Berbel and Gabriella Morreale de Escobar 136 Tryptophan intake and the influence of serotonin on development and plasticity of sensory circuits............................ 2135 Claudio A. Serfaty
Contents
Contents
xxiii
137 Behavioral Aspects of Failure to Thrive in Infants and Young Children............................................................................... 2153 Robert Drewett 138 Video Game Play, Behavior, and Dietary Health................................ 2169 Mary Ballard 139 Disinhibited Eating and Body Weight in Youth.................................. 2183 Lauren B. Shomaker, Marian Tanofsky-Kraff, and Jack A. Yanovski 140 Gender-Based Food Stereotypes Among Young Japanese................. 2201 Atsushi Kimura, Yuji Wada, and Ippeita Dan
Volume 4 Part XXIV: Starvation and Nutrient Deficiency 141 Diet-Related Behavioral Mechanisms in Times of Economic Constraint............................................................................................... 2217 A.R. Kelles, M. Shroff, and A. Rinehart 142 Food Deprivation: A neuroscientific perspective............................... 2239 Harald T. Schupp and Britta Renner 143 Symptoms of Starvation in Eating Disorder Patients......................... 2259 Riccardo Dalle Grave, Elettra Pasqualoni, and Giulio Marchesini 144 Epigenetics and Nutrition: B-Vitamin Deprivation and its Impact on Brain Amyloid......................................................... 2271 Sigfrido Scarpa 145 Reinforcement and Food Hedonics: A Look at How Energy Deprivation Impacts Food Reward...................................................... 2285 Jameason D. Cameron and Éric Doucet 146 Nutritional Deficiencies and Spatial Memory Function..................... 2307 Sayali C. Ranade 147 Arguments for a Relationship Between Malnutrition and Epilepsy........................................................................................... 2329 Sabrina Crepin, Bertrand Godet, Pierre-Marie Preux, and Jean-Claude Desport 148 Cortical Spreading Depression: A Model for Studying Brain Consequences of Malnutrition.............................................................. 2343 Rubem Carlos Araújo Guedes 149 Dietary Zinc and the Brain................................................................... 2357 Mohammad Tariqur Rahman
xxiv
150 Dietary Copper and the Brain.............................................................. 2375 Helma Antony and Ian G. Macreadie 151 Fetal–Neonatal Iron Deficiency Affects Neurotrophic Factor Expression, Neural Differentiation, and Neuroplasticity in the Rat Hippocampus........................................................................ 2393 Michael K. Georgieff and Phu V. Tran 152 Iodine and brain metabolism............................................................... 2411 R.H. Verheesen and C.M. Schweitzer 153 Riboflavin Deficiency, Brain Function, and Health............................ 2427 Rita Sinigaglia-Coimbra, Antonio Carlos Lopes, and Cicero G. Coimbra 154 Brain Mechanisms Involved in the Detection and Adaptation to Lysine Deficiency................................................................................ 2451 Takashi Kondoh and Kunio Torii Part XXV: Anorexia Nervosa 155 Genotypes and Phenotypes of Anorexia Nervosa................................ 2471 Janet Treasure, Natalie Kanakam, and Christine-Johanna Macare 156 Metabolic Consequences in Anorexia Nervosa................................... 2491 Daniel Rigaud and Marie-Claude Brindisi 157 Application of Personal Construct Theory to Understanding and Treating Anorexia Nervosa............................................................ 2503 Malgorzata Starzomska and Marek Smulczyk 158 Comorbidity in Anorexic Adolescents: Assessment Through ASEBA System and Semistructured Interviews................................. 2517 Filippo Muratori, Valentina Viglione, Chiara Montalto, and Sandra Maestro 159 ACTH, Cortisol, Beta Endorphin, Catecholamines, and Serotonin in Anorexia Nervosa: Implications for Behavior................................ 2529 Marie-Claude Brindisi and Daniel Rigaud Part XXVI: Bulimia Nervosa and Night Eating Syndrome 160 Ethnicity in Bulimia Nervosa and Other Eating Disorders............... 2539 Athena Robinson and W. Stewart Agras 161 Dyscontrol in Women with Bulimia Nervosa: Lack of Inhibitory Control over Motor, Cognitive, and Emotional Responses in Women with Bulimia Nervosa.......................................................... 2549 Sonia Rodríguez-Ruiz, Silvia Moreno, M. Carmen Fernández, Antonio Cepeda-Benito, and Jaime Vila
Contents
Contents
xxv
162 Experiences of Women with Bulimia Nervosa.................................... 2565 Kathryn Proulx 163 The Night Eating Syndrome: An Overview......................................... 2583 Jennifer D. Lundgren Part XXVII: Metabolic Syndrome and Non-obese Overweight 164 Metabolic Syndrome as a Disorder of the Brain with Its Origins in the Perinatal Period............................................... 2597 Undurti N. Das 165 Schizophrenia and the Metabolic Syndrome....................................... 2617 Jared Edward Reser 166 Nutrition, Behavior, and the Developmental Origins of the Metabolic Syndrome................................................................... 2627 Jared Edward Reser Part XXVIII: Obesity 167 The Relationship Between Television Viewing and Overweight and Obesity in Young Children: A Review of Existing Explanations........................................................................................... 2641 Vickii B. Jenvey 168 The Dopamine Transporter Gene (DAT1) in Obesity and Binge Eating Disorders.................................................................. 2659 Karen Wight, Caroline Reid-Westoby, and Caroline Davis 169 Feeding and Satiety Signals in Prader-Willi Syndrome: Relation to Obesity, Diet, and Behavior............................................... 2673 Maithe Tauber, Emmanuelle Mimoun, Patrick Ritz, and Gwenaelle Diene Part XXIX: Diabetes 170 Insulin and Clinical Eating Disorders in Diabetes.............................. 2693 Masato Takii 171 Comparing Abnormal Eating Behavior in Type 1 and 2 Diabetic Patients......................................................... 2713 Patrick Ritz, Monelle Bertrand, and Hélène Hanaire 172 Binge Eating in Overweight and Obese Individuals with Type 2 Diabetes...................................................................................... 2721 Amy A. Gorin, Heather M. Niemeier, and Anna Schierberl Scherr
xxvi
Contents
Part XXX: Nutrient Excess and Toxicity 173 High Blood Glucose and Damage to Neuronal Tissue........................ 2733 Robert R. Miller Jr. 174 Neurological Aspects of Dietary Lead.................................................. 2755 Kim M. Cecil and Diana M. Lindquist 175 Diet- and Mercury-induced Visual Loss.............................................. 2775 Cian E. Collins Part XXXI: Aging and Dementia 176 Soy, Tofu and Brain Function in the Elderly....................................... 2783 Amina Yesufu-Udechuku, Tri Budi W. Rahardjo, and Eef Hogervorst 177 Nutritional Risk in the Elderly with Cognitive Impairment: A Far Eastern Perspective..................................................................... 2817 Kang Soo Lee and Chang Hyung Hong 178 Aluminium in the Diet, Cognitive Decline and Dementia.................. 2829 Vincenza Frisardi, Vincenzo Solfrizzi, Patrick G. Kehoe, Bruno P. Imbimbo, Gianluigi Vendemiale, Antonio Capurso, and Francesco Panza 179 Dietary Fatty Acids, Cognitive Decline, and Dementia...................... 2851 Vincenzo Solfrizzi, Vincenza Frisardi, Cristiano Capurso, Alessia D’Introno, Anna M. Colacicco, Gianluigi Vendemiale, Antonio Capurso, and Francesco Panza 180 Nutritional Issues for Older People and Older People with Dementia in Institutional Environments..................................... 2885 Angela B. Kydd
Volume 5 Part XXXII: Alcohol 181 Diffusion Tensor Imaging of the Brain in Fetal Alcohol Spectrum Disorder................................................................................. 2897 Catherine Lebel, Carmen Rasmussen, and Christian Beaulieu 182 Neuronal Cell Migration in Fetal Alcohol Syndrome......................... 2915 Tatsuro Kumada, Yutaro Komuro, Ying Li, Yoav Littner, and Hitoshi Komuro
Contents
xxvii
183 Folate and the Effects of Prenatal Alcohol on the Brain.................... 2931 Yajun Xu and Jie Zhao 184 Decreased Appetite for Food in Alcoholism........................................ 2949 Anna Kokavec 185 Gender Aspects in the Comorbidity of Eating Disorders and Alcohol Use Disorders.................................................................... 2963 Tahany M. Gadalla 186 The Great Disinhibitor: Alcohol, Food Cues, and Eating Behavior.................................................................................................. 2977 Wilhelm Hofmann, Georg Förster, Wolfgang Stroebe, and Reinout W. Wiers 187 Brain Atrophy in Alcoholics.................................................................. 2993 E. González-Reimers and F. Santolaria-Fernández 188 Alcohol Consumption in Predementia and Dementia Syndromes.... 3011 Francesco Panza, Vincenza Frisardi, Patrick G. Kehoe, Cristiano Capurso, Alessia D’Introno, Anna M. Colacicco, Gianluigi Vendemiale, Antonio Capurso, and Vincenzo Solfrizzi 189 Sweet Preference and Mood: Implications for the Risk of Alcoholism.......................................................................................... 3045 Alexei B. Kampov-Polevoy 190 Anxiety and Self Medication with Alcohol.......................................... 3061 Carmen C. Moran and Anthony J. Saliba Part XXXIII: Quality of Life 191 Developmental Aspects of Health Related Quality of Life (HRQL) in Food Related Chronic Disease: The Example of Food Allergy....................................................................................... 3077 Audrey DunnGalvin and Jonathan O’B. Hourihane 192 Nutrition and Quality of Life in Older People.................................... 3099 Salah Gariballa 193 Biopsychosocial, Behavioural Aspects and Quality of Life with Home Enteral Nutrition................................................................ 3115 Agostino Paccagnella, Alessandra Mauri, Gessica Schiavo 194 Quality of Life, Diet, and Behavior in Cancer..................................... 3137 Brenda Larson and Aminah Jatoi
xxviii
195 Quality of Life Assessment in Prader–Willi Syndrome...................... 3153 Pietro Caliandro, Graziano Grugni, Domenica Taruscio, Yllka Kodra, and Luca Padua 196 Dysphagia, Behavior, and Quality of Life............................................ 3163 D.A. de Luis, Mick P. Fleming, and Colin R. Martin 197 Bariatric Surgery and Health-Related Quality of Life....................... 3173 Raed Tayyem, Abdulmajid Ali, John Atkinson, and Colin R. Martin 198 The Impact of Dietary Restrictions on Quality of Life in Kidney Disease................................................................................... 3187 Cheryl Glover, Pauline Banks, Amanda Carson, Mick P. Fleming, and Colin R. Martin 199 Huntington’s Disease: Quality of Life and Diet.................................. 3199 Glenn R. Marland and Colin R. Martin Part XXXIV: Body Image 200 Personal Values, Vanity, Physical Health, and Perceived Body Image Influences in Food-Purchasing and Consumption Decisions.................................................................................................. 3211 Barry O’Mahony and John Hall 201 Factors Influencing Body Image During Adolescence........................ 3221 Rheanna N. Ata, Ariz Rojas, Alison Bryant Ludden, and J. Kevin Thompson 202 Body Image and Eating Disorders Among Immigrants..................... 3241 Nan M. Sussman and Nhan Truong Part XXXV: The Young and Adolescents 203 “My Body Is My Template”: Why Do People Suffering from Anorexia Nervosa See Their Bodies Differently?...................... 3257 Naresh Mondraty and Perminder Sachdev 204 Motor Learning Approaches for Improving Negative Eating-related Behaviors and Swallowing and Feeding Skills in Children................ 3271 Justine Joan Sheppard 205 The Young and Adolescents: Initiating Change in Children’s Eating Behavior...................................................................................... 3285 Tom Baranowski, Teresia O’Connor, and Janice Baranowski
Contents
Contents
xxix
206 Eating Disorders and Behavioral Aspects of School-Based Prevention Programs............................................................................. 3295 Riccardo Dalle Grave, Lucia Camporese, and Elettra Pasqualoni 207 Treating Obesity in Childhood: Behavioral Considerations.............. 3307 Fiona Davies and Louise A. Baur Part XXXVI: Adults and the Elderly 208 The Potential of Internet-Based Programs for Eating Disorder Prevention in Students........................................................................... 3329 Katajun Lindenberg, Markus Moessner, and Stephanie Bauer 209 Behavioral Interventions for Preventing and Treating Obesity in Adults.................................................................................................. 3343 Manoj Sharma and Melinda J. Ickes 210 The Use of a Cognitive Behavioral Program for Diabetes and Cardiovascular Risk Reduction.................................................... 3361 Jeroen Lakerveld, Sandra D.M. Bot, and Giel Nijpels 211 Treatment of Diet-Related Disorders in Adult Diabetes.................... 3375 Ying-Xiao Li, Kai-Chun Cheng, Akihiro Asakawa, and Akio Inui Part XXXVII: Other General or Specific Conditions 212 The Impact of Health-Promoting Media-Literacy Education on Nutrition and Diet Behavior............................................................ 3391 Lynda Bergsma and Elizabeth Ferris 213 Treatment of the Night Eating Syndrome............................................ 3413 Albert J. Stunkard and Kelly C. Allison Part XXXVIII: Selective Methods 214 Psychological Assessment of Eating Disorders.................................... 3425 Wayne A. Bowers and Alissa A. Haedt-Matt 215 Behavioral Assessment of Pediatric Feeding Problems...................... 3437 Colleen Taylor Lukens 216 Use of the International Personality Disorder Examination (IPDE) and the Millon Clinical Multiaxial Inventory (MCMI) to Assess Personality Disorders in Eating Disorders.......................................... 3453 Izaskun Marañon and Iratxe Gonzalez
xxx
217 The Assessment of Bulimic Symptomatology: Factorial Structure and Measurement Invariance of the Bulimic Investigatory Test, Edinburgh Across Gender and Age..................... 3471 Eduardo Fonseca-Pedrero, Susana Sierra-Baigrie, Mercedes Paíno, and Serafín Lemos-Giráldez, José Muñiz 218 Weight-Related Eating Behavior Questionnaires: Applying Theory to Measurement........................................................................ 3487 Susan M. Schembre 219 Ambient, On-Body, and Implantable Monitoring Technologies to Assess Dietary Behavior.................................................................... 3507 Oliver Amft Index ................................................................................................................. 3527
Contents
Contributors
Gian Franco Adami, M.D. Department of Surgery, University of Genova, Largo Rosanna Benzi 8, Genova 16132, Italy W. Stewart Agras, M.D. Department of Psychiatry and Behavioral Sciences, Stanford University, School of Medicine, Stanford, CA, USA Rexford S. Ahima, M.D., Ph.D. Institute for Diabetes, Obesity and Metabolism, School of Medicine, University of Pennsylvania School of Medicine, 712A Clinical Research Building, 415 Curie Boulevard, Philadelphia, PA 19104 Jennie C. Ahrén Center for Health Equity Studies, CHESS, Karolinska Institutet/Stockholm University, Sveavagen 160, 106 91, Stockholm, Sweden Akira Akabayashi Graduate School of Medicine, Department of Stress Sciences and Psychosomatic Medicine, The University of Tokyo, Tokyo, Japan Takashi Akamizu Translational Research Center, Kyoto University Hospital, Kyoto University School of Medicine, 54 Shogoin-Kawaharacho, Sakyo-ku, Kyoto 606-8507, Japan Samir Al-Adawi, M.Sc., Ph.D. Department of Behavioral Medicine, College of Medicine and Health Sciences, Sultan Qaboos University, Al-Khoudh 123, Muscat, Sultanate of Oman Abdulmajid Ali General Surgery Department, Ayr Hospital, Ayr, UK J. Alaghband-Zadeh Department of Clinical Biochemistry, King’s College Hospital, London, UK Thomas R. Alley Department of Psychology, 418 Brackett Hall, Clemson University, Clemson, SC 29634-1355, USA
xxxi
xxxii
Contributors
Kelly C. Allison, Ph.D. Department of Psychiatry, Center for Weight and Eating Disorders, University of Pennsylvania School of Medicine, Philadelphia, PA, USA Ibrahim Al-Zakwani, Ph.D. Department of Pharmacy College of Medicine and Health Sciences, Sultan Qaboos University, Muscat, Sultanate of Oman Oliver Amft, Ph.D. ACTLab, Signal Processing Systems, Faculty of Electrical Engineering, Eindhoven University of Technology, NL-5600 MB Eindhoven, The Netherlands Joel G. Anderson Department of Nutrition, University of North Carolina at Greensboro, Greensboro, NC, USA Helma Antony School of Medical Science, Griffith University, Gold Coast Campus, Southport, Queensland, Australia Gastón Ares Sección Evaluación Sensorial, Departamento de Ciencia y Tecnología de Alimentos, Facultad de Química. Universidad de la República (UdelaR), Gral. Flores 2124. C.P. 11800, Montevideo, Uruguay Akihiro Asakawa Department of Psychosomatic Internal Medicine, Kagoshima University Graduate School of Medical and Dental Sciences, Kagoshima, Japan Rheanna N. Ata, B.A. Department of Psychology, University of South Florida, 4202 E. Fowler Avenue, PCD 4118G, Tampa, FL 33620-8200 John Atkinson West of Scotland University, Paisley, UK Nicole M. Avena, Ph.D. Department of Psychiatry, University of Florida, Gainesville, FL 32608, USA and Department of Psychology and Program in Neuroscience, Princeton University, Princeton, NJ 08540, USA Mary Ballard, Ph.D. Department of Psychology, Appalachian State University, Boone, NC 28608, USA Pauline Banks School of Health, Nursing and Midwifery, University of the West of Scotland, Beech Grove, Ayr, UK Tom Baranowski, Ph.D. Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, 1100 Bates Street, Houston TX 77030, USA Stephanie Bauer Center for Psychotherapy Research, University Hospital Heidelberg, Heidelberg, Germany
[AU1]
Contributors
xxxiii
Louise A. Baur, B.Sc.(Med), M.B.B.S. (Hons), Ph.D., FRACP Discipline of Paediatrics and Child Health, University of Sydney, Sydney, NSW 2006, Australia and Weight Management Services, The Children’s Hospital at Westmead, Westmead, NSW 2145, Australia Christian Beaulieu Department of Biomedical Engineering, 1098 Research Transition Facility, University of Alberta, Edmonton, Alberta T6G 2V2, Canada Stephen C. Benoitm, Ph.D. Department of Psychiatry, University of Cincinnati, Cincinnati, OH, USA David Benton Department of Psychology, University of Wales Swansea, Swansea SA2 8PP, Wales, UK Pere Berbel Instituto de Neurociencias, Universidad Miguel Hernández and Consejo Superior de Investigaciones Científicas, Campus de Sant Joan, Apartado de Correos 18, Sant Joan d’Alacant, 03550 Alicante, Spain Lynda Bergsma, Ph.D. Mel & Enid Zuckerman College of Public Health, University of Arizona, P.O. Box 245209, Tucson, Arizona 85724 Daniel M. Bernstein Department of Psychology, Kwantlen Polytechnic University, 12666 – 72nd Avenue, Surrey, BC, Canada Monelle Bertrand Unité de Nutrition, CHU Larrey, Toulouse cedex Carole A. Bisogni Division of Nutritional Sciences, 183 MVR Hall, Cornell University, Ithaca, NY 14853, USA Christine E. Blake Department of Health Promotion, Education, and Behavior, Arnold School of Public Health, University of South Carolina, Columbia, SC, USA Eva Blomstrand The Åstrand Labaratory, The Swedish School of Sport and Health Sciences, Box 5626 114 86 Stockholm, Sweden Miriam E. Bocarsly Department of Psychology, Princeton University, Princeton, NJ, USA Sandra D.M. Bot, Ph.D. Department of General Practice and the EMGO Institute for Health and Care Research, VU University Medical Center, Amsterdam, the Netherlands Wayne A. Bowers, Ph.D. Department of Psychiatry, University of Iowa, 2916 John Pappajohn Pavilion, Iowa City, IA 52242
xxxiv
L. Mallory Boylan, Ph.D., RD, LD Department of Nutrition, Hospitality & Retailing, Texas Tech University, Lubbock, TX, USA Emma J. Boyland Kissileff Laboratory for the Study of Human Ingestive Behaviour, School of Psychology, Eleanor Rathbone Building, Bedford Street South, University of Liverpool, Liverpool, L69 7ZA, UK Christophe Breton Neurostress (EA4347), Equipe Dénutritions Maternelles Périnatales, Bâtiment SN4, Université des Sciences et Technologies de Lille, France Marie-Claude Brindisi, M.D. Endocrinology and Nutrition Department, CHU le Bocage (Dijon University Hospital), BP 77908, 21079 Dijon Cedex Russell Brown Department of Obstetrics & Gynaecology, IWK Health Centre, Dalhousie University, Halifax, NS, Canada Thomas A. Brunner ETH Zurich, Institute for Environmental Decisions, Consumer Behavior, Universitaetsstrasse, Zurich, Switzerland Eleanor J. Bryant Centre for Psychology Studies, Social Sciences and Humanities, Richmond House, University of Bradford, West Yorkshire, BD7 1DP, UK S. Alexandra Burt, Ph.D. Department of Psychology, Michigan State University, East Lansing, MI, USA Pietro Caliandro Institute of Neurology, Largo F. Vito 1, 00168, Roma, Italy and Fondazione Pro Iuventute Don Carlo Gnocchi, Roma, Italy Maria Valeria Camboni, Bpsych Department of Psychology, University of Cagliari, Cagliari, Italy Jameason D. Cameron Behavioral and Metabolic Research Unit, School of Human Kinetics, University of Ottawa, Ottawa, Ontario, Canada L. Arthur Campfield, Ph.D. Department of Food Science and Human Nutrition, College of Applied Human Sciences, Colorado State University, Fort Collins, CO 80523-1571, USA Lucia Camporese, PsyD AIDAP Verona, Via Sansovino, Verona, Italy Mehmet Cansev Uludag University Medical School, Department of Pharmacology, Gorukle, 16059 Bursa, Turkey
Contributors
Contributors
xxxv
Antonio Capurso, M.D. Department of Geriatrics, Center for Aging Brain, Memory Unit, University of Bari, Bari, Italy Cristiano Capurso Department of Geriatrics, University of Foggia, Foggia, Italy Céline Caquineau Centre for Integrative Physiology, University of Edinburgh, Edinburgh, UK Daniel P. Cardinali, M.D., Ph.D. Departamento de Docencia e Investigación, Pontificia Universidad Católica Argentina, Buenos, Aires, Argentina Amanda Carson School of Health, Nursing and Midwifery, University of the West of Scotland, Beech Grove, Ayr, UK Natalie A. Ceballos Department of Psychology, Texas State University, San Marcos, TX, USA Kim M. Cecil, Ph.D. Cincinnati Children’s Environmental Health Center at the Cincinnati Children’s Hospital Medical Center, Departments of Radiology, Pediatrics, Neuroscience and Environmental Health, University of Cincinnati College of Medicine, Cincinnati, OH, USA and Cincinnati Children’s Hospital Medical Center, Department of Radiology/Imaging Research Center MLC 5033, 3333 Burnet Avenue, Cincinnati, OH 45229, USA Hellas Cena, M.D. Department of Health Sciences, Section of Human Nutrition, University of Pavia, Pavia, Italy Elena Centis, Ph.D. Clinical Dietetics, University of Bologna, Bologna, Italy Antonio Cepeda-Benito Department of Psychology, Henderson Hall, Texas A&M University, College Station, TX, USA Seung Hun Cha School of Medicine, Department of Biological Chemistry, The Johns Hopkins University, Baltimore, MD, USA Evonne J. Charboneau, M.D. School of Medicine, Vanderbilt University, Nashville, TN, USA Christine Charrueau, PharmD, Ph.D. Laboratoire de Pharmacie Galénique EA 2498, Faculté des Sciences Pharmaceutiques et Biologiques, Université Paris Descartes, Paris Cedex, France
xxxvi
Catherine Chaumontet INRA, CNRH-IdF, UMR914 Nutrition Physiology and Ingestive Behavior, Paris, France and AgroParisTech, CNRH-IdF, UMR914 Nutrition Physiology and Ingestive Behavior, Paris, France Anastasia Chellino, B.S. Health Psychology, Ingestive Behavior Laboratory, Pennington Biomedical Research Center, Baton Rouge, LA, USA and Department of Natural Sciences, University of Education Heidelberg, Heidelberg, Germany Hui Chen Faculty of Science, Department of Medical and Molecular Bioscience, University of Technology, Sydney, Australia and School of Medical Sciences, Department of Pharmacology, University of New South Wales, Sydney, Australia Kai-Chun Cheng Department of Psychosomatic Internal Medicine, Kagoshima University Graduate School of Medical and Dental Sciences, Kagoshima, Japan Sergio Chieffi Department of Experimental Medicine, Section of Human Physiology, and Clinical Dietetic Service, Second University of Naples, Naples, Italy W.I. Cho Department of Food and Animal Biotechnology, College of Agriculture and Life Sciences, Seoul National University, San 56-1 Sillim-dong, Gwanak-gu, Seoul 151-921, Korea Derrick L. Choi, B.S. Department of Psychiatry, University of Cincinnati, Cincinnati, OH, USA Chin Moi Chow Delta Sleep Research Unit, Discipline of Exercise and Sport Science, The University of Sydney, Lidcombe NSW 1825, Australia Yvonne Christley School of Health, Nursing and Midwifery, University of the West of Scotland, Beech Grove, Ayr, UK J.K. Chun Department of Food and Animal Biotechnology, College of Agriculture and Life sciences, Seoul National University, Seoul, Korea Cicero G. Coimbra Laboratory of Clinical and Experimental Pathophysiology, Department of Neurology and Neurosurgery, Federal University of São Paulo (UNIFESP), São Paulo SP BRAZIL
Contributors
Contributors
xxxvii
Anna M. Colacicco, Ph.D. Department of Geriatrics, Center for Aging Brain, Memory Unit, University of Bari, Bari, Italy Nelson Barros Colauto Laboratório de Biologia Molecular, Universidade Paranaense, Praça Mascarenhas de Moraes, Umuarama-PR, Brazil Cian E. Collins, FRCSI (Ophth), MRCOphth Princess Alexandra Eye Pavilion, Chalmers Street, Edinburgh EH3 9HA Mark Conner Institute of Psychological Sciences, University of Leeds, Leeds, UK James R. Connor, Ph.D. Milton S. Hershey Medical Center, Pennsylvania State University, Hershey, PA, USA Richard W.I. Cooke School of Reproductive and Developmental Medicine, University of Liverpool, First Floor, University Department, Liverpool Women’s Hospital, Crown Street, Liverpool, L8 7SS, UK John Correa, B.S. Health Psychology, Ingestive Behavior Laboratory, Pennington Biomedical Research Center, Baton Rouge, LA, USA Vassiliki Costarelli, BSc, MSc, Ph.D, R.Nutr. Human Ecology Laboratory, Department of Home Economics and Ecology, Harokopio University, 70 El. Venizelou Ave, 17671 Kallithea, Athens, Greece Mihai Covasa INRA, Écologie et Physiologie du Système Digestif, 78350 Jouy-en-Josas, France Ronald L. Cowan M.D., Ph.D. School of Medicine, Vanderbilt University, Nashville, TN, USA Sabrina Crepin Service de Toxicologie et Pharmacologie - Pharmacovigilance, Centre Hospitalier Universitaire, F-87042 Limoges, France Kristen M. Culbert, M.A. Department of Psychology, Michigan State University, East Lansing, MI, USA Maria Czyzewska Department of Psychology, Texas State University, 601 University Drive, San Marcos, TX, USA Alessia D’Introno, Ph.D. Department of Geriatrics, Center for Aging Brain, Memory Unit, University of Bari, Bari, Italy Ippeita Dan Sensory & Cognitive Food Science Laboratory, National Food Research Institute, Kannondai, Tsukuba, Ibaraki, Japan
xxxviii
Daniel Rigaud, M.D. Department of Endocrinology and Nutrition, CHU Le Bocage (Dijon University Hospital), BP 77908, 21079 Dijon Cedex, France Nicolas Darcel INRA, CNRH-IdF, UMR914 Nutrition Physiology and Ingestive Behavior, Paris, France and AgroParisTech, CNRH-IdF, UMR914 Nutrition Physiology and Ingestive Behavior, Paris, France Undurti N. Das, M.D., FAMS UND Life Sciences 13800 Fairhill Road, #321, Shaker Heights, OH, 44120, USA Jawaharlal Nehru Technological University, Kakinada- 533 003, Andhra Pradesh, India Christopher Davids Department of Psychology, Cornell College, Mt. Vernon, IA, USA Fiona Davies, B.A. (Hons), M.Psych. (Applied) Weight Management Services, The Children’s Hospital at Westmead, Westmead, Australia and Gymea Lily Psychotherapy Centre, Sutherland, Australia Caroline Davis Department of Psychiatry, University Health Network, Toronto, Ontario, Canada Jon F. Davis, Ph.D. Department of Psychiatry, University of Cincinnati, Genome Research Institute Building E, Lab 334, 2170 East Galbraith Road, 45237 Cincinnati, OH, USA Gwenaelle Diene Centre de référence du syndrome de Prader-Willi, Hôpital des Enfants, Toulouse Cedex, France Bruno De Luca Department of Experimental Medicine, Section of Human Physiology, and Clinical Dietetic Service, Second University of Naples, Naples, Italy Mary Jane De Souza, Ph.D., FACSM Women’s Health and Exercise Laboratory, Noll Laboratory, Department of Kinesiology, Penn State University, University Park, PA 16802, USA Fabien Delahaye Neurostress (EA4347), Equipe Dénutritions Maternelles Périnatales, Bâtiment SN4, Université des Sciences et Technologies de Lille, France Sylvie Deloof Neurostress (EA4347), Equipe Dénutritions Maternelles Périnatales, Bâtiment SN4, Université des Sciences et Technologies de Lille, France
Contributors
Contributors
xxxix
Jean-Claude Desport Université de Limoges, Institut de Neuroépidémiologie et de Neurologie Tropicale, EA 3174 NeuroEpidémiologie Tropicale et Comparée, Limoges, France and Service de Gastro-Hépato-Entérologie, Unité de Nutrition, Centre Hospitalier Universitaire, Limoges, France Silvia Di Domizio, R.D. Clinical Dietetics, University of Bologna, Bologna, Italy D.A.J. Dijck-Brouwe University Medical Center Groningen, Groningen, the Netherlands Atsu S.S. Dorvlo, Ph.D. Department of Mathematics and Statistics, Sultan Qaboos University, Muscat, Sultanate of Oman Éric Doucet Behavioral and Metabolic Research Unit, School of Human Kinetics, University of Ottawa, Ontario, Ottawa, Canada, K1N 6N5 Alison J. Douglas Centre for Integrative Physiology, University of Edinburgh, Hugh Robson Building, George Square, Edinburgh, EH8 9XD, UK Ronald G. Downey, Ph.D. Kansas State University, Manhattan, KS, USA Vicky Drapeau Department of physical education, Laval University, Quebec City, Canada Robert Drewett Science Laboratories, Durham University, South Road, Durham, DH1 3LE, UK Tim Duffy School of Health, Nursing and Midwifery, University of the West of Scotland, Beech Grove, Ayr, UK Audrey DunnGalvin, Ph.D., Reg.Psychol.Ps.S.I. Clinical Investigations Unit, Department of Paediatrics and Child Health, Cork University Hospital, University College Cork (UCC), Wilton, Cork, Ireland Holiday Durham, Ph.D., RD Pennington Biomedical Research Center, Louisiana State University, Baton Rouge, LA, USA Isabelle Dutriez-Casteloot Neurostress (EA4347), Equipe Dénutritions Maternelles Périnatales, Bâtiment SN4, Université des Sciences et Technologies de Lille, France Louise Dye Human Appetite Research Unit, Institute of Psychological Sciences, University of Leeds, LS2 9JT, UK
xl
Nina Eikelis, Ph.D. Human Neurotransmitters Laboratory, Vascular and Hypertension Division, Baker IDI Heart and Diabetes Institute, Melbourne, Victoria 8008, Australia Mark Ellrichmann, M.D. Department of Medicine I, University Hospital Schleswig Holstein, Campus Kiel, Schittenhelmstr 12, 24105 Kiel, Germany Enzo Emanuele, M.D. Department of Health Sciences, Section of Psychiatry, University of Pavia, Pavia, Italy Keith M. Erikson Department of Nutrition, University of North Carolina at Greensboro, Greensboro, NC 27402-6170, USA Charlotte Erlanson-Albertsson, M.D., Ph.D. Appetite Control Unit, Division of Diabetes, Endocrinology and Metabolism, Department of Experimental Medical Science, Lund University, BMC B11, SE-221 84, Lund, Sweden Gabriella Morreale de Escobar Instituto de Investigaciones Biomédicas Alberto Sols, CSIC and Universidad Autónoma de Madrid, and Center for Biomedical Research on Rare Diseases (CIBERER), Madrid, Spain Ana I. Esquifino, Ph.D. Departamento de Bioquímica y Biología Molecular III, Universidad Complutense, 28040 Madrid, Spain Christine Feinle-Bisset Discipline of Medicine, Royal Adelaide Hospital, Adelaide, South Australia, Australia Dorianne Feldman, M.D., MSPT Department of Physical Medicine and Rehabilitation, School of Medicine, Johns Hopkins University, Baltimore, MD Marlís González Fernández, M.D., Ph.D. Department of Physical Medicine and Rehabilitation, School of Medicine, Johns Hopkins University, Baltimore, MD M. Carmen Fernández Departamento de Personalidad, Evaluación y Tratamiento Psicológico, Facultad de Psicología, Universidad de Granada, Granada, Spain Elizabeth Ferris University of Arizona, Mel & Enid Zuckerman College of Public Health, Tucson, AZ, USA and Department of Psychiatry, Center for Weight and Eating Disorders, University of Pennsylvania School of Medicine, Philadelphia, PA, USA Mick P. Fleming School of Health, Nursing and Midwifery, University of the West of Scotland, Beech Grove, Ayr, UK
Contributors
Contributors
xli
Eduardo Fonseca-Pedrero Department of Psychology, University of Oviedo, Centro de Investigación Biomédica en Red de Salud Mental, CIBERSAM, Plaza Feijoo, s/n, Oviedo, 33003, Spain Georg Förster Department of Psychology, University of Würzburg, Würzburg, Germany Joanna S. Fowler, Ph.D. Medical Department, Brookhaven National Laboratory, Upton, New York, USA Vincenza Frisardi, M.D. Department of Geriatrics, Center for Aging Brain, Memory Unit, University of Bari, Bristol, UK Gilles Fromentin INRA, CNRH-IdF, UMR914 Nutrition Physiology and Ingestive Behavior, F-75005 Paris, France and AgroParisTech, CNRH-IdF, UMR914 Nutrition Physiology and Ingestive Behavior, F-75005 Paris, France and UMR 914 INRA/AgroParisTech, 16 rue Claude Bernard, 75231 Paris Cedex 05, France Toshikatsu Fujii Department of Behavioural Neurology and Cognitive Neuroscience, Tohoku University Graduate School of Medicine, Sendai, Japan Tahany M. Gadalla, Ph.D., M.Sc., MMath Factor-Inwentash Faculty of Social Work, University of Toronto, 246 Bloor Street West, Toronto, Ontario, Canada, M5S 1V4 Salah Gariballa, M.D., FRCP Department of Internal Medicine, Faculty of Medicine and Health Sciences, UAE University, Al-Ain, United Arab Emirates Michael K. Georgieff, M.D. Department of Pediatrics, Neonatology Division & Center for Neurobehavioral Development, School of Medicine, University of Minnesota, 420 Delaware St. SE, MMC 39 (for mail), D-136 Mayo Building (for courier), Minneapolis, MN 55455, USA E.L. Gibson Clinical and Health Psychology Research Centre, Department of Psychology, Whitelands College, Roehampton University, Holybourne Avenue, London SW15 4JD, UK Giuseppe Derosa, M.D., Ph.D. Department of Internal Medicine and Therapeutics, University of Pavia, Pavia, Italy Cheryl Glover School of Health, Nursing and Midwifery, University of the West of Scotland, Ayr Campus, Beech Grove, Ayr, UK
[AU2]
xlii
Bertrand Godet Service de Neurologie, Centre Hospitalier Universitaire, Limoges, France Dina Gohar Department of Psychology, University of Pennsylvania, Philadelphia, PA, USA Iratxe Gonzalez Oscar Gonzalez-Perez, M.D., Ph.D Laboratory of Neuroscience, School of Psychology, University of Colima, Av. Universidad 333, Colima, Col 28040, México and Neuroscience Department, Centro Universitario de Ciencias de la Salud, University of Guadalajara, Guadalajara, Jal 44340, México E. González-Reimers Servicio de Medicina Interna, Hospital Universitario de Canarias, Ofra s/n 38320, La Laguna, Tenerife, Canary Islands, Spain S.A. van Goor University Medical Center Groningen, Nijlandspark 26, 9301 Bz Roden, The Netherlands Amy A. Gorin, Ph.D. Department of Psychology, Center for Health, Intervention, and Prevention, University of Connecticut, 2006 Hillside Road, Storrs, CT 06269-1248, USA Alja Gössler Department of Pediatric and Adolescent Surgery, General Hospital Klagenfurt, St.Veiter Str. 47, 9020 Klagenfurt, Austria Reiko Graham Department of Psychology, Texas State University, San Marcos, TX, USA Riccardo Dalle Grave, M.D. Department of Eating and Weight Disorder, Villa Garda Hospital, Via Montebaldo 89, I-37016 Garda (VR), Italy Michael W. Green Nutrition and Behaviour Laboratory, Psychology Department, School of Life and Health Sciences, Aston University, Birmingham, UK, B4 7ET Melinda A. Green Department of Psychology, Cornell College, 106E Law Hall, 600 First Street West, Mt. Vernon, IA 52314 G. Grugni Department of Auxology, IRCCS ‘S. Giuseppe Hospital’-Verbania, Roma, Italy Rubem Carlos Araújo Guedes, M.D., Ph.D. Departamento de Nutrição, Universidade Federal de Pernambuco, BR-50670901, Recife, PE, Brazil
Contributors
Contributors
xliii
Matthieu J. Guitton, Ph.D. Centre de Recherche Université Laval Robert-Giffard (CRULRG), 2601 Chemin de la Canardière F-6517, Québec, QC, G1J 2G3, Canada and Faculty of Pharmacy, Laval University, Quebec City, QC, Canada Alissa A. Haedt, M.A. Department of Psychiatry, University of Iowa, Iowa City, IA, USA Jason C.G. Halford Kissileff Laboratory for the Study of Human Ingestive Behaviour, School of Psychology, University of Liverpool, Liverpool, UK John Hall, Ph.D. Deakin Business School, Deakin University, Melbourne, Victoria, Australia Hélène Hanaire Unité de Nutrition, CHU Larrey, Toulouse cedex Rhona M. Hanning, Ph.D., RD Department of Health Studies and Gerontology, University of Waterloo, Waterloo, Ontario, Canada Joanne A. Harrold Kissileff Laboratory for the Study of Human Ingestive Behavior, School of Psychology, University of Liverpool, Liverpool, UK Sybil L. Hart, Ph.D. Texas Tech University, Department of Human Development and Family Studies, Lubbock, TX 79409-1230 Jenni Harvey Division of Medical Sciences, Centre for Neuroscience, Ninewells Hospital and Medical School, University of Dundee, Dundee DD1 9SY, UK Kenji Hashimoto Division of Clinical Neuroscience, Chiba University Center for Forensic Mental Health, 1-8-1 Inohana, Chiba 260-8670, Japan Michio Hashimoto Department of Environmental Physiology, Shimane University Faculty of Medicine, Izumo, Shimane 693–8501, Japan Remco C. Havermans Department of Clinical Psychological Science, Maastricht University, Maastricht, 6200 MD, The Netherlands Narasimha Hegde, Ph.D. Department of Nutritional Sciences, Pennsylvania State University, University Park, PA, USA Christopher Herrera Delta Sleep Research Unit, Discipline of Exercise and Sport Science, The University of Sydney, Lidcombe, Australia
xliv
Marion Hetherington Institute of Psychological Sciences, University of Leeds, Leeds, West Yorkshire, UK Takashi Higuchi Department of Integrative Physiology, University of Fukui, Eiheij-cho, Matsuoka, Fukui, 910-1193, Japan Anke Hinney Klinik für Psychiatrie und Psychotherapie des Kindes- und Jugendalters, LVRKlinikum Essen, Kliniken und Institut der Unisitat Duisburg-Essen, Essen, Germany Angelica Lindén Hirschberg, M.D., Ph.D. Department of Obstetrics and Gynaecology, Karolinska University Hospital, Stockholm, Sweden Claude Marcel Hladik UMR CNRS/MNHN 7206, Eco-Anthropologie et Ethnobiologie, Brunoy, France Bartley G. Hoebel Department of Psychology, Princeton University, Princeton, NJ, USA Wilhelm Hofmann University of Chicago, Booth School of Business, 5807 South Woodlawn Avenue, Chicago, IL 60615 Pleunie Hogenkamp Division of Human Nutrition, TI Food and Nutrition, Wageningen University and Research Centre, Wageningen, the Netherlands Eef Hogervorst, Ph.D. School of Sport, Exercise and Health Sciences, Loughborough University, Ashby Road, Loughborough, Leicestershire, United Kingdom R.H. Holloway, B.Sc. (Med), M.B.B.S., M.D., FRACP Discipline of Medicine, University of Adelaide, Adelaide, Australia Chang Hyung Hong Department of Psychiatry and Ajou Institute of Aging, Ajou University School of Medicine, San 5, Wonchun-dong, Youngtong-gu, Suwon-si, 443-749, Korea Mary Hostler Program Secretary, Cancer Research Center Hawaii, Prevention and Control Program, University of Hawaii at Manoa, Honolulu, HI, USA Jonathan O’B. Hourihane, M.B., D.M., MRCPI, FRCPCH Clinical Investigations Unit, Department of Paediatrics and Child Health, Cork University Hospital, University College Cork (UCC), Wilton, Cork, Ireland Alexa Hoyland Human Appetite Research Unit, Institute of Psychological Sciences, University of Leeds, Leeds, UK Wallace Huffman, Ph.D. Department of Economics, Iowa State University, Ames, IA 50011, USA
Contributors
Contributors
xlv
Ethan E. Hull, Ph.D. Children’s Hospital of Pittsburgh, Pittsburgh, PA Michael Huncharek, MD, MPH Meta-Analysis Research Group, 10 Sasanqua Circle, Columbia, SC 29209, USA Teresa Iannaccone Department of Experimental Medicine, Section of Human Physiology, and Clinical Dietetic Service, Second University of Naples, Naples, Italy Melinda J. Ickes, Ph.D. Health Promotion & Education, University of Cincinnati, Cincinnati, OH, USA Bruno P. Imbimbo, Ph.D. Research and Development Department, Chiesi Farmaceutici, Parma, Italy and Department of Geriatrics, University of Foggia, Foggia, Italy Syed A. Imran Department of Physiology & Biophysics, Dalhousie University, Halifax, NS, Canada Sheila M. Innis Department of Paediatrics, Child and Family Research Institute, University of British Columbia, 950 West 28th Avenue, Vancouver, BC V5Z 4H4, Canada Akio Inui Department of Psychosomatic Internal Medicine, Kagoshima University Graduate School of Medical and Dental Sciences 8-35-1 Sakuragaoka, Kagoshima 890-8520, Japan Shera C. Jackson, M.S., CLC Department of Human Development and Family Studies, Texas Tech University, Lubbock, TX, USA Sanjay Jaju, M.D. M.Phil. Directorate of Research & Studies, DG Planning, Ministry of Health (HQ), Muscat, Sultanate of Oman Anita Jansen Department of Clinical Psychological Science, Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands Margaret Jastran Division of Nutritional Sciences, Cornell University, Ithaca, NY, USA Aminah Jatoi, M.D. Department of Oncology, Mayo Clinic, 200 First Street SW, Rochester, Minnesota, USA Vickii B. Jenvey School of Psychology, Psychiatry, Monash University, Building 17, Clayton, Victoria 3800, Australia
xlvi
Amanda Jepson Department of Psychology, Cornell College, Mt. Vernon, IA, USA and Neuroscience Department, Centro Universitario de Ciencias de la Salud, University of Guadalajara, Guadalajara, Jal, México Amy R. Johnson Department of Nutrition, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA Fiona Jones Department of Psychology, University of Bedfordshire, Luton, LU1 3JU, UK Michelle P. Judge, Ph.D., RD School of Nursing, University of Connecticut, 231 Glenbrook Rd, Unit 2026, Mansfield, CT, USA Alexei B. Kampov-Polevoy, M.D., Ph.D. Department of Psychiatry, University of North Carolina at Chapel Hill, 237 Medical School Wing B, Campus Box 7160, Chapel Hill, NC 27599-7160 Natalie Kanakam Institute of Psychiatry, King’s College London, London, UK and Department of Academic Psychiatry, Eating, Disorder Research Unit, Guy’s Hospital, London, UK Masanori Katakura Department of Environmental Physiology, Shimane University Faculty of Medicine, Izumo, Shimane, Japan and Meta-Analysis Research Group, Columbia, SC, USA Patrick G. Kehoe, Ph.D. Dementia Research Group, Institute of Clinical Neurosciences, The John James Building, Frenchay Hospital, University of Bristol, Bristol, UK A.R. Kelles, Ph.D. Department of Nutrition, New York Chiropractic College, Seneca Falls, New York, USA Eva Kemps School of Psychology, Flinders University, GPO, Adelaide, SA 5001, Australia Alexandra P.F. Key, Ph.D. Vanderbilt Kennedy Center for Research on Human Development, 230 Appleton Place, Peabody Box 74, Vanderbilt University, Nashville, TN 37203 Joan Khoo Discipline of Medicine, Royal Adelaide Hospital, Adelaide, South Australia, Australia Hiroe Kikuchi Educational Physiology Laboratory, Graduate School of Education, The University of Tokyo, Tokyo, Japan
Contributors
Contributors
xlvii
Atsushi Kimura Sensory & Cognitive Food Science Laboratory, National Food Research Institute, Kannondai, Tsukuba, Ibaraki, Japan Neil King Human Movement Studies & Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Queensland, Australia Kelly L. Klump, Ph.D. Department of Psychology, Michigan State University, East Lansing, MI, USA Y. Kodra National Centre for Rare Diseases, Istituto Superiore di Sanità, Roma, Italy Anna Kokavec, Ph.D. School of Psychological Science, La Trobe University, Bendigo, Australia, 3552 Hitoshi Komuro Department of Neurosciences/NC30, Lerner Research Institute, The Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, OH 44195, USA Yutaro Komuro Department of Neurosciences/NC30, Lerner Research Institute, The Cleveland Clinic Foundation, Cleveland, OH, USA Takashi Kondoh, Ph.D. Institute of Life Sciences, Ajinomoto Co., Inc., Kawasaki, Japan and AJINOMOTO Integrative Research for Advanced Dieting, Graduate School of Agriculture, Kyoto University, Kyoto, Japan and Institute of Psychiatry, King’s College London, London, UK and Department of Academic Psychiatry, Eating Disorders Research Unit, Guy’s Hospital, London, UK Karel Krafka Department of Pediatric and Adolescent Surgery, General Hospital Klagenfurt, Klagenfurt, Austria David Kugler Department of Psychology, Cornell College, Mt. Vernon, IA, USA Jyrki T. Kuikka Imaging Center, Kuopio University Hospital, FIN-70211 Kuopio, Finland Tatsuro Kumada Department of Neurosciences/NC30, Lerner Research Institute, The Cleveland Clinic Foundation, Cleveland, OH, USA Angela B. Kydd School of Health Nursing and Midwifery, University of the West of Scotland, Hamilton Campus, Almada Street, Hamilton, Lanarkshire, ML3 0JB, UK
xlviii
Christine Laborie Neurostress (EA4347), Equipe Dénutritions Maternelles Périnatales, Bâtiment SN4, Université des Sciences et Technologies de Lille, France Virginie F. Labrousse Psynugen, Université Bordeaux 2, INRA, UMR1286, CNRS, UMR5226, Bâtiment UFR Pharmacie – 2ème Tranche – 2ème Etage, Case courrier 34, 33076 BORDEAUX Cedex Jeroen Lakerveld Department of General Practice and the EMGO Institute for Health and Care Research, VU University Medical Center, van der Boechorststraat 7, 1081 BT, Amsterdam, The Netherlands Sylvia M.S. Chung Chun Lam Riddet Institute, Massey University, Private Bag 11-222, Palmerston North, New Zealand Gavin Lambert, Ph.D. Human Neurotransmitters Laboratory, Vascular and Hypertension Division, Baker IDI Heart and Diabetes Institute, Melbourne, Australia Carol J. Lammi-Keefe, Ph.D., RD Human Nutrition and Food, School of Human Ecology, Louisiana State University, Baton Rouge, LA, USA Daniel Lamport Human Appetite Research Unit, Institute of Psychological Sciences, University of Leeds, Leeds, UK M. Daniel Lane Department of Biological Chemistry, School of Medicine, The Johns Hopkins University, Baltimore, Maryland 21205 Brenda Larson, M.D. Department of Oncology, Mayo Clinic, 200 First street SW, Rochester, Minnesota, USA Antonio Laverde Laboratório de Química de Produtos Naturais, Universidade Paranaense, Praça Mascarenhas de Moraes, Umuarama-PR, Brazil Clare Lawton Human Appetite Research Unit, Institute of Psychological Sciences, University of Leeds, Leeds, UK Sophie Laye, Ph.D. Psychoneuroimmunology, nutrition and genetics (Psynugen), Université Bordeaux 2, INRA, UMR1286, CNRS, UMR5226, Bâtiment UFR Pharmacie – 2ème Tranche – 2ème Etage, Case courrier 34, 146 rue Léo Saignat, 33076 BORDEAUX, Cedex Catherine Lebel Department of Biomedical Engineering, 1098 Research Transition Facility, University of Alberta, Edmonton, Alberta
Contributors
Contributors
xlix
Edward B. Lee, M.D., Ph.D. Institute for Diabetes, Obesity and Metabolism, School of Medicine, University of Pennsylvania, Philadelphia, PA, USA Kang Soo Lee Department of Psychiatry, Bundang Cha hospital, CHA University, Bundang-gu, Seongnam-si, Kyounggi-do, Korea Louis Lefebvre Department of Biology, McGill University, Montreal, Quebec, Canada David S. Leland, Ph.D. Department of Psychology, University of Wisconsin - Eau Claire, 105 Garfield Ave, Eau Claire, WI 54702, USA Serafín Lemos-Giráldez Department of Psychology, Centro de Investigación Biomédica en Red de Salud Mental (CIBERSAM), University of Oviedo, Oviedo, Spain Gareth Leng Centre for Integrative Physiology, University of Edinburgh, George Square, Hugh Robson Building, EH8 9XD Edinburgh, UK William R. Leonard Department of Anthropology, Northwestern University, 1810 Hinman Avenue, Evanston, IL 60208, USA Jean Lesage Neurostress (EA4347), Equipe Dénutritions Maternelles Périnatales, Bâtiment SN4, Université des Sciences et Technologies de Lille, France Moira S. Lewitt Faculty of Science & Technology, University of the West of Scotland, Paisley Campus, PA1 2BE, Paisley, Scotland, UK Ying Li Department of Neurosciences/NC30, Lerner Research Institute, The Cleveland Clinic Foundation, Cleveland, OH, USA Ying-Xiao Li Department of Psychosomatic Internal Medicine, Kagoshima University Graduate School of Medical and Dental Sciences, Kagoshima, Japan S.W. Lim Department of Biological System Engineering, University of Wisconsin-Madison, Madison, WI, USA Giani Andrea Linde Laboratório de Biologia Molecular, Universidade Paranaense, Praça Mascarenhas de Moraes, 4282, CEP 87.502-210, Umuarama-PR, Brazil Katajun Lindenberg Center for Psychotherapy Research, University Hospital Heidelberg, Bergheimer Straße 54, 69115 Heidelberg, Germany
l
Diana M. Lindquist, Ph.D. Department of Radiology/Imaging Research, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA Yoav Littner Department of Neonatology, Cleveland Clinic Children’s Hospital, The Cleveland Clinic Foundation, Cleveland, OH, USA Elizabeth F. Loftus Department of Psychology, University of California at Irvine, Irvine, CA, USA Antonio Carlos Lopes Department of Medicine, Federal University of São Paulo (UNIFESP), São Paulo, Brazil Alison Bryant Ludden, Ph.D. Psychology Department, College of the Holy Cross, Worcester, MA, USA D. A. de Luis Institute of Endocrinology and Nutrition, Medicine School and Unit of Investigation, Hospital Rio Hortega. Hospital Clinico, University of Valladolid, Valladolid, Spain Marie-Amélie Lukaszewsk Neurostress (EA4347), Equipe Dénutritions Maternelles Périnatales, Bâtiment SN4, Université des Sciences et Technologies de Lille, France Colleen Taylor Lukens, Ph.D. Pediatric Feeding and Swallowing Center, The Children’s Hospital of Philadelphia, 34th Street & Civic Center Boulevard, Philadelphia, PA 19104, USA Jennifer D. Lundgren, Ph.D. Department of Psychology, University of Missouri-Kansas City, 4825 Troost Avenue, Ste. 124, Kansas City, Missouri, 64110, USA Christine-Johanna Macare Institute of Psychiatry, King’s College London, London, UK and Department of Academic Psychiatry, Eating, Disorder Research Unit, Guy’s Hospital, London, UK Ian G. Macreadie Bio21 Institute, University of Melbourne, 30 Flemington Road, Melbourne, VIC 3010, Australia Sandra Maestro Division of Child Neuropsichiatry, Stella Maris Scientific Institute, University of Pisa, Italy Mario Maj Department of Psychiatry, University of Naples SUN, Naples, Italy Alexandros Makriyannis Center for Drug Discovery, Northeastern University, Boston, MA, USA
Contributors
Contributors
li
Izaskun Marañon Navarra Health Service (Osasunbidea), Itsasargi 11-3, 20280 Hondarribia, Spain Glenn R. Marland School of Health, Nursing and Midwifery, University of the West of Scotland, Dumfries Campus, Dudgeon House, Bankend Road, Dumfries, DG1 4ZN, UK Monica Mars Division of Human Nutrition, TI Food and Nutrition/Wageningen University and Research Centre, Wageningen, the Netherlands Corby K. Martin, Ph.D. Ingestive Behavior Laboratory, Pennington Biomedical Research Center, 6400 Perkins Rd., Baton Rouge, LA, USA Colin R. Martin School of Health, Nursing and Midwifery, University of the West of Scotland, Ayr Campus, Beech Grove, Ayr, KA8 0SR, UK Catia Martins Department of Cancer Research and Molecular Medicine, Faculty of Medicine, Norwegian University of Science and Technology, Trondheim, Norway Giulio Marchesini, M.D. Unit of Metabolic Diseases & Clinical Dietetics, “Alma Mater Studiorum” University, University of Bologna, Policlinico S. Orsola, Via Massarenti 9, I-40138, Bologna, Italy Rebecca Marzocchi, M.D. Clinical Dietetics, University of Bologna, Bologna, Italy Alessandra Mauri Nutrition, Metabolism and Diabetes Unit, Ospedale Ca’ Foncello,Treviso, Italy Sylvain Mayeur Neurostress (EA4347), Equipe Dénutritions Maternelles Périnatales, Bâtiment SN4, Université des Sciences et Technologies de Lille, France F. Joseph McClernon, Ph.D. Health Behavior Neuroscience Research Program, Investigator, Center for Nicotine and Smoking Cessation Research, Duke University Medical Center, Durham, NC, USA Patrick O. McGowan Department of Psychiatry, McGill University, Montreal, QC, Canada Peter J. McLaughlin Department of Psychology, University of Connecticut, Storrs, CT, USA and Department of Psychology, Edinboro University of Pennsylvania, Edinboro, PA, USA James S. McTaggart Department of Physiology, Anatomy and Genetics, University of Oxford, Oxford, UK
lii
Michael J. Meaney Sackler Program for Epigenetics and Psychobiology, Douglas Institute – Research, Montreal, QC, Canada Giovanni Messina Department of Experimental Medicine, Section of Human Physiology, and Clinical Dietetic Service, Second University of Naples, Naples, Italy Jessie L. Miller Department of Psychiatry and Behavioural Neurosciences, Offord Centre for Child Studies, McMaster University, Chedoke Site, Central Building, 3rd Floor, 1200 Main Street West, Hamilton, Ontario, Canada, L8N 3Z5 Robert R. Miller, Jr. Biology Department, Hillsdale College, 278 N. West St., Dow Science 213, Hillsdale, MI 49242-1205, USA Emmanuelle Mimoun Centre de référence du syndrome de Prader-Willi/Hôpital des Enfants/ 330 av de Grande Bretagne/TSA 70034/31059 Toulouse Cedex 9/France Paola Miotto, M.D. Eating Disorders Unit, Department of Psychiatry, Conegliano, TV, Italy Katsumi Mizuno, M.D., Ph.D. Department of Pediatrics, Showa University of Medicine, 1-5-8 Hatanodai, Shinagawa-ku, Tokyo, 142-8666, Japan Markus Moessner Center for Psychotherapy Research, University Hospital Heidelberg, Heidelberg, Germany Christophe Moinard, Ph.D. Laboratoire de Biologie de la Nutrition EA 4466, Faculté des Sciences Pharmaceutiques et Biologiques, Université Paris Descartes, 4 avenue de l’Observatoire 75270 Paris Cedex 06, France Marcellino Monda, M.D. Department of Experimental Medicine, Section of Human Physiology, and Clinical Dietetic Service, Second University of Naples, Via Costantinopoli 16, 80138 Naples, Italy Naresh Mondraty Wesley Eating Disorders Centre, Wesley Hospital, 85 Milton Street, Ashfield, Sydney NSW, 2131 Australia Marie-Odile Monneuse Centre National de la Recherche Scientifique, UMR CNRS/MNHN 7206: Eco-Anthropologie et Ethnobiologie, Musée National d’Histoire Naturelle Département HNS - CP135, 57 rue Cuvier, 75231 Paris Cedex 05, France Chiara Montalto Division of Child Neuropsichiatry, Stella Maris Scientific Institute, University of Pisa, Italy
Contributors
Contributors
liii
Palmiero Monteleone Department of Psychiatry, University of Naples SUN, Largo Madonna delle Grazie, 80138 Naples, Italy Carmen C. Moran School of Psychology & National Wine and Grape Industry Centre, Charles Sturt University, Wagga Wagga, NSW, Australia Silvia Moreno Departamento de Psicología, Facultad de Humanidades y Ciencias de la Educación, Universidad de Jaén, Jaén, Spain Linda Morgan Division of Nutritional Sciences, Faculty of Health and Medical Sciences, University of Surrey, Guildford, UK Etsuro Mori Department of Behavioural Neurology and Cognitive Neuroscience, Tohoku University Graduate School of Medicine, Sendai, Japan Béatrice Morio, Ph.D. UMR1019 Nutrition Humaine, INRA, 63120 Saint Genès Champanelle, France and Université Clermont 1, UFR Médecine, 63000 Clermont-Ferrand, France Margaret J. Morris Department of Pharmacology, School of Medical Sciences, University of New South Wales, Sydney, NSW 2052, Australia Paul J. Moughan Riddet Institute, Massey University, Palmerston North, New Zealand Leonardo Munari Dipartimento di Farmacologia Preclinica e Clinica, Universita’ di Firenze, Florence, Italy Filippo Muratori Division of Child Neuropsichiatry, Stella Maris Scientific Institute, University of Pisa, IRCCS Stella Maris, Via dei Giacinti, 2-56018 Calambrone, Pisa, Italy F.A.J. Muskiet University Medical Center Groningen, Groningen, the Netherlands Sabine Naessén, M.D., Ph.D. Department of Obstetrics and Gynaecology, Karolinska University Hospital, SE-171 76 Stockholm, Sweden Naoko Narita, M.D., Ph.D. Institute of Education, Bunkyo University, 3337 Minamiogishima, Koshigaya-City, Saitama, 343-8511, Japan Masaaki Narita, M.D., Ph.D. Developmental and Regenerative Medicine, Mie University, Tsu, Mie, Japan
liv
Chantal Nederkoorn Department of Clinical Psychological Science, Maastricht University, Maastricht, the Netherlands N.Q. Nguyen, M.B.B.S., FRACP, Ph.D. Department of Gastroenterology and Hepatology, Royal Adelaide Hospital, North Terrace, Adelaide, SA, 5000, Australia Heather M. Niemeier, Ph.D. Department of Psychology, University of Wisconsin Whitewater, Whitewater, WI, USA Giel Nijpels, M.D., Ph.D. Department of General Practice and the EMGO Institute for Health and Care Research, VU University Medical Center, Amsterdam, the Netherlands Chiara Nuccitelli, Ph.D. Clinical Dietetics, University of Bologna, Bologna, Italy Daryl B. O’Connor Institute of Psychological Sciences, University of Leeds, LS2 9JT, UK Teresia O’Connor, M.D., MPH Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Houston, TX, USA Barry O’Mahony, Ph.D. School of Hospitality, Tourism and Marketing, Victoria University, Melbourne City, Vic 8001, Australia Kaeko Ogura Department of Pediatrics, National Rehabilitation Center for Persons with Disabilities, 1, Namiki 4-chome, Tokorozawa, Saitama 359-8555, Japan and Department of Behavioural Neurology and Cognitive Neuroscience, Tohoku University Graduate School of Medicine, Sendai, Japan Ken Ohashi Graduate School of Medicine, Department of Metabolic Diseases, The University of Tokyo, Tokyo, Japan Chuma O. Okere Department of Biological Sciences, Clark Atlanta University, Atlanta, GA, USA Leonardo A. Ortega Department of Psychology, Texas Christian University, Fort Worth, TX, USA Sarah E. Overington, Ph.D. Department of Biology, McGill University, 1205 Avenue Docteur Penfield, Montreal, Quebec, Canada H3A 1B1 Agostino Paccagnella Nutrition, Metabolism and Diabetes Unit, Ospedale Ca’ Foncello, 31100 Treviso, Italy
Contributors
Contributors
lv
L. Padua Institute of Neurology, Università Cattolica, Roma, Italy and Fondazione Pro Iuventute Don Carlo Gnocchi- Roma, Italy Mercedes Paíno Department of Psychology, Centro de Investigación Biomédica en Red de Salud Mental (CIBERSAM), University of Oviedo, Oviedo, Spain Francesco Panza, M.D., Ph.D. Department of Geriatrics, Center for Aging Brain, Memory Unit, University of Bari, Policlinico, Piazza Giulio Cesare, 11, 70124 Bari, Italy Mauricio R. Papini Department of Psychology, Texas Christian University, Fort Worth, TX 76129 Timo Partonen, M.D., Ph.D. Mood, Depression and Suicidal Behaviour Unit, Department of Mental Health and Substance Abuse Services, National Institute for Health and Welfare, Helsinki, Finland Elettra Pasqualoni, R.D. Department of Eating and Weight Disorder, Villa Garda Hospital, Garda, Italy Elettra Pasqualoni, R.D. AIDAP Verona, Via Sansovino, Verona, Italy Patrick Pasquet UMR CNRS/MNHN 7206: Eco-Anthropologie et Ethnobiologie, Brunoy, France Maria Beatrice Passani Dipartimento di Farmacologia Preclinica e Clinica, Universita’ di Firenze, Viale Pieraccini 6, Florence, Italy Kendra Patrick Michigan State University, East Lansing, MI, USA Karen Patte Faculty of Health, York University, Toronto, Ontario, Canada Nicole L.M. Pernat Department of Psychology, Kwantlen Polytechnic University, Surrey, BC, Canada Jennifer E. Phillips, Jaime A. Pineda, Ph.D. Department of Cognitive Science and Group in Neuroscience, University of California, San Diego, CA, USA J. Pinkney, M.D., FRCP Department of Diabetes and Endocrinology, Royal Cornwall Hospital, Peninsula College of Medicine and Dentistry, Truro, Cornwall, UK Pierluigi Politi Department of Health Sciences, Section of human nutrition and dietetics, University of Pavia, Via Bassi 21, I-27100 Pavia, Italy
lvi
Kathleen A. Potter Department of Biochemistry and Molecular Biology, Medical University of South Carolina, Charleston, SC, USA Michael L. Power, Ph.D. Research Department, American College of Obstetricians and Gynecologists, Washington, DC 20090–6920 and Nutrition Laboratory, Conservation Ecology Center, Smithsonian National Zoological Park, MRC 5503 Washington, DC 20013-7012 Antonio Preti, M.D. Department of Psychology, University of Cagliari, via is Mirrionis 1, 09123 Cagliari, Italy and Centro Medico Genneruxi, Via Costantinopoli 42, 09129 Cagliari, Italy Pierre-Marie Preux Université de Limoges, Institut de Neuroépidémiologie et de Neurologie Tropicale, EA 3174 NeuroEpidémiologie Tropicale et Comparée, Limoges, France Kathryn Proulx, Ph.D., PMHCNS-BC Mental Health Services, University of Massachusetts, Hills North, Amherst, MA 01103 Sarah E. Racine, M.A. Department of Psychology, Michigan State University, East Lansing, MI, USA Tri Budi W. Rahardjo, Ph.D. Center for Health Research, University of Indonesia, Jakarta, Indonesia Sayali C. Ranade National Brain Research Centre, National Highway-8, Near NSG Campus, Nainwal Mode, Manesar, Haryana 122050, India Patrick A. Randall Department of Psychology, University of Connecticut, Storrs, CT, USA Christoph Randler Biology, Department of Natural Sciences, University of Education Heidelberg, Im Neuenheimer Feld 561–2, D-69120 Heidelberg, Germany L.R. Ranganath, FRCPE, FRCPath, Ph.D. Department of Clinical Chemistry, Royal Liverpool University Hospital, Prescot Street, Liverpool, L7 8XP, UK Mohammad Tariqur Rahman Department of Biomedical Science, Kulliyyah of Science, International Islamic University Malaysia (IIUM), Jalan Istana, Bandar Indera Mahkota, 25200 Kuantan, Malaysia Carmen Rasmussen Department of Pediatrics, Glenrose Rehabilitation Hospital, Edmonton, Alberta
Contributors
Contributors
lvii
Christopher K. Rayner Discipline of Medicine, Royal Adelaide Hospital, Adelaide, South Australia, Australia Katherine Read Department of Psychology, Cornell College, Mt. Vernon, IA, USA Caroline Reid Centre for Addiction and Mental Health, University of Toronto, Toronto, Ontario, Canada Thomas Reinehr Department of Paediatric Nutrition Medicine, Vestische Hospital for Children and Adolescents, University of Witten/Herdecke, Dr. F. Steiner Str. 5, 45711 Datteln, Germany Britta Renner Department of Psychology, University of Konstanz, Konstanz, Germany Jared Edward Reser University of Southern California, Psychology Department, SGM 501, 3620 South McClintock Ave., Los Angeles, CA 90089-1061, USA Caroline Reverdy AUXIME - ODOROSMÊ, Les Grandes Roches, 69490, Saint Romain de Popey, France and PANCOSMA, R&D Department, 6 Voie des Traz, 1218 Le Grand Saconnex, Switzerland Paul Richardson Brain, Behaviour & Cognition Research Group, Psychology, Sheffield Hallam University, Collegiate Campus, Sheffield, UK S10 2BP A. Rinehart New York Chiropractic College, Seneca Falls, New York, USA Reeta Rintamäki, M.D., Ph.D. National Institute for Health and Welfare, Department of Mental Health and Substance Abuse Services, Mood, Depression and Suicidal Behaviour Unit, 30 (Mannerheimintie 166), FI-00271 Helsinki, Finland Patrick Ritz, M.D., Ph.D. Unité de Nutrition, CHU Larrey, TSA 30030, F-31059 Toulouse cedex 9, France Rachel E. Roberts Discipline of Medicine, School of Medicine, King’s College London, 2nd Floor, Henriette Raphael House, Guy’s Campus, London Bridge, London, SE1 1UL, UK Marcia L. Robertson Department of Anthropology, Northwestern University, Evanston, IL, USA Denise Robertson Division of Nutritional Sciences, Faculty of Health and Medical Sciences, University of Surrey, Guildford, UK
lviii
Athena Robinson, Ph.D. Department of Psychiatry and Behavioral Sciences, Stanford University, School of Medicine, 401 Quarry Road, Stanford, CA 94305-5722 Sonia Rodríguez-Ruiz Departamento de Personalidad, Evaluación y Tratamiento Psicológico, Facultad de Psicología, Universidad de Granada, Campus de la Cartuja s/n, 18071, Granada, Spain Dana L. Rofey, Ph.D. Children’s Hospital of Pittsburgh, 3414 Fifth Avenue, Room 128, Pgh, PA 15213, USA Ariz Rojas, M.A. Department of Psychology, University of South Florida, Tampa, FL, USA Susanne de Rooij, Ph.D. Department of Clinical Epidemiology and Biostatistics, Room J1b 210.1, Academic Medical Centre, University of Amsterdam, Meibergdreef 9, 1100 DD, Amsterdam, The Netherlands Matthew Rousu, Ph.D. Department of Economics, Susquehanna University, Selinsgrove, PA, USA C.W. le Roux Department of Metabolic Medicine, Imperial College Faculty of Medicine, London, UK Paul Rozin Department of Psychology, University of Pennsylvania, 3720 Walnut St, Philadelphia, PA 19104-6241, USA J.A. Rycroft Unilever Research and Development, Colworth House, Sharnbrook, Bedford, UK and Department of Experimental Medicine, Section of Human Physiology, and Clinical Dietetic Service, Second University of Naples, Naples, Italy Perminder Sachdev University of New South Wales, Neuropsychiatric Institute, The Prince of Wales Hospital, Randwick, NSW, Australia John D. Salamone Department of Psychology, University of Connecticut, 406 Babbidge Road, Storrs, CT 06269-1020, USA Anthony J. Saliba National Wine and Grape Industry Centre, Charles Sturt University, Locked Bag 588, Wagga Wagga, NSW, 2678, Australia F. Santolaria-Fernández Servicio de Medicina Interna, Hospital Universitario de Canarias, Tenerife, Canary Islands, Spain Fumiyo Sato Corporate Research & Development Group, SHARP Corporation
Contributors
Contributors
lix
Nūn Sava-Siva Amen-Ra, M.A., Ph.D. Amenta Academy of Theoretical Sciences (AATS), 25101Chimney-House Court, Damascus, MD, USA Donna Scarborough, Ph.D., CCC-SLP Department of Speech Pathology and Audiology, Miami University, 26 Bachelor Hall, Oxford, OH 45056, USA Sigfrido Scarpa Dip. di Chirurgia “P. Valdoni”, Centro di Ricerca in Neurobiologia “Daniel Bovet” CriN, Sapienza Università di Roma, via Antonio Scarpa, 14, 00161 Roma, Italy Jennifer L. Scheide Women’s Health and Exercise Laboratory, Noll Laboratory, Department of Kinesiology, Penn State University, University Park, PA, USA Susan M. Schembre, Ph.D., RD Cancer Research Center Hawaii, Prevention and Control Program, University of Hawaii, Manoa, 677 Ala Moana Blvd. Suite 200, Honolulu, Hawaii 96813 Anna Schierberl Scherr, B.A. Center for Health, Intervention, and Prevention, Department of Psychology, University of Connecticut, Storrs, CT, USA Gessica Schiavo Nutrition, Metabolism and Diabetes Unit, Ospedale Ca’ Foncello, Treviso, Italy Jay Schulkin Nutrition Laboratory, Conservation Ecology Center, Smithsonian National Zoological Park, Washington DC Harald T. Schupp Department of Psychology, University of Konstanz, 78457 Konstanz, Germany C.M. Schweitzer, M.D., Ph.D. Department of Internal Medicine, TweeSteden Ziekenhuis, Tilburg Christina L. Scott, Ph.D. Whittier College, 13406 E. Philadelphia St. Whittier, CA 90608, USA Kristen N. Segovia Department of Psychology, University of Connecticut, Storrs, CT, USA Claudio A. Serfaty Laboratory of Neural Plasticity, Fluminense Federal University, 100180, Niterói, RJ, 24001-970, Brazil Justine Joan Sheppard, Ph.D. Department of Biobehavioral Sciences, Teachers College, Columbia University, 525 W. 120th Street, New York, NY 10027
lx
Hossain Md Shahdat Department of Environmental Physiology, Shimane University Faculty of Medicine, Izumo, Shimane, Japan and Department of Biochemistry and Molecular Biology, Jahangirnagar University, Savar, Dhaka, Bangladesh Manoj Sharma, M.B.B.S, Ph.D. Health Promotion & Education, University of Cincinnati, Cincinnati, OH 45221-0068, USA Jason Shogren, Ph.D. Department of Economics and Finance, University of Wyoming, Laramie, WY, USA Lauren B. Shomaker, Ph.D. Department of Medical and Clinical Psychology, Uniformed Services University of the Health Sciences, Bethesda, MD, USA and Unit on Growth and Obesity, Program in Developmental Endocrinology and Genetics, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health Department of Health and Human Services, Bethesda, MD, USA M. Shroff, Ph.D. Department of Health Promotion, Education, and Behavior, University of South Carolina, Columbia, SC, USA Susana Sierra-Baigrie Department of Psychology, University of Oviedo, Plaza Feijoo, s/n, Oviedo 33003 Spain Kelly Siglin Department of Psychology, Cornell College, Mt. Vernon, IA, USA Bruno Simmen UMR CNRS/MNHN 7206, Eco-Anthropologie et Ethnobiologie, Brunoy, France and Department of Biochemistry and Molecular Biology, Jahangirnagar University, Savar, Dhaka, Bangladesh Julia Simner School of Philosophy, Psychology and Language Sciences, University of Edinburgh, 7 George Square, Edinburgh, EH8 9JZ, UK Rita Sinigaglia-Coimbra Electron Microscopy Center, Federal University of São Paulo (UNIFESP), São Paulo, Brazil and Laboratory of Clinical and Experimental Pathophysiology, Department of Neurology and Neurosurgery, Federal University of São Paulo (UNIFESP), São Paulo, Brazil
Contributors
Contributors
lxi
Kelly Sink Department of Psychology, University of Connecticut, Storrs, CT, USA Gudrun V. Skuladottir Department of Physiology, Faculty of Medicine, School of Health Sciences, University of Iceland, Vatnsmyrarvegi 16, IS-101 Reykjavik, Iceland Alexandra C. Smith Department of Food Science and Human Nutrition, College of Applied Human Sciences, Colorado State University, Fort Collins, CO, USA Veronique de Smedt-Peyrusse Psynugen, Université Bordeaux 2, INRA, UMR1286, CNRS, UMR5226, Bâtiment UFR Pharmacie – 2ème Tranche – 2ème Etage, Case courrier 34, 33076 BORDEAUX Cedex Françoise J. Smith Department of Food Science and Human Nutrition, College of Applied Human Sciences, Colorado State University, Fort Collins, CO, USA Marek Smulczyk The Maria Grzegorzewska Academy of Special Education, Institute of Applied Psychology, 40 Szczesliwicka Street, 02-353 Warsaw, Poland J. Josh Snodgrass Department of Anthropology, University of Oregon, Eugene, OR, USA Harriëtte M. Snoek Behavioural Science Institute, Radboud University Nijmegen, The Netherlands and Agricultural Economics Research Institute, Wageningen University and Research Center, P.O. Box 8130, 6700 EW Wageningen, The Netherlands Vincenzo Solfrizzi, M.D., Ph.D. Department of Geriatrics, Center for Aging Brain, Memory Unit, University of Bari, Policlinico, Piazza Giulio Cesare, 11, 70124 Bari, Italy Annette Stafleu TI Food and Nutrition/TNO Quality of Life, 3700 AJ Zeist, The Netherlands J.E.R. Staddon Department of Psychology and Neuroscience, Duke University, Durham, NC, USA and Department of Psychology, University of York, York, UK Malgorzata Starzomska The Maria Grzegorzewska Academy of Special Education, Institute of Applied Psychology, 40 Szczesliwicka Street, 02-353 Warsaw, Poland Andreas Stengel, M.D. CURE: Digestive Diseases Research Center, David Geffen School of Medicine, University of California, Los Angeles, Building 115, Room 117, Los Angeles, CA, USA
lxii
Ashley Stillman Department of Psychology, Cornell College, Mt. Vernon, IA, USA Wolfgang Stroebe Department of Social and Organizational Psychology, Utrecht University, Utrecht, the Netherlands James Stubbs Slimming World, Alfreton, Derbyshire, UK Albert J. Stunkard, M.D. Center for Weight and Eating Disorders, Department of Psychiatry, School of Medicine, University of Pennsylvania, 3535 Market Street, Philadelphia, PA, 19104-3309, USA Shiro Suda, M.D., Ph.D. Research Center for Child Mental Development, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan Alessandro Suppini, Ph.D. Clinical Dietetics, University of Bologna, Bologna, Italy Nan M. Sussman, Ph.D. Department of Psychology, College of Staten Island, City University of New York, Staten Island, New York Timothy Swartz INRA, Écologie et Physiologie du Système Digestif, 78350 Jouy-en-Josas, France Moshe Szyf Sackler Program for Epigenetics and Psychobiology, and Department of Pharmacology and Therapeutics, McGill University, 3655 Sir William Osler Promenade, room 1309, Montreal, QC, Canada H3G 1Y6 Yvette Taché, Ph.D. CURE: Digestive Diseases Research Center and Center for Neurovisceral Sciences & Women’s Health, Digestive Diseases Division, David Geffen School of Medicine at UCLA and VA Greater Los Angeles Healthcare System, Los Angeles, CA, USA Domenico Tafuri Department Studies of Institutions and Territorial Systems, Faculty of Movement Science, University of Naples “Parthenope”, Naples, Italy and Faculty of Movement Science, University of Naples “Parthenope”, Naples, Italy Nori Takei, M.D., M.Sc., Ph.D. Research Center for Child Mental Development and United Graduate School of Child Development, Hamamatsu University School of Medicine, 1-20-1 Handayama, Higashi-ku Hamamatsu Shizuoka, 431-3192 Japan and Division of Psychological Medicine, Institute of Psychiatry, London, SE5 8AF, UK
Contributors
Contributors
lxiii
Masato Takii Department of Psychosomatic Medicine, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka, 812-8582, Japan Yoshiyuki Takimoto Graduate School of Medicine, Department of Stress Sciences and Psychosomatic Medicine, The University of Tokyo, Tokyo, Japan Marian Tanofsky-Kraff, Ph.D. Department of Medical and Clinical Psychology, Uniformed Services University of the Health Sciences, Bethesda, MD, USA and Unit on Growth and Obesity, Program in Developmental Endocrinology and Genetics, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Department of Health and Human Services, Bethesda, MD, USA D. Taruscio National Centre for Rare Diseases, Istituto Superiore di Sanità, Roma, Italy and Fondazione Pro Iuventute Don Carlo Gnocchi, Roma, Italy Maithe Tauber Centre de référence du syndrome de Prader-Willi, Hôpital des Enfants, 330 av de Grande Bretagne, TSA 70034, 31059 Toulouse Cedex 9, France Raed Tayyem General surgery department, Ayr Hospital, Ayr, UK Mami Tazoe, M.S. The Department of Clinical Psychology, Japan Lutheran College, Mitaka, Tokyo, Japan Ababyue Tegene US Department of Agriculture, Economic Research Service, Washington DC Frank Telang, M.D. Neuroimaging Laboratory, National Institute on Alcohol Abuse and Alcoholism, Upton, New York, USA Salvadeo Sibilla Anna Teresa, M.D. Department of Internal Medicine and Therapeutics, University of Pavia, Pavia, Italy Panayotis K. Thanos, Ph.D. Behavioral Neuropharmacology & Neuroimaging Laboratory, National Institute on Alcohol Abuse and Alcoholism, Upton, New York, USA and Medical Department, Brookhaven National Laboratory, Upton, New York, USA David R. Thompson Department of Health Sciences and Department of Cardiovascular Sciences, University of Leicester, Leicester, UK
lxiv
J. Kevin Thompson, Ph.D. Department of Psychology, University of South Florida, Tampa, FL, USA Marika Tiggemann School of Psychology, Flinders University, Adelaide, Australia Daniel Tomé INRA, CNRH-IdF, UMR914 Nutrition Physiology and Ingestive Behavior, Paris, France and AgroParisTech, CNRH-IdF, UMR914 Nutrition Physiology and Ingestive Behavior, Paris, France Kunio Torii Institute of Life Sciences, Ajinomoto Co., Inc., Suzuki-cho 1-1, Kawasaki-ku, Kawasaki 210-8681, Japan Phu V. Tran, Ph.D. Center for Neurobehavioral Development, Neonatology Division & Center for Neurobehavioral Development, School of Medicine, Department of Pediatrics, University of Minnesota, MN, USA Janet Treasure Institute of Psychiatry, King’s College London, London, United Kingdom and Department of Academic Psychiatry, Eating Disorder Research Unit, Guy’s Hospital, Bermondsey Wing, SE1 9RT, London, United Kingdom Angelo Tremblay Division of Kinesiology, Laval University, Quebec City, QC, Canada Nhan Truong, Ph.D. Ismail H. Ulus Department of Pharmacology, Acibadem University Medical School, Gulsuyu Mahallesi, Maltepe, Istanbul, Turkey Erica L. Unger, Ph.D. Department of Nutritional Sciences, Pennsylvania State University, University Park, PA 16802 Tracy Vaillancourt Faculty of Education and School of Psychology, University of Ottawa, Ottawa, Ontario, Canada and Department of Psychology, Neuroscience, & Behaviour, McMaster University, Hamilton, Ontario, Canada Olga van den Akker Department of Psychology, Middlesex University, The Town Hall, The Burroughs, Hendon, London, NW4 4BT, UK
Contributors
Contributors
Koert van Ittersum, Ph.D. Georgia Institute of Technology, Atlanta, GA, USA V. Kiran Vemuri Center for Drug Discovery, Northeastern University, Boston, MA, USA Gianluigi Vendemiale Department of Geriatrics, University of Foggia, Foggia, Italy and Internal Medicine Unit, IRCSS Casa Sollievo dalla Sofferenza, San Giovanni Rotondo, Puglia, Italy. R.H. Verheesen, M.D. Regionaal Reuma Centrum Z.O. Brabant, Máxima Medisch Centrum, Ds. Th. Fliednerstraat 1, 5631 BM Eindhoven, The Netherlands Claudia Vicidomini Department of Experimental Medicine, Section of Human Physiology, and Clinical Dietetic Service, Second University of Naples, Naples, Italy Didier Vieau Environment Périnatal et Croissance (EA4489), Equipe Dénutritions Maternelles Périnatales, Bâtiment SN4, 2ème étage, Université des Sciences et Technologies de Lille, 59655 Villeneuve d’Ascq, France Andrea Viggiano Department Studies of Institutions and Territorial Systems, Faculty of Movement Science, University of Naples “Parthenope”, Naples, Italy and Faculty of Movement Science, University of Naples “Parthenope”, via Medina, Naples, Italy Valentina Viglione Division of Child Neuropsichiatry, Stella Maris Scientific Institute, University of Pisa, Italy Jaime Vila Departamento de Personalidad, Evaluación y Tratamiento Psicológico, Facultad de Psicología, Universidad de Granada, Granada, Spain Piergiuseppe Vinai, M.D. “GNOSIS” No Profit Research Group, V Langhe 64, 12060, Magliano Alpi, CN, Italy Vivianne H.M. Visschers ETH Zurich, Institute for Environmental Decisions, Consumer Behavior, Universitaetstrasse 22 CHN J75.2, 8092 Zurich, Switzerland Nora D. Volkow, M.D. Office of Director, National Institute on Drug Abuse, Rockville, MD, USA Yuji Wada Sensory & Cognitive Food Science Laboratory, National Food Research Institute, 2-1-12, Kannondai, Tsukuba, Ibaraki 305-8642, Japan
lxv
lxvi
Contributors
Bai-Ren Wang Institute of Neuroscience, Fourth Military Medical University, Xi’an, China Gene-Jack Wang, M.D. Medical Department, Building 490, Brookhaven National Laboratory, Upton, New York, 11973 USA Brian Wansink, Ph.D. Cornell University, 110 Warren Hall, Ithaca, NY 14853-7801, USA Carolina Werle, Ph.D. Grenoble Ecole de Management CERAG, Grenoble, France Stephen Whybrow Faculty of Health and Medical Sciences, University of Surrey, Guildford, GU2 7XH, UK Reinout W. Wiers Department of Psychology, University of Amsterdam, Amsterdam, the Netherlands Karen Wight Faculty of Health, York University, 223 Bethune College, 4700 Keele Street, Toronto, Canada and Centre for Addiction and Mental Health, University of Toronto, Toronto, Canada Michael Wilkinson Department of Obstetrics & Gynaecology, IWK Health Centre, Dalhousie University, University Avenue, Halifax, NS, Canada, B3K 6R8 and Department of Physiology & Biophysics, Dalhousie University, Halifax, NS, Canada, B3H 1X5 and Division of Endocrinology and Metabolism, Victoria General Hospital, Department of Medicine, Dalhousie University, Halifax, NS, Canada, B3H 2Y9 Gary Wittert Discipline of Medicine, Royal Adelaide Hospital, Adelaide, South Australia, Australia Sarah J. Woodruff, Ph.D., CEP Department of Kinesiology, University of Windsor, 401 Sunset Avenue, Windsor, Ontario N9B 3P4, Canada Li-Ze Xiong Department of Anesthesiology, Xijing Hospital, Fourth Military Medical University, 15 West Changle Rd, Xi’an, 710032, P.R. China Yajun Xu Department of Nutrition and Food Hygiene, School of Public Health, Peking University Health Science Center, No. 38 Xue Yuan Road, Hai Dian District, Beijing 100191, China Yoshiharu Yamamoto Educational Physiology Laboratory, Graduate School of Education, The University of Tokyo
Contributors
lxvii
Jack A. Yanovski, M.D., Ph.D. Section on Growth and Obesity, Program in Developmental Endocrinology and Genetics, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Department of Health and Human Services, 9000 Rockvile Pike, MSC 1103, Hatfield CRC, room IE-3330, Bethesda, Maryland 20892-1103, USA Amina Yesufu-Udechuku, Ph.D. CORE, Clinical Health Psychology, University College London, London, UK Kazuhiro Yoshiuchi Graduate School of Medicine, Department of Stress Sciences and Psychosomatic Medicine, The University of Tokyo, Tokyo, Japan B.S. Zanutto IIBM-Universidad de Buenos Aires, and IBYME-CONICET, Argentina Steven H. Zeisel, M.D., Ph.D. University of North Carolina at Chapel Hill, Nutrition Research Institute, 500 Laureate Way, Kannapolis, North Carolina 28081 Jie Zhao Department of Nutrition and Food Hygiene, School of Public Health, Peking University Health Science Center, Beijing, China Zheng-Hua Zhu Department of Anesthesiology, Xijing Hospital, Fourth Military Medical University, Xi’an, China and Centro Medico Genneruxi, Cagliari, Italy Nicolien Zijlstra Division of Human Nutrition, TI Food and Nutrition/Wageningen University and Research Centre, Wageningen, the Netherlands
wwwwwwwwwwwwwww
Part I
General and Normative Aspects: Evolutionary and Genetic
wwwwwwwwwwwwwww
Chapter 1
Diet and Brain Evolution: Nutritional Implications of Large Human Brain Size William R. Leonard, J. Josh Snodgrass, and Marcia L. Robertson
Abbreviations AA DHA DQ IGF-1 LC-PUFA MYA RMR RQ
Arachidonic acid Docosahexaenoic acid Dietary quality Insulin-like growth factor I Long-chain polyunsaturated fatty acid Million years ago Resting metabolic rate Respiratory quotient
1.1 Introduction The evolution of the human nutritional requirements is now receiving ever-greater attention among scientists from a variety of different fields, including nutritional science, anthropology and exercise science (Crawford 1992; Leonard et al. 1992, Leonard and Robertson 1994; Aiello and Wheeler 1995; Cordain et al. 2005). We are increasingly coming to realize that many of the key features that distinguish humans from other primates have important implications for our distinctive nutritional needs (Leonard 2002). The most notable of these features is our relatively large brain sizes, which are ~3 times the size our nearest primate relatives, the great apes (Martin 1989; McHenry and Coffing 2000). Because neural tissue has very high energy demands (~16 times that of muscle tissue; Kety 1957), our large brains exact a high metabolic cost. On average, adult humans spend some 350–400 kcal/ day on brain metabolism (Holliday 1986). Yet, despite the fact that humans have much larger brains per body weight than other primates or terrestrial mammals, the resting energy demands for the human body are no more than for any other mammal of the same size (Leonard and Robertson 1994). As a consequence, humans expend a much larger share of their resting metabolic rate (RMR) to “feed their brains” than other primates or non-primate mammals (Leonard et al. 2003).
W.R. Leonard (*) Department of Anthropology, Northwestern University, 1810 Hinman Avenue, Evanston, IL 60208, USA e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_1, © Springer Science+Business Media, LLC 2011
3
4
W.R. Leonard et al.
To support the high metabolic demands of our large brains, humans have diets of much higher quality – more dense in calories and nutrients – than other primates (Leonard and Robertson 1994). On average, we consume higher levels of dietary fat than other primates (Popovich et al. 1997), and much higher levels of key long-chain polyunsatured fatty acids (LC-PUFAs) that are critical to brain development (Cordain et al. 2001; Crawford et al. 1999). Moreover, humans also appear to be distinctive in their developmental changes in body composition. We have higher levels of body fatness than other primate species, and these differences are particularly evident in early in life. This chapter draws on both analyses of living primate species and the human fossil record to explore the influence of brain evolution on human nutritional needs. We begin by examining comparative dietary data for modern human groups and other primate species to evaluate the influence that variation in relative brain size has on dietary patterns among modern primates. We then turn to an examination of the human fossil record to examine when and under what conditions in our evolutionary past key changes in brain size and diet likely took place. Finally, we explore how the evolution of large human brains was likely accommodated by distinctive aspects of human growth and development that promote increased levels of body fatness from early in life.
1.2 Comparative Perspectives on Primate Dietary Quality The high energy costs of large human brains are evident in Fig. 1.1 which shows the allometric (scaling) relationship between brain weight (g) and RMR (kcal/day) for humans, 36 other primate species, and 22 non-primate mammalian species. The solid line denotes the best-fit regression for nonhuman
Fig. 1.1 Log-log plot of brain weight (BW;g) versus resting metabolic rate (RMR; kcal/day) for humans, 36 other primate species, and 22 non-primate mammalian species. The primate regression line is systematically and significantly elevated above the non-primate mammal regression. For a given RMR, primates have brain sizes that are three times those of other mammals, and humans have brains that are three times those of other primates
1 Diet and Brain Evolution
5
primate species, and the dashed line denotes the best-fit regression for the non-primate mammals. The data point for humans is denoted with a star. The slopes of the primate and non-primate mammalian log-log regressions are comparable (0.93 vs 0.90, respectively), whereas the Y-intercept of the primate regression is significantly greater than that of the non-primate mammals (−0.38 vs −0.83; P nonfood comparison (Reprinted from Kaurijoki et al. (2008). With permission of Wiley-Blackwell)
4.5.2 High- and Low-calorie Food > Nonfood Visual Stimuli Killgore and Yurgulun-Todd (2006) examined whether affect ratings predicted regional cerebral responses to high- and low-calorie food pictures among normal-weight adult women. Positive and negative affect had different effects on appetite-related regions depending on the calorie content of the food images. When viewing high-calorie images, positive affect was associated with increased activity in satiety-related regions (lateral orbitofrontal cortex), but when viewing low-calorie food, affect was associated with increased activity in hunger-related regions (medial orbitofrontal cortex and insular cortex).
4.5.3 Food > Nonfood Visual Stimuli While Fasting > 12 h St-Onge et al. (2005) employed fMRI to isolate cortical sites involved in the appreciation of food and nonfood pictures in healthy, normal-weight, and fasting individuals. Food, and nonfood images were presented to subjects both visually and tactilely. Brain regions that were significantly activated to a
56
J.T. Kuikka
Fig. 4.5 Association between genotype and brain activation. Effects of serotonin transporter promoter regulatory region polymorphism (5-HTTLPR) variation on the response for the fMRI signal intensity change food > nonfood in the left posterior cingulate cortex (PCC). The subjects with two copies of the long allele variant have the strongest activation (Reprinted from Kaurijoki et al. (2008). With permission of WileyBlackwell)
greater extent during the presentation of foods compared with nonfood items included the anterior cingulate cortex, superior temporal gyrus, parahippocampal gyrus, hippocampus, and the insula. This study as well as many others, suffer that in a group analysis alone, all findings are not necessarily observed in all subjects.
4.5.4 Cerebral Activity in Hunger and Satiety Führer et al. (2008) explored brain activity during hunger versus satiety using visual simulation. During the hunger condition, significantly enhanced brain activity was found in the left striatum and extrastriatal cortex, the inferior parietal lobe, and the orbitofrontal cortex. Stimulation with food images was associated with increased activity in insula, the left striatum, and extrastriatal cortex, and the anterior midprefrontal cortex. A significant interaction in activation pattern between the states of hunger and satiety and stimulation with food and nonfood images was found for the left anterior cingulate cortex, the superior occipital sulcus, and in the vicinity of the right amygdala. They emphasized that central nervous system activation is not only altered with hunger and satiety but that food and nonfood images have also specific effects on regional cerebral activity if exposure takes place in different states of satiety (Führer et al. 2008; Chechlacz et al. 2009).
4 Cerebral Activity to Visual Presentation of Food
57
4.6 Applications to Other Areas of Health and Disease The primary field of fMRI applications remains the cognitive neurosciences, but several applications are also found in surgical treatment planning (Vlieger et al. 2004), vascular diseases (Sharma et al. 2009) and preclinical studies (Marota et al. 2000).
4.7 Conclusions The imaging methods (fMRI, MEG, PET, and SPECT) for evaluating cerebral activity to visual presentation of food have been in use for a decade. The versatility of fMRI makes it a powerful tool and BOLD imaging has proven to have the capacity to objectively measure the response changes, despite the fact that the underlying mechanisms are still not fully understood. The increased sensitivity and specificity of higher-field (³3 T) MRI systems will improve our understanding of the underlying mechanisms related to the regulation of food intake and that way also on the pathophysiology of eating disorders. Further studies on regional activation and connectivity, coupled with psychological observations of the particular affective processes involved, will improve our understanding of food-related disorders.
Summary Points • The block design is the most efficient paradigm for studying activated areas. Blocks of stimulation allow more time to the BOLD signal to build up and thus statistical certainty between the task minus another condition increases. • In healthy subjects, a stronger focus locates on the left hemisphere (such as anterior cingulate cortex, insula, medial prefrontal cortex, orbitofrontal cortex, and posterior cingulate cortex). • The subjects with eating disorder exhibit greater activation in response to food pictures in a large number and usually bilateral regions than controls. • The present results suggest the possible genetically driven variation in the response of the certain brain regions to visual presentation of food. • However, the results are still very preliminary and heterogeneous due to the several confounding factors such as subjects’ selection, their nutritional and hormonal status, study protocol, statistical settings, etc. Acknowledgment The author thanks Dr. Leila Karhunen, Salla Kaurijoki, and Mervi Könönen for their valuable comments during preparation of this manuscript.
References Amaro Jr E, Barker GJ. Brain Cogn. 2006;60:220–32. Bellgowan PS, Saad ZS, Bandettini PA. Proc Natl Acad Sci USA. 2003;100:1415–9. Belliveau JW, Kennedy Jr DN, McKinstry RC, Buchbinder BR, Weiskoff RM, Cohen MS, et al. Science. 1991;254:716–9. Chechlacz M, Rotshtein P, Klamer S, Porubská K, Higgs S, Booth D, et al. Diabetologia. 2009;52:524–33. Cornier AM, von Kaenel SS, Bessesen DH, Tregellus JR. Am J Clin Nutr. 2007;86:965–71.
58
J.T. Kuikka
Del Pagiri A, Gautier JF, Chen K, Salbe AD, Ravussin E, Reiman E, et al. Ann NY Acad Sci. 2002;967:389–97. Frank GK, Kaye WH. Nucl Med Biol. 2005;32:755–61. Friston KJ, Zarahn E, Josephs O, Henson RN, Dale AM. Neuroimage. 1999;10:607–19. Führer D, Zysset S, Stumvoll M. Obesity. 2008;16:945–50. Genovese CR, Lazar NA, Nichols T. Neuroimage. 2002;15:870–8. Karhunen LJ, Lappalainen RI, Vanninen EJ, Kuikka JT, Uusitupa MI. Brain. 1997;120:1675–84. Kaurijoki S, Kuikka JT, Niskanen E, Carlson S, Pietiläinen KH, Pesonen U, et al. Clin Physiol Funct Imaging. 2008;28:270–6. Killgore WDS, Yurgelun-Todd DA. Int J Eat Disord. 2006;39:357–63. Killgore WD, Young AD, Femia LA, Bogorodzki P, Rogowska J, Yurgelun-Todd DA. Neuroimage. 2003;19:1381–94. Kringelbach ML. Neuroscience. 2004;126:807–19. Kuikka JT, Belliveau JW, Hari R. Eur J Nucl Med. 1996;23:737–40. Lang PJ, Bradley MM, Cuthbert, BN. International affective picture system (IAPS): instruction manual and affective ratings. In: The center for research in psychopathology. Gainesville: University of Florida; 2001. Marota JJA, Mandeville JB, Weisskoff RM, Moskowitz MA, Rosen BR, Kosofsky BE. Neuroimage. 2000;11:13–23. Morris JS, Dolan RJ. J Neurosci. 2001;21:5304–10. Norris DG. J Magn Reson Imaging. 2006;23:794–807. Ogawa S, Lee TM, Kay AR, Tank DW. Proc Natl Acad Sci USA. 1990;87:9868–72. Pauling L, Coryell CD. Proc Natl Acad Sci USA. 1936;22:210–6. Riera JJ, Schousboe A, Waagepetersen HS, Howarth C, Hyder F. Neuroimage. 2008;40:1436–59. Schienle A, Schäfer A, Hermann A, Vaitl D. Biol Psychiatry. 2009;65:654–61. Sharma N, Simmons LH, Jones PS, Day DJ, Carpenter TA, Pomeroy VM, et al. Stroke. 2009;40:1315–24. Siep N, Roefs A, Roebroeck A, Havermans R, Bonte ML, Jansen A, Behav Brain Res. 2009;198:149–58. Stice E, Spoor S, Bohon C, Small DM. Science. 2008;322:449–52. Stoeckesl LE, Weller RE, Cook III EW, Twieg DB, Knowlton RC, Cox JE. Neuroimage. 2008;41:636–47. St-Onge MP, Sy M, Heymsfield SB, Hirsch J. J Nutr. 2005;135:1014–8. Talairach J, Tournoux P. Co-planar stereotaxic atlas of the human brain. Stuttgart, New York: Georg Thieme Verlag; 1988. Uher R, Murphy T, Brammer MJ, Dalgleish T, Phillips ML, Ng VW, et al. Am J Psychiatry. 2004;161:1238–46. Vlieger EJ, Majoie CB, Leenstra S, Den Heeten GJ. Eur Radiol. 2004;14:1143–53.
Chapter 5
Infant Visual Acuity and Relationships with Diet and Nutrition Michelle P. Judge, Carol J. Lammi-Keefe, and Holiday Durham
Abbreviations ACP ARASCO DHA DHASCO ERG FPL GPCR Kg LCPUFA Mg SCO VEP
Acuity card procedure Arachidonic acid single-cell oil Docosahexaenoic acid Docosahexaenoic acid single-cell oil Electroretinogram Forced-choice preferential looking G-protein coupled reaction Kilogram Long chain polyunsaturated fatty acid Milligrams Single-cell oil Visual evoked potential
5.1 Introduction The visual system comprises a complex signaling system involving the retina, thalamus and primary visual cortex. In addition, other cortical regions are involved in the cognitive integration of visual information. Researchers have assessed the impact of long-chain polyunsaturated fatty acids (LCPUFA) on infant visual development by evaluating two main aspects of the infant’s visual system; retinal development using electroretinogram (ERG) and visual processing at the level of the primary visual cortex. Cortical measures of grated visual acuity include sweep visual evoked potential (VEP) and behavioral assessments such as the forced-choice preferential looking (FPL). Of the grated acuity measures, VEP measures cortical response directly and has less inherent variability. However, behavioral acuity measures provide the most direct measure of what an infant actually perceives and normative data are well established for these procedures (Neuringer 2000).
M.P. Judge (*) School of Nursing, University of Connecticut, 231 Glenbrook Rd, Unit 2026, Mansfield, CT, USA e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_5, © Springer Science+Business Media, LLC 2011
59
60
M.P. Judge et al.
Compared to other cells in the body, the retinal photoreceptors are known to have the highest content of the LCPUFA docosahexaenoic acid (DHA, 22:6 n-3) (Neuringer 2000). The gray matter in the visual cortex and multiple other cortical regions also have high levels of DHA. In rhesus monkeys, a diet deficient in LCPUFA during pregnancy impeded the accumulation of DHA in these tissues in the offspring. A 50% reduction of DHA in retinal tissue and 25% reduction in cerebral cortex DHA were reported for the offspring of rhesus monkeys fed a diet deficient in a-linonenic acid (18:3, n-3) during pregnancy compared to controls (Neuringer et al. 1986). Linolenic acid is the precursor to along the biosynthetic pathway. Neuringer and colleagues also demonstrated that these deficient animals had lower visual acuity measured using preferential looking and a prolonged recovery time of the dark-adapted ERG after a saturating flash. Behavioral methods for assessment of visual acuity have repeatedly shown that human infants have better visual function with the provision of DHA during the postnatal period (Birch et al. 1993a; Makrides et al. 1993, 1995; Carlson et al. 1996; Jorgensen et al. 1996; Hoffman et al. 2003). To date, there is one report (Judge et al. 2007) of improved infant visual acuity (assessed using the Acuity Card Procedure, ACP) related to maternal dietary DHA intake during pregnancy. Malcolm et al. used VEP (Malcolm et al. 2003a) and ERG (Malcolm et al. 2003b) to assess visual acuity related to maternal DHA supplementation, but no relationship was apparent. There was no significant treatment effect for ERG in the first postnatal week (Malcolm et al. 2003a) or VEP at 2.5 or 6.5 months of age (Malcolm et al. 2003b). Pregnant women in the USA and Canada have DHA intakes below the current recommended amount of 200 mg/day for optimal fetal development (Koletzko et al. 2008; Lewis et al. 1995; Innis and Elias 2003; Loosemore et al. 2004; Denomme et al. 2005). As the developing fetus has extremely limited capacity for deriving DHA from the 18 carbon precursor, the fetus is clearly at risk for DHA deficiency when maternal DHA intake is low. Given this risk for deficiency, marine, eggphospholipid, and single-cell oils (SCO) have been developed and explored for maternal and infant supplementation. Increased efforts toward community-based educational programs targeting women of child-bearing age with the goal of increasing DHA intake are recommended. Interruptions in visual development can delay the achievement of other developmental milestones with important long-term implications. This chapter will outline the contemporary research related to infant visual development, contrast assessment methodologies of visual acuity, and explore dietary factors that are fundamental to neurodevelopment.
5.2 Neuroanatomy of the Visual System The cerebral cortex is comprised of two distinct tissue layers, gray and white matter. The cortex surrounds the entire perimeter of the brain and comprises approximately 80% of the volume of the human brain (Kolb and Whishaw 2003). Neuronal cell bodies are found primarily in the gray matter. The white matter of the brain is comprised primarily of the myelinated nerve axons. Different areas of the cortex are responsible for receiving and processing stimuli. The four functional areas of the cerebral cortex include: motor, sensory, visual and auditory. The primary functional areas of the cerebral cortex receive initial sensory information and the secondary and tertiary areas are responsible for association and processing. Visual information goes to the primary visual cortex and processing related to assessment of shape, color and categorization occur in secondary and tertiary visual areas. These secondary and tertiary areas, include, motor, sensory and auditory cortical areas; all communicate with central brain structures for further processing.
5 Infant Visual Acuity and Relationships with Diet and Nutrition
61
5.3 The Visual System Vision and behavioral responses to visual stimuli rely on a complex system of communication between the retina, thalamus and multiple regions of the cerebral cortex. Visual processing is initiated when the photoreceptors of the retina receive a visual stimulus that triggers signal transduction. The impulse travels down the optic tract where signal transductions are sent to the opposite cerebral hemisphere after crossing at the optic chiasm (Fig. 5.1). After passing the optic chiasm, the optic tract sends the signal to the lateral geniculate nucleus of the thalamus, which relays visual signaling to the primary visual cortex. When visual information is sent to the primary visual cortex it is disseminated to various cortical areas by two main pathways, the dorsal and ventral streams, for higher cortical processing. The dorsal stream projections are sent to the parietal lobe to process information pertaining to location and movement (Zigmond et al. 1999). The ventral stream projects to the temporal lobe where more integrative processing occurs, e.g., object and facial recognition (Fig. 5.2). The process of facial recognition relating to memory is conducted primarily by the limbic system which involves the entorhinal cortex, hippocampus and amygdala. At the level of the membrane, G protein-coupled receptor (GPCR) signal transduction occurs in the retina, brain, and nervous system and has been attributed to signaling pathways leading to visual and cognitive function (Salem et al. 2001; Nui et al. 2004). G proteins are binding proteins that use neurotransmitters to collectively activate receptors (referred to as coupled activation) (Zigmond et al. 1999). Action of the neurotransmitter utilizes a second messenger system for the regulation of the
Visual Stimulus
Retina
Fig. 5.1 Visual processing is initiated when the photoreceptors of the retina receive a visual stimulus that triggers signal transduction. The impulse travels down the optic tract where signal transductions are sent to the opposite cerebral hemisphere after crossing at the optic chiasm. After passing the optic chiasm, the optic tract sends the signal to the lateral geniculate nucleus of the thalamus, which relays visual signaling to the primary visual cortex
Optic Chiasm
Lateral Geniculate Nuclei
Primary Visual Cortex
62
M.P. Judge et al.
Dorsal Stream: Parietal Lobe Location & Movement
Ventral Stream: Temporal Lobe
Integrative Processing & Memory Entorhinal Cortex (ER) Hippocampus (H) Amygdala (A)
Primary Visual Cortex
Fig. 5.2 When visual information is sent to the primary visual cortex it is disseminated to various cortical areas by two main pathways, the dorsal and ventral streams. The dorsal stream projections are sent to the parietal lobe to process information pertaining to location and movement. The ventral stream projects to the temporal lobe where more integrative processing occurs, e.g., object and facial recognition involving the entorhinal cortex (ER), hippocampus (H) and amygdale (A)
synthesis of protein kinases. The protein kinase produced as a result of the GPCR depends upon the neurotransmitter coupling the reaction (Zigmond et al. 1999). The specific type of protein kinase generated will initiate a phosphorylation reaction at the site of the ion channel causing excitation or inhibition. In the visual system, the rod photoreceptor is stimulated or hyperpolarized due to the GPCR involving the receptor metarhodopsin II (triggered by rhodopsin) coupled with transducin (the visual G protein) activating the production of the protein kinase phosphodiesterase (Nui et al. 2004). Fatty acid composition of the membrane bilayer has been shown to alter the rate of GPCR coupling (Niu et al. 2001; Mitchell et al. 2003; Nui et al. 2004). In those investigations GPCR coupling was reduced in the absence of DHA.
5.4 Testing and Assessment of the Visual System in Infants and Toddlers A robust area of research with regard to nutrition and infant development is in the area of visual processing. The majority of the visual research linked to cognitive functioning has been reported during the past three decades (Kellman and Banks 1998). ERG, FLP and VEP are the three main techniques employed for the assessment of infant visual function. The consideration of visual acuity is important to the assessment of cognitive development. There is a need for good vision in order to perform optimally on other developmental tests. If an infant scores poorly on a given developmental test, results could be clouded by poor vision and be mistakenly classified as poor mental processing. Measures of infant visual development have focused on two main aspects of the visual system; retinal development using ERG and visual processing at the level of the primary visual cortex. Cortical measures of grated visual acuity include sweep VEP and behavioral assessments such as FLP (Fig. 5.3). Of the grated acuity measures, VEP measures cortical response directly and has less inherent variability. However, behavioral acuity measures provide the most direct measure of what an infant actually perceives.
5 Infant Visual Acuity and Relationships with Diet and Nutrition
63
Fig. 5.3 Visual acuity measures. Outline of methodologies measuring cerebral response to visual stimuli: Central response as in visual evoked potential is a direct measure of cortical activity. Behavioral response methodologies like forced-choice preferential looking rely on movement or behavior associated with the cortical stimuli
5.4.1 Electroretinogram The ERG is a direct measure of the retinal response to a visual stimulus prior to reaching any of the cortical areas. In this procedure, electrodes placed on the cornea and surrounding eye tissue measure cellular electrical reactivity related to visual stimulus. ERG measurements therefore do not provide information regarding higher-level (i.e., cortical) visual processing (Table 5.1). The work of Malcolm et al. (2003a) was focused on maternal DHA supplementation and infant ERG. These investigators supplemented pregnant women with fish oil from 15 weeks gestation to delivery. ERG assessments were performed in the early postnatal period circumventing the potential impact of infant feeding method or other socioeconomic variables. There were no significant differences in retinal reactivity between infants of mothers who were supplemented compared to controls. Maternal DHA concentrations were higher in the experimental group when compared to control, however the relationship between infant DHA status (umbilical cord blood) and maternal status was not significant. The finding that maternal supplementation of DHA was not related to infant status is contrary to the view that a relationship exists and may be related to the dose (200 mg DHA daily) chosen for the investigation. Wijendran et al. (2000) demonstrated a strong relationship between maternal and infant DHA status in pregnancies not complicated with gestational diabetes mellitus (GDM). In that investigation, in which control subjects were compared to women with GDM, the cord vein erythrocyte phospholipid DHA and arachidonic acid (wt%) were higher than the maternal phospholipid DHA and arachidonic acid in the control group with significantly positive correlations between maternal and fetal erythrocyte phospholipid for arachidonic acid and DHA (r = 0.83, P = 0.003; r = 0.62, P = 0.04, respectively) (Tables 5.2).
5.4.2 Forced-Choice Preferential Looking (FPL) Assessment of visual acuity in infancy is complicated by the need for gathering nonverbal information. For this reason, forced-choice preferential looking procedures were developed in the area of visual acuity assessment to better understand and quantify the behavioral cues of infants. Preferential looking procedures use black and white stripes that are grated in a series of frequencies that correspond with the number of stripes per visual angle (commonly termed as cycles/degree) (Fig. 5.4). Table 5.1 Key features of the electroretinogram 1. Electroretinogram measures retinal response to a visual stimulus. 2. Electrodes are placed directly on the cornea and surrounding eye tissue. 3. The eye is exposed to a visual stimulus, i.e. light. 4. Electrical reactivity of the eye cells is recorded and quantified. This table lists the key facts of electroretinogram
64
M.P. Judge et al.
Table 5.2 Key points on the role of DHA in maternal and infant nutrition 1. The essential fatty acid DHA has a central role in membrane fluidity and synaptic signaling impacting retinal reactivity and visual processing. DHA accumulates in high concentrations in specific regions of the brain, including cerebral cortex, synapses and retinal rod photoreceptors in mice, rats, baboons, and other mammals (Bazan and Scott 1990; Bowen and Clandinin 2002; Sarkadi-Nagy et al. 2003). DHA deficiency induced during the gestational period in rats significantly impact the amount of DHA that was incorporated into the neuronal growth cone (Auestad and Innis 2000). 2. The current recommendation for DHA intake during pregnancy that stemmed from an expert panel is 200 mg/day (Koletzko et al. 2008). 3. Single-cell oils (SCO) and marine sources are safe and can adequately provide DHA supplementation to both the mother and infant to prevent deficiencies. DHA deficiency can interfere with optimal infant visual and cognitive development. This table lists key points relating to the role of DHA in maternal and infant health including adverse developmental outcomes associated with DHA deficiency during the gestational period, current intake recommendations for pregnant women, and safe supplementation sources
Fig. 5.4 Forced-choice preferential looking. In the Teller Acuity Card procedure, stripes are flipped from right (a) to left (b) with the location of the stripe unknown to the tester. Tester looks for a behavioral response (i.e., head movement toward the striped patch) (MPJ Original)
The point at which an infant is unable to resolve or see the contrasting stripes in a grated acuity procedure is considered the visual threshold. The term “forced choice” refers to the fact that the observer must choose or assess if the infant is looking to the left or right corresponding with the visual stimulus. The Teller Acuity Card Procedure (ACP) is a FPL procedure that has been standardized for use in a U.S. population. The ACP is portable and easily administered making it ideal in the research setting. As with any measure that relies on the interpretation of behavioral responses, there are a number of confounding variables that could skew findings. For example, a sleepy or fussy infant may lack interest in the visual stimulus and therefore alter results. There are multiple published studies utilizing ACP/FPL procedures for the investigation of the relationship between DHA status and visual development in term infants (Innis et al. 1994, 1996, 1997; Carlson et al. 1996; Jorgensen et al. 1996; Auestad et al. 1997, 2001; Birch et al. 2000; O’Connor et al. 2001). Carlson et al. (1996) investigated term infants fed formula with varying levels of DHA. Using FPL, the supplemented group performed similarly to the group of infants fed human milk, which contains variable amounts of DHA depending upon the maternal diet and/or supplementation. Three of the studies in term infants relating to DHA status have shown statistically significant differences using ACP/FPL procedures (Birch et al. 1993a, b; Carlson et al. 1996; Jorgensen et al. 1996). Contrary to the findings of
5 Infant Visual Acuity and Relationships with Diet and Nutrition
65
Carlson et al., Auestad et al. (1997, 2001) reported no differences between supplemented and control groups using the ACP procedure. It is noteworthy to mention that the amount of DHA in these studies (Auestad et al. 1997, 2001) was considerably lower than those used in other studies (Birch et al. 1993a, b; Carlson and Werkman 1996; Jorgensen et al. 1996). Only one investigation has reported on the impact of maternal DHA supplementation during pregnancy and infant visual acuity measured using the ACP. Judge et al. (2007) investigated the role of a DHA-functional food consumed during pregnancy on infant visual acuity at 4 and 6 months of age measured behaviorally using the ACP. In this randomized, longitudinal, double-blinded, and placebocontrolled trial, 30 pregnant women received either the DHA-functional food (n = 16) or the placebo (n = 14). There were significant main effects for visual acuity at 4 months of age (P = 0.018). The mean acuity score were, 3.8 ± 1.1 cycles/degree in the DHA group versus 3.2 ± 0.7 cycles/degree in the placebo group. At 6 months there was no group difference (Judge et al. 2007). Based on these results, it was concluded that DHA supplemented during pregnancy plays a role in the maturation of the visual system. This finding combined with evidence from studies conducted evaluating DHA supplementation during the postnatal period provide compelling evidence that DHA supplementation during pregnancy can have a profound impact on infant visual acuity development.
5.4.3 Visual Evoked Potential VEP is a direct measure of cortical response and has less inherent subjectivity than the behavioral methods of visual acuity assessment. In this procedure, the infant is given an electroencephalogram by placing electrodes on the scalp to measure neural cell reactivity related to a visual stimulus. The majority of the studies performed using VEP have shown a positive relationship between DHA status and visual development (Makrides et al. 1993, 1995; O’Connor et al. 2001; Hoffman et al. 2003). Hoffman et al. (2003) investigated term infants who fed human milk for 4–6 months then fed either formula supplemented with DHA and arachidonic acid or control formula. Infants in the supplemented group had significantly better VEP than the control group. Makrides et al. (1995) also investigated the impact of DHA supplementation on visual processing in term infants and found that the supplemented and human milk–fed groups performed similarly on VEP measures and better than the group of infants who did not receive DHA. Similar findings have been reported for preterm infants. O’Connor et al. (2001) investigated preterm infants and reported significantly better VEP in DHA supplemented groups. In summary, the majority of studies to assess vision in term and preterm infants have demonstrated a positive association between processing and DHA status. With regard to fetal development, Malcolm and coworkers (Malcolm et al. 2003a) examined maternal DHA supplementation and infant VEP. They supplemented pregnant women with fish oil from 15 weeks gestation to delivery. VEP assessments were performed at 2.5 and 6.5 months of age and there were no significant differences between infants of mothers who were supplemented compared to controls. This finding is in contrast to the study of visual acuity carried out by Judge et al. (2007).
5.5 Dietary Factors Influencing Visual Development The last trimester of pregnancy is a critical interval for fetal neurological development. During this time DHA accumulates in neural tissue at an accelerated rate (Fox 1998). Reports of maternal DHA intake during pregnancy suggest that intakes are generally far below the recommended level of intake
66
M.P. Judge et al. Herring Salmon Mackerel Swordfish Tuna Smelt Trout Sardine Shark Squid Halibut Flounder/Sole Haddock Lobster Whitefish Cod Chicken Egg Turkey Duck Pork
0
200
400
600
800
1000
1200
DHA (mg/100 gm edible portion) Fig. 5.5 Dietary sources of docosahexaenoic acid (mg DHA/100 g edible portion). Herring, salmon and mackerel are examples of cold-water marine fish that are a rich dietary source of DHA. Source: U.S.D.A., Human Nutrition Information Service
which suggests that many infants are at risk for associated impairments in development (Cheruku et al. 2002; Olsen and Secher 2002; Loosemore et al. 2004). Figure 5.5 outlines common food sources of DHA and points to cold water marine fish as the best dietary sources. It is unknown how concerns about contaminants in fish relate to low DHA intake during pregnancy (Cheruku et al. 2002; Olsen and Secher 2002; Loosemore et al. 2004). Compounding the problem of reduced maternal intake of DHA, the overall intake of fatty acids in the n-6 family is typically high in western industrialized countries such as the U.S. High intake of n-6 fatty acids is related to an abundance of processed and fried foods containing plant oils with linoleic acid. Linolenic and linoleic both compete for the ∆-6 desaturase and high intake of linoleic acid can inhibit conversion of linolenic acid to DHA (Simopoulos et al. 2000). In 2008 a panel of experts led by Koletzko et al. (2008) concluded that the optimal level of DHA intake during pregnancy is 200 mg/day. DHA has a central role in membrane fluidity and synaptic signaling in neural tissue and DHA accumulates in high concentrations in specific regions of the brain, including the cerebral cortex, synapses, and retinal rod photoreceptors in mice, rats, baboons, and other mammals (Bazan and Scott 1990; Bowen and Clandinin 2002; Sarkadi-Nagy et al. 2003). DHA deficiency induced during the gestational period in rats significantly reduce the amount of DHA that was incorporated into the neuronal growth cone (Auestad and Innis 2000). In humans, autopsy studies have demonstrated the importance of DHA in brain tissue. Clandinin et al. (1980) investigated fatty acid components of the fetal brain during the third trimester. Brain tissue fat content was analyzed upon autopsy at varying stages of gestation to determine the pattern of fatty acid accretion during pregnancy. That study demonstrated that the accretion of the essential
5 Infant Visual Acuity and Relationships with Diet and Nutrition
67
fatty acids linolenic acid and linoleic acid (the metabolic precursors for DHA and arachidonic acid, respectively) were low during the last trimester and that there is a substantial accretion of DHA and arachidonic acid. Evidence shows the importance of DHA intake for both maternal and infant health (Table 5.2).
5.5.1 Dietary Sources of DHA for Supplementation Three main sources of DHA and arachidonic acid oil that have been considered for use in infant formulas are: marine, egg phospholipid, and SCO. Commercially available maternal supplements contain either marine or SCO. There are potential drawbacks associated with the use of marine and egg phospholipid oils in infant formulas.
5.5.1.1 Marine Oil Early studies focused on the addition of DHA and arachidonic acid to human infants via marine oil which indicated a potential negative impact on growth. This adverse outcome was attributed to the eicosapentaenoic acid content of the oil and a possible competitive inhibition of the n-6 pathway responsible for the production of arachidonic acid (Carlson et al. 1992a, b, 1993). Makrides et al. investigated DHA vis-a-vis marine oil supplementation in term infants and reported a significantly lower DHA erythrocyte concentration in the experimental group receiving marine oil compared to the breast milk or placebo group (Makrides et al. 1995). An additional safety concern for marine oil supplementation is increased bleeding time related to the eicosapentaenoic acid, a reflection of cyclooxygenase production. Cyclooxygenase inhibits platelet aggregation and causes vasodilatation. Impaired platelet aggregation decreases blood clotting and resulting in increased bleeding time. Scientific evidence exists that lends credence to the notion that marine oil supplementation does not have a negative impact on growth or bleeding time. In a series of studies, Carlson et al. investigated the safety and efficacy of supplementation with marine oils in preterm infants (Carlson et al. 1992a, b). In study I, infants were given a nasogastric bolus of marine oil in the dose of 750 mg/kg/day for 4 weeks. In study II, the marine oil was mixed into the formula at doses of 116 mg/ kg/day (0.2% DHA and 0.3% eicosapentaenoic acid) and 290 mg/kg/day (0.5% DHA and 0.75% eicosapentaenoic acid) of marine oil for a 2-week period. In study III, the dose of 116 mg/kg/day of marine oil was investigated for a 9-month period. The bolus dose of marine oil in study I resulted in poor fatty acid absorption. In studies I and II phosphatidylethanolamine arachidonic acid, a primary phospholipid component of the cell membrane, declined significantly over time for all infants (Carlson et al. 1992a, b). In study II, the infants in the marine oil group had significantly lower phosphatidylethanolamine arachidonic acid than the controls; however, the marine oil groups also had lower baseline arachidonic acid levels, which likely contributed to these findings. Conversely, eicosapentaenoic acid and DHA increased in the marine oil groups and decreased in the control groups. The infants receiving 290 mg/kg/day of marine oil had significantly lower phosphatidylcholine, a primary phospholipid component of the cell membrane also commonly known as lecithin, than the control and lower dose groups (i.e., 116 mg/kg/day). In study III, there were positive correlations between arachidonic acid and DHA in the marine oil group through 9 months of age.
68
M.P. Judge et al.
Uauy et al. (1994) compared four groups of very low birth weight infants given human milk (control), formula with corn oil, formula with soy oil, and formula with soy and marine oil. In this study bleeding times were evaluated and no significant differences in bleeding time were reported for any of the groups and bleeding times were within normal ranges for all groups. Additionally, Uauy et al. considered anthropometric data for each of the groups. Anthropometric data included weight, length, head circumference, triceps skinfold, and subscapular skinfold. Measurements were collected at birth, 34, 40, 48, and 57 weeks’ gestational age. There were no differences among any of the study groups, i.e., weight gain and growth were similar for all four groups. Based on the few studies carried out in human infants, there are no associated adverse impacts on growth or bleeding time in a dose of fish oil below 116 mg/kg/d. It appears that DHA absorption is better with marine oil if mixed into the formula compared to a bolus dose. Likewise, multiple large studies have been conducted that provided marine oil during the gestational period and no negative pregnancy outcomes were reported. In fact, marine oil supplemented during pregnancy has been demonstrated to increase the average length of gestation resulting in significantly higher birth weight (Olsen et al. 1995).
5.5.1.2 Egg Yolk Phospholipid Egg yolk phospholipid is composed primarily of phosphatidlycholine. A drawback of egg-derived oils is that it contains relatively low amounts of DHA and arachidonic acid. Therefore, large amounts need to be added to infant formula to reach the desired DHA and arachidonic acid levels. Given the limitations of supplementation, egg-derived phospholipid has not been studied for commercial use in pregnancy. Consuming large quantities of egg phospholipids cause concern because of the exposure to large doses of sterols and egg-related allergens. In the context of this review, few research groups have used egg-derived phospholipids as DHA and arachidonic acid sources. Of the studies that have been reported, none have reported adverse outcomes with regard to toxicity or growth (Carlson et al. 1996; Auestad et al. 1997, 2001).
5.5.1.3 Single-Cell Oils DHA and arachidonic acid oils, single-cell oils (SCO), are produced via the fermentation, extraction, and purification of microalgae and microfungus (Boswell et al. 1996). These oils are used as a maternal supplement during pregnancy and in infant formulas. The marine alga used for the production of DHA is Crypthecodinium cohinii and the oil is 40–50% DHA with no Eicosapentaenoic acid. This DHA single-cell oil is commonly referred to as DHASCO (Boswell et al. 1996). Arachidonic acid is formed from the unicellular fungus strain Mortierella alpina and is 40–50% arachidonic acid. This arachidonic acid single-cell oil is commonly referred to as ARASCO (Boswell et al. 1996). The microorganisms used to produce SCO had never before been used in food products and therefore required Food and Drug Administration approval for use in infant formulas. The natural toxins inherent in the two microorganisms were of concern with regard to use in food. Sufficient evidence exists regarding the safety of SCO as these have been investigated extensively for use in infant formulas (Boswell et al. 1996). In summary, SCO and marine oil are both considered safe and efficacious for consumption during pregnancy; however, research points to the efficacy of a low eicosapentaenoic acid formulation for this group (Carlson et al. 1992a, b, 1993). Single-cell oils have been investigated extensively for use in infant formulas and provide the best alternative to marine and egg-derived sources.
5 Infant Visual Acuity and Relationships with Diet and Nutrition
69
5.6 Summary In summary, visual signaling involves a complex interaction between the retina, thalamus, and primary visual cortex. Other cortical regions of the brain as well as central brain structures are involved in processing visual stimuli. Measures of visual development in infancy assess retinal reactivity or cortical processing of the visual stimulus. Behavioral measures of visual acuity give insight on the processing of visual information. Retinal photoreceptors have a high DHA concentration and deficiency during fetal development and the postnatal period have been associated with interruptions in visual development. Maternal intake of DHA during pregnancy is significantly below the recommended 200 mg/day for healthy pregnancy outcomes. In turn, the developing fetus is at risk for negative outcomes. Optimal visual development enables the infant to move toward more integrative cognitive processing and the achievement of established developmental milestones. Additional community-based efforts are necessary to educate communities regarding the important role of DHA for optimal developmental outcomes during both the prenatal and postnatal periods. Future work should focus on the evaluation of maternal DHA status and its relationship to visual development in older children and associations with other cognitive processes.
5.7 Applications to Other Areas of Health and Disease The consideration of visual acuity is very important to the assessment of cognitive development for a number of reasons. First, there is a need for good vision in order to perform optimally on other developmental tests. If an infant scores poorly on a given developmental test, results could be clouded by poor vision and mistakenly classified as poor mental processing. Additionally, early visual perception probably guides the development of action systems to promote learning about the infant’s physical and social worlds (Von Hofsten 1980; Bertenthal 1996). For example, the development of infant reaching is prompted by the visual stimulus of an object that an infant is interested in exploring. Future work should focus on (1) the evaluation of maternal DHA status and the infant’s ability to integrate visual stimuli, as in face and object recognition, and (2) the relationship of acuity assessed during infancy with later vision and other cognitive processes (Neuringer and Jeffrey 2003). Focused community-based programs are necessary for promoting adequate intake of DHA during pregnancy and ensuring optimal infant visual and cognitive development.
Summary Points • Visual signaling involves the retina, thalamus, and primary visual cortex. • Other cortical regions of the brain as well as central brain structures are involved in processing visual stimuli. • Measures of visual development assess retinal stimulation (ERG) or cortical processing (VEP and FPL). • Forced-choice preferential looking procedures are behavioral measures of visual acuity and lend insight to the integration of visual information. • Retinal photoreceptors and neural cell bodies are comprised predominantly of the essential fatty acid docosahexaenoic acid.
70
M.P. Judge et al.
• Deficiencies of docosahexaenoic acid during fetal and infant development have been associated with interruptions in normal visual processing. • Alterations in G protein coupled signal transduction related to docosahexaenoic acid provides a plausible mechanism for observations of alterations in visual functioning. • Maternal docosahexaenoic acid during pregnancy is significantly below recommendations (200 mg/day) for optimal pregnancy outcomes placing the developing fetus at risk. • Dietary sources of docosahexaenoic acid for the purpose of supplementation include: Marine and single-cell oils. • Poor visual processing can impede infant attainment of later developmental milestones that rely on integrative processing. • Additional educational efforts are necessary to educate the community regarding the important role of docosahexaenoic acid in infant development during both the prenatal and postnatal periods.
Definitions of Key Terms Photoreceptors: A neural cell that is specific to the retina with specificity for visual stimuli. Visual cortex: The portion of the brain’s cortex where visual information is sent for further processing. Electroretinogram: A direct measure of the retinal response to a visual stimulus prior to reaching any of the cortical areas. Forced-choice preferential looking: Preferential looking procedures use black and white stripes that are grated in a series of frequencies that correspond with the number of stripes per visual angle (commonly termed as cycles/degree). The point at which an infant is unable to resolve or see the contrasting stripes in a grated acuity procedure is considered the visual threshold. Visual evoked potential: A direct measure of cortical response to a visual stimulus. Docosahexaenoic acid: An essential long-chain polyunsaturated fatty acid, which is a metabolic end product of the omega-3 fatty acid biosynthetic pathway. Marine oil: Oil derived from edible fish sources for the purpose of supplementation. Egg yolk phospholipid: Phospholipid derived from egg yolk. Single-cell oils: Oils that are produced from single-cell sources including microalgae and microfungus.
References Auestad N, Innis SM. Am J Clin Nutr. 2000;71:312S–4. Auestad N, Montalto MB, Hall RT, Fitzgerald KM, Wheeler RE, Connor WE, et al. Pediatr Res. 1997;41:1–10. Auestad N, Halter R, Hall RT, Blatter M, Bogle ML, Burks W, et al. J Pediatr. 2001;108:372–81. Bazan NG, Scott BL. J Med Sci. 1990;48:97–107. Bertenthal BI. Annu Rev Psychol. 1996;47:431–59. Birch E, Birch DG, Hoffman D. J Pediatr Opthalmol Strabismus. 1993a;30:33–8. Birch E, Birch DG, Hoffman D, Everett M, Uauy R. J Pediatr Ophthalmol Strabismus. 1993b;30:33–8. Birch EE, Garfield S, Hoffman DR, Uauy R, Birch DG. Dev Med Child Neurol. 2000;42:174–81. Boswell K, Koskelo EK, Carl L, Glaza S, Hensen DJ, Williams KD, et al. Food Chem Toxicol. 1996;34:585–93. Bowen RA, Clandinin MT. J Neurochem. 2002;83:764–74. Carlson SE, Werkman SH. Lipids. 1996;31:85–90. Carlson SE, Cooke R, Werkman S, Tolley EA. Lipids. 1992a;27:901–7.
5 Infant Visual Acuity and Relationships with Diet and Nutrition
71
Carlson SE, Cooke RJ, Rhodes JM, Peeples J, Werkman SH. J Pediatr. 1992b;120:159–67. Carlson SE, Werkman S, Peeples JM, Cooke RJ, Tolley EA. Proc Natl Acad Sci. 1993;90:1073–7. Carlson SE, Ford AJ, Werkman SH, Peeples JM, Koo WK. Pediatr Res. 1996;39:882–8. Cheruku SR, Montgomery-Downs HE, Farkas SL, Thoman EB, Lammi-Keefe CJ. Am J Clin Nutr. 2002;76:608–13. Clandinin MT, CJ LS, Heim T, Sawyer PR, Chance GW. Early Hum Dev. 1980;4:121–9. Denomme J, Stark KD, Holub BJ. J Nutr. 2005;135(2):206–11. Fox P. Fetal and neonatal physiology, vol 1. Philadelphia: W.B. Saunders Company; 1998. p. 489–504. Hoffman DR, Birch EE, Castaneda YS, Fawcett SL, Wheaton DH, Birch DG, et al. J Pediatr. 2003;142:669–77. Innis SM, Elias SL. Am J Clin Nutr. 2003;77:473–8. Innis SM, Nelson CM, Rioux FM, King JD. Am J Clin Nutr. 1994;60:347–52. Innis SM, Nelson CM, Lwanga D, Rioux FM, Walsen P. Am J Clin Nutr. 1996;64:40–6. Innis SM, Akrabawi SS, Diersen-Schade DA, Dobson VM, Guy DG. Lipids. 1997;32:63–72. Jorgensen HM, Hernell O, Lund P, Holmer G, Michaelsen KF. Lipids. 1996;31:99–105. Judge MP, Harel O, Lammi-Keefe CJ. Lipids. 2007;42:117–22. Kellman PJ, Banks MS. Infant visual perception. In: Kuhn D, Siegler RS, editors. Handbook of child psychology: cognition, perception, and language, vol 2. New York: Wiley; 1998. p. 103–46. Kolb B, Whishaw IQ. Organization of the nervous system. In: Atkinson RC, Lindzey G, Thompson RF, editors. Fundamentals of human neuropsychology. New York: Worth Publishers; 2003. p. 46–74. Koletzko B, Lien E, Agostoni C, Böhles H, Compoy C, Cetin I, et al. J Perinat Med. 2008;36:5–14. Lewis NM, Widga AC, Buck JS, Frederick AM. J Agromedicine. 1995;2:49–56. Loosemore ED, Judge MP, Lammi-Keefe CJ. Lipids. 2004;39:421–4. Makrides M, Simmer K, Goggin M, Gibson RA. Pediatr Res. 1993;33:425–7. Makrides M, Neumann M, Simmer K, Pater J, Gibson R. Lancet. 1995;345:1463–8. Malcolm CA, Hamilton R, McCulloch DL, Montgomery C, Weaver LT. Invest Opthalmol Vis Sci. 2003a;44:3685–91. Malcolm CA, McCulloch DL, Montgomery C, Shepherd A, Weaver LT. Arch Dis Child Fetal Neonatal Ed. 2003b;88:F383–90. Mitchell DC, Niu SL, Litman BJ. J Pediatr. 2003;143:880–6. Neuringer M. Am J Clin Nutr. 2000;71:256S–67. Neuringer M, Jeffrey BG. J Pediatr. 2003;143:S87–95. Neuringer M, Connor WE, Barstad L, Luck S. Proc Natl Acad Sci. 1986;83:4021–5. Niu SL, Mitchell DC, Litman BJ. J Biol Chem. 2001;276:42807–11. Nui SL, Mitchell DC, Lim SY, Wen ZM, Kim HY, Salem N, et al. J Biol Chem. 2004;279:31098–104. O’Connor DL, Hall R, Adamkin D, Auestad N, Castillo M, Connor WE, et al. J Pediatr. 2001;108:359–71. Olsen SF, Secher NJ. Br Med J. 2002;324:447–50. Olsen SF, Hansen HS, Secher NJ, Jensen B, Sandstrom B. Br. J Nutr. 1995;73:397–404. Salem Jr N, Litman B, Kim HY, Gawrisch K. Lipids. 2001;36:945–59. Sarkadi-Nagy E, Wijendran V, Diau GY, Chao AC, Hsieh AT, Turpeinen A, et al. Pediatr Res. 2003;54:244–52. Simopoulos AP, Leaf A, Salem N. Prostaglandins Leukot Essent Fatty Acids. 2000;83:119–21. Uauy R, Hoffman DR, Birch EE, Birch DG, Jameson DM, Tyson J. J Pediatr. 1994;124:612–20. Von Hofsten C. J Exp Child Psychol. 1980;30:369–82. Wijendran V, Bendel RB, Couch SC, Philipson EH, Cheruku S, Lammi-Keefe CJ. Lipids. 2000;35:927–31. Zigmond MJ, Bloom FE, Landis SC, Roberts JL, Squire LR. Fundamental Neuroscience. San Diego: Academic; 1999.
wwwwwwwwwwwwwww
Chapter 6
Acquired Tastes: Establishing Food (Dis-)Likes by Flavour–Flavour Learning Remco C. Havermans and Anita Jansen
Abbreviations ANC BMI CS OFC PDD UCS
Amygdalar nuclear complex Body mass index Conditioned stimulus Orbitofrontal cortex Pervasive developmental disorder Unconditioned stimulus
6.1 Introduction It is a truism that eating behaviour (i.e. food choice and food intake) is determined by many different factors. Nonetheless, within the normal physiological boundaries of satiety and hunger one may argue that people simply eat what they like and avoid foods they do not like (Eertmans et al. 2001). Such hedonic eating behaviour is particularly apparent in children. For example, in the early twentieth century, most Dutch children (in addition to children in many other countries) still received a daily spoonful of cod liver oil before bedtime. Many of these individuals, now adults, still shiver at the memory of the often rancid taste of cod liver oil. However, there was good reason to subject children to this cod liver oil ordeal, as this oil was known to somehow prevent rickets (or rachitis; the softening of bones misshaping knees, wrists, and ankles). It wasn’t until the 1930s that it was recognised that rickets is caused by vitamin D deficiency and that cod liver oil is rich in such vitamin D. This knowledge eventually allowed for the development of vitamin D supplementation and with it the standard practise of administering children cod liver oil dissipated. One will cease the consumption of aversive tastes when there is no pressing need for the consumption of such tastes. People are born with a preference for sweet tastes and an aversion against bitter tastes, but more specific flavour preferences are developed during later childhood. Specific food likes and dislikes can differ between individuals and this suggests that these preferences are acquired through personal experience with certain flavours. In other words, one acquires a (dis)taste for certain foods by trying out these foods (Capaldi and Vandenbos 1991). How then does this work exactly? A prominent form R.C. Havermans (*) Department of Clinical Psychological Science, Maastricht University, Maastricht, 6200 MD, The Netherlands e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_6, © Springer Science+Business Media, LLC 2011
73
74
R.C. Havermans and A. Jansen
of such purported experiential learning involves Pavlovian conditioning, the learning of an association between a neutral conditioned stimulus (CS) (e.g. a flavour) and a biologically relevant unconditioned stimulus (UCS). In the case of conditioned taste aversion learning, the UCS usually comprises some degree of gastrointestinal discomfort. By pairing a specific flavour with such discomfort, one comes to associate the flavour with the discomfort and this then leads to future avoidance of the flavour (Pelchat and Rozin 1982). Pavlovian conditioning has then endowed the flavour CS a signalling function, it ‘predicts’ internal malaise (the UCS). Likewise, it was found in animals that a taste preference is conditioned for flavours paired with the recovery of illness (Green and Garcia 1971). Another way of conditioning a food preference was demonstrated by Holman (1975). In one of his experiments, he gave rats on alternate days either a banana- or almond-flavoured mash. To this mash, he added saccharin to make it taste sweet. Rats, like humans, have an innate preference for sweet tastes. Holman varied the concentration of the saccharin of the different mashes. Rats were divided into two groups: A and B. Group A received 60 min access to the almond paired with concentrated saccharin and banana paired with diluted saccharin solution. For group B, the flavour to saccharin concentration (concentrated versus diluted) assignment was reversed. At test, the rats received 30 min access to 40 mL of both flavours now with an equal saccharin concentration. The rats showed a clear preference for the flavour previously paired with the concentrated saccharin. Figure 6.1 displays an illustration of the procedure, design, and results of this experiment. On the basis of these and similar findings, Holman concluded that the apparent reinforcing effect of saccharin is the hedonic quality of its sweet flavour. This form of learning has thus been termed flavour–flavour learning as the apparent
Fig. 6.1 Design, procedure and results of Eric Holman’s (1975) experiment on flavour–flavour learning. Eric Holman (1975; Experiment 2) gave rats during a 20-day training period one flavour paired with concentrated saccharin on even days and another flavour paired with diluted saccharin on uneven days. On a first test, rats were given simultaneous access to both flavours, now both with diluted saccharin. Rats drank much more of the flavour previously paired with the concentrated saccharin. On a second test, after another 4-day period of training, the same pattern of results was obtained when both flavours were now presented with concentrated saccharin
6 Acquired Tastes: Establishing Food (Dis-)Likes by Flavour–Flavour Learning
75
Table 6.1 Key facts concerning flavour–flavour learning and its applications A neutral flavour becomes more liked or less liked after having been paired with, respectively, an already liked or disliked flavour Children can come to better like the flavour of fruits and vegetables through flavour–flavour learning
shift in flavour preference arises from the acquisition of an association between two flavours: a neutral flavour (e.g. almond) and an already liked flavour (e.g. the sweet taste of concentrated saccharin) (Fanselow and Birk 1982; Myers and Sclafani 2006). Some researchers argue that a flavour–flavour association is qualitatively different from a preference arising from flavour–consequence learning. As the latter form of association is recognised as the result of Pavlovian conditioning, this implies that flavour–flavour learning may not be considered as a form of Pavlovian associative learning (Hermans et al. 2002) and this gives rise to the question what mechanism then underlies flavour–flavour learning. Other questions raised and discussed in the present chapter concern the generality of flavour–flavour learning, its neurological underpinnings and its application to promoting the development of healthy dietary habits in children (Table 6.1).
6.2 What is Flavour–Flavour learning? Flavour–flavour learning comprises the transfer of affect to a neutral flavour CS by pairing that particular flavour with another already liked or disliked flavour UCS. The established shift in liking is always in the direction of the affective value of the evaluative UCS. This is why flavour–flavour learning is generally regarded as a form of evaluative conditioning. But what exactly is evaluative conditioning?
6.2.1 Evaluative Conditioning In a particular study, Levey and Martin (1975) had participants categorise pictures of paintings as liked, neutral, or disliked. Subsequently, the participants received exposure to several neutral pictures paired with other liked pictures, or neutral pictures, or disliked pictures. At test, participants had to evaluate all pictures and it was found that they now rated the neutral pictures that had previously been paired with the disliked pictures more negatively and the neutral pictures that had been paired with the liked pictures as more positive. Martin and Levey (1978) termed this transfer of affect to an originally neutral stimulus due to pairings of this stimulus with another affective (positive or negative) stimulus evaluative conditioning. Procedurally, evaluative conditioning is very similar to Pavlovian conditioning. With evaluative conditioning, the neutral stimulus is usually referred to as the CS and the affective stimulus as the UCS. Pairing these stimuli should then lead to the formation of an association between these two stimuli, allowing for the transfer of affect from the UCS to the CS. Despite the procedural similarity there are notable discrepancies between evaluative conditioning and other more typical Pavlovian conditioning paradigms such as fear conditioning. In the case of fear conditioning, the CS, a neutral stimulus (e.g. a tone or a picture), is paired with aversive stimulation, such as the administration of an electric shock or a very loud noise. Hermans et al. (2002) exposed participants to such a procedure, pairing pictures of faces (the CSs) with electrocutaneous stimulation (the UCS; i.e. an aversive electric shock). The participant learned to anticipate this adverse stimulation that was reflected by a strong expectation of the UCS when presented with the CS previously paired with the UCS. Hermans and colleagues also showed that participants came to dislike the CSs that had been paired with the
76
R.C. Havermans and A. Jansen
UCS, indicating evaluative conditioning. UCS expectations extinguish easily with non-reinforced exposures to the CS (Dibbets et al. 2008), but an acquired evaluative shift is highly resistant to such an extinction treatment (De Houwer et al. 2000). Pavlovian conditioning can be described in terms of signal learning; one learns to recognise the CS as a signal for the UCS, and hence, when exposed to the CS, one expects the UCS. No such learning is required in case of evaluative conditioning; the CS then merely has to refer to the UCS (see also De Houwer et al. 2000). In line with the reasoning that evaluative conditioning reflects some form of referential learning, Havermans and Jansen (2007a) argue that evaluative conditioning more specifically reflects a stimulus generalisation process. Within Pavlovian conditioning, it is well known that an acquired associative strength can generalise from one CS to other CSs. The extent of transfer of associative value is determined by the degree of similarity between the CS and the novel stimulus (Pearce 2002). According to Havermans and Jansen, such transfer is probably not limited to associative value. Indeed, it may also comprise the transfer of other relevant stimulus characteristics, such as affective value. This then means that pairing a neutral CS with an evaluative UCS merely provides one the opportunity for determining (or just passively experiencing) the similarity between the two stimuli. If there is some degree of stimulus similarity, and if this similarity is somehow noted, this may be enough to induce the transfer of affect from the UCS to the CS (Davey 1994; Field and Davey 1999). Figure 6.2 represents the stimulus generalisation model of evaluative conditioning as outlined by Havermans and Jansen.
6.2.2 The Flavour–Flavour Paradigm Evaluative conditioning is not limited to the transfer of affect between visual stimuli such as the pictures of paintings. Indeed, it was soon recognised that flavour–flavour learning can also be understood in terms of evaluative conditioning. In fact the flavour–flavour paradigm is recognised as one of the more robust forms of evaluative conditioning (Field et al. 2008).
Fig. 6.2 A stimulus generalisation model of evaluative conditioning. Havermans and Jansen (2007a) argue that when a neutral CS X is paired with an affective UCS Y a configural X-Y representation is formed in memory. When subsequently X is presented, the representation of the UCS Y will be activated directly through its similarity with X and indirectly through the X-Y configural representation. The degree to which the representation of Y is activated determines the extent of affective value being transferred from Y to CS X
6 Acquired Tastes: Establishing Food (Dis-)Likes by Flavour–Flavour Learning
77
As described above, Eric Holman (1975) tested whether a rat’s flavour preference could be influenced by the sweetness of saccharin associated with the flavour. It could and such flavour–flavour learning is not limited to rats; it has also been demonstrated in humans. Baeyens et al. (1995) found that pairing a flavour with the aversive taste of polysorbate-20 (the emulsifier Tween) led to decreased liking of that particular flavour (see also Baeyens et al. 1990, 1996, 1998a, b). These researchers, however, did not find evidence of positive flavour–flavour learning with pairings of a specific neutral flavour with a sweet taste. This may mean that negative flavour–flavour learning is more easily established than positive flavour–flavour learning, but the failure to induce positive flavour–flavour learning could also be attributed to the fact that the participants only moderately liked the sweet taste. It was already shown that it certainly is not impossible to induce positive flavour–flavour learning. Zellner et al. (1983) had students drink different flavours of tea. Some of these teas had to be tasted several times with sweet-tasting sucrose. At test, all participants again had to taste and evaluate the different teas, now left unsweetened, and the participants clearly and specifically liked the previously sweetened teas better. Yeomans et al. (2008) similarly found evidence of positive flavour–flavour learning when a specific flavour of dessert was paired with the sweet taste of either sucrose or aspartame. To ensure that the sweet taste (the flavour UCS) was well liked, Yeomans took care to only select self-identified sweet likers for participation in their study. Interestingly, Brunstrom et al. (2001) found that unrestrained eaters showed enhanced preference for drinks most often paired with a sweet reward (e.g. chocolate chips, puffed rice or raisins), but this was not found among restrained eaters, suggesting that dietary restraint somehow devalues the sweet reward. This is precisely what Brunstrom et al. (2005) found in a series of two experiments. In one of these experiments, Brunstrom and colleagues paired different fruit juices of with a sweet reward (i.e. chocolate chips). One of the juices was paired on just 10% of trials with the sweet and another juice was paired 90% of the trials with the sweet reward. Unrestrained eaters came to prefer the 90% paired picture, but the restrained eaters came to prefer the less-often paired pictures. Brunstrom and colleagues thus argued that negative beliefs and attitudes regarding the UCS devalue the UCS and within restrained eaters may in fact function as an aversive stimulus, thus promoting the acquisition of a dislike. In a more recent study, however, Brunstrom and Fletcher (2008) failed to replicate this effect of dietary restraint on flavour–flavour learning, but instead found an effect of hunger state. When pairing tea with a non-caloric sweetener, only hungry participants came to like the flavour of this particular tea relative to other unsweetened teas. Similarly, Mobini et al. (2007) found that pairing a peach-flavoured tea with another sucrose drink, increased liking for the tea particularly when participants had been trained and tested hungry. Further, Yeomans and Mobini (2006) demonstrated that increased liking for odours paired with a sweet sucrose UCS was apparent only when participants were hungry. One may argue that hunger increases attention and liking for caloric cues, such as a sweet taste that generally signals carbohydrate energy (e.g. as in the case of sucrose, fructose and glucose), hence increasing the salience of the flavour UCS allowing for larger transfer of affective value to the CS flavour (see Fig. 6.3).
6.2.3 Neural Correlates of Flavour–Flavour Learning Cues associated with a certain food UCS can come to affect subsequent food choice and food intake. The brain structures known to be involved in the expression of cue-induced food selection and acceptance, such as the hypothalamus, amygdala and OFC probably also play a role in the associative acquisition of food likes and dislikes (see Holland and Petrovich 2005). Indeed, although many brain areas are activated with exposure to the sight, smell and taste of food, studies on the evaluation of food stimuli ubiquitously point to the specific involvement of the amygdala, the OFC and the insula.
78
R.C. Havermans and A. Jansen
Fig. 6.3 Flavour–flavour learning as stimulus generalisationIn terms of the stimulus generalisation model of evaluative conditioning as put forward by Havermans and Jansen (2007a), pairing a neutral flavour X with an already liked flavour Y, leads to the formation of a configural X-Y representation in memory. Subsequent presentation of X alone then leads to the activation of the configural X-Y unit and the representation of flavour Y. If one is hungry, the salience of the representation of Y is hypothetically bigger and allows for more transfer of affective value to flavour X
With oral taste neurons projecting directly to the insula, activation of this area has been associated with representations of taste intensity and quality (e.g. texture, taste intensity, bitterness, sweetness, sourness or saltiness) and is thus referred to as the primary gustatory cortex (Small et al. 1999; De Araujo and Rolls 2004). Adjacent to this area, the OFC is involved in representations of the purported acquired valence of food cues (e.g. the pleasure derived from the sight, smell and taste of food). This region, therefore, has been termed the secondary taste cortex. Rolls and colleagues showed that decreasing the subjective pleasure derived from eating a specific food item through repeated exposure to that food leads to a corresponding decrease in the activation within the OFC when presented with that specific food item (see Rolls 2000). The amygdala is recognised to be activated by both pleasurable and aversive food cues (O’Doherty et al. 2001) and thus appears to respond nonspecifically to any valenced food stimulus. Recent research suggests that such activation of the amygdala requires deliberate and attentive processing of the presented food stimulus (Siep et al. 2009). Considering the presumed involvement of the amygdala in the conditioning of likes and dislikes, Coppens et al. (2006) examined whether patients with a unilateral section of the temporal lobe (which includes the amygdala) exhibit impaired flavour–flavour learning as compared to a control. Participants received two flavours, one of which was served with the addition of Tween. The researchers hypothesised that given the fact that even unilateral damage to the amygdala (i.e. the ANC) attenuates fear conditioning; such damage would also disturb evaluative conditioning. However, all participants demonstrated clear negative flavour–flavour learning, acquiring a dislike for the Tweenpaired flavour CS relative to the unpaired CS. The authors correctly point out that this does not mean that the amygdala does not play any role in the acquisition of food likes and dislikes as all the patients included in the study had only unilateral damage to the ANC. Clearly, more research is needed to determine the exact neurological underpinnings of flavour–flavour learning.
6 Acquired Tastes: Establishing Food (Dis-)Likes by Flavour–Flavour Learning
79
6.3 Applying Flavour–Flavour Learning to Increase Liking for Vegetables Overweight (BMI > 25 kg/m2) is a rapidly increasing worldwide health problem. Especially obesity (i.e. severe overweight; BMI > 30) poses several health risks, such as coronary heart diseases and type II diabetes. It is estimated that currently in the USA approximately 25% of total health care costs are associated with obesity, a percentage representing billions of dollars (Levi et al. 2009). Currently, more than two-thirds of the US population is overweight or obese. Table 6.2 displays the present top five states with the highest obesity rates of the USA. The number of overweight children has also increased dramatically in the past few decades. The health risks associated with obesity at a young age are less dire, but it is more difficult to attain a normal weight if one has been obese since childhood. Moreover, the health risks for obese adults who have been obese since their childhood are greater than for people who became obese in later adulthood (see e.g. Visscher et al. 2002). Overweight and obesity are the result of a positive energy balance; more energy is consumed than expended. Particularly excessive caloric intake is now thought to have contributed to the steep rise in the incidence of obesity (Swinburn et al. 2009). Healthier eating, that is, consuming less high-calorie products should lower the prevalence of overweight. Indeed, Raynor and Epstein (2001) demonstrated that whereas consuming many different snacks is related to obesity, the consumption of a large variety of fruits and vegetables is associated with lean body weight. Therefore, getting people (especially children) to eat ample amounts of fruits and vegetables may prove to be an effective strategy in curbing the present obesity epidemic. Recently, we reasoned that flavour–flavour learning might be a powerful technique to increase children’s liking of vegetables, and hence their consumption of vegetables. To test whether such flavour–flavour learning actually increases children’s liking of vegetables, we conducted an experiment. In this experiment 4- to 6-year-old children evaluated and rank ordered six different vegetables. Next, they were instructed to repeatedly consume small amounts of two of the six specific vegetables. These two flavours served as the flavour CSs and one of the two vegetables was now sweetened with glucose. After this repeated exposure procedure, all children again were instructed to evaluate and rank order the six vegetables. At this posttest, the children now specifically ranked the previously sweetened vegetable as better liked than before (Havermans and Jansen 2007b; see Fig. 6.4). This positive flavour–flavour learning effect was also demonstrated in a more recent study by Capaldi and Privitera (2008). In a first experiment they had 2- to 5-year-old children repeatedly taste grapefruit juice mixed with the sweet taste of sucrose. This led to increased liking of unsweetened grapefruit juice. Moreover, this positive shift in liking proved stable for weeks. In a second experiment, undergraduate students were presented with several occasions in which they were instructed to consume one small stalk of cauliflower and another stalk of broccoli. One of the two vegetables Table 6.2 Current obesity rates in the USA State Obesity rate (%) 1 Mississippi 32.5 2 Alabama 31.2 3 West Virginia 31.1 4 Tennessee 30.2 5 South Carolina 29.7 Top five states with the most obese people in the USA, as reported by the Trust for America’s Health in their sixth annual issue report (Levi et al. 2009) on the obesity epidemic in the USA
80
R.C. Havermans and A. Jansen
Fig. 6.4 Children’s changes in hedonic ranking of vegetables as the result of flavour–flavour learning. Mean shift in ranking score (from pretest to posttest) for the CS paired with glucose (CS+) compared with the shift in ranking for the unpaired CS (CS−), as demonstrated by Havermans and Jansen (2007b) in the conditioning of vegetable flavour preferences in children
would be sweetened by having it dipped in sugar water. The assignment of vegetable to sugar water was counterbalanced between participants. Capaldi and Privitera found that the pairings of either cauliflower or broccoli with sugar increased liking of the taste of these vegetables. In sum, the results from the studies discussed above suggest that the flavour–flavour learning paradigm can be applied to increase children’s liking of fruits and vegetables. Importantly, this can be achieved even when the taste of the CS is initially disliked (e.g. the bitter taste of grapefruit), plus the achieved hedonic shift appears to be stable over a longer term.
6.4 Applications to Other Areas of Health and Disease Children can develop strong food preferences and as a result may be particularly finicky about having to eat certain foods (Dovey et al. 2008). When such picky eating becomes a longer term habit of eating a very limited variety of food items, this food selectivity can form a serious health risk. One way to treat such food selectivity is through mere exposure. Williams et al. (2008), for example, used this method in six children being treated for extreme food selectivity. One case concerned a young girl diagnosed with autism who during treatment was exposed regularly to foods presented in meals and taste sessions. Meals would contain three table spoons of about three different foods (fruits, vegetables, meat, starch or dairy product), and taste sessions were used to introduce the child to a specific novel food. This mere exposure worked very well. She learned to accept and eat many different foods. Of the 49 different novel foods she learned to eat, she still ate 47 at 3 months after treatment. Mere exposure thus seems to be a viable method to treat food selectivity. However, flavour–flavour learning may be even more powerful in the treatment of such food selectivity. Compared with a mere
6 Acquired Tastes: Establishing Food (Dis-)Likes by Flavour–Flavour Learning
81
exposure procedure, flavour–flavour learning requires relatively few learning trials. Such flavour–flavour learning may involve the pairing of a novel or nonpreferred food with an already liked food. Furthermore, this pairing should be simultaneous, that is, the two foods (liked and disliked) should be presented together, rather than sequentially, as was shown by Holman (1975) in his rat studies. In one experiment, Holman found that pairing a flavour CS (CS+; cinnamon or wintergreen) with saccharin induced a flavour preference relative to another unpaired flavour (CS−; cinnamon when CS+ was wintergreen, and vice versa). Holman, however, failed to induce a flavour preference for the CS+ in a following experiment when he inserted a 30 min interval between the consumption of the CS+ and the ingestion of a saccharin solution. The procedure, design and results of these experiments are illustrated in Fig. 6.5. Piazza et al. (2002) compared simultaneous and sequential pairings of non-preferred foods with preferred foods in three children treated for food selectivity. One 11-year-old girl diagnosed with PDD ate lettuce with salad dressing and a few other creamy foods, but not much else. She received exposure to foods from different food groups (fruits and vegetables, starches and protein-rich foods) and these foods were paired with salad dressing. One group of foods (A) was always presented simultaneously with the salad dressing. Another group of foods (B) was paired sequentially with the dressing. The girl would only come to eat the foods from group A, not the foods from group B, corroborating the notion and previous findings that simultaneous pairings are superior in inducing an evaluative shift. Similar findings were reported for the other two cases in this study. It thus seems that food selectivity can be treated by means of evaluative conditioning, but note that the researchers did
Fig. 6.5 Holman experiments indicating that only simultaneous CS−UCS pairings induce a shift in CS flavour preference. An illustration of the procedure, design and main results of two experiments by Holman (1975; Experiments 3 and 4). Both experiments comprised comparing a shift in liking/preference between a CS+ paired with saccharin (the UCS) and another explicitly unpaired CS flavour (i.e. CS−). At test, the animals received the opportunity to drink from both (now unsweetened) flavours. The tests results from both experiments taken together clearly demonstrate that simultaneous (Experiment 3) but not sequential pairing (Experiment 4) of the CS+ with saccharin induces a flavour preference relative to CS−
82
R.C. Havermans and A. Jansen
not present the ‘non-preferred’ foods separately without the unconditioned flavour after the repeated pairings. This makes it impossible to assess any positive shift in liking of the flavour of the nonpreferred food itself. Furthermore, as the authors themselves note, they did not conduct a follow-up test. Whether flavour–flavour learning is indeed beneficial and perhaps even superior to a mere exposure procedure in the treatment of food selectivity thus requires further empirical validation.
6.5 Conclusion Evaluative conditioning is the transfer of affective value from an affective stimulus to another stimulus, due to pairing of the latter stimulus with the affective stimulus. As such, flavour–flavour learning – the transfer of affective value to a flavour being paired with an already liked/disliked flavour – can be regarded as a specific instance of evaluative conditioning. Flavour–flavour learning is a robust form of evaluative conditioning and has been demonstrated in both animals and humans. Humans appear especially sensitive to the reinforcing sweet taste of sugar (Yeomans et al. 2008). The reason for this is unknown, but recent research suggests that humans possess orosensory receptors that specifically function to detect the presence of carbohydrates (apart from sweet taste). Oral maltodextrin (not sweet) and glucose (sweet) both directly activated brain regions known to be involved in the processing of the reward value of food, such as the insula, OFC and striatum (Chambers et al. 2009). One may hypothesise then that flavour–flavour learning in humans is probably much more effective when using some form of sugar and may even be evident when using non-sweet carbohydrates as UCS. Flavour–flavour learning has proven to be a powerful technique to change someone’s preference for a flavour and hence food. It is rapid and requires relatively little experience with the flavours themselves. In terms of nutrition and health, it has been shown that it can be applied to increase children’s liking of fruits and vegetables even when the foods are initially disliked (Capaldi and Privitera 2008). However, whether the established positive change in hedonics also leads to a corresponding positive change in behaviour is an effect that is predicted but as of yet has not been examined empirically. The full benefits of flavour–flavour learning to nutrition thus still wait to be examined.
Summary Points • Food choice and intake can be understood in terms of hedonic behaviour; that is, one prefers consuming foods one likes and tries to avoid having to eat foods one does not like. • Food likes and dislikes are mostly acquired through direct experience with food. • Flavour–flavour learning, a form of evaluative conditioning, comprises the transfer of affect (positive or negative) to a flavour by pairing that particular flavour with another already liked or disliked flavour. • In humans, positive flavour–flavour learning seems to be mediated by hunger state. • The amygdala is thought to play a critical role in the acquisition of flavour (dis)likes through flavour–flavour learning. • Flavour–flavour learning can be applied to increase children’s liking of fruits and vegetables and may be a beneficial technique in the treatment of severe cases of food selectivity.
6 Acquired Tastes: Establishing Food (Dis-)Likes by Flavour–Flavour Learning
83
Key Terms Evaluative conditioning: A learning paradigm in which pairing a CS with a positive UCS results in a positive shift in liking for the initially neutral CS, whereas conversely, pairing such a stimulus with a negative UCS results in a negative shift in liking. Flavour–flavour learning: A form of evaluative conditioning that allows for the learning of flavour preferences by pairing a flavour with an already liked flavour, such as the sweet taste of saccharin or sugar. Food selectivity: Feeding disorder diagnosed in infants or children when the child is not eating adequately and the insufficient food intake cannot be attributed to any specific medical condition. Hedonic eating behaviour: The principle that anticipated and derived pleasure guide food choice and food intake, respectively. Pavlovian conditioning: The learning of a response to a CS due to its pairings with a biologically relevant UCS. This conditioned response typically reflects the anticipation of the UCS. Stimulus generalisation: Transfer of prominent qualities of one stimulus – such as its associative and affective value – to another stimulus.
References Baeyens F, Eelen P, Van den Bergh O, Crombez G. Learn Motiv. 1990;21:434–55. Baeyens F, Crombez G, Hendrickx H, Eelen P. Learn Motiv. 1995;26:141–60. Baeyens F, Crombez G, De Houwer J, Eelen P. Learn Motiv. 1996;27:200–41. Baeyens F, Hendrickx H, Crombez G, Hermans D. Appetite. 1998a;31:185–204. Baeyens F, Vanhouche W, Crombez G, Eelen P. Psychologica Belgica. 1998b;38:83–108. Brunstrom JM, Fletcher HZ. Physiol Behav. 2008;93:13–9. Brunstrom JM, Downes CR, Higgs S. Appetite. 2001;37:197–206. Brunstrom JM, Higgs S, Mitchell GL. Physiol Behav. 2005;85:524–35. Capaldi ED, Privitera GJ. Appetite. 2008;50:139–45. Capaldi ED, VandenBos GR. Hosp Community Psychiatry. 1991;42:787–9. Chambers ES, Bridge MW, Jones DA. J Physiol. 2009;587:1779–94. Coppens E, Vansteenwegen D, Baeyens F, Vandenbulcke M, van Paesschen W, Eelen P. Neuropsychologia. 2006;44:840–3. Davey GCL. A reply to Martin and Levey (1994). Behav Res Ther. 1994;32:307–10. De Araujo IE, Rolls ET. J Neurosci. 2004;24:3086–93. De Houwer J, Thomas S, Baeyens F. Psych Bull. 2000;127:853–69. Dibbets P, Havermans R, Arntz A. Behav Res Ther. 2008;46:1070–7. Dovey TM, Staples PA, Gibson EL, Halford JCG. Appetite. 2008;50:181–93. Eertmans A, Baeyens F, Van den Bergh O. Health Educ Res. 2001;16:443–56. Fanselow MS, Birk J. Anim Learn Behav. 1982;10:223–8. Field AP, Davey GCL. J Exp Psychol Anim Behav Proc. 1999;25:211–24. Field AP, Lascelles KRR, Lester KJ, Askew C, Davey GCL. Neth J Psychol. 2008;64:46–64. Green KF, Garcia J. Science. 1971;173:749–51. Havermans RC, Jansen A. Neth J Psychol. 2007a;63:38–49. Havermans RC, Jansen A. Appetite. 2007b;48:259–62. Hermans D, Vansteenwegen D, Crombez G, Baeyens F, Eelen P. Behav Res Ther. 2002;40:217–34. Holland PC, Petrovich GD. Physiol Behav. 2005;86:747–61. Holman EW. Learn Motiv. 1975;6:91–100. Levey AB, Martin I. Behav Res Ther. 1975;4:205–7. Levi J, Vinter S, Richardson L, St Laurent R, Segal L. Trust for America’s Health & Robert Wood Johnson Foundation; 2009.
84
R.C. Havermans and A. Jansen
Martin I, Levey AB. Adv Behav Res Ther. 1978;1:57–102. Mobini S, Chambers LC, Yeomans MR. Appetite. 2007;48:20–8. Myers KP, Sclafani A. Dev Psychobiol. 2006;48:380–8. O’Doherty J, Rolls ET, Francis S, Bowtell R, McGlone F. J Neurophysiol. 2001;85:1315–21. Pearce JM. Anim Learn Behav. 2002;30:73–95. Pelchat ML, Rozin P. Appetite. 1982;3:341–51. Piazza CC, Patel MR, Santana CM, Goh HL, Delia MD, Lancaster BM. J Appl Behav Anal. 2002;35:259–70. Raynor HA, Epstein LH. Psychol Bull. 2001;127:325–41. Rolls ET. Cereb Cortex. 2000;10:284–94. Siep N, Roefs A, Roebroeck A, Havermans R, Bonte M, Jansen A. Behav Brain Res. 2009;198:149–58. Small DM, Zald DH, Jones-Gotman M, Zatorre RJ, Pardo JV, Frey S, et al. NeuroReport. 1999;10:7–14. Swinburn BA, Sacks G, Lo SK, Westerterp KR, Rush EC, Rosenbaum M, et al. Am J Clin Nutr. 2009;89:1723–8. Visscher TLS, Kromhout D, Seidell JC. Int J Obes. 2002;26:1218–24. Williams KE, Paul C, Pizzo B, Riegel K. Appetite. 2008;51:739–42. Yeomans MR, Mobini S. J Exp Psychol Anim Behav Process. 2006;32:460–6. Yeomans MR, Leitch M, Gould NJ, Mobini S. Physiol Behav. 2008;93:798–806. Zellner DA, Rozin P, Aron M, Kulish C. Learn Motiv. 1983;14:338–50.
Chapter 7
Personality Traits in the Context of Sensory Preference: A Focus on Sweetness Paul Richardson and Anthony Saliba
Definitions Agreeableness: Degree toward being considerate, cooperative, and accommodating Character: A combination of traits that are learned (e.g., responsibility) Conscientiousness: Degree of self-discipline, persistence, and perfectionism Disposition: The predominant tendency or pattern Empathy: Capacity to recognize and share the feelings and emotions of others Extraversion: Outgoing and highly sociable personality dimension Harm Avoidance: Anticipatory worry and fear of uncertainty – similar to neuroticism Impulsiveness: Inclination to act on the spur of the moment without considering long-term implications Neuroticism: Propensity to experience emotional events negatively; linked with levels of anxiety and stress Novelty Seeking: Sensation seeking – similar to venturesomeness and impulsivity Openness: Willingness to explore new and unfamiliar experiences Personality: An enduring pattern of a person’s psychological and behavioral characteristics Temperament: The collection of unconscious dispositions Trait: A distinguishing characteristic of someone’s personality Venturesomeness: Thrill-seeking behavior, particularly for high-risk and potentially hazardous situations
P. Richardson (*) Brain, Behaviour & Cognition Research Group, Psychology, Sheffield Hallam University, Collegiate Campus, Sheffield, UK S10 2BP e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_7, © Springer Science+Business Media, LLC 2011
85
86
P. Richardson and A. Saliba
7.1 Introduction The consumption of food and drink is necessary to maintain life. However, there is widespread variation not only in the amount of food eaten, but also in the types of food consumed; the presence of preferences lies behind the mixed patterns of consumption. Certain tastes may be disliked by the vast majority of the population (e.g., bitter-tasting foods) while others (particularly sweet tasting), may be actively preferred (Mennella et al. 2005). The processes associated with these preferences are complex; multifactorial models have been developed in order to delineate the influences underlying the preferences but these can be broadly defined as food effects, that is, the physical and sensory properties of food and non-food effects, for example, personspecific reasons.
7.2 Sensory Properties of Food The food-related effects (the sensory properties of food such as texture, appearance, and taste) have been of prime interest in choice behavior research; intuitively, any food perceived to be unpleasant along these dimensions is unlikely to be consumed. The texture of an edible substance (e.g., its viscosity, crispness, or grittiness) is known to relate directly to food selection in both primates and humans (Araujo and Rolls 2004). Similarly, the color and appearance of food has been demonstrated to interfere with judgments of flavor intensity and is known to influence the pleasantness and acceptability of foods. Children seem to be particularly sensitive to these effects, often refusing to consume a particular food if it has bits in it (De Moura 2007) or happens to be green (see later section on “Sweet Taste Preference”). However, it is how a food product tastes that has been the main focus of the food-choice research.
7.3 The Basic Tastes The human tongue has receptors for what have been described as sweet, sour, salty, and bitter tastes. These are classically described as the four basic food tastes, and each have been associated with variable levels of like/dislike among the general population. Umami (savoriness) and fat both have a growing recognition in the literature as basic tastes that have corresponding receptors on the tongue and unique neural responses (Bachmanov and Beauchamp 2007). Other “tastes” such as spiciness are not unique, and what may be (erroneously) perceived as taste (the burning irritation) can then be construed as a pleasant or unpleasant sensation.
7.4 Individual Differences in Taste Preference The sensation of taste arises from chemical stimulation of specialized cells, taste receptors, which are grouped in small clusters called taste buds. Although located throughout the oral cavity, taste buds are in the highest concentration on the human tongue in structures called papillae (see Fig. 7.1 below).
7 Personality Traits in the Context of Sensory Preference: A Focus on Sweetness
87
Fig. 7.1 A diagram of the human tongue. The location of the circumvallate, foliate, and fungiform papillae are indicated in bold (Amended from Gray 2000. Reproduced with permission)
Individual differences exist for the exact number present on the tongue and intuitively, taste intensity perception is correlated with the number of papillae present (Miller and Reedy 1990). This biological basis for individual difference in taste preference has thus far been difficult to study; however, with recent advancements in genetic research, it is now becoming clear that genetic differences account for some individual difference in taste preference (Drewnowski et al. 2001). This research thrust is exciting because a genetic basis for preference would transcend cultural boundaries, however it will be several decades before results can be used for any serious applied outcome. Individual differences could also potentially stem from idiosyncratic likes and dislikes, demographics, psychological, and personality traits. It is entirely plausible that such individual differences could predict taste preferences, however in many published articles these individual differences are often considered “annoying sources of variance in research…” and “referred to as nuisance variables…” (Stevens 1996 p.303). Obvious differences such as gender and age tend to be included in research, but the less-obvious individual differences, particularly along the psychological dimensions are not. Yet, in other research fields such as health psychology, the opposite pattern is found and individual differences are taught and studied. The emphasis is one of inclusion, and variables such as personality are regarded as potentially important distal determinants of health behavior. They can provide “useful evidence about the nature of mechanisms underlying sensory phenomena and thus are important in the generation of research hypotheses” (Stevens 1996, pp. 303).
88
P. Richardson and A. Saliba
7.5 Gender Differences Sex differences have been established in various food-related investigations such as cravings for particular food types, proportion of fruit and vegetables consumed, and avoidance of fats from meat (Goldberg and Strycker 2002). These effects appear to exist as early as primary school (Caine-Bish and Scheule 2009). In relation to sweetness, women do tend to rate sweeter tasting products as more pleasant than men are reported to do (e.g., Laeng 1993); in a study investigating sweet taste preference in obesity, nearly 32% of the female participants rated themselves as having a strong sweet taste preference in comparison to less than 13% of the male participants (Elfhag and ErlansonAlbertsson 2006). However, this apparent gender preference for sweet tasting products does not translate to females consuming greater quantities of sweet snacks (Grogan et al. 1997). This same study reported that women were more ambivalent toward eating sweet snacks than men (perceiving them to be significantly less healthy) and were also more influenced by social pressures than men (Grogan et al. 1997). It is not uncommon for research studies to employ single-sex studies to avoid such entanglements: in a recent study investigating sweet intake, liking, and the urge to eat, in relation to alcohol dependence, the authors decided not to recruit female participants due to the social ambivalence regarding sweet snacks (Krahn et al. 2006). We would recommend against this type of subject selection where it can be avoided, instead, we recommend examining the potential determinants of variation. Males tend to be less sensitive to bitter and sour tastes than females. Acidity is a natural preservative and is an important ingredient in a range of products; bitterness tends to be minimized but is nevertheless a natural flavor in some foods (e.g., the important antioxidant activity in olive oil is related to polyphenols that elicit a bitter taste). Sweetness may be used to mask certain levels of sour and bitter tastes. It follows that females may seek higher levels of sweetness, firstly because of a natural preference and secondly to mask perceived acidity and bitterness given higher acuity than reported for males.
7.6 Personality Personality can be described as the relatively enduring pattern of an individual’s unique psychological and behavioral characteristics, which can exert some degree of influence over the thoughts and behavior of the individual (e.g., Dworetzky 1999). Different models of personality abound, but those based on trait theory suggest that these dispositions are relatively stable, enabling a high degree of predictability in a person’s behavioral repertoire. Secondary traits (or states) tend to be more transient or only appear under certain conditions and can relate to attitudes and preferences. Biological models argue that particular aspects of personality are heritable (see Fig. 7.2) and that these can map onto specific neurotransmitters in the brain with predictable behavioral responses. One theory of personality that is commonly reported in research is Cloninger’s temperament and character model (Table 7.1), based on his work with people who possessed “extreme” personality profiles (Cloninger 1987). Temperament is depicted as a collection of unconscious dispositions that can differentiate people quantitatively. The four constructs of “temperament” include novelty seeking, harm avoidance, reward dependence, and persistence. In contrast, “character” refers to constructs that demonstrate an acquired or learnt component, such as the degree of self-directedness, co-cooperativeness, and self-transcendence (Cloninger 1993, 1994). See Table 7.2 for an explanation of these terms.
7 Personality Traits in the Context of Sensory Preference: A Focus on Sweetness
89
Fig. 7.2 Estimated heritability of personality factors from the Big five questionnaire. Similar data from two independent studies indicate a high degree of consistency in this estimation. Key: O openness, C conscientiousness, E extraversion, A agreeableness, N neuroticism
7.7 Temperament and Eating Behavior It is the constellation of temperament qualities that has received most attention in research due to the forging of a direct link between personality and biological mechanisms. Temperament is assumed to have a constitutional or biological basis strongly influenced by heredity. Three strands of Cloninger’s theory of temperament have been linked to neurotransmitter functioning in the brain, and are known to modulate eating behavior and the regulation of feeding. Specifically these are: dopamine (novelty seeking), serotonin (harm avoidance), and noradrenalin (reward dependence). Cloninger posits that high novelty seeking is a proxy measure of dopamine activity within the mesolimbic regions of the brain (Cloninger 1994). Dopamine is a neurotransmitter thought to modulate the rewarding aspects of “motivating” stimuli, particularly during the learning of new likes and dislikes, the seeking out of new and novel experiences. Animal studies indicate that dopamine activity is linked more to the anticipatory desire and motivation (the wanting) rather than pleasure derived from consummation (the liking) (Robinson et al. 2005). Increases in dopamine activity in the mesolimbic and orbitofrontal regions of the brain have been associated with response to pleasant tastes and these same regions show a higher degree of activity when hungry compared to satiated (Tataranni et al. 1999). As such, dopamine is a primary candidate for involvement in food reinforcement, and by extension, in mediating taste preference. Low levels of serotonin are typically reported in patients with clinical levels of anxiety and depression. Subscales of harm avoidance include anticipatory worry, fear of uncertainty, and shyness, all of which can be considered prototypical features of clinical anxiety (Cloninger 1993). Those with high levels of harm avoidance may actively avoid novel situations or those that may involve confrontation or punishment. More specifically, raising serotonin levels has been demonstrated to suppress appetite, lower food intake, and lower fat intake (for a review see Lam et al. (in press)). Classical and operant conditioning are two processes that underpin the ability to learn, essentially the pairing of conditioned stimuli/cues with responses. The acquisition of paired associations is thought to be modulated by noradrenaline. Atypical noradrenaline functioning has
90
P. Richardson and A. Saliba Table 7.1 Key terms of Cloninger’s temperament and character theory of personality (Cloninger 1998) Descriptors of extreme variants Temperament dimension High Low Harm avoidance Pessimistic Optimistic Fearful Daring Shy Outgoing Fatigable Energetic Novelty seeking Exploratory Reserved Impulsive Rigid Extravagant Frugal Irritable Stoic Reward dependence Sentimental Critical Open Aloof Warm Detached Sympathetic Independent Industrious Lazy Determined Spoiled Ambitious Underachieving Perfectionistic Pragmatic Character dimension Self-directedness Responsible Blaming Purposeful Aimless Resourceful Inept Self-accepting Vain Disciplined Undisciplined Cooperative Tenderhearted Intolerant Empathic Insensitive Helpful Hostile Compassionate Revengeful Principled Opportunistic Self-transcendent Self-forgetful Unimaginative Transpersonal Controlling Spiritual Materialistic Enlightened Possessive Idealistic Practical
been demonstrated in patients with posttraumatic stress disorder, who are overly sensitive to trauma-related cues (for review, see Yehuda 2002). Typically, it is thought of as governing the body’s “fight or flight” response in times of acute stress, and on the TCI, its subscales are sentimentality, attachment, and dependence. Low levels of reward dependence are associated with being aloof, withdrawn, and detached; high levels are synonymous with being open, revealing, and dedicated (Cloninger 1993). Drugs that target noradrenaline receptors have been demonstrated to promote weight loss by suppressing appetite and producing early satiety.
7.8 Sweet Taste Preference Sweetness has been referred to as an innate or naturally preferred taste, even in very young children, while salty, sour, and bitter are acquired tastes that need some perseverance, often extending into late childhood before being liked. Foods that are rated as pleasant and consumed in great quantities by
7 Personality Traits in the Context of Sensory Preference: A Focus on Sweetness
91
Table 7.2 Temperament personality constructs indicated with neurotransmitter systems and associated subsequent behavioral response (Amended from Cloninger 1998. The Genetics and Psychobiology of the Seven-Factor Model of Personality pp.68). In Kenneth R. Silk (1998) Biology of Personality Disorders Temperament construct Principal neurotransmitter Relevant stimuli Behavioral response Novelty seeking Dopamine Novelty Exploratory pursuit CS of reward Appetitive approach Active avoidance, escape CS or UCS of relief of monotony or punishment Passive avoidance, extinction Harm avoidance Serotonin Conditioned signals for punishment, novelty, or frustrative nonreward Formation of appetitive CS Reward dependence Noradrenaline Reward conditioning (pairing CS and UCS) CS conditioned stimulus, UCS unconditioned stimulus
children in most western countries are those with a high sugar and fat content, while vegetables are regarded as universally disliked. This pattern of taste preferences appears to be universal and transcends cultures (Prescott et al. 1992). Additional evidence for the innate preference of sweet tastes comes from studies involving neonates. A common methodology has been adopted by several studies – neonates are given solutions containing the basic tastants and facial expressions are observed immediately after ingestion (see Fig. 7.3). By observing positive facial expressions, sweet tastes were found to be universally accepted, while sour and bitter tastes were associated with tell-tale facial responses indicating an aversion (e.g., Steiner 1977). Recent research suggests this innate preference for sweet tastes may have a partial genetic basis (Mennella et al. 2005), and this regulates the consumption of sweet foods (Reed et al. 1997). Keskitalo and colleagues (2007) found a relationship between the frequency of consuming sweet foods and a genetic marker on chromosome 16. They had participants rate the pleasantness of various tastants, and reported preference for the extremely sweet solution, reporting an estimated heritability factor of 40% (Keskitalo et al. 2007). However, two caveats to this study are (1) the unequal gender split in favor of women (68% vs. 32%), and (2) the age range of participants (18–78 years). No data was published describing the demographics, but demographic variables were used as covariates in the analysis. The innate tendency for a sweet preference might confer an adaptive advantage – a sweet taste signals the presence of sugars and thus of immediate high calorific value. In contrast, sour and bitter tastes are perhaps indicative of noxious substances and the decomposition of food and thus best avoided. Infants and older children often appear to be picky with food, particularly if they have not tasted it before. This mild neophobia illustrates the process of children learning what is safe to eat and what is not. Neophobia has been identified in adults and can predict behavior pertaining to ingestion of novel vs. familiar foods and willingness to eat novel foods (Pliner and Hobden 1992). The authors categorized neophobia as a personality trait, and report that it correlates positively with measures of trait anxiety but negatively with novelty/sensation seeking (Pliner and Hobden 1992). Both of these traits have genetic links, suggesting that neophobia is also heritable. Indeed, one twin study estimated the heritability of neophobia at 67%, though this was only with female participants (Knaapila et al. 2007). This evidence is supportive for the notion that likes and dislikes of specific tastes and foods can be mediated by personality traits.
92
P. Richardson and A. Saliba
Fig. 7.3 Neonates displaying variable facial responses to basic tastes (Reproduced from Steiner 1997. With permission). Key: 1 rest, 2 water, 3 sweet, 4 sour, 5 bitter
One early study to investigate personality traits and taste preferences was completed in India by Venkatramaiah and Baby Devaki (1990). They asked 38 female postgraduate students to self-report food preferences via checklist, and measured personality traits with the Indian Personality Inventory (unpublished measure created by the same authors in 1980). The personality traits were derived from the Indian philosophy of Samkhya. They reported main effects for personality and food types, speculating that sweet foods were preferred most by those with high “tamas” scores (a trait that maps onto Cloninger’s harm-avoidance trait). However, no post hoc analyses were ran to assess group differences for any tastes. A recent study investigated the purported influence of trait anxiety on food taste preferences in Japan (Kato et al. 2006) (Fig. 7.4). The authors divided 73 university students into 2 groups according to the trait scores from the State-Trait Anxiety Inventory, and then tested for sensitivity to sweet, salty, and sour tastes. Of prime interest was the finding that the trait-anxiety group had a higher sensitivity to sweet tastes but not salty or sour. Kato and colleagues (2006) posited that people with high trait anxiety experience more stress on a daily basis, and that this was instrumental in causing the change to eating habits. However, the research was based on young university undergraduate students with no breakdown of gender. Given that the study did not employ a longitudinal design and no premorbid dietary measures were taken, it may be speculative to conclude that trait anxiety caused any changes to eating habits. The link between trait anxiety and sweet preference is supported by an earlier study investigating personality and dietary habits (Kikuchi and Watanabe 2000). They tested 470 university undergraduate students who were required to keep daily records of their dietary habits recorded over a 2-year period, designed to capture the frequency and amount of 40 pre-specified food types consumed. Personality was measured using a modified version of the NEO Five Factor Inventory (Costa and McRae 1992). This provided scores on the “big five” factors of personality: openness, conscientiousness, extraversion, agreeableness, and neuroticism. High scores on each of the five personality factors
7 Personality Traits in the Context of Sensory Preference: A Focus on Sweetness
93
Fig. 7.4 A taste preference experiment in progress. This may usually take place in a partially secluded booth, in which the participant is able to make judgments on preferred food or drink tastes by sampling the products; this permits a fine degree of control and manipulation of several variables – in contrast to interview-based research, which relies on participants recollecting taste preferences in the absence of any actual tasting
had characteristic features; of specific interest was the finding that those with high neuroticism had a pronounced sweet taste preference irrespective of gender (Kikuchi and Watanabe 2000). This finding was supported by Elfhag and Erlanson-Albertsson (2006) in a study of 60 clinically obese patients awaiting treatment at a clinic. Sweet (and fat) taste preference was assessed via structured interviews; on this basis, patients were assigned to one of four groups: strong sweet preference, strong fat preference, strong preference for sweet and fat, and no strong taste preference. Personality was measured with the Swedish University Scales of Personality. The Three Factor Eating Questionnaire was also used to provide measures of cognitive restraint (attempts to limit food intake), disinhibition (difficulties in limiting food intake), and hunger experience (subjective feelings of hunger). The results indicated that those who demonstrated a strong preference for sweet tastes (as opposed to no strong taste preference) showed higher levels of neuroticism; in particular, the neuroticism subscales that were critical for this distinction were lack of assertiveness and embitterment. Sweet taste preference bore no relation to any of the three eating characteristics (Elfhag and Erlanson-Albertsson 2006). The authors suggest these imply low self-confidence and self-esteem, and an externalized locus of control for difficulties. However, one earlier study (Stone and Pangborn 1990) instead found that preference for sweetness was higher in university students with an “outgoing” personality style, a trait that Elfhag and Erlanson-Albertsson conceded was the “opposite of neuroticism and rather implies being more unconcerned and carefree” (p. 64). However, the authors commented that the relation between the outgoing trait and liking for sweetness was not straightforward, as participants who did prefer sweeter drinks did not differ on personality measures from those who did not prefer the sweeter drinks (Stone and Pangborn 1990). Overall, these recent studies all indicate that those with high levels of trait anxiety and neuroticism exhibit a sweet taste preference. It has been proposed that this may be characterized by “comfort eating,” indicative of some reward deficiency syndrome (Wang et al. 2003) or possibly reflective of some “self-medicating purpose that exceeds basic energy requirements” (Davis et al. 2004 p. 132). There are indications that the opioid neurotransmitters may be involved in the rating of sweet tastes as pleasant and in reducing observed behavioral measures of stress after eating sweet tasting foods
94
P. Richardson and A. Saliba
(Dallman et al. 2003). Dallman and colleagues (2003) proposed that symptoms of chronic stress could be attenuated by the consumption of sucrose (a sweet tasting sugar) by way of dampening sympathetic activity in the hypothalamo–pituitary–adrenal (HPA) axis, a feedback system known to be dysfunctional in anxiety disorders (e.g., Yehuda 2002); it is thought that this is mediated through opioidergic pathways (Dallman et al. 2003). We recently published a study investigating personality traits associated with a preference for a sweet white wine (Saliba, Wragg and Richardson 2009). Forty-five participants (with a near-equal gender split) were categorized as preferring either a base or sweet taste on the basis of their hedonic response to a dry white wine and another sample with fructose added at 20 g/L. Personality traits were assessed with the big five personality inventory and the Impulsiveness Venturesomeness and Empathy Scale (IVE - Eysenck and Eysenck 1991). We found that the sweet preference group had significantly lower levels of the openness to experiences trait but higher levels of impulsivity (see Fig. 7.5). Openness to experience has been portrayed as a proxy measure of the willingness to explore new and unfamiliar experiences, ideas, and feelings (Costa and McCrae 1992) and may be akin to the novelty-seeking trait in Cloninger’s measure of temperament (Cloninger 1993). Previous studies suggest that those with high openness report healthier dietary practices (Goldberg and Strycker 2002). We speculated that infants with a preference for non-sweet taste do not receive sufficient dietary stimulation and so are open to new experiences in order to fulfill this. The finding that sweet taste was preferred by those with high levels of impulsivity is consistent with previous research (Davis et al. 2004). Animal studies have reported a positive association between preference for sweet solutions and the subsequent intake of alcohol (Sinclair et al. 1992). In a study conducted in patients with alcohol dependency, Kampov-Polevoy and colleagues reported a preference for a solution containing the
Fig. 7.5 Graph depicting mean average scores of the dry vs. sweet wine taste preference groups on the Impulsiveness, Venturesomeness and Empathy (IVE) and Big five (OCEAN) personality questionnaires (data from Saliba et al. 2009). Note: the scores from the two personality tests are not comparable with each other, and are shown for illustrative purposes only. Key: *denotes significance at p juices > control. Total lunch energy intake (including the preload) was reduced after eating the apple (solid) as preload. Mourao et al. (2007) studied the effect of food form on appetite and energy intake in matched beverage and solid food forms. They compared watermelon and watermelon juice (high carbohydrate), cheese and milk (high protein), and coconut meat and coconut milk (high fat). In all cases, the beverage food form elicited a weaker compensatory dietary response than the matched solid food form. Total energy intake was 12–19% higher on the beverage days compared to the solid food days, which was due more to a weak effect on satiety than on satiation. Not all studies showed an effect in the same direction. Santangelo et al. (1998) reported that meal consistency influences both gastric emptying rate and satiety sensation. In their study a homogenized vegetable-rich meal was found to be more satiating than when the meal was offered in clearly separated solid and liquid components. The overall gastric emptying rate was slowed after the homogenized meal (followed by a CCK peak later). Flood and Rolls (2007) investigated meal intake after consuming preloads of different soups with the same energy density: broth and vegetables served separately, chunky vegetable soup, chunky-pureed vegetable soup, and pureed vegetable soup. Subjects consumed less of the test food after eating the soup compared to no soup, but the type of soup did not significantly influence test meal intake.
10.8 Mechanism The mechanisms underlying the effect of texture on satiety are not well understood. Physical properties such as viscosity and texture could affect chewing, oro-gastric handling of foods, and absorption (Rolls 2009). Chewing solid foods may give a satiety signal, which is not induced by swallowing a liquid (Haber et al. 1977; Mattes 1996; Rolls 2009). From a recent paper by Cassady et al. (2009) it was concluded that masticating nuts, and likely other foods and nutrients, results in important differences in appetitive and physiologic responses.
10 Texture and Diet Related Behavior: A Focus on Satiation and Satiety
139
10.8.1 Cognitive and Sensory Factors Cognitive and sensory factors could play a role in the effect of food form on satiety. People might have different beliefs about the satiating capacity of solid foods compared to beverages. Expected satiation might influence portion sizes (Brunstrom and Shakeshaft 2009). It may be that the mechanisms through which texture affects food intake work through differences in perception due to differences in oral processing. Also the satiating capacity of soups is thought to be in part cognitive: a soup is believed to be nutritive and impacts appetite more than thirst (Mattes 2006). Eating rates of liquid foods are higher than eating rates of (semi-)solid foods (Haber et al. 1977; Kissileff et al. 1980; Zijlstra et al. 2008). Based on the study of Zijlstra et al. (2008) it may be hypothesized that the mechanisms through which texture affects food intake works through a higher/ longer sensory exposure time and/or a longer transit time of the product in the oral cavity. A liquid remains a short period in the mouth, while a semisolid product is eaten more slowly and thus stays longer in the mouth. This increases the exposure time to sensory receptors in the oral cavity and therefore there is more opportunity for exposure to taste, smell, texture, and so on (Zijlstra et al. 2008). It is shown that bite size and oral processing indeed affect ad libitum food intake of a semisolid product (Zijlstra et al. 2009b). Also, smaller food sizes (nibbles versus bars) resulted in lower intake (Weijzen et al. 2008) and small sip sizes reduced the ad libitum intake of soft drinks compared to large sip sizes (Weijzen et al. 2009). A factor that plays a role in the termination of food intake is the degree of sensory-specific satiety for that food. This refers to a decrease in the reward derived from consumption for the food eaten compared to a not eaten food (see, e.g., Rolls et al. 1981). For example, after consumption of a plate of macaroni, one’s pleasure for macaroni is decreased, while the pleasure of custard remains the same or increases. Foods may differ in their degree of sensory-specific satiety. For some foods, the pleasure derived from consumption will decline sooner than for others. Texture-specific satiety might play a role in meal termination. Guinard and Brun (1998) showed that pleasantness of texture, and desire to eat hard test foods decreased after eating a hard lunch food. This might also have played a role in the study of Weenen et al. (2005). They suggested that eating cheese biscuits might have given rise to a higher degree of boredom than eating pears in light syrup, caused by the dryness and thus amount of saliva required to form a bolus that can be swallowed.
10.8.2 Gastrointestinal Processes Gastrointestinal processing can partly explain the effect of viscosity on intake. Soluble dietary fiber may alter viscosity of gastrointestinal contents and this may result in effects on satiety and food intake (Dikeman and Fahey 2006). Juvonen et al. (2009) did find an effect of bran viscosity on gastric emptying and an effect on satiety hormones. Also Santangelo et al. (1998) found effects of physical state of a meal on gastric emptying and CCK release. An effect of viscosity on satiety hormones could not be found in the study of Zijlstra et al. (2009a). An explanation might be that the starch used by Zijlstra et al. (2009a) was already broken down to glucose by enzymes in saliva before it reached the stomach. The mechanism of how viscosity, gastric emptying, and satiety interact is not clear yet. Tieken et al. (2007) examined whether solid meal-replacement products differed from liquid meal-replacement products in appetite and appetite-regulating hormones. They did find differences in appetitive responses, insulin and ghrelin, but not in CCK and leptin, between the products. However, as the authors mentioned, although the products had a similar energy content, the macronutrient compositions differed, which might have affected the appetite and hormonal responses.
140
A. Stafleu et al.
10.8.3 Learned Associations Zijlstra et al. (2008) suggest that differences in satiety responses between liquids and solids are partly based on learned behavior during infancy. Since the viscosity and caloric density of human breast milk appear to vary together, breast feeding may provide an important initial exposure to a general rule that thicker substances contain more calories than thinner substances (McDaniel et al. 1989; Davidson and Swithers 2004). This indicates that the mouth feel of a product already could have an effect on the relationship between viscosity and intake. Mars et al. (2009) indicated that the higher viscosity of a food, and thus the longer orosensory stimulation, may facilitate the learned association between sensory signals and metabolic consequences. However, current trends in eating patterns and food supply (high energy drinks and low-energy solids) may alter learned associations between hunger or thirst and the post-ingestive consequences of eating and drinking (McKiernan et al. 2008).
10.9 Conclusions Most studies on the effect of food form on satiety compared liquids with a solid form. From these studies it appeared that liquids are less satiating than solids. In addition, there seems to be an independent role of viscosity on satiation. More research is necessary to study textures with less pronounced differences than liquid versus solid foods, for example, the effects of texture differences within solid products. Also, the gastric role in food texture warrants further investigation.
10.10 Applications to Other Areas of Health and Disease Several studies showed that viscosity and texture differences influence satiation (meal termination) and to a lesser extent satiety. Consumption of an energy-yielding beverage poses a greater risk for consuming more energy than a semisolid or solid food. From a review of Malik et al. (2006) it appeared that both short-term and long-term studies have shown that energy ingested from sugarsweetened beverages add to the total energy intake during the day. Liquid calories may lead to a positive energy balance and subsequent weight gain (DiMeglio and Mattes 2000). This knowledge can be applied both in the underweight and overweight situation.
Summary Points • Food properties that increase the process of satiation (meal termination) and/or induce longerterm feelings of satiety (increase the intermeal interval) may help to control weight. • Liquid foods elicit weaker suppressive appetite responses and a weaker compensatory response in energy intake than semisolid foods or solid foods. • Ad libitum intake of foods decreases as viscosity increases. • The mechanisms underlying the effect of texture on satiation and satiety are not well understood. • It may be hypothesized that the mechanisms through which texture affects food intake work through a higher/longer sensory exposure time and/or a longer transit time of the product in the oral cavity.
10 Texture and Diet Related Behavior: A Focus on Satiation and Satiety
141
Definitions of Key Terms Mastication: The process by which food taken into the mouth is converted into a form suitable for swallowing. Satiation: The process that determines when we stop eating, and therefore determines meal size. Satiety: The process that suppresses the internal drive to eat (appetite). Sensory-specific satiety: The decrease in pleasantness of a product after eating that product to satiety, compared to the pleasantness of selected uneaten control products. Texture: The sensory and functional manifestation of the structural, mechanical, and surface properties of foods detected through the senses of vision, hearing, touch, and kinesthetics. For definitions see Rogers and Blundell (1979), Rolls et al. (1981), Szczesniak (2002).
References Bleich SN, Wang YC, Wang Y, Gortmaker SL. Am J Clin Nutr. 2009;89:372–81. Blundell JE, Green S, Burley V. Am J Clin Nutr. 1994;59:728S–34. Brunstrom JM, Shakeshaft NG. Appetite. 2009;52:108–14. Cassady BA, Hollis JH, Fulford AD, Considine RV, Mattes RD. Am J Clin Nutr. 2009;89:794–800. Davidson TL, Swithers SE. Int J Obes Relat Metab Disord. 2004;28:933–5. De Castro JM. Physiol Behav. 1993;53:1133–44. De Graaf C, Blom WAM, Smeets PAM, Stafleu A, Hendriks HFJ. Am J Clin Nutr. 2004;79:946–61. De Wijk RA, Zijlstra N, Mars M, de Graaf C, Prinz JF. Physiol Behav. 2008;95:527–32. Dikeman CL, Fahey GC. Crit Rev Food Sci Nutr. 2006;46:649–63. DiMeglio DP, Mattes RD. Int J Obes. 2000;24:794–800. Flood JE, Rolls BJ. Appetite. 2007;49:626–34. Flood-Obbagy JE, Rolls BJ. Appetite. 2009;52:416–22. Guinard JX, Brun P. Appetite. 1998;31:141–57. Haber GB, Heaton KW, Murphy D, Burroughs LF. Lancet. 1977;2:679–82. Hulshof T, De Graaf C, Westrate JA. Appetite. 1993;21:273–86. Hutchings JB, Lillford P. J Texture Stud. 1988;19:103–15. Juvonen KR, Purhonen AK, Salmenkallio-Marttila M, Lähteenmäki L, Laaksonen DE, Herzig KH, et al. J Nutr. 2009;139:461–6. Kissileff HR, Klingsberg G, Van Itallie TB. Am J Physiol. 1980;238:R14–22. Malik VS, Schulze MB, Hu FB. Am J Clin Nutr. 2006;84:274–88. Mars M, Hogenkamp PS, Gosses AM, Stafleu A, de Graaf C. Physiol Behav. 2009;98:60–6. Mattes RD. Physiol Behav. 1996;59:179–87. Mattes R. Physiol Behav. 2005;83:739–47. Mattes R. Physiol Behav. 2006;89:66–70. Mattes RD, Cambell WW. J Am Diet Assoc 2009;109:430–437. Mattes RD, Rothacker D. Physiol Behav. 2001;74:551–7. McDaniel MR, Barker E, Lederer CL. J Dairy Sci. 1989;72:1149–58. McKiernan F, Houchins JA, Mattes RD. Physiol Behav. 2008;94:700–8. McKiernan F, Hollis JH, McCabe GP, Mattes RD. J Am Diet Assoc. 2009;109:486–90. Mourao DM, Bressan J, Campbell WW, Mattes RD. Int J Obes. 2007;31:1688–95. Prinz JF, Lucas PW. Proc R Soc Lond B. 1997;264:1715–21. Rogers PJ, Blundell JE. Psychopharmacology. 1979;66:159–65. Rolls BJ. Physiol Behav. 2009;97:609–15. Rolls BJ, Rolls ET, Rowe EA, Sweeney K. Physiol Behav. 1981;27:137–42. Russell K, Delahunty C. Food Qual Prefer. 2004;15:743–50. Santangelo A, Peracchi M, Conte D, Fraquelli M, Porrini M. Br J Nutr. 1998;80:521–7. Szczesniak AS. Food Qual Prefer. 2002;13:215–25. Tieken SM, Leidy HJ, Stull AJ, Mattes RD, Schuster RA, Campbell WW. Horm Metab Res. 2007;39:389–94. Tordoff MG, Alleva AM. Am J Clin Nutr. 1990;51:963–9.
142
A. Stafleu et al.
Tournier A, Louis-Sylvestre J. Appetite. 1991;16:17–24. Tsuchiya A, Almiron-Roig E, Lluch A, Guyonnet D, Drewnowski A. J Am Diet Assoc. 2006;106:550–7. Van der Bilt A, Engelen L, Pereira HW, van der Glas HW, Abbink JH. Physiol Behav. 2006;89:22–7. Van Vliet T, van Aken GA, de Jongh HHJ, Hamer RJ. Adv Colloid Interface Sci. 2009;150:27–40. Vuksan V, Panahi S, Lyon M, Rogovik AL, Jenkins AL, Leiter LA. Nutr Metab Cardiovasc Dis. 2009;19:498–503. Weenen H, Stafleu A, de Graaf C. Food Qual Prefer. 2005;16:528–35. Weijzen PLG, Liem DG, Zandstra EH, De Graaf C. Appetite. 2008;50:435–42. Weijzen PLG, Smeets PA, de Graaf C. Br J Nutr. 2009;102:1091–7. Wolf A, Bray GA, Popkin BM. Obes Rev. 2008;9:151–64. Zijlstra N, Mars M, de Wijk RA, Westerterp-Plantenga MS, De Graaf C. Int J Obes. 2008;32:676–83. Zijlstra N, Mars M, de Wijk RA, Westerterp-Plantenga MS, Holst JJ, De Graaf C. Physiol Behav. 2009a;97:68–75. Zijlstra N, de Wijk RA, Mars M, Stafleu A, de Graaf C. Am J Clin Nutr. 2009b;90:269–75.
Chapter 11
Sensory Education: French Perspectives Caroline Reverdy
Abbreviations EPODE Ensemble Prévenons l’Obésité des Enfants: Preventing Childhood Obesity Together PNNS Programme National Nutrition Santé: French national diet and health program
Family (meal)
Sensory education
EFFECTS
NO
Food habits
Description of food Food neophobia Identification of taste and odor Categorization of odor Liking of more complex products
YES School (meal)
Society (program of health) Culture (cooking)
Effects in children food preferences and behaviours via EXPOSURE to food Effects in children food preferences and behaviours via INFORMATION on food : children
11.1 Introduction Eating is one of the most common actions in everyday life. What about tasting food? Is there any such thing as sensory education or educating the sense of taste? We shall take the example of one country, France – well known for its culinary culture – to see what factors implicitly or explicitly
C. Reverdy (*) AUXIME - ODOROSMÊ, Les Grandes Roches, 69490, Saint Romain de Popey, France and PANCOSMA, R&D Department, 6 Voie des Traz, 1218 Le Grand Saconnex, Switzerland e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_11, © Springer Science+Business Media, LLC 2011
143
144
C. Reverdy
construct our sense of taste. Then we shall see how explicit sensory education at school could change eating preferences and behavior. And finally, we shall compare sensory education to other forms of education and lay out the prospects of this method.
11.2 Can the Sense of Taste Be Educated? 11.2.1 Learning in Life As with other omnivores, human eating preferences and habits are mainly learned. Learning varies with the particular food culture, but the general mechanisms are comparable. Children’s tastes are influenced by their mother’s feeding habits, first of all by straightforward conditioning in the womb (Schaal et al. 2000), and then via the mother’s milk and gradual diversification of diet (Maier et al. 2006). Later, they are shaped by what is eaten at family meals, then in playgroups and at school meals. To pure conditioning are thus added mechanisms such as incident and non-intentional learning by simple exposure to new foodstuffs (Zajonc 1968; Pliner et al. 1993), parental teaching, whether to encourage or to discourage (Birch 1980; Hanse 1994), and imitation of peer behavior (Birch 1980). With the exception of parental teaching, these mechanisms influence behavior in an implicit way, without reaching the cognitive level and without the child being conscious of them. Thus the variety of culinary experience stimulates the development of preferences more or less positively depending on the atmosphere of the feeding situation and on the individual’s own olfacto-gustatory sensitivity. This process develops and enriches sensory experience in terms of what flavors are known. The range of sensory experiences thus depends on circumstances, but also on individual personality (Pliner and Salvy 2006), as in neophobia – where the individual is afraid to try foods that are new to him or her (Loewen and Pliner 1999). Children are dependent on what foodstuffs they happen to be given; teenagers, on the other hand, become more independent and make their own choices. They can get past their neophobia and extend their sensory experience. This then stabilizes in adulthood, in the encounter with the partner’s eating habits, and then those adopted by the couple (Köster 1990). Simple exposure (Zajonc 1968) to traditional foodstuffs and learning mechanisms (i.e., culture) strongly influence the development of preferences; other mechanisms determine the orientation and dynamics of preference. These influences are the subject of the psychology of exploratory behavior and motivation (Dember and Earl 1957; Berlyne 1970; Walker 1980) and are shown schematically in Fig. 11.1. Berlyne (1970) explained how a stimulated organism expresses maximal preference for an optimal stimulus activation level, preference being lesser for any other stimulus level. Dember and Earl (1957) further showed how the optimal level changes after exposure to stimuli that are more stimulating than the individual’s previous optimum level (pacer): the optimum shifts toward the level of the more stimulating stimuli. Exposure to stimuli that are less stimulating than the optimal level, on the other hand, do not alter the optimum but rather lead to boredom, as predicted by Walker (1980). These theories, based on visual stimuli, have been confirmed using olfacto-gustatory stimuli (fruity drinks) (Lévy et al. 2006). Walker (1980) showed that stimulus complexity as perceived by the subject decreases with exposure – which would explain why novelties at first seem too complex but then come to be more appreciated with time; with yet more exposure, they begin to lose their perceived complexity and may become commonplace.
11 Sensory Education: French Perspectives
145
Fig. 11.1 Inverted-U relationship between liking and the arousal potential of a stimulus, as suggested by Berlyne’s (1967) arousal theory (solid curve), and the shift (broken curve) of the original inverted-U curve and the optimal individual level of psychological complexity upon exposure to a “Pacer” (B), as suggested by the arousal theories of Dember and Earl (1957) and Berlyne (1967) (From Lévy et al. 2006). Liking changes with perceived complexity depending on the optimum of the individual
This phenomenon, known as product boredom, is well known in marketing (Köster and Mojet 2007). Thus, taste can evolve under various mechanisms and according to experience. Food preferences are flexible and subject to mainly implicit learning. The question therefore arises as to whether food preferences might also be modifiable over a shorter term by explicit learning such as sensory education.
11.2.2 What Is Sensory Education? 11.2.2.1 Definition Sensory education develops the senses by focusing attention on them. The sense of taste is developed by information regarding taste perception (which involves all five senses: taste, smell, touch, sight, and hearing) and by practical training to enhance sensory acuity. Sensory education is to be distinguished from sensory training. The former concerns the sense of taste in general and is intended for nonexpert consumers, whereas the latter applies to a particular type of foodstuff and is intended for expert analysts.
11.2.2.2 Examples of Sensory Education Encouraging children to explore their bodies is not a new idea. As early as 1930, Rampillon & Gruyer-Wantrin already explained that “young children should learn to use their senses, just as they need to learn to acquire other skills”.
146
C. Reverdy
In France, many nonprofit organizations, such as Agropolis-Museum, Pomme et Sens, and (the oldest) the Institut du Gout, work in schools and playgroups, running taste education sessions, with a hedonistic approach, developing the pleasure associated with tasting. The Institut du Goût (Taste Institute) introduces teachers to the Classes du Goût (Taste Classes) method developed in France by Jacques Puisais in the 1970s. This learning method, aimed at children around the age of 9, seeks to develop food curiosity, refine taste, and enrich food vocabulary. In 2001, Daviet explained the three main lines of this method: the first is “knowledge of the senses, balanced diet, production of some foods, and vocabulary enrichment”; the second is “know-how – for example, how to compose meals and menus”; the third is “self-awareness of the body and of sensations”. The first version of the module, in 10 sessions, came out in 1984, after more than 10 years of trials and prototypes run in schools in Tours and Paris (Puisais and Pierre 1987; Puisais 1999). It was then applied in schools in many regions of France, for more than 20 years. For example, the method is used by the association Les Sens du Goût (Senses of Taste), from the Avesnois region, which runs a comprehensive, permanent, and extensive program of developing and education of taste in the local area. In particular, the association runs information, education, and training campaigns for the population as a whole, to raise awareness of local identity and the sense of belonging to a specific area (Pautrel 2002). Subsequently, the creation of the Institut du Goût in 1999 gave the method a new lease of life, while respecting its original nature. Thus, in 2002, a CD-ROM1 (Puisais et al. 2002) presented a second version, in 12 sessions. Transmission to teachers is based not only on the CD-ROM, but also by training courses run by the Institut du Goût. In recent years, this French teaching method has been translated and adapted for other European countries (Sweden, The Netherlands, Finland). Thus, the transmission of taste education is now developing in research programs such as Sapere. This program, set up in 1994, is based in Brussels. Sapere – from the Latin for both knowledge and taste – began as an international association and went on to become an international non-profitmaking foundation. It brings together experts from the human sciences, research, industry, and the education and communication sectors. Its partners include many celebrities, scientific institutes, corporations, and associations such as EPODE (Ensemble Prevenons l’Obésité Des Enfants: Preventing Childhood Obesity Together). Since its inception, Sapere has promoted Jacques Puisais’s method throughout Europe – for instance, in the Scandinavian countries (Jonsson et al. 2005). In other countries (The Netherlands, Switzerland), the focus has been on implicit methods for 6–9 year olds. This type of sensory education for healthy children is quite recent. Sensory education also exists as a form of therapy.
11.2.3 Application to Other Areas of Health and Disease 11.2.3.1 Obesity and Sensoriality Sensory education of taste can further be expected to improve food habits and therefore help to limit overweight and the prevalence of obesity. The new generation of dietary education (e.g., EPODE) encourages adding a sensory dimension in their programs, with tastings and cooking. Nevertheless, it must be stressed that the effect of sensoriality on limiting overweight and obesity is quite indirect and needs further investigation to be proven.
1
This program is only available in the form of a compact disc – read only memory.
11 Sensory Education: French Perspectives
147
11.2.3.2 Disability and Sensoriality General sensory education, not specifically targeting taste, has been extended to work with disabled (blind, deaf) children. In France, Patty Canac, a perfume specialist, runs rehabilitation for cranial trauma, cerebral hemorrhage, and prolonged coma victims. She helps these patients find a taste for life again, using olfactory stimulation to awaken memory by the pleasure of playful learning (Canac et al. 2008).
11.3 What Factors Educate Our Sense of Taste? 11.3.1 Culture It is well known that food preferences are linked to exposure to a given foodstuff (Wardle et al. 2003) and that we tend to consume the food that is part of our particular culinary culture. We are thus conditioned to appreciate more highly the foodstuffs of our own culture.
11.3.2 Family Food education within the family involves exposure to foods: children are encouraged to taste. The parents play a role and may reinforce neophobia to a greater or lesser extent, whether directly or by showing a strong emotional reaction to their children’s not eating (Hanse, l994). Such parental pressure and restriction regarding food are measurable on the Kids’ Child Feeding Questionnaire (KCFQ) (Carper et al. 2000).
11.3.3 School Meals Half of French children take lunch at school (Stratégies 2006). This plays a role in their food education. However, it amounts to only a small number of their annual meals and cannot in itself ensure balanced diet. The elementary taste education provided at school is regulated by the French Education Ministry’s Le Bulletin Officiel (2001), which encouraged diversity of diet and the development of taste. It is also suggested the following main lines of activities around the issue of food (see Table 11.1). These recommendations advise providing children with explicit information, but also enriching their implicit experience via tastings and education concerning odors and spices. Table 11.1 Key points of the main lines suggested by the French Ministry of Education (2001) for activities around the issue of food 1. “Educating pupils’ taste and promoting culinary heritage and products of good gustatory and nutritional quality 2. Highlighting precise vocabulary concerning flavors, especially for children who confuse terms, so that they are able to define their sensations clearly 3. Explaining the secrets of the production and composition of foods 4. Tasting local specialties 5. Discovering odors, spices, and essences” The French Ministry of Education suggested five main lines for activities around the issue of food from food vocabulary to culture of food
148
C. Reverdy
Even so, food education as provided by the family and school meals may prove insufficient to ensure healthy eating; dietary education programs have therefore been set up by the authorities.
11.3.4 French Ministry of Health Taste education and dietary education are often associated. At the present time – in France as in other industrial countries – there is a worrying increase in the prevalence of childhood overweight and obesity in the 5–12-year-old age group from 3% in 1965 to 16% in 2000 (Rolland-Cachera et al. 2001). Daniel Thomas (2003) stressed the importance of changing behavior with respect to obesityrelated cardiovascular risk. It was with this in mind that the French authorities set up the PNNS in 2001, to improve public health by acting on the main determinants of diet. Information and education for young people is one strategic axis of the PNNS (Programme National Nutrition Santé: French national diet and health program), stressing the importance of early and lasting dietary education, concerning both food and physical exercise. The Ministry of Health and the Ministry of Education are working together on this. One of the priorities of the circular on School children’s health: 5-year prevention and education program (dated December 1st, 2003) was dietary education and the prevention of overweight and obesity. For this, dietary education and taste and consumption education are to be included in curricula as of primary school, as supports for teaching and in conjunction with school activities in general. Several programs are already being implemented in schools. Information documents and PNNS-approved intervention tools have been made available (Kerneur et al., information on the Internet). Each guide contains simple and accessible information and practical advice adapted to the individual’s habits so as to combine health, pleasure, and the requirements of day-to-day living in meeting the objectives of the PNNS. The guide to children’s diet is called Health comes by eating and moving: a child and teens diet guide for all parents. Traditional dietary education also seeks to provide information, but often in terms of good and bad, which is liable to give rise to feelings of guilt. The environmental factors influencing food preference are summarized in Fig. 11.2.
Fig. 11.2 Environmental factors impacting food preference. Culture, society, school, and family have influences on the food preferences
11 Sensory Education: French Perspectives
149
11.3.5 School: Classes Du Goût As described above, Classes du Goût modules can be run in schools. We shall now detail their principles and content. The teaching approach in each session is structured according to the following model. First, there is a questions phase, helping the children explore the theme of the day; then, a tasting phase, which provides information and enables the children to reach a practical understanding of how the gustatory system works; finally, the children’s various responses and answers are brought together, with complementary input on the initial questions. The module comprises 12 sessions of 90 min, on the following topics: “the five senses”; “taste”; “vision”; “olfaction”; “touch and hearing”; “aroma”; “flavor”; “preparation of a dish”; “food preferences”; “regional specialties”; “recapitulation”; and “the festive meal”. Table 11.2 presents the program of this sensory education module. Table 11.2 Key points of the description of the sensory education program (Reverdy et al. 2008) Title and aim of the Organization of the Conclusion and/or contribution of the session session session − All five senses are necessary to The teacher presents a 3-step Lesson 1 “The five senses” establish contact with the world tasting procedure (before, This session evokes the of foods. during, and after) using pupils’ understanding the five senses to describe − This contact always comes about of the way in which a food. This session in three phases (before, during, one makes contact allows children to enrich and after the tasting). with foods. their vocabulary on what they feel sensorially and emotionally. − Food offer a wide variety of tastes The teacher presents varied Lesson 2 “Taste” and chemical sensations in the tastings to let pupils This session lets the mouth. discover their own taste pupils discover the perception and to find out − Knowing how to express different basic tastes (sweet, that the answers they give tastes and sensations. salty, sour, bitter, and are similar to others’, yet umami) and other − Showing the diversity between still very personal. mouth sensations individuals in gustative perception. (prickling, burning, and astringency). − Colors have an impact on other Tasting making use of Lesson 3 “Vision” sensory perceptions. colorants, to show how This session lets the visual perception creates − Establishment of precise vocabulary pupils understand expectations that can for visual perception. how vision creates modify other perceptions. expectations that allow them to anticipate food taste. − Showing the relationship between Odor bottles (more or less Lesson 4 “Olfaction” an odor and the evocation of its familiar odors, fruity This session shows pupils source and associated memories. odors, etc.) are presented how difficult it is to to evoke pupils’ memories − Identifying certain familiar odors. recognize odors. and lead to identification. − Various sensations are linked to Samples of different Lesson 5 “Touch and hearing” touch during tasting: tactile, materials (silk, wool, This session aims to auditory, and thermal perception. velvet, etc.) are presented. enrich vocabulary for Foods of varying − Enriching vocabulary on touch and touch and to show hardness, crispness, and hearing. how touch and crumbliness are tasted. hearing complement − Linking together food texture, each other. consistence, and temperature. (continued)
150
C. Reverdy
Table 11.2 (continued) Title and aim of the session Lesson 6
Lesson 7
“Aroma” This session shows how aromas in mouth are perceived by retro-nasal olfaction and how the same food eaten cold or warm gives off different aromas. “Flavor” This session studies the sensations of sessions 2, 4, and 6 simultaneously.
Organization of the session
Conclusion and/or contribution of the session
Various highly aromatic foods are tasted, and the teacher explains how one can get rid of the aromatic sensations by using ortho- or retro-nasal routes.
− Distinction between direct (odor) and indirect (aroma) olfaction. − The temperature of food modifies sensation.
The pupils learn to identify these different sensations when simultaneously present in complex foods, and to identify their development over time. This cooking workshop takes place with a professional.
− Definition of flavor. − Forms of interaction between the senses involved. − Synthesis of all themes dealt with so for (important information and vocabulary).
− Experiencing the pleasure of “Preparing a dish” preparing a dish. In this session, with the − Understanding the variety of help of a professional, possible ways to eat a given food. the pupils prepare food from a recipe. − Discussion of eating habits. − Individual food preferences differ Tasting unfamiliar food: each Lesson 9 “Food preferences” a lot. group is required to The pupils are asked to search for taste informa− Talking about your preferences and argue for their tion about one fruit and arguing for them. preferences and to then to encourage the stimulate their − Accepting the taste of new foods. others to taste it. curiosity for new foods. − Presenting regional culinary The teacher and pupils bring Lesson 10 “Regional specialties” specialties brought in by the pupils, in local specialties for The pupils note and thinking about their origin. tasting. differences in local − Enlarging knowledge by tasting and international specialties from other regions and specialties and how cultures. cultural history may explain them. Quiz, synthesizing acquired − Remembering acquired knowledge Lesson 11 “Recapitulation” knowledge. that can be of help in assessing food. In this session, the pupils have to remember the knowledge built up during the previous lessons. The pupils eat a meal at a − Applying acquired knowledge. Lesson 12 “The festive meal” restaurant. − Taking part in a relaxing experience: This last session is an the pleasure of eating together. opportunity for the pupils to transfer the − Developing skills linked to table things they have manners and savoir-vivre. learned during the − Assessing the effect of preservation previous lessons by on the taste of food. evaluating a meal prepared for them. The sensory education program of Classes du goût contains 12 lessons with theoretical knowledge and practical activities Lesson 8
11 Sensory Education: French Perspectives
151
Our cultural, family, social, and school environments can implicitly influence food behavior. How can explicit sensory education at school alter food preferences and behavior?
11.4 H ow Far Can Sensory or Dietary Education Change Food Preferences and Behavior? 11.4.1 Edusens in France A program over a period of years to assess the benefits of sensory education has been set up by the European Center of Taste Science in Dijon, France. The first part of the study focused on young adults, while the rest concerned 8–11 year olds. 11.4.1.1 Assessment of Sensory Education in Young Adults The aim of the study was to run a sensory education program for young adults and then to assess its impact on their consumer behavior, complexity perception, and preferences. The assumption was that educating attention to taste would shift preferences from simple to more complex versions of food, due on the one hand to improved sensory performance (sensitivity, identification, discrimination, and description) and on the other hand to a reduction in the perceived complexity of complex variants, making them less unsure and more pleasant (in line with the theories of Berlyne (1970) and Walker (1980)). In the first step, a panel of 67 consumers assessed perceived complexity and stated their preferences for sets of chocolate, coffee, and tea of varying aromatic complexity. Sensory acuity was measured on a battery of tests developed for the study. Then, in the second step, the subjects were divided into two groups of equal size. One of the groups underwent 12 sessions of sensory education, with practical exercises and theoretical data, while the other group did not. At the end of the module, the two groups returned to the laboratory to do the same tests as in the first step. The results confirmed the initial hypotheses that complexity defined a priori by the formulae corresponded to perceived complexity for all the foodstuffs, and that initial preferences correlated negatively with aromatic complexity. Evolution following sensory education showed a change in the experimental group’s assessments. Their descriptive performance was improved. The perceived complexity of the two most complex variants decreased, particularly with regard to variables that were hard to describe and identify. These changes led to increased preference for these two items, whereas no change was observed in the control group (Reverdy et al. 2004). 11.4.1.2 Evaluation of Sensory Education in Children This study was run with a panel comprising an experimental group of 100 children and a control group of the same size. The experimental group took part in 12 one-and-a-half-hour sessions of taste education in class (J. Puisais’s Classes du Goût) during the primary school year. The entire panel took part in three measurement sessions comprising three laboratory assessments each, before the sensory education module (T0), just after it (T1), and in the following school year, some 9 or 10 months after the module (T2), in order to test durability of impact. The results showed increased liking in both groups for the more aromatic and intense variants at T1, with this increase persisting in time (T2) only in the experimental group. Thus, repeat assessment
152
C. Reverdy
(or exposure) had a stronger effect than education initially (T1), the education effect appearing only later (T2) in the form of a consolidation of the exposure effect (Reverdy et al. 2010a). Education increased children’s neophilia, but only temporarily (Reverdy et al. 2008). Education improved description of tasting experience toward more objective rather than subjective criteria, and this improvement was lasting. Finally, education shifted the categorization strategy for unknown odors toward a strategy based on less hedonic criteria (Reverdy et al. 2010b, c). A new method for measuring food choice was set up for the study, but showed no sensory education effect on choice behavior. In conclusion, sensory education as carried out here showed some impact on food preference and behavior, but not durably, and mainly affected description of tasting experience (Reverdy 2008). Some of these conclusions were anticipated by Ton Nu (1996), who assessed the first version of the Classes du Goût in France. In 144 9–12 year olds, 69 of whom had had Classes du Goût, she found a positive education impact on verbalization of sensation and on attention to food quality, but none on the desire to taste new foods or on consumption patterns for unfamiliar and familiar items. Even so, the children who had been in the Classes du Goût claimed to be more tempted to taste new foods and were interested in the history of foodstuffs. These findings put the validity of declarative assessment in doubt when not followed by congruent behavior. The validity of Ton Nu’s results is moreover limited by the lack of any pretest to enable intra-subject comparison before and after education.
11.4.2 Classes Du Goût in Finland A second study, conducted in Finland (Mustonen et al. 2009), evaluated the effect of sensory education on taste and odor awareness and food ratings in school children. Two hundred and forty-four school children, aged 7–11 years, from two schools in Helsinki area, were involved. In each school, two distinct treatments – educated (96 children) and uneducated (79 children) – were applied for a 2-year period. The sensory education contained ten lessons of the Classes du Goût and five lessons familiarizing the children with different food categories. During the 2-year period, the two groups were assessed four times each on the following parameters: free odor naming, taste identification of six solutions, descriptive characterization of two breads, rating attention paid to the sensory properties of food, willingness-to-try rating for unfamiliar vs. familiar foods, and aided odor naming (5 odors, 10 verbal labels). During the test period, the “educated” children improved their skills to identify tastes and odors and to characterize foods, while the control group exhibited no evolution. However, “education” effects were mainly to be found in younger children, and were small and not always consistent over the 2-year period. However, children activated their odor and taste perceptions and improved their ability to describe sensory properties of food after sensory education. Finally, Classes du Goût sensory education can influence food preferences and behavior, whether in France or in Finland, as summarized in Fig. 11.3.
11.4.3 Dietary Education The effects of dietary education were measured in a French study conducted in the towns of Fleurbaix and Laventie, and called Fleurbaix – Laventie Ville Santé. This epidemiological study started in 1992 in these two towns in the Nord-Pas-de-Calais region of France, with the initial objective of assessing dietary education in young children. The first phase consisted in a dietary survey in the region, which had been chosen due to the high local prevalence of overweight. Children were then
11 Sensory Education: French Perspectives
153
Fig. 11.3 Measured effects of the Classes du Goût. The effects of Classes du Goût were measured on neophobia, description, liking of complex food, odor categorization, food characterization, and taste and odor identifications
given dietary information. Subsequent change in food behavior in the entire family was observed, validating the hypothesis that children who received dietary education at school would encourage the whole family to adopt better food consumption behavior (Borys et al. 1993; Borys and Lafay 2000; Borys 2003). A positive approach, based on the pleasure of eating, was found to have more effect on families’ food behavior than the restrictive approach of other dietary interventions. This first phase (1992–1997) was followed up by a second, of 5 more years (1997–2002), to explore the determinants of weight gain and the respective impacts of diet, hormones, biological and genetic factors, and physical activity. In phase 3 (2002–2007), the population was offered the possibility of health coaching. Twice-yearly follow-up of a cohort extended to all age groups assessed dietary and behavioral status in the population in comparison with dietary recommendations (Borys et al. 1993; Basdevant et al. 1999; Brouet 2003). Moreover, initial results showed the EPODE approach to have a positive influence, reducing childhood obesity in the local areas involved, compared to increasing obesity in control areas. In the pre-EPODE era, traditional dietary education was mainly based on restriction and avoidance. Unfortunately, it failed to include sufficiently the sensory pleasure associated with mealtimes. This may explain the poorer effectiveness of dietary campaigns in the UK, based almost exclusively on restriction and the analysis of dietary risk.
11.5 Conclusion and Perspectives 11.5.1 C omparison Between Sensory Education at School and Other Forms of Education (Dietary and Family) 11.5.1.1 Comparison with Dietary Education The explicit information and implicit experience provided by Classes du Goût have a more or less lasting effect on children depending on the particular behavior being studied. This teaching method is particularly rich in sensory experience, with attention to each child’s individual differences. It is nuanced, unlike dietary education which tends to involve a dualistic approach in terms of good and bad behavior. Moreover, dietary education is often limited to simple information transmission, divorced from any sensory experience. Classes du Goût offer implicit sensory experience in a positive context enriched with explicit information as in dietary education.
154
C. Reverdy
11.5.1.2 Comparison with Family Education Within the family, children are often encouraged to extend their range by tasting new foods, which they learn, by repetition, to like over time. Such exposure to foods, however, does not always come with a positive feeling, if it involves parental pressure. Classes du Goût offer tastings, introduced by prior information. Moreover, they take place in a context of playful, peer-group learning. Like at home, children extend their food experience in Classes du Goût. However, the method associates implicit experience to explicit information. Moreover, how a food is perceived in the context of an exchange with the parents may be very different from how it is experienced in an exchange with the teacher in the presence of the other children. The comparison of food education types are described in Table 11.3. Table 11.3 Key points of the comparison between types of food education: sensory, dietary, and parental Characteristics of the education type Sensory education at school Dietary education Parental education Sensory experience Yes No Yes Tasting introduced by prior information Yes No No Information without tasting No Yes No Highlighting individual differences Yes No No Exchange Peer groups and teachers Nutritionists Parents Dualistic approach (good, bad) No Yes Yes/no Parental pressure No No Yes Sensory education at school focus on experience and information; dietary education focus on information; parental education focus on experience
11.5.2 Advantages and Limits of Sensory Education at School 11.5.2.1 Advantages of the Method This teaching method has the interest of focusing the children’s attention on their own sensations, finding their own responses rather than some preestablished or conventional response. Its originality compared to other school subjects (such as mathematics) enables children who are doing badly at school to find a space for expression and success, as Öström and Annett (2008) reported in Swedish children. Classes du Goût may not really change food behavior, but do provide general knowledge of the raw materials of food and food education including how dishes are cooked. Given the contemporary food system which encourages consumption of processed foods, this approach seeks to promote healthier eating, closer to the natural products. In this way, it helps combat the fast-food culture. Advertising for processed foods targeting children, such as biscuits, has an impact on the type of consumption. The Classes du Goût culture favors a counterculture. However, as we have seen, its impact is slight in comparison with the effectiveness of implicit learning by food exposure.
11.5.2.2 Limits of the Method The great effectiveness of exposure puts into perspective the effects of explicit education of the Classes du Goût type, which finally provides mainly a gastronomic culture for the children without actually changing their eating habits.
11 Sensory Education: French Perspectives
155
Moreover, while the Classes du Goût approach stresses a combination of the pleasure of eating and explicit learning, it must be said that implicit learning by exposure also involves the postingestion effects of food, which are probably more important than explicit information for the formation of eating habits and the development of food appreciation. Aversion subsequent to bad post-ingestion experiences is much more persistent than attraction caused by good experience, as the avoidance reaction provoked by aversion reduces the chances of correction by new experience, whereas attraction increases the chance of coming up against a bad experience.
11.5.3 Action! The findings of EduSens, showing increased appreciation of complex food items following a small exposure (just three times is enough), should encourage the authorities to seek to change eating behavior by means of exposure to healthy food. A method focusing on the effects of exposure to varied and complex flavors and odors has the advantage over the Classes du Goût approach of being applicable in young children whose language development is incomplete. School meal campaigns, such as the traditional Taste Week in October, are to be encouraged, even if the benefit has yet to be properly assessed. Finally, in the light of the success of the session in which the children got actively involved in preparing a dish and learned to adjust the quantities of ingredients according to their preferences, it would seem that food education should involve active participation in learning to cook, especially as the capability and pleasure of cooking well will be essential when the child reaches adulthood, in order to be able to prepare enjoyable and balanced meals. The kitchen is surely a wonderful place to discover and transmit the pleasure of eating well, as Marie-Claire Thareau Dupire (2006) suggests. This finding was further confirmed in the Let’s make a meal together session in phase 3 of the EduSens project, which got the families involved. This is why we strongly encourage any cooking activity that enables children to put different ingredients together so as to make a dish adjusted to their own tastes.
Key Point • Sensory education: context, courses of action, and effects
Definitions Classes du Goût: (Taste Class) is a method developed in France by Jacques Puisais in the 1970s. This learning method, aimed at children around the age of 9, seeks to develop food curiosity, to refine taste, and enrich food vocabulary Food preferences: Preferences for food are obtained by comparison, whereas liking is absolute information. Thus, preferences can be used to compare several variants of a range, as in the EduSens project. Neophobia: The fear to try foods that are new to the subject Sensory education: Develops the senses by focusing attention on them. The sense of taste is developed by information regarding taste perception (which involves all five senses: taste, smell, touch, sight, and hearing) and by practical training to enhance sensory acuity. It is general and intended for nonexpert consumers. Sensory training: Is training with food, odor, or solutions in order to improve taste ability and is intended for expert analysts in sensory analysis.
156
C. Reverdy
Summary Points • The content of sensory education of taste at school is information regarding taste perception and food processes, and practical training to enhance sensory acuity and to enrich food vocabulary. • Scientific studies showed the following effects of sensory education in children: • Improved description and characterization of food • Improved identification of odor and taste • Temporary decrease in neophobia • Increased liking for more complex food products • More expert odor categorization • The advantages of sensory education at school are to focus on children’s own sensations and responses and provide knowledge concerning food. • Sensory education at school does not really change food habits and its explicit information seems less important than the post-ingestive effects of food. • An alternative to sensory education could be cooking lessons to encourage better food habits. Acknowledgments My warmest thanks to Egon Peter Köster, who guided and supported me throughout my research on sensory education for children. Warm thanks to Pascal Schlich, Christine Lange, and the Edusens team, who welcomed me in their dynamic team and with whom we set up the EduSens program. Warm thanks to Dominique Montoux for reading over the text. She has used the Classes du Goût method with her pupils for 20 years, and is now running training in it as part of the Sapere and EPODE programs. Many thanks to David Bravo, R&D Director for Pancosma, for encouraging me to write this chapter and for his precious corrections.
References Basdevant A, Boute D, Borys JM. Int J Obes. 1999;23(4):10–3. Berlyne DE. In: Nebraska symposium on motivation. Arousal and reinforcement. University of Nebraska Press; 1967. p.1-100. Berlyne DE. Percept Psychophys. 1970;8:279–86. Birch LL. Child Dev. 1980;51:489–96. Borys JM. Nutr Services, Jan 2003:1–2 Borys JM, Lafay L. Rev Méd Suisse Romande. 2000;120:207–9. Borys JM, Boute D, Thomas F, Fontbonne A, Eschwege E. Cahiers Nutri Diet. 1993;28(3):177–80. Brouet I. Imp Médec, 16 May 2003 Canac P, Samuel C, Socquet-Juglard S. InterEditions; 2008. Carper JL, Orlet Fisher J, Birch LL. Appetite. 2000;35:121–9. Daviet C. La Santé de l’Homme. 2001 March-April: 352 Dember WN, Earl RW. Psychol Rev. 1957;64(2):91–6. Hanse L (1994) Unpublished Ph.D. thesis. Paris X-Nanterre University, Paris, France. Jonsson IM, Ekström MP, Gustafsson IB. Int J Consumer Stud. 2005;29:78–85. Kerneur C, Duchène C, Noirot L http://www.mangerbouger.fr/pro/education/enseignants/intro.html Köster EP. Food ingredients Europe conference proceedings 1990. The Netherlands: Expoconsult, Maarssen; 1990. p. 32–7. Köster EP, Mojet J. In: MacFie HJH, editor. Consumer-led food product development. Abington, Cambridge: Woodhead Publishing; 2007. p. 262–80. Lévy CM, MacRae A, Köster EP. Acta Psychol (Amst). 2006;123:394–413. Loewen R, Pliner P. Appetite. 1999;32(3):351–66. Maier AS, Leathwood P, Chabanet C, Issanchou S, Schaal B. Chem Senses. 2006;31:E5. Ministère de l’éducation nationale (2001) BO spécial n°9, 28 June 2001: 6 Mustonen S, Rantanen R, Tuorila H. Food Qual Prefer. 2009;20(3):230–40.
11 Sensory Education: French Perspectives
157
Öström A, Annett J. A sense of innovation, 3rd European conference on sensory and consumer research, Hamburg, Germany; 7–10 September 2008. Pautrel D. Cahier Espaces. December 2002;76:27–30. Pliner P, Salvy SJ. In: Shepherd R, Raats M, editors. The psychology of food choice. Wallingford, Oxon: CABI; 2006. p. 75–92. Pliner P, Pelchat M, Grabski M. Appetite. 1993;20:111–23. Puisais J. Le goût chez l’enfant, l’apprentissage en famille. Flammarion; 1999. Puisais J, Pierre C. Le goût et l’enfant. Flammarion; 1987 Puisais J, Mac Leod P, Politzer N. Le Goût et les 5 Sens, Compact disc – read only memory, Odile Jacob MultimédiaSCEREN; 2002. Rampillon L, Gruyer-Watrin J. Jeux et exercices en vue de l’éducation sensorielle et mentale des jeunes enfants et qui est destiné aux institutrices maternelles et aux mères de famille. Librairie de l’enfance, Paris, 2nd edition; 1930. Reverdy C. Unpublished Ph.D. thesis, Université de Bourgogne, Dijon, France; 2008. Reverdy C, Schlich P, Köster EP, Ginon E, Lange C. Food Qual Prefer. 2010a;21:794-804. Reverdy C, Lange C, Thibaut A, Schlich P, Köster EP. 32nd Achems Annual Meeting, 21–25 April 2010, St. Pete Beach, USA; 2010b. Reverdy C, Lange C, Thibaut A, Schlich P, Köster EP. 20th ECRO Congress, 14–19 September 2010, Avignon, France; 2010c. Reverdy C, Lange C, Schlich P. A sense of identity, 1st European conference on sensory science of food and beverages, 26–29 September 2004, Florence, Italy; 2004. Reverdy C, Chesnel F, Schlich P, Köster EP, Lange C. Appetite. 2008;51:156–65. Rolland-Cachera MF, Bellisle F. In: Burniat W, Lissau I, Cole TJ, editors. The obese and overweight child. Cambridge: Cambridge University Press; 2001. p. 69–92. Schaal B, Marlier L, Soussignan R. Chem Senses. 2000;25:729–37. Stratégies. 1428, 5 Oct. 10, 2006:24. Thareau Dupire MC. Mon enfant mange de tout. Paris: Leduc S Editions; 2006. Thomas D. Pano Méd. 2003;4874:40. Ton Nu C. Unpublished Ph.D. thesis, ENGREF, Paris, France; 1996. Walker EL. Enduring effects of occurrence of psychological events, psychological complexity and preference: a hedgehog theory of behavior. Monterey: Brook/Cole; 1980. p. 129–96. Wardle J, Cooke LJ, Gibson EL, Sapochnik M, Sheiman A, Lawson M. Appetite. 2003;40:155–62. Zajonc RB. J Personality Social Psychol Monogr Suppl. 1968;9(2):1–27.
wwwwwwwwwwwwwww
Part III
General and Normative: Endocrine and Neuroendocrine
wwwwwwwwwwwwwww
Chapter 12
The Role of Cholecystokinin (CCK) in Eating Behavior Mihai Covasa and Timothy Swartz
Abbreviations AP CCK CCK–1R CCK–2R CCK-8 DIO DR DVC Fos-Li GI HF ICV IP LETO LF mRNA NTS OLETF
Area postrema Cholecystokinin Cholecystokinin 1 receptor Cholecystokinin 2 receptor Cholecystokinin octapeptide Diet-induced obesity Diet-induced obesity resistant Dorsal vagal complex Fos-like immunoreactivity Gastrointestinal High-fat Intraventricular Intraperitoneal Long-Evans Tokushima Otuska Low-fat messenger RNA Nucleus tractus solitarius Otsuka Long-Evans Tokushima Fatty
12.1 Introduction A growing number of humoral factors are released from the gut during feeding and they play a prominent role in the cascade of events bringing a meal to an end. The chief among them is cholecystokinin (CCK), the first gastrointestinal peptide implicated in the control of food intake and subsequently coined “satiation signal.” Detected for the first time in 1928 by Ivy and Oldberg and later characterized by Mutt and Jorpes (Jorpes and Mutt 1956), it was not until 36 years ago when Gibbs et al. published the landmark paper showing that the biologically active, synthetic, CCK octapeptide (CCK-8) reduced food intake in the rat (Gibbs et al. 1973). Since then, the M. Covasa (*) INRA, Écologie et Physiologie du Système Digestif, 78350 Jouy-en-Josas, France e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_12, © Springer Science+Business Media, LLC 2011
161
162
M. Covasa and T. Swartz
i nvestigation into the role of CCK on food intake continued unabated making CCK one of the most intensely studied gut peptide. Consequently, its suppressive effects on food intake have been demonstrated in several species including humans (see Ritter 2004a, b). CCK controls food intake by coordinating visceral functions to optimize digestion and absorption and by interacting with other short- and long-term meal-related signals. CCK may also contribute to satiation by reducing caloric consumption, thus exerting its role in the control or regulation of other systems, such as body adiposity. This chapter addresses: (1) the role of CCK on alimentary organs that participate in control of food intake, (2) mechanisms by which CCK controls meal size, (3) interactions of CCK and other hormones that control food intake, and (4) disruptions in CCK signaling pathways leading to disordered phagia.
12.2 CCK Controls Gastrointestinal Functions to Optimize Digestion In addition to its role in controlling food intake, CCK elicits multiple effects on the GI tract including stimulation of pancreatic secretion and gallbladder contraction, bile secretion into the duodenum, inhibition of gastric secretion and emptying, as well as motor functions like lower oesophageal sphincter relaxation and intestinal and colonic motility (see Little et al. (2005)). Together, these actions promote reducing the rate of passage of nutrients in the interest of efficient and complete digestion. That is, CCK may control food intake, particularly protein and fats solely in the interest of GI functions. For a list of main functions of CCK, see Table 12.1.
12.3 CCK Controls Food Intake via Vagal Afferents Vagal sensory afferents synapsing from the gastric, celiac, and intestinal branches of the vagus to the hindbrain are the pathway both endogenous and exogenous CCK use to reduce food intake (Smith et al. 1981; Yox and Ritter 1988). Evidence for this comes from the fact that vagal afferent fibers express CCK receptors and CCK application to vagal afferent preparations results in vagal discharge (Ritter et al. 1989; Raybould and Lloyd 1994). Additionally, sensory fiber removal either by surgical or chemical means attenuates CCK and intestinal nutrient-induced suppression of food intake (Ritter 2004a, b). Finally, vagal sensory fibers innervating gastric and intestinal mucosa are sensitive to CCK (Ritter et al. 1989). These effects are largely due to a paracrine action of CCK on peripheral,
Table 12.1 Key features of CCK in eating behavior 1. CCK is released from the duodenal enteroendocrine I-cells of the small intestine. 2. All three macronutrient classes stimulate release of CCK in humans; however in rodents, only fats and protein cause CCK secretion. 3. CCK’s primary pathway to induce satiation is through a paracrine mode of action through CCK-1Rs. 4. CCK elicits physiological effects on a variety of digestive organs including the stomach, intestine, colon, gallbladder, and pancreas. 5. CCK activates hindbrain neurons through vagal afferent sensory nerve fibers. 6. CCK serves as a satiation signal to limit short-term food intake; however its synergistic effects with other long-term homeostatic energy signals implicate it in reducing bodyweight. Table illustrating the key points of CCK in controlling eating behavior summarizing stimulation of secretion, release, and effects of the hormone
12 The Role of Cholecystokinin (CCK) in Eating Behavior
163
capsaicin-sensitive vagal afferent fibers. However, CCK-producing cells are strategically positioned in very close proximity with vagal afferent terminals in lamina propria, and few synapses-like appositions between them have been reported. Therefore, a neural mode of action has been suggested (Reidelberger et al. 2004) although the largely held view is that locally released CCK diffuses to its action sites through a paracrine mechanism. In addition to a vagally mediated pathway, CCK sites acting through non-vagal mechanisms may also serve to decrease food intake (Reidelberger 1992; Blevins et al. 2000). CCK localized in the central nervous system also has been reported to decrease food intake (Blevins et al. 2000). Released by neurons from the spinal cord to forebrain, CCK is a potent neuropeptide (Hokfelt et al. 2002). CCK-1 and CCK-2 receptors also are highly expressed in many brain regions, including areas known to control food intake and regulate energy balance. Intraventricular (ICV) or parenchimal administration of CCK into several hypothalamic and hindbrain areas decreases food intake (Blevins et al. 2000) while central infusion of CCK receptor antagonists (Corp et al. 1997; Dorre and Smith 1998) or CCK antisera (Della-Fera et al. 1981) increases food intake. Also, mechanical and chemical stimulation of the GI tract induces release of CCK from brain regions controlling food intake (Schick et al. 1989). Further, reduction of intake by infusion of carbohydrates or feeding of diets that do not result in substantial CCK release can be attenuated or reversed by antagonists of CCK receptors (Brenner and Ritter 1996). Finally, whereas vagal afferents mediate CCK-induced suppression of food intake, they are not required for the increased food intake following systemic administration of CCK-1R antagonists (Reidelberger 1992). Therefore, CCK from other sources can reach CCK-1Rs present outside the abdominal vagal terminals and inhibit food intake. Indeed, administration of devazepide, a CCK-1R antagonist that crosses the BBB, results in increased food intake in both vagotomized and non-vagotomized rats, while A-70104 another CCK-1R antagonist that does not penetrate the BBB increased intake only in non-vagotomized rats (Reidelberger et al. 2004). These results suggest that increased food intake is due to an action of the antagonists at CCK receptors located on peripheral sites. However, it also suggests that devazepide increases food intake by acting either on remnant vagal afferents or by accessing other non-vagal CCK receptors (Ritter 2004a, b). It is known that CCK receptors and CCK-containing neurons and terminals are present in the hindbrain that may also participate to CCK’s effects on the vagal hindbrain (Hill and Woodruff 1990).
12.4 CCK Controls Food Intake via Intestinal Nutrients CCK is released from discrete enteroendocrine I-cells concentrated primarily along the proximal duodenal and jejunal mucosa. The apical surface of the CCK cells comes in contact with food components triggering a series of intracellular events resulting in the peptide release from the basolateral cell membrane into the circulation (Liddle 1997) (see Fig. 12.1). Several molecular forms of CCK have been identified and they are derived from the 95 amino acid pro-CCK (CCK-5 to CCK-83) with CCK-8 and CCK-58 being the most biologically potent in suppression of food intake (Glatzle et al. 2008). Reduction of food intake by intraintestinal nutrient infusions is thought to exercise controls of food intake, which normally are activated when components of a meal enter the duodenum from the stomach. In the rat and other mammals, plasma CCK concentrations are elevated in response to intraintestinal products of fat digestion, unhydrolyzed protein, and inhibitors of pancreatic trypsin (Liddle et al. 1986; Weller et al. 1992; Brenner et al. 1993). Some, but not all, intestinal nutrients stimulate secretion of the gut peptide, cholecystokinin. For example, carbohydrates and amino acids do not release CCK in the rat (Brenner et al. 1993); although in humans, CCK is released in response
164
M. Covasa and T. Swartz
Fig. 12.1 Diagram depicting the stimulation of release of cholecystokinin (CCK) (small black circles) from a duodenal I-cell by luminal nutrients (small open circles) on the apical portion of the cell. CCK is released on the basolateral side of the cell into the circulation or binds to localized CCK-1R on vagal afferent terminals, which activate upstream hindbrain neurons
to both l-phenylalanine (Ballinger and Clark 1994) and glucose (Parker et al. 2005). CCK reduces food intake by acting at CCK-1Rs, located on small unmyelinated vagal sensory neurons (South and Ritter 1988), indicating that the substrate that mediates CCK-induced satiation is similar, if not identical, to that which mediates reduction of food intake by intestinal nutrients. Participation of CCK-1Rs in reduction of food intake by intestinal nutrients is well supported by the fact that CCK-1R antagonists attenuate or abolish reduction of food intake by intraintestinally infused triglycerides (Woltman et al. 1995), long chain fatty acids (Yox et al. 1992), oligosaccharides (Brenner and Ritter 1996), and protein (Woltman and Reidelberger 1999). In addition to reversing the reduction of food intake observed following exogenous CCK (Brenner and Ritter 1995) injection or intestinal nutrient infusion, CCK-1R antagonists increase food intake when they are administered alone (Moran et al. 1992; Brenner and Ritter 1995). Taken together, these results suggest a direct relationship between CCK-1Rs and control of food intake by intestinal nutrients. There is accumulating convincing evidence indicating that CCK mediates nutrient-suppression of food intake mainly through a paracrine rather than an endocrine mode of action. For example, both suppression of food intake and inhibition of gastric emptying by intraintestinal carbohydrate infusions, which do not elevate plasma CCK, are attenuated by a CCK-1R antagonist (Brenner et al. 1993). On the other hand, infusions of proteins that markedly increase plasma CCK concentrations have a marginal effect on reduction of sham feeding (Brenner et al. 1993). Furthermore, administration of a peptide CCK-1R antagonist impermeable to the BBB attenuated the satiating effects of CCK and intestinal nutrients (Brenner and Ritter 1995) and increased food intake when given alone (Brenner and Ritter 1995). Finally, the endocrine source of CCK must undergo hepatic portal degradation, which renders most of the biologically active CCK ineffective in suppressing food intake (Reeve et al. 2003). Consistent with this, very low concentrations of CCK (1–5 picomolar range) is detectable in the blood circulation and the only endocrine form of CCK found in the rat is CCK-58 (Reeve et al. 2003). Together, these data overwhelmingly point to a peripheral site of action whereby a local source of CCK acting in proximity of vagal afferent fibers may be sufficient to mediate reduction of food intake. This is also supported by studies showing that low doses of intraperitoneal (IP) which likely mimic the paracrine mode of action on vagal afferents (Canova and Geary 1991), or near-arterial administration of CCK-8 reduces food intake more than higher doses when administered intravenously (Cox et al. 1995). Whether sufficient amounts of systemically administered CCK
12 The Role of Cholecystokinin (CCK) in Eating Behavior
165
are able to penetrate the lamina propria and elicit a physiological and behavioral effect is unknown. Thus, the exact identity of this source of CCK outside “I” cell remains to be located. Studies examining receptor affinity indicate a relationship between local concentrations of CCK in the GI tract and CCK-1R affinity. It is thought that low affinity receptors are localized near areas with high CCK concentration while high affinity receptors are in areas with low CCK levels (Pandya et al. 1994; Talkad et al. 1994). For example, administration of an agonist of high affinity CCK-1Rs does not decrease food intake (Weatherford et al. 1993). Additionally, work done in vagal afferent preparations shows the majority of CCK-1Rs expressed on vagal afferent neurons are thought to be low affinity CCK-1Rs (Simasko et al. 2002). Together, these data suggest that gastrointestinal CCK acting locally on vagal afferent sensory fibers is responsible for nutrient induced satiation. Although the paracrine effects of CCK on peripheral vagal afferents are well documented, a direct action of CCK on the nucleus tractus solitarius (NTS) neurons have been reported. This source of CCK could originate either from CCK-producing neurons located in the hindbrain or from the systemic circulation through the leaky portion of the BBB, including NTS (Baptista et al. 2007).
12.5 CCK Reduces Food Intake via CCK-1 Receptors The targets of CCK action, CCK-1 and CCK-2 receptors (formerly CCK-A and CCK-B), belong to the G-protein coupled transmembrane classification (Kopin et al. 1992; de Weerth et al. 1993). Several studies have identified and cloned the genes encoding these receptors (Kopin et al. 1992; Wank et al. 1994). For a list of sites where CCK-1Rs are located, see Table 12.2. Selective CCK-1 and CCK-2 receptor antagonist studies indicate that reductions of food intake by systemic injections of CCK are mediated primarily via CCK-1Rs (Melville et al. 1992; Moran et al. 1992). Accordingly, reductions in food intake by systemic CCK administration can be significantly attenuated or abolished by administration of a selective CCK-1R antagonist [for review see Ritter et al. (1999)]. Activation of these receptors is mediated via vagal sensory neurons primarily innervating the proximal duodenum and the stomach (Moran et al. 1990, 1997). In addition, independent administration of CCK-1R antagonists increase food intake, supporting CCK as a physiological satiety signal. At the cellular level, binding sites for CCK receptors are identified on vagal sensory neurons, and 30–40% of vagal sensory neurons express CCK-1R mRNA. Furthermore, use of isolated vagal sensory neuron preparations demonstrates that calcium influx occurs after CCK application and is attenuated by CCK-1R antagonist (Simasko et al. 2002). Finally, electrophysiological data show that CCK-1Rs also mediate activation of vagal afferent fibers by intestinal stimuli (Eastwood et al. 1998). Table 12.2 Distribution of CCK-1R • Peripheral ° Gallbladder ° Pancreas ° Pylorus of stomach ° Vagal afferents • Central ° Area postrema (AP) ° Dorsal medial hypothalamus (DMH) ° Median raphe nucleus ° Nucleus accumbens ° Nucleus tractus solitarius (NTS) Peripheral and central locations where CCK-1Rs have been detected
166
M. Covasa and T. Swartz
Together, these findings are strong evidence that CCK-1Rs are necessary for suppression of food intake by CCK. Indeed, the fact that exogenous administration of CCK is ineffective in animal models lacking CCK-1Rs has demonstrated unequivocally the contribution of CCK-1R in control of food intake by CCK and nutrients.
12.6 Changes in CCK Sensitivity by Nutrients Responses to gastrointestinal satiation signals are not fixed. Rather they appear to vary widely with changes in the dietary and endocrine milieu. For example, studies in both rats and mice adapted to a high-fat (HF) diet become less sensitive to both exogenous (Covasa and Ritter 1998; Nefti et al. 2009) (Fig. 12.4), and endogenous CCK via intestinal infusion of oleate (Covasa et al. 2000). Similarly, human subjects adapted to a HF diet have increased plasma CCK following a standard meal compared to control subjects and reported greater hunger following intestinal lipid infusion (Boyd et al. 2003). These effects were associated with an increase in the average daily food consumption and body weight. The exact mechanisms by which maintenance of a HF diet leads to a reduction in sensitivity to satiation signals such as CCK and nutrients are not known. However, similarly, reduction in sensitivity to the anorectic effects of acute CCK injection was also demonstrated in rats receiving chronic infusion of CCK via osmotic minipump (Covasa et al. 2001). This persistent elevated plasma CCK, accompanying long-term exposure to a diet rich in fat leads to several adaptive changes in nutrientand CCK-signaling pathways. For example, relative to rats fed a low-fat (LF) diet, rats fed a HF diet exhibit increase pancreatic secretory and plasma CCK responses to intestinal fat. They also secrete reduced amounts of pancreatic amylase in response to CCK (Chowdhury et al. 2000) and increased amounts of pancreatic lipase (Gidez 1973). In normal Sprague-Dawley rats, chronic HF feeding selectively reduces vagal and enteric neuronal sensitivity to intestinal oleic acid or CCK injection (Covasa et al. 2000). Because fat is a potent stimulus for CCK release, it may be that modifications at the level of CCK-1Rs play an important physiologic role. This is supported by evidence that continuous CCK infusion leads to downregulation of the receptor gene expression in rat pancreatic acinar cells (Ohlsson et al. 2000) and hypothalamo–pituitary–adrenal axis (Malendowicz et al. 2003). Indeed, mice maintained on a HF diet for 15 days have a decreased Fos expression in the NTS in response to CCK or nutrient load and a decreased mRNA level of CCK-1R transcripts in nodose ganglia (Nefti et al. 2009). However, in the rat, Broberger and colleagues (Broberger et al. 2001), using in situ hybridization techniques, reported that feeding an HF diet does not appear to alter vagal CCK-1R mRNA expression in the nodose ganglia. It remains to be determined whether alterations in CCK-1Rs in other tissues accompany the diminished sensitivity of endogenous satiety mechanisms resulting from chronic consumption of dietary fat. CCK-1R affinity and capacity are also reduced in rats adapted to an energy-restricted diet (Kawano et al.). Also, rat pancreatic acinar tumor AR42J cells express the CCK-1R subtype, and CCK-8 reduced CCK-1R mRNA expression to 56% after exposure (Kawano et al. 1992). It is also possible that reducing the number of binding sites at the neuronal membrane surface downregulates vagal sensitivity. In addition to the abnormalities in the binding of CCK to its receptors in genetically obese rats, there are also CCK neuronal changes associated with dietaryinduced obesity. For example, early signs of obesity in neonatal overfed weanling rats are associated with a significant decreased number of CCK-positive neurons in the paraventricular hypothalamic nuclei (Plagemann et al. 1998). Studies of CCK-receptor function in pancreatic acini and Chinese hamster ovary cells (Rao et al. 1997) indicate that receptor internalization and
12 The Role of Cholecystokinin (CCK) in Eating Behavior
167
phosphorylation are important mechanisms for CCK-induced desensitization in vitro. Therefore, reduced sensitivity to CCK could be mediated either by altered receptor protein translation or increased sequestration of previously translated receptors. Downregulation of transduction cascades has also been associated with CCK-induced desensitization of pancreatic amylase secretion (Otsuki and Williams 1983). Therefore, a change in postreceptor transduction is yet another potential mechanism for reduced vagal sensory response to CCK. The current data suggests that modifications at the level of the CCK-1Rs plays an important physiological role in the adaptation of feeding behavior; however, the relationship between decreased sensitivity to CCK or nutrient infusion in animals adapted to a HF diet and the functional characteristic changes occurring in the peripheral CCK-1Rs has not been elucidated.
12.7 CCK Interacts with Other Hormones to Control Food Intake CCK-induced satiation is enhanced when combined with other anorectic signals. For example, serotonin or 5-hydroxytryptamine (5-HT) released from the enterochromaffin (EC) cells in response to carbohydrates serves to terminate a meal in coordination with CCK. While cholecystokinergic and serotonergic systems independently control meal size, when both systems are concomitantly activated, it results in an enhanced suppression of food intake (Burton-Freeman et al. 1999; Helm et al. 2003; Hayes et al. 2004a, b). This enhanced suppression of intake is reversed by simultaneous blockade of CCK-1 and 5HT3R (Hayes et al. 2004a, b; Hayes and Covasa 2005). Further, concomitant blockade of CCK-1 and 5HT3R synergistically enhances food intake and suppression of food intake by CCK is attenuated following blockade of peripheral or central 5-HT3 receptor antagonists (Hayes and Covasa 2006a, b) (Fig. 12.2). These findings demonstrate that CCK and 5-HT systems cooperate interdependently to control food intake. 5HT3 receptor mediated CCK-induced satiation is thought to occur through indirect mechanisms as part of a feedback cascade via inhibition of gastric emptying. Evidence of this comes from the inability of 5-HT receptor antagonists to attenuate CCK-induced reduction of sham feeding. Further, 5-HT receptor antagonism attenuates CCK-induced gastric distention and inhibition of gastric emptying (Hayes and Covasa 2006a, b). Finally, mice that lack specific subtypes of 5-HT receptors, which are localized only in the CNS, have a decreased responsiveness to exogenously administered CCK (Asarian 2009). The interactions between CCK and 5-HT are vagally mediated, which is based on the following evidence: (1) both CCK and 5-HT cause a profound activation of proximal small intestine afferent fibers, which is blocked by vagotomy or capsaicin; (2) 5 out of 9 nodose ganglion neurons that were activated by CCK-8 respond to intraluminal perfusion of 5-HT; (3) a subthreshold dose of CCK-8 that produced no measurable responses augmented the neuronal response to luminal 5-HT perfusion. This potentiation effect was eliminated by a CCK-1R antagonist. In addition to the interaction between CCK and 5-HT, CCK and mechanical stimuli, such as gastric distension, exert synergistic or cooperative effects on the control of food intake in a variety of species, including rats, monkeys, and humans (Moran and McHugh 1982; Feinle et al. 1996). For example, intake after either a gastric distention, a test meal, or a preload is reduced when combined with lower doses of CCK (Moran and McHugh 1982; van de Wall et al. 2005). These effects are mediated by vagal CCK-1Rs. Furthermore, gastric distension enhances CCK-induced hindbrain c-Fos expression, and vagal excitation (Wang et al. 2007). Thus, CCK-1Rs may act by enhancing responses to gastric detention. Given the interaction between leptin and CCK discussed above and that all leptin-responsive gastric vagal afferents are responsive to CCK, leptin may enhance the responses of gastric afferents to distension or CCK. Other anorexigenic and orexigenic factors involved in the control of food intake and bodyweight also interact with CCK either to enhance or inhibit its satiating effects. First, apolipoprotein A-IV
168
M. Covasa and T. Swartz
Fig. 12.2 5-HT3 receptor antagonism attenuates cholecystokinin (CCK)-induced satiation. Intraperitoneal administration of cholecystokinin (CCK, 1.0 mg/kg) significantly reduced rat chow intake at 30, 60, and 120 min compared with control in food-deprived rats. Concomitant administration of ondansetron, a 5-HT3 receptor antagonist (Ond; 1.0 mg/ kg ip) and CCK significantly attenuated CCK-induced reduction of 30-, 60-, and 120-min rat chow intake. *P G 939C>T 957C>T 725 bp 3’ G>T
NS NS NS NS S
Bergen et al. (2005) Bergen et al. (2005) Bergen et al. (2005) Bergen et al. (2005) Bergen et al. (2005)
Dopamine D3 receptor (DRD3) Dopamine D4 receptor (DRD4)
10,620C>T TaqA1 Bal-I
S NS NS
The association was statistically significant in ANBP but not in ANR
3-bp deletion
NS
Hinney et al. (1999b)
48-bp repeat D4pr C(521)T
NS S
D4pr C(616)G
NS
D4pr A(809)G
NS
D4pr 120 repeat
S
D4 exon III repeat 4bp Del/Ins (promoter)
NS
The association was statistically significant in ANBP but not in ANR
Hinney et al. (1999b) Bachner-Melman et al. (2007) Bachner-Melman et al. (2007) Bachner-Melman et al. (2007) Bachner-Melman et al. (2007)
b3-adrenergic Trp64Arg receptor Val158Met Catechol-O(472G/A) methyltransferase (COMT)
NS NS
The association was statistically significant in ANR but not in ANBP patients
S
−1,219A/G 186C/T 408C/G
NS NS NS S
ARVCF
826InsC 659C/T
NS NS
The association was statistically significant in the ANR but not in the ANBP group
Noradrenaline transporter (NET)
S
Bergen et al. (2005) Nisoli et al. (2007) Bruins-Slot et al. (1998)
Bachner-Melman et al. (2007) Urwin et al. (2002)
Hu et al. (2007) Hinney et al. (1997c); Miyasaka et al. (2006) Frisch et al. (2001); Michaelovsky et al. (2005); Micolajczyk et al. (2006) Gabrovsek et al. (2004) Michaekowsky et al. (2005) Michaekowsky et al. (2005) Michaekowsky et al. (2005)
Michaekowsky et al. (2005) Michaekowsky et al. (2005) (continued)
76 The Role of Gene Polymorphisms in Susceptibility to Anorexia Nervosa and Bulimia Nervosa
1181
Table 76.2 (continued) Analyzed Candidate gene polymorphism
Statistical significance Note
References
524T/C
S
Michaekowsky et al. (2005)
Monoamino oxidase A (MAOA) Brain-derived neurotrophic factor (BDNF)
MAOA-uVNTR
NS
The association was statistically significant in the ANR but not in the ANBP group
196G/A (Val66Met)
S
Ribasès et al. (2003, 2004, 2005a); Koizumi et al. (2004); DmitrzakWeglarz et al. (2007)
NS
The Val66Met SNP of the BDNF gene was found quite consistently although not specifically linked to ANR
−270C/T
NS
Neurotrophic tyrosin kinase receptor 2 (NTRK2)
−69C>G
S
IVS2+40C>T NS IVS13+40G>A NS IVS17+125T>C NS IVS18+13G>A NS 2,785– NS 2,785insC Pstl NS
The association was statistically significant in the ANBP but not in the ANR group
Rosenkranz et al. (1998a)
Gly426Gly
NS
Rosenkranz et al. (1998a)
G760A
S
G526A C659T 80T>G
S NS NS
Vink et al. (2001); Dardennes et al. (2007) Vink et al. (2001) Vink et al. (2001) Bergen et al. (2003)
8,214T>C 23,340A>G 47,821A>G 51,502A>T rs569356 rs521809
NS NS S NS S S
rs4654327
S
The association was statistically significant in the ANR but not in the ANBP group The association was statistically significant in the ANR but not in the ANBP group
Neuropeptide Y Y1 receptor Neuropeptide Y Y5 receptor Agouti-related protein (AGRP) Opioid receptor delta-1
Urwin et al. (2003b)
Friedel et al. (2005); Rybakowski et al. (2007); Dardennes et al. (2007) Ribasès et al. (2003, 2004, 2005a); Koizumi et al. (2004); Dardennes et al. (2007) Ribasès et al. (2005b)
Ribasès et al. (2005b) Ribasès et al. (2005b) Ribasès et al. (2005b) Ribasès et al. (2005b) Ribasès et al. (2005b)
Bergen et al. (2003) Bergen et al. (2003) Bergen et al. (2003) Bergen et al. (2003) Brown et al. (2007) Brown et al. (2007)
Brown et al. (2007)
(continued)
1182
P. Monteleone and M. Maj
Table 76.2 (continued) Analyzed Candidate gene polymorphism Cannabinoid receptor 1 (CNR1)
rs204055 rs204047 rs2298896 AAT 7,9–15 repeats
Statistical significance Note
References
NS NS NS S
Brown et al. (2007) Brown et al. (2007) Brown et al. (2007) Siegfried et al. (2004)
The 13-repeat allele was preferentially transmitted in ANBP patients while the 14-repeat allele was preferentially transmitted in the ANR group
NS Muller et al. (2008) −22,959A/G NS Muller et al. (2008) −6,274A/T NS Muller et al. (2008) −6,215T/G NS Muller et al. (2008) −5,489T/C NS Muller et al. (2008) −1,359G/A NS Muller et al. (2008) −272G/A NS Muller et al. (2008) Fatty acid amide hydrolase (FAAH) 10,741C/A NS Muller et al. (2008) 11,966G/A NS Muller et al. (2008) 13,883G/A NS Muller et al. (2008) 19,542C/A NS Muller et al. (2008) NS Muller et al. (2008) N-acylethanolamine- 368A/G hydrolyzing acid amidase (NAAA) 9,263A/T NS Muller et al. (2008) 19,229G/T NS Muller et al. (2008) This table lists published association studies of polymorphic variants of genes involved in the physiology of central regulators (neurotransmitters, neuropeptides and neurohormones) of feeding and energy homeostasis in anorexia nervosa AN anorexia nervosa, ANR anorexia nervosa restricted subtype, ANBP anorexia nervosa binge-purging subtype, 5-HT serotonin, NS not significant, S significant, VNTR variable number of tandem repeats, ARVCF armadillo repeat gene deleted in velocardiofacial syndrome, MAOA-uVNTR MAOA-upstream variable number of tandem repeats
76.2.1.2 Serotonin Receptors 5-HT2A and 5-HT2C receptors play a role in the serotonergic control of appetite, and polymorphisms of genes coding these two receptors have been identified. The -1438G/A polymorphism in the promoter region of the 5-HT2A receptor gene is of particular interest, since functional activity of the promoter and 5-HT2A receptor activation have been reported to be lower for the G allele and higher for the A allele (Shimizu et al. 2003; Parsons et al. 2004). A significantly higher frequency of both the AA genotype and the A allele of the -1438G/A polymorphism was found in AN by some studies (Collier et al. 1997; Collier 1999; Sorbi et al. 1998; Enoch et al. 1998; Nacmias et al. 1999; Ricca et al. 2002, 2004), but not confirmed by others (Hinney et al. 1997b; Campbell et al. 1998; Ziegler and Görg, 1999; Ando et al. 2001a; Nishiguchi et al. 2001; Kipman et al. 2002; Gorwood et al. 2002; Rybakowski et al. 2006) (Table 76.2). Four of the positive studies, all of the same research group (Sorbi et al. 1998; Nacmias et al. 1999; Ricca et al. 2002, 2004), reported that the -1438G/A polymorphism was specifically associated to ANR subtype. The A allele has been reported to be linked with some AN-related phenotypic traits such as an older age at onset, higher levels of weight
76 The Role of Gene Polymorphisms in Susceptibility to Anorexia Nervosa and Bulimia Nervosa
1183
Table 76.3 Association studies of genes coding central regulators of feeding in bulimia nervosa Analyzed Statistical Candidate gene polymorphism significance Note References Di Bella et al. (2000); 5-HT Transporter 44 bp Del/Ins S One study reported a higher Monteleone et al. (promoter) frequency of the S allele (Di (2006a) Bella et al. 2000); the other one reported a higher frequency of the L allele (Monteleone et al. 2006a) NS Lauzurica et al. (2003); Matsushita et al. (2004) Nishiguchi et al. (2001); −1,438G/A S One study (Nishiguchi et al. 5-HT2A receptor Ricca et al. (2002, (promoter) 2001) found a significant 2004) association with the G allele instead of the A allele NS Enoch et al. (1998); Ziegler and Gorge (1999); Fuentes et al. (2004); Bruce et al. (2005) Thr25Asn NS Nacmias et al. (1999) His452Tyr NS Nacmias et al. (1999) 102T/C NS Nacmias et al. (1999) 516T/C NS Nacmias et al. (1999) 5-HT2C receptor Cys23Ser NS Nacmias et al. (1999); Burnet et al. (1999) 5-HT1Db receptor G861C NS Levitan et al. (2001, 2006) Tryptophan A218C NS The A allele was associated with Monteleone et al. (2007) hydroxylase-1 a more severe bulimic symptomatology VNTR NS Shinohara et al. (2004) Dopamine Transporter (DAT1) Dopamine D2 TaqA1 NS Nisoli et al. (2007) receptor (DRD2) b3-adrenergic Trp64Arg NS Miyasaka et al. (2006) receptor Val158Met NS Micolajczyk et al. (2006) Catechol-O(472G/A) methyltransferase (COMT) 196G/A S Ribasès et al. (2004) Brain-derived (Val66Met) neurotrophic factor (BDNF) Ribasès et al. (2003, NS In the study of Koizumi et al. 2005a); Koizumi et al. (2004) a significant (2004); Monteleone association was found in the et al. (2006b) BNNP group −270C/T NS Ribasès et al. (2003, 2004, 2005a); NS Ribasès et al. (2005b) Neurotrophic tyrosin −69C>G kinase receptor 2 (NTRK2) IVS2+40C>T NS Ribasès et al. (2005b) IVS13+40G>A S Ribasès et al. (2005b) IVS17+125T>C S Ribasès et al. (2005b) (continued)
1184 Table 76.3 (continued) Analyzed Candidate gene polymorphism
P. Monteleone and M. Maj
Statistical significance Note
References
IVS18+13G>A NS Ribasès et al. (2005b) 2,785– NS Ribasès et al. (2005b) 2,785insC This table lists published association studies of polymorphic variants of genes involved in the physiology of central regulators (neurotransmitters, neuropeptides, and neurohormones) of feeding and energy homeostasis in bulimia nervosa BNNP bulimia nervosa nonpurging subtype, 5-HT serotonin, NS not significant, S significant, VNTR variable number of tandem repeats
and shape concerns and total Eating Disorder Examination scores, lower levels of harm avoidance, and reward dependence (Kipman et al. 2002; Gorwood et al. 2002; Ricca et al. 2004; Rybakowski et al. 2006). Three studies reported a significant association of the -1438G/A polymorphism of the promoter of the 5-HT2A receptor gene with BN (Nishiguchi et al. 2001; Ricca et al. 2002, 2004), whereas five others did not (Enoch et al. 1998; Ziegler and Görge 1999; Fuentes et al. 2004; Bruce et al. 2005) (Table 76.3). One of the positive studies performed in an Asian population (Nishiguchi et al. 2001) found a nominal association between BN and the GG genotype/G allele instead of the AA genotype/A allele as observed in the other two studies (Ricca et al. 2002, 2004). Interestingly, the GG genotype and the G allele were more frequent in the group with binge-purging phenotype (ANBP and BNNP), and BN patients with GG genotype were characterized by blunted prolactin response to meta-chlorophenylpiperazine and higher levels of impulsivity (Bruce et al. 2005), which suggests a reduced serotonergic activity of the wild type allele. Therefore, the higher 5-HT2A receptor activity associated to the A allele, could result in a potentiation of the anorectic 5-HT signal explaining the above-reported association of this allele with ANR and with reduced levels of harm avoidance, whereas the G allele could sustain binge eating behavior in BN because of lower 5-HT2A receptor activity. The Thr25Asn, His452Tyr, 102T/C, and 516T/C SNPs of the 5-HT2A receptor gene were found not significantly associated with both AN and BN (Hinney et al. 1997b; Nacmias et al. 1999) (Tables 76.2 and 76.3). The Ser23 allele of the Cys23Ser SNP of the 5HT2C receptor gene is associated with a higher 5-HT2C receptor expression (Sodhi et al. 1999), and two out of three studies (Nacmias et al. 1999; Westberg et al. 2002; Hu et al. 2003) reported higher frequencies of the Ser23Ser genotype and the Ser23 allele in Caucasian AN patients (Table 76.2). Furthermore, in the study of Hu et al. (2003), including also 43 AN trios, the TDT showed a preferential transmission of the Ser23 allele, which was also associated with lower BMI. Therefore, the putative enhanced activity of 5-HT2C receptors in carriers of Ser23 allele would sustain the restrictive behavior of AN patients and could explain the lower BMI. Recently, a higher frequency of the Ser23 allele has been found also in teenage girls reporting weight loss but not affected by AN (Westberg et al. 2002), which suggests that this SNP might be not specifically associated with AN, but with a general proneness of young women to experience weight loss through reducing food intake. No association between the Cys23Ser SNP of the 5-HT2C receptor gene and BN has been detected (Nacmias et al. 1999; Burnet et al. 1999) (Table 76.3). Some SNPs and the haplotypes -1123T>C/1080C>T and 1080C>T/2190A>G of the 5-HT1Db receptor gene were found significantly associated with AN or ANR (Bergen et al. 2003; Brown et al.
76 The Role of Gene Polymorphisms in Susceptibility to Anorexia Nervosa and Bulimia Nervosa
1185
2007) (Table 76.2). These associations are intriguing, since the 5-HT1Db receptor gene is positioned under an observed linkage peak on chromosome 1 for ANR (Grice et al. 2002; Bergen et al. 2003). Therefore, the 5-HT1Db receptor gene certainly is worth of further investigation. The G861C SNP of the 5-HT1Db receptor gene has been investigated in BN patients (Levitan et al. 2001, 2006) (Table 76.3); although no significant association emerged between this SNP and BN, bulimic individuals with GG genotype had a minimum lifetime BMI significantly lower than GC and CC genotypes and a more severe comorbid obsessive-compulsive disorder. No significant association has been detected between the Pro279Leu SNP of 5-HT7 receptor gene and AN (Hinney et al. 1999a).
76.2.1.3 Serotonin Biosynthetic Enzyme The A218C SNP of the tryptophan-hydroxylase-1 (TPH-1) gene is endowed with functional effects, since the A allele has been found to be associated with lower cerebrospinal fluid 5-hydroxyindolacetic acid levels in healthy volunteers (Jonsson et al. 1997). Although no significant association was found between the A218C SNP of the TPH-1 gene and BN (Table 76.3), BN women with the AA genotype displayed a more severe bulimic symptomatology (as measured by the weekly frequency of binge-purging episodes and the Bulimia Investigation Test Edinburgh) and higher levels of harm avoidance as compared to AC and CC genotypes (Monteleone et al. 2007). Therefore, it could be hypothesized that BN individuals with the A variant of the TPH-1 A218C SNP have reduced concentrations of 5-HT at their central synapses, which would represent a vulnerability factor for binge eating behavior and higher harm avoidance. No significant association between the Y1095C SNP of the TPH-1 gene and AN was reported in a small sample study (Han et al. 1999) (Table 76.2).
76.2.2 Dopamine Dopaminergic neurotransmission modulates feeding, thinking processes, motor activity, rewardmotivated and drug-seeking behaviors. AN and BN patients, aside from their well-known disturbances of eating behavior, exhibit physical hyperactivity, distortion of thinking and body image, and obsessive–compulsive behaviors. Therefore, genes of dopamine transmission are candidate genes in EDs.
76.2.2.1 Dopamine Transporter The dopamine transporter (DAT) protein is a critical regulator of synaptic dopamine. The DAT1 gene, encoding the DAT protein, is polymorphic, and in most human beings occurs with greatest frequency in the 9- and 10-repeat forms. The 10-repeat variable number of tandem repeats (VNTR) polymorphism has been shown to be associated with an approximately 50% increase of DAT binding sites as compared to the 9-repeat allele (vanNess et al. 2005). One study investigated the VNTR polymorphism of the DAT1 gene in small groups of ANBP and BN patients and found a higher frequency of short alleles (seven and nine repeats) as compared to long alleles (10 and 11 repeats) in ANBP but not in BN (Shinohara et al. 2004) (Tables 76.2 and 76.3).
1186
P. Monteleone and M. Maj
76.2.2.2 Dopamine Receptors Bergen et al. (2005) tested seven SNPs within the dopamine D2 receptor gene (DRD2) and found significant associations of the 725bp3’G>T and 10620C>T SNPs with ANBP (Table 76.2). Moreover, the haplotype Indel+939C>T was significantly associated with both AN and ANR; the haplotypes Indel +957C>T and 939C>T+725bp3’G>T were significantly associated with AN, and the haplotype 939C>T+10520C>T was significantly associated with ANR. Another variant of the DRD2 gene is represented by the polymorphic Taq1A restriction endonuclease site that has been shown to reduce D2 receptor synthesis (Laakso et al. 2005). No association between the DRD2 Taq1A polymorphism and AN or BN was detected in a very small sample of patients, including also obese individuals (Nisoli et al. 2007) (Tables 76.2 and 76.3), although it was found significantly associated with the Eating Disorder Inventory Subitems that characterize the drive for thinness and ineffectiveness, which suggests a possible role of this polymorphism in ED-related phenotypic traits. SNPs of the dopamine D3 (DRD3) and D4 (DRD4) receptor genes have been investigated for association with AN and results were generally negative (Bruins-Slot et al. 1998; Hinney et al. 1999b; Bachner-Melman et al. 2007) (Table 76.2). A functional polymorphism of the DRD4 gene is represented by the D4prC(521)T, where the C allele has been found to be responsible for less transcriptional activity of the DRD4 gene (Okuyama et al. 2000). The recent study of Bachner-Melmam et al. (2007), who explored five SNPs of the DRD4 gene, showed that the C allele of the D4pr C(521) T SNP was preferentially transmitted to AN individuals and that the D4pr C(521)T SNP, the D4pr 120 repeat, and several two, three, four, and five locus haplotypes were significantly associated with AN (with some differences between ANR and ANBP) and with the socially proscribed perfectionism subscale and/or the self-oriented perfectionism subscale of the Children and Adolescent Perfectionism Scale in AN. These findings suggest a possible involvement of DRD4 gene in ED-related phenotypic traits such as perfectionism.
76.2.3 Noradrenaline Decreased levels of norepinephrine have been detected in the blood and cerebrospinal fluid of long-term weight-restored patients with AN (Kaye et al. 1985), which supports an involvement of noradrenergic transmission in AN. A repeat polymorphism in the promoter region of the norepinephrine transporter (NET) gene (NETpPR), characterized by a 4bp deletion (NETpPR-S4) or insertion (NETpPR-L4), has been characterized. This polymorphism results in an alteration of a potential transcription factor binding site. In a sample of 101 AN Australian trios, including 14 ANBP trios and 87 ANR trios, a higher frequency of L4/L4 genotype and L4 allele was found in ANR patients. Moreover, the TDT showed a preferential transmission of the NETpPR-L4 allele from L4/S4 heterozygous parents to their ANR children (Urwin et al. 2002). This finding was not confirmed in a subsequent study (Hu et al. 2007) (Table 76.2). In humans, the beta-3 adrenoceptor is primarily expressed in visceral fat and its activation by catecholamines promotes lipolysis. Recently, the Trp64Arg SNP of the beta-3 adrenergic receptor has been suggested to be a predisposing factor for abdominal obesity and metabolic alterations, especially in women (Kurabayashi et al. 1996); therefore, this SNP has been considered to be of interest also in EDs. However, no association between the Trp64Arg SNP of the beta-3 adrenergic receptor gene and AN or BN has been detected (Hinney et al. 1997c; Miyasaka et al. 2006) (Tables 76.2 and 76.3).
76 The Role of Gene Polymorphisms in Susceptibility to Anorexia Nervosa and Bulimia Nervosa
1187
76.2.4 Monoamine Degrading Enzymes Catechol-O-methyltransferase (COMT) and monoamine oxidase A (MAOA) are the two main enzymes involved in the degradation of brain monoamines. An SNP at position 472 (472G/A) of the COMT gene creates an amino acid change (158 Val>Met) leading to two forms of the mature protein: the 158Val protein is provided of high enzymatic activity, which is about 4 times greater than the 158Met protein (Lachman et al 1996). A nominal association of the ValVal genotype and the Val allele with AN has been shown in two studies (Frisch et al. 2001; Micolajczyk et al. 2006) but has not been confirmed in a large case-control study of individuals from six European countries (Gabrovsek et al. 2004) (Table 76.2). Similarly, a preferential transmission of the Val allele was detected in a sample of 66 ANR trios (Michaelovsky et al. 2005), but not confirmed in a larger study including 372 AN trios (Gabrovsek et al. 2004). Moreover, four SNPs of the COMT gene other than the Val158Met SNP were investigated together with two SNPs in the adjacent armadillo repeat gene deleted in velocardiofacial syndrome (ARVCF) gene, and the haplotype COMT-186C-408G-472G-ARVFC-659C-524T was identified as a “risk haplotype,” whereas the haplotype COMT-186T-408C-472A-ARVFC659T-524C was identified as a “protective haplotype” for AN (Michaekowsky et al. 2005) (Table 76.2). Moreover, the TDT showed a preferential transmission of the C allele of the 408C/G and of the T allele of the 524T/C SNPs. Finally, the Met allele of the 158 Val>Met SNP of the COMT gene was found to be associated with ED-related phenotypic traits, such as higher scores on the subscales bulimia, ineffectiveness, interoceptive awareness, maturity fears, and impulse regulation of the Eating Disorder Inventory (Frieling et al. 2006). The promoter of the MAOA gene has a functional MAOA-upstream VNTR (MAOA-uVNTR) polymorphism, which consists of 3 (3-allele) or 4 (4-allele) copies of a 30-bp sequence, or rarely 2 (2-allele) or three copies plus the first 18-bp of the same 30-bp sequence (3a-allele), or five copies (5-allele). The 3a-allele and 4-allele (MAOA-L) are transcribed more efficiently than the shorter 3-allele (MAOA-S) (Sabol et al. 1998). One study found no significant association between the MAOA-uVNTR polymorphism and AN (Urwin et al. 2003b) (Table 76.2). No study has been performed to assess specifically the association of SNPs of the COMT and MAOA genes with BN. Only Mikolajczyk et al. (2006) reported no significant association between the 472G/A SNP of the COMT gene and BN in a study including only 28 BN individuals.
76.2.5 Epistasis Epistasis, that is gene–gene interaction, between the 5-HTTLPR polymorphism and polymorphisms of the NET or the MAOA gene have been investigated in AN. No significant interaction between 5-HTTLPR and a polymorphism within the NET gene promoter polymorphic region was found in a sample of 106 Australian trios (Urwin et al. 2003a). On the contrary, the same research group detected a significant synergistic epistatic interaction between the 5-HTTLPR and a polymorphism in the MAOA gene, since the risk of developing AN was up to eight times greater that the risk imposed by the MAOA gene variant alone when the MAOA variant was transmitted together with the 5-HTTLPR SS genotype (Urwin and Nunn 2005). Finally, an epistatic interaction between the MAOA-uVNTR and the NETpPR polymorphisms was identified, since receiving a MAOA-L allele was found to more than double the risk for developing ANR, conditional on an individual also being a NETpPR-L4 homozygote (Urwin et al. 2003b).
1188
P. Monteleone and M. Maj
76.2.6 Brain-Derived Neurotrophic Factor It has been widely demonstrated that BDNF has a role in the regulation of eating behavior, acting as a satiety factor in the experimental animal (Tsuchida et al. 2001). Moreover, in symptomatic patients with AN or BN, circulating levels of BDNF have been detected to be decreased (Monteleone et al. 2004, 2005b) and only partially restored after body weight rehabilitation in AN (Nakazato et al. 2006). The Val66Met (196G/A) SNP of the BDNF gene, which has an effect on the intracellular processing and secretion of BDNF, has been assessed for association with AN yielding somewhat conflicting, yet generally, promising results. Indeed, a small study and a large case-control study of individuals from five European countries showed that ANR and ANBP individuals had a higher frequency of the AA genotype and the A allele of the Val66Met (196G/A) SNP of the BDNF gene (Ribasès et al. 2003, 2004), and that patients carrying at least one copy of the A allele had lower values of lifetime minimum BMI (Table 76.2). In a subsequent study (Ribasès et al. 2005a) including 359 family trios recruited from seven European countries, the previously reported higher frequency of the AA/AG genotypes and the A allele in ANR was confirmed, but no preferential transmission of the A or G allele was detected, although in trios with ANR the A allele was transmitted with a significant lower minimum BMI than the G allele. An association between the Val66Met SNP and ANR or AN was reported by two further independent research groups (Koizumi et al. 2004; DmitrzakWeglarz et al. 2007). Three studies, instead, provided negative results (Friedel et al. 2005; Rybakowski et al. 2007; Dardennes et al. 2007); however, in one of them AN patients with the A allele of the Val66Met SNP had higher levels of harm avoidance than GG homozygotes (Rybakowski et al. 2007), suggesting an association between this SNP and personality traits in AN (Table 76.2). As for BN, Ribasès et al. (2004) found a higher frequency of the AA/AG genotype and the A allele of the Val66Met SNP in 389 BN individuals recruited from three European countries, but this was not confirmed in their subsequent family trios study (Ribasès et al. 2005a) and in a small study including only 70 BN patients (Ribasès et al. 2003). Negative findings were also reported by Koizumi et al. (2004) in a sample of 118 BN patients; however, when patients were stratified into BNP and BNNP subtypes, a significantly higher frequency of the AG genotype was detected in BNP patients while the A allele was significantly associated with BNNP subtype (Table 76.3). Monteleone et al. (2006b) assessed the association of the Val66Met SNP with both BN and binge eating disorder, and results were negative; however, patients carrying the AA genotype displayed a higher severity of binge-eating (assessed by the weekly frequency of bingeing and the Bulimia Investigation Test Edinburgh symptom and total scores), which suggests a possible involvement of this SNP in the susceptibility to the aberrant eating behavior. A second common SNP of the BDNF gene is represented by the -270C/T SNP. No significant association between the -270 C/T SNP and AN was founded in five studies (Ribasès et al. 2003, 2004, 2005a; Koizumi et al. 2004; Dardennes et al. 2007) (Table 76.2). However, in a study including 359 AN trios recruited from seven European countries and investigating also the Val66Met SNP, it was found that the -270C/T/Met66 haplotype was preferentially transmitted to the affected ANR offspring (Ribasès et al. 2005a) while Rybakowski et al. (2007) reported that AN patients with the T allele had higher levels of persistence and harm avoidance than CC homozygotes. No significant association of the -270C/T SNP with BN has been detected (Ribasès et al. 2003, 2004, 2005a); however, BN individuals carrying the T allele exhibited an earlier age at onset of weight loss and higher maximum BMI (Ribasès et al. 2004) (Table 76.3). Very recently, epistasis between the BDNF gene and the DRD4 gene has been assessed in a sample of 162 female probands with BN (Kaplan et al. 2008). Probands carrying both the hypofunctional 7R allele of DRD4 gene and the Met allele of BDNF gene had significantly higher maximum BMI than probands in the other gene–gene interaction groups.
76 The Role of Gene Polymorphisms in Susceptibility to Anorexia Nervosa and Bulimia Nervosa
1189
76.2.6.1 BDNF Receptor SNPs of the NTRK2 gene, which encodes a BDNF receptor, have been investigated for association with AN and BN in a study that screened the entire NTRK2 gene and identified 14 SNPS (Ribasès et al. 2005b). The -69C >G SNP was found nominally associated with ANBP and the IVS13+40G>A and the IVS17+125T>C SNPs were found significantly associated with BN, whereas no significant association emerged between the IVS2+40C>T, IVS18+13G>A and 2784–2785insC with both AN and BN (Tables 76.2 and 76.3). Moreover, a strong association was detected between the C-A-insC haplotype and ANBP. Finally, BN patients carrying the C-A-insC haplotype showed higher scores in the harm avoidance dimension of the personality. These results need replication.
76.2.7 Neuropeptide Y and the Agouti-Related Protein NPY is a highly potent stimulator of hunger in the hypothalamus, where it is co-secreted with another potent orexigenic peptide, the agouti-related protein, which is an inverse agonist at melanocortinergic receptors where melanocortins derived from proopiomelanocortin act as anorexigenic agonists. NPY exerts its effects in the regulation of food ingestion mainly through NPY 1 and NPY 5 receptors. No significant associations between SNPs of NPY Y1 and Y5 receptor genes and AN were detected in one study (Rosenkranz et al. 1998a), whereas the G760A and G526A, but not the C659T SNP of the agouti-related protein gene were found nominally associated to AN (Vink et al. 2001) (Table 76.2). The 760A and 526A alleles were in complete linkage disequilibrium and the mutant allele was preferentially transmitted to ANBP offspring in a family trios study (Dardennes et al. 2007). Finally, it was found that AN individuals carrying the mutant allele had a younger age at onset and lower sense of interpersonal distrust (Dardennes et al. 2007).
76.2.8 Opioids and Endocannabinoids Opioid peptides and endocannabinoids have been demonstrated to be involved in the control of both quantitative and hedonic/rewarding aspects of food choice and consumption. In particular, endocannabinoids, such as anandamide and 2-arachidonylglycerol, control food intake at two levels. First, they tonically reinforce the motivation to find and consume food with a high incentive value, possibly by interacting with the mesolimbic pathway involved in reward mechanisms. Second, they are activated “on demand” in the hypothalamus after short-term food deprivation and transiently regulate the levels and/or action of other orexigenic and anorectic mediators to modulate appetite (Matias and Di Marzo 2007). Alterations in circulating levels of anadamide and opioid peptides have been detected in ED patients (Brambilla et al. 1991; Monteleone et al. 2005c); so the investigation of genes involved in the physiology of those substances seems to be promising in EDs. Bergen et al. (2003) found that the 47821A>G SNP and the haplotypes 8214T>C/47821A>G and 80A>G/8214T>C/23340A>G/47821T>G/51502A>T of the OPRD1 gene were associated to AN. Recently, Brown et al. (2007) found three other SNPs of the OPRD1 gene to be associated with both ANR and/or ANBP (Table 76.2). It is of note that the OPRD1 gene is positioned under the observed linkage peak on chromosome 1 for ANR close to the 5HTR1D gene (Grice et al. 2002; Bergen et al. 2003).
1190
P. Monteleone and M. Maj
Polymorphisms of the endocannabinoid CB1 receptor gene (CNR1) and of the gene coding the fatty acid amide hydrolase (FAAH) and the N-acylethanolamine-hydrolyzing acid amidase (NAAA), the major degrading enzymes of endogenous cannabinoids, have been studied for association with AN (Table 76.2). Specifically, it was demonstrated that the 13-repeat allele of the (AAT)n triplet repeat polymorphism of the CNR1 gene was preferentially transmitted in ANBP patients while the 14-repeat allele was preferentially transmitted in the ANR group (Siegfried et al. 2004), but this was not confirmed in a larger study showing also no significant association or transmission of different SNPs of CNR1, FAAH, and NAAA genes with AN in 91 German AN trios (Muller et al. 2008).
76.3 Peripheral Regulators of Eating Behavior A number of substances secreted by the gut, the pancreas, and the adipose tissue have been identified and characterized as modulators of food intake and/or energy expenditure. Alterations in circulating levels of those substances and/or change in their secretory responses to physiological stimuli have been detected in both AN and BN patients (Monteleone et al. 2008). At the moment, there are no conclusive data as to whether secretory alterations of feeding regulatory substances precede the appearance of an ED or are the consequence of the nutritional changes occurring with the disorder. However, it has been suggested, although not proved, that even if those alterations are secondary phenomena disappearing after the recovery from the ED, they may hypothetically contribute to the maintenance of both aberrant eating behaviors and/or other symptomatic changes, thus affecting the course and the outcome of the ED. Therefore, genetic association studies have been performed to evaluate the role of polymorphic variants of genes coding ghrelin, cholecystokinin, leptin, and their receptors, as well as adiponectin and resistin in the genetic susceptibility to AN and/or BN. No significant associations between the Arg51Gln and/or the Gln90Leu SNPs of the ghrelin gene with AN have been reported in four independent studies (Ando et al. 2006; Cellini et al. 2006; Dardennes et al. 2007; Monteleone et al. 2006c) (Table 76.4); the Leu72Met SNP, instead, was found associated with ANBP in a family trios study reporting a preferential transmission of the Met allele and of the haplotype 90Gln/72Met to ANBP offspring (Dardennes et al. 2007). However, two casecontrol studies and a large family trios study did not confirm those results (Ando et al. 2006; Cellini et al. 2006; Monteleone et al. 2006c) (Table 76.4). No significant association was found between the Arg51Gln and the Gln90Leu SNPs of the ghrelin gene and BN (Cellini et al. 2006; Monteleone et al. 2006c), although the haplothype Gln90/Leu72/Arg51 was detected preferentially transmitted to BN offspring (Cellini et al. 2006). Similar negative findings were reported for the Leu72Arg SNP in two different BN samples (Cellini et al. 2006; Monteleone et al. 2006c) (Table 76.5). Ando et al. (2006), instead, reported a nominal association of the Leu72Met and 3056T>C SNPs with BNP and a higher frequency in this group of the haplotype Met72/3056C. Finally, the 171T/C SNP of the ghrelin receptor gene was found associated with BN, but not with AN in a Japanese population (Miyasaka et al. 2006). Only the rs11129946 SNP of the cholecystokinin gene was found nominally associated with AN in a single study (de Krom et al. 2006) (Table 76.4). No associations were found between SNPs of the cholecystokinin-A receptor, leptin, leptin receptor, adiponectin, and resistin genes with AN and/ or BN in single small case–control studies (Hinney et. 1998; Quinton et al. 2004; Dolinkova et al. 2006; Miyasaka et al. 2006) (Tables 76.3 and 76.4). The only interesting finding that emerged from those studies was that AN patients carrying the minor T allele of the 276G>T SNP of the adiponectin gene had increased cholesterol levels (Dolinkova et al. 2006), which might help to explain hypercholesterolemia of underweight AN patients.
76 The Role of Gene Polymorphisms in Susceptibility to Anorexia Nervosa and Bulimia Nervosa
1191
Table 76.4 Association studies of genes coding peripheral regulators of feeding in anorexia nervosa Analyzed Statistical Candidate gene polymorphism Significance Note References Ghrelin Arg51Gln NS Ando et al. (2006); Cellini et al. (2006); Monteleone et al. (2006b) Gln90Leu NS Ando et al. (2006); Cellini et al. (2006); Dardennes et al. (2007) Dardennes et al. (2007) Leu72Met S Significantly associated to ANBP NS Ando et al. (2006); Cellini et al. (2006); Monteleone et al. (2006b) 3,056T>C NS Ando et al. (2006) 3,083A>G NS Ando et al. (2006) 3,615A>C NS Ando et al. (2006) 171T/C NS Miyasaka et al. (2006) Growth hormone secretagogue receptor (GHSR, ghrelin Receptor) Cholecystokinin rs11129946 S De Krom et al. (2006) (CCK) rs6791019 NS De Krom et al. (2006) rs7611677 NS De Krom et al. (2006) rs6809785 NS De Krom et al. (2006) rs6801844 NS De Krom et al. (2006) CCK-A Receptor −81A>G NS Miyasaka et al. (2006) −128G>T NS Miyasaka et al. (2006) Leptin −1,387 G/A NS Hinney et al. (1998) Leptin receptor Gln223Arg NS Quinton et al. (2004) Lys109Arg NS Quinton et al. (2004) Lys656Asn NS Quinton et al. (2004) Adiponectin 45T>G NS Dolinkovà et al. (2006) 276G>T NS Dolinkovà et al. (2006) Resistin 62G>A NS Dolinkovà et al. (2006) 180C>G NS Dolinkovà et al. (2006) This table lists published association studies of polymorphic variants of genes involved in the physiology of peripheral regulators (peptides and hormones) of feeding and energy homeostasis in anorexia nervosa ANBP anorexia nervosa binge-purging subtype, NS not significant; S significant
76.4 Other Candidate Genes Other genes that have been supposed to be likely involved in the biological vulnerability to EDs are those coding uncoupling proteins (UCP), tumor necrosis factor-a, estrogen receptors, phospholipase A2, the small-conductance calcium-activated potassium channel-3 (KCNN3), and the circadian locomotor output cycles kaput (CLOCK). UCP2 and UCP3 are ubiquitous proteins promoting thermogenesis and modulating metabolic adaptation during fasting. Allele 13 of the microsatellite marker D11S911 of the UCP2/UCP3 gene was found associated with AN in one study (Campbell et al. 1999), but this was not confirmed by another research group (Hu et al. 2002). One study performed in an Asian population reported no
1192
P. Monteleone and M. Maj
Table 76.5 Association studies of genes coding peripheral regulators of feeding in bulimia nervosa Analyzed Statistical Candidate gene polymorphism significance Note References Ghrelin Arg51Gln NS Cellini et al. (2006); Ando et al. (2006); Monteleone et al. (2006b) Gln90Leu NS Ando et al. (2006); Cellini et al. (2006) Ando et al. (2006) Leu72Met S Significant association with BNP NS Cellini et al. (2006); Monteleone et al. (2006b) Ando et al. (2006) 3056T>C S Significant association with BNP 3,083A>G NS Ando et al. (2006) 3,615A>C NS Ando et al. (2006) 171T/C S Miyasaka et al. (2006) Growth hormone secretagogue receptor (GHSR, ghrelin receptor) CCK-A receptor −81A>G NS Miyasaka et al. (2006) −128G>T NS Miyasaka et al. (2006) Leptin −1,387 G/A NS Hinney et al. (1998) This table lists published association studies of polymorphic variants of genes involved in the physiology of peripheral regulators (peptides and hormones) of feeding and energy homeostasis in bulimia nervosa BNP bulimia nervosa purging subtype, CCK colecystokinin, NS not significant, S significant
association between the -866G>A SNPs of the UCP2 gene and the -55cT SNP of the UCP3 gene with AN (Ando et al. 2004) (Table 76.6). Tumor necrosis factor-a is a cytokine that decreases the activity of lipogenic enzymes including the intracellular phospholipase A2 enzyme and induces lipolysis-promoting cachexia. Four SNPs of the tumor necrosis factor-a gene and the intPla polymorphism of the Phospholipase A2 gene were found not associated with AN in two case–control studies (Ando et al. 2001b; Slopien et al. 2004) (Table 76.6). EDs occur predominantly in women with a female to male ratio of 9:1. This female predominance suggests a role for sex hormones, especially estrogens, in the etiopathogenesis of AN and BN. The genes coding estrogen receptor type 1 (ESR1) and estrogen receptor type 2 (ESR2) have been analyzed for their putative association with AN and BN. The 1082G>A SNP but not the 1730A>G SNP of the ESR2 gene has been found nominally associated with AN in two independent case-control studies (Rosenkranz et al. 1998; Eastwood et al. 2002) (Table 76.6). The 1730A>G SNP of the ESR2 gene, instead, has been found nominally associated with BN in one study (Nilsson et al. 2004), but not in another one (Rosenkranz et al. 1998) (Table 76.7). Finally, a statistically significant higher frequency of the mutant A allele of the ERbcx+56G>A polymorphism of the ESR2 gene has been detected in a small sample of BN individuals including also patients with partial syndrome BN (Nilsson et al. 2004). No association of polymorphisms of the ESR1 gene with AN was detected in a single case–control study (Eastwood et al. 2002). The KCNN3 regulates ion flow through the NMDA-glutamate receptor and dampens neuronal excitability. An association between the CAG repeat polymorphism of the KCNN3 gene and AN has
76 The Role of Gene Polymorphisms in Susceptibility to Anorexia Nervosa and Bulimia Nervosa Table 76.6 Association studies of other candidate genes in anorexia nervosa Analyzed Statistical Candidate gene polymorphism significance Note D11S911 S Allele 13 of D11S911 Uncoupling protein microsatellite marker 2, 3 (UCP2/ was significantly over UCP-3) represented in AN NS D11S916 NS Tumor necrosis factor-a (TNFa) Phospholipase A2 Estrogen receptor-1 (ESR1) Estrogen receptor-2 (ESR2 or ESR-b) Calcium-activated potassium channel (KCNN3) Circadian Locomotor Output Cycles Kaput (CLOCK)
1193
References Campbell et al. (1999)
NS
Hu et al. (2002) Campbell et al. (1999); Hu et al. (2002) Ando et al. (2004)
NS
Ando et al. (2004)
NS
Ando et al. (2001b)
−863C>A −857C>T 308G/A intPLA2 ESR1-PvuI
NS NS NS NS NS
Ando et al. (2001b) Ando et al. (2001b) Slopien et al. (2004) Slopien et al. (2004) Eastwood et al. (2002)
ESR1-XbaI Dinucleotide repeat 1,082G>A
NS NS
Eastwood et al. (2002) Eastwood et al. (2002)
S
Rosenkranz et al. (1998); Eastwood et al. (2002)
1,730A>G
NS
Dinucleotide repeat CAG repeat
NS
Rosenkranz et al. (1998); Eastwood et al. (2002) Eastwood et al. (2002)
S
Alleles longer than 19 repeats were more frequent in AN
−866G/A (UCP-2) −55C/T (UCP-3) −1,031T>C
Koronyo-Hamaoui et al. (2002, 2004)
Tortorella et al. (2007) AN subjects with at least one copy of the C allele exhibited a minimum past BW significantly lower than those with T/T genotype This table lists published association studies of polymorphic variants of genes involved in the physiology of other regulators (peptides and hormones) of feeding and energy homeostasis in anorexia nervosa AN anorexia nervosa, BW body weight, NS not significant, S significant 3,111T/C
NS
been recently investigated in two studies by the same group (Koronyo-Hamaoui et al. 2002, 2004) (Table 76.6). Patients with AN were more likely than controls to manifest genotypes with alleles longer than 19 CAG repeats (L alleles). The TDT showed a preferential transmission of the L alleles to AN offspring. Moreover, AN patients with comorbid obsessive-compulsive disorder had a higher frequency of L repeats than those without an obsessive–compulsive disorder (Koronyo-Hamaoui et al. 2004). Feeding is subjected to circadian regulation; therefore, changes in the components of the endogenous oscillator regulating circadian rhythms may be involved in disordered rhythmicity of eating
1194
P. Monteleone and M. Maj
Table 76.7 Association studies of other candidate genes in bulimia nervosa Analyzed Statistical Candidate gene polymorphism significance Note References Estrogen receptor-2 1,082G>A NS Rosenkranz et al. (1998); (ESR2 or ESR-b) Nilsson et al. (2004) 1,730A>G S Nilsson et al. (2004) NS Rosenkranz et al. (1998); ERb cx+56G>A S Nilsson et al. (2004) Tortorella et al. (2007) NS BN subjects with at least Circadian Locomotor 3,111T/C one copy of the C allele Output Cycles exhibited a minimum Kaput (CLOCK) past BW significantly lower than those with T/T genotype This table lists published association studies of polymorphic variants of genes involved in the physiology of other regulators (peptides and hormones) of feeding and energy homeostasis in bulimia nervosa BN bulimia nervosa, BW body weight, NS not significant, S significant
behavior as it occurs in EDs. Although several genes have been identified in the endogenous machinery modulating circadian rhythms, to date only the CLOCK gene has been investigated in EDs. No significant associations of the 3111T/C SNP of the CLOCK gene with AN or BN has emerged (Table 76.7); however, AN and BN subjects with at least one copy of the C allele exhibited a minimum past BW significantly lower than those with the T/T genotype, which suggests a possible involvement of the CLOCK gene in the regulation of BW (Tortorella et al. 2007).
76.5 Concluding Remarks It is evident from the above that genetic association studies of AN and BN are in an early phase. Although in the last 10 years, polymorphic variants of different candidate genes have been assessed for an association with AN and BN, results have been often inconsistent. There are several reasons that may explain at least part of such an inconsistency. First of all, the majority of association studies has been performed on small subject samples (often in groups of less than 100 subjects) and suffered from insufficient statistical power and lack of correction for multiple testing. Second, differences in the ethnicity of populations included and the use of different criteria to diagnose AN and BN may further contribute to the discrepancies among the studies. Third, genetic heterogeneity and population stratification may provide false positive results. Furthermore, EDs have a high rate of comorbidity with other psychiatric conditions, including affective, anxiety, and personality disorders, which may further contribute to the clinical heterogeneity of studied samples and may partially account for current discrepant results. Finally, the effects of environmental risk factors on gene expression have been so far completely neglected in genetic association studies on EDs. Life events, nutrition, cultural, and social risk factors by acting through epigenetic mechanisms may influence the activation and/or deactivation of genes and modify the risk of developing an ED. As outlined by Moffitt et al. (2005), discrepancies among association studies may be the result of still unknown gene x environmental interactions, in which genes have a susceptibility role for an ED only in those patients who are exposed to those putative environmental risk factors. Nonetheless, some intriguing conclusions can be drawn across association studies in AN and BN. The -1438G/A SNP of the 5-HT2A receptor gene and the Val66Met SNP of the BDNF gene have been
76 The Role of Gene Polymorphisms in Susceptibility to Anorexia Nervosa and Bulimia Nervosa
1195
found quite consistently although not specifically linked to ANR and/or to phenotypic traits associated with the disorder in large sample studies. Therefore, the 5-HT2A receptor gene and the BDNF gene are promising candidates for genetic influences on AN. The second intriguing finding is that in both AN and BN SNPs have been found frequently associated with ED-related phenotypic traits rather that to the full syndromes as currently categorized in DSM-IV. Although these results need replication and confirmation in larger studies, they underline the importance of focusing on more homogeneous subgroups, either relying on specific ED traits or identifying endophenotypes. For instance, six quantitative phenotypic traits (obsessionality, age at menarche, anxiety, lifetime minimum BMI, concern over mistake, and food-related obsessions) have been characterized as specific traits to be used for both linkage analyses and association studies (Bulik et al. 2005), and future researches should consider them. An endophenotype, instead, is a measurable trait that may be physical (neurophysiological, biochemical, neuroanatomical), cognitive, or neuropsychological and that is associated with the related illness, is heritable, and primarily state-independent. The identification of endophenotypes in EDs will help to identify more homogenous subgroups of patients in order to reduce the potential obscuring effects of focusing on currently categorized complex syndromes. It is clear that the genetics of EDs needs more and larger association studies with adequate sample sizes. To pursue this goal, multicenter collaborative studies should be encouraged. Moreover, a promising area for future research is to examine gene–gene interactions in light of the fact that biological susceptibility to EDs has a multigenic nature. Epistatic studies have produced some interesting findings, but the results are still too preliminary to draw any significant conclusion. In summary, genetic association studies of EDs have produced a plethora of data but little conclusive knowledge. We are hopeful that, in the future, a more homogeneous characterization of clinical phenotypes and the promotion of multicenter collaborative studies will help to identify the genes likely involved in the heritable transmission of biological vulnerability to these serious and debilitating conditions.
76.6 Applications to Other Area of Health and Disease Anorexia nervosa and bulimia nervosa are psychiatric disorders with physical consequences that often may threat the survival of the patients. The etiology of these conditions is not know, but it is commonly believed that a genetic predisposition to the disorders is transmitted to vulnerable individuals and may interact with personality and environmental risk factors to generate anorexia or bulimia nervosa. Some phenotypic aspects, such as low body weight, bingeing behavior and personality dimensions, seems to be more strictly associated with polymorphic gene variants. Moreover, preliminary evidence has been provided that polymorphisms of certain candidate genes may predict the outcome to treatments. Putting together genetic research and clinical intervention studies will help clinicians to improve the outcome of anorexia and bulimia nervosa and to organize adequate prevention. Summary Points • Epidemiological, family and molecular genetic studies suggest a strong genetic component in anorexia nervosa and bulimia nervosa. • Association studies of candidate genes indicate the 5-HT2A receptor gene and the BDNF gene as promising candidates for genetic susceptibility to AN. • Association studies of candidate genes have provided inconsistent results in BN.
1196
P. Monteleone and M. Maj
• The following reasons may explain at least part of discrepancies among the studies: a) the majority of studies has been performed on small subject samples and suffered from insufficient statistical power and lack of correction for multiple testing, b) different ethnic populations have been studied; c) different criteria to diagnose AN and BN have been used; d) genetic heterogeneity and population stratification could provide false positive results; e) comorbidity of AN and BN with other psychiatric conditions has been not taken into account; f) the interaction gene- environment has been completely neglected. • Several intriguing associations have emerged between polymorphisms of candidate genes and eating disorder-related phenotypic traits. • In the future, the identification of endophenotypes and the promotion of multicenter collaborative studies will lead to more consistent results.
Key Terms Binge eating: An episode of binge eating is characterized by both of the following: o Eating, in a fixed period of time, an amount of food that is definitely larger than most people would eat under similar circumstances. o A lack of control over eating during the episode: a feeling that one cannot stop eating or control what or how much one is eating. Candidate gene: gene coding a protein that is likely involved in the etiopathogenesis of a disease as supported by clinical, biochemical and other research data. Compensatory behaviors: acts by which the person tries to compensate for the effects of overeating. Examples of such acts are purging (induced vomiting or laxative abuse), fasting, and heavy exercising. Eating-disorder related phenotypic traits: clinical symptoms or personality characteristics that are typical of patients with anorexia nervosa or bulimia nervosa. Epistasis: is the interaction between genes. Epistasis takes place when the effects of one gene are modified by one or several other genes. Single-nucleotide polymorphism: is a DNA sequence variation occurring when a single nucleotide in the genome differs between members of a species (or between paired chromosomes in an individual). In this case two alleles are identified: the wild type allele and the polymorphic (or mutant) allele.
References American Psychiatric Association. Diagnostic and statistical manual of mental disorders. Washington DC: American Psychiatric Press; 1990. Ando T, Ishikawa T, Kawamura N, Karibe M, Oba M, Tatsuta N, Hara S, Takii M, Naruo T, Takei M, Kurokawa N, Nozoe S, Kubo C, Komaki G. Psychiatr Genet. 2001a;11:161–4. Ando T, Komaki G, Karibe M, Kawamura N, Hara S, Takii M, Naruo T, Kurokawa N, Takei M, Tatsuta N, Ohba M, Nozoe S, Kubo C, Ishikawa T. Psychiatr Genet. 2001b;11:157–60. Ando T, Kodama N, Ishikawa T, Naruo T, Tachikawa N, Nozaki T, Okabe K, Takeuchi K, Masuda A, Kawamura N, Komaki G. Psychiatr Genet. 2004;14:215–8. Ando T, Komaki G, Naruo T, Okabe K, Takii M, Kawai K, Konjiki F, Takei M, Oka T, Takeuchi K, Masuda A, Ozaki N, Suematsu H, Denda K, Kurokawa N, Itakura K, Yamaguchi C, Kono M, Suzuki T, Nakai Y, Nishizono-Maher A, Koide M, Murakami K, Nagamine K, Tomita Y, Ookuma K, Tomita K, Tonai E, Ooshima A, Ishikawa T, Ichimaru Y. Am J Med Genet B Neuropsychiatr Genet. 2006;141:929–34.
76 The Role of Gene Polymorphisms in Susceptibility to Anorexia Nervosa and Bulimia Nervosa
1197
Bachner-Melman R, Lerer E, Zohar AH, Kremer I, Elizur Y, Nemanov L, Golan M, Blank S, Gritsenko I, Ebstein RP. Am J Med Genet B Neuropsychiatr Genet. 2007;144:748–56. Bergen AW, van den Bree MB, Yeager M, Welch R, Ganjei JK, Haque K, Bacanu S, Berrettini WH, Grice DE, Goldman D, Bulik CM, Klump K, Fichter M, Halmi K, Kaplan A, Strober M, Treasure J, Woodside B, Kaye WH. Mol Psychiatry. 2003;8:397–406. Bergen AW, Yeager M, Welch RA, Haque K, Ganjei JK, van den Bree MB, Mazzanti C, Nardi I, Fichter MM, Halmi KA, Kaplan AS, Strober M, Treasure J, Woodside DB, Bulik CM, Bacanu SA, Devlin B, Berrettini WH, Goldman D, Kaye WH. Neuropsychopharmacology. 2005;30:1703–10. Brambilla F, Ferrari E, Petraglia F, Facchinetti F, Catalano M, Genazzani AR. Psychiatry Res. 1991;39:115–27. Brown KM, Bujac SR, Mann ET, Campbell DA, Stubbins MJ, Blundell JE. Biol Psychiatry. 2007;61:367–73. Bruce KR, Steiger H, Joober R, Ng Ying Kin NMK, Israel M, Young SN. Am J Med Genet B Neuropsychiatr Genet. 2005;137:40–4. Bruins-Slot L, Gorwood P, Bouvard M, Blot P, Adès J, Feingold J, Schwartz JC, Mouren-Siméoni MC (1998) Biol. Psychiatry 43: 76–78. Bulik CM, Bacanu SA, Klump KL, Fichter MM, Halmi KA, Keel P, Kaplan AS, Mitchell JE, Rotondo A, Strober M, Treasure J, Woodside DB, Sonpar VA, Xie W, Bergen AW, Berrettini WH, Kaye WH, Devlin B. Am J Med Genet B Neuropsychiatr Genet. 2005;139:81–7. Burnet PWJ, Smith KA, Cowen PJ, Fairburn CG, Harrison PJ. Psychiatr Genet. 1999;9:101–4. Campbell DA, Sundaramurthy D, Markham AF, Pieri LF. Lancet. 1998;351:499. Campbell DA, Sundaramurthy D, Gordon D, Markham AF, Pieri LF. Mol Psychiatry. 1999;4:68–70. Cellini E, Nacmias B, Brecelj-Anderluh M, Badía-Casanovas A, Bellodi L, Boni C, Di Bella D, Estivill X, FernandezAranda F, Foulon C, Friedel S, Gabrovsek M, Gorwood P, Gratacos M, Guelfi J, Hebebrand J, Hinney A, Holliday J, Hu X, Karwautz A, Kipman A, Komel R, Rotella CM, Ribases M, Ricca V, Romo L, Tomori M, Treasure J, Wagner G, Collier DA, Sorbi S; EC Framework V ‘Factors in Healthy Eating’ consortium. Psychiatr Genet. 2006;16:51–2. Cloninger CR, Svrakic DM, Przbeck TR. Arch Gen Psychiatry. 1993;50:975–90. Collier DA, Arranz MJ, Li T, Mupita D, Brown N, Treasure J. Lancet. 1997;350:412. Collier DA, on behalf of all authors. Authors’ reply. Lancet. 1999;353:929. Dardennes RM, Zizzari P, Tolle V, Foulon C, Kipman A, Romo L, Iancu-Gontard D, Boni C, Sinet PM, Thérèse Bluet M, Estour B, Mouren MC, Guelfi JD, Rouillon F, Gorwood P, Epelbaum J. Psychoneuroendocrinology. 2007;32:106–13. de Krom M, Hendriks J, Hillebrand J, van Elburg A, Adan R. Psychiatr Genet. 2006;16:239. Di Bella DD, Catalano M, Cavallini MC, Riboldi C, Bellodi L. Mol Psychiatry. 2000;5:233–4. Dmitrzak-Weglarz M, Skibinska M, Slopien A, Szczepankiewicz A, Rybakowski F, Kramer L, Hauser J, Rajewski A. Psychiatr Genet. 2007;17:245–6. Dolinková M, Krízová J, Lacinová Z, Dolezalová R, Housová J, Krajícková J, Bosanská L, Papezová H, Haluzík M. Cas Lek Cesk. 2006;145:562–6. Eastwood H, Brown KM, Markovic D, Pieri LF. Mol Psychiatry. 2002;7:86–9. Enoch MA, Kaye WH, Rotondo A, Greenberg BD, Murphy DL, Goldman D. Lancet. 1998;351:1785–6. Erzegovesi S, Riboldi C, Di Bella D, Di Molfetta D, Mapelli F, Negri B, Tenace M, Bellodi L. J Clin Psychopharmacol. 2004;24:680–2. Friedel S, Horro FF, Wermter AK, Geller F, Dempfle A, Reichwald K, Smidt J, Brönner G, Konrad K, HerpertzDahlmann B, Warnke A, Hemminger U, Linder M, Kiefl H, Goldschmidt HP, Siegfried W, Remschmidt H, Hinney A, Hebebrand J. Am J Med Genet B Neuropsychiatr Genet. 2005;132:96–9. Frieling H, Römer KD, Wilhelm J, Hillemacher T, Kornhuber J, de Zwaan M, Jacoby GE, Bleich S. Psychiatr Genet. 2006;16:205–8. Frisch A, Laufer N, Danziger Y, Michaelovsky E, Leor S, Carel C, Stein D, Fenig S, Mimouni M, Apter A, Weizman A. Mol Psychiatry. 2001;6:243–5. Fuentes JA, Lauzurica N, Hurtado A, Escartí A, Barrios V, Morandé G, Soriano J, Jáuregui I, González-Valdemoro MI, García-Camba E. Psychiatr Genet. 2004;14:107–9. Fumeron F, Betoulle D, Aubert R, Herbeth B, Siest G, Rigaud D. Mol Psychiatry. 2001;6:9–10. Gabrovsek M, Brecelj-Anderluh M, Bellodi L, Cellini E, Di Bella D, Estivill X, Fernandez-Aranda F, Freeman B, Geller F, Gratacos M, Haigh R, Hebebrand J, Hinney A, Holliday J, Hu X, Karwautz A, Nacmias B, Ribases M, Remschmidt H, Komel R, Sorbi S, Tomori M, Treasure J, Wagner G, Zhao J, Collier DA. Am J Med Genet B Neuropsychiatr Genet. 2004;124:68–72. Gorwood P, Adès J, Bellodi L, Cellini E, Collier DA, Di Bella D, Di Bernardo M, Estivill X, Fernandez-Aranda F, Gratacos M, Hebebrand J, Hinney A, Hu X, Karwautz A, Kipman A, Mouren-Siméoni MC, Nacmias B, Ribasés M, Remschmidt H, Ricca V, Rotella CM, Sorbi S, Treasure J; EC Framework V ‘Factors in Healthy Eating’ consortium. Mol Psychiatry. 2002;7:90–4.
1198
P. Monteleone and M. Maj
Gorwood P. Am J Pharmacogenomics. 2004;4:9–17. Greenberg BD, Tolliver TJ, Huang SJ, Li Q, Bengel D, Murphy DL. Am J Med Genet. 1999;88:83–7. Grice DE, Halmi KA, Fichter MM, Strober M, Woodside DB, Treasure JT, Kaplan AS, Magistretti PJ, Goldman D, Bulik CM, Kaye WH, Berrettini WH. Am J Hum Genet. 2002;70:787–92. Han L, Nielsen DA, Rosenthal NE, Jefferson K, Kaye W, Murphy D, Altemus M, Humphries J, Cassano G, Rotondo A, Virkkunen M, Linnoila M, Goldman D. Biol Psychiatry. 1999;45:615–9. Heils A, Teufel A, Petri S, Stöber G, Riederer P, Bengel D, Lesch KP. J Neurochem. 1999;66:2621–4. Hinney A, Barth N, Ziegler A, von Prittwitz S, Hamann A, Hennighausen K, Pirke KM, Heils A, Rosenkranz K, Roth H, Coners H, Mayer H, Herzog W, Siegfried A, Lehmkuhl G, Poustka F, Schmidt MH, Schäfer H, Grzeschik KH, Lesch KP, Lentes KU, Remschmidt H, Hebebrand. J Life Sci. 1997a;61:295–303. Hinney A, Ziegler A, Nöthen MM, Remschmidt H, Hebebrand J. Lancet. 1997b;350:1324–5. Hinney A, Lentes KU, Rosenkranz K, Barth N, Roth H, Ziegler A, Hennighausen K, Coners H, Wurmser H, Jacob K, Römer G, Winnikes U, Mayer H, Herzog W, Lehmkuhl G, Poustka F, Schmidt MH, Blum WF, Pirke KM, Schäfer H, Grzeschik KH, Remschmidt H, Hebebrand J. Int J Obes Relat Metab Disord. 1997c;21:224–30. Hinney A, Bornscheuer A, Depenbusch M, Mierke B, Tölle A, Middeke K, Ziegler A, Roth H, Gerber G, Zamzow K, Ballauff A, Hamann A, Mayer H, Siegfried W, Lehmkuhl G, Poustka F, Schmidt MH, Hermann H, HerpertzDahlmann BM, Fichter M, Remschmidt H, Hebebrand J. Mol Psychiatry. 1998;3:539–43. Hinney A, Herrmann H, Löhr T, Rosenkranz K, Ziegler A, Lehmkuhl G, Poustka F, Schmidt MH, Mayer H, Siegfried W, Remschmidt H, Hebebrand J. Int J Obes Relat Metab Disord. 1999a;23:760–3. Hinney A, Schneider J, Ziegler A, Lehmkuhl G, Poustka F, Schmidt MH, Mayer H, Siegfried W, Remschmidt H, Hebebrand J. Am J Med Genet. 1999b;88:594–7. Hu X, Murphy F, Karwautz A, Li T, Freeman B, Franklin D, Giotakis O, Treasure J, Collier DA. Mol Psychiatry. 2002;7:276–7. Hu X, Giotakis O, Li T, Karwautz A, Treasure J, Collier DA. Neuroreport. 2003;14:781–3. Hu X, Karwautz A, Wagner G, Holliday J, Li T, Treasure J, Collier DA. Psychiatr Genet. 2007;17:247–8. Jönsson EG, Goldman D, Spurlock G, Gustavsson JP, Nielsen DA, Linnoila M, Owen MJ, Sedvall GC. Eur Arch Psychiatry Clin Neurosci. 1997;247:297–302. Kaplan AS, Levitan RD, Yilmaz Z, Davis C, Tharmalingam S, Kennedy JL. Int J Eat Disord. 2008;41:22–8. Kaye WH, Jimerson DC, Lake CR, Ebert MH. Psychiatry Res. 1985;14:333–42. Kipman A, Bruins-Slot L, Boni C, Hanoun N, Adès J, Blot P, Hamon M, Mouren-Siméoni M, Gorwood P. Eur Psychiatry. 2002;17:227–9. Koizumi H, Hashimoto K, Itoh K, Nakazato M, Shimizu E, Ohgake S, Koike K, Okamura N, Matsushita S, Suzuki K, Murayama M, Higuchi S, Iyo M. Am J Med Genet B Neuropsychiatr Genet. 2004;127:125–7. Koronyo-Hamaoui M, Danziger Y, Frisch A, Stein D, Leor S, Laufer N, Carel C, Fennig S, Minoumi M, Apter A, Goldman B, Barkai G, Weizman A, Gak E. Mol Psychiatry. 2002;7:82–5. Koronyo-Hamaoui M, Gak E, Stein D, Frisch A, Danziger Y, Leor S, Michaelovsky E, Laufer N, Carel C, Fennig S, Mimouni M, Apter A, Goldman B, Barkai G, Weizman A. Am J Med Genet B Neuropsychiatr Genet. 2004;131:76–80. Kurabayashi T, Carey DGP, Morrison NA. Diabetes. 1996;45:1358–63. Laakso A, Pohjalainen T, Bergman J, Kajander J, Haaparanta M, Solin O, Syvälahti E, Hietala J. Pharmacogenet Genomics. 2005;15:387–91. Lachman HM, Papolos DF, Saito T, Yu YM, Szumlanski CL, Weinshilboum RM. Pharmacogenetics. 1996;6:243–50. Lauzurica N, Hurtado A, Escartí A, Delgado M, Barrios V, Morandé G, Soriano J, Jàurequi I, Gonzàles-Valdemoro MI, Garcià-Camba E, Fuentes JA. Neurosci Lett. 2003;352:226–30. Levitan RD, Kaplan AS, Masellis M, Basile VS, Walker ML, Lipson N, Siegel GI, Woodside DB, Macciardi FM, Kennedy SH, Kennedy JL. Biol Psychiatry. 2001;50:640–3. Levitan RD, Kaplan AS, Masellis M, Basile VS, Richter MA, Kennedy JL. Eur Neuropsychopharmacol. 2006;16:1–6. Lucki I. Biol Psychiatry. 1998;44:151–62. Matias I, Di Marzo V. Trends Endocrinol Metab. 2007;18:27–37. Matsushita S, Suzuki K, Murayama M, Nishiguchi N, Hishimoto A, Takeda A, Shirakawa O, Higuchi S. Am J Med Genet B Neuropsychiatr Genet. 2004;128:114–7. Michaelovsky E, Frisch A, Leor S, Stein D, Danziger Y, Carel C, Fennig S, Mimouni M, Klauck SM, Benner A, Poustka A, Apter A, Weizman A. Am J Med Genet B Neuropsychiatr Genet. 2005;139:45–50. Mikolajczyk E, Smiarowska M, Grzywacz A, Samochowiec J. Neuropsychobiology. 2006;54:82–6. Miyasaka K, Hosoya H, Sekime A, Ohta M, Amono H, Matsushita S, Suzuki K, Higuchi S, Funakoshi A. J Neural Transm. 2006;113:1279–85. Moffitt TE, Caspi A, Rutter M. Arch Gen Psychiatry. 2005;62:473–81.
76 The Role of Gene Polymorphisms in Susceptibility to Anorexia Nervosa and Bulimia Nervosa
1199
Monteleone P, Tortorella A, Martiadis V, Serritella C, Fuschino A, Maj M. Psychosom Med. 2004;66:744–8. Monteleone P, Santonastaso P, Tortorella A, Favaro A, Fabrazzo M, Castaldo E, Caregaro L, Fuschino A, Maj M. Mol Psychiatry. 2005a;10:716–8. Monteleone P, Fabrazzo M, Martiadis V, Serritella C, Pannuto M, Maj M. Psychol Med. 2005b;35:897–905. Monteleone P, Matias I, Martiadis V, De Petrocellis L, Maj M, Di Marzo V. Neuropsychopharmacol. 2005c;30:1216–21. Monteleone P, Santonastaso P, Mauri, Bellodi L, Erzegovesi S, Fuschino A, Favaro A, Rotondo A, Castaldo E, Maj M. Psychosom Med. 2006a;68:99–103. Monteleone P, Zanardini R, Tortorella A, Gennarelli M, Castaldo E, Canestrelli B, Maj M. Neurosci Lett. 2006b;406:133–7. Monteleone P, Tortorella A, Castaldo E, Di Filippo C, Maj M. Neurosci Lett. 2006c;398:325–7. Monteleone P, Tortorella A, Martiadis V, Serino I, Di Filippo C, Maj M. Neurosci Lett. 2007;421:42–6. Monteleone P, Castaldo E, Maj M. Regulat Peptides. 2008;149:39–50. Muller TD, Reichwald K, Bronner G, Kirschner J, Nguyen TT, Scherag A, Herzog W, Herpertz Dahlmann B, Lichtner P, Meitinger T, Platzer M, Schäfer H, Hebebrand J, Hinney A. Child Adol Psychiatry Ment Health. 2008;2:33–9. Nacmias B, Ricca V, Tedde A, Mezzani B, Rotella CM, Sorbi S. Neurosci Lett. 1999;277:134–6. Nakamura M, Ueno S, Sano A, Tanabe H. Mol Psychiatry. 2000;5:32–8. Nakazato M, Hashimoto K, Yoshimura K, Hashimoto T, Shimizu E, Iyo M. Prog Neuro-Psychopharmacol Biol Psychiatry. 2006;30:1117–21. Nilsson M, Naessén S, Dahlman I, Lindén Hirschberg A, Gustafsson JA, Dahlman-Wright K. Mol Psychiatry. 2004;9:28–34. Nishiguchi N, Matsushita S, Suzuki K, Murayama M, Shirakawa O, Higuchi S. Biol Psychiatry. 2001;50:123–8. Nisoli E, Brunani A, Borgomainerio E, Tonello C, Dioni L, Briscini L, Redaelli G, Molinari E, Cavagnini F, Carruba MO. Eat Weight Disord. 2007;12:91–6. Okuyama Y, Ishiguro H, Nankai M, Shibuia H, Watanabe A, Arinami T. Mol Psychiatry. 2000;5:64–9. Parsons MJ, D’Souza UM, Arranz MJ, Kerwin RW, Makoff AJ. Biol Psychiatry. 2004;56:406–10. Quinton ND, Meechan DW, Brown K, Eastwood H, Blakemore AI. Psychiatr Genet. 2004;14:191–4. Ribasés M, Gratacòs M, Armengol L, de Cid R, Badía A, Jiménez L, Solano R, Vallejo J, Fernández F, Estivill X. Mol Psychiatry. 2003;8:745–51. Ribasés M, Gratacòs M, Fernández-Aranda F, Bellodi L, Boni C, Anderluh M, Cavallini MC, Cellini E, Di Bella D, Erzegovesi S, Foulon C, Gabrovsek M, Gorwood P, Hebebrand J, Hinney A, Holliday J, Hu X, Karwautz A, Kipman A, Komel R, Nacmias B, Remschmidt H, Ricca V, Sorbi S, Wagner G, Treasure J, Collier DA, Estivill X. Hum Mol Genet. 2004;13:1205–12. Ribasés M, Gratacòs M, Fernández-Aranda F, Bellodi L, Boni C, Anderluh M, Cristina Cavallini M, Cellini E, Di Bella D, Erzegovesi S, Foulon C, Gabrovsek M, Gorwood P, Hebebrand J, Hinney A, Holliday J, Hu X, Karwautz A, Kipman A, Komel R, Nacmias B, Remschmidt H, Ricca V, Sorbi S, Tomori M, Wagner G, Treasure J, Collier DA, Estivill X (2005a) Eur. J. Hum. Genet. 13:428–434. Ribases M, Gratacos M, Badia A, Jimenez L, Solano R, Vallejo J, Fernandez-Aranda F, Estivill X. Mol Psychiatry. 2005b;10:851–60. Ribasés M, Fernández-Aranda F, Gratacòs M, Mercader JM, Casasnovas C, Núñez A, Vallejo J, Estivill X. J Psychiatr Res. 2008;42:50–7. Ricca V, Nacmias B, Cellini E, Di Bernardo M, Rotella CM, Sorbi S. Neurosci Lett. 2002;323:105–8. Ricca V, Nacmias B, Boldrini M, Cellini E, di Bernardo M, Ravaldi C, Tedde A, Bagnoli S, Placidi GF, Rotella CM, Sorbi S. Neurosci Lett. 2004;365:92–6. Rosenkranz K, Hinney A, Ziegler A, Hermann H, Fichter M, Mayer H, Siegfried W, Young JK, Remschmidt H, Hebebrand J. J Clin Endocrinol Metab. 1998;83:4524–7. Rosenkranz K, Hinney A, Ziegler A, von Prittwitz S, Barth N, Roth H, Mayer H, Siegfried W, Lehmkuhl G, Poustka F, Schmidt M, Schäfer H, Remschmidt H, Hebebrand J. Int J Obes Relat Metab Disord. 1998;22:157–63. Rybakowski F, Slopien A, Dmitrzak-Weglarz M, Czerski P, Rajewski A, Hauser J. Neuropsychobiology. 2006;53:33–9. Rybakowski F, Dmitrzak-Weglarz M, Szczepankiewicz A, Skibinska M, Slopien A, Rajewski A, Hauser J. Neuroendocrinol Lett. 2007;28:153–8. Sabol SZ, Hu S, Hamer D. Hum Genet. 1998;103:273–9. Serretti A, Artioli P. Psychopharmacology. 2004;174:490–503. Shimizu M, Kanazawa K, Matsuda Y, Takai E, Iwai C, Miyamoto Y, Hashimoto M, Akita H, Yokoyama M. Thromb Res. 2003;112:137–42. Shinohara M, Mizushima H, Hirano M, Shioe K, Nakazawa M, Hiejima Y, Ono Y, Kanba S. J Psychiatry Neurosci. 2004;29:134–7. Siegfried Z, Kanyas K, Latzer Y, Karni O, Bloch M, Lerer B, Berry EM. Am J Med Genet B Neuropsychiatr Genet. 2004;125:126–30.
1200
P. Monteleone and M. Maj
Slopien A, Rybakowski F, Dmitrzak-Weglarz M, Czerski P, Hauser J, Komorowska-Pietrzykowska R, Rajewski A. Acta Neuropsychiatr. 2004;16:290–4. Sodhi M, Kerwin R, Campbell I, Makoff A. Mol Psychiatry. 1999;4:94. Sorbi S, Nacmias B, Tedde A, Ricca V, Mezzani B, Rotella CM. Lancet. 1998;351:1785. Steiger H, Joober R, Israël M, Young SN, Ng Ying Kin NM, Gauvin L, Bruce KR, Joncas J, Torkaman-Zehi A. Int J Eat Disord. 2005;37:57–60. Steiger H, Joober R, Gauvin L, Bruce KR, Richardson J, Israel M, Anestin AS, Groleau P. J Clin Psychiatry. 2008;69:1565–71. Sundaramurthy D, Pieri LF, Gape H, Markham AF, Campbell DA. Am J Med Genet. 2000;96:53–5. Tortorella A, Monteleone P, Martiadis V, Perris F, Maj M. Am J Med Genet B Neuropsychiatr Genet. 2007;144:992–5. Tsuchida A, Nonomura T, Ono-Kishino M, Nakagawa T, Taiji M, Noguchi H. Int J Obes Relat Metab Disord. 2001;25:1286–93. Urwin RE, Bennetts B, Wilcken B, Lampropoulos B, Beumont P, Clarke S, Russell J, Tanner S, Nunn KP. Mol Psychiatry. 2002;7:652–7. Urwin RE, Bennetts BH, Wilcken B, Beumont PJ, Russell JD, Nunn KP. Neuropsychopharmacology. 2003a;28:1351–5. Urwin RE, Bennetts BH, Wilcken B, Lampropoulos B, Beumont PJ, Russell JD, Tanner SL, Nunn KP. Eur J Hum Genet. 2003b;11:945–50. Urwin R, Nunn K. Eur J Hum Genet. 2005;13:370–5. VanNess SH, Owens MJ, Kilts CD. BMC Genet. 2005;6:55. Vink T, Hinney A, van Elburg AA, van Goozen SH, Sandkuijl LA, Sinke RJ, Herpertz-Dahlmann BM, Hebebrand J, Remschmidt H, van Engeland H, Adan RA. Mol Psychiatry. 2001;6:325–8. Westberg L, Bah J, Råstam M, Gillberg C, Wentz E, Melke J, Hellstrand M, Eriksson E. Neuropsychopharmacology. 2002;26:789–93. Ziegler A, Görg T. Lancet. 1999;353:929.
Chapter 77
Changes in Brain Gene Expression in Nutrient Deficiencies: An Example with Iron Erica L. Unger, Narasimha Hegde, and James R. Connor
Abbreviations CSF IF Tim-2 IRP IRE UTR mTOR CamK2a BDNF ARA MBP MOG MAL MOBP PMP22 PLP DAT GABA GAT1 VMAT2 trkB
Cerebrospinal fluid Interstitial fluid T-cell immunoglobulin and mucin domain-containing protein 2 Iron regulatory protein Iron responsive element Untranslated region Mammalian target of rapamycin Calcium/calmodulin-dependent protein kinase II alpha Brain derived neurotrophic factor Arachidonic acid Myelin basic protein Myelin-oligodendrocyte glycoprotein Myelin and lymphocyte protein Myelin-associated oligodendrocytic basic protein Peripheral myelin protein 22 Proteolipid protein Dopamine transporter Gamma-aminobutyric acid Gamma-aminobutyric acid transporter type 1 Vesicular monoamine transporter 2 Tyrosine kinase B
77.1 Introduction Iron is an indispensable element in biological systems and is responsible for the function of many prosthetic groups, including heme and iron–sulfur clusters. Cellular iron uptake, distribution, and export must be tightly regulated, as iron deficiency impairs the function of many proteins, and excess E.L. Unger (*) Department of Nutritional Sciences, Pennsylvania State University, University Park, PA 16802 e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_77, © Springer Science+Business Media, LLC 2011
1201
1202
E.L. Unger et al.
iron can oxidize and damage the protein, nucleic acid, and lipid content of the cell. A well-regulated feedback control functions to enhance iron absorption when the body is deficient and to limit excessive iron accumulation when iron is abundant. Major sites of iron regulation are the doudenum where dietary iron is absorbed, the bone marrow where red blood cells are produced, the liver where most excess iron is stored, and the spleen and reticuloendothelial system where red blood cells are catabolized and iron is extracted for reuse. The importance of iron homeostasis becomes clear when one considers that iron deficiency anemia is the most widespread nutritional disorder affecting over 30% of the world’s population, and that iron overload can lead to diabetes and heart disease and is a contributing factor to neurodegenerative diseases. The proper distribution and sequestering of iron is a challenging task at both the cellular and organismic levels due to the involvement of many organs and their differential requirement for iron that differs by age and sex. In the brain, iron is required for a multitude of cellular processes; thus it is the organ most vulnerable to iron deficiency during critical periods of development, particularly during the last trimester of fetal life and during the period of brain growth spurt and differentiation. This chapter will discuss iron regulation in the brain and changes in gene expression in the condition of iron deficiency.
77.2 Iron Transport in the Brain Iron is required as a cofactor in central nervous system metabolic processes such as oxygen transport, neurotransmitter production, nitric oxide metabolism, and oxidative phosphorylation (Ponka 2004). The neuronal need for iron is obvious considering the fact that about 20% of the oxygen consumption of the body relies on mitochondrial respiration in the brain. During development, embryonic and dividing neurons require iron to support the activity of the enzyme ribonucleotide reductase, which is critical for DNA replication. Further, the role of iron in neuroectodermal development was demonstrated by mutating the gene encoding the transferrin receptor, which provides transferrin-bound iron to the brain, resulting in nonviable embryos with severe malfunctions of the central nervous system (Levy et al. 1999). Evidence over the last 2 decades suggests that mechanisms of iron transport across the blood–brain barrier involve a transferrin–transferrin receptor pathway. (Bradbury 1997; Malecki et al. 1999; Burdo and Connor 2003). Uptake of transferrin-bound iron by transferrin receptor-mediated endocytosis from the blood into the cerebral endothelial cells is similar to iron uptake by other cell types. This process involves binding, endocytosis, acidification, and dissociation, and finally translocation of iron across the endosomal membrane via a process that may involve divalent metal transporter 1 (Fleming et al. 1997). A second proposed mechanism of iron transport across the abluminal membrane involves astrocytes, through their endfoot processes on the capillary endothelium (Malecki et al. 1999; Oshiro et al. 2000). A small portion of transferrin–iron complex is also known to cross the blood–brain barrier by receptormediated transcytosis (Moos and Morgan 1998). In addition to these pathways, it has been proposed that the lactoferrin receptor-lactoferrin and glycosylphosphatidylinositol (GPI)- anchored p97-secreted pathways may play a role in iron transport across the blood–brain barrier (Faucheux et al. 1995). When iron has been transported across the blood–brain barrier, it binds to transferrin that is secreted from the oligodendrocytes and epithelial cells of the choroid plexus (Bradbury 1997). Transferrin in cerebrospinal fluid (CSF) and interstitial fluid (IF) is fully saturated with iron under normal conditions, and several experiments suggest that the iron concentration exceeds that of the binding capacity of transferrin in the CSF and IF. The nontransferrin-bound iron in the brain is likely associated with citrate, ascorbate, albumin, lactoferrin, and secreted p97. The presence of nontransferrin-bound iron in brain extracellular fluids and the absence of transferrin receptors in astrocytes, oligodendricytes, and microglia show that glia acquire iron via mechanisms independent
77 Changes in Brain Gene Expression in Nutrient Deficiencies
1203
of transferrin/transferrin receptor mediation (Moos 1996; Moos and Morgan 1998). For instance, oligodendrocytes express T-cell immunoglobulin and mucin domain-containing protein 2 (Tim-2), which binds and internalizes the iron storage protein H-ferritin. This interaction is likely the main mechanism for iron acquisition into oligodendrocytes (Todorich et al. 2008).
77.3 Iron Homeostasis Given the fact that iron requirements far exceed the gastrointestinal absorption capacity, iron utilized on a daily basis is continuously recycled from internal storage. This process is regulated by the hepatocyte-derived peptide hepcidin, which maintains systemic iron homeostasis by regulating gastrointestinal iron absorption and iron release from reticuloendothelial stores via the plasma membrane iron exporter ferroportin (Nemeth et al. 2004). Ceruloplasmin maintains a rate of plasma iron oxidation sufficient for the continued release of this essential metal from the storage site (Harris et al. 1998). Iron levels in the blood are not static and are known to fluctuate with the diurnal cycle, which has been attributed to fluctuating iron release from reticuloendothelial cells (Scales et al. 1988; Uchida et al. 1983). The rationale for variations in plasma iron relative to fluctuations in tissue needs for iron is interesting but not elucidated by current theories of iron requirements. Brain iron accounts for less than 2% of the total body iron content. Iron content in the brain is highest in basal ganglia, with substantia nigra, red nucleus, and hippocampus also having significant iron contents (Hill and Switzer 1984; Morris et al. 1992). Similar to peripheral iron levels, regional brain iron levels fluctuate with the diurnal cycle, with as much as 30% differences in ventral midbrain (substantia nigra/ventral tegmentum) iron between the light and dark phases of the diurnal cycle in several inbred strains of mice (Unger et al. 2009). This flux in iron levels has not yet been attributed to specific cell types and the mechanisms surrounding this observation remain unknown. This is a paradigm shifting observation that establishes a dynamic state of iron flux rather than the static concepts that are entrenched in the literature to date (Dallman and Spirito 1977). In the developing brain, iron uptake is greatest during the prenatal period when the brain is undergoing rapid growth. Iron deficiency in infancy is now known to inhibit cognitive development and alter affective behavior. Iron supplementation is ineffective in correcting many of these deficits (Lozoff 2007; Lozoff et al. 2000; Pinero et al. 2001). Similar to humans, rodents that were iron deficient during gestation and early postnatal life show similar deficits in cognition and attention after iron repletion. Thus, an understanding of the effects of early iron deficiency on brain structure and function and of the inability of the brain to correct iron content during postnatal iron repletion has critical implications. Iron regulatory proteins: Regulation of iron metabolism occurs through iron regulatory proteins (IRPs) and iron-responsive elements (IREs). IRP1 and IRP2 register cytosolic iron concentrations and posttranscriptionally regulate expression of iron metabolism genes. These two proteins are ubiquitously expressed mammalian members of the aconitase gene family (Gruer et al. 1997), and bind to IRE with high affinity when intracellular iron levels are low. The IRE is a highly conserved nucleotide hairpin (IRP binding regions) located in the untranslated region (UTR) in mRNAs encoding specific proteins of iron metabolism (Henderson et al. 1994). Among the genes known to take part in iron homeostasis, the mRNA for ferritin, ferroportin, and HIF2alpha contain IREs in the 5¢ untranslated region while transferrin receptor 1 and DMT1 mRNA contain IREs in the 3¢ untranslated regions. In states of iron depletion, each IRP responds by binding to IREs. When the IRE is located in the 3¢ UTR, the binding of IRPs to IRE stabilizes the mRNA, and hence more protein is translated (Haile 1999). Alternatively, by binding to a single IRE located in the 5¢ UTR of mRNA, the IRP prevents translation. In cells that are iron-repleted, IRPs do not bind IREs, and ferritin and other transcripts that have an IRE in the 5¢UTR are freely translated (Klausner et al. 1993).
1204
E.L. Unger et al.
77.4 Effect of Dietary Iron on Gene Expression in the Brain 77.4.1 Iron Regulatory Proteins Abnormal amounts of iron in the brain have been documented in a number of age-related neurodegenerative disorders including Alzheimer’s (Bishop et al. 2002; Connor et al. 1992a, b 1995) and Parkinson’s diseases (Kaur and Andersen 2004). It is generally accepted that this excess iron catalyzes the formation of reactive oxygen species and induces oxidative damage to which the brain is sensitive (Robb and Connor 1998). Iron deficiency-related research conducted to date suggests that dietary iron deficiency during critical periods of development also adversely affects brain function (Beard et al. 2002, 2003a; Felt et al. 2006; Felt and Lozoff 1996). Although a large body of literature is available that focuses on dietary iron deficiency and cognitive and behavioral deficits, only recently have studies focused on gene expression profiling in the brain during prenatal and postnatal iron deficiency (Carlson et al. 2007; Clardy et al. 2006; Liu et al. 2007). Alterations in gene expression of the iron transporter transferrin in brain in response to iron deficiency has been observed in several rodent models. Iron deficiency from early gestation to postanatal day 21 produces a downregulation in transferrin gene expression (Clardy et al. 2006), which was also observed in a postweaning animal model where iron deficiency continued for six weeks (Han et al. 2003 #184). These studies further indicate that transferrin receptor and divalent metal transporter 1 are upregulated by gestational/lactational iron deficiency, while transferrin receptor gene expression is not changed by postnatal iron deficiency (Carlson et al. 2007; Clardy et al. 2006; Han et al. 2003; Siddappa et al. 2002). In pre and postnatal models of iron deficiency, transferrin protein levels in the brain and periphery are typically upregulated, which is an indication that transferrin protein could be acquired from sources outside the brain (Clardy et al. 2006; Crowe and Morgan 1992; Han et al. 2003). In contrast to the previously mentioned studies, iron deficiency during gestation and during the first week of lactation followed by 1 week of iron repletion increases transferrin mRNA levels in hippocampus (Carlson et al. 2007). Oligodendrocytes are the predominant source of transferrin mRNA in the brain, suggesting that transferrin mRNA production by oligodendrocytes may be upregulated by small periods of adequate iron during early postanal life.
77.4.2 Energy Metabolism and Cell Growth Since iron homeostasis is coupled to energy metabolism, it is expected that dietary iron deficiency would bring about changes in the expression pattern of genes involved in energy metabolism and cellular growth. Mammalian target of rapamycin (mTOR) is a serine/threonine protein kinase that integrates several pathways to regulate cell growth, differentiation and survival. Importantly, the expression of genes in the mTOR signaling cascade including Fkbp 1a, Gltscr 2, Ddit4, and Tsc2 are altered in hippocampus from gestational and early lactational iron-deficient rats (Carlson et al. 2007). A significant upregulation of the gene Slc2a1, the major brain glucose (energy) transporter, is observed in this same animal model (Carlson et al. 2007). Altered regulation of genes involved in energy metabolism and cell growth suggests that neural dendritic growth and synaptic transmission may be affected by iron deficiency. Iron deficiency also induced an upregulation in the hippocampal expression of the calcium/calmodulin-dependent protein kinase II alpha (CamK2a) gene, which is regulated by brain derived neurotrophic factor (BDNF) through the mTOR pathway, further implies an altered density of glutamatergic synapses and changes in synaptic plasticity (Schratt et al. 2004).
77 Changes in Brain Gene Expression in Nutrient Deficiencies
1205
These specific neuronal changes may underlie some of the disturbed spatial recognition and learning abilities in the iron-deficient animal models. Genes involved in the cell division cycle are also responsive to iron depletion. Microarray analyses of the whole brain from rats that are iron deficient during early gestation until weaning show more than a 1-fold increase in the cyclin B1, cyclin D1, cyclin L, and cyclin-dependent kinase four genes, indicating that iron deficiency affects the cell cycle progression during this period. To date, there is no evidence as to which cell type(s) show more active cycling after this period of iron deficiency. Studies using isolated hippocampal cells cultured in iron-depleted conditions show altered mRNA levels of Bs69, Pdcd5, Anapc8, and Ruvbl2 (Liu et al. 2007). These proteins function during cell cycle arrest and can lead to apoptotic events in the brain. The direct relationship, if any, between iron status of cells and cell cycle activity is not known. Does iron status drive cell division or is iron only important for the cell to meet metabolic needs during division? One possible direct mechanism is through IRP2 which has a phosphorylation site, Ser157, that is phosphorylated by cyclin-dependent kinase/cyclin B1 during G(2)M. Thus phosphorylation of this site provides a mechanism by which ferritin and transferrin receptor production can be regulated during the cell cycle (Wallander et al. 2008). Microarray analyses of whole brain (Clardy et al. 2006) and hippocampus (Carlson et al. 2007) suggest that several genes involved in cellular signal transduction mechanisms are altered by prenatal/ lactational iron deficiency. Both of these studies have reported nearly a 3-fold increase (highest among all the genes represented in the array) in the expression of arachidonate 12-lipoxygenase and arachidonate 15-lipoxygenase. These enzymes belong to the family of oxydoreductases and are iron-dependent. 12/15-lipoxygenases are the major source of oxidative stress during pathological conditions including Alzheimer’s disease (Pratico et al. 2004), which suggests that upregulation of these two enzymes could be indicative of brain oxidative stress induced by dietary iron deficiency. Arachidonate 12-lipoxygenase and arachidonate 15-lipoxygenase also participate in arachidonic acid (ARA) metabolism. ARA is a polyunsaturated fatty acid that is present in the phospholipids of membranes and is abundant in the brain (Rapoport 2008). Functions of arachidonic acid in the brain include maintaining hippocampal cell membrane fluidity (Fukaya et al. 2007), protecting the brain from oxidative stress by activating perioxisomal proliferator-activated receptors (Wang et al. 2006), and activating syntaxin-3 that is involved in neuron growth and repair (Darios and Davletov 2006). The involvement of ARA in cellular signaling as a lipid second messenger and as an inflammatory intermediate has been well documented (Darios and Davletov 2006; Abe et al. 1992). It is worth mentioning at this point that upregulation of these oxydoreductases in the ARA pathway are a good indication of altered lipid synthesis during developmental iron deficiency. The brain has to depend on de novo synthesis of lipids and cholesterols for structural changes such as dendritic growth, synapse formation, and myelination, which reach their peak during early postnatal life (Brody et al. 1987). Changes in the structural composition of the brain are likely to alter many processes including synaptic neurotransmission between and within brain regions and cause potentially irreparable behavioral deficits.
77.4.3 Myelin-Related Genes Oligodendrocytes produce myelin in the CNS and are the most robustly iron stained cells in the healthy adult brain (Benkovic and Connor 1993; Todorich et al. 2009). A deficiency in iron increases the latency of auditory brain stem responses in human infants, an indirect indication of hypomyelination (Roncagliolo et al. 1998). Moreover, gestational and lactational iron deficiency in rodents reduces myelin formation in the spinal cord and the presence of integral myelin proteins proteolipid protein and myelin basic protein isoforms in whole brain homogenates (Ortiz et al. 2004; Yu et al. 1986).
1206
E.L. Unger et al.
Table 77.1 Myelin-related genes downregulated by early iron deficiency in whole brain Name Known or proposed function Myelin basic protein Myelin compaction and stability Myelin-oligodendrocyte glycoprotein Myelin stability, maintenance Myelin and lymphocyte protein Myelin biogenesis, stabilization Myelin-associated oligodendrocyte basic protein Myelin sheath maintenance Peripheral myelin protein 22 Myelin compaction Proteolipid protein Myelin compaction, oligodendrocyte maturation Myelin is a fatty insulating sheath that is wrapped around the nerve process termed the axon. Mylein increases the speed at which signals are conducted within the brain. A reduction in myelin associated genes suggests that the myelin content of the brain is reduced, thus affecting the transmission of signals
Further, hindbrain CNPase and myelin basic protein levels, indicators of reduced myelin production, are reduced in rats only exposed to postweaning iron deficiency, indicating that sufficient iron is required by oligodendrocytes for maintenance of myelin even after the period of peak myelination from P8-P12 (Beard et al. 2003b).The question of whether these changes in myelination are due to posttranslational events in iron deficiency or whether gene expression is also altered was examined as part of a study that investigated gene responses to gestational and early postnatal iron deficiency in rats (Clardy et al. 2006). This model of early iron deficiency showed a downregulation of myelin basic protein (MBP) and myelin-oligodendrocyte glycoprotein (MOG) that are important in myelin stability, myelin and lymphocyte protein (MAL) that is involved in myelin biogenesis, and myelinassociated oligodendrocytic basic protein (MOBP) that may be important in myelin sheath maintenance. Other downregulated genes include peripheral myelin protein 22 (PMP22), proteolipid protein (PLP), fibroblast growth factor, and chimerin 2. The decrease in myelin-related gene expression is consistent with a delay in oligodendrocyte maturation or with a reduced number of oligodendrocytes in the iron-deficient brain (Clardy et al. 2006). Importantly, iron supplementation from weaning until 6 months of age can correct the deficits in whole brain myelin gene expression (Clardy et al. 2006). Despite this normalization of gene expression, myelin deficits remain at 6 months of age, indicating that there is a critical time point during which iron is required for proper myelination in adulthood (Ortiz et al. 2004). Given these recent findings, future studies aimed at investigating the critical period during which iron supplementation is required for proper myelination are warranted. The peak period of myelination is P8-P12 in the rat, but the question remains as to how late in postnatal life that iron supplementation can begin to have proper myelin production (Table 77.1).
77.5 Iron Deficiency and Neurotransmitter Systems 77.5.1 Dopamine Extensive studies during the last several decades have indicated that iron deficiency has deleterious effects on the dopamine neurotransmitter system. The nigrostriatal dopamine pathway that regulates movement and the mesolimbic pathway that mediates the reinforcing effects of drug addiction appear to be the most affected regions of the brain (Beard 2003; Beard et al. 2003a; Erikson et al. 2000; Pinero et al. 2001; Youdim et al. 1984; Youdim). The 12 transmembrane domain protein, the dopamine transporter (DAT), serves as a key regulator of dopamine neurotransmission and is encoded by the Slc6A3 gene. Importantly, the solute carrier (Slc) gene family encodes membrane transport
77 Changes in Brain Gene Expression in Nutrient Deficiencies
1207
Table 77.2 Genes affected by iron deficiency that encode neurotransmitter transporters and receptors Name Period of ID Brain region(s) Upregulated genes Norepinephrine transporter GABA transporter type 1 Serotonin transporter Vesicular monoamine transporter 2 Downregulated genes
P21 – P63 P21 – P63 G5 – P21 G5 – P21
Globus pallidus Substantia nigra Whole brain Whole brain
Dopamine receptor 1A G5 – P21 Whole brain Norepinephrine transporter P21 – P63 Locus ceruleus, substantia nigra, cerebellum Norepinephrine alpha2 receptor P21 – P63 Locus ceruleus, cerebellum GABAA receptor delta G5 – P21 Whole brain GABAA receptor alpha6 G5 – P21 Whole brain GABAA receptor P21 – P63 Substantia nigra GABAB receptor P21 – P63 Substantia nigra Many neurotransmitter systems including dopamine, serotonin, norepinephrine and GABA are altered by iron deficiency. The timing (postnatal versus gestational) and the extent of iron deficiency are likely important factors in producing these outcomes. Further, the effect of iron deficiency on the brain is region dependent, i.e. all brain regions are not affected equally by iron deficiency Period of iron deficiency (ID); G gestational day; P postnatal day
proteins that regulate neurotransmitter and nutrient transport, and deficits in the expression of these genes have serious health implications. In most rodent models of iron deficiency, regional DAT protein expression is downregulated, but the limited studies to date do not indicate that there is a similar change in Slc6A3 gene expression (Carlson et al. 2007; Clardy et al. 2006). The protein levels of two subtypes of dopamine receptors, D1 and D2, are also affected by iron deficiency, although only a downregulation in dopamine receptor 1A has been described in whole brain. Gene expression of DAT and dopamine receptor subtypes have not been evaluated in brain regions that typically show the largest reductions in DAT protein (i.e., striatum and nucleus accumbens). Thus, studies evaluating dopamine-related gene expression within these regions are necessary to conclude that changes in protein levels are a consequence of posttranslational modifications and/or changes in gene expression in the iron-deficient brain (Table 77.2).
77.5.2 Norepinephrine There is a large body of literature that implicates norepinephrine in the modulation of attention, memory, emotion, and drug addiction. Deficits in these behaviors have been linked to iron deficiency in humans and rodents (Unger et al. 2007; Lozoff et al. 2006; Lozoff and Brittenham 1987; Lozoff et al. 1991, 2007). Norepinephrine cell bodies project from the locus coeruleus in the pons to diverse regions in the forebrain (cortex, caudate putamen, and hippocampus) or to the hindbrain cerebellum. Regulation of norepinephrine signaling occurs via the norepinephrine transporter, a member of the Slc6 family of transporters that has a similar structure to the previously mentioned dopamine transporter, and two distinct classes of receptors, alpha and beta, that are further divided into subtypes. Iron depletion in the brain typically reduces regional norepinephrine transporter protein expression and intracellular and extracellular norepinephrine levels in striatum (Anderson et al. 2009; Beard et al. 2006b; Bianco et al. 2009). Two studies have specifically investigated the effects of iron deficiency on norepinephrine transporter and receptor mRNA levels in the brain using a postweaning animal model (iron deficient from
1208
E.L. Unger et al.
postnatal day 21 to day 63). Iron deficiency causes a downregulation of norepinephrine transporter mRNA in locus ceruleus, substantia nigra, and cerebellum and produces an upregulation in globus pallidus compared to iron-sufficient rats (Anderson et al. 2009; Beard et al. 2006b). Likewise, alpha2 receptor expression is reduced in locus ceruleus and cerebellum. These changes in the norepinephrine neurotransmitter system are not surprising given the extensive changes in the dopamine system in iron deficiency and the anatomical connections between the dopaminergic and noradrenergic systems. There is mounting evidence that the locus ceruleus can mediate dopamine release in the nucleus accumbens via direct and indirect projections to the ventral tegmental area, and that dopamine afferents in the locus ceruleus originate in the ventral tegmental area (Guiard et al. 2008). These projections allow a communication between neurotransmitter systems that together modulate cognition, attention, affect, drug addiction, and many other behaviors. As research into the gene expression profile of iron deficiency continues, elucidating the response of key genes involved in these pathways will become increasingly important to understand the persistent effects associated with iron deficiency.
77.5.3 Serotonin The serotonin neurotransmitter system is widely known for its involvement in anxiety- and depression-related disorders. Slc6a4 encodes the serotonin transporter, which moves serotonin into the presynaptic neuron to terminate the serotonin signal, and is the site of action of many antidepressant/ antianxiety drugs. Gestational and early postnatal iron deficiency causes an upregulation of Slc6a4 in whole brain (Clardy et al. 2006). Several postweaning models of iron deficiency have indicated that regional serotonin transporter protein levels are reduced, which creates a potential disconnect between gene and protein data (Beard 2003; Burhans et al. 2005). It is likely that the effects of preversus postweaning iron deficiency on the serotonin neurotransmitter system are quite diverse. A more in depth study investigating regional serotonin-related gene expression during iron deficiency will facilitate our knowledge as to the impact of iron regulation on the serotonin neurotransmitter system, and perhaps offer clues as to the relationship between iron deficiency, serotonin, and affective disorders.
77.5.4 GABA Gamma-aminobutyric acid (GABA) is the most abundant inhibitory neurotransmitter in the adult mammalian central nervous system. Similar to monoamine neurotransmitters, the action of GABA is terminated by uptake via a membrane transporter protein and transmitted via distinct transmembrane receptors. The GABA neurotransmitter system has 4 known subtypes of transporters named type 1 (GAT1), type 2, type 3, and type 4 GABA transporters that are located on neurons and glia (Beleboni et al. 2004). The receptors are divided into two classes, metabotropic and ionotropic receptors. GABAB receptors are metabotropic and characterized by their ability to produce a slow inhibitory response through activation of G proteins (G-protein coupled receptors) and subsequent opening of ion channels. GABAA and GABAC ionotropic receptors are directly coupled to transmembrane ion channels (ligand gated ion channels) and produce fast responses. An understanding of the effect of iron deficiency on the GABA neurotransmitter system has become increasingly important because of behavioral abnormalities that are characteristic of deficits in dopamine/GABA interactions in the substantia nigra, striatum, and nucleus accumbens (Beard et al. 2006a; Erikson et al. 2000, 2001;
77 Changes in Brain Gene Expression in Nutrient Deficiencies
1209
Pinero et al. 2001). These brain regions have proven to be sensitive to iron loss and to dopamine system dysregulation in iron deficiency (Beard et al. 2007; Erikson et al. 1997; Unger et al. 2007). GABA-related gene expression in iron deficiency has been investigated in gestational and postweaning animal models. Microarray analysis of whole brain from rats that were deficient in iron from gestational day 5 to postnatal day 21 indicates that GABAA receptor delta and GABAA receptor alpha 6 are downregulated after early iron deficiency (Clardy et al. 2006). Since this study was performed with whole brain samples, we are unable to pinpoint these changes to distinct brain regions. Importantly, it appears likely that long-term iron repletion can correct this deficit since 5 months of iron supplementation did restore whole brain GABA receptor mRNA to control levels. Iron deficiency begun at postnatal day 21 and continuing for 6 weeks reduces substantia nigra GABAA and GABAB receptor expression compared to iron sufficient rats, with other regions including caudate and cerebellum not being significantly affected (Anderson et al. 2008). GAT-1 mRNA expression was increased 3-fold in substantia nigra, again with little affect in other brain regions. Protein concentrations of GAT-1 were decreased and GABAA were increased in the iron-deficient substantia nigra suggesting that the gene response is a mechanism aimed at maintaining normal levels of these proteins or that there is a change in transcription efficiency. There are GABAergic striatal neurons that project to the substantia nigra (striatonigral neurons) that act to inhibit dopaminergic output from the dopamine call body containing substantia nigra. Given the consistent changes in GABA gene and protein expression in the substantia nigra, this provides a mechanism by which GABA may be involved in regulating dopamine and the related behaviors observed in iron deficiency.
77.5.5 Other Transporters Two additional solute carrier genes that are important in the regulation of neurotransmitter signaling, Slc18a2 and Slc7a1 are upregulated by gestational/lactational iron deficiency (Clardy et al. 2006). Slc18a2, or vesicular monoamine transporter 2 (VMAT2), accumulates cytosolic monoamines including dopamine, serotonin, and norepinephrine into synaptic vesicles via an ATP dependent mechanism for release into the synapse. As noted previously, neurotransmitter system functioning is compromised in iron deficiency, and altered VMAT2 expression may contribute to the observed deficits in iron-deficient rodents. The gene encoding Slc7a1, or cationic amino acid transporter 1, is also upregulated in the same iron deficiency model and is responsible for arginine, lysine, ornithine, and histidine transport in the brain. Its role in arginine transport has implications for cell proliferation and for the production of nitric oxide that is necessary for many biological processes and has detrimental effects if not present at appropriate levels (Lameu et al. 2009). The altered expression of these transporter genes further indicates that proper movement of molecules and ions in the iron-deficient brain cannot be maintained.
77.6 Iron Deficiency and Neurotrophic Factors Neurotrophic factors are a family of proteins that promote the growth and maintenance of developing neurons and the survival of mature neurons. Probably the best studied class of neurotophic factors are the neurotrophins, that include brain derived neurotrophic factor (BDNF) and nerve growth factor (NGF) among others. Perinatal iron deficiency from gestational day 2 to postnatal day 7 reduces levels of BDNF transcript III and BDNF transcript IV mRNA at postnatal days 7, 15, and 30, and these effects persist through postnatal day 65 (Tran et al. 2008, 2009). The gene expression of the
1210
E.L. Unger et al.
primary binding site of BDNF, tyrosine kinase B (trkB) receptor, is also reduced at P65. The BDNF signaling mechanism has several downstream targets including the activity-dependent immediate early transcription factors that assist in the regulation of neurite outgrowth and synaptic plasticity. Similar to BDNF, these transcriptional targets, early-growth-response-gene 1 (ERG1) and 2 (ERG2), cfos, and 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR), are reduced at P65. The literature suggests that a reduction in the expression of BDNF and activity dependent immediate early transcription factors in hippocampus can result in a loss of neuronal plasticity. Indeed, early iron deficiency in rats does produce long-term changes in the morphology of dendrites and alters synaptic function in the CA1 area of the hippocampal dendrites (Jorgenson et al. 2003, 2005). These studies illustrate that perinatal iron deficiency reduces the action of BDNF through early adult life and perhaps throughout the lifespan. Importantly, peak import of iron into the hippocampus occurs between postnatal day 5 and postnatal day 15 (Siddappa et al. 2002), and dramatic changes in gene expression in the rat hippocampus occur during the first 2 weeks of postnatal life (Stead et al. 2006). This period also encompasses the stage of maximal neuron growth and differentiation. Detrimental neurochemical and behavioral outcomes resulting from altered BDNF gene expression have been revealed in several mouse models and include deficits in maintenance of long term potentiation (Barco et al. 2005; Patterson et al. 1996), learning (Gorski et al. 2003; Monteggia et al. 2004) and affective behaviors (MacQueen et al. 2001). These findings may have direct relevance to the prolonged and irreversible losses in hippocampal-based learning and memory functioning and emotional responses that have been observed in human infants with iron deficiency and iron-deficient rodent models (Felt et al. 2006; Lozoff 2007; Lozoff et al. 2006).
77.7 A pplications to Other Areas of Health and Disease – Alzheimer’s Disease Dysregulation of iron homeostasis has been implicated in several disease states including Parkinson’s and Alzheimer’s diseases (Connor et al. 1992a, b). An upregulation of several genes known to be implicated in the etiology in Alzheimer’s disease has been observed in the developing hippocampus of mice that were iron deficient during gestation and early lactation (Carlson et al. 2007). These genes include connective tissue growth factor, fibronectin 1, cystatin C, cathepsin B, cathepsin S, amyloid beta precursor-like protein 1, amyloid beta precursor protein-binding familyB member 1, clusterin, and NDRG2. These data suggest there could be iron deficiency in the Alzheimer’s brain despite the reported increase in iron. The latter concept may not be contradictory; much of the elevated iron in the Alzheimer’s disease brain may be bound to plaques and thus, not bioavailable (Connor et al. 1992a; Meadowcroft et al. 2009). Moreover, there is a similar pattern of expression in genes associated with oxidative stress in the hippocampus of iron-deficient rodents and brains from Alzheimer’s patients. These data suggest that we should re-evalute the interpretation of some of these data to consider that iron deficiency could generate a gene/protein profile of oxidative stress
Summary Points • The brain is the organ most vulnerable to iron deficiency during critical periods of development that include the stage of brain growth spurt and differentiation. • Genes that are altered by brain iron deficiency are integral to processes including signal transduction, myelin formation, cell growth, energy metabolism, and neurotransmitter signaling.
77 Changes in Brain Gene Expression in Nutrient Deficiencies
1211
• While the expression level of some genes can be corrected with iron supplementation, other genes may not be responsive to iron treatment strategies. • Timing of iron deficiency is an important factor when considering treatment intervention strategies. Critical to this idea is to determine the point in development when iron supplementation is required such that the negative behavioral effects associated with early iron deficiency do not persist through adulthood. • The novel concept that iron deficiency may generate a stress profile in the brain similar that that observed in Alzheimer’s Disease is an area that warrants future studies.
Definitions of Key Terms and Words Iron deficiency anemia: is a late stage of iron depletion characterized by low plasma iron levels, a low serum transferrin saturation, elevated transferrin receptor levels and low serum ferritin concentrations. Transferrin: is a glycoprotein that binds ferric iron for transport to target tissues. The transferrin receptor: is a carrier protein that binds transferrin for movement of iron across membranes. Transferrin receptor levels are regulated by iron content in the body via iron regulatory proteins and iron response elements. An oligodendrocyte: in a type of glial cell located within the central nervous system that is responsible for myelination of axon processes. Oxidative stress: is a condition defined by elevated levels of reactive oxygen species, including peroxidases and free radicals, which the body is not able to eliminate efficiently. Dopamine: is an inhibitory neurotransmitter synthesized from the amino acid tyrosine. This neurotransmitter is the precursor of the neurotransmitter norepinephrine. Dopamine transporter: is a 12 transmembrane domain protein that transports dopamine from the extracellular space into the cytoplasm of a presynaptic neuron. Brain-derived neurotrophic factor: is a member of the neurotrophin family of growth factors that supports neuron growth and survival.
Key Points of Iron Deficiency • The World Health Organization (WHO) recognizes iron deficiency as the most common and widespread nutrient disorder in the world. Iron deficiency is currently the only nutrient deficiency that is prevalent in both nonindustrialized and industrialized countries, where it is prevalent in mainly women and children. • Iron deficiency is believed to be the most common cause of anemia, which afflicts more than two billion people worldwide. Anemia is a condition in which the number of red blood cells in the body is reduced or the ability of the red blood cells to carry oxygen is diminished and cannot meet the body’s needs. • Iron deficiency in infancy has been consistently associated with adverse changes in development and in behavior, including impaired motor skills and increased fearfulness and unhappiness. Many of these behavioral outcomes continue to be observed for years after iron status has been improved in these children. • Iron deficiency anemia can have a significant impact on an individual’s work capacity by causing severe fatigue and producing affective difficulties, decreased attention, and reduced cognitive abilities. These negative outcomes can have an economical impact, particularly in regions where iron deficiency anemia is rampant. • There is considerable evidence that iron plays an important role in many processes in the brain including neurotransmitter metabolism, myelin formation, and brain energy metabolism. These physiological consequences have been linked to the behavioral and development observations in iron deficiency.
1212
E.L. Unger et al.
References Abe K, Araki T, Kawagoe J, Aoki M, Kogure K. Adv Exp Med Biol. 1992;318:183–95. Anderson JG, Fordahl SC, Cooney PT, Weaver TL, Colyer CL, Erikson KM. Neurotoxicology. 2008;29:1044–53. Anderson JG, Fordahl SC, Cooney PT, Weaver TL, Colyer CL, Erikson KM. Brain Res. 2009;1281:1–14. Barco A, Patterson S, Alarcon JM, Gromova P, Mata-Roig M, Morozov A, Kandel ER. Neuron. 2005;48:123–37. Beard J. J Nutr. 2003;133:1468S–72S. Beard J, Erikson KM, Jones BC. J Nutr. 2003a;133:1174–9. Beard JL, Erikson KM, Jones BC. Behav Brain Res. 2002;134:517–24. Beard JL, Felt B, Schallert T, Burhans M, Connor JR, Georgieff MK. Behav Brain Res. 2006a;170:224–32. Beard JL, Unger EL, Bianco LE, Paul T, Rundle SE, Jones BC. J Nutr. 2007;137:1176–82. Beard JL, Wiesinger JA, Connor JR. Dev Neurosci. 2003b;25:308–15. Beard JL, Wiesinger JA, Jones BC. Brain Res. 2006b;1092:47–58. Beleboni RO, Carolino RO, Pizzo AB, Castellan-Baldan L, Coutinho-Netto J, dos Santos WF, Coimbra NC. Cell Mol Neurobiol. 2004;24:707–28. Benkovic SA, Connor JR. J Comp Neurol. 1993;338:97–113. Bianco LE, Unger EL, Earley CJ, Beard JL. Chronobiol Int. 2009;26:447–63. Bishop GM, Robinson SR, Liu Q, Perry G, Atwood CS, Smith MA. Dev Neurosci. 2002;24:184–7. Bradbury MW. J Neurochem. 1997;69:443–54. Brody BA, Kinney HC, Kloman AS, Gilles FH. J Neuropathol Exp Neurol. 1987;46:283–301. Burdo JR, Connor JR. Biometals. 2003;16:63–75. Burhans MS, Dailey C, Beard Z, Wiesinger J, Murray-Kolb L, Jones BC, Beard JL. Nutr Neurosci. 2005;8:31–8. Carlson ES, Stead JD, Neal CR, Petryk A, Georgieff MK. Hippocampus. 2007;17:679–91. Clardy SL, Wang X, Zhao W, Liu W, Chase GA, Beard JL, True Felt B, Connor JR. J Neural Transm Suppl. 2006;173–96. Connor JR, Menzies SL, St Martin SM, Mufson EJ. J Neurosci Res. 1992a;31:75–83. Connor JR, Snyder BS, Arosio P, Loeffler DA, LeWitt P. J Neurochem. 1995;65:717–24. Connor JR, Snyder BS, Beard JL, Fine RE, Mufson EJ. J Neurosci Res. 1992b;31:327–35. Crowe A, Morgan EH. Brain Res. 1992;592:8–16. Dallman PR, Spirito RA. J Nutr. 1977;107:1075–81. Darios F, Davletov B. Nature. 2006;440:813–7. Erikson KM, Jones BC, Beard JL. J Nutr. 2000;130:2831–7. Erikson KM, Jones BC, Hess EJ, Zhang Q, Beard JL. Pharmacol Biochem Behav. 2001;69:409–18. Erikson KM, Pinero DJ, Connor JR, Beard JL. J Nutr. 1997;127:2030–8. Faucheux BA, Nillesse N, Damier P, Spik G, Mouatt-Prigent A, Pierce A, Leveugle B, Kubis N, Hauw JJ, Agid Y, et al. Proc Natl Acad Sci U S A. 1995;92:9603–7. Felt BT, Beard J, Schallert T, Shao J, Aldridge JW, Connor JR, Georgieff MK, Lozoff B. Behav Brain Res. 2006;171:261–70. Felt BT, Lozoff B. J Nutr. 1996;126:693–701. Fleming MD, Trenor CC, 3rd, Su MA, Foernzler D, Beier DR, Dietrich WF, Andrews NC. Nat Genet. 1997;16:383–6. Fukaya T, Gondaira T, Kashiyae Y, Kotani S, Ishikura Y, Fujikawa S, Kiso Y, Sakakibara M. Neurobiol Aging. 2007;28:1179–86. Gorski JA, Balogh SA, Wehner JM, Jones KR. Neuroscience. 2003;121:341–54. Gruer MJ, Artymiuk PJ, Guest JR. Trends Biochem Sci. 1997;22:3–6. Guiard BP, El Mansari M, Blier P. Mol Pharmacol. 2008;74:1463–75. Haile DJ. Am J Med Sci. 1999;318:230–40. Han J, Day JR, Connor JR, Beard JL. Nutr Neurosci. 2003;6:1–10. Harris ZL, Klomp LW, Gitlin JD. Am J Clin Nutr. 1998;67:972S–7S. Henderson BR, Menotti E, Bonnard C, Kuhn LC. J Biol Chem. 1994;269:17481–9. Hill JM, Switzer RC, 3rd. Neuroscience. 1984;11:595–603. Jorgenson LA, Sun M, O’Connor M, Georgieff MK. Hippocampus. 2005;15:1094–1102. Jorgenson LA, Wobken JD, Georgieff MK. Dev Neurosci. 2003;25:412–20. Kaur D, Andersen J. Ageing Res Rev. 2004;3:327–43. Klausner RD, Rouault TA, Harford JB. Cell. 1993;72:19–28. Lameu C, de Camargo AC, Faria M. Recent Pat CNS Drug Discov. 2009;4:137–42. Levy JE, Jin O, Fujiwara Y, Kuo F, Andrews NC. Nat Genet. 1999;21:396–9. Liu M, Xiao DS, Qian ZM. Mol Cell Biochem. 2007;300:139–47. Lozoff B. Food Nutr Bull. 2007;28:S560–71.
77 Changes in Brain Gene Expression in Nutrient Deficiencies
1213
Lozoff B, Beard J, Connor J, Barbara F, Georgieff M, Schallert T. Nutr Rev. 2006;64:S34–43; discussion S72–91. Lozoff B, Brittenham GM. Hematol Oncol Clin North Am. 1987;1:449–64. Lozoff B, Corapci F, Burden MJ, Kaciroti N, Angulo-Barroso R, Sazawal S, Black M. J Nutr. 2007;137:683–9. Lozoff B, Jimenez E, Hagen J, Mollen E, Wolf AW. Pediatrics. 2000;105:E51. Lozoff B, Jimenez E, Wolf AW. N Engl J Med. 1991;325:687–94. MacQueen GM, Ramakrishnan K, Croll SD, Siuciak JA, Yu G, Young LT, Fahnestock M. Behav Neurosci. 2001;115:1145–53. Malecki EA, Devenyi AG, Beard JL, Connor JR. J Neurosci Res. 1999;56:113–22. Meadowcroft MD, Connor JR, Smith MB, Yang QX. J Magn Reson Imaging. 1999;29:997–1007. Monteggia LM, Barrot M, Powell CM, Berton O, Galanis V, Gemelli T, Meuth S, Nagy A, Greene RW, Nestler EJ. Proc Natl Acad Sci U S A. 2004;101:10827–32. Moos T. J Comp Neurol. 1996;375:675–92. Moos T, Morgan EH. J Neurosci Res. 1998;54:486–94. Morris CM, Candy JM, Oakley AE, Bloxham CA, Edwardson JA. Acta Anat (Basel). 1992;144:235–57. Nemeth E, Tuttle MS, Powelson J, Vaughn MB, Donovan A, Ward DM, Ganz T, Kaplan J. Science. 2004;306:2090–3. Ortiz E, Pasquini JM, Thompson K, Felt B, Butkus G, Beard J, Connor JR. J Neurosci Res. 2004;77:681–9. Oshiro S, Kawahara M, Kuroda Y, Zhang C, Cai Y, Kitajima S, Shirao M. Biochim Biophys Acta. 2000;1502:405–14. Patterson SL, Abel T, Deuel TA, Martin KC, Rose JC, Kandel ER. Neuron. 1996;16:1137–45. Pinero D, Jones B, Beard J. J Nutr. 2001;131:311–8. Ponka P. Ann N Y Acad Sci. 2004;1012:267–81. Pratico D, Zhukareva V, Yao Y, Uryu K, Funk CD, Lawson JA, Trojanowski JQ, Lee VM. Am J Pathol. 2004;164:1655–62. Rapoport SI. J Nutr. 2008;138:2515–20. Robb SJ, Connor JR. Brain Res. 1998;788:125–32. Roncagliolo M, Garrido M, Walter T, Peirano P, Lozoff B. Am J Clin Nutr. 1998;68:683–90. Scales WE, Vander AJ, Brown MB, Kluger MJ. J Appl Physiol. 1988;65:1840–6. Schratt GM, Nigh EA, Chen WG, Hu L, Greenberg ME. J Neurosci. 2004;24:7366–77. Siddappa AJ, Rao RB, Wobken JD, Leibold EA, Connor JR, Georgieff MK. J Neurosci Res. 2002;68:761–75. Stead JD, Neal C, Meng F, Wang Y, Evans S, Vazquez DM, Watson SJ, Akil H. J Neurosci. 2006;26:345–53. Todorich B, Pasquini JM, Garcia CI, Paez PM, Connor JR. Glia. 2009;57:467–78. Todorich B, Zhang X, Slagle-Webb B, Seaman WE, Connor JR. J Neurochem. 2008;107:1495–505. Tran PV, Carlson ES, Fretham SJ, Georgieff MK. J Nutr. 2008;138:2495–501. Tran PV, Fretham SJ, Carlson ES, Georgieff MK. Pediatr Res. 2009;65:493–8. Uchida T, Akitsuki T, Kimura H, Tanaka T, Matsuda S, Kariyone S. Blood. 1983;61:799–802. Unger EL, Paul T, Murray-Kolb LE, Felt B, Jones BC, Beard JL. J Nutr. 2007;137:118–24. Unger EL, Earley CJ, Beard JL. J Appl Physiol. 2009;106:187–93. Wallander ML, Zumbrennen KB, Rodansky ES, Romney SJ, Leibold EA. J Biol Chem. 2008;283:23589–98. Wang ZJ, Liang CL, Li GM, Yu CY, Yin M. Chem Biol Interact. 2006;163:207–17. Youdim MB, Ashkenazi R, Ben-Shachar D, Yehuda S. Adv Neurol. 1984;40:159–70. Youdim MB, Ben-Shachar D. Isr J Med Sci. 1987;23:19–25. Yu GS, Steinkirchner TM, Rao GA, Larkin EC. Am J Pathol. 1986;125:620–4.
wwwwwwwwwwwwwww
Part XII
Pathology and Abnormal Aspects: Sensory
wwwwwwwwwwwwwww
Chapter 78
When Taste Triggers Sociophobia Matthieu J. Guitton
Abbreviations 5-HT2C CTA ER ISI mCPP PTSD LiCl WD
Serotonin receptor 2C Conditioned taste aversion Estrogen receptor Interstimulus interval Meta-chlorophenylpiperazine Post-traumatic stress disorder Lithium chloride Water deprivation
78.1 Introduction Taste perception, food ingestion behaviors, and social behaviors are intimately linked together in mammals. The search for comestible food is one of the principal behavioral drives for animals. Transmission of information related to food has a critical importance in strongly socially organized species. Accurate social perception – necessary to maintain hierarchical rankings and satisfying group cohesion – is by essence multimodal. If the influence of smell on interindividual communication is obvious, the involvement of taste-related information in the regulation of interindividual interaction appears somehow less direct. It is however not necessarily less important. In addition, these links are pertinent – and can be actualized – not only in normal physiological situations, but also in pathological ones. Philogenetical and ecological considerations can explain why the gustatory modality deserves a particular treatment, compared to other sensory modalities. Taste and smell are the only two sensory modalities using chemodetection directed toward external stimuli; they thus are the heirs of the most archaic forms of perception. Taste-dependent learning differs from other forms of associative (classical or instrumental) learning in a critical point: the time interval separating the conditioned stimulus M.J. Guitton (*) Centre de Recherche Université Laval Robert-Giffard (CRULRG), 2601 Chemin de la Canardière F-6517, Québec, QC, G1J 2G3, Canada and Faculty of Pharmacy, Laval University, Quebec City, QC, Canada e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_78, © Springer Science+Business Media, LLC 2011
1217
1218
M.J. Guitton
from the unconditioned stimulus (the interstimulus interval or ISI). Whereas in most cases the two stimuli must appear almost together in time in order to be correctly associated, taste-dependent learning tolerates incredibly long ISI, up to several hours. This is due in part to the fact that, in contrast to the vast majority of negative reinforcers, ingestion-induced malaises (which unfortunately represents one of the main ecological outcomes of taste experience) usually require some time to develop. Stimuli that are less closely associated with food (such as auditory, visual, or tactile cues) are not able to support the same long intervals than taste (Garcia et al. 1968). In the context of conditioning, socially oriented behavioral sequences represent conditioned responses much longer than the ones classically observed for other forms of learning. Thus, the very long ISI of taste-dependent learning could support associations with social behaviors. In this text, we will focus on the alteration to social behavior which can be induced by presentation of a taste. We will describe whether presentation of taste can induce direct social withdrawal, how taste presented by a conspecific can socially transmit food-related information, and how taste can reactivate off-line negative emotional states to trigger marked social withdrawal. The implications of these findings for human eating disorders will be extensively discussed, as well as some promising topics such as the influence of sexual hormones and the exploration of alterations in hedonism.
78.2 Food Ingestion and Social Withdrawal The search for food is one of the most basic motivational drives of animal behaviors. Indeed, the capacity of survival of the individual is a direct function of the amount of food which can be found – and successfully ingested. However, one of the inherent problems of ingestive behavior is the risk of absorbing toxic or poisonous food. But luckily enough for animals, toxic food does not necessarily have a lethal effect. In most cases, the ingestion of toxic food only translates into transient malaise, often characterized by visceral pain. If the toxicity is averred, the individual won’t be able to efficiently react to challenging situations. Thus, intoxicated animals should, through their behavior, avoid encountering predators – even more than in normal conditions. One of the best ways not to be captured by a predator is for the individual to avoid being in the same location than other potential preys, i.e., other conspecifics. However, the increased vulnerability to predators does not represent the only risk for poisoned individuals. In the context of social organization, the maintenance of hierarchized ranking is a dynamic process. Animals sickened by inadapted food are likely to be unable to fight to maintain their hierarchical position regarding their subordinates. After ingesting poisonous food, the animal may well survive, but to fully avoid subsequent problems, the process should not be accompanied by a loss of social status within the hierarchy. In the context of social groups based on hierarchical organization, weakened animals should thus not only avoid encountering predators, but also avoid encountering conspecifics (Fig. 78.1). Isolation periods following toxic food ingestion could thus rely on at least two evolutionary justifications: first, directed toward nonconspecific – to avoid predators while unable to defend efficiently – second, directed toward conspecifics – to avoid losing social status in a highly hierarchized context. Worthy of note is that this global social withdrawal following food poisoning is very close to the general sickness-associated social isolation (Bluthé et al. 1991; Dantzer 2001). In the case of unknown food – i.e., food with “new” taste – postingestion social withdrawal may therefore represent a natural predisposition of ecological significance (Fig. 78.1). By actively avoiding unnecessary social contacts, the individual may secure recently acquired food and avoid putative risks linked to postingestion. This direct effect represents a first level of food related taste-triggered social withdrawal in animals.
78 When Taste Triggers Sociophobia
1219
Fig. 78.1 Social withdrawal provoked by poisoning. After ingestion of poisonous or toxic food, the animal (white animal) will experience delayed noxious effects (such as visceral pain), and will adapt its social behavior in order to actively decrease its social interactions with conspecifics (grey animal)
78.3 Socially Transmitted Food Preference The social withdrawal triggered by absorption of toxic food represents a direct effect which can be associated with taste perception. Surprisingly, the effect that tastes can have on social behaviors go much further: in addition to effects on the social behavior of the animal which actually ate the food, passive information carried following food ingestion can trigger massive effect on the social behavior of other conspecifics as well (Galef and Wigmore 1983; Clipperton et al. 2008). Indeed, the taste of food attached to the animal can also carry information which will be socially transmitted to conspecifics (Galef and Wigmore 1983; Valsecchi and Galef 1989). This social transmission of information related to taste takes advantage of the fact that taste and smell can persist long time after their original cause (in this case, the food) disappeared. Once an animal has eaten something, the smell of the food will stay on body parts surrounding the mouth. Actually, even the taste of the ingested food will stay and could be detected if another conspecific would be able to lick these body areas. While saying that may at first glance sound surprising, it is not. Indeed, most mammals do have interactions bringing into play these specific zones. For instance, in the case of rodents, whiskers represent one of the main organs used to investigate the world (Brecht et al. 1997). This also includes social interindividual exploratory behaviors – such as sniffing or grooming the partner (Guitton et al. 2008). There are thus possibilities for direct exchange of taste related information between two animals during the process of social interactions. Generally speaking, alterations of perception can trigger clear modifications of social behavior. For instance, the occurrence of tinnitus – perception of sound in the absence of auditory stimulation – triggers in mice the full spectrum of social alterations
1220
M.J. Guitton
(Guitton 2009). As for other sensory modalities, such links exist between taste and social behavior, as it has been described here above. But, in contrast to other sensory information, taste can be directly transmitted from one individual to another. Due to the possibility of direct transmission of taste information between two individuals after physical contact, a same taste can elicit a second order, indirect effect in an individual which was not at first exposed to the original food. As we mentioned above, one of the inherent problems of ingestive behavior is the risk to consume poisonous or toxic food. A major advantage for animals to live in a social context is the possibility for any given member of the group to use the experience of others. Taste information related to food can be transmitted without having to experience the real ingestion process; in contrast, effects of the food can be inferred by assessing the physical state of the “demonstrator” conspecific (Fig. 78.2). Thus, emerged the mechanism of social acquired food preference referred to as socially transmitted food preference (Galef and Wigmore 1983; Valsecchi and Galef 1989). The logic of this particular form of social learning is based on the principle that if an individual survived after ingesting a given food, then this food should not be so bad to eat (Fig. 78.2). On the other hand, for this mechanism to correctly work, if a newly ingested food is toxic, it is crucial for the poisoned individual not only to avoid taking it again, but also to transmit this information to its conspecifics by not letting them validate the food as putatively good by assessing its presence (“taste”) on him. The obvious ecological advantage of acquiring food preferences from conspecifics is the possibility to somehow “bypass” the danger linked to the potential ingestion of toxic food (Galef and Wigmore 1983; Valsecchi and Galef 1989). Thus, in addition to the two first ecological justifications underlying social withdrawal following the ingestion of toxic food, a third one would be not to transmit to other conspecifics the false information of the relative safety of the ingested food. Social withdrawal observed after the ingestion of toxic
Fig. 78.2 Socially transmitted food preference. (a) When an animal is confronted with unknown food, it will hesitate before eating it (neophobia), since the new food may be potentially dangerous. (b) If a conspecific (white animal) has already eaten this food and “presents” the taste on its whiskers, the animal which was not confronted with the food (grey animal) will display a marked preference for this food when it encounters it. Indeed, the animal will “infer” from the fact that the demonstrator is still alive that the food can be ingested safely. This form of socially acquired knowledge is referred as socially transmitted food preference
78 When Taste Triggers Sociophobia
1221
food can therefore have at least three ecological reasons. As we will discuss below, the combination of these three postulated causes is highly interesting in the context of the ontogeny of food disorders observed in humans. What we just discussed however only represents the most obvious part of taste-triggered social withdrawal. Indeed, in this case, the signals – both the taste (either directly from food, or indirectly via presentation of the taste by a conspecific) and the effects of food ingestion (either direct if the animal ate the food, or indirect if the animal is assessing the state of its conspecific) – are “on-line.” But can taste trigger alterations of social behaviors if stimuli are not present, in other words “off-line” (Guitton et al. 2008)? The next paragraph will attempt to answer this question by examining how taste exposure and association with emotional states can produce long-term effects on social behaviors.
78.4 Taste-Dependent Social Withdrawal Up to now, the effects described here where only “on-line” effects, meaning effects observed when the two stimuli (the taste and the effects of food) are still salient. The relationship between on-line taste information of social behavior is obvious, either in the case of immediate taste distress-provoked social withdrawal with direct taste detection and effect of the food, or in the case of immediate taste-dependent social learning with indirect presentation of the taste via a conspecific and evaluation of the health status of the individual. But in addition to these responses, taste can also trigger massive alterations of social behavior when the reinforcer is not there, i.e., when the original stimuli are “off-line” (Guitton et al. 2008). When an animal is first exposed to a poisonous food, the odds that this animal will ingest this food if presented again are rather low. Indeed, the animal is likely to avoid it (Garcia et al. 1966; Berman and Dudai 2001; Guitton and Dudai 2004). This taste-dependent learning, in which the subject associates a conditioned stimulus (the taste) with a delayed unconditioned stimulus (visceral pain) is called a conditioned taste aversion (CTA, for a classical protocol used to induce CTA in rats, see Fig. 78.3) (Garcia et al. 1966; Bures et al. 1998). Following such a conditioning, the conditioned response will be a massive aversion for the food displaying the same taste. In the case of a classical CTA (while the new taste is associated with clear-cut visceral pain), there is no real need to engram the information necessary to trigger complex social withdrawal next time the taste will be presented, since once the taste will be recognized anew, the corresponding food will not be eaten again. And indeed, if animals present a CTA for a given taste, a subsequent presentation of this taste will not trigger any modification of their social behavior (Guitton et al. 2008). They will clearly avoid the food, but no further social effect will be observed (Fig. 78.4). However, the situation is drastically different if the food did not provoke massive visceral pain, but rather elicited a diffuse and vague emotional state. Anxiety and anxiety-like state result from situations that represent either real or imaginary danger for the individual (Marks 1987; LeDoux 1998). In contrast to visceral pain, anxiety is a rather diffuse internal state. But can anxiety act as unconditioned stimulus in the context of taste-dependent learning? This question was addressed using pharmacological tools to experimentally induce anxiety-like state in rats (Guitton and Dudai 2004). For instance, metachlorophenylpiperazine (mCPP), a serotoninergic agent acting as a 5-HT2C receptor agonist, is known to be able to provoke, both in humans and animals, anxiety-like state when injected (Bilkei-Gorzo et al. 1998; Guitton and Dudai 2004). Delayed anxiety provoked by intraperitoneal injection of mCPP was able to be specifically associated with taste information by rats, and this association was strong enough to be detected by behavioral measurements as a CTA (Guitton and Dudai 2004).
1222
M.J. Guitton
Fig. 78.3 A classical protocol to induce a CTA. Here is exemplified a “classical” protocol which can be used to induce conditioned taste aversion (CTA) in rats. During the time course of the protocol, animals are under water deprivation (WD), meaning that their access to water is restrained to 10 min per day (except the first day, for which the time allowed to access water is 30 min). At Day 4 (Conditioning), animals are allowed to drink a solution with a new taste instead of water for 10 min. About 15–45 min after the offset of the drinking period, animals are exposed to an agent inducing visceral pain (classically, a single intraperitoneal injection of LiCl, 0.15M, 2% of body weight). Three days after the conditioning (Day 7, Test), animals are given free choice between water and the conditioned taste during their 10 min daily drinking period. Conditioned animals are expected to massively display a behavioral preference toward water over the taste associated with visceral pain
Using a combination of anxiety-induced CTA (using mCPP injections as inductor of anxiety-like state) and social interaction measurements, we recently reported that taste-dependent conditioning procedure can trigger a marked impairment in social behavior – corresponding to a massive and active social withdrawal, or a “sociophobia” (Fig. 78.4) – in response to the conditioned taste (Guitton et al. 2008). When compared to control animals (or animals subjected to a classical CTA with visceral pain as unconditioned stimulus), animals treated with mCPP completely diverged concerning their behavioral outcomes when presented anew with the conditioned taste. Animals for which taste was initially paired with delayed anxiety presented a drastic decrease of the social events they displayed compared to any of the other experimental groups (Guitton et al. 2008). The number of behavioral escapes stayed very low for experimental animals whatever the group, except for the animals initially conditioned with mCPP-induced anxiety. In a phenotyping point of view, the modifications observed for the escape responses are of major importance to explain the deterioration of social interactions, since they confirm an active participation of the treated animal. Thus, the decrease of the number of social events was definitively not merely passive, but rather an active phenomenon. Therefore, this decrease of social interactions was really isomorphic to a social withdrawal (Guitton et al. 2008). The abnormal (anti)social behavior displayed by these animals is likely to affect the behavior of their naïve partner. And that is what actually occurred. Indeed, naïve partners strongly increased their following behaviors, mirroring the evolution of the escape behaviors
78 When Taste Triggers Sociophobia
1223
Fig. 78.4 Taste-dependent sociophobia. (a) Consumption of food does usually not induce any alteration of social behavior. (b) If an animal experiences the taste of a new food and is subjected later on to visceral pain (for instance using LiCl injection), the association of these two stimuli will produce a conditioned taste aversion (CTA). Such association will result in a strong avoidance of the conditioned taste (and so, food). However, if the animal eats this food again (which is unlikely to happen, discontinued arrow), the perception of this taste will not induce any modification of social behavior. (c) However, if the animal experiences a new taste followed by a delayed negative emotional state (such as anxiety-like state), the association will be more diffuse and the behavioral outcomes will be totally different. First, the animal may still want to eat the target food. However, after ingesting it, social behavior will be drastically affected. Indeed, the animal will actively avoid social contact (social withdrawal)
observed in mCPP-treated animals. However, since they were receiving less social inputs, naïve partners were ultimately also decreasing slightly their social activity in response to the sociophobic behavior displayed by mCPP-treated animals (Guitton et al. 2008) (Fig. 78.5). This behavioral response observed once the taste was presented again was clearly not a generalized avoidance, since almost no reduction in approach behavior of the spout, or in consumption was observed, translating the fact that anxiety was able to produce only a limited CTA (Guitton and Dudai 2004; Guitton et al. 2008). Furthermore, taste–anxiety associations are not generalized (Fig. 78.6): only the taste on which the association has been made can trigger social withdrawal, other tastes being unable to do so (Guitton et al. 2008). Thus, if anxiety-like states have a relatively low power to induce CTA, this taste–emotion association has remarkable outcomes at the social behavior level (Guitton et al. 2008). The retrieval-induced “sociophobic” state lingers long after a single exposure to the nonreinforced taste. By predicting the unconditioned stimulus (which is in this case a purely emotional state), the gustatory cue appears to trigger a persistent emotional and social response. And that could be happening in the case of human beings. Indeed, taste could get associated with the emergence of a feeling, even unrelated. Since taste-dependent conditioning tolerates long ISI, the emotional feeling could well be associated with the taste. The association of taste with delayed negative internal states could generate conditioned response different from taste aversion, such as in this case active social withdrawal. It is clear that such forms of memory of association may contribute to the ontogenesis, reinforcement, and symptoms of some types of taste- and food-related disorders. The further association of a taste with social withdrawal might represent a conditioned augmentation of the natural predisposition of social withdrawal following ingestion of new food described in the beginning of this text.
1224
M.J. Guitton
Fig. 78.5 Social interactions and social withdrawal. (a) Social interactions are active processes in which each of the two animals exhibits social behavior. In other words, there are interactions because each animal displays and initiates social events directed toward its partner – the events can be elicited either originally, or in response to the previous event emitted by the partner. (b) If one of the animal does not respond to the social stimuli sent by its partner (for instance, in the case of social withdrawal), there are no “interactions.” So, the social dialog evolves toward a monolog. Finally, the partner will lose interest in the social communication with the unresponding animal, and will in turn also decrease the number of social behaviors emitted
Fig. 78.6 Taste specificity in association leading to taste-dependent sociophobia. (a) Associations able to lead to a taste-dependent sociophobia (association between a new taste and a negative emotional state) are not generalized. Only the taste on which was made the association can potentially trigger alterations of social behavior such as social withdrawal. (b) In contrast, if another taste is presented to the animal, no alterations of social behavior will be assessed
78.5 Relevance for Human Eating Disorders As we suggested above, this last effect of taste-triggered social withdrawal, described in animals, has significant implications for the understanding of ontogenesis and maintenance of human food disorders. Even if clearly based on abnormal and in adapted behaviors triggered by altered perception of food or food-related
78 When Taste Triggers Sociophobia
1225
elements (including body image – anorexic patients commonly present distortions in the perception of the size and shape of their own body), the social component of anorexia or bulimia is obvious and cannot be denied (Klein and Walsh 2004). According to current definitions, anorexia could be characterized by social withdrawal triggered by food consumption itself, or by the fact (averred or inferred) that others know that food was ingested. Numerous physiological, psychological, and social alterations are associated with eating disorders, and deciphering which are the factors predisposing or accelerating the apparition of the disorder is far to be an easy task (Klein and Walsh 2004). The ontogenesis of eating disorders is likely to be multifactorial, with massive interactions between internal (biological or psychological) and external (social or environmental) risk factors (Walsh and Devlin 1998; Klein and Walsh 2004). Links between food disorders and alterations of social behavior clearly exist (Kaye et al. 2004; Guitton et al. 2008). Indeed, clinical studies demonstrated that both anorexia and bulimia nervosa present a significant comorbidity with social phobia (Kaye et al. 2004). A study performed during the Second World War on healthy conscientious war objectors showed that starvation provoked in humans increased irritability and diminished social interest (Franklin et al. 1948). Irritability is also one of the psychological symptoms frequently reported by anorexic patients in the acutely underweight state (Kaye 1997). Thus, progressive weight loss facilitates social avoidance. But since decrease of social interactions also oftenly triggers further weight loss, these two elements may well draw a vicious circle. Furthermore, food disorders have a strong co-occurrence with anxiety (Godart et al. 2000; Klein and Walsh 2004), with the onset of anxiety states often preceding the onset of anorexia or bulimia nervosa (Deep et al. 1995; Bulik et al. 1997; Godart et al. 2000). These privileged relations between negative emotions, taste perception, and social behavior may well represent a key element in the processes at the origin of food disorders. In this view, the taste–anxiety–sociophobia association described above may be highly relevant to eating disorders (Guitton et al. 2008). Psychological stress may play some role in the initiation of the disorder. However, it is important to note that despite the high frequency of anxiety and mood alterations, purely antidepressant medications generally fail to help during the acutely underweight state of anorexia (Jones et al. 1991), suggesting that anxiety plays a role of reinforcer rather than being the real origin of the disorder. In the case of bulimia, negative emotions (such as feelings of anxiety, rejection, frustration, or low mood) often precede binges, which seem to become learned responses to negative emotional states (Abraham and Beumont 1982; Heatherton and Baumeoster 1991; Klein and Walsh 2004). A striking characteristic of eating disorders is the high persistence of disordered eating and/or abnormal dieting behavior once it has begun (Klein and Walsh 2004). This long-term persistence also strongly echoes the notion of strong association assessed in the particular form of memory that is a CTA (Bures et al. 1998; Berman and Dudai 2001; Guitton and Dudai 2004; Guitton et al. 2008). Finally, in CTA, the context actualized when the subject tasted the food has also been shown to be able to act as a cue to elicit latter taste aversion (Desmedt et al. 2003), echoing the importance of the contextual environment in food disorders. In conclusion, association between food perception system and social behavior, and particularly the possibility of delayed social withdrawal triggered by taste, may be highly relevant for eating disorders (Table 78.1). An interesting possibility could be to consider anorexia as an epiphenomenon of social disturbance triggered by taste and food perception systems.
78.6 Influence of the Sex In humans, eating disorders clearly affect disproportionately women (Klein and Walsh 2004), with a lifetime prevalence among women estimated at 1–3% for bulimia nervosa and at 0.5–2% for anorexia nervosa (Kendler et al. 1991; Favaro et al. 2003; Klein and Walsh 2004). Cultural reasons
1226
M.J. Guitton
Table 78.1 Key features of taste-dependent sociophobia 1. Social withdrawal occurs when an animal actively decreases the emission of social behaviors directed toward partners. 2. The regulation of social interactions and social ranking is dependent on sensory information. 3. Taste information related to food has clear ecological significance and can induce marked alterations of social behavior. 4. Social withdrawal can be triggered in mammals after association between taste and visceral pain or after retrieval of association between taste and negative emotional state. 5. The understanding of the links between taste and social behavior (especially taste-dependent social withdrawal) has significant implications for the understanding of ontogenesis and maintenance of human food disorders. 6. Eating disorders are associated with numerous physiological, psychological, and social alterations. 7. Anorexia and bulimia nervosa present a significant comorbidity with social phobia. 8. Anorexia may be characterized by social withdrawal triggered by food consumption, or by the fact (averred or inferred) that others know that food was ingested. This table lists the key facts of taste-dependent sociophobia, including basic concept of social withdrawal, and how taste-dependent sociophobia evidenced in animals can be relevant for human food disorders
have often been mentioned to explain this situation. However, when considered with a sociobiological point-of-view – meaning, when considering that social behaviors are grounded by biological mechanisms – these sex-dependent differences observed in humans can find some elements of explanation at the molecular level. Indeed, estrogens are known to be able to modulate social learning. Both estrogen receptors alpha and beta have been demonstrated to be necessary for optimal social recognition in mice (Imwalle et al. 2002; Choleris et al. 2003, 2006). This role of estrogens in social learning is particularly true for the social transmission of food preferences, and has been demonstrated using mice at various phases of the estrous cycle (SánchezAndrade et al. 2005; Clipperton et al. 2008). When tested immediately after socially acquired preference, the preference for the “unknown” food from which the taste was presented by a conspecific lasted longer in females in diestrus or proestrus females, than in females in estrus or ovariectomized (Clipperton et al. 2008). In addition to this effect on acquisition phase and short-term behavioral expression of this taste-dependent social learning, there is also an effect on the maintenance phase. Indeed, when compared to females in estrus or ovariectomized, females in proestrus and diestrus display a prolonged preference for the food targeted by the socially modulated taste-transmission (Clipperton et al. 2008). Finally, only mice in proestus clearly show a massive food preference when tested 24 h after the exposition to the taste through a conspecific having eaten the target food (Sánchez-Andrade et al. 2005). A recent pharmacological study performed on animal model investigated more deeply the molecular basis of this modulation of taste-related social learning by estrogens, by focusing on the effects of selective agonists of the estrogen receptors (ERs) on socially transmitted food preferences. This study unveiled a differential role of ER alpha and ER beta in the acquisition and maintenance of socially transmitted food preference (Clipperton et al. 2008). When treated with the ER alpha selective agonist PPT (1,3,5-tris(4-hydroxyphenyl)-4-propyl-1H-pyrazole), ovariectomized mice are not able to acquire socially transmitted food preference – the alterations of this specific learning being neither due to the quality of interactions, nor to the food (Clipperton et al. 2008). In contrast, ovariectomized mice treated with the ER beta selective agonist WAY-200070 (7-Bromo-2-(4hydroxyphenyl)-1,3-benzoxazol-5-ol) presented a massive increase of the duration of food preference (maybe partially through effects on submissive behavior), with the higher doses being able to trigger preferences similar to those observed in intact female mice during the proestrus and diestrus phases (Clipperton et al. 2008). This suggests the possibility of a balance between the activation of estrogen receptors – with ER beta activation countering ER alpha effects – in the modulation of taste-dependent
78 When Taste Triggers Sociophobia
1227
social learning. Further investigations may contribute to a better understanding of fine mechanisms of sexual regulation of taste-dependent withdrawal in human.
78.7 Perception of Hedonic Valence Among the alterations of taste perception observed in humans suffering from eating disorders, those touching hedonism are highly interesting. In anorexic patients, hedonic self-rated perception of tastes are shifted toward a marked preference for sweetness, while high-fat solutions become aversive (Sunday and Halmi 1990). These alterations of hedonic valence do not seem however to stem from global disturbances of taste perception, since other dimensions of taste perception, such as taste intensity, are generally intact in anorexic patients (Klein and Walsh 2004). The mechanisms underlying these alterations in hedonism are still unknown. Some electrophysiological studies suggested that the brain processing of gustatory stimuli could be altered in anorexic patients (Tóth et al. 2004). Aversion to some particular kind of food (such as high-fat foods) might well reflect a form of conditioned aversion to “perceived” calories. Clearly, the “negative status” of some food is likely to relate at least to some extend to a learned association between the consumption of this food and subsequent disinhibition of eating behavior. Interestingly, the alterations of hedonism perception seem to somehow persist in anorexic patients, even following treatment (Sunday and Halmi 1990). Such, these shifts in hedonic perception could witness predisposing traits, rather than effects of the periods of self-imposed starvation. Anorexic patients may present alterations of the reward value of food (Klein and Walsh 2004). However, food still represents a relevant stimulus for these patients, since food-related cues were demonstrated in neuropsychological experiments to be able to act as distractors (Sackville et al. 1998). Similarly, experiments realized with bulimic patients confirm this view, with cue reactivity demonstrated for taste, but also smell and sight, of typical binge foods (Staiger et al. 2000). Such a role of hedonism seems to be also pertinent for animal models of food disorders. Actually, a recent work even suggests the possibility of a direct behavioral measure of hedonism in rats in the context of food disorders (Guitton et al. 2008). Indeed, once the conditioning has been performed using anxiety as unconditioned stimulus, consumption of beverage with the conditioned taste strongly correlates with the time of social interactions spent by the animal, following a linear regression with a different slope for each taste (Guitton et al. 2008). It has been suggested that the variation of the slope – which reflects at the individual level the relationship between active taste consumption and social behavior – could be a direct indicator of the hedonic valence of the considered taste (Guitton et al. 2008). In rats, food deprivation enhances the hedonic value of nonfood rewards, such as psychostimulants (Cabeza de Vaca and Carr 1998) and intracranial self-stimulation (Carr 2002). This suggests that starvation may have a “reward-potentiating effect,” which may strongly reinforce abnormal behaviors in the case of human eating disorders. Further research is still needed to fully understand this phenomenon, but understanding the mechanisms underlying this process of reward-potentiation may further our knowledge on ontogeny of food disorders.
78.8 Conclusion Either after a direct or indirect association between taste and on-line noxious state, or off-line retrieval of association between taste and negative emotional state, taste presentation can trigger social withdrawal in mammals. Even if clear demonstration of this phenomenon has been obtain on rats, these
1228
M.J. Guitton
properties of association between taste and social behavior have tremendous interest in the understanding of eating disorders in humans (Table 78.1). Modeling the cognitive aspects of eating disorders in animal clearly represents a highly challenging issue – as it is the case for any psychiatric disorder. However, the use of adapted animal models unveils considerable promise as a mean to gain further insight into the molecular and biological mechanisms sustaining these abnormal eating behaviors. Further investigations will of course be needed, but the kind of behavioral approach described in this text on animal models may well contribute to the identification of new molecular targets, and to the development of innovative pharmacological strategies to treat human patients suffering from food disorders. Molecular regulation of the biological systems underlying this phenomenon is still far to be fully understood. However, understanding the fine regulation of the behaviorally observed forms of tastedependent social withdrawal may well be one of the keys to understand the ontogeny of human food disorders.
78.9 Applications to Other Areas of Health and Disease The main applications of research on taste-dependent sociophobia in animals are obviously the understanding of food and eating disorders, particularly anorexia nervosa and bulimia nervosa. However, applications to other areas of health, as well as to other specific diseases should not be neglected. In particular, working on this topic may help to understand some of the mechanisms modulating metabolic disorders such as diabetes. By exploring complex conditioned responses different than the one classically analyzed in CTA, the interactions between taste detection, social behavior, and the association between these two components could help to further the analogies that several researchers have suggested between CTA and post-traumatic stress disorder (PTSD). Given the significance in terms of public health of PTSD, as well as the present lack of pharmacological agents to treat this pathology, the contribution of research performed on sociophobia induced following taste presentation may be crucial in the middle to long term. Rats – like humans – are social animals and display a rich repertoire of social behaviors. In this view, testing social interactions in rats offers a possibility to test high-level behaviors, in addition to the simple aversive behavior classically observed with paradigms such as CTA. Such, investigation of social interactions in rodents (rats or mice) may represent an interesting tool to help further our understanding of pathologies strongly affecting behavior, and particularly psychiatric disorders. Finally, understanding how taste perception can influence social behavior and the maintenance of dynamic social ranking may account for a significant contribution in the research field of multimodal interactions in perception and cognition.
Summary Points • • • • •
Taste perception may have repercussions on social behaviors. Ingestion of poisonous or toxic food induces social withdrawal. Social withdrawal provoked by presentation of a taste is an active phenomenon. Food-related taste information can be transmitted from one animal to another conspecific. The social transmission of taste information between two animals can lead to a subsequent preference for the related taste in the animal which was not initially confronted to it.
78 When Taste Triggers Sociophobia
1229
• The association of a new taste and visceral pain results in a strong avoidance of the conditioned taste; however, if the animal detects this taste again, its perception will not induce modification of social behavior. • In contrast, the association of a new taste and delayed anxiety-like state is more diffuse: the animal may have only a slight avoidance for the taste, but perceiving this taste anew will trigger massive alterations of social behavior, characterized by social withdrawal. • The effects of this last form of association are rather specific to the conditioned taste and are not generalized. Definitions of Key Terms Anxiety-like state: Negative emotional state isomorphic to what is referred as anxiety in humans, the emotional state triggered by the perception of a potential danger (real or inferred). Conditioned taste aversion: A particular form of associative learning between a new taste and an aversive delayed stimulus (usually visceral pain), which leads to a marked subsequent aversion for this taste and tolerates a very long interstimulus interval compared to other forms of associative learning. Estrogen receptors: Endogenous receptors specifically activated by the binding with estrogenic hormones. Neophobia: A form of novelty avoidance which occurs when an animal is confronted to an unknown taste, which reflects the inference of a putative toxicity of the food. Poisonous or toxic food: Food which provokes deleterious effects when ingested by an animal; these effects can be either lethal or nonlethal (commonly provoking in this case transient malaise). Socially transmitted food preference: Food preference acquired through social interactions with a conspecific which already tried this food without experiencing poisoning and display the taste on its whiskers. Socially transmitted food preference represents in rodents a socially acquired knowledge based on taste information. Social withdrawal: Active (not a merely passive phenomenon) adaptation of behavior in order to decrease the occurrence and/or the duration of social interactions with conspecifics. Acknowledgments MJG holds a Career Grant from the “Fonds de la Recherche en Santé du Québec” (FRSQ), and is supported by the Canadian Institutes of Health Research (CIHR – MOP 89699) and the Natural Sciences and Engineering Research Council of Canada (NSERC – grant number 371644).
References Abraham SF, Beumont PJ. Psychol Med. 1982;12:625–35. Berman DE, Dudai Y. Science. 2001;291:2417–9. Bilkei-Gorzo A, Gyertyan I, Levay G. Psychopharmacology. 1998;136:291–8. Bluthé RM, Parnet P, Dantzer R, Kelley KW. Neurosci Res Commun. 1991;9:151–8. Brecht M, Preilowski B, Merzenich MM. Behav Brain Res. 1997;84:81–97. Bulik CM, Sullivan PF, Fear JL, Joyce PR. Acta Psychiatr Scand. 1997;96:101–7. Bures J, Bermudez-Rattoni F, Yamamoto T. In: Bures J, Bermudez-Rattoni F, Yamamoto T, editors. Conditioned taste aversion, memory of a special kind. Oxford: Oxford University Press; 1998. pp. 1–25. Cabeza de Vaca S, Carr KD. J Neurosci. 1998;18:7502–10. Carr KD. Physiol Behav. 2002;76:353–64.
1230
M.J. Guitton
Choleris E, Gustafsson J-Å, Korach KS, Muglia LJ, Pfaff DW, Ogawa S. Proc Natl Acad Sci USA. 2003;100:6192–7. Choleris E, Ogawa S, Kavaliers M, Gustafsson JÅ, Korach KS, Muglia LJ, Pfaff DW. Genes Brain Behav. 2006;5:528–39. Clipperton AE, Spinato JM, Chernets C, Pfaff DW, Choleris E. Neuropsychopharmacology. 33:2362–75. Dantzer R. Brain Behav Immun. 2001;15:7–24. Deep AL, Nagy LM, Weltzin TE, Rao R, Kaye WH. Int J Eat Disord. 1995;1:291–7. Desmedt A, Hazvi S, Dudai Y. J Neurosci. 2003;23:6102–10. Favaro A, Ferrara S, Santonastaso P. Psychosom Med. 2003;65:701–8. Franklin JC, Schiele BC, Brozek J, Key A. J Clin Psychology. 1948;4:28–45. Galef BG Jr, Wigmore SW. Anim Behav. 1983;31:748–58. Garcia J, Ervin FR, Koeling RA. Psychon Sci. 1966;5:121–2. Garcia J, McGowan BK, Erzin FR, Koelling RA. Science. 1968;160:794–5. Godart NT, Flament MF, Lecrubier Y, Jeammet P. Eur Psychiatry. 2000;15:38–45. Guitton MJ, Dudai Y. Biol Psychiatry. 2004;56:901–4. Guitton MJ, Klin Y, Dudai Y. Behav Brain Res. 2008;191:148–52. Guitton MJ. J Psychosom Res. 2009; doi:10.1016/j.jpsychores.2008.10.017 Heatherton TF, Baumeoster RF. Psychol Bull. 1991;110:86–108. Imwalle DB, Scordalakes EM, Rissman EF. Horm Behav. 2002;42:484–91. Jones BP, Duncan CC, Brouwers P, Misrky AF. J Clin Exp Neuropsychol. 1991;13:711–28. Kaye WH. Psychopharmacol Bull. 1997;33:335–44. Kaye WH, Bulik CM, Thornton L, Barbarich N, Masters K. Am J Psychiatry. 2004;161:2215–21. Kendler KS, MacLean C, Neale M, Kessler R, Heath A, Eaves L. Am J Pyschiatry. 1991;148:1627–37. Klein DA, Walsh BT. Physiol Behav. 2004;81:359–74. LeDoux JE. Biol Psychiatry. 1998;44:1229–38. Marks I. Fears, phobias and rituals: panic, anxiety and their disorders. New York: Oxford University Press; 1987. Sackville T, Schotte DE, Touyz SW, Griffiths R, Beumont PJ. Int J Eat Disord. 1998;23:77–82. Sánchez-Andrade G, James BM, Kendrick KM. J Reprod Dev. 2005;51:54–558. Staiger P, Dawe S, McCarthy R. Appetite. 2000;35:27–33. Sunday SR, Halmi KA. Physiol Behav. 1990;48:587–94. Tóth E, Túry F, Gáti A, Weisz J, Kondákor I, Molnár M. Int J Psychophysiol. 2004;52:285–90. Valsecchi P, Galef BG Jr. Int J Comparat Psychol. 1989;2:245–56. Walsh BT, Devlin MJ. Science. 1998;280:1387–90.
Chapter 79
Why Do We Dislike So Many Foods? Understanding Food Aversions Christina L. Scott and Ronald G. Downey
Abbreviations CR CS UCS UCR
Conditioned response Conditioned stimulus Unconditioned stimulus Unconditioned response
79.1 Introduction Like breathing and drinking, eating is critical to our long-term survival. Given this fact, why do we avoid so many edible foods? This chapter will explore the major psychological factors that drive our eating behaviors (see Table 79.1). While there are clear physiological constraints to what we can eat (e.g., allergic reaction to peanuts or eating foods that contain large amounts of salmonella), this chapter will focus on why we avoid foods that we could eat with little danger to our health or survival. We will first focus on four major ways that these aversive behaviors develop (Passer and Smith 2007). These are: • • • •
Classical conditioning Operant conditioning Observational learning Conscious choice
After briefly describing each of the four major ways that aversions to foods can develop, we will: • Provide relevant examples of how aversive behaviors develop and are sustained over long period of time in our daily life, in spite of many efforts to change the behaviors • Explain how our basic senses (e.g., vision, etc.) often drive learning • Discuss how a variety of conscious choices impact out eating behaviors and • Review the current methods that are used to try and change disruptive aversion behaviors
C.L. Scott (*) Whittier College, 13406 E. Philadelphia St. Whittier, CA 90608, USA e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_79, © Springer Science+Business Media, LLC 2011
1231
1232
C.L. Scott and R.G. Downey
Table 79.1 Key features of food aversion 1. Food aversions develop quickly through the pairing of food with very negative events; e.g., foods that are too hot or foods that taste “bad.” 2. Over time, we learn to seek out certain foods when we are reinforced by the positive nature of the food and we avoid those foods that produce a negative response. 3. We model our food preferences by observing other people; e.g., family, friends, etc. 4. Our senses of the smell, sight, feel (texture), and/or taste of food are all driving forces for eating or avoiding food. 5. We often consciously decide to avoid foods to obtain certain goals or avoid bad outcomes; e.g., a healthier lifestyle or manage health conditions. 6. Changing food aversions (or attractions) is enormously difficult. This table lists the key facts of food aversion, including how they develop, interact with our basic senses, and affect our eating behaviors
79.2 Four Models of Food Aversion 79.2.1 Classical Conditioning Classical conditioning is the most basic process of learning (Pavlov 1928). As is shown in Fig. 79.1, all living organisms have an innate response (e.g., approach or avoidance) when they encounter certain stimuli (e.g., heat, light, movement, taste), When babies are fed food that is too hot, they spit it out and cry. Thus, the association between the heat (unconditioned stimulus, UCS) and pain (unconditioned response, UCR) is established by the baby. Problems can occur when the baby pairs heat (UCS) and other properties of the food together (conditioned stimulus, CS) with the pain producing a conditioned response (CR). For example, if the food was strained carrots, the taste (or color, or texture, etc.) of carrots is now associated with heat/pain. When the baby eats carrots (hot or cold) in the future, the conditioned response is to spit them out.
79.2.2 Operant Conditioning Operant conditioning is a more complex leaning process where the stimuli and responses are more complex, develop over a much more extended time, and involve reinforcement (positive or negative). Figure 79.2 outlines the basic process. Through operant conditioning we learn to effectively operate within a complex and ever changing environment through reinforcement (Skinner 1938). Simply put, when we are reinforced for a behavior, we are more likely to repeat the action, but when we are punished, we are less likely to repeat the behavior in the future. When offered a candy bar, a child is immediately reinforced with the sweet taste of milk chocolate and seeks to repeat the experience; however a bitter taste (a negative experience) of dark chocolate may serve as a disincentive and thus discourages the child from trying that food again.
79.2.3 Observational Learning Observational learning occurs when we perceive and mirror the behaviors of others. No immediate reinforcement of the new behavior is required, but the perception exists that the modeled behavior will be rewarded in the future, either through social acceptance or direct reinforcement (Bandura 1969). A child who observes his father pushing away his plate with uneaten broccoli is more likely
79 Why Do We Dislike So Many Foods? Understanding Food Aversions
1233
Before Conditioning Carrots
No fear, no anxiety
During Conditioning Carrots
+
Fear, anxiety, and avoidance
Hot food Burns tongue.
After Conditioning Carrots
Fear, anxiety, and avoidance
Fig. 79.1 Classical conditioning of food aversion. This figure illustrates how food aversions can develop by associating a negative event with a specific food
to emulate his father’s behavior and push aside his/her own vegetables (see Fig. 79.3). This aversion to broccoli has not developed as a dislike for any specific properties of broccoli, but rather a desire to model the behavior of the father. Social learning plays a very powerful role in our eating behaviors, whether it is a toddler avoiding vegetables, or a teenager trying diet soda to “fit in” with the crowd. Social learning transcends a wide variety of situations and foods.
79.2.4 Conscious Choice Finally, our eating behaviors can be directly influenced by our own judgments and decisions. Conscious choice influences us when we decide to avoid a food based on our own reasoning and when coupled with our past experience, and the outcomes we have witnessed in others (see Fig. 79.4). Suppose a teenage girl has a date planned for the following weekend and decides she wants to lose weight. Past
1234
C.L. Scott and R.G. Downey
Fig. 79.2 Operant c onditioning of food aversion. This figure illustrates how reinforcement creates a food preference and punishment leads to a food aversion.
REINFORCEMENT PROCESS
BEHAVIOR
Positive
Response occurs
Stimulus is presented
CONSEQUENCE
Response increases
RESULT
(milk chocolate is consumed)
(sweet taste of milk Chocolate is enjoyed)
(more milk chocolate is eaten)
PUNISHMENT
Punishment
Child observes father avoiding broccoli
Response occurs
Aversive stimulus Response is presented increases
(dark chocolate is consumed)
(bitter taste is delivered)
Child wants to model father
(dark chocolate is avoided)
Child avoids broccoli to model father
Fig. 79.3 Observational learning of food aversion. This figure illustrates how we avoid specific foods by observing other people Wants to lose weight for upcoming date.
Recognizes that fatty foods will cause weight gain.
Makes conscious choice to avoid fatty foods.
Fig. 79.4 Conscious choice model of food aversion. This figure illustrates how we consciously process information about our food choices
experience suggests that certain foods (e.g., fatty foods) may cause weight gain. Thus, the teenager makes a conscious choice to avoid fatty foods and opts for a garden salad. When choosing between foods we like, many of our food choices are determined by conscious decisions to select one food over another; sometimes even if we do not “like” the food we have chosen.
79.2.5 Summary of Learning Models Thus, our responses to food are complex and determined by a variety of past events. Some of the responses (see Table 79.2) are very basic and almost instinctive (classical leaning). Others are acquired over a long time and require repeated experiences (operant learning). Equally complex
79 Why Do We Dislike So Many Foods? Understanding Food Aversions
1235
Table 79.2 Four models of food aversion. This figure outlines the four basic learning models associated with the development of food aversion
What behavior is required?
How does learning occur?
What is the sequence of events? What is the outcome?
Accidently Eat a piece of Based on eat hot food chocolate. watching other (UCS) and people, the pair it with same the specific behaviors are taste of replicated. carrots (CS). Pairing of the Reinforcement Observe UCS and CS encourages behavior and leading to a future behavior motivated to conditioned and punishment replicate it. response. discourages future behavior. CS occurs Reward Observation before the (punishment) occurs before CR. occurs after the the behavior. behavior. Strong Strong positive Strong negative or negative positive or reaction to reaction to negative specific foods foods due to response to due to expected specific foods unconscious outcome. based on prior pairing. observation.
A decision is made to avoid a specific food.
Conscious motivation to avoid specific foods.
Conscious decision is made, behavior follows. Strong positive or negative response to specific foods based on a series of thought processes and decisions.
are the ways we learn from other individuals within our immediate environment (observational learning). Finally, as we grow and mature we make choices about what we do and do not want to eat and these choices may not be directly related to the sensory properties of the food. In the next sections we will discuss in detail how each of these processes can lead to potentially dysfunctional eating behaviors.
79.3 Sensory Processes and Eating As will be outlined in the next sections, our sensory processes unconsciously determine many of our reactions. These include sight (appearance), smell, taste, and feel. While these will be discussed as separate senses, there is ample evidence (Martins and Pliner 2005) that they can be closely interconnected. For example, it is difficult for people to determine when taste and smell are different.
79.3.1 Smell A food’s aroma is one of the strongest signals we evaluate when facing an unfamiliar dish (Halpern 2002). If the scent “reminds” us of a similar food that we like, such as chicken or steak, we are more likely to respond favorably to the new dish. However, if the scent is strong and
1236
C.L. Scott and R.G. Downey
unrecognizable, such as sauerkraut or fish, we tend to reject the new food without trying it. One of the strongest components of food aversion is the overlap between the way a food smells and its taste. If you’ve ever tried eating your favorite foods with a head cold, you quickly realized that the flavor of the food is diminished (Beauchamp and Bartohshuk 1997). Children, who are instructed to finish their vegetables, often plug their noses, to reduce the smell and, they assume, flavor of the offending vegetable. Memories of unpleasant smells in childhood are likely to follow us into adulthood and smells that remind us of disliked foods (such as Brussels sprouts) will be immediately rejected.
79.3.2 Sight (Appearance) Go to a five star restaurant and you will be dazzled by the presentation of the food you are served, because “presentation is everything.” Imagine two chicken dishes are placed before you; one features a plump, juicy, white chicken breast surrounded by colorful vegetables and a mound of mashed potatoes with a dollop of butter balanced on top; while the second dish features a dry, thin, graying chicken breast, next to a pile of green beans, and a baked potato wrapped in foil. The classical conditioning model suggests that we are immediately drawn to the more attractive meal because we associate good taste with attractive presentation that leads to satisfaction in our food choice. However, it is important to realize that our appreciation for the appearance of food is not innate. One look at the contents of baby food will remind us that as infants, our food preferences were mostly based on taste (and temperature) and not esthetics. However, as we grow older, we learn to value the appearance of our food. Toddlers find the round shape of Cheerios easy to grasp, which is reinforcing as they learn to manage finger foods, and in time, they will prefer whole crackers and cookies over broken pieces, even if the taste is identical.
79.3.3 Taste From the time we are babies, we respond to food positively or negatively based on taste (Halpern 2002). The human tongue has approximately 9,000 taste buds that are devoted to recognizing the five elements of taste perception: sweet, sour, bitter, salty, and savory (Halpern 2002). Sweet tastes such as sugar are immediately reinforcing, while strong or bitter flavored foods such as cabbage and grapefruit are initially disliked. College students learn to consume large amounts of coffee by masking the bitter taste with sugar. Eventually many college students will desensitize to the bitter taste and they may require less sugar because the actual or perceived positive cognitive effect of the caffeine has become a stronger reinforcement.
79.3.4 Feel (Texture and Temperature) As we demonstrated (Scott and Downey 2007) texture is strongly associated with food aversions and choice. On a hot summer’s day nothing is more appealing that the cool creamy consistency of ice cream as it cools a burning throat. Although the flavor of the ice cream is important, the texture and
79 Why Do We Dislike So Many Foods? Understanding Food Aversions
1237
temperature are equally essential in our enjoyment. Melting ice cream loses its texture and although the taste remains the same, we may reject it. Foods that are not served at their ideal temperature are immediately judged as being less appealing. Cold coffee and warm white wine are immediately rejected as being unfit for human consumption. Tepid vegetables often remained untouched on a plate, as compared with the same vegetables served steaming hot. Anyone who has indulged in McDonald’s French fries will attest to the fact that “fresh French fries” far outweigh the taste of lukewarm ones! Foods with smooth even textures signal us at an unconscious level that what we are consuming is pure and consistent. Imagine eating a cup of yogurt or enjoying a fruit smoothie. Lumps or chunks produce an aversive response and we evaluate them as evidence that our food hasn’t been blended properly. From our earliest experiences with food, we were served well-blended baby food, which featured a smooth and even consistency. We come to expect a certain texture from our food, such as creamy mashed potatoes and if we are served something outside of our expectations, we may find it aversive. A classic example would be the texture of soggy bread. As children we learn the ideal texture for bread by watching our parents. If it becomes too firm (stale) or too soggy (wet) we respond negatively and leave the bread untouched. Slimy and/or slippery foods such as oysters, mussels, and internal organs (e.g., liver) feature an unfamiliar and initially unappealing texture that results in most people avoiding these foods without trying them.
79.3.5 Summary of Sensory Processes and Eating Our five senses play a critical role in our food preferences and aversions. While a large number of food choices, including avoidance of certain foods or types of foods can be traced back to sensory types of processes, a large number are related to the human processes of choice and decision making. The underlying processes determining these choices and decisions are complex and multidetermined. The next section will explore some of the major ways we decide to avoid certain foods. However, it should be noted that these decisions may be unconsciously associated with some of the sensory properties of the foods.
79.4 A pplications of Food Aversions to Other Areas of Health and Wellness While many food aversions are unconscious reactions to something unappealing, unfamiliar, or unpleasant, a large portion of our food choices are made based on conscious choice (see Table 79.3). We may be operating based on previous negative experience or observing the behaviors of others, but the outcome is the same, we avoid certain foods for the following reasons: • • • • • •
Questionable or unsafe foods Temperature Media warnings Dietary restrictions (health) Dieting and diet fads Going organic: organic foods, vegetarians, and vegans budget
1238
C.L. Scott and R.G. Downey Table 79.3 Applications of food aversions to other areas of health and wellness. This figure outlines how we make a decision to avoid specific foods in order to achieve a healthier lifestyle Questionable Temperature or unsafe foods
Media Dietary warnings restrictions (health)
Dieting & diet fats
Going organic
Budget
What is the thought process?
That food doesn’t look/smell familiar!
That It’s unsafe temperature or isn’t unhealthy appealing. to eat that!
Eating that That food Artificially That food may is too grown specific put my fattening. foods food is too health at aren’t as expensive. risk. healthy as organic foods.
Food example
Oysters
Warm Beer Corn syrup
Glutens or sugar
What is the outcome?
Avoidance
Temporary Temporary avoidance avoidance
Temporary Temporary Temporary Temporary avoidance avoidance avoidance avoidance
Pizza
Produce
Lobster
79.4.1 Questionable or Unsafe Foods At the most basic level, we may choose to avoid specific foods because they appear questionable or unsafe, even though we have enjoyed similar foods in the past. Most of us have sampled milk that has begun to spoil or been taken aback at the green fuzzy mold growing on our cheese. Although we enjoy these foods on a regular basis, we detect that the foods may be “spoiled” or “unsafe” to consume and we avoid them for that one instance. The same might hold true if we are served an undercooked steak (although some people ask for rare steak) or we are served raw fish (however, many people like Sushi). This tendency may be enhanced by our previous experiences with undercooked foods, such as a case of food poisoning, but we might be equally reluctant to eat an undercooked piece of chicken if we saw a recent news report cautioning the dangers of salmonella or E Coli contamination. The texture of a food may enhance this concern. If we faced with slippery or slimy foods such as runny eggs or raw oysters, we may be more reluctant to try them than firmer textures such as steak or chicken.
79.4.2 Temperature Although certain foods seem to taste better when served hot, such as fajitas, French fries or filet mignon, other foods, such as ice cream and shrimp seem to taste better cold. Many of these preferences for temperature are based on culture and develop through social learning. In the United States, we prefer our milk and dairy products to be ice cold, while in Europe consuming milk and yogurt served at room temperature is commonplace and milk straight from the cow is desired in many cultures (Rappoport 2003). Imagine you are served a bowl of leek soup and it arrives cold. Although this recipe is both valid and delicious, we may find the cold soup aversive and turn it away, because we expect soup to be served hot. In the Hofbrauhaus in Germany, warm beer is served intentionally to enhance the flavor of the beverage; however many American tourists would be offended if their beer was not served chilled: a conscious choice.
79 Why Do We Dislike So Many Foods? Understanding Food Aversions
1239
79.4.3 Media Warnings Although many food aversions develop through direct experience (classical or operant conditioning) or direct observation (social learning), many food aversions are a conscious choice based on media warnings. With recent warnings about transfat, MSG, and other additives/preservatives in food, the public is more health conscious than ever. In the 1980s a media campaign warning about the dangers of high cholesterol suggested that consumption of eggs could cause serious health risks. Consumers made a conscious choice to cut back on their egg consumption in order to be health conscious and new food avoidance behavior was created. In the late 1990s, consumers were surprised to see a series of new commercials with a very different message advertising “The incredible, edible eggs!” as an excellent source of protein. For many consumers, the decision to avoid or enjoy eggs became a combination of the conflicting messages they received and a conscious choice based their own personal preference for eggs. In 2008, commercials speaking out against products containing corn syrup surfaced featuring concerned mothers protecting their children from foods made from corn syrup. No specific claims were made about the possible health risks of corn syrup, but the commercials clearly suggested that giving your child a Popsicle containing unnatural chemicals such as red dye and corn syrup was an irresponsible decision. These commercials were followed almost immediately by a rebuttal campaign, using the same social learning technique featuring a mother correcting another woman’s misconceptions about corn syrup arguing that it is produced naturally from corn and in moderation is “the same as sugar.” As consumers, we have to determine which foods contain high fructose corn syrup (read the label) and make a conscious choice whether we want to give them up based on the media warnings, or decide to just enjoy them.
79.4.4 Dietary Restrictions (Health) For people diagnosed with Diabetes, Celiac Disorder, or Food Allergies, the conscious choice to avoid specific foods is a necessary decision to protect overall health and wellness. Diabetics may watch the high fructose corn syrup debate with interest, but ultimately realize that any form of sugar, for their well being, will need to be monitored carefully. They may actively seek out foods with less sugar or sugar substitutes, such as Splenda or Equal, while restricting their intake of foods high in carbohydrates or sugars. There may be instances when they make a conscious choice to not restrict their dietary selections, which could result in higher blood sugar levels, and possible medical complications. Food allergies can range from mild to severe and most appear in early childhood. Children and adults with Celiac Disorder are allergic to Glutens and therefore need to adhere to a strict GlutenFree diet, which means no foods made with wheat. This would systematically eliminate most breads, pastas, and pizza from their diet, unless they were made with special Gluten-free ingredients. Wanting to enjoy a piece of cake at a classmate’s birthday party or a slice of pizza with friends afterschool becomes increasingly more frustrating for children with Celiac Disorder as they may associate specific foods with “fitting in” and social acceptance. The conscious choice to avoid specific foods can be a matter of life and death. Individuals with peanut or chocolate allergies can find themselves suffering from migraines to severe anaphylactic shock depending on the level of their exposure and the severity of their allergies.
1240
C.L. Scott and R.G. Downey
79.4.5 Dieting & Diet Fads While we may be drawn to “comfort foods” such as sweets or starchy foods (Conneret al. 1988), many of us make a conscious choice for a limited period of time to avoid specific foods in order to lose weight. Freshmen college students, who have packed on the “freshmen fifteen” over the course of their first semester of college, may suddenly restrict their fried food intake and late night pizza binges in order to wear swimsuits over spring break. Although most dieticians suggest that “lifelong changes” are necessary to lose weight and maintain the weight loss, most Americans “crash diet” (Crowet al. 2006),which means a drastic restriction in calorie intake, combined with a long list of foods that must be avoided for quick weight loss. The promise of quick weight loss has led to an ever-increasing number of “fad diets” such as the SouthBeach and Atkins diets, which suggest a lowcarbohydrate and high-protein regimen. Although both programs promise immediate results, most people prefer carbohydrates and find it difficult to avoid them for a prolonged period of time. Although most of us recognize that sugary foods are higher in calories, we don’t want to give up the sweet taste we enjoy, so we make compromises such as switching to diet sodas or using sugar substitute products (Appleton and Conner 2001). We want the benefits of a lower calorie lifestyle, but are unwilling to sacrifice our preference for sweet tastes. A recent trend in wellness has been a detoxifying procedure, where people will consume only fruit juice or green tea; some people make the conscious choice to avoid all foods for 3–10 days in order to clear the impurities from their system (Glittleman 2006).
79.4.6 Going Organic: Organic Foods, Vegetarians, and Vegans With the recent warnings about food chemicals leading to cancer, many Americans are turning to organic products. The decision to avoid foods grown with pesticides, growth hormones, or preservatives is becoming increasingly more popular, especially with parents seeking to protect their children (Burke 2007). Specialized grocery stores such as Whole Foods and Trader Joe’s offer a variety of organic options, including dairy products and produce and most large grocery chains now have organic food sections. The only known downside to organic products is their higher cost, which means the availability of organic foods is restricted to those who can afford higher grocery bills. The decision to avoid meat (vegetarians) or the decision to avoid all foods that contain either meat or dairy products (vegans) can be made for a variety of reasons. Some people may wish to avoid chemicals and preservatives and opt for organic fruits and vegetables, while others may wish to make a statement about the treatment of animals raised for food. The recent bestseller, The Omnivore’s Dilemma (Pollan 2006), highlights the overcrowding and mistreatment of cattle raised for consumption, and led thousands to swear off meat in support of animal rights. Many such individuals report that although they enjoy the taste of a good steak, they are making a conscious choice not to support an industry that mistreats animals.
79.4.7 Budget Although we may enjoy the “surf and turf” option at a five star restaurant, our budgets may restrict the foods we choose to eat. As any college student can tell you, the conscious choice to eat Top
79 Why Do We Dislike So Many Foods? Understanding Food Aversions
1241
Ramen, is not one made for taste preference, but generally due to the low cost. Dollar menus have become increasingly more common at fast food restaurants, including McDonald’s, Taco Bell, and Burger King. Recognizing the need for lower priced offerings, establishments such as Applebee’s and Chili’s have started offering $7 menu items, or combination dinner specials which feature a shared appetizer, two entrees, and a dessert for a lower price. Senior menus and children’s menus are often coveted for not only their smaller portions, but specifically their lower prices. Limited budgets required conscious choices to find lower priced food.
79.4.8 Summary of Conscious Choice A conscious choice to avoid a specific food is made after considering a wide range of evidence. We may already have a negative association/experience with a specific food or we may we evaluating the health risks associated with a food we have enjoyed for many years. Our choice may be permanent, such as making a lifestyle change to promote better health or it may be temporary because we watching our budget or restricting our calories. We are constantly evaluating new evidence regarding our food choices and making conscious choices about the foods we consume.
79.5 Changing Food Aversion Behaviors As most of us can attest, change is not easy. The older we are, the more likely we are to be set in our ways, which is why changes in food aversions are best introduced in infancy or early childhood (Chatoor and Ganiban 2003). Such changes are therefore most commonly introduced by parents and not based on the child’s desire to overcome a food aversion. The most common methods of changing food habits in children include food pairings (classical conditioning), rewards for eating specific foods (operant conditioning), and food blending (masking foods within liked foods) and can be seen in Table 79.4. Havermans and Jansen (2006) suggested a classical conditioning approach for introducing vegetables to young children. Parents paired a positive flavor (sweet taste of sugar) with a neutral flavor (vegetables) in order to help children develop a positive response to vegetables. It is important to note that the children in the study saw the vegetables as a neutral stimulus and had not developed a strong aversion to any of the vegetables they were asked to consume. The stronger and more long-term someone’s aversion to a specific food is, the more difficult it is to override the initial negative response, no matter how strong the pairing. Brunstrom and Fletcher (2008) sought to explore the effects of flavor–flavor pairings with young adults (18–25 years), but discovered that the pairings were only successful when the participants were hungry, and therefore more motivated to eat. Another approach to changing food aversions is through rewards, which utilizes an operant conditioning model. Hendy et al. (2005) rewarded elementary school children for eating fruits and vegetables in the cafeteria, by giving them tokens for each food consumed. Children exchanged tokens for small prizes and 2 weeks after the study, food preferences for fruits and vegetables increased. However, the effect was not long-lasting and when children were sampled seven months after the study, their interest in fruits and vegetables had returned to the baseline. Parents can encourage their children to eat less appealing foods through reinforcement; however the change is generally only
1242
C.L. Scott and R.G. Downey Table 79.4 Changing food aversion behaviors. This figure reviews the structure, process, and limitations of the four primary methods for changing food aversion behaviors Classical conditioning Positive pairing is required. Sweet taste (UCS) is paired with vegetables (CS).
Operant conditioning Reinforcement of desired behaviors is required. Receive rewards for eating vegetables.
How does learning occur?
Pairing of the UCS and CS leading to a new conditioned response.
Reinforcement encourages future behavior and punishment discourages future behavior.
What is the sequence of events?
CS occurs before the CR.
Reward occurs after the behavior.
What is the outcome?
Limited positive outcome if initiated early in infancy or childhood.
No permanent change, behavior terminates without reward.
What behavior is required?
Masking Small amounts of a new food are hidden within an enjoyed food. (e.g., putting sweet potatoes in pancake batter). Remember behavior and have motivation to replicate it.
Lewin’s change theory model Repeated exposure, time and commitment to adapt to a new food behavior.
Combination of classical, operant and observational learning, but increased conscious motivation to avoid or approach specific foods. Increased Three stage process: exposure to Unfreezing –override new food aversion hidden in Change–consume preferred food. new food Freezing–adopt food into permanent lifestyle. High level of Limited psychological distress positive involved with limited outcome if positive outcome. initiated early in infancy or Tendency to revert childhood. back to original behavior.
temporary. When the reinforcement is withdrawn, it is likely that children return to their original eating behaviors and the new foods are abandoned. Another suggested technique for parents to change children’s choices involves masking less appealing foods with preferred foods (Mueller et al. 2004). Recipes such as sweet potato pancakes, allow parents to “sneak in” a serving of vegetables into an already accepted food. In order to promote healthier eating habits, parents may switch to wheat bread when serving peanut butter and jelly sandwiches, or incorporate whole grain tortillas when serving burritos. Occasionally these pairings are successful with adults, such as the acceptance of green beans, when they are buried deep within a casserole, consisting of a sauce of mushroom soup and fried onions. However, most adults are fully conscious that undesirable foods, such as vegetables are being blended into their food preferences. For both adults and children, if we “know” a specific food is “in there” we may be unable to convince ourselves to try the new food, especially when strong aversions are present. As we discussed previously, we may decide to make a conscious choice to change our food behaviors, but change does not come quickly or easily. In 1947, Kurt Lewin introduced his
79 Why Do We Dislike So Many Foods? Understanding Food Aversions
1243
“change theory” which specifically identified three major stages necessary for a change in food habits to occur. The initial step involved changing a person’s current mindset, and bypassing their defense mechanisms; a process he called “unfreezing.” Suppose a person has a strong aversion for tofu, but has recently starting dating a vegetarian and is trying to make a change in his food habits. Suppose he has promised to give up red meat in exchange for tofu burgers. Unfreezing would involve trying to override the initial aversion to tofu, including the texture, taste, and smell. Lewin’s second stage was the “change” where the individual would begin eating tofu, even if their initial response was negative. Lewin notes that this stage requires a substantial amount of time because the new food is still unfamiliar and the person may find himself tempted to return to old habits, such as eating red meat. The final stage is “freezing” in which the new mindset is locked in place and a person’s comfort level is restored once the new food has been accepted. Lewin’s model would argue that with time, tofu would become more familiar and a part of the individual’s eating routine. Lewin’s model requires repeated exposure, time, and a commitment to changing a specific food habit before a new behavior is adopted and the process is not without emotional consequences. Lewin notes that anxiety and discomfort are hallmarks of the second stage “change” when we are learning to accept unfamiliar foods. Although we may make a conscious choice to change our eating patterns, all three stages are limited by the very fact that food aversions often exist at a basic level. We know vegetables are good for us and that tofu is a health alternative to red meat, but we “just don’t like the taste, texture, and/or smell” and whether it is food blending, food pairing, or trying to adopt a new food habit altogether, our aversions persist. Most adults can’t articulate why they avoid specific foods, but despite our best efforts to reverse them, the long-term changes are minimal.
79.5.1 Summary Changing Food Aversion Behaviors The older we are the more resistant we become to changing our food aversions. Although we may believe that our eating behaviors are predominantly a conscious choice, there are many unconscious influences that make certain foods unpalatable to us (see Table 79.4). Masking the flavor of disliked foods, such as putting cheese and butter on Brussel sprouts is only effective if we have a mild aversion. Unlike young children, it may not matter to us as adults how the food is masked or rewarded; we are still unlikely to eat it.
Overall Summary It is almost always impossible to pinpoint the exact moment when food aversions develop for individuals. As we have discussed they can develop through a variety of learning and conscious choices. These include: classical conditioning, operant conditioning, and observational learning. Many aversions are driven by our sensory processes of smell, sight, taste, and texture. We can also make decisions to avoid specific foods whether they are for health, diet, or cost. Once we have developed these avoidance (or approach) behaviors, changing them is enormously challenging. Attempts to modify our food choices typically have not resulted in long-term changes in our eating behaviors. In essence, we are “what we eat” because our food preferences have become an ingrained and long-term part of who we are as unique individuals.
1244
C.L. Scott and R.G. Downey
Definitions Celiac disorder: A chronic nutritional disturbance, usually of young children, caused by the inability to metabolize gluten. The disorder can be controlled by a special diet that emphasizes the elimination of all foods containing gluten. Classical conditioning: A neutral stimulus (the conditioned stimulus) is repeatedly presented in association with a neutral stimulus (the unconditioned stimulus) that elicits a natural response (the unconditioned response) until the neutral stimulus alone elicits the same response (now called the conditioned response). Glutens: The tough, viscid, nitrogenous substance remaining when the flour of wheat or other grain is washed to remove the starch. Diabetes: A medical condition in which the body is unable to control the level of sugar in the blood. E. coli: A dangerous form of Escherichia coli, a bacterium that normally lives in the human colon. Observational learning: Behavior occurs as a function of observing, retaining, and replicating behavior observed in others. Omnivore: An organism that eats both plants and animals Operant conditioning: A behavior is either reinforced or punished, which directly affects the likelihood of the behavior being repeated again. Organic: Raised or conducted without the use of drugs, hormones, or synthetic chemicals. Vegan: A person who does not eat meat, fish, or any animal products such as cheese, butter, etc. Vegetarian: A person who does not eat meat or fish.
Acknowledgment We wish to thank Heather Heintz and Angelberto Cortez, Jr. for their assistance in designing the tables for this manuscript.
References Appleton KM, Conner MT. Appetite. 2001;37:225–230. Bandura A. New York: Holt, Rinehart, & Winston; 1969. Beauchamp GK, Bartohshuk L (editors). Philadelphia: Academic Press; 1997. Brunstrom JM, Fletcher HZ. Physiol. Behav. 2008;93:13–19. Burke C. New York: Marlowe; 2007. Chatoor I, Ganiban J. Cognitive Behav Practice. 2003;10:138–46. Conner M, Haddon A, Pickering E, Booth D. J Appl Psychol. 1988;73(2):275–80. Crow S, Eisenberg M, Story M, Neumark-Sztainer D. J Adoles Health. 2006;38:569–74. Glittleman AL New York: Doubleday Broadway; 2006. Halpern B. Taste. In: Pashler H, Yantis S, editors. 3rd ed. New York: Wiley; 2002. Havermans RC, Jansen A Appetite. 2006;48:259–62. Hendy HM, Williams KE, Camise TS. Appetite. 2005;45:250–63. Lewin K. Washington, DC: National Research Council; 1943. pp. 35–65. Martins Y, Pliner P. Appetite. 2005;45:214–24. Muller MM, Piazza CC, Patel MR, Kelly ME, Pruett A. J Appl Behav Anal. 2004;37:159–70. Passer MW, Smith RE. New York: McGraw-Hill; 2007. Pavlov IP. New York: International Publishers; 1928. Pollan M. New York: Penguin Press; 2006. Rappoport L. Ontario: ECW Press; 2003. Scott CL, Downey RG. J Psychol: Interdisciplinary Appl. 2007;141(2):127–34. Skinner BF. New York: Aplleton-Century-Crofts; 1938.
Chapter 80
Influence of Cognitive Biases in Visual Evaluation of Food Amount in Patients Affected by Eating Disorders Piergiuseppe Vinai
Abbreviations AN AW BED BMI BN FARS FFQ FPPB IPSAS PSMA
Anorexia nervosa Adjustable wedge Binge eating disorder Body mass index Bulimia nervosa Food amount rating scale Food frequency questionnaire Food portion photograph book Interactive portion size assessment system Portion size measurement aids
80.1 Introduction Accurate portion estimation is the key to controlling the amount of food ingested, a key element in the treatment of obesity and eating disorders (Dohm and Striegel-Moore et al. 2002), but it is a common clinical experience to find under-reporting of food intake among obese subjects and over-reporting among patients affected by anorexia nervosa (AN). Blundell (2000) provocatively affirmed that asking nutritionists to report information regarding their extramarital sexual encounters there would probably be a significant under-reporting, while asking on their donations to charity there would probably be an over-reporting. Certainly, social acceptability contributes to food reporting but many other factors influence food amount estimation. Current levels of hunger/satiety (Beasley et al. 2004) and food palatability play a role in the evaluation as do the subjects’ gender (Yuhas et al. 1989; Mac Diarmid and Blundell 1998), age (Frobisher and Maxwell 2003), and Body Mass Index (BMI) (Klesges et al. 1995). Children are less able to evaluate than adults; and women and overweight/obese subjects underreport their food intake more than men and normal-weight patients do.
P. Vinai (*) “GNOSIS” No Profit Research Group, V Langhe 64, 12060, Magliano Alpi, CN, Italy e-mail:
[email protected];
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_80, © Springer Science+Business Media, LLC 2011
1245
1246
P. Vinai
Table 80.1 Factors influencing food amount evaluation among general population and subjects affected by eating disorders • Cognitive biases • Social acceptability • Perceptive deficits • Hunger and satiety • Emotional states • Reward sensibility • Food palatability • Eating disorders • Subjects’ age Many factors influence food amount evaluation; among them cognitive biases and emotional states play a pre-eminent role in subjects affected by eating disorders. Moreover, perceptive biases affect the evaluation of subjects not affected by any eating disorder
In this chapter we will discuss how perceptive, cognitive, emotional, and social factors influence the evaluation of the amount of food consumed by patients affected by eating disorders. Moreover, the efficacy of proposed tools to improve the subjects’ ability to evaluate their own food intake will be evaluated (Table 80.1).
80.2 Food Is Always Too Much. Food Amount Evaluation of AN Patients Frequently, patients affected by AN claim to be unable to evaluate the amount of food served, engaging the therapists in endless Socratic debates regarding their real food intake, but the hypothesis that they really misperceive a food amount has been poorly studied. More than 20 years ago, Yellowlees et al. (1988) surprisingly noted that in the many studies on AN only a minimal part was focused on food perception and little attempt was made to explain why they eat in such an unusual manner. So, they performed the first study regarding food perception in a sample of twenty patients affected by AN. Five items of food and four neutral objects of a similar size were video recorded. In the first experiment the objects were alternately placed in the middle of a dummy television screen, near a real TV screen. Each subject could compare the size of the real object with its image on the screen. The subjects were asked to manipulate the size of the object on the video screen until each image was, in their opinion, as close as possible to the size of the real object. No significant difference was found between AN patients and the control group in their ability to measure size. Both groups did, however, noticeably exaggerate the size of food, but patients with anorexia nervosa perceived it as 12% bigger overall than the control group members. The results did not change when the objects were placed in the dummy television screen for 10 s and then removed. The subjects were subsequently asked to adjust the video image from what they remembered. To our knowledge no further research regarding food amount evaluation in patients affected by AN has been done in the subsequent 20 years. All of the studies on the perceptive abilities of these patients focused on their own body shape and weight estimation (Epstein et al. 2001). Their body size misperception seems to be more based on cognitive factors than by a perceptive bias (Skrzypek et al. 2001). Given that there is a tendency to a top-down process when judging food (Urdapilleta et al. 2005) and Stroop interference for food-related words (Overduin et al. 1995), cognitive factors could probably also influence their food amount evaluation. To test their ability in estimating food amount, thus avoiding these interferences, our research group (Vinai et al. 2007) carried out
80 Influence of Cognitive Biases in Visual Evaluation of Food Amount
1247
a study in an experimental situation without any relation to food intake. Patients were informed that there was no effect on the therapy induced by their answer and particular attention was paid to avoid emotional and social influences on evaluation. We showed them subsequently a dish containing 27 high caloric and tasty yellow candies (A stimulus) and 27 plastic yellow LEGO® bricks of the same size and shape as the candies (B stimulus), placed in the same position on the plate to minimize the influence of shape, size, and disposition of the objects on number perception. Moreover, to avoid misperception through hunger on the evaluation, all the subjects had breakfast before being tested. No significant difference between AN patients and the control group in evaluating the number of the two stimuli has been found (both significantly underestimated the number of presented stimuli): 78% of AN patients and 78.6% of control subjects underestimated the number of candies, while 18.7% of the AN group and 21.4% of the control group overestimated them. Two AN patients (3.4%) and no control subjects guessed the correct number of candies. We found no correlation between the evaluation of number and the age of onset, or the duration of the illness, or the current or minimum BMI of the patients. The plastic bricks were evaluated in the same way as the candies (Fig. 80.1). These results are apparently in contrast with the common clinical finding of overestimation of food intake among AN patients and with those of the study of Yellowlees et al. (1988), but the discrepancy can be explained by the differences between the experimental settings. In real life, as in the Yellowlees’ experiment, the evaluation is not only the result of a flow of sensory information from periphery to brain, but also involves a top-down processing of a selection of inputs considered relevant by the subjects. In our study these processes have been minimized. Summarizing, AN patients do not seem to be affected by a real perceptive deficit; they share the tendency of the general population to underestimate food amount. Top-down processes in
Anorexic patients' evaluation
underestimation correct evaluation
18,7 3,4
overestimation
78
21,4 0
Control subjects evaluation
78,6%
Fig. 80.1 Estimation of a number of tasty candies among patients affected by Anorexia Nervosa (AN) and control subjects. Seventy-eight percent of AN patients underestimate the number of tasty candies presented to them: 18.7% overestimated them and 3.4% correctly guessed the number. Patients affected by AN share the tendency to underestimate the amount of food with the general population. There are no significant differences between AN patients and control subjects in this evaluation. There is a general tendency among the normal weight population to underestimate the amount of food presented
1248
P. Vinai
judging food amounts seem to play a dominant role among them; e.g., when asked to categorize aliments, they cluster food on the basis of the consequences of ingestion (in terms of health, digestion, and weight gain), rather than on visual characteristics, as normal weight subjects do (Urdapilleta et al. 2005).
80.3 S urely I Have Eaten Too Much! Evaluation of Food in Patients Affected by Bulimia Nervosa While there is a multitude of research about the eating behavior of patients affected by bulimia nervosa (BN), few studies focus on bulimics’ perceptions of their peculiar eating behavior. Several findings led to the hypothesis that a distortion of what is a normal meal may represent a central cognitive mechanism of the disorder. Bulimic patients overrate the amount of food presented to them. This effect increases as caloric intake or actual amount increases at a greater rate than control subjects do (Gleaves et al. 1993). Overestimation of food amount could induce a more restrictive diet – the major risk factor for the onset of bulimic behavior. Moreover, the evaluation of caloric content of a food is subject to bias among these patients, worsening their tendency to reduce food intake. Their rating of the amount eaten is predicted by the estimate of the actual amount eaten, by the kind of food, and by mood level before eating (Keel et al. 2001). In a study including 165 females, those with bulimic behaviors and restrained eating were more likely to overestimate caloric content of the food presented (Stanton and Tips 1990). Bulimic patients (Williamson et al. 1991) report overeating at a higher rate than control subjects, as they are affected by a bias distorting their perception of the amount of food eaten. Such cognitive biases could be the effect of an excessive concern about eating and dieting. As excessive weight is a threatening situation for these patients, they have interpretational biases toward foods that could induce weight gain (Williamson et al. 1991; Gleaves et al. 1993). Caloric intake during binging episodes vary widely (Telch et al. 1990) and patients base their perceptions of having binged more on the type of food consumed than on the amount of food eaten (Rosen et al. 1985). Bulimic individuals make cognitive attributions about different foods on a continuum from safe to forbidden, i.e., those high in calories and fat. The evaluation of an eaten food as forbidden, despite its caloric value, can trigger a binge episode in these patients and consumption of even small amounts of it may lead subjects to perceive their consumption as overeating (Herman and Polivy 1984; Rosen et al. 1985; Kales 1991). It has been reported (Kales 1990) that 69% of the episodes they evaluate as a binge contained forbidden foods, i.e., having a higher caloric density, whereas only 15% of the episodes defined as nonbinge contained them (Table 80.2). Knight and Boland (1989) asked a group of women to consume either cottage cheese (nonforbidden food) or a chocolate milkshake (forbidden food) preload before tasting ice cream, informing participants that both foods contained the same number of calories. Highly restrained participants subsequently ate more ice cream when having first eaten the forbidden milkshake than they did after eating the cottage cheese. (Guertin and Conger 1999). Furthermore, these patients have perceptual distortions of their own body image in response to eating forbidden foods (Rosen et al. 1985) (Fig. 80.2). Summarizing, subjective perceptions of overeating and binging are core factors in the onset of bulimic behavior; the perception of having overeaten and the self-reporting of having binged, rather than actual overeating, lead to disinhibition and binging (Polivy and Herman 1985). Given the lack of studies on the possibility of improving the ability of patients affected by BN to evaluate their food intake, further researches on this topic are needed.
80 Influence of Cognitive Biases in Visual Evaluation of Food Amount
1249
Table 80.2 Key features of cognitive distortion regarding food intake among bulimic patients • It is perceived as overeating, despite its caloric value. • It can trigger a binge episode despite its caloric value. • Most episodes they evaluate as binge eating contain forbidden foods. • When they eat forbidden foods they have perceptual distortions of their own body image. This table shows the effect of eating a forbidden food on patients affected by Bulimia Nervosa and its relationship with their eating behavior
I’m too fat !
I have broken my diet
Today it’s useless to restrain my food intak e
I have eaten two chocolates To taste chocolates is a binge , I’ll vomit to avoid weight gain.
so I can continue to binge
Fig. 80.2 Cognitive and behavioral consequences of the intake of a forbidden food. The picture shows the worry induced by forbidden food intake in a patient affected by Bulimia Nervosa. Cognitive biases regarding their own body image, the amount of eaten food, and the consequences of its intake frequently induce the patients to overeat and to vomit
80.4 P robably I Didn’t Eat So Much. Visual Evaluation of Food in Patients Affected by Binge Eating Disorder Patients affected by Binge Eating Disorder (BED) frequently fail in evaluating the quantity of food when introducing large quantities of food in a short time, but it is not yet clear whether this is due to either a perceptive or a cognitive bias. Greeno et al. (1999) tested whether BED patients have an altered perception of food. They asked obese women either affected or not by BED to report their typical and largest-ever portions of foods. Moreover, they had to report the minimum amount of each of eight foods they considered a binge. Neither evidence of cognitive distortions among BED patients, nor significant differences between the groups have been found. However, binge eaters reported that their typical and largest-ever portions of foods were larger in a very significant way, confirming the need for focus on the global eating behaviors of these patients even in the absence of perceptual distortions.
1250
P. Vinai
80.5 E yes Are Bigger Than Stomach. Perceptive and Social Factors Influencing Evaluation of Food Amount in the General Population The difficulty in evaluating the amount of a presented food is not a peculiar characteristic of eating disordered patients. Normal-weight subjects’ ability to describe an amount of food they have just chosen, without the aid of measuring devices, is poor (Guthrie 1984) and both overweight and normal-weight subjects fail in estimating food portions (Blake et al. 1989). Many perceptive biases explain these results. It has been highlighted by Ginsburg in studies on general population since 1978 that the arrangement of the stimuli influences the evaluation of their amount. Using circular lighting dots he found that regularly arranged stimuli are overestimated while those randomly disposed are underestimated; these results have been confirmed using stimuli of different shapes (Ginsburg 1980). Usually food is randomly disposed into the dish so it gets underestimated. The area collectively occupied by dots also provides a basis for judging their numerousness: the larger the occupancy pattern, the higher the number estimated (Allik and Tuulmets 1991). Moreover, item size influences the evaluation: there is an inverse relation between item dimension and estimation of numerousness (Ginsburg and Nicholls 1988). In addition, form and energy density of food may affect its assessment; people have more difficulties in estimating the correct serving size for amorphous foods (e.g., apple sauce) compared to solid foods (Yuhas et al. 1989; Ayala 2006) (Table 80.3). In a recent study our research group (Vinai et al. 2008) investigated the influence of socioeconomic factors and food palatability on food amount evaluation among children. We tested 94 ten- to fifteen-year old children living in Mali in western Africa, and 124 subjects comparable in age and gender in northern Italy. They were asked to evaluate an amount of palatable candies. Both Italian and Malian children significantly underestimated them. This characteristic could represent a protective factor against food shortage, but in regions with food abundance it could play a role in the onset and maintenance of an obesity epidemic (Fig. 80.3). Subsequently we asked the same subjects to evaluate an amount of altered food (toffees dyed with ink and ammonia so as not to be palatable), and found a significant difference between the groups. African children do not underestimate the number of toffees presented to them while Italian children continued to underestimate them as palatable ones. Food amount evaluation seems to be asymmetrically regulated. Food shortage influences the evaluation of nonedible food but current food abundance has no effect on the evaluation of palatable food. There is no downregulation caused by food abundance in evaluating the amount of tasty food. We can hypothesize that the few generations in which there has been food abundance in Italy are not enough to modify the tendency to underestimate a palatable food; a characteristic developed over centuries of shortage, while they are enough to reduce the emotional impact of seeing wasted food (Fig. 80.4). Table 80.3 Factors inducing perceptive biases in food amount evaluation among general population • Disposition (randomly/regularly) • Dimensions • Area occupied • Form (solid/amorphous) • Energy density • Food disposability When food is randomly disposed it is frequently underestimated, such as when it is cut in small pieces, or when occupies a small area. Subjects fail more frequently in evaluation of the amount of amorphous food and of aliments with a high energy density
80 Influence of Cognitive Biases in Visual Evaluation of Food Amount Fig. 80.3 Evaluation of the number of tasty candies in different cultural contexts. Children living in different cultural contexts (Mali in Africa/Italy in Europe) with different food availability share a tendency to underestimate the number of tasty candies presented
100
1251
Tasty candies 83
80
76
60 40 23
20
15 1
2
0 under estimation
Fig. 80.4 Evaluation of the number of not edible candies in different cultural contexts. There are significant differences in evaluating an amount of altered, so no longer edible, candies between children living in different cultural contexts with different food availability. The 90% of the children living in Italy underestimate the number of candies. Only the 47% of the children living in Mali in Western Africa underestimate them
100
90
correct estimation
overe stimation
Not edible candies
80
Children living in Mali (Africa) Children living in Italy
60
52 47
40
20
10 1
0
underestimation
0
correct estimation
overestimation
Summarizing cognitive and perceptive biases influence food amount evaluation among the general population inducing underestimation of food amount. In a context of food abundance they can represent a risk factor for overeating.
80.6 A pplications to Other Areas of Health and Disease, Improving Food Amount Evaluation to Deal with an Obesity Epidemic It is a common experience among people living in Europe to be surprised by the size of the portions as they enter a restaurant in the USA. The repeated exposure to extra-large portion sizes could have fostered the normalization of a high-energy intake. What was seen as an enormous portion 25 years ago may have down-sized to an ‘average-sized’ portion in the eyes of today’s consumers (Dohm et al. 2005).
1252
P. Vinai
Table 80.4 Factors influencing food amount estimation among obese subjects • Sensitivity to reward • Large portions size • Body Mass Index • Preference for fatty foods • Underestimation of portion size Among factors influencing the food amount evaluation of obese subjects, the tendency to consider enormous amounts of food as normal, and the high caloric concentration of food seem to play a pre-eminent role
Mistakes in evaluating food intake have been reported by the Special Supplemental Food Program for Women, Infants, and Children (Webb and Yuhas 1988) and by several research programs focusing on obese subjects affected by type II diabetes (Rapp et al. 1986) or currently dieting (Howat et al. 1994). Obese persons generally under-report food intake and fail in estimating food quantity and calories more than normal-weight subjects do (Bandini et al. 1990); such mistakes are a barrier to the control of food intake (Davis et al. 2007). The largest miscalculation in reported caloric intake seems to be due to errors in portion-size estimation (Beasley et al. 2005), but it is still an open question whether obese patients actually have a perceptive bias needing specific training to accurately selfreport their food intake during a weight loss program (Lansky and Brownell 1982), or intentionally under-report their food intake for social desirability reasons (Muhlheim et al. 1998). Moreover, they make great mistakes in converting food quantity to actual caloric values; indeed foods with the highest caloric density had the highest error level (Lansky and Brownell 1982). Among obese women (Davis et al. 2007) an interesting interaction between obesity and sensitivity to reward on food amount ratings was found: high reward sensitivity was associated with the underestimation of food amounts. The authors hypothesize that a subject highly attuned to the rewarding properties of food will tend to have a biased perception of food amounts. In other words, larger portion sizes may be judged more normal by those who are easily rewarded in comparison to subjects with a more nonhedonic reaction to food. They also found that women with a strong preference for high-fat foods were more likely to underestimate the size of portions. Reward sensitivity is positively associated with BMI and predisposes individuals to overeating (Loxton and Dawe 2001; Davis et al. 2004). Subjects’ BMI seems to be a strong predictor of larger than recommended amounts of food. Individuals with a higher BMI may view a larger portion as typical and thus eat significantly larger portions of food (Burger et al. 2007). A strong preference for high-fat foods predisposes subjects to choose fast-food restaurants where portion sizes are typically larger than meals cooked at home (Briefel and Johnson 2004). Moreover, larger portion sizes tend to be underestimated more than smaller ones (Harnack et al. 2004) (Table 80.4). The knowledge regarding food amount evaluation can be used in assessments of meal size and caloric intake of obese subjects and to develop psycho-educational programs to heighten awareness about correct portion sizes and to teach strategies for improving abilities to estimate food amounts (Davis et al. 2007) (Fig. 80.5).
80.7 I nstruments to Improve Ability in Food Amount Evaluation Among Adults Food diaries are widely used for dietary survey. One of the main problems of their use is to establish the size of consumed portions and the caloric value of food. Since 1984, Marilyn and Zegman found that 68% of a sample of 43 obese female subjects failed in evaluating calories using the Handbook 456, a
80 Influence of Cognitive Biases in Visual Evaluation of Food Amount
1253
Studies reporting biases in the evaluation of food amount 14
14
perceptive biases
12
cognitive biases
10
number of studies
9
8
8
6 4 1
1 0
0 Binge Eating Disorder
1
Bulimia Nervosa
2
0
0
0 normalweight subjects
obese subjects
Anorexia Nervosa
Fig. 80.5 Perceptive and cognitive biases in the evaluation of food amount. The graph reports the number of studies regarding food amount evaluation found in Medline® and their results. In the graph, the studies evaluating perceptive and cognitive biases are reported among patients affected by eating disorders, obesity, and normal-weight subjects. Several studies found perceptive deficits in the general population. Very few studies focus on food amount evaluation in BED and AN patients. Cognitive biases seem to play a primary role among patients affected by BN
small text showing caloric value of commonly used aliments. In the same year Guthrie (1984) suggested the use of tools to help patients in assessment of portion sizes. Given the importance of food amount evaluation in the assessment and therapy of ED and overweight patients, an objective measure that presents standard stimuli to assess an individual’s judgment of food amounts is needed (Table 80.5). Many instruments have been developed to improve subjects’ ability in evaluating their own food intake. Portion Size Measurement Aids (PSMA) include two dimensional models: drawings of real foods, abstract shapes, household measures, food photographs, computer graphics, and package labels; and three-dimensional tools such as household measures, real samples, and models of food. Kircaldy et al. (1989) did not find any significant difference by comparing the accuracy of portion size estimates obtained with four different PSMA: color pictures of foods, abstract shapes, black and white drawings, preweighted amount of food. Their results were subsequently confirmed (Posner et al. 1992) comparing the effect of two-dimensional and three-dimensional models on evaluation. PSMA have been used separately and together but the validity of this tool is not yet clear (Cypei et al. 1997; Matheson et al. 2002). The most common dietary assessment tool used in large epidemiologic studies of diet and health are the Food Frequency Questionnaires (FFQ); self-administered booklets asking participants to report the frequency of consumption and portion size of many foods over a period of time (Willet 2002). They contain sets of pictures of different portion sizes of foods. Each line item is defined by a series of foods or beverages (Table 80.6).
1254
P. Vinai
Table 80.5 Instruments to improve ability in food amount evaluation among adults • Portion Size Measurement Aids (PSMA) • Food Frequency Questionnaires (FFQ) • Adjustable Wedge • Food Portion Photograph Book (FPPB) • Food Amount Rating Scale (FARS) • Interactive Portion Size Assessment System (IPSAS) Several instruments have been developed to improve subjects’ ability in evaluating their food intake; among them photographs of food and a food amount rating scale seem to be the most effective. Among children, IPSAS and food photograph books adapted for children seem to be more effective than the instruments used for adults
Table 80.6 Key features of Food Frequency Questionnaires (FFQ) FFQs ask participants to report the frequency of consumption and portion size of a series of foods and beverages (usually 100–150 items) over a defined period of time, e.g., the last month. Several questions on food preparation methods enable the researchers to further refine nutrient calculations. Food composition tables listing the average nutrient content of listed foods have been developed to estimate nutrient intake from SFFQ. Food Frequency Questionnaires (FFQ) are self-administered booklets, commonly used in epidemiologic studies, developed with the purpose of obtaining a measure of usual diet
Kuehneman (1994) found no difference in the accuracy of food amount evaluation between subjects using PSMA and FFQ. An Adjustable Wedge (AW) was proposed to assess wedge-shaped foods (e.g, pie, cake, and pizza) and tested by Godwin in 2006, among 320 subjects of both sexes, ranging from 18 to 65 years, using multiple sizes and types of wedge-shaped foods. The evaluation made with the AW was more accurate (P < 0.05) than the ruler in approximately one third of comparisons, but regardless of the aid used, some people have difficulty in estimating portions of wedge-shaped foods. Photographs of food are useful instruments which are also widely used among people of different cultures, but despite the apparent simplicity of these tools, three skills are involved in their use: perception, memory, and cognitive functions. The subject has to be able to perceive the amount of eaten food, to remember it, to make a mental construction of an amount which is not currently present, and to relate it to the photograph. Any impairment in each of these skills can reduce the efficacy of the instruments. Regarding their efficacy, Nelson et al. (1994) found that photographs of small portions of food are overestimated while large ones are underestimated. To assess individuals’ estimation of food amounts Dohm and Striegel-Moore (2002) developed the Food Amount Rating Scale (FARS). It asks the subjects to rate 24 different portions of food described in the questionnaire (e.g., 1 bowl spaghetti, ½ cup tomato sauce, 1 bowl tossed green salad, 1/8 cup salad dressing, 2 pieces garlic bread) by checking via five definitions: small amount, moderate amount, large amount, enormous amount, beyond enormous, the definition which best describes each portion.
80.8 Food Amount Evaluation Among Children Children also fail in estimating food amount and portion size and several research programmes tested the efficacy of the above described tools to improve their food amount evaluation. Matheson et al. (2002) evaluated errors associated with children’s portion size estimates by means of 2 food portion measurement aids: the standard 2-dimensional food portion visuals and manipulative props.
80 Influence of Cognitive Biases in Visual Evaluation of Food Amount
1255
Errors in quantitative estimates of gram weight of foods and energy intakes were large with both methods. Frobisher and Maxwell (2003) compared the ability of adults and children in evaluating portion size using food photographs and descriptions of portions. Forty-seven adults and 37 children were asked to describe a portion of nine food items using the terms small, medium, and large and to choose the photo in a photographic food atlas which best represented the portion size they had just served themselves. All subjects were asked to recall their food intake 3–4 days after the meal. There were very few differences in the estimation of portion sizes between the two testing periods but among children there was greater error than among adults. Children overestimate portions probably because they usually have smaller portion sizes than those depicted in the smallest portion size photograph in the food atlas. The findings suggested that for young subjects it was necessary to modify the tools for estimating portion sizes. Lillegaard et al. (2005) researching whether age influences the ability to estimate food portion sizes by viewing photographs of food did not find any age-related difference, but portion size was estimated more accurately when the actual served portion of food had exactly the same appearance as the foods on the photograph booklet. Foster et al. (2006) found that children’s estimates of portion sizes were significantly more accurate (an underestimate of 1% on average) using age-appropriate food photographs than photographs designed for use with adults (an overestimate of 45% on average). Accuracy of children’s estimates of portion size using ageappropriate tools was similar to that of adults; children overestimated a food weight by 18% on average and adults underestimated by 5%. In conclusion, providing children with food photographs depicting age-appropriate portion sizes greatly increases the accuracy of portion size estimates compared with estimates using photographs designed for use with adults. New technologies have recently been used to help children’s evaluation of the quantity of food. The Interactive Portion Size Assessment System (IPSAS) is a computer-based system to allow the child or interviewer to scroll through images of food depicting increasing portion size. These images are photographs of real foods that are used to indicate the portion served and any food left over. The system automatically records the portion size selected and stores this and related data, including the subjects’ details. This data can easily be exported to a database or statistical software. Foster et al. (2008, 2009) assessed the accuracy of children’s estimates of portion size using food photographs, food models, and the IPSAS. Significant differences were found between the accuracy of estimates using the three tools. Children of all ages performed better using the IPSAS and food photographs than when using food models. The authors suggest that IPSAS has potential for the assessment of dietary intake with children. Before its practical application it needs to be expanded to cover a wider range of foods and to be validated in a “real-life” situation. In conclusion, color food photography and the atlas seem to be useful tools for quantifying food portion size but new instruments are needed to improve food amount evaluation among children.
80.9 Transcultural Studies Several studies assessed the utility of PSMA in different cultural contexts. Venter et al. (2000) developed the Food Portion Photograph Book (FPPB) in Southern Africa. Commonly eaten foods, preparation methods, recipes, and portion sizes were collected in a pilot study; then color photographs taken of foods prepared by the researchers and measured into three or four portion sizes were put together in the FPPB. One hundred and sixty-nine adult African volunteers were tested by presenting them with a portion of real food and asking them to estimate its size by matching it with one of the photographs. Of 2,959 portions tested, 68% were accurately estimated. The estimation was not affected by gender, age, or education of the participants.
1256
P. Vinai
In Italy, Turconi et al. (2005) developed a color photographs food atlas by weighing and taking digital photographs of three portion sizes of 434 foods and beverages typical of the Italian diet. They tested 448 volunteers ranging from 6 to 60 years old from a wide variety of social backgrounds, assessing 9,075 food portions eaten at lunch and dinner in relation to a set of color food photographs during 8 weeks of investigation. The results show that weights of portion sizes chosen from the set of photographs are significantly associated to weights of eaten portions and are regardless of age, gender, and BMI. Huybregts et al. (2008) assessed the utility of photography in food recall after 24 h among women of Burkina Faso. Each photo album contains four photos for each item. Subjects were shown two previously weighed foods every morning, and another in the afternoon. The day after, the subjects were shown the pictures and had to choose the representative food of the day before. The use of images was effective for the estimation of food.
80.10 Conclusion In conclusion, our findings suggest that assessment of what constitutes a binge or overeating, and judgment of food amounts in general, may be influenced by an individual’s gender, ethnicity, dietary restraint status, level of depressive symptomatology, and current weight (Fig. 80.6).
10
Results of the studies evaluating food amount estimation 9
8
9
underestimation correct evaluation overestimation
6 Number of studies 4
3 2
2 1
1
0
0
0
0
0 normalweight subjects
obese subjects
Binge Eating Disorder
Bulimia Nervosa
Anorexia Nervosa
Fig. 80.6 Errors in food amount evaluation. The graph reports the number of studies regarding food amount evaluation found in Medline® and their results. There is a tendency to underestimate food amounts among the general population, obese subjects, and patients affected by BED. The studies involving bulimic subjects found an overestimation of food amount and the few researches regarding AN patients had controversial results
80 Influence of Cognitive Biases in Visual Evaluation of Food Amount
1257
Researchers studying issues related to dietary intake (e.g., obesity) or disordered eating (e.g., binge eating) need to account and check for the possible confounding effects of these characteristics on self-reports of food intake. Furthermore, using this information to educate study participants and treatment clients about how their perceptions may be biased could improve future food reporting (Dohm et al. 2005). Future investigations should include an income measure to account for the effects of socioeconomic status on portion-size estimation. In addition, we suggest that interventions for preventing weight gain in young adults should focus on the importance of portion sizes for all foods, regardless of macronutrient content, and on increasing the awareness of eating habits in response to media messages and product packaging (Burger et al. 2007). It is noteworthy that training improves ability in estimating food. Yuhas et al. (1989) tested two groups of students in a nutrition course, dividing them into two groups: only one group (76 subjects) received training on estimating food quantities. The training consisted of 10-min sessions in which subjects passed around and viewed food models labeled with their portion sizes while one of the researchers verbally indicated the quantities. There were two solid foods (meat loaf and fish), two liquids (milk and soup), and two amorphous items (spaghetti and apple sauce). Immediately after, subjects individually estimated portion sizes of one food on a display. The second group of subjects, who had received no training, also estimated the same foods on the same displays. Training improved estimations and women tended to estimate more accurately than men. These results suggest the importance of using specific training in order for ED and obese subjects to evaluate their food intake better.
Summary Points • Perceptive and cognitive biases influence food amount evaluation in the general population and among patients with eating disorders. • The general population tends to underestimate food amounts. • Patients affected by Anorexia Nervosa are not affected by perceptive deficit – it is cognitive and emotional factors which influence their evaluation of food amount. • distortion of what is a normal meal represents a central cognitive mechanism of Bulimia Nervosa. • Bulimic patients over-rate the amount of food presented to them; this can induce a more restrictive diet, a risk factor for the onset of bulimic behavior. • Consumption of small amounts of forbidden food may lead bulimic patients to perceive their consumption as a binge. • The perception of having overeaten rather than actual overeating leads to binging in BN. • No evidence of perceptive distortion among BED patients has been found. • It is possible to improve the subjects’ ability in evaluating food amount. • Food photographs are useful tools to train subjects in estimating their food intake.
Key Terms Anorexia nervosa: An eating disorder characterized by markedly reduced weight and aversion to food. Binge eating disorder: An eating disorder characterized by episodes of over-eating with a sensation of loss of control, not followed by purging behaviors. Body mass index (BMI): An index for relating a subject’s body weight to his/her height. It represents the subject’s weight in kilograms divided by his/her height in meters squared.
1258
P. Vinai
Bulimia nervosa: An eating disorder characterized by episodes of binge-eating followed by purging behaviors, such as self-induced vomiting, abuse of laxatives or diuretics, or excessive physical exercise. Cognitive biases: The tendency to make errors in judgment based on the processing of information, applying knowledge and changing preferences; it includes errors in statistical judgment, social attribution, and memory. Downregulation of food: The process by which a subject underestimates a food amount in response to an external variable such as food abundance. Food amount rating scale (FARS): A multiple choice questionnaire which asks for a rating by checking as: small, moderate, large, enormous, and beyond enormous – the amount of 24 different portions of food described in it. Food frequency questionnaires (FFQ): Booklets asking participants to report the frequency of consumption and portion size of many foods over a period of time. Malian: People living in Mali in Western Africa. Perceptive biases: Mistakes in evaluation due to a sensorial deficit. Portion size measurement aids (PSMA): Tools useful for improving accuracy in food amount evaluation. They include household measures, drawings, abstract shapes, photographs, package labels, real samples, and models of food. Stroop interference: When a word such as green, red, blue etc. is printed in a color differing from the color expressed by the word’s semantic value (e.g. the word green printed in red ink), naming the color of the word takes longer and is more prone to errors than when the meaning of the word is congruent with its ink color. Patients affected by Anorexia Nervosa show interference on the Stroop color naming task for food. Type II diabetes: An illness characterized by high blood glucose due to the inability of body cells to respond appropriately to insulin, a hormone produced by the pancreas to reduce blood glucose levels. It is distinguished from type I diabetes due to reduced production of insulin by the pancreas.
References Allik J, Tuulmets T. Percept Psychophys. 1991;49(4):303–14. Ayala GX. Health Educ Res. 2006;21:133–45. Bandini LG, Schoeller DA, Cyr HN, Dietz WH. Am J Clin Nutr. 1990;52:421–5. Beasley LJ, Hackett AF, Maxwell SM, Stevenson L. J Hum Nutr Diet. 2004;17:219–25. Beasley J, Riley WT, Jean-Mary J. Nutrition. 2005;21:672–7. Blake AJ, Guthrie HA, Smiciklas-Wright H. J Am Diet Assoc. 1989;89:962–4. Blundell JE. Am J Clin Nutr. 2000;71:3–5. Briefel RR, Johnson CL. Annu Rev Nutr. 2004;24:401–31. Brunstrom JM, Rogers PJ, Pothos EM, Calitri R, Tapper K. Appetite. 2008;51:296–301. Burger KS, Kern M, Coleman KJ. J Am Diet Assoc. 2007;107:611–8. Cypei YS, Guenthet PM, Petot GJ. J Am Diet Assoc. 1997;97:289–92. Davis C, Levitan RD, Muglia P, Bewell C, Kennedy JL. Obes Res. 2004;12:929–35. Davis C, Curtis C, Tweed S, Patte K. Eat Behav. 2007;8:170–6. Dohm FA, Cachelin FM, Striegel-Moore RH. Obes Res. 2005;13:1061–9. Dohm FA, Striegel-Moore RH. Obes Res. 2002;10:1173–9. Epstein J, Wiseman CV, Sunday SR, Klapper F, Alkalay L, Halmi KA.. Eat Weight Disord. 2001;6:140−7. Foster E, Matthews JN, Nelson M, Harris JM, Mathers JC, Adamson AJ. Public Health Nutr. 2006;9:509–14. Foster E, Mattheews JN, Lloyd J, Marshall L, Mathers JC, Nelson M, Barton KL, Wrieden WL, Cornelissen P, Harris J, Adamson AJ. Br J Nutr. 2008;99:175–84.Foster E, Adamson AJ, Anderson AS, Barton KL, Wrieden WL. Eur J Clin Nutr. 2009;63:45–9.
80 Influence of Cognitive Biases in Visual Evaluation of Food Amount
1259
Frobisher C, Maxwell SM. J Hum Nutr Diet. 2003;16:181–8. Ginsburg N. J Gen Psychol. 1980;103:211–6. Ginsburg N. Am J Psychol. 1978;91:267–73. Ginsburg N, Nicholls A. Percept Mot Skills. 1988;67:656–8. Gleaves DH, Williamson DA, Barker SE. Addict Behav. 1993;18:299-309. Godwin S, Chambers E, Cleveland L, Ingwersen L. J Am Diet Assoc. 2006;106:1246–50. Greeno CG, Wing RR, Marcus MD. Addict Behav. 1999;24:299–303. Guertin TL, Conger AJ. Addict Behav. 1999;24:175–93. Guthrie HA. J Am Diet Assoc. 1984;84:1440–4. Harnack L, Steffen L, Arnett DK, Gao S, Luepker RV. J Am Diet Assoc. 2004;104:804–6. Herman CP, Polivy J. Res Publ Assoc Res Nerv Ment Dis. 1984;62:141–56. Howat PM, Mohan R, Champagne C, Monlezun C, Wozniak P, Bray GA. J Am Diet Assoc. 1994;94:169–73. Huybregts L, Roberfroid D, Lachat D, Van Camp J, Kolsteren P. Public Health Nutr. 2008;11:581–7. Kales EF. Physiol Behav. 1990;48:837–40. Kales EF. Appetite. 1991;17:241–2. Keel PK, Mayer SA, Harnden-Fischer JH. Int J Eat Disord. 2001;29:294–301. Kirkcaldy-Hargreaves M, Lynch GW, Santor C. J Can Diet Assoc. 1989;41:102–10. Klesges RC, Eck LH, Ray JW. J Consult Clin Psychol. 1995;63:438–44. Knight LJ, Boland FJ. J Abnorm Psychol. 1989;98:412–20. Kuehneman T, Stanek K, Eskridge K, Angle C. J Am Diet Assoc. 1994;94(5):548–51. Lansky D, Brownell KD. Am J Clin Nutr. 1982;35:727–32. Lillegaard IT, Overby NC, Andersen LF. Eur J Clin Nutr. 2005;59:611–17. Loxton NJ, Dawe S. Int J Eat Disord. 2001;29:455–62. Macdiarmid J, Blundell JE. Nutr Res Rev. 1998;11:231–53. Marilyn AZ. Addictive Behav. 1984;9:347–50. Matheson DM, Hanson KA, McDonald TE, Robinson TN. Arch Pediatr Adolesc Med. 2002;156:867–71. Muhlheim LS, Allison DB, Heshka S, Heymsfield SB. Int J Eat Disord. 1998;24:259–66. Nelson M, Atkinson M, Darbyshire S. Br J Nutr. 1994;72:649–63. Overduin J, Jansen A, Louwerse E. Int J Eat Disord. 1995;18:277−85. Polivy J, Herman CP. Am Psychol. 1985;40:193–201. Posner BM, Smigelski C, Duggal A, Morgan JL, Cobb J, Cupples LA. J Am Diet Assoc. 1992;92:738–41. Rapp SR, Dubbert PM, Burkett PA, Buttross Y. J Am Diet Assoc. 1986;86:249–51. Robson Pj, Livingstone MB. Public Health Nutr. 2000;3:183–92. Rosen JC, Leitenberg H, Gross J, Willmuth M. Adv Behav Res Ther. 1985;7:181–97. Scagliusi FB, Polacow VO, Artioli GG, Benatti FB, Lancha AHJ. J Am Diet Assoc. 2003;103:1306–13. Skrzypek S, Wehmeier PM, Remschmidt H. Eur Child Adolesc Psychiatry. 2001;10:215−21. Stanton AL, Tips TA. Int J Eat Disord. 1990;9:387–93. Telch CF, Agras WS, Rossiter EM, Wilfley D, Kenardy J. J Consult Clin Psychol. 1990;58:629–35. Turconi G, Guarcello M, Berzolari FG, Carolei A, Bazzano R, Ruggeri C. Eur J Clin Nutr. 2005;59:923–31. Urdapilleta I, Mirabel-Sarron C, Meunier JM, Richard JF. Encephale. 2005;31:82−91. Venter CS, MacIntyre UE, Vorster HH. J Hum Nutr Diet. 2000;13:205–18. Vinai P, Cardetti S, Ferrato N, Vallauri P, Carpegna G, Masante D, Luxardi G, Todisco P, Manara F, Ruggiero GM, Sassaroli S. Eat Behav. 2007;8:291–5. Vinai P, Carpegna G, Cardetti S, Ferrato N, Vallauri P, Masante D, Meister A, Ruggiero GM, Sassaroli S. Eat Disord. 2008;16:128–35. Webb CA, Yuhas JA. J Am Diet Assoc. 1988;88:601–2. Willett WC. In: Willett WC, editor. Nutritional epidemiology. New York: Oxford University Press; 2002. pp. 74–100. Yellowlees PM, Roe M, Walker MK, Ben-Tovim DI. Br Med J. 1988;296:1689–90. Yuhas JA, Bolland JE, Bolland TW. J Am Diet Assoc. 1989;89:1473–7. Williamson DA, Gleaves DH, Lawson OJ. J Psychopathol Behav Assess. 1991;13:257–68.
wwwwwwwwwwwwwww
Chapter 81
Visual Processing, Food Cravings and Weight-Loss Dieters Eva Kemps and Marika Tiggemann
Abbreviations DVN EM ST TS
Dynamic visual noise Eye movements Spatial tapping Thought suppression
81.1 Introduction The act of dieting involves the deliberate restriction of food intake in an attempt to achieve weight loss. Although moderate weight reduction in obese individuals can have clear health benefits, such as reduced risk of cardiovascular disease and increased aerobic and anaerobic aptitudes, the actual practice of weight-loss dieting is associated with a number of negative outcomes. For example, people who are on a diet often also exhibit mood swings, depression, lowered self-esteem and impaired cognition. Another unintended negative consequence of weight-loss dieting is the experience of unwanted food cravings. In fact, recurrent food cravings have the potential to disrupt and thwart dieting attempts. This highlights the need for effective techniques for reducing food cravings. However, contemporary intervention tools involving either suppression of craving-related thoughts or unreinforced exposure to food cues have shown only mixed success. This chapter describes a radically different approach to controlling food cravings based on converging evidence that mental imagery is a key component of the craving experience. Empirical data, which show that interfering with the cognitive processes that support craving-related mental images can suppress food cravings in dieters, are reviewed.
81.2 Food Cravings The term “craving” refers to a motivational state whereby an individual feels compelled to seek and ingest a particular substance, usually cigarettes, alcohol or drugs (Baker et al. 1986). More recently, however, interest has fallen on food cravings. These have been described as an intense desire or urge to E. Kemps (*) School of Psychology, Flinders University, GPO, Adelaide, SA 5001, Australia e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_81, © Springer Science+Business Media, LLC 2011
1261
1262
E. Kemps and M. Tiggemann
eat a specific food (Weingarten and Elston 1990). Food cravings are not to be equated with hunger, for they generally occur in the absence of hunger (Cornell et al. 1989). Hunger signals a physiological need to eat, which can be satisfied with any food. Cravings, however, are for a particular food, and can only be satisfied by eating that food. For example, one craves chocolate ice-cream, rather than something generically sweet. In particular, there is no substitute for chocolate when it is craved; other sweet foods, including white ‘chocolate’, cannot fully satisfy chocolate cravings (Michener and Rozin 1994). Food cravings are a common and everyday experience (Lafay et al. 2001), with prevalence rates varying according to gender and age. Women tend to crave more than men (Weingarten and Elston 1991), and younger people crave more than older people (Pelchat 1997). Not surprisingly, there exist marked cross-cultural differences in the kind of foods craved (Hawks et al. 2003). For example, chocolate is the most commonly craved food in Western societies (Hetherington and Macdiarmid 1993), followed by chips, pizza, cake, and ice-cream. In Egypt, on the other hand, the most craved foods are vegetable dishes, some stuffed or cooked with meat (Parker et al. 2003). The origin of food cravings has been attributed to a range of physiological and psychological factors. At a physiological level, food cravings have been linked to nutritional deficiencies (Wardle 1987), and in women to hormonal changes, such as menstrual-related changes (Dye et al. 1995) and pregnancy (Dickens and Trethowan 1971). At a psychological level, food cravings can be triggered by negative mood states, including feelings of boredom, loneliness, depression, anxiety, as well as stress (Hill et al. 1991). Food cravings can also by elicited by exposure to environmental cues, such as the sight or smell of tasty food (Fedoroff et al. 2003).
81.3 Food Cravings and Weight-Loss Dieting Although food cravings are not invariably linked to dietary restraint, weight-loss dieting has been associated with an increased occurrence of cravings (Hetherington and Macdiarmid 1993; Pelchat 1997). Food cravings are thought to be part of the preoccupying cognitions concerning food, weight and body shape that accompany dieting behaviour (Green 2001). These dieting-related cravings can give rise to a range of negative consequences. In particular, food cravings have been implicated in the early dropout from weight-loss programs. For example, Sitton (1991) reported that attrition rates were almost three times higher for carbohydrate cravers than for non-cravers during the first month of a prescribed diet. Cravings for food can also lead to feelings of guilt and shame following consumption if dietary restriction fails (MacDiarmid and Hetherington 1995). Importantly, dieting-related cravings are known to adversely affect cognitive processing. For example, dieters exhibit biased attentional processing of food and body shape words on the modified Stroop task (Green and Rogers 1993). Additionally, experimentally induced food cravings have been shown to slow responses on a simple reaction time task in highly restrained eaters and dieters (Green et al. 2000). This finding has important implications in that speeded responding to a visual probe (as in a reaction time task) is vital in vigilance tasks, such as inspecting items on an industrial production line, or manoeuvring through dense traffic. Slowed reaction times arising from food cravings could compromise performance of such everyday tasks, thereby reducing work efficiency or increasing accidents. Finally, cravings triggered by chronic dietary restriction have been identified as a precursor to binge eating, particularly in the context of obesity (Schlundt et al. 1993) and eating disorders, such as bulimia nervosa (Mitchell et al. 1985). Obesity and eating disorders are themselves risk factors for the development of other serious medical and psychological problems, such as cardiovascular disease, Type 2 diabetes, depression and low self-esteem.
81 Visual Processing, Food Cravings and Weight-Loss Dieters
1263
The incidence of such dieting-related food cravings is likely to be on the rise, fuelled by a continuing increase in obesity rates (Wadden et al. 2002) and body image concerns (Cash and Pruzinsky 2002). The advent of public health campaigns targeting the health risks associated with being overweight is intended to result in more people trying to lose weight. In parallel, the continual and pervasive presentation by the media of the thin ideal for women and the lean but muscular ideal for men can only serve to increase rates of body dissatisfaction, which many individuals attempt to resolve by dieting in an attempt to more closely approach the elusive societal ideal. As more people adopt food restriction (dieting) as the means for attempting weight loss in response to these messages, there will be a concomitant increase in accompanying levels of food craving. Thus the availability of a technique for curbing dieting-related food cravings is of considerable practical importance.
81.4 Reduction of Food Cravings 81.4.1 Thought Suppression and Cue Exposure Response Prevention The techniques currently available for reducing food cravings involve either suppression of dietingrelated thoughts or cue exposure response prevention. The rationale behind thought suppression is that people can avoid performing an unwanted behaviour by deliberately not thinking about it. Harnden et al. (1997) used this paradigm to investigate the effect of thought suppression on weightrelated preoccupations in dieters. They instructed female dieters and non-dieters to either suppress or express thoughts about weighing themselves and to subsequently verbalise all thoughts that came to mind during a 5-min period. Although dieters mentioned more weight-related thoughts overall than did non-dieters, suppression reduced the frequency of such thoughts in both groups. The magnitude of suppression was, however, less for dieters than for non-dieters. Oliver and Huon (2001) further showed that women who reported high levels of disinhibited eating were more successful at suppressing thoughts about food and eating than were women low on disinhibition. However, high disinhibitors were more likely to use punishment and worry strategies to control their thoughts. These thought control strategies were themselves related to feelings of anxiety and distress (Table 81.1).
Table 81.1 Schematic representation of the craving reduction techniques of thought suppression and cue exposure response prevention Thought suppression Cue exposure response prevention Period 1: Thought monitoring Exposure 1 to craved food while resisting consumption, followed by a rating of food craving Participants are instructed to think about anything they like, and to press a button if they think about the craved food Period 2: Thought suppression Participants are instructed to not think about the craved food, and to press a button if they do Period 3: Thought monitoring
Exposure 2 to craved food while resisting consumption, followed by a rating of food craving…
Exposure n to craved food while resisting consumption, followed by a rating of food craving Participants are again instructed to think about anything they like, and to press a button if they think about the craved food This table presents a step-by-step overview of the techniques of thought suppression and cue exposure response prevention as applied to the reduction of food cravings
1264
E. Kemps and M. Tiggemann Table 81.2 Mean pre- and post-imagery task food craving levels as a function of imagery task and dieting status (standard deviations in parentheses) Dieters Non-dieters Imagery task Pre Post Pre Post Visual 71.93 (24.11) 38.80 (26.32) 45.13 (29.95) 29.31 (25.02) Auditory 49.40 (35.79) 35.96 (36.83) 28.93 (26.08) 21.11 (20.22) This table shows that the visual imagery task reduced participants’ level of food craving more than did the auditory imagery task, particularly for dieters. Craving was rated on a 100-point scale ranging from “no desire or urge to eat” to “extremely strong desire or urge to eat”
More generally, suppression of unwanted thoughts about food, eating or weight can paradoxically lead to an increase in such thoughts. For example, Johnston et al. (1999) showed that participants who had been instructed to suppress chocolate-related thoughts made fewer mentions of chocolate during a 5-min articulated thoughts task that involved planning a dessert menu for a dinner party than participants who had received no such instruction. However, those participants who engaged in thought suppression actually worked harder (performed better) on a subsequent task that yielded chocolate awards, indicating an increase in drive for chocolate. Although such a rebound effect is not necessarily always seen, ironic effects of thought suppression have been documented in a number of other areas, such as mood control, stereotyping and traumatic memories (Abramowitz et al. 2001). Furthermore, an increased occurrence of unwanted thoughts following attempts at suppression can itself lead to perceived loss of control over one’s thoughts, and thereby give rise to feelings of failure and distress (Kelly and Khan 1994). The second currently available technique, cue exposure response prevention, involves successive exposures to an appetitive target (food), with instructions to attempt to resist consumption. The elicited craving is expected to diminish across exposures because the appetitive cue is not reinforced by consumption. Tuomisto et al. (cited in Hetherington 2001) used this paradigm in an attempt to reduce food cravings in obese women. Following a baseline measurement of craving, the women were presented with warm pizza. They were asked to look at, smell and imagine eating the pizza, but not to eat it. Ratings of food craving were taken every 10 min. Although cravings decreased somewhat across exposures, they remained high compared to baseline. At a general level, cue exposure response prevention has been used with mixed success in the treatment of alcoholism (Conklin and Tiffany 2002), phobias and binge eating in patients with bulimia nervosa (Jansen 2001). Thus, neither thought suppression nor cue exposure response prevention holds much promise as a technique for reliably reducing dieting-related food cravings. We reasoned that in order to develop a different (and potentially more successful) kind of craving-reduction technique, we first need to gain insight into the nature of the actual craving experience (Table 81.2).
81.4.2 The Nature of Food Cravings: A Role for Mental Imagery There is accumulating evidence from a number of different sources pointing to a role for mental imagery in craving episodes. First, anecdotal accounts describe the experience of desire-related images in naturally occurring cravings (Salkovskis and Reynolds 1994). These accounts have now been corroborated by more formal surveys about everyday food cravings that show that respondents readily use imagery terms to describe their cravings. For example, Tiggemann and Kemps (2005) reported that 30% of their undergraduate student sample used phrases such as ‘I could picture the pizza in my mind, picture eating it’ when asked to write a short paragraph describing a previous food craving episode. In addition, when presented with a list of descriptive statements, respondents
81 Visual Processing, Food Cravings and Weight-Loss Dieters
1265
strongly endorse imagery-based descriptors as characteristic of their cravings. Imagery descriptors in the visual modality (e.g. ‘I am visualising it’) in particular are rated highly; in contrast, auditory descriptors (e.g. ‘I imagine the sound of myself having it’) are not highly rated. Furthermore, when asked to assign specific percentages to each of the five sensory modalities involved in an imagined food craving experience, as shown in Fig. 81.1, the visual modality was the highest. These findings indicate that the imagery basis of food cravings is predominantly visual in nature (Table 81.3). Second, instructing participants to imagine urge-related scenarios can induce cravings in the laboratory. For example, using a methodology derived from cigarette craving research, Green et al. (2000) reported a positive correlation between latency on a simple reaction time task following instructions to imagine a food scenario (‘Imagine you are eating your favourite food’) and self-reported desire to eat. Harvey et al. (2005) subsequently showed that craving levels increased following instructions to imagine a food scenario, but not a nonfood scenario. They also noted a positive correlation between participants’ self-reported vividness of the image of the food scenario and their level of craving, indicating that stronger food cravings are associated with more vivid images. Finally, a recent theoretical account of cravings, the elaborated intrusion theory of desire, proposes that sensory images are a central feature of any craving experience (Kavanagh et al. 2005).
Table 81.3 Key facts of food cravings 1. A food craving is an intense desire or urge to eat a specific food 2. The most commonly craved food in Western society is chocolate, followed by chips, pizza, cake and ice cream 3. Food cravings are a common, everyday experience, particularly in women and younger adults 4. Food cravings generally occur in the absence of hunger; they are instead triggered by hormonal changes, negative mood states and exposure to environmental food cues 5. Food cravings have been associated with a number of negative outcomes (particularly in weight-loss dieters), such as early dropout from weight-loss programs, feelings of guilt and shame, poor cognitive task performance and binge eating 6. Mental imagery is a key component of food cravings; specifically, when people crave they have vivid images of the craved food This table lists the key characteristics of food cravings, including the various causes of food cravings, the potential consequences of food cravings, and the role of mental imagery in the experience of food cravings
Fig. 81.1 Mean percentages of sensory modalities involved in craving-related food images. This figure shows the relative involvement of each of the five sensory modalities in craving-related images of food. These are in order: visual (sight), gustatory (taste), olfactory (smell), tactile (touch) and auditory (hearing)
Sensory modality
Visual
Gustatory
Olfactory
Tactile
Auditory 0
10
20
30
Percentage
40
50
1266
E. Kemps and M. Tiggemann
81.4.3 A Working Memory Approach to Reducing Food Cravings In recent years, cognitive psychologists have adopted a working memory approach to conceptualise the phenomenological experience of mental imagery (Richardson 1999). A widely used and successful account of working memory is that originally proposed by Baddeley and Hitch (1974). The working memory model has undergone revision and elaboration over time but in its current version comprises an overarching central executive responsible for coordinating the activities of a number of modality-specific slave systems and an episodic buffer, a multimodal interface with long-term memory (Baddeley 2000). The two initially proposed slave systems, which have attracted the most research attention, are the visuo-spatial sketch pad and the phonological loop. The visuo-spatial sketch pad maintains visual and spatial material and is involved in visual imagery. The phonological loop analogously maintains verbal material and is involved in auditory imagery. These slave systems are argued to have limited capacity. Thus a concurrent visual (or verbal) task will interfere with the content of the visuo-spatial sketch pad (phonological loop) by competing for the limited visual (verbal) storage capacity. In support, Baddeley and Andrade (2000) employed dual-task methodology to show that the vividness of visual imagery was reduced by tasks that selectively loaded the visuo-spatial sketch pad, whereas the vividness of auditory imagery was reduced by tasks that specifically utilised the phonological loop. Inspired by recent developments in cigarette-craving research (Panabokke 2004), we (Harvey et al. 2005) applied this analysis to investigate whether interference from a competing task that requires the same working memory resources as those used to create and maintain craving-related images could suppress food cravings in dieters. As questionnaire data on the subjective experience of cravings have shown that craving-related food images involve visual rather than auditory content, we predicted that a competing visual task (loading the visuo-spatial sketch pad) would reduce food cravings relative to a comparable auditory task (loading the phonological loop). In our study, undergraduate women (self-identified dieters and non-dieters) underwent an imaginal food craving induction procedure and subsequently rated their level of craving. They were then cued to vividly imagine a series of nonfood items. A random half of each group formed visual images of common objects and scenes (e.g. a rainbow), whereas the other half formed auditory images of everyday sounds (e.g. a siren). Following the imagery task, all participants again rated their level of food craving. As predicted, the visual imagery task was superior to the auditory task in reducing the women’s food cravings. Moreover, as can be seen in Table 18.2, dieters reported stronger cravings overall, and the magnitude of craving reduction was greater for them than for non-dieters. Thus visual imagery techniques offer potential scope for reducing dieting-related food cravings. While forming a series of visual images and holding them in mind constitutes an excellent laboratory technique for investigating cravings, it is not only an elaborate, but also a cognitively effortful, procedure and therefore unlikely to be an effective craving reduction technique in a practical sense. Instead, we need to find a simple and relatively undemanding visual task. Fortunately, the working memory literature boasts several simple tasks known to load the visuo-spatial sketch pad. These include saccadic eye movements (visually tracking a rapidly moving stimulus; Idzikowski et al., cited in Baddeley 1986), dynamic visual noise (watching a flickering pattern of random black and white dots, similar to snow on an untuned television screen; Quinn and McConnell 1996) and spatial tapping (tapping four keys arranged in a square pattern; Farmer et al. 1986). To investigate the craving reduction capacity of such simple repetitive tasks, we (Kemps et al. 2004) adopted a paradigm derived from experimental analogues of post-traumatic stress disorder (Table 81.4). These have shown that concurrent visual processing in the form of simple eye or arm movements, or watching a flickering pattern of random black and white dots, reduced the vividness and consequent emotional impact of distressing images (Andrade et al. 1997; Kavanagh et al. 2001). Accordingly, in view of the association between imagery vividness and food craving intensity,
81 Visual Processing, Food Cravings and Weight-Loss Dieters
1267
Table 81.4 Key features of visual working memory based craving reduction 1. The visual component of working memory (i.e. visuo-spatial sketch pad) is involved in the generation and maintenance of visual images 2. Food cravings involve primarily images in the visual sensory modality 3. In working memory terms, food cravings engage the visuo-spatial sketch pad 4. The visuo-spatial sketch pad has limited storage capacity 5. Loading the visuo-spatial sketch pad with a concurrent task when experiencing a food craving reduces the intensity of the craving, because the task and the visual imagery associated with the craving compete for the same limited storage capacity 6. The intensity of a food craving is related to the vividness with which the craved food is imagined, such that stronger food cravings are associated with more vivid images of the craved food 7. Concurrent visual processing has its specific craving reducing effect by diminishing the vividness of craving-related food images This table outlines the theoretical rationale behind the utility of visual working memory-based techniques for reducing food cravings, with emphasis on the visual imagery basis of food cravings, the mutual competition between food cravings and concurrent visual tasks for limited visuo-spatial sketch pad storage capacity, and the relationship between imagery vividness and craving intensity
No task
EM
DVN
ST
No task
DVN
ST
55 50 Food craving rating
Imagery vividness rating
90
EM
80
70
60
45 40 35 30 25
50
20 Dieters
Non-dieters
Dieters
Non-dieters
Fig. 81.2 Mean imagery vividness and food craving levels (with standard errors) for dieters and non-dieters in each of the concurrent task conditions. This figure shows that the concurrent visual tasks of eye movements, dynamic visual noise and spatial tapping made food images less vivid and food cravings less intense for both dieters and non-dieters. Imagery vividness and craving were rated on a 100-point scale ranging from “no image at all” to “image perfectly clear and vivid”, and from “no desire or urge to eat” to “extremely strong desire or urge to eat”, respectively. EM eye movements, DVN dynamic visual noise, ST spatial tapping
we investigated whether engaging in relatively effortless visual activity could reduce food cravings in dieters (via the analogous mechanism of decreasing the vividness of the accompanying images). In our investigations, dieting and non-dieting female undergraduate students formed and maintained images of commonly craved foods (e.g. chocolate, cake, ice-cream) elicited by pictures or verbal cues, while performing one of the three aforementioned visual tasks: saccadic eye movements, dynamic visual noise or spatial tapping. Participants rated the vividness of their images and the intensity of their food cravings. As predicted, all three tasks made participants’ food images less vivid. Importantly, as can be seen in Fig. 81.2, this reduction in imagery vividness was accompanied by a reduction in participants’ level of food craving. Dieters and non-dieters showed the same pattern of results. Thus simple, repetitive visual tasks hold promise as a practical technique for controlling food cravings in dieters.
1268
E. Kemps and M. Tiggemann
No task
DVN
No task
TS
TS
50
60
40 Food craving rating
Imagery vividness rating
70
DVN
50 40 30 20 10 0
30 20 10 0
Dieters
Non-dieters
Dieters
Non-dieters
Fig. 81.3 Mean imagery vividness and food craving levels (with standard errors) for dieters and non-dieters in each of the concurrent task conditions. This figure shows that dynamic visual noise was a more effective technique than thought suppression for reducing the vividness of food images and the intensity of food cravings in dieters; however, the techniques were equally effective for non-dieters. Imagery vividness and craving were rated on a 100-point scale ranging from “no image at all” to “image perfectly clear and vivid”, and from “no desire or urge to eat” to “extremely strong desire or urge to eat”, respectively. DVN = Dynamic visual noise; TS = Thought suppression
In a subsequent study, we (Kemps et al. 2008) examined whether these findings in dieting samples of normal weight university students extended to community samples of overweight people who are actively trying to lose weight. Overweight women following a prescribed weight-loss diet and non-dieting controls formed and maintained images of highly desired foods (e.g. chocolate, chips, pizza) while watching the dynamic visual noise display or suppressing thoughts about food. Both techniques successfully reduced food cravings for dieters as well as non-dieters. However, as shown in Fig. 81.3, their relative effectiveness differed markedly depending on the women’s dieting status. While both techniques reduced cravings equally well for non-dieters, dynamic visual noise was clearly the more effective technique for dieters. Thus the utility of visually based tasks for curbing dieting-related food cravings clearly extends to overweight individuals who are trying to lose weight.
81.5 Conclusion The working memory approach outlined here presents a radically different strategy for reducing dieting-related food cravings, one that is based on the visual imagery nature of food cravings. According to working memory theory, food cravings engage the visuo-spatial sketch pad, and can thus be reduced by loading this component with a concurrent task. Specifically, concurrent visual tasks have their craving reducing effect by diminishing the vividness of desire-related food images held in the limited capacity visuo-spatial sketch pad. Unlike thought suppression and cue exposure response prevention, concurrent visual processing does not focus directly on the craved food. As a result, it is likely to be found less aversive, making it not only an effective but also an attractive technique for reducing dieting-related food cravings.
81 Visual Processing, Food Cravings and Weight-Loss Dieters
1269
81.6 Applications to Other Areas of Health and Disease The next challenge is to see how well these techniques involving concurrent visual activity generalize outside the laboratory, how well they are able to control unwanted and problematic food cravings in clinically overweight and obese individuals, as well as in patients diagnosed with binge eating disorder and bulimia nervosa. These techniques may also have utility beyond the food domain, to cravings for other substances, such as nicotine, caffeine, alcohol and drugs. Deprivation of these substances, like food restriction, can lead to cravings. Such cravings are also imagery based, and involve, primarily, visual images (May et al. 2004; Kemps and Tiggemann 2009). In fact, emerging evidence shows that visual imagery tasks can similarly suppress cigarette cravings in smokers (Versland and Rosenberg 2007) and caffeine cravings in habitual coffee drinkers (Kemps and Tiggemann 2009). Future research could determine whether simple visual tasks, like dynamic visual noise, could also reduce these cravings, as well as cravings for alcohol and other drugs. More generally, mental imagery has been shown to play a role in the development and maintenance of a variety of psychological disorders, including post-traumatic stress disorder, agoraphobia, body dysmorphic disorder and mood disorders. Imagery-based interventions have been used successfully in the treatment of these psychopathologies (Hackman and Holmes 2004). More recently, Lilley et al. (in press) showed that making simple eye movements while imagining a traumatic event reduced the vividness and emotional impact of distressing, intrusive images in a clinical sample with post-traumatic stress symptoms. Summary Points • Dieting to lose weight is associated with a number of negative outcomes, including the experience of unwanted food cravings. • Although food cravings are a common everyday experience, dieting-related cravings can give rise to a range of negative consequences, such as binge eating, early dropout from weight-loss programs, feelings of guilt and shame and poor cognitive task performance. • The prevalence of dieting-related cravings is likely to increase, because of increasing rates of obesity and body image concerns, highlighting the need for effective craving reduction techniques. • The currently available techniques for reducing dieting-related food cravings involving either thought suppression or cue exposure response prevention have not proven very successful. • Based on converging evidence pointing to mental imagery as a key component of the craving experience, recent findings show that interfering with the cognitive processes that support craving-related images reduces food cravings in weight-loss dieters. • As food cravings involve images primarily in the visual sensory modality, engaging in a concurrent visual task is most effective in reducing dieting-related cravings. Definitions and Explanations of Key Terms Binge eating: The consumption of an excessively large amount of food in a single sitting, accompanied by feelings of lacking control over one’s eating. Cue exposure response prevention: A psychological intervention derived from conditioning models of psychopathology that involves successive exposures to an appetitive target, with instructions to attempt to resist consumption. Food craving: A strong urge or desire to eat a specific food.
1270
E. Kemps and M. Tiggemann
Mental imagery: The mental representation of stimuli that are not physically present. Preoccupying cognitions: A chronic preoccupation with personally relevant or motivationally salient thoughts. Thought suppression: A cognitive control strategy whereby an individual avoids thinking about a behaviour in order to prevent performing that behaviour. Weight-loss dieting: The deliberate restriction of food intake in an attempt to lose weight. Working memory: A system for the temporary storage of information in the face of ongoing processing.
Acknowledgements This work was supported under the Australian Research Council’s Discovery Project funding scheme (project number DP0664435).
References Abramowitz JS, Tolin DF, Street GP Clin Psychol Rev. 2001;21:683–703. Andrade J, Kavanagh D, Baddeley A Brit J Clin Psychol. 1997;36:209–23. Baddeley A. Trends Cogn Sci. 2000;4:417–23. Baddeley AD. Working memory. Oxford: Oxford University Press; 1986. —Baddeley AD, Hitch GJ In: Bower G, editor. The psychology of learning and motivation. London: Academic; 1974. p. 47–90. Baddeley AD, Andrade J J Exp Psychol Gen. 2000;129:126–45. Baker TB, Morse E, Sherman JE In: Rivers PC, editor. The Nebraska symposium on motivation: alcohol use and abuse. Lincoln: University of Nebraska Press; 1986. p. 257–323. Cash TF, PruzinskyT editors. Body image: a handbook of theory, research and clinical practice. New York: Guilford; 2002. Conklin CA, Tiffany ST Addiction 2002;97:155–67. Cornell CE, Rodin J, Weingarten H Physiol Behav. 1989;45:695–704. Dickens G, Trethowan WH J Psychosom Res. 1971;15:259–68. Dye L, Warner P, Bancroft J J Affect Disorders. 1995;34:157–64. Farmer EW, Berman JVF, Fletcher YL Q J Exp Psychol. 1986;38A:675–88. Fedoroff IC, Polivy J, Herman CP Appetite 2003;41:7–13. Green MW, Rogers PJ Int J Eat Disorder. 1993;14:515–7. Green MW, Rogers PJ, Elliman NA Int J Eat Disorder. 2000;27:419–27. Hackmann A, Holmes EA Memory 2004;12:389–402. Harnden JL, McNally RJ, Jimerson DC Int J Eat Disorder. 1997;22:285–90. Harvey K, Kemps E, Tiggemann M Brit J Health Psych. 2005;10:49–56. Hawks SR, Madanat HN, Merrill RM, Goudy MB, Miyagawa T Health Promot Int. 2003;18:153–62. Hetherington MM. Food cravings and addiction. Surrey: Leatherhead; 2001. Hetherington MM, Macdiarmid JI Appetite 1993;21:233–46. Hill AJ, Weaver CF, Blundell JE Appetite 1991;17:187–97. Jansen AIn: Hetherington MM, editor. Food cravings and addiction. Surrey: Leatherhead; 2001. p. 549–65. Johnston L, Bulik CM, Anstiss V Int J Eat Disorder. 1999;26:21–7. Kavanagh D, Andrade A, May J Psychol Rev. 2005;112:446–7. Kavanagh DJ, Freese S, Andrade J, May J Brit J Clin Psychol. 2001;40:267–80. Kelly AE, Khan JH J Pers Soc Psychol. 1994;66:998–1006. Kemps E, Tiggemann M (2009) Exp. Clin. Psychopharm. 17: 43–50 Kemps E, Tiggemann M, Woods D, Soekov B Int J Eat Disorder. 2004;36:31–40. Kemps E, Tiggemann M, Christianson R J Behav Ther Exp Psy. 2008;39:177–86. Lafay L, Thomas F, Mennen L, Charles MA, Eschwege E, Borys J, Basdevant A Int J Eat Disorder. 2001;29:195–204. Lilley SA, Andrade J, Turpin G, Sabin-Farrell, Holmes E, Brit. J. Clin. Psychol. 2009;48:309–321. Macdiarmid JI, Hetherington MM Brit J Clin Psychol. 1995;34:129–38. May J, Andrade J, Panabokke N, Kavanagh D Memory 2004;12:447–61.
81 Visual Processing, Food Cravings and Weight-Loss Dieters Michener W, Rozin P Physiol Behav. 1994;56:419–22. Mitchell JE, Hatsukami D, Eckert ED, Pyle RL Am J Psychiat. 1985;142:482–5. Oliver KG, Huon KG Int J Eat Disorder. 2001;30:329–37. Panabokke N. The role of visual imagery in craving. PhD University of Sheffield; 2004. Parker S, Kamel N, Zellner D Appetite 2003;40:193–5. Pelchat ML. Appetite 1997;28:103–13. Quinn JG, McConnell J Q J Exp Psychol. 1996;49A:200–15. Richardson JTE. Imagery. Hove: Psychology; 1999. Salkovskis PM, Reynolds M Behav Res Ther. 1994;32:193–201. Schlundt DG, Virts KL, Sbrocco T, Pope-Cordle J Addict Behav. 1993;18:67–80. Sitton SC. Psychol Rep. 1991;69:683–6. Tiggemann M, Kemps E Appetite 2005;45:305–13. Versland A, Rosenberg H Addict Res Theory. 2007;15:177–87. Wadden TA, Brownell KD, Foster GD J Consult Clin Psych. 2002;70:510–25. Wardle J. Brit J Clin Psychol. 1987;26:47–55. Weingarten HP, Elston D Appetite 1990;15:231–46. Weingarten HP, Elston D Appetite 1991;17:167–75.
1271
wwwwwwwwwwwwwww
Chapter 82
Abnormal Physiologic Responses to Touch in Feeding Difficulties Donna Scarborough
Abbreviations APIB Assessment of Preterm Infants’ Behavior NBAS Neonatal Behavioral Assessment Scale
82.1 Introduction Development of early feeding skills in full-term infants is complex for a number of reasons: • An infant undergoes a dramatic transformation during the first year of life in all aspects of development including: neurologic, anatomic, motor, sensory, physiologic, feeding/swallowing skills, communication, cognitive, social–emotional, and play. Each of these areas is interdependent, and, like the pieces of a jigsaw puzzle, can be viewed piece by piece or as a whole (Fig. 82.1). • The transition of feeding behaviors from primarily physiologic responses to cognitively learned responses, or combinations of both. • Infants and young children are dependent upon others during the feeding process and express themselves by indirect, nonverbal means. Thus, successful feeding interactions and development are dependent upon appropriate interpretations of behavior. One aspect of feeding development that reflects this complexity is the ability of an infant to appropriately respond to touch input (Table 82.1). This section will examine early infant development, identify behaviors related to touch processing during the earliest stages of development, and examine long-term feeding complications and clinical applications related to an interruption of these early stages.
82.2 Early Developmental Levels Since the 1950s, a number of researchers have described the childhood maturation process as proceeding in a sequential manner. Many of the early developmental theories proposed by Piaget (1952), Mahler et al. (1975), Greenspan and Lourie (1981), and Kopp (1982) have historically been important D. Scarborough (*) Department of Speech Pathology and Audiology, Miami University, 26 Bachelor Hall, Oxford, OH 45056, USA e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_82, © Springer Science+Business Media, LLC 2011
1273
1274
D. Scarborough
Fig. 82.1 The complex puzzle of development. For a child to successfully develop each individual component must cohesively connect
Table 82.1 Key features of touch (Henricson et al. 2008; Pihko and Lauronen 2004; Jean 2001; Hendry et al. 1999) • Touch or mechanoreception is only one component of the somatosensory system (Fig. 82.2) • Touch includes perception of pressure, texture, form, and vibration • Touch receptors are found at varying depths below the skin, and include Pacinian corpuscles, Ruffini corpuscles, Meissner corpuscles, and Merkel cells • Each type of touch receptor responds differently to varying types of input • Touch information from the body and limbs is carried to the dorsal root ganglia cells of the spinal cord, through the dorsal column–medial lemniscus pathway to the thalamus. From the thalamus, the information is then carried to the cortex. • Touch from the face, oral cavity, and pharynx is carried from the trigeminal, glossopharyngeal, and vagus nerves to the thalamus. From the thalamus the information is then carried to the cortex. • Touch is organized throughout each level with a clearly defined map of the body • As early as the seventh month of pregnancy, touch pathways are intact enough to produce cortical responses • The speed of processing and timing of processing improves as mylenation occurs • Sleep stages impact processing of touch input in newborns • The touch of firm pressure has a calming effect on the sympathetic nervous system
for classifying feeding behaviors. For the purpose of this section, a compilation and summary of these four theories targeting the first year of development is described utilizing the nomenclature of Greenspan and Lourie (1981) and Kopp (1982) (Fig. 82.3): • The period of “homeostasis” or “neurophysiologic maturation,” is a critical period from birth to 3 months of life when an infant’s biological functions stabilize as a result of autonomic nervous system maturation. Examples of stabilizing biologic functions include temperature regulation, respiratory rhythms, sleep–wake cycles, and feeding and excretion patterns. Touch processing during the beginning of this stage involves generalized autonomic system responses, such as startling,
82 Abnormal Physiologic Responses to Touch in Feeding Difficulties
1275
Fig. 82.2 Somatosensory system and feeding. The somatosensory system is very complex and critical for successful feeding. Touch or mechanoreception is only one component of this system and will be the focus of this section
Fig. 82.3 Developmental levels for feeding. Homeostasis is the first developmental level for term infants. In typical development, this stage flows into the concurrent stages of attachment and neurophysiologic modulation. Preterm development is assessed through the NBAS (which highlights state behavior changes) and other models of preterm development
gagging, or state behavior changes. As the child progresses through this period of development touch responses become more appropriate such that an infant begins to differentiate between the soothing touch of being swaddled and the painful touch of a needle (Scarborough et al. 2006). • The periods of “attachment” and “sensorimotor modulation” follow “homeostasis.” During the period of attachment infants respond to the outside world, thus beginning the process of forming relationships and differentiating between caregivers. Concurrent with the period of attachment is the period of sensorimotor modulation. During this period of development the infant begins to utilize multiple sensory systems (i.e., touch, vision, balance) during interactions in order to modulate motor responses such that movements become voluntary rather than reflexive. An important component of this stage is that children will repeat behaviors or modulate movements without purposeful or conscious reflection. By the end of the first year of life, these periods of development should be complete.
1276
D. Scarborough
82.3 U nderstanding Infant Behaviors During the Period of “Homeostasis” Successful interpretation of infant behavior is one of the greatest challenges for developmental feeding specialists, particularly in very young infants. Although the period of “homeostasis” is considered the first stage of development for full-term infants, we know that development is on a continuum, thus later behaviors are dependent on successful achievement of previous neurologic maturation. Therefore, much of our understanding of behaviors of very young-term infants may be extrapolated from the healthy preterm infant population. In the 1970s, Brazelton published the Neonatal Behavioral Assessment Scale (NBAS) that classified infant behaviors while experiencing periods of stimulation or stress (Brazelton 1973) (Fig. 82.3). As part of this original research, it was postulated that the baby’s state of consciousness or state behavior was the single most important element of the NBAS examination. State behaviors are seen as reflections of autonomic nervous system maturation, as well as reactions to incoming stimuli (Brazelton and Cramer 1990). Since that time, seven distinct states behaviors have emerged in infants older than 36 weeks gestational age (Brazelton 1984; Prechtl 1987). Furthermore, it is recognized that only in an alert, bright state does the newborn orient to external sources of input and takes in information (Brazelton 1990; Kaufmann-Hayoz 1987). The “behavioral organization” model was proposed in the 1980s (Fig. 82.3). This model separates infant behavior according to distinct but interdependent neurologically based subsystems. Such subsystems included not only state behavior organization, but also autonomic, motor, attention, interaction, self-regulation, and balance systems (Als 1982, 1986; Als et al. 2005). The behavioral organization model also assumes that neurologic maturation will occur with increasing gestational age, which in turn, allows for improved interaction between systems. Although this model has primarily been applied to infants through 42 weeks gestation, we know that the modulation these subsystems continue through the period of homeostasis, which should be complete by 3 months of age (Chatoor et al. 1984; Porges 1996; Greenspan 1990). As a result of the behavioral organization model, an Assessment of Preterm Infants’ Behavior (APIB) was developed, which allows clinicians to observe and score unique physio logic behaviors related to disorganization from each of the specific subsystems. Gagging, for example, is considered a sign of autonomic stress, whereas changes in posture and muscle tone are reflective of motoric stress (Als et al. 2005). Understanding which specific subsystems are disorganized allows clinicians and caregivers to provide the appropriate support to assist the infant in regulation.
82.4 A utonomic Homeostasis, Impact on Emotional and Behavioral Development The Polyvagal Theory describes two distinct types of vagal responses that infants may present to stressors: the mammalian system and the reptilian vagal system (Porges et al. 1996). A mammalian response system is considered the optimal means for an infant to maintain physiologic homeostasis in response to stress by means of an increased arousal state. However, if this system is regularly challenged (i.e., medical interventions such as gavage feeding) then the heightened physiologic response pattern could become detrimental to the developing infant. The second type of vagal response has been deemed as the reptilian vagal system. Basically with this type of response pattern to stress, the autonomic nervous system shuts down. Clinical observations of infants who are utilizing reptilian strategies in response to stress can involve potentially lethal bradycardia (slow heart rate) and/or apnea. Therefore, depending upon the vagal system that an
82 Abnormal Physiologic Responses to Touch in Feeding Difficulties
1277
Fig. 82.4 Infant stress responses. Basic representation of the Polyvagal theory, which indicates that two distinct types of vagal response patterns will evoke a heightened or reduced physiologic state
infant uses to respond to stress, either heightened or reduced physiologic states will be observed (Porges 1996, 2009) (Fig. 82.4). Based on the Polyvagal Theory, a hierarchical model was developed that identified the importance of neural regulation of autonomic homeostasis as a precursor to emotional, cognitive, and behavioral regulation (Porges 1996). At the foundation of this model (Level 1), infants must be able to achieve internal physiologic homeostasis through autonomic regulation between the brainstem and peripheral organs, before attempting to cope with external factors. Once infants are able to successfully regulate single physiologic systems, the infants can then begin to coordinate multiple systems and respond to external challenges (Level II) (Fig. 82.3). Thus, the successful completion of a complex activity (such as feeding), which requires an infant to both negotiate internal homeostasis and with external demands, are dependent on regulation of autonomic system. Further, this model also has provided a means for studying long-term effects of abnormal autonomic nervous system functioning particularly related emotional dysregulation in preschool to adolescence (Beauchaine et al. 2007) Thus, the polyvagal theory has been ground-breaking for providing a means to identify autonomic nervous system regulation as an integral part of healthy cognitive, emotional, and behavioral development and for showing that difficulties with physiologic homeostasis can persist well beyond the first year of life.
82.5 Touch and Feeding and Interruption of the Period of “Homeostasis” The behavioral response pattern of neonates and full-term infants younger than 3 months of age includes a variety of autonomic nervous system responses. For example, it is not uncommon to observe gagging and/or “state” behavior changes (such as uncontrolled crying, fussiness, drowsiness, or falling asleep) in young infants while processing touch input. Touch input refers both to touch to the body such as general physical handling and/or oral touch associated with feeding. Over time as the infant develops beyond the period of “homeostasis” (3 months of life) these autonomic nervous system
1278
D. Scarborough
Fig. 82.5 Interrupting homeostasis. If normal development is interrupted at any point during the period of homeostasis (including the preterm period) then aberrant autonomic behaviors during feeding may persist well beyond the first year of life
responses become less generalized and more appropriate to environmental situations. For instance, gagging is a normal protective reflex when elicited by touch to the posterior oral regions; however, gagging to anterior oral touch or touch to more peripheral areas of the body is seen as a sign of immature physiological (autonomic) processing. Thus, in typical infants older than 3 months of age autonomic nervous system responses to touch are unusual (Scarborough et al. 2002, 2006). Full-term infants and children who have had an interruption of oral feeding during at least 2 weeks during the period of “homeostasis” (first 3 months of life) have been found to respond to graded tactile input with signs of poor autonomic regulation (such as gagging and major state changes) (Fig. 82.5). In particular, these aberrant autonomic responses have been documented from touch to both nonoral body regions (i.e., shoulders) and to regions in the anterior portion of the mouth and are not evident in age matched healthy peers. Further, these aberrant autonomic behaviors have been observed in children as old as 18 months of age and tend to persist even with behavioral intervention (Scarborough et al. 2006).
82.6 Clinical Application Like many areas of pediatrics, one of the challenges for the professional is making clinical judgments based on observed behaviors that could be the result of more than one etiology. As an example, a “hypersensitive” gag reflex is a regularly observed behavior within the pediatric feeding population (Fig. 82.6). A hypersensitive gag reflex will be defined for the purpose of this section as triggering a gag reflex from nonoral regions of the body or within the anterior 2/3 of the oral cavity regardless of the strength of the motor response (Scarborough et al. 2008). In the case of a hypersensitive gag reflex, obtaining the child’s past medical history is a critical component to determining the cause of the behavior. For instance, a hypersensitive gag reflex observed in premature and full-term medically fragile infants who have a history of tube feedings during the period of homeostasis is likely the result of abnormal autonomic nervous system development (Scarborough and Isaacson 2006; Scarborough et al. 2006). In contrast, children who have a history of traumatic brain injury, specifically involving bilateral corticobulbar tracts demonstrate a “hypersensitive” gag reflex due to the loss of upper motor neuron inhibition (Schulze-Delrieu and Miller 1997). In other children, a hypersensitive gag reflex has been reported as a result of maladaptive parent–child interactions through conditioned negative responses (Byars et al. 2003). Therefore, when a hypersensitive gag reflex is observed during a feeding trial, it is critical that the clinician not assume an underlying cause of this behavior regardless of the age of the child without a complete medical history. Once the cause(s) of the hypersensitive gag reflex is found, then a targeted treatment plan can be developed
82 Abnormal Physiologic Responses to Touch in Feeding Difficulties
1279
Fig. 82.6 Clinical decision flow chart. Clinical decisions for treatment need to look beyond age-related behaviors. This is to ensure that the underlying cause of a behavior is addressed
that appropriately remediates or offsets this negative feeding behavior. Further, if a single treatment technique is not effective for correcting the behavior, other treatments should be explored that target different levels of development.
Summary Points The period of “homeostasis” is critical for modulation of touch responses through the autonomic nervous system. • Normal early infant behaviors (under 3 months of age) related to general physical handling or oral touch during feeding may involve state behavior changes or gagging. These autonomic responses modulate to more appropriate behavioral patterns during the first 3 months of life in healthy infants. • Developmental theories of preterm infants can provide valuable information regarding full-term infant behavior. • Interruption of normal oral tactile input (i.e., tube feedings) in full-term infants during the period of homeostasis and earlier can lead to abnormal processing of touch and persistent behaviors such as gagging and extreme state behavior changes. • Behaviors related to autonomic regulation can persist well past the first year of life. • The same negative feeding behavior observed in older children may be the result of more than one cause acquired during early childhood. Therefore, assessment of feeding behaviors, such as the hypersensitive gag reflex, should include a comprehensive past medical history of the child. • Treatment techniques should be geared to remediate the underlying cause(s) of an observed feeding behavior.
1280
D. Scarborough
Explanation of Key Terms Homeostasis: A critical developmental period of neurophysiologic modulation from birth to approximately 3 months of life when an infant’s biological functions stabilize as a result of autonomic nervous system maturation. Attachment: The developmental period following homeostasis when infants respond to the outside world, thus beginning the process of forming relationships and differentiating between caregivers. Neurophysiologic modulation: The period of development that is concurrent with attachment when an infant begins to utilize multiple sensory systems to modulate motor responses such that movements become voluntary rather than reflexive without purposeful or conscious reflection. Polyvagal theory: Proposes that successful adaptation of mammals is dependent on systematic and reliable withdrawal and reengagement of the vagus nerve as a mechanism to respond to environmental demands (Porges et al. 1996). This theory describes two distinct types of vagal responses that infants may present to stressors, the mammalian system and the reptilian vagal system. For a detailed review of this theory, see Porges 1996 and Porges et al. 1996. Behavioral organization model: Separates infant behavior according to distinct but interdependent neurologically based subsystems. Such subsystems included not only state behavior organization, but also autonomic, motor, attention, interaction, self-regulation, and balance systems. The behavioral organization model also assumes that neurologic maturation will occur with increasing gestational age, which in turn, allows for improved interaction between systems. For a detailed review, see Als 1982, 1986; and Als et al. 2005. State behaviors: Different levels of infant alertness that are considered as reflections of autonomic nervous system maturation, as well as reactions to incoming stimuli. For a detailed review of state behaviors, see Brazelton 1973, 1984. Hypersensitive gag reflex: Triggering a gag reflex from nonoral regions of the body or within the anterior 2/3 of the oral cavity regardless of the strength of the motor response.
References Als H. Inf Mental Health J. 1982;3:229–43. Als H. Phys Occup Ther Pediatr. 1986;6:3–55. Als H, Butler S, Kosta S, McAnulty G. Ment Retard Dev D R. 2005;11:94–102. Beauchaine TP, Gatzke-Kopp L, Mead HK. Biol Psychol. 2007;74:174–84. Brazelton TB. Clin Dev Med. 1973;50:1–66. Brazelton TB. Clinics in developmental medicine. Philadelphia: J.B. Lippencott; 1984. Brazelton TB. Child Dev. 1990;61:1661–71. Brazelton TB, Cramer BG. The earliest relationship: parents, infants, and the drama of early attachement. Reading: Addison-Wesley; 1990. Byars KC, Burklow KA, Ferguson K, OFlaherty T,Santoro K, Kaul A. J Pediatr Gastr Nutr. 2003;37:473–80. Chatoor I, Schaefer S, Dickson L, Egan J. Pediatr Ann. 1984;13:829–43. Greenspan SI. In:Shonkoff M, editor. Handbook of early childhood intervention. New York: University Press; 1990. pp. 150–72. Greenspan SI, Lourie RS. Am J Psychiat. 1981;138:725–35. Hendry SHC, Hsiao SS, Bushnell MC. In: Zigmond MJ, Bloom FE, Landis SC, Roberts JL, Squire LR, editors. Fundamental neuroscience. San Diego: Academic; 1999. pp. 761–89. Henricson M, Berglund A-L, Maatta S, Ekman R, Segesten K. J Clin Nurs. 2008;17:2624–33. Jean A. Physiol Rev. 2001;81:929–69. Kaufmann-Hayoz R. In: GEV Stelmach PA, editor. Advances in psychobiology: psychobiology and early development. North-Holland: Elsevier Science; 1987. pp. 117–27.
82 Abnormal Physiologic Responses to Touch in Feeding Difficulties
1281
Kopp CB. Dev Psychobiol. 1982;18:199–214. Mahler M, Pine F, Bergman A. The psychological birth of the human infant. New York: Basic Book; 1975. Piaget J. The origins of intelligence in children. New York: International Universities Press; 1952. Pihko E, Lauronen L. Exp Neurol. 2004;190:S2–S7. Porges SW. Dev Psychopathol. 1996;8:43–58. Porges SW. Clev Clin J Med. 2009;76:S86–S90. Porges SW, Doussard-Roosevelt JA, Portales AL, Greenspan SI. Dev Psychobiol. 1996;29:697–712. Prechtl H. In: HS Rauh HC, editor. Psychobiology and early development. North-Holland: Elsevier Science; 1987. pp. 231–8. Scarborough DR. Consequences of interrupting normal neurophysiologic development: Impact on prefeeding skills. (Dissertation) University of Cincinnati; 2002. Scarborough DR, Boyce S, McCain G, Oppenheimer S, August A, Strinjas J. Dev Med Child Neurol. 2006;48:460–4. Scarborough DR, Isaacson LG. Clin Anat. 2006;19:640–4. Scarborough DR, Bailey-Van Kuren M, Hughes M. J Am Dent Assoc. 2008;139:1365–72. Schulze-Delrieu K, Miller RM. In: Perlman AL, Schulze-Delrieu K, editors. Deglutition and its disorders. San Diego: Singular; 1997. pp. 144–5.
wwwwwwwwwwwwwww
Part XIII
Pathology and Abnormal Aspects: Endocrine and Neuroendocrine
wwwwwwwwwwwwwww
Chapter 83
Gut Peptides and Enteral Feeding in Critically Ill Patients: Implications for Gastric Dysmotility and Appetite N.Q. Nguyen and R.H. Holloway
Abbreviations ICU CCK PYY GLP-1 IPPW MMC IL
Intensive care unit Cholecystokinin Peptide YY Glucagons-like peptide 1 Isolated pyloric pressure wave Migrating motor complex Interleukin
83.1 Introduction Nutritional deprivation or malnutrition in critically ill patients have been shown to be associated with impairment of immunological function, prolongation of mechanical ventilation, increased length of intensive care unit (ICU) and hospital stay, increased infective complications, and ultimately higher mortality (Harrington 2004). Although the practice of nutritional support by either enteral or parenteral routes has now become standard treatment, for many years it was frequently either not provided routinely (Berger et al. 1997) or inadequately delivered to meet metabolic requirements (Heyland et al. 1995, 2003). Previously, the incidence of malnutrition amongst long-stay critically ill patients has been as high as 50% (Quirk 2000). Currently, enteral nutrition is the preferred mode of nutritional delivery in critically ill patients due to the ease of administration, reduced health-care costs, lower rate of sepsis, lack of requirement for central venous access, and enhancement of gastrointestinal barrier function with the potential reduction in bacterial translocation (Heyland et al. 1998, 2003). Adequate delivery of enteral feeds to critically ill patients, however, is frequently hampered by a variety of factors; the most frequent is gastric dysmotility. A number of factors including mechanical ventilation, drugs (especially opiates and catecholamines), hyperglycemia, shock, circulating inflammatory cytokines, and the admission diagnosis have been implicated in the etiology of slow gastric emptying in critically ill patients. However, the majority of these factors were inferred from studies
N.Q. Nguyen (*) Department of Gastroenterology and Hepatology, Royal Adelaide Hospital, North Terrace, Adelaide, SA, 5000, Australia e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_83, © Springer Science+Business Media, LLC 2011
1285
1286
N.Q. Nguyen and R.H. Holloway
performed in either animals (Dubois et al. 1975) or a non-critically ill population (Yuan et al. 1998). More recently, there is evidence to suggest that gastric dysmotility of critical illness may relate to disturbance of the complex interaction between enteral nutrition and gastrointestinal hormones that involved in the enterogastric feedback regulation. The aim of this chapter is to review the gastrointestinal motor and hormonal responses to enteral nutrition in critically ill patients, and how the complex interactions between these factors can potentially give insights into the pathogenesis of impaired gastric motor function in this population.
83.2 The Role of Enteral Nutrition in Critically Ill Patients The impact of nutritional support on patients’ outcomes has been examined extensively over last two decades. While many studies have documented that nutritional support changes metabolic outcomes, such as amino acid profile and nitrogen balance, the impact of this intervention on clinically important end points such as infection, length of stay, and mortality is less clear-cut. There are a number of methodological reasons that might explain the discrepancies among the reported studies. First, over 60% of trials that have examined this issue lacked sufficient statistical power (Heyland 1998; Heyland et al. 2003; Doig et al. 2005). This is further confounded by the heterogeneity of the critically ill population and differences in the mix of patients between the studies (Doig et al. 2005). In addition, the quality of many trials was suboptimal with a lack of randomization, controls, and blinding (Doig et al. 2005). Meta-analyses that have examined only high-quality randomized–controlled (level 2) trials suggest that, compared with patients who received standard care (i.e., IV fluid support only), nutritional support via either the enteral or parenteral route is associated with a 5–12% reduction in mortality (Heyland et al. 1995, 2003; Doig and Simpson 2005). In surgical and multitrauma patients, nutritional support improves wound healing, decreases the catabolic response to injury, and reduces infective complications (Kudsk et al. 1992; Heyland 1998; Heyland et al. 1998, 2003). Similar reductions in septic complications and a shorter length of stay in ICU and hospital have also been demonstrated in medical critically ill patients (Heyland et al. 1995, 2003; Doig and Simpson 2005). The optimal time to start nutritional support in critical illness, however, is unknown. It is recognized that patients should not be deprived of nutrients for more than 5 days as earlier studies have demonstrated a higher mortality in surgical patients who had been deprived of nutrition for longer than this (Thomas and Robert 1979). In animals, nutrient deprivation for durations as short as 24 h may lead to intestinal mucosal atrophy and malabsorption (Alpers 2002). These abnormalities can be reversed or prevented by the early introduction of luminal nutrients (Alpers 2002). In a more recent randomized–controlled study, patients with nutrient deprivation for a mean duration of 4 days had a significant reduction in the villous height and crypt ratio and an increased gut permeability as assessed by the lactulose–mannitol test (Hernandez et al. 1999), compared those who received enteral feed within 24 h of admission. Until recently, the concept of early aggressive feeding was not universally accepted by many critical care staff. Despite recognizing the importance of early feeding, a number of nutritional surveys in intensive care units have reported that less than 50% of eligible patients received enteral nutrition within 48 h of admission (McClave et al. 1999; Montejo 1999; Heyland et al. 2003), placing these patients at risk of intestinal mucosal atrophy, increase gut permeability, and malnutrition (Alpers 2002). As the integrity of the gastrointestinal tract is thought to be important for the prevention of bacterial translocation and subsequent infective complications (Alpers 2002), these findings suggest that delayed enteral nutrition may increase the risk of sepsis. The impact of early enteral nutrition on outcomes in critically ill patients has been investigated in a number of randomized controlled trials (Minard et al. 2000; Artinian et al. 2006). These have produced conflicting results possibly due to a lack of adequate statistical power. A meta-analysis using
83 Gut Peptides and Enteral Feeding in Critically Ill Patients: Implications for Gastric Dysmotility
1287
these data has also given an inconclusive answer with only a trend toward a reduction in infective complications and mortality, and no difference in length of stay (Heyland et al. 2003). Most recently, Artinian et al. (2006) conducted a large study that included 4,049 medical critically ill patients. This study showed that early enteral nutrition was associated with a 20% reduction in ICU mortality and a 25% reduction in hospital mortality. The benefit to mortality was most pronounced among patients with the highest Acute Physiology and Chronic Health Evaluation II (APACHE II) scores (Artinian et al. 2006), although a higher incidence of ventilated associated pneumonia was reported in patients who had received early feeding. The reason for this finding is unclear but may relate to the more prolonged exposure to nasogastric feeds, a known risk factor for gastroesophageal reflux and aspiration (Montejo 1999; Mentec et al. 2001). Based on current evidence, a number of international working committees have published guidelines for nutritional support in critically ill patients (Heyland et al. 2003; Doig and Simpson 2005). Of these, the Canadian Clinical Practice guideline is best known and has been most widely adopted (Heyland et al. 2003). Although there are minor variations among these guidelines, common themes include (Heyland et al. 2003): • Nutritional support, either in the form of total parenteral nutrition or enteral nutrition, should be considered in all patients who are admitted to critical care. • Enteral nutrition is the preferred modality of nutritional support in critically ill patients, particularly in those with intact gastrointestinal tract. • In patients whom enteral nutrition is indicated, it should be commenced within 24–48 h of admission.
83.3 Regulation of Gastric Emptying and Enterogastric Feedback in Health In order to appreciate the potential significances of gastric dysmotility and hormonal abnormalities in response to enteral nutrition in critically ill patients, an understanding of gastric emptying regulation in health, especially those related to “enterogastric” feedback mechanisms, is required. After ingestion of a meal, the rate of gastric emptying is tightly regulated by the presence of intraluminal nutrients in the duodenum and small intestine, a process known as the enterogastric reflex (Lin et al. 1989). This feedback response is important in preventing excessive dumping of gastric contents into the small intestine and allows nutrients to enter the small intestine at a rate of 2–3 kcal/min (Lin et al. 1989).
83.3.1 Gastric Motor Responses of Enterogastric Feedback After ingestion of nutrients, the proximal stomach relaxes and acts as a reservoir. Subsequently, the fundus undergoes slow sustained contractions to distribute contents distally and is believed to have a major role in controlling emptying of liquids (Kelly 1980). In contrast, motor activity in the distal stomach is characterized by irregular contractions, which aid mixing and propagation of nutrient along the gastrointestinal tract. Mixing is due to intermittent isolated antral waves contracting against a closed pylorus. Gastric emptying of nutrient occurs predominantly in a pulsatile fashion when peristaltic wave activity continues through an open pylorus, aiding movement of contents into the duodenum (White et al. 1981). The peristaltic wave is dependent on the integration of motor activity in the proximal and distal stomach, as well as in the proximal small intestine (Horowitz et al. 1994). The rate of transpyloric flow is tightly regulated by the “enterogastric” feedback (Brener et al. 1983), whereby duodenal nutrient triggers a neurohumoral response that reduces antral waves, increases basal pyloric pressure and increases frequency of isolated pyloric pressure waves (IPPWs).
1288
N.Q. Nguyen and R.H. Holloway
83.3.2 Humoral Response of Enterogastric Feedback A large number of hormones, including cholecystokinin (CCK), peptide YY (PYY), motilin, GLP-1, and ghrelin, are involved in the “humoral” regulation of the enterogastric feedback responses (King et al. 1985; Horowitz et al. 1994). Of these, CCK and PYY are the best studied. The best-characterized CCK in humans is cholecystokinin-octapeptide (CCK-8) (Liddle et al. 1986). This peptide hormone is secreted by cells in the duodenum and upper jejunum in response to duodenal acid and nutrients, particularly lipids and proteins (Liddle et al. 1986). In humans, exogenous CCK-8 infusions that mimic plasma concentrations in the postprandial range induce significant fundic relaxation (Straathof et al. 1998), inhibit antral motility, increase basal pyloric pressure, and stimulate IPPWs (Fraser et al. 1993). These responses are abolished by coadministration of a CCK antagonist, loxiglumide (Mesquita et al. 1997). The effects of CCK on gastric emptying in humans are, however, conflicting. Although intravenous administration of both exogenous CCK-8 and CCK-3 at physiological concentrations is associated with a decrease in gastric emptying of both liquid and semisolid meals (Moran and McHugh 1982; Liddle et al. 1986), data on the impact of specific CCK antagonists on gastric emptying in humans remain controversial, which may be related to the different formulation of the CCK antagonists used among the studies. All studies that have used the intravenous form of the specific CCK antagonists, loxiglumide and lintitript, have reported an accelerated gastric emptying of both liquid and solid meal, as well as nutrient and nonnutrient meal (Fried et al. 1991; Kreiss et al. 1998). On the other hand, the two studies that used the orally administered CCK antagonists, such as MK-320 and loxiglumide, have failed to demonstrate a change in the gastric emptying rate of either solid or liquid meals, despite a positive effect on gallbladder motility (Liddle et al. 1989; Corazziari et al. 1990). Peptide YY, also known as the “ileal-brake” hormone that regulates both gastric emptying and intestinal transit, is secreted by the endocrine L cells of the small and large bowel (Adrian et al. 1985). After an intraduodenal meal, plasma PYY increases even before nutrients reach the PYYcontaining cells in the ileum, suggesting that PYY release is neurally mediated, probably via the vagus (Greeley et al. 1989). In humans, PYY 3-36 has been consistently reported to inhibit gastric emptying of liquids in a dose-dependent manner (Pironi et al. 1993). In contrast, there are limited data regarding the effects of PYY antagonists on gastric emptying in humans. Unlike CCK, the motor mechanisms underlying the slowing of gastric emptying of PYY 3-36 are less well studied. The impact of PYY on proximal gastric motor activity has not been examined. In humans, elevated fasting PYY concentrations are associated with less frequent phase 3 activity of the MMC arising from the antroduodenal region (Naslund et al. 1998), a reduction in the frequency of antroduodenal contractions (Ledeboer et al. 1999) and an increase in IPPW (MacIntosh et al. 1999). Currently, there are no studies that have examined the impact of PYY antibody on gastric motility in humans.
83.4 Upper Gastrointestinal Motility During Critical Illness Disturbances in gastrointestinal motor activity occur in up to 70% of the critically ill patients who require mechanical ventilation (Heyland et al. 1996; Kao et al. 1998; Ritz et al. 2001; Nguyen et al. 2007b, d, 2008a, b). Delayed gastric emptying occurs in 40–80% of patients, and the prevalence depends on factors such as admission diagnosis as well as the techniques used to assess gastric emptying (Heyland et al. 1996; Kao et al. 1998; Ritz et al. 2001; Nguyen et al. 2007d, 2008a). Using gastric scintigraphy, up to 80% of critically ill patients admitted with head injury have slow gastric emptying (Kao et al. 1998; Nguyen et al. 2008a). In contrast, delayed gastric emptying occurs in only 40–60% of patients in unselected cohort, when assessed by either the paracetamol absorption test or 13C-octanoic
83 Gut Peptides and Enteral Feeding in Critically Ill Patients: Implications for Gastric Dysmotility
1289
Table 83.1 Gastric motor abnormalities reported during critical illness Region of stomach Gastric motor abnormalities Proximal stomach • Delayed and prolonged fundal relaxation • Reduced fundic volume waves • Greater proximal meal retention Distal stomach • Persistent MMC during feeding • Reduction in antral contractions and antral motility index • Increased isolated pyloric contractions and pyloric tone • Inverse correlation between pyloric motor activity and gastric emptying Integration between • Lack of fundoantral motor association gastric regions • Antral contractions frequently occur in close association with pyloric contractions Duodenum • Disorganized duodenal contractions • Increased number of retrograded duodenal contractions Enterogastric feedback • Increased fasting plasma CCK and PYY levels responses • Increased nutrient-stimulated plasma CCK and PYY levels • Plasma CCK and PPY inversely correlated with rate of gastric emptying
acid breath test (Heyland et al. 1996; Tarling et al. 1997; Ritz et al. 2001; Nguyen et al. 2007d, 2008b). Delayed gastric emptying of critical illness is associated with impaired motor function in all regions of the stomach during both fasting and fed state (Dive et al. 1994; Chapman et al. 2005; Nguyen et al. 2006a, 2007a, 2008c; Chapman et al. 2008) (Table 83.1). Clinical indicators of upper gut dysmotility include high gastric residual volumes (GRV), vomiting, reflux, and aspiration (Montejo 1999).
83.4.1 Fasted State During fasting, the gastric phase of the migrating motor complex (MMC) in the antroduodenal region is significantly shorter in mechanically ventilated, critically ill patients than in controls (32 vs. 101 min) (Bosscha et al. 1998). In part, this appears to be related to a virtual absence of phase 2 of the MMC and a marked reduction in antral pressure waves (Dive et al. 1994; Bosscha et al. 1998; Chapman et al. 2005), whereas the lengths of phase 1 and 3 do not differ significantly from those of healthy volunteers (Bosscha et al. 1998). Although proximal gastric volume, pyloric tone, and the frequency of IPPWs are similar between critically ill patients and healthy subjects, the frequency of fundic volume waves is significantly lower in critically ill patients (Dive et al. 1994; Chapman et al. 2005; Nguyen et al. 2006a, 2007a).
83.4.2 Intragastric Nutrient Stimulation During intragastric feeding, there is persistence of the gastric and small intestinal interdigestive pattern but less than 10% of the phase 3 activity fronts of MMC originate in the stomach (Bosscha et al. 1998). In a small observational study, a persistent fasting pattern was observed in one half of the patients and a mixed pattern of fasting and postprandial motility was seen in the other half (Bosscha et al. 1998). The number of antral pressure waves (Dive et al. 1994; Chapman et al. 2005) and the antral motility index (Bosscha et al. 1998) are also markedly reduced in critically ill patients during gastric meal, and there is a negative correlation between the antral motility index and gastric retention during gastric feeds (Bosscha et al. 1998).
1290
N.Q. Nguyen and R.H. Holloway
83.4.3 Intraduodenal Nutrient Stimulation More marked gastrointestinal motor abnormalities are observed during direct small intestinal feeding (Chapman et al. 2005). Relaxation of the proximal stomach is delayed although the magnitude of the relaxation is normal. Fundic slow wave activity is reduced and the recovery of proximal gastric volume to prestimulation levels is delayed. Antral motility is also reduced (Chapman et al. 2005) and is associated with an increase in isolated pyloric activity (Chapman et al. 2005). There is an inverse correlation between pyloric motor activity and gastric emptying (Chapman et al. 2005), suggesting that the increase in localized pyloric motor activity may contribute to the slowing of gastric emptying in critical illness. More importantly, these adverse motor changes are still observed in the critically ill patients even when the infusion rate is as low as 1 kcal/min, which does not alter distal gastric motor activity in healthy subjects (Chapman et al. 2005). These findings suggest that the gastrointestinal tract of critically ill patients responds to small intestinal nutrients differently from normal and the sensitivity of enterogastric feedback response to small intestinal nutrients in critical illness is enhanced.
83.4.4 Antrofundic Motor Integration In addition to the generalized gastric dysmotilities, the association between the two gastric regions (i.e., antrofundic motor integration) has been shown recently to be disturbed. Antral contractions frequently occur in close association with pyloric contractions, particularly during infusion of a high nutrient load (2 kcal/min) (Nguyen et al. 2008c). Given the importance of this gastric regional association to the redistribution of proximal gastric content distally, the disruption to this antrofundic integration is likely to provide an explanation for the significantly greater meal retention in the proximal stomach in these patients (Nguyen et al. 2008a), especially in those with delayed gastric emptying.
83.4.5 Duodenal Motility Although duodenal activity usually persists, the organization of the waves is abnormal with an increased proportion of retrograde contractions (Chapman et al. 2008). This has been reported to be associated with impaired transpyloric flow and slowing of gastric emptying (White et al. 1981). Disruption of the organization of duodenal contractions may also contribute to duodenogastric reflux and thereby, bile-induced esophagitis (Dive et al. 1999), which is common in these patients.
83.5 Hormonal Responses to Enteral Nutrition in Critically Ill Patients Given the known abnormalities of gastric motor activity and enterogastric feedback responses in critically ill patients, it is not surprising that the gastrointestinal hormonal responses that regulate gastric motility through the enterogastric feedback mechanisms are also significantly disturbed (Nguyen et al. 2006b, 2007c, 2008b). Compared to healthy volunteers, both fasting and duodenal nutrientstimulated plasma CCK and PYY concentrations are significantly elevated (Fig. 83.1) (Nguyen et al. 2006b, 2007c, 2008b). More importantly, plasma concentrations of CCK and PYY are greatest in patients with delayed gastric emptying (Fig. 83.2) and who do not tolerate enteral feeding (Fig. 83.3),
83 Gut Peptides and Enteral Feeding in Critically Ill Patients: Implications for Gastric Dysmotility
1291
Fig. 83.1 Plasma CKK (a) and PYY (b) concentrations during fasting and 60 min after duodenal nutrient infusion at 1 and 2 kcal/min in healthy subjects () and critically ill patients (). *P < 0.05, and **P < 0.001, vs. healthy subjects (Data adapted from Nguyen et al. 2006b, 2007c)
a clinical manifestation of slow gastric emptying (Nguyen et al. 2006b, 2007c, 2008b). As observed in healthy subjects (Nguyen et al. 2006b, 2008b), the elevation of plasma PYY concentration is observed within 20 min of nutrient stimulation in critically ill patients (Nguyen et al. 2006b), suggesting that the elevated PYY concentrations are most probably mediated by factors in the proximal small intestine rather than direct nutrient stimulation of the distal ileum (Adrian et al. 1985). CCK is likely to be an important “proximal” mediator given its known stimulatory effect on the release of PYY in the small intestine (Lin et al. 2000) and a positive correlation between the hormones have been demonstrated in these patients (Fig. 83.4) (Nguyen et al. 2006b, 2008b). More importantly, the elevated fasting plasma PYY concentrations in these patients during the first week of admission have been shown to normalize 3 weeks after discharge from the ICU (Nematy et al. 2005). Preliminary data also suggest that plasma concentration of ghrelin, an enterogastrone that stimulates gastric motility and increases gastric emptying (Murray et al. 2005), is also disturbed in these patients with a reduced fasting level during critical illness and returned to normal level as the patient recovers from their illness (Nematy et al. 2005).
1292
N.Q. Nguyen and R.H. Holloway
Fig. 83.2 Plasma concentrations of CCK (a) and PYY (b) during fasting and 120 min after an intragastric feed (100 kcal) in critically ill patients with normal () and delayed () gastric emptying (GE). *P < 0.05, vs. patients with normal GE (Data adapted from Nguyen 2008b)
Fig. 83.3 Plasma CCK (a) and PYY (b) concentrations 60 min after intraduodenal nutrient infusions, at rate of 1 and 2 kcal/min, in critically ill patients who tolerated () and did not tolerate () enteral feeding. *P < 0.01 vs. feed tolerance (Data adapted from Nguyen et al. 2006b, 2007c)
The mechanisms responsible for the abnormally high plasma CCK and PYY concentrations during fasting and in response to small intestinal nutrient infusion in critically ill patients remain unclear. Recent data suggest that the presence of inflammation may have a role in the regulation of entero endocrine cells in the small intestine (McDermott et al. 2006). In a mouse model, upper gut inflammation has been reported to increase plasma CCK concentrations and reduce energy intake, via an
83 Gut Peptides and Enteral Feeding in Critically Ill Patients: Implications for Gastric Dysmotility
b
Fasting [PYY] (pmol/L)
80
P=0.02 r = 0.52 60
40
20
0
0
5
10
15
Fasting [CCK] (pmol/L)
20
Integrated plasma [PYY] (pmol/L.min)
a
1293
6000
P 40 kg/m² without comorbidities or between 35 and 40 kg/m² in patients with significant comorbidities. In view of its beneficial effects in improving glycemic control, it may be appropriate to offer bariatric surgery to patients with lower BMIs who have comorbid T2DM, especially in cases of significant insulin resistance or when simple weight-loss methods and pharmacological treatment have proved ineffective. In future pharmacological approaches that mimic the physiological changes seen after bariatric surgery could prove beneficial in treating patients with obesity and T2DM. For example, therapeutic agents based on targeted bile acid delivery to the distal small intestine may reduce appetite, stimulate weight loss, and improve glycemic control in patients with obesity and T2DM. Pharmacological agonists of receptors activated by bile acid-dependent pathways, for example TGR5 and FXRa could be successful in improving glycemic control. Activation of TGR5 increases energy expenditure in vivo and in cultured cells (Watanabe et al. 2006) and increases GLP-1 in cell lines (Katsuma et al. 2005). Further work in vivo is required to determine whether such agents are efficacious in humans. One limitation of many studies that demonstrate increased gut hormone concentrations after bariatric surgery is that they have been conducted on nondiabetic subjects. These results have then been used as a possible explanation for the mechanisms of improved glycemic control in diabetic patients after bariatric surgery. Therefore, further research needs to be done directly on diabetic subjects to validate this mechanism in the diabetic population and to investigate the role that duration and severity of diabetes may play in determining the success of treating T2DM with bariatric surgery.
85 The Role of Bile Acids in Gut-Hormone-Induced Weight Loss After Bariatric Surgery
1327
Summary Points • Gut hormones play an integral role in appetite control and food intake. PYY and GLP-1 act in the hypothalamus to suppress appetite. GLP-1 also has an important incretin effect, thereby improving glycemic control. • PYY and GLP-1 are both secreted by endocrine L-cells predominantly found in the distal small intestine. Both nutrients and bile acids are known stimulators of PYY and GLP-1 secretion. • Roux-en-Y gastric bypass (RYGBP) is the most frequently performed and most successful surgery to treat obesity. It results in food bypassing 95% of the stomach, the entire duodenum, and a section of the jejunum. The mechanisms of weight loss following RYGBP are incompletely understood. • RYGBP can dramatically improve glycemic control in subjects with T2DM. These improvements occur before significant weight loss has occurred. It has been proposed that these immediate improvements in diabetic control may be the result of altered gut hormone profiles following bariatric surgery. • Following RYGBP PYY and GLP-1 concentrations are increased. It is likely that altered intestinal anatomy following RYGBP enables quicker transit and greater exposure of both nutrients and bile acids to the distal terminal ileum resulting in increased L-cell stimulation and PYY and GLP-1 secretion. • Understanding the mechanism of weight loss and improved glycemic control after bariatric surgery has important implications for the future treatment of obesity and T2DM. Pharmacological compounds that stimulate bile acid-dependent pathways are a possible therapeutic option in improving metabolic conditions such as T2DM. Definitions and Explanations of Key Terms and Words Anorexigenic: Decreasing or inhibiting appetite. Bariatric surgery: Surgical intervention for the treatment of morbid obesity. Bile acids: Steroid acids found predominantly in bile. The two major bile acids are cholic acid and chenodeoxycholic acid. The main function of bile acids is to facilitate the formation of micelles, which promotes processing of dietary fat. Bile salts: Bile acids conjugated to glycine or taurine. In humans, taurocholic acid and glycocholic acid (derivatives of cholic acid) represent approximately 80% of all bile salts. Bile salts have an enhanced amphipathic nature to enable greater emulsifying activity for more effective dietary lipid absorption in the GI tract. Ghrelin: A 28-amino acid peptide produced predominantly by X/A-like cells of the mucosal layer in the fundus of the stomach. Ghrelin is the only known orexigenic hormone, and plays a role in regulating premeal hunger and meal initiation as well as long-term energy balance. Ghrelin also stimulates growth hormone release, lactotroph and corticotroph secretion, gastric motility and has cardiovascular effects. Glucagon-like peptide-1: A 30-amino acid peptide, which is cleaved from its precursor preproglucagon. GLP-1 inhibits gastric motility, reduces gastric emptying, and inhibits gastric acid secretion. GLP-1 also has an important incretin effect and suppresses glucagon secretion, enhances glucose disposal, and inhibits food intake. Incretins: A group of GI hormones that enhance glucose-dependent insulin secretion from beta cells of the islets of Langerhans after eating, thereby improving glucose tolerance. L-cells: The most abundant endocrine cell type in the distal small intestine. The apical surface of L-cells has microvilli, which are in contact with the intestinal lumen allowing them to sense nutrients.
1328
R.E. Roberts et al.
The basal surface of L-cells is rich in endocrine granules allowing secretion of hormones into the circulation. L-cells secrete peptide YY (PYY), proglucagon-derived peptides such as glucagonlike peptide 1 (GLP-1), glucagon-like peptide 2 (GLP-2), glicentin, and oxyntomodulin (OXM). Orexigenic: Increasing or stimulating appetite. Peptide YY: A 36-amino acid GI hormone, where Y depicts the abbreviation for tyrosine. It is a member of the pancreatic polypeptide family and mediates its effects through G-protein linked NPY receptors. PYY has an important role in appetite control. It stimulates satiety and reduces food intake. It also inhibits transit through the proximal small intestine, delays gastric emptying, and inhibits gallbladder emptying. Roux-en-Y gastric bypass: Division of the stomach into a small proximal pouch and a larger distal portion, which is bypassed. The small pouch is anastomosed to the distal jejunem. The remaining distal portion of the stomach, the duodenum, and proximal jejunem are reattached to the GI tract distal to the gastrojejunostomy to allow for excretion of GI and pancreatic juices. It is a type of bariatric surgery. Type 2 diabetes mellitus: A disorder characterized by high blood glucose in the context of insulin resistance and relative insulin deficiency.
Key Facts About PYY • PYY is a 36-amino acid GI hormone, where Y depicts the abbreviation for tyrosine. PYY exists in two forms – the full-length peptide PYY1-36, which is truncated to the biologically active PYY3-36 by dipeptidyl peptidase-IV. PYY3-36 is the predominant form. • PYY is a member of the pancreatic polypeptide family, which includes pancreatic polypeptide (PP) and neuropeptide Y (NPY), which mediate their effects through G-protein-linked NPY receptors of which there are several subtypes (Y1R, Y2R, Y4R, and Y5R represent fully defined subtypes). • PYY is secreted by L-cells of the distal gut, together with glucagon-like peptide and oxyntomodulin. Peripheral neurons, especially enteric neurons, also express PYY, as do a restricted set of central neurons. • Secretion of PYY in the GI tract is primarily stimulated by the presence of nutrients (mainly lipids and protein) in the gut lumen, and is proportional to the caloric density of the meal ingested. Other stimulants of PYY release include intraluminal bile acids, gastric acid, and cholecystokinin. Peak plasma concentrations of PYY occur in the second hour following food ingestion. • PYY reduces food intake, inhibits transit through the proximal small intestine, delays gastric emptying, and inhibits gallbladder emptying. • PYY secretion stimulates vagal and somatosensory afferent fibers arising in the GI tract and terminating at the nucleus tractus solitarius (NTS) of the brainstem, which transmit information pertaining to recent food intake. Neurons from the NTS can then relay this information to the ARC of the hypothalamus, a key component of the forebrain pathway involved in appetite control. PYY can also act directly in the brain via the blood, entering at areas where the blood– brain barrier is deficient. • In obese humans, plasma concentrations of PYY are reduced. Genetic variations in PYY and NPY receptor Y2R genes may contribute to obesity and are associated with the severe obesity of Pima Indian men. • In contrast, PYY concentrations have been found to be increased in disease states characterized by significant weight loss, such as anorexia nervosa, celiac disease, inflammatory bowel disease, and cardiac cachexia.
85 The Role of Bile Acids in Gut-Hormone-Induced Weight Loss After Bariatric Surgery
1329
References Adrian TE, Ballantyne GH, Basson MD, Bilchik AJ, Graham S, Longo WE, Modlin IM, Tierney RP Gut 1993;34:1219–24. Angelin B, Björkhem I, Einarsson K, Ewerth S J Clin Invest. 1982;70:724–31. Anini Y, Chariot J, Cuber JC, Fu-Cheng X, Kervran A, Roz C Pflugers Arch. 1999;438:299–306. Aponte GW, Fink AS, Meyer JH, Tatemoto K, Taylor IL Am J Physiol. 1985;249:G745–50. Armstrong DN, Ballantyne GH, Krenz HK, Modlin IM Gut 1993;34:483–8. Ballantyne GH, Adrian TE, Bilchik AJ, Longo WE, Modlin IM, Savoca PE, Sussman J, Vukasin AP Am J Physiol. 1989;257:G715–24. Batterham RL, Bloom SR, Cohen MA, Ellis SM, Frost GS, Ghatei MA, le Roux CW, Withers DJ N Engl J Med. 2003;349:941–8. Biertho L, Gagner M, Herron D, Horber FF, Inabnet WB, Pomp A, Ricklin T, Steffen R J Am Coll Surg. 2003;197:536–44. Borg CM, Aylwin SJB, Bloom SR, Ghatei MA, le Roux CW, Patel AGBr J Surg. 2006;93:210–5. Bray GA, Gallagher TFJ Lancet. 1968;1:1066–7. Brolin RE. JAMA 2002;288:2793–96. Brolin RE, Cody RP, Kenler HA, LaMarca LB J Gastrointest Surg. 2002;6:195–203. Bryant MG, Bloom SR, Demling L, Domschke S, Domschke W, Hobbs S, Mitznegg P, Polak JM, Ruppin H Gut 1983;24:114–9. Buchwald H, Avidor Y, Braunwald E, Fahrbach K, Jensen MD, Pories W, Schoelles K JAMA 2004;292:1724–37. Chaikomin R, Doran S, Feinle-Bisset C, Horowitz M, Jones KL, O’Donovan D, Rayner CK Am J Physiol Endocrinol Metab. 2005;289:E504–7. Champe PC, Harvey RA, Ferrier DR In: Champe PC, Harvey RA, editors. Biochemistry. Philadelphia: Lippincott Williams & Wilkins; 2005. p. 217–42. Chan JL, Goldfine AB, Mantzoros CS, Mun EC, Stoyneva V Obesity (Silver Spring) 2006;14:194–8. Clements RH, Gonzalez QH, Laws HL, Long CI, Wittert G Am J Surg. 2004;70:1–4. Cohen DE. In: Zakim D, Boyer TD, editors. Hepatology: a textbook of liver disease. Philadelphia: Saunders; 2003. p. 1713–43. Copeland P, Davis P, Kaplan L Obes Res. 2003;11:A17. Couce M, Burguera B, Cottam D, Esplen J, Schauer PR, Teijeiro R Obes Res. 2003;11:A35. Cummings DE, Breen PA, Dellinger EP, Frayo RS, Ma MK, Purnell JQ, Weigle DS N Engl J Med. 2002;346:1623–30. Cummings DE, Foster-Schubert KE, Overduin J J Clin Endocrinol Metab. 2004;89:2608–15. DePaula AL, Halpern A, Macedo AL, Machado CA, Rassi N, Schraibman V, Silva LQ Surg Endosc. 2008;22:706–16. Dietschy JM. J Lipid Res. 1968;9:297–309. Faraj M, Cianflone K, Jones P, Sniderman AD J Lipid Res. 2001;42:571–80. Faraj M, Blank D, Cianflone K, Havel PJ, Phelis S, Sniderman AD J Clin Endocrinol Metab. 2003;88:1594–602. Fonseca VA, Jones MR, Rosenstock J, Truitt KE, Wang AC Diabetes Care. 2008;31:1479–84. Fruhbeck G, Diez CA, Gil MJ N Engl J Med. 2004;350:308–9. Garg A, Grundy SM Ann Intern Med. 1994;121:416–22. Geloneze B, Geloneze SR, Pareja JC, Pilla VF, Repetto EM, Tambascia MA Obes Surg. 2003;13:17–22. Goldberg RB, Fonseca VA, Jones MR, Truitt KE Arch Intern Med. 2008;168:1531–40. Gupta S, Dent P, Hylemon PB, Stravitz RT J Biol Chem. 2001;11:276. Halmi KA, Falk JR, Mason E, Stunkard A Int J Obes. 1981;5:457–64. Han SI, Dent P, Fang Y, Grant S, Gupta S, Hylemonm PB, Li W, Qiao L, Studer E Hepatology 2004;39:456–63. Hertz AF. Ann Surg. 1913;58:466–72. Holdstock C, Engstrom BE, Karlsson FA, Lind L, Obrvall M Sundbom M J Clin Endocrinol Metab. 2003;88:3177–83. Holt JA, Billin AN, Bisi J, Donahee M, Goodwin B, Jones SA, Kliewer SA, Kozarsky KF, Luo G, Mansfield TA, McNeill YY, Wang DY Genes Dev. 2003;17:1581–91. Houten SM, Auwerx J, Watanabe M EMBO J. 2006;25:1419–25. Ikemoto S, Ezaki O, Higuchi M, Itakura H, Kawanaka K, Maruyama K, Tabata I, Takahashi M, Tange T, Tsunoda N, Yamamoto TT Am J Physiol. 1997;273:E37–45. Inagaki T, Choi M, Cummins CL, Gerard RD, Goodwin B, Jones SA, Kliewer SA, Luo G, Mangelsdorf DJ, McDonald JG, Moschetta A, Peng L, Repa JJ, Richardson JA Cell Metab. 2005;2:217–25. Katsuma S, Hirasawa A, Tsujimoto G Biochem Biophys Res Commun. 2005;329:386–90. Kawamata Y, Fujii R, Fujino M, Fujisawa Y, Fukusumi S, Habata Y, Harada M, Hinuma S, Hosoya M, Itoh T, Miwa M, Shintani Y, Yoshida H J Biol Chem. 2003;278:9435–40. Kenler HA, Brolin RE, Cody RP Am J Clin Nutr. 1990;52:87–92. Korner J, Bessler M, Cirilo LJ, Conwell IM, Daud A, Restuccia NL, Wardlaw SL J Clin Endocrinol Metab. 2005;90:359–65.
1330
R.E. Roberts et al.
Korner J, Bessler M, Conwell IM, Daud A, Inabnet W, Olivero-Rivera L, Restuccia NL, Taveras C Obesity 2006;14:1553–61. Korner J, Inabnet W, Febres G, Conwell IM, McMahon DJ, Salas R, Taveras C, Schrope B, Bessler M Int J Obes. 2009;33:786–95. Leonetti F, Bacci V, Basso MS, Basso N, Di Mario U, Lacobellis G, Lannucci CV, Perrotta N, Ribaudo MC, Silecchia G, Tiberti C, Zappaterreno A J Clin Endocrinol Metab. 2003;88:4227–31. le Roux CW, Bloom SR, Fandriks L, Ghatei MA, Kokkinos A, Laurenius A, Lonroth H, Olbers T, Osborne A, Welbourn R, Werling M Ann Surg. 2007;246:780–5. MacLean LD, Nohr CW, Rhode BM J Gastrointest Surg. 2001;5:525–30. Makishima M, Hull MV, Learned RM, Luk A, Lustig KD, Mangelsdorf DJ, Okamoto AY, Repa JJ, Shan B, Tu H Science 1999;284:1362–5. Mangelsdorf DJ, Beato M, Blumberg B, Chambon P, Evans RM, Herrlich P, Kastner P, Mark M, Schutz G, Thummel C, Umesono K Cell 1995;83:835–9. Maruyama T, Itadani H, Miyamoto Y, Nakamura T, Okada H, Tamai Y, Tanaka K, Sugiyama E Biochem Biophys Res Commun. 2002;298:714–9. Mason EE. Obes Surg. 1999;9:223–8. Moo TA, Rubino F Curr Opin Endocrinol Diabetes Obes. 2008;15:53–8. Naslund E, Backman L, Gryback P, Hellstrom PM, Holst JJ, Jacobsson H, Theodorsson E Int J Obes Relat Metab Disord. 1997;21:387–92. Parks DJ, Blanchard SG, Bledsoe RK, Chandra G, Consler TG, Kliewer SA, Lehmann JM, Moore DD, Stimmel JB, Willson TM, Zavacki AM Science 1999;284:1365–8. Patriti A, Donini A, Facchiano E Ann Surg. 2004a;240:388–9. Patriti A, Donini A, Facchiano E, Gulla N, Sanna A Obes Surg. 2004b;14:840–8. Patriti A, Annetti C, Aisa MC, Donini A, Facchiano E, Fanelli C, Galli F Obes Surg. 2005;15:1258–64. Patriti A, Aisa MC, Annetti C, Donini A, Ferri I, Galli F, Gullà N, Sidoni A Surgery 2007;142:74–85. Patti ME, Houten SM, Bianco AC, Bernier R, Larsen PR, Holst JJ, Badman MK, Maratos-Flier E, Mun EC, Pihlajamaki J, Auwerx J, Goldfine AB. Obesity 2009; Epub ahead of print. Plaisancié P, Chayvialle JA, Cuber JC, Dumoulin V J Endocrinol. 1996;151:421–9. Pories WJ, Barakat HA, Brown BM, deRamon RA, Dolezal JM, Israel G, Long SB, MacDonald KG, Morris PG, Swanson MS Ann Surg. 1995;222:339–50. Qiao L, Amorino G, Dent P, Engelhardt JF, Fang Y, Gilfor D, Grant S, Gupta S, Han SI, Hylemon PB, Park JS, Sealy L, Valerie K Mol Cell Biol. 2003;23:3052–66. Rand CS, Hankins GC, Macgregor AM South Med J. 1987;80:961–4. Saber AA, Elgamal MH, McLeod MK Obes Surg. 2008;18:121–8. Schwartz MW, Morton GJ Nature 2002;418:595–7. Shaham O, Carr SA, Gerszten RE, Lewis GD, Mootha VK, Ricciardi C, Thadhani R, Vasan RS, Wang TJ, Wei R Mol Syst Biol. 2008;4:214. Soper NJ, Brown ML, Chapman NJ, Go VL, Kelly KA, Phillips SF Gastroenterology 1990;98:111–6. Stoeckli R, Chanda R, Keller U, Langer I Obes Res. 2004;12:346–50. Strader AD, D’Alessio DA, Jandacek RJ, Seeley RJ, Vahl TP, Woods SC Am J Physiol Endocrinol Metab. 2005;288:E447–53. Tadross JA, le Roux CW Int J Obes. 2009;33:S28–32. Tomlinson E, French D, Fu L, Huang X, Hultgren B, John L, Powell-Braxton L, Renz M, Stephan JP, Stewart TA, Tsai SP Endocrinology 2002;143:1741–7. Tritos NA, Bertkau A, Goldfine A, Grayson R, Maratos-Flier E, Mun E Obes Res. 2003;11:919–24. Trostler N, Avinoach E, Charuzi II, Mann A, Zilberbush N Obes Surg. 1995;5:39–51. Vendrell J, Broch M, Gómez JM, Gutiérrez C, Molina A, Richart C, Simón I, Soler J, Vilarrasa N Obes Res. 2004;12:962–71. Vidal J, Casamitjana R, Gomis R, Moize V, Morinigo R Obes Res. 2003;11:A9. Vincent RP, le Roux CW Clin Endocrinol (Oxf). 2008;69:173–9. Watanabe M, Auwerx J, Bianco AC, Christoffolete MA, Ezaki O, Harney JW, Houten SM, Kim BW, Kodama T, Mataki C, Messaddeq N, Sato H, Schoonjans K Nature 2006;439:484–9. Wen J, Holst JJ, Kost LJ, Phillips SF, Sarr MG Am J Physiol Gastrointest Liver Physiol. 1995;269:G945–52. WHO. Obesity and overweight: fact sheet No 3111. 2006.http://www.who.int/mediacentre/factsheets/fs311/en/index. html. Yamagata K, Daitoku H, Fukamizu A, Hirota K, Ishida J, Matsuzaki H, Shimamoto Y J Biol Chem. 2004;279:23158–65.
Chapter 86
Orlistat and the Influence on Appetite Signals Mark Ellrichmann
Abbreviations AgRP Alpha-MSH Apo A-IV BMI BNRP CART CCK CCK1R CNS CRF DPP-4 FFA GHRH GHS GI GIP GIPR GLP-1 GLP-2 kJ MCH MG NPY OLETF OXM PP POMC PYY
Agouti-related peptide Alpha-melanocyte-stimulating hormone Apolipoprotein A-IV Body mass index Bombesin/bombesin-related peptide Cocaine- and amphetamine-regulated transcript Cholecystokinin CCK-1 receptor Central nervous system Corticotropin-releasing factor Dipeptidyl-peptidase-4 Free fatty acid Growth hormone-releasing hormone Growth hormone secretagogue Gastrointestinal Gastric inhibitory polypeptide Gastric inhibitory polypeptide receptor Glucagons-like peptide 1 Glucagon-like peptide 2 Kilo joule Melanin-concentrating hormone Monoglyceride Neuropeptide Y Otsuka-Long-Evans-Tokushima-Fatty rat Oxyntomodulin Pancreatic polypeptide Pro-opiomelanocortin Peptide YY
M. Ellrichmann (*) Department of Medicine I, University Hospital Schleswig Holstein, Campus Kiel, Schittenhelmstr 12, 24105 Kiel, Germany e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_86, © Springer Science+Business Media, LLC 2011
1331
1332
THL TG VAS
M. Ellrichmann
Tetrahydrolipstatin Triglyceride Visual analogue scales
86.1 Introduction Over the past decades, the prevalence of overweight and obesity worldwide has reached epidemic proportions. This has been correlated with various comorbidities, the most relevant ones being diabetes mellitus, arterial hypertension, and cardiovascular diseases. The management of obesity has become a modern challenge since nonpharmacological methods have demonstrated only short-term efficacy. Current obesity guidelines recommend that drug therapy should be considered in conjunction with nonpharmacological strategies for patients with a body mass index (BMI) greater than 30 kg/m2 or a BMI of 25–30 kg/m2 with one or more obesity-related disorders (Pi-Sunyer 1998). Antiobesity drugs can be classified into two categories: (1) drugs that suppress appetite, increase satiety or thermogenesis, primarily by modifying central nervous system (CNS) neurotransmitters; and (2) inhibitors of intestinal fat absorption.
86.2 Orlistat Orlistat, also known as tetrahydrolipstatin (THL), is a covalent inhibitor of digestive lipases derived from lipstatin, a natural product from Streptomyces toxytricini. It is an active site-directed inhibitor that reacts with the nucleophilic serine residue from the catalytic center of intestinal lipases. By covalently blocking the active site of lipases, orlistat inhibits the hydrolysis of dietary triglycerides (TG) and thus reduces the intestinal absorption of the lipolysis products monoglycerides (MG) and free fatty acids (FFAs) (Lucas et al. 2001) (Fig. 86.1). The formulation of the drug was licensed by the European Union in 1998 and in the United States in 1999 for the treatment of morbid obesity. At the standard prescription dose of 120 mg three times daily before a meal, orlistat prevents approximately 30% of dietary fat from being absorbed (Zhi et al. 1999). Historically, orlistat has been available by prescription only, whereas nowadays certain formulations of orlistat at a dose of 60 mg have been approved in Australia, the European Union, and the Unites States (Table 86.1). The primary side effects of the drug are gastrointestinal (GI)-related, and include steatorrhea, fecal incontinence, frequent or urgent bowel movements, and flatulence. Side effects are most severe when beginning therapy and may decrease in frequency with time. This is supported by the XENDOS study, which showed that only 36% of people had GI adverse effects during their fourth year of taking orlistat, whereas 91% of the study population had experienced at least one GI-related side effect within the first month of treatment (Torgerson et al. 2004). A recent meta-analysis by Padwal et al. included 11 orlistat studies with a follow-up period of at least 1 year. These trials comprised a total of 6,021 participants with an average BMI of 35.7 kg/m2. The dose of orlistat used in all studies was 120 mg t.i.d. according to the recommended standard dose in clinical practice. In all reported studies a greater weight loss was reported in the orlistat group compared to placebo. Pooled results revealed a weight loss of 2.7 kg (95% CI: 2.3–3.1 kg; 11 studies) or 2.9% (95% CI: 2.3–3.4%; ten studies) with orlistat treatment (Padwal et al. 2003).
86 Orlistat and the Influence on Appetite Signals
1333
Fig. 86.1 Orlistat – mode of action. Oral administration of orlistat inhibits intestinal lipase and thus prevents dietary triglycerides (TG) from being hydrolyzed in absorbable free fatty acids (FFA) and monoglycerides (MG). About 30% of dietary TG are excreted undigested Table 86.1 Key facts of orlistat Orlistat • Tetrahydrolipstatin • Antiobesity drug • Covalently blocks intestinal lipases • Reduces hydrolysis of TG in MG and FFA by 30% • 120 mg t.i.d. prescription dose • 60 mg t.i.d. over-the-counter dose • Negligible bioavailability • Approved in the EU in 1998, in the US 1999 TG triglycerides, MG monoglycerides, FFA free fatty acids
86.3 Orlistat and Satiety The regulation of body weight is precisely controlled by a variety of gut hormones and peripheral as well as central signals that influence hypothalamic, limbic, and brainstem circuits controlling appetite and energy expenditure. These include cholecystokinin (CCK), glucagon-like peptide (GLP)-1, insulin, and ghrelin among others (Wren et al. 2007). These and various other hormones contributing to the regulation of hunger and satiety are presented in Table 86.2. While CCK, YY, and GLP-1 have been shown to enhance satiety and reduce the amount of caloric intake, ghrelin has been identified as an endogenous orexigenic factor (Wren et al. 2007). The hypothalamus also receives mechanical signals indicating distension of the stomach, and signals transmitting smell, sight, and social context of food. In addition, velocity of gastric emptying, pyloric pressure, and antro-pyloro-duodenal motility have been shown to modify appetite and satiety (Chaudhri et al. 2006). The secretion of the GI hormones from enteroendocrine cells is primarily controlled by the absorption of nutrients from the gut and can be modulated by variations in the velocity of gastric emptying (Meier et al. 2005). Therefore, inhibition of intestinal lipase would be expected to have extensive effects on GI hormone levels and gastric emptying. However, there has
1334
M. Ellrichmann Table 86.2 Overview of hormones and peptides regulating energy intake Hormones/peptides involved in hunger Hormones/peptides involved in satiety Ghrelin CCK Orexin A and B PP NPY PYY AgRP GLP-1 MCH Leptin Galanin OXM Noradrenalin Amylin ß-Endorphin Enterostatin Dynorphin Somatostatin GHRH BNRP Apo A-IV Thyrotropin-releasing hormone Calcitonin-gene-related peptide Serotonin Alpha-MSH CART POMC CRF Urocortins Hormones levels that are attenuated by orlistat administration according to the current data are underlined NPY Neuropeptide Y, AgRP Agouti-related peptide, MCH Melanin-concentrating hormone, GHRH Growth hormone-releasing hormone, CCK Cholecystokinin, PP Pancreatic polypeptide, PYY Peptide YY, OXM Oxyntomodulin, GLP-1 Glucagon-like peptide 1, BNRP Bombesin/bombesin-related peptide, Apo A-IV Apolipoprotein A-IV, Alpha-MSH Alphamelanocyte-stimulating hormone, CART Cocaine- and amphetamine-regulated transcript, POMC Pro-opiomelanocortin, CRF Corticotropin-releasing factor
been some controversy regarding the influence of orlistat treatment on the postprandial levels of CCK, PYY, and GLP-1, with some studies reporting significant reductions in the respective hormone levels and others showing normal or even increased concentrations. In a similar way, the effects of intestinal lipase inhibition on appetite control, food intake, and GI motility have been widely debated within the literature. The reasons underlying these discrepancies are yet unclear. In our own tudy, 25 healthy volunteers were examined with a solid–liquid test meal following oral administration of 120 mg orlistat or placebo in a randomized fashion. Visual analogue scales (VAS) were used to assess feelings of hunger, satiety, fullness, and prospective food consumption over a period of 2 h after meal ingestion. As shown, the feelings of satiety and fullness rose significantly after ingestion of the test meal and declined during the subsequent study period ( p < 0.0001). Both satiety and fullness were significantly reduced by orlistat administration from 15 to 120 min after the test meal (Fig. 86.2). The mean reduction of satiety by orlistat treatment was 15% ( p < 0.0001), whereas fullness was lowered by 12% ( p < 0.0001). Conversely, appetite and prospective food consumption were lowered immediately after meal ingestion and increased subsequently throughout the observation period ( p < 0.0001). Orlistat significantly increased the mean ratings for appetite and prospective food consumption by 24% or 31%, respectively ( p < 0.0001) (Ellrichmann et al. 2008). Previous studies have also examined the effects of orlistat on appetite sensations and the corresponding secretion of GI hormones, but conflicting results have been reported. Feinle et al. could show that lipase inhibition prevented the reduction in scores for prospective food consumption and hunger induced by duodenal fat infusion and increased energy intake at a free buffet meal (Feinle et al. 2003). These findings are
86 Orlistat and the Influence on Appetite Signals
1335
Fig. 86.2 Mean ratings for satiety (a), hunger (b), fullness (c) and prospective food consumption (d), as determined by visual analogue scales (VAS), after ingestion of a solid–liquid test meal following oral administration of 120 mg orlistat or placebo in 25 healthy volunteers. Data are presented as means ±SEM. p-Values were calculated using paired repeated measures ANOVA and denote A: difference between the experiments, B: differences over time and AB: differences due to the interaction of experiment and time. Asterisks indicate significant difference ( p < 0.05) versus placebo at individual time points (one-way ANOVA)
consistent with other observations that the decrease in hunger and increase in fullness induced by gastric distension are diminished by lipase inhibition (Feinle et al. 2001). Moreover, these data are in line with other studies indicating that the suppression of food intake by duodenal infusion of olive oil is attenuated when fat digestion is inhibited, underlining the importance of the products of fat digestion in the regulation of appetite perception and food intake (Matzinger et al. 2000) (Table 86.3). In contrast, Goedecke and colleagues found no effect of orlistat treatment on appetite ratings in healthy male subjects. Orlistat did not alter the ratings of hunger, satiety, and prospective food consumption, when examining changes over time in response to a high-fat meal. In addition, there were no differences in post-test meal intake between orlistat and placebo trials. The reasons underlying the discrepant results of some of these studies are difficult to explain, but certainly methodological issues need to be considered. In particular, different meal compositions and the respective routes of nutrient administration (direct intraduodenal infusions versus oral meal ingestion) might have had an impact on the conflicting results (Goedecke et al. 2003). Nevertheless, the majority of studies support the notion that appetite is acutely increased by orlistat treatment, most likely as a consequence of impaired secretion of anorexigenic GI hormones. The significant increase in appetite and prospective food consumption observed in our as well as previous studies (O’Donovan et al. 2003) may counteract the primary therapeutic indication of orlistat, i.e., the treatment of morbid obesity. Against this, intestinal lipase inhibition has proven to efficiently lower body weight in obese patients over chronic treatment durations (Kelley et al. 2002). Furthermore, in a recent prospective trial over 3 years eating behavior was not affected in a negative way by orlistat
1336
M. Ellrichmann
Table 86.3 Effects of orlistat administration on gastrointestinal (GI) hormone regulation Change in Hormone Study author (year) Test meal hormone release Comment CCK Ellrichmann Oral administration 25 healthy volunteers, inverse ↓ et al. (2008) correlation with hunger scores Feinle et al. (2001) Intraduodenal infusion ↓ Intraduodenal infusion of lipids Borovicka Intraduodenal infusion ↓ Acceleration of gastric emptying, et al. (2000) increase in postprandial gastric acidity Hildebrand Intraduodenal infusion ↓ Intraduodenal infusion of fat, et al. (1998) significant CCK suppression by orlistat GLP-1 Ellrichmann Oral administration ↓ Orally ingested test meal, et al. (2008) significant suppression postprandial GLP-1 levels by orlistat Oral administration ↑ Only 8 patients with type 2 O’Donovan et al. (2004) diabetes Damci et al. (2004) Oral administration ↑ Increased GLP-1 plasma levels, decreased food intake Feinle et al. (2003) Intraduodenal infusion ↓ Significantly attenuated GLP-1 levels by orlistat Pilichiewicz Oral administration ↓ Reduced GLP-1 levels after orlistat et al. (2003) administration in type 2 diabetes Oral administration ↓ Signification attenuation of PYY PYY Ellrichmann et al. (2008) by orlistat Degen et al. (2007) Intraduodenal infusion ↓ CCK1R blockade reduced postprandial PYY concentrations Feinle et al. (2005) Intraduodenal infusion ↓ Fat digestion required for PYY secretion Feltrin et al. (2006) Oral administration ↓ CCK-8 infusions stimulate PYY concentrations PP Feinle et al. (2005) Intraduodenal infusion ↓ Fat digestion required for PP secretion GIP Enc et al. (2009) Oral administration ↓ Attenuation of GIP, no significant changes of GLP-1, PYY by orlistat Pilichiewicz Oral administration ↓ Only 7 patients with type 2 et al. (2003) diabetes, attenuation of GIP Leptin Sahin et al. (2008) Oral administration unchanged No effect of single dose of orlistat on serum leptin levels, 34 patients Dimitrov et al. (2005) Oral administration ↓ Long-term therapy with orlistat reduces leptin leves, mediated by weight loss Ghrelin Ellrichmann Oral administration unchanged Ghrelin unchanged, due to et al. (2008) accelerated gastric emptying Degen et al. (2007) Intraduodenal infusion ↑ Effect mediated indirectly via CCK1R Feinle et al. (2005) Intraduodenal infusion ↑ Elevated ghrelin levels by orlistat, intraduodenal lipid infusion Mohlig et al. (2002) Intraduodenal infusion ↑ Elevated ghrelin levels by orlistat, intraduodenal lipid infusion Overview of current studies evaluating the effect of orlistat on postprandial hormone levels, i.e., cholecystokinin (CCK), glucagon-like peptide (GLP)1, peptide YY (PYY), pancreatic polypeptide (PP), gastric inhibitory polypeptide (GIP), leptin, and ghrelin
86 Orlistat and the Influence on Appetite Signals
1337
treatment, although hunger ratings were still significantly increased at the end of the observation period. Over 300 obese subjects with a mean BMI of 37.5 ± 4.1 kg/m2 were randomized to either 120 mg t.i.d or placebo. The scores for hunger were significantly increased in the orlistat group compared to placebo in male subjects after 33 months of treatment. Similar tendencies were observed in women, but did not achieve statistical significance (Svendsen et al. 2008). The acute and long-term effects of orlistat on satiety are mediated by several satiety signals, i.e., GI orexigenic and anorexigenic hormones.
86.4 Orlistat and CCK Gibbs et al. first demonstrated a dose-dependent effect of exogenous CCK on reducing food intake in rats. This effect was specific to food intake, with CCK having no effect on water intake in waterdeprived rats (Gibbs et al. 1973). This finding was subsequently confirmed in humans, in whom an intravenous infusion of the terminal octapeptide of CCK reduced meal size and duration. It has been proposed that the inhibitory effect of CCK on gastric motility might contribute to its inhibitory actions on feeding. CCK may promote excitation of gastric mechanoreceptors, and thus evoke a neural feedback from the gut to appetite centers of the brain (Moran et al. 1988). Similarly, gastric distension in humans was found to augment the reduction of nutrient intake effected by intravenous CCK-8 (Kissileff et al. 2003). CCK also alters food intake through pathways that are independent of its effect on the stomach. While the induction of satiety at higher doses of CCK may be attenuated by surgical removal of the pyloric sphincter, lower doses continue to be effective in inhibiting food intake. Lesioning of the vagus nerve abolishes the effects of CCK (Moran and Kinzig 2004). The induction of satiety by CCK at physiological concentrations may therefore rely on direct activation of vagal afferent fibers. In addition, CCK-1 receptors (CCK1Rs) were found on afferent fibers of the vagus nerve, and also in the brainstem and dorsomedial nucleus of the hypothalamus. The use of specific CCK-1 and CCK-2 receptor antagonists has implicated the CCK1R in the reduction of food intake (Moran et al. 1992). The Otsuka-Long-Evans-Tokushima Fatty (OLETF) rat, which lacks CCK1R, is both hyperphagic and obese, supporting the role of CCK in regulating energy balance (Schwartz et al. 1999). CCK is released by intestinal I-cells in response to adequate hydrolysis of dietary TG. Hildebrand et al. showed in human studies that endogenous CCK release depends on adequate digestion of dietary fat by pancreatic lipase in the intestinal lumen (Hildebrand et al. 1998). As noted, administration of orlistat covalently inhibits the hydrolysis of dietary TG into MG and FFAs. We have recently established that orlistat significantly impairs CCK release, thereby accelerating gastric emptying and altering ratings of satiety and hunger. In our experiments healthy volunteers were investigated on two occasions, either with orlistat or placebo. Meal ingestion elicited an immediate and pronounced rise in CCK concentrations in the placebo experiments, whereas the increase in CCK levels was largely delayed and significantly blunted after orlistat treatment ( p < 0.0001). Correlation analysis revealed a positive correlation between integrated CCK serum levels and hunger ratings, and an inverse correlation to satiety levels (Fig. 86.3). We also demonstrated a strong correlation between CCK serum levels and gastric emptying (Ellrichmann et al. 2008). Borovicka et al. (2000) showed that intragastric administration of orlistat with a stable emulsion of TG, which reduced lipolysis by 75%, inhibited plasma CCK release. When administered intraduodenally, orlistat attenuated the CCK response to a triglyceride emulsion (Hildebrand et al. 1998). These studies are consistent with our findings that intragastric and intraduodenal administration of orlistat attenuates CCK release, via its inhibitory effects on lipolysis.
1338
M. Ellrichmann
Fig. 86.3 (a) Plasma concentrations of cholecystokinin (CCK) after ingestion of a solid–liquid test meal following oral administration of 120 mg orlistat or placebo in 25 healthy volunteers. Data are presented as means ± SEM. p-Values were calculated using paired repeated measures ANOVA and denote A: difference between the experiments, B: differences over time and AB: differences due to the interaction of experiment and time. Asterisks indicate significant difference (p < 0.05) versus placebo at individual time points (one-way ANOVA). (b, c) Linear regression analysis between the integrated incremental plasma concentration of CCK and mean ratings for hunger (b) and satiety (c) as determined using visual analogue scales (VAS) (n = 50). Dashed lines indicate the respective upper and lower 95% confidence interval. r2 = correlation coefficient squared, p < 0.05 statistically significant
86.5 Orlistat and GLP-1 Glucagon-like peptide (GLP)-1 is cleaved from preproglucagon within the intestine, where it is co-localized in the endocrine L-cells of the distal gut with OXM, PYY, and GLP-2. GLP-1 mediates glucose-dependent insulinotropic effects in a number of species, inhibits gastric acid secretion and gastric emptying, as well as suppresses glucagon release and promotes an increase in pancreatic b-cell mass (Meier and Nauck 2005). Consistent with its role as an incretin, GLP-1 is released into the circulation in response to a meal in proportion of calories ingested (Orskov et al. 1994). In common with other gut peptides, GLP-1 works as a neurotransmitter within the CNS. It is present within the dorsovagal complex, the thalamus, and the pituitary, in key areas of the hypothalamus involved
86 Orlistat and the Influence on Appetite Signals
1339
in appetite regulation GLP-1-immunoreactive neurons were found (Larsen et al. 1997). Food intake data from human studies are concordant with animal studies. GLP-1 dose-dependently decreases appetite and caloric intake in lean and obese subjects and in patients with type 2 diabetes (Gutzwiller et al. 1999). A meta-analysis by Verdich et al. concluded that infusion of GLP-1 reduces both appetite and food intake with the magnitude of this reduction being similar in lean and obese men (Verdich et al. 2001). The impact of orlistat administration on postprandial GLP-1 secretion has been debated in the literature. Damci et al. found that lipase inhibition stimulates the meal-mediated secretion of GLP-1. In this study, only a single blood sample was obtained 60 min after a meal in diabetic patients (Damci et al. 2004). O’Donovan et al. revealed that orlistat administration may exacerbate postprandial glycemia as a result of accelerated gastric emptying (O’Donovan et al. 2004). These observations need to be critically considered. In a study conducted by Feinle et al., the effect of lipase inhibition by orlistat on gastric emptying, antro-pyloro-duodenal motility, satiety, and GLP-1 levels was investigated. Healthy volunteers were given a TG emulsion via 120 min duodenal infusion either with orlistat or placebo. Immediately after the duodenal infusion, food intake at an ad libitum buffet was quantified. Lipase inhibition with orlistat resulted in significantly reduced postprandial GLP-1 levels. In addition, administration of orlistat was associated with markedly higher scores of prospective food consumption and hunger and greater food intake at the subsequent buffet. They conclude that GLP-1 secretion depends on adequate hydrolysis of TG into MG and FFA in the upper part of the small intestine (Feinle et al. 2003). In patients with type 2 diabetes, the ingestion of orlistat together with a liquid meal containing oil and glucose also accelerates gastric emptying and attenuates the postprandial rise in GLP-1 (Pilichiewicz et al. 2003). Furthermore, Beysen et al. demonstrated differential stimulation of GLP-1 release by different types of oral fat (monounsaturated, olive oil; polyunsaturated, softflower oil; saturated, palm stearin). The greatest increase in GLP-1 was observed after fat containing monounsaturated fatty acids (Beysen et al. 2002). These findings are in line with our own data showing that in healthy volunteers peak GLP-1 levels following orlistat treatment were 20% lower compared to placebo. Likewise, the integrated concentrations of GLP-1 were significantly lower following orlistat administration. These lowered GLP-1 levels were associated with reduced satiety ratings (Fig. 86.4).
86.6 Orlistat and PYY Peptide YY (PYY) is released into the circulation in response to a meal from L-cells of the GI tract in proportion to the calories ingested and in relation to the meal composition (Wren et al. 2007). Higher plasma levels are seen following isocaloric meals of fat compared to meals consisting of protein or carbohydrate. The release of PYY in response to fat in the proximal intestine is atropinesensitive, raising the possibility that a neural reflex involving the vagus nerve may mediate PYY release (Lin and Taylor 2004). Other stimulants of PYY release include intraluminal bile acids, gastric acid, and CCK (Onaga et al. 2002). PYY3-36 has a central role in the control of appetite in humans as supported by a number of observations. In disease states characterized by weight loss PYY3-36 levels are elevated (Le Roux et al. 2005). Conversely, in obese humans fasting plasma levels of PYY3-36 are reduced in overweight patients having a relative deficiency of postprandial PYY3-36 release associated with reduced satiety (Batterham et al. 2003b). Degen et al. used orlistat to determine whether inhibition of fat hydrolysis affects the release of PYY. In their study, PYY secretion in response to intraduodenal fat in the small intestine was completely abolished by orlistat administration (Degen et al. 2007). The data are in line with previous
1340
M. Ellrichmann
Fig. 86.4 (a) Plasma concentrations of glucagonlike peptide (GLP)1 after ingestion of a solid–liquid test meal following oral administration of 120 mg orlistat or placebo in 25 healthy volunteers. Data are presented as means ± SEM. p-Values were calculated using paired repeated measures ANOVA and denote A: difference between the experiments, B: differences over time and AB: differences due to the interaction of experiment and time. Asterisks indicate significant difference ( p < 0.05) versus placebo at individual time points (one-way ANOVA). (b, c) Linear regression analysis between the integrated incremental plasma concentration of GLP-1 and mean ratings for hunger (b) and satiety (c) as determined using visual analogue scales (VAS) (n = 50). Dashed lines indicate the respective upper and lower 95% confidence interval. r2 = correlation coefficient squared, p < 0.05 statistically significant
studies: suppression of fat hydrolysis by orlistat inhibits the release of PYY (Feinle et al. 2005). The crucial importance of fat hydrolysis on digestive functions is illustrated by the effects of these products on exocrine pancreatic responses in animals and humans. Only duodenal infusion of long-chain FFAs can stimulate maximal pancreatic enzyme secretion, whereas undigested long-chain TG are ineffective (Lin and Taylor 2004). Inhibition of lipolysis by orlistat reduces the amount of FFAs in the small intestine with a subsequent reduction in PYY release. The reduction in PYY release attenuates inhibitory appetite circuits resulting in further food intake. This hypothesis was confirmed by our work. Meal ingestion elicited a significant rise in PYY levels in both experiments, but the increase in PYY levels was significantly attenuated by orlistat treatment. The reduction in PYY responses resulted in increased ratings for hunger compared to control, though correlation analysis failed to reach statistical significance (Fig. 86.5) (Ellrichmann et al. 2008). This indicates that PYY may play a minor role in enteroendocrine appetite regulation.
86 Orlistat and the Influence on Appetite Signals
1341
Fig. 86.5 (a) Plasma concentrations of peptide YY (PYY) after ingestion of a solid–liquid test meal following oral administration of 120 mg orlistat or placebo in 25 healthy volunteers. Data are presented as means ± SEM. p-Values were calculated using paired repeated measures ANOVA and denote A: difference between the experiments, B: differences over time and AB: differences due to the interaction of experiment and time. Asterisks indicate significant difference ( p < 0.05) versus placebo at individual time points (one-way ANOVA). (b, c) Linear regression analysis between the integrated incremental plasma concentration of PYY and mean ratings for hunger (b) and satiety (c) as determined using visual analogue scales (VAS) (n = 50). Dashed lines indicate the respective upper and lower 95% confidence interval. r2 = correlation coefficient squared, p < 0.05 statistically significant
PYY secretion is initiated either directly via luminal contact of nutrients to the endocrine cells or indirectly through neurohumoral signals such as CCK. Since a CCK1R antagonist (Dexlox) completely abolished PYY response to intraluminal FFAs, it is conceivable that the products of fat digestion stimulate CCK release, which in turn regulates PYY secretion via CCK1Rs (Degen et al. 2007).
86.7 Orlistat and PP Pancreatic polypeptide (PP) is produced largely not only in the PP-cells of the endocrine pancreas, but also in the exocrine pancreas, colon, and rectum. Like PYY, PP is released into the circulation in response to a meal in proportion to the caloric intake. Low levels of PP have been found in obese
1342
M. Ellrichmann
patients (Glaser et al. 1988), and high levels in patients with anorexia nervosa (Fujimoto et al. 1997). Both basal and postprandial PP release is subject to control by the vagus nerve as proved by truncal vagotomoy. Both truncal vagotomy and atropine have been shown to reduce meal-induced PP release in humans (Meguro et al. 1995). PP secretion is also controlled by other gut hormones, including ghrelin, motilin, and secretin, all of which stimulate PP release, and somatostatin, which potently inhibits (Gomez et al. 1997). A recent study evaluating the effects of duodenal infusion of a long-chain TG infusion with or without 120 mg THL revealed that lipase inhibition completely abolishes the postprandial PP release. In response to a subsequent free buffet meal, PP levels were stimulated in both groups with and without application of THL. In addition, the THL group consumed significantly more food at the buffet than the control group. This may suggest that, in contrast to PYY, gastric distension, enhanced by the greater amount of food eaten, may be a more potent stimulus for PP (Feinle-Bisset et al. 2005). Moderate gastric distension with a 600 mL balloon has been shown to cause a substantial increase in PP secretion in healthy volunteers (Koop et al. 1990). Alternatively, it is possible that other macronutrients, carbohydrate and protein, are more potent stimuli than fat. Batterham et al. demonstrated that PP infusion reduces both appetite and food intake in healthy subjects. In addition, inhibition of food intake was sustained, such that energy intake, assessed by food diaries, was significantly reduced both in the evening of the study and the following morning. Plasma levels of PP remain elevated for up to 6 h postprandially suggesting that PP may regulate meal-to-meal intervals (Batterham et al. 2003). Conversely, reduced PP levels induced by orlistat treatment may shorten meal-to-meal intervals and increase food uptake per meal by attenuating satiety signals. Further studies are required to establish these potential mechanisms.
86.8 Orlistat and GIP The first incretin hormone to be identified was isolated from porcine small intestine and was initially named gastric inhibitory polypeptide (GIP), based on its ability to inhibit gastric acid secretion in dogs. Since the inhibitory effect on gastric acid secretion was only observed at pharmacological doses, whereas its incretin action occurred at physiological levels, GIP was renamed as glucosedependent insulinotropic polypeptide (Meier et al. 2002). In accordance with its role as an incretin hormone, GIP is released from K-cells of the small intestine in response to glucose and fat ingestion, and thus potentiates glucose-stimulated insulin secretion. While the effects of GIP on insulin secretion have been studied extensively, only limited information is available about its effects on lipid homeostasis. The potent stimulation of GIP secretion after high-fat meals, as well as the observation of increased GIP plasma levels in obesity, suggested a role of GIP as an anabolic regulator of fat metabolism. The anabolic effects of GIP in fat include stimulation of fatty acid synthesis, enhancement of insulin-stimulated incorporation of fatty acids into TG, upregulation of lipoprotein lipase synthesis, and reduction of glucagon-stimulated lipolysis (Meier et al. 2002). Ob/ob mice with a GIP receptor (GIPR) knock-out are resistant to diet-induced obesity and exhibit reduced adipocyte mass. GIPR-/- mice expend more energy and use fat as preferred energy substrate, thereby preventing the accumulation of fat in adipocytes. Food intake is comparable in GIPR-/- and wild-type mice (Miyawaki et al. 2002). In humans, no direct links between GIP and obesity as well as appetite regulation have been demonstrated. Since GIP is released into the circulation in response to fat ingestion, the role of lipase inhibition on postprandial GIP response was investigated by several groups. According to Enc et al., orlistat administration induced profound attenuation of postprandial GIP response. With ingestion of the control mixed meal, plasma GIP levels increased significantly and
86 Orlistat and the Influence on Appetite Signals
1343
remained elevated throughout the experimental period. Orlistat significantly lowered GIP response as expressed by integrated areas under the curve (Enc et al. 2009). Pilichiewicz et al. evaluated the effect of orlistat on gastric emptying and plasma GIP response in patients with type 2 diabetes. This study also showed reduced postprandial GIP secretion and accelerated gastric emptying by orlistat treatment (Pilichiewicz et al. 2003). It is therefore likely that orlistat by its ability to inhibit intestinal lipid absorption attenuates postprandial GIP release. Despite the well-characterized effects of GLP-1 on appetite and food intake, the role of GIP in feeding regulation remains unclear. Peripheral infusion of GIP at a rate shown to elicit supraphysiological plasma concentrations increases feelings of hunger in healthy lean volunteers (p < 0.05). Comparing hunger responses between healthy lean subjects and obese patients with type 2 diabetes during GIP infusion, a trend for higher hunger scores was observed in lean subjects, though this trend failed to reach statistical significance (p > 0.05). Despite these findings, ad libitum food intake during a buffet lunch at the end of the infusions was similar in the GIP and placebo group (Daousi et al. 2008). Whether these potential effects of GIP on appetite regulation are relevant under physiological conditions remains questionable.
86.9 Orlistat and Leptin Adipose tissue, once considered to be a relatively passive site of lipid storage, is known to carry a number of complex metabolic and endocrine functions. One important hormonal factor is the protein leptin, which was isolated and synthesized following positional cloning of the gene responsible for obesity in the ob/ob mouse strain. Leptin is synthesized and released from white adipose tissue and circulates to the brain where it binds to its receptor. It acts to decrease weight and adipose tissue mass through reduction in appetite and food intake. Detected synthesis of leptin within the gastric mucosa raised the possibility that it may also play a role in meal termination (Peters et al. 2005). Markedly elevated leptin levels have been shown in obese humans compared to nonobese, conversely leptin levels markedly decline in underweight patients compared to normal-weight subjects (Haluzik et al. 1999). Only two studies have investigated the effect of lipase inhibition on serum leptin levels. Sahin et al. studied the acute effects of orlistat on postprandial leptin levels in nondiabetic obese patients. Although serum leptin levels showed a more horizontal and delayed increase after a mixed meal in patients in the orlistat group than they had in the placebo group, there were no statistical differences between these two groups (Sahin et al. 2008). A long-term, 3-month prospective study was conducted in 40 patients with clinical features of metabolic syndrome either with or without coexisting type 2 diabetes. In addition to a hypocaloric diet, these patients were given orlistat t.i.d. They observed beneficially enhanced weight loss by orlistat accompanied by decreased serum leptin concentrations (Dimitrov et al. 2005). These changes cannot be regarded as direct effects by lipase inhibition but as indirect effects due to the significant weight loss.
86.10 Orlistat and Ghrelin Ghrelin acts as the endogenous ligand for the growth hormone secretagogue (GHS) receptor secreted primarily by the stomach to the circulation although ghrelin mRNA is present throughout the whole GI tract. Ghrelin is a peripherally active appetite-stimulating gut hormone. Ghrelin levels rise in the fasting state and fall upon eating, which has led to the suggestion that ghrelin may be involved in meal initiation. The postprandial suppression of ghrelin does not require the vagus nerve, but vagal
1344
M. Ellrichmann
afferents are necessary for the rise of ghrelin in the fasted state (Wren et al. 2007; Cummings et al. 2001). The exact mechanisms responsible for postmeal suppression of ghrelin are not known. Both carbohydrate and fat, when ingested orally, suppress ghrelin secretion, whereas protein may stimulate ghrelin release or may have no effect (Greenman et al. 2004). Postprandial decline of ghrelin levels significantly depends on the caloric load of the meal: a larger caloric load produces a lower nadir in ghrelin levels after a meal than a smaller one of similar volume (Callahan et al. 2004). Recent evidence in humans suggests that both PYY and oxyntomodulin (OXM) suppress concentrations of ghrelin when given peripherally, PP having no effect (Batterham et al. 2003). Feinle et al. documented that duodenal infusion of a long-chain TG emulsion potently suppresses ghrelin secretion in healthy young man. This emulsion given with THL completely abolished ghrelin suppression, indicating that ghrelin secretion is sensitive to digested fat, an effect apparently not mediated by an increase in blood FFAs (Feinle et al. 2005). These findings are consistent with the data obtained by Degen et al. Intraduodenal infusion of fat induced a significant inhibition in ghrelin levels ( p < 0.01) and a significant increase in PYY concentrations (p < 0.02). Inhibition of lipolysis by orlistat completely abolished both effects. In addition, administration of a specific CCK1R antagonist together with intraduodenal infusion of fat abolished the effect of long-chain fatty acids on ghrelin and PYY secretion. These authors conclude that fat hydrolysis in the proximal small intestine plays a crucial role for digestive processes. Free fatty acids stimulate the release of CCK, which in turn acts on CCK1Rs thereby initiating a series of digestive functions, including modulation of CCK-1dependent ghrelin secretion (Degen et al. 2007). The data are contrasted by our own results showing no effect of orlistat on postprandial ghrelin concentrations (Ellrichmann et al. 2008). The size, composition, and route of administration of the respective test meals might explain these inconsistencies.
86.11 Applications to Other Areas of Health and Disease CCK does not only regulate appetite sensation and gastric emptying but also regulates gallbladder contraction and emptying of bile during the postprandial state by binding to G-protein-coupled CCKA-receptors on the smooth muscle of the gallbladder thereby mediating gallbladder emptying. As already mentioned, CCK is released by intestinal I-cells in response to an adequate hydrolysis of dietary TG. In addition to the previous studies, we assessed whether oral administration of orlistat inhibits CCK release in response to a test meal. In the same setting, healthy volunteers were given a test meal with 120 mg orlistat or placebo visualizing gallbladder volume by abdominal ultrasound. Orlistat administration resulted in significantly reduced gallbladder emptying through meal-mediated CCK release (Ellrichmann et al. 2008) (Fig. 86.6). Since reduced gallbladder contractility leading to bile stasis is one of the major risk factors for the development of cholesterol gallstones, in a 1-year followup Mathus-Vliegen et al. investigated the risk of gallstone occurrence by orlistat administration. After 1 month of diet, obese patients were treated with either placebo or 60 or 120 mg orlistat t.i.d. One month and 12 months after randomization, gallbladder kinetics and CCK levels were assessed. In the 60-mg group, no significant changes of the gallbladder motility were observed, whereas 120 mg orlistat resulted in a significant decrease in gallbladder emptying. The residual gallbladder volumes and percentage of gallbladder emptying are comparable to our findings. After 1 year of treatment, disturbed gallbladder emptying persisted, 16.7% of patients in the placebo group developed gallstones, whereas only 7.1% of gallstone formation was seen in the 60 mg orlistat group. In contrast, no gallstone formation was observed in the 120 mg orlistat group (Mathus-Vliegen et al. 2004). These results were rather unexpected and might indicate that altered gallbladder kinetics is outweighed by other factors, such as a reduction of cholesterol supernaturation and nucleation factors (Fig. 86.7).
Fig. 86.6 Gallbladder volume after ingestion of a solid–liquid test meal following oral administration of 120 mg orlistat or placebo in 25 healthy volunteers. Data are presented as means ± SEM. p-Values were calculated using paired repeated measures ANOVA and denote A: difference between the experiments, B: differences over time and AB: differences due to the interaction of experiment and time. Asterisks indicate significant difference ( p < 0.05) versus placebo at individual time points (one-way ANOVA). (b, c) Linear regression analysis between the integrated incremental plasma concentration of PYY and mean ratings for hunger (b) and satiety (c) as determined using visual analogue scales (VAS) (n = 50). Dashed lines indicate the respective upper and lower 95% confidence interval. r2 = correlation coefficient squared, p < 0.05 statistically significant
Fig. 86.7 Effects of orlistat administration on postprandial gastrointestinal (GI) hormone levels. Orlistat reduces leptin levels secreted by the adipose tissue. In addition, the secretion of cholecystokinin (CCK), glucagon-like peptide (GLP)1, peptide YY (PYY), pancreatic polypeptide (PP), and gastric inhibitory polypeptide (GIP) from pancreas and the small intestine is attenuated. Ghrelin levels, primarily secreted by the stomach, are elevated. The changes in these GI hormone concentration lead to increased appetite sensation in the central nervous system (CNS) and thus to increased food intake
1346
M. Ellrichmann
86.12 Conclusion In conclusion, intestinal lipase inhibition by orlistat accelerates gastric emptying and markedly reduces the postprandial secretion of the anorexic hormones CCK, GLP-1, PYY, PP, and GIP, whereas ghrelin levels are elevated. Taken together, these alterations in GI hormone concentrations can increase appetite sensations leading to greater food consumption. These findings underline the importance of GI hormones in the central nervous control of energy homoeostasis. Increased appetite sensations, a potentially increased risk of developing gallbladder stones, and alterations in postprandial glucose regulation should be considered as potential side effects when applying lipase inhibitors for the treatment of morbid obesity.
Summary Points • Orlistat inhibits intestinal lipases and thus reduces hydrolysis of dietary TG in MG and FFAs. • Orlistat leads to a marked acceleration of gastric motility and to an inhibition of gallbladder emptying. • Lipase inhibition significantly attenuates appetite ratings and increases food intake. • Increased appetite induced by orlistat is mediated by changes in GI hormone levels, especially CCK, GLP-1, PYY and ghrelin. • Changes in enteroendocrine hormone levels regulating appetite counteract the weight-lowering effect of the drug and the treatment of morbid obesity.
Definitions and Explanations of Key Terms Cholecystokinin (CCK): CCK is derived from a 115-amino acid precursor, pro-CCK. Selective cleavage results in several circulating isoforms; the major forms in man are CCK-58, -33, -22, and -8. Biological activity resides in the amidated C-terminus of the peptide; all active species of CCK share a C-terminal heptapeptide sequence including O-sulfated tyrosine. CCK physiologically mediates gallbladder contraction and inhibits gastric emptying and gastric acid secretion. In addition, CCK induces satiating effects in the CNS. Glucagon-like peptide (GLP)1: Post-translational processing of preproglucagon in enteroendocrine L-cells yields the peptides glicentin, OXM, GLP-2, and GLP-1. Multiple forms of GLP-1 are secreted in vivo, including GLP-1(1–37) and GLP-1(1–36)NH2, which are thought to be inactive, and GLP-1(7–37) and GLP-1(7–36), which are biologically active. In humans, the majority of GLP-1 in the circulation is GLP-1(7–36)NH2. GLP-1 possesses a variety of physiological properties comprising increase in insulin secretion, decrease of glucagon secretion, inhibition of gastric emptying, and attenuation of food intake. Peptide YY (PYY): PYY is a 36-amino acid linear peptide and is a member of the pancreatic peptide family. The main of PYY is stored and secreted to the circulation as PYY(3–36), which is the N-terminally truncated form of the peptide. PYY is secreted from the endocrine L-cells of the small and large bowel, with high concentrations at the terminal ileum and colon and maximum concentration in the rectum. PYY exerts its action through NPY receptors thereby inhibiting gastric motility and pancreatic secretion. Furthermore, PYY has been shown to reduce to appetite and prospective food consumption via NPY receptor within the CNS.
86 Orlistat and the Influence on Appetite Signals
1347
Pancreatic polypeptide (PP): PP is a 36-amino acid peptide and as well a member of the pancreatic peptide family. It is produced in the endocrine type F cells located in peripheries of the pancreatic islets in response to food ingestion, hypoglycemia, and exercise. The physiological mechanisms of PP still remain unclear. Several studies suggest that vagal nerve activity can be estimated by PP serum levels. Gastric inhibitory peptide (GIP): GIP is the first incretin hormone to be identified isolated from porcine intestine. Based on its ability to inhibit gastric acid secretion it was named “gastric inhibitory polypeptide.” Subsequent studies demonstrated a glucose-dependent stimulation of insulin secretion by GIP. Accordingly, the term “glucose-dependent insulinotropic polypeptide” was proposed to more accurately define the acronym GIP. GIP is synthesized in K-cells of the upper small intestine. The mature bioactive 42-amino acid form of GIP is cleaved from its 153amino acid proGIP precursor via post-translational processing. GIP is known to stimulate insulin secretion from the endocrine pancreas. It is also thought to have significant effects on fatty acid metabolism by stimulation of lipoprotein lipase in adipocytes. Leptin: Leptin is a 16 kDa protein that plays a key role in regulating energy intake and expenditure. The ob gene located on chromosome 7 in humans is expressed in adipose tissue and encodes for leptin. In addition for being a biomarker for body fat composition, serum leptin levels reflect individual energy balance. By binding to NPY neurons in the arcuate nucleus leptin induces satiety in the CNS. Ghrelin: Ghrelin has a 28-amino acid structure, with an octanoylated serine residue at position 3. This unique side chain appears to be essential for the effects on appetite regulation and GH release. The X/A-like cells in the fundus of the stomach are the major source of circulating ghrelin. To date, ghrelin is the only endogenous peripheral hormone that has been show to induce hunger and increase food intake.
Acknowledgments Juris J. Meier and Wolfgang E. Schmidt are acknowledged for helpful comments and critical discussion of the manuscript.
References Batterham RL, Cowley MA, Small CJ, Herzog H, Cohen MA, Dakin CL, Wren AM, Brynes AE, Low MJ, Ghatei MA, Cone RD, Bloom SR. Gut hormone PYY(3-36) physiologically inhibits food intake. Nature 2002;418:650–4. Batterham RL, Cohen MA, Ellis SM, Le Roux CW, Withers DJ, Frost GS, Ghatei MA, Bloom SR. To eat or not to eat - how the gut talks to the brain. N Engl J Med. 2003;349:926–8. Batterham RL, Le Roux CW, Cohen MA, Park AJ, Ellis SM, Patterson M, Frost GS, Ghatei MA, Bloom SR. Pancreatic polypeptide reduces appetite and food intake in humans. J Clin Endocrinol Metab. 2003;88:3989–92. Beysen C, Karpe F, Fielding BA, Clark A, Levy JC, Frayn KN. Interaction between specific fatty acids, GLP-1 and insulin secretion in humans. Diabetologia. 2002;45:1533–41. Borovicka J, Schwizer W, Guttmann G, Hartmann D, Kosinski M, Wastiel C, Bischoff-Delaloye A, Fried M. Role of lipase in the regulation of postprandial gastric acid secretion and emptying of fat in humans: a study with orlistat, a highly specific lipase inhibitor. Gut. 2000;46:774–81. Callahan HS, Cummings DE, Pepe MS, Breen PA, Matthys CC, Weigle DS. Postprandial suppression of plasma ghrelin level is proportional to ingested caloric load but does not predict intermeal interval in humans. J Clin Endocrinol Metab. 2004;89:1319–24. Chaudhri O, Small C, Bloom S. Gastrointestinal hormones regulating appetite. Philos Trans R Soc Lond B Biol Sci. 2006;361:1187–209. Creutzfeld W, Ebert R, Willms B, Frerichs H, Brown JC. Gastric inhibitory polypeptide (GIP) and insulin in obesity: increased response to stimulation and defective feedback control of serum levels. Diabetologia. 1978;14:15–24.
1348
M. Ellrichmann
Cummings DE, Purnell JQ, Frayo RS, Schmidova K, Wisse KE, Weigle DS. A preprandial rise in plasma ghrelin levels suggests a role in meal initiation in humans. Diabetes. 2001;50:1714–19. Damci T, Yalin S, Balci H, Osar Z, Korugan U, Ozyazar M, Ilkova H. Orlistat augments postprandial increases in glucagon-like peptide 1 in obese type 2 diabetic patients. Diabetes Care. 2004;27:1077–80. Daousi C, Wilding JP, Aditya S, Durham BH, Cleator J, Pinkney JH, Ranganath LR. Effects of peripheral administration of synthetic human glucose-dependent insulinotropic peptide (GIP) on energy expenditure and subjective appetite sensations in healthy normal weight subjects and obese patients with type 2 diabetes. Clin Endocrinol. 2009;71:195–201. Degen L, Drewe J, Piccoli F, Gräni K, Oesch S, Bunea R, D’Amato M, Beglinger C. Effect of CCK-1 receptor blockade on ghrelin and PYY secretion in men. Am J Physiol Regul Integr Comp Physiol. 2007;292:R1391–9. Dimitrov D, Bohchelian H, Koeva L. Effect of orlistat on plasma leptin levels and risk factors for the metabolic syndrome. Metab Syndr Relat Disord. 2005;3:122–9. Ellrichmann M, Kapelle M, Ritter PR, Holst JJ, Herzig KH, Schmidt WE, Schmitz F, Meier JJ. Orlistat inhibition of intestinal lipase acutely increases appetite and attenuates postprandial glucagon-like peptide-1-(7-36)-amide-1, cholecystokinin, and peptide YY concentrations. J Clin Endocrinol Metab. 2008;93:3995–8. Enc FY, Ones T, Akin HL, Dede F, Turoglu HAT, Ulfer G, Bekiroglu N, Haklar G, Rehfeld JF, Holst JJ, Ulusoy NB, Imeryüz N. Orlistat accelerates gastric emptying and attenuates GIP release in healthy subjects. Am J Physiol Gastrointestinal Liver Physiol. 2009;296:G482–9. Feinle C, Rades T, Otto B, Fried M. Fat digestion modulates gastrointestinal sensations induced by gastric distention and duodenal lipid in humans. Gastroenterology 2001;120:1100–7. Feinle C, O’Donovan D, Doran S, Andrews JM, Wishart J, Chapman I, Horowitz M. Effects of fat digestion on appetite, APD motility, and gut hormones in response to duodenal fat infusion in humans. Am J Physiol Gastrointest Liver Physiol. 2003;284:G798–807. Feinle-Bisset C, Patterson M, Ghatei MA, Bloom SR, Horowitz M. Fat digestion is required for suppression of ghrelin and stimulation of peptide YY and pancreatic polypeptide secretion by intraduodenal lipid. Am J Physiol Endocrinol Metab. 2005;289:E948–53. Figlewicz DP, Nadzan AM, Sipolis AJ, Green PK, Liddle RA, Porte D Jr, Woods SC. Intraventricular CCK-8 reduces single meal size in the baboon by interaction with type-A CCK receptors. Am J Physiol. 1992;263:R863–7. Flint A, Raben A, astrup A, Holst JJ. Glucagon-like peptide 1 promotes satiety and suppresses energy intake in humans. J Clin Invest. 1998;101:515–20. Fujimoto S, Inui A, Kiyota N, Seki W, Koide K, Takamiya S, Uemoto M, Nakajima Y, Baba S, Kasgua M. Increased cholecystokinin and pancreatic polypeptide responses to a fat-rich meal in patients with restrictive but not bulimic anorexia nervosa. Biol Psychiatry. 1997;41:1068–70. Gibbs J, Young RC, Smith GP. Cholecystokinin decreases food intake in rats. J Comp Physiol Psychol. 1973;84:488–95. Glaser B, Zoghlin G, Pienta K, Vinik AI. Pancreatic polypeptide response to secretin in obesity: effects of glucose intolerance. Horm Metab Res. 1988;20:288–92. Goedecke JH, Barsdorf M, Beglinger C, Levitt NS, Lambert EV. Effects of a lipase inhibitor (Orlistat) on cholecystokinin and appetite in response to a high-fat meal. Int J Obes Relat Metab Disord. 2003;27:1479–85. Gomez G, Udupi V, Greeley GH Jr. Comparison of somatostatin and pancreastatin on secretion of gastrin, pancreatic polypeptide, and peptide YY. Proc Soc Exp Biol Med. 1997;215:165–7. Greenman Y, Golani N, Gilad S, Yaron M, Limor R, Stern N. Ghrelin secretion is modulated in a nutrient- and genderspecific manner. Clin Endocrinol. 2004;60:382–8. Gutzwiller JP, Drewe J, Göke B, Schmidt H, Rohrer B, Lareida J, Beglinger C. Glucagon-like peptide-1 promotes satiety and reduces food intake in patients with diabetes mellitus type 2. Am J Physiol. 1999;276:R1541–4. Haluzik M, Papezova M, Nedvidkova J, Kabrt J. Serum leptin levels in patients with anorexia nervosa before and after partial refeeding, relationships to serum lipids and biochemical nutritional parameters. Physiol Res. 1999;48:197–202. Hildebrand P, Petrig C, Burckhardt B, Ketterer S, Lengsfeld H, Fleury A, Hadvary P, Beglinger C. Hydrolysis of dietary fat by pancreatic lipase stimulates cholecystokinin release. Gastroenterology. 1998;114:123–9. Kelley DE, Kuller LH, McKolanis TM, Harper P, Mancino J, Kalhan S. Effects of moderate weight loss and orlistat on insulin resistance, regional adiposity, and fatty acids in type 2 diabetes. Diabetes Care. 2004;27:33–40. Kissileff HR, Carretta JC, Geliebter A, Pi-Sunyer FX. Cholecystokinin and stomach distension combine to reduce food intake in humans. Am J Physiol Regul Integr Comp Physiol. 2003;285:R992–8. Koop I, Ruppert-Seipp G, Koop H, Schafmayer A, Arnold R. Cholecystokinin release by gastric distension--an atropine-sensitive mechanism. Digestion. 1990;46:220–7. Larsen PJ, Tang-Christensen M, Holst JJ, Orskov C. Distribution of glucagon-like peptide-1 and other preproglucagonderived peptides in the rat hypothalamus and brainstem. Neuroscience. 1997;77:257–70. Le Roux CW, Ghatei MA, Gibbs JS, Bloom SR. The putative satiety hormone PYY is raised in cardiac cachexia associated with primary pulmonary hypertension. Heart 2005;91:241–2.
86 Orlistat and the Influence on Appetite Signals
1349
Lin HC, Taylor IL. Release of peptide YY by fat in the proximal but not distal gut depends on an atropine-sensitive cholinergic pathway. Regul Pept. 2004;117:73–6. Lucas KH, Kaplan-Machlis B. Orlistat--a novel weight loss therapy. Ann Pharmacother. 2001;35:314–28. Mathus-Vliegen EM, Van Ierland-van Leeuwen ML, Terpstra A. Lipase inhibition by orlistat: effects on gall-bladder kinetics and cholecystokinin release in obesity. Aliment Pharamcol Ther. 2004;19:601–11. Matzinger D, Degen L, Drewe J, Meuli J, Duebendorfer R, Ruckstuhl N, D`Amato M, Rovati L, Beglinger C. The role of long chain fatty acids in regulating food intake and cholecystokinin release in humans. Gut. 2000; 46: 688-93. Meguro T, Shimosegawa T, Kikuchi Y, Koizuma M, Toyota T. Effects of cisapride on gallbladder emptying and pancreatic polypeptide and cholecystokinin. J Gastroenterol. 1995;30:237–43. Meier JJ, Nauck MA. Glucagon-like peptide 1(GLP-1) in biology and pathology. Diabetes Metab Res Rev. 2005;21:91–117. Meier JJ, Nauck MA, Schmidt WE, Gallwitz B. Gastric inhibitory polypeptide: the neglected incretin revisite. Regul Pept. 2002;107:1–13. Meier JJ, Kemmeries G, Holst JJ, Nauck MA. Erythromycin antagonizes the deceleration of gastric emptying by glucagon-like peptide 1 and unmasks its insulinotropic effect in healthy subjects. Diabetes 2005;54:2212–8. Miyawaki K, Yamada Y, Ban N, Ihara Y, Tsukiyama K, Zhou H, Fujimoto S, Oku A, Tsuda K, Toyokuni S, Hiai H, Mizunoya W, Fushiki T, Holst JJ, Makino M, Tashita A, Kobara Y, Tsubamoto Y, Jinnouchi T, Jomori T, Seino Y. Inhibition of gastric inhibitory polypeptide signaling prevents obesity. Nat Med. 2002;8:738–42. Moran TH, Kinzig KP. Gastrointestinal satiety signals II. Cholecystokinin. Am J Physiol Gastrointestinal Liver Physiol. 2004;286:G183–8. Moran TH, McHugh PR. Gastric and nongastric mechanisms for satiety action of cholecystokinin. Am J Physiol. 1988;254:R628–32. Moran TH, Ameglio PJ, Schwartz GJ, McHugh PR. Blockade of type A, not type B, CCK receptors attenuates satiety actions of exogenous and endogenous CCK. Am J Physiol. 1992;262:R46–50. Murakami N, Hayashida T, Kuroiwa T, Nakahara K, Ida T, Mondal MS, Nakazato M, Kojima M, Kangawa K. Role for central ghrelin in food intake and secretion profile of stomach ghrelin in rats. J Endocrinol. 2002;174: 283–8. O’Donovan D, Feinle-Bisset C, Wishart J, Horowitz M. Lipase inhibition attenuates the acute inhibitory effects of oral fat on food intake in healthy subjects. Br J Nutr. 2003;90:849–52. O’Donovan D, Horowitz M, Russo A, Feinle-Bisset C, Murolo N, Gentilcore D, Wishart JM, Morris HA, Jones KL. Effects of lipase inhibition on gastric emptying of, and on the glycaemic, insulin and cardiovascular responses to, a high-fat/carbohydrate meal in type 2 diabetes. Diabetologia. 2004;47:2208–14. Onaga T, Zabielski R, Kato S. Multiple regulation of peptide YY secretion in the digestive tract. Peptides 2002;23:279–90. Orskov C, Rabenhoj L, Wettergren A, Kofod H, Holst JJ. Tissue and plasma concentrations of amidated and glycineextended glucagon-like peptide I in humans. Diabetes 1994;43:535–9. Padwal R, Li SK, Lau DC. Long-term pharmacotherapy for overweight and obesity: a systematic review and metaanalysis of randomized controlled trials. Int J Obes Relat Metab Disord. 2003;27:1437–46. Peters JH, McKay BM, Simasko SM, Ritter RC. Leptin-induced satiation mediated by abdominal vagal afferents. Am J Physiol Regul Integr Comp Physiol. 2005;288:R879–84. Pilichiewicz A, O’Donovan D, Feinle C, Lei Y, Wishart JM, Bryant L, Meyer JH, Horowitz M, Jones KL. Effect of lipase inhibition on gastric emptying of, and the glycemic and incretin responses to, an oil/aqueous drink in type 2 diabetes mellitus. J Clin Endocrinol Metab. 2003;88:3829–34. Pi-Sunyer X, Kissileff HR, Thornton J, Smith GP. C-terminal octapeptide of cholecystokinin decreases food intake in obese men. Physiol Behav. 1982;29:627–30. Sahin M, Tanaci N, Yucel M, Kutlu M, Tutuncu NB, Pamuk B, Guvener ND. Acute effects of orlistat on postprandial serum leptin levels in nondiabetic obese patients. Minerva Endocrinol. 2008;33:169–73. Schwartz GJ, Whitney A, Skoglund C, Castonguay TW, Moran TH. Decreased responsiveness to dietary fat in Otsuka Long-Evans Tokushima fatty rats lacking CCK-A receptors. Am J Physiol. 1999;277:R1144–51. Svendsen M, Rissanen A, Richelsen B, Rössnr S, Hansson F, Tonstad S. Effect of orlistat on eating behavior among participants in a 3-year weight maintenance trial. Obesity (Silver Spring). 2008;16:327–33. Torgerson JS, Hauptman J, Boldrin MN, Sjöström L. XENical in the prevention of diabetes in obese subjects (XENDOS) study: a randomized study of orlistat as an adjunct to lifestyle changes for the prevention of type 2 diabetes in obese patients. Diabetes Care. 2004;27:155–61. Track NS, McLeod RS, Mee AV. Can J Physiol Pharmacol. 1980;58:1484–9. Verdich C, Flint A, Gutzwiller JP, Näslund E, Beglinger C, Hellström PM, Long SJ, Morgan LM, Holst JJ, Astrup A. A meta-analysis of the effect of glucagon-like peptide-1 (7-36) amide on ad libitum energy intake in humans. J Clin Endocrinol Metab. 2001;86:4382–9. Williams DL, Grill HJ, Cummings DE, Kaplan JM. Endocrinology 2003;144:5184–7.
1350
M. Ellrichmann
Wren AM, Bloom SR. Gut hormones and appetite control. Gastroenterology 2007;132:2116–30. Zhi J, Melia AT, Guerciolini R, Chung J, Kinberg J, Hauptman JB, Patel NJ. Retrospective population-based analysis of the dose-response (fecal fat excreation) relationship of orlistat in normal and obese volunteers. Clin Pharmacol Ther. 1994;56:82–5. Zhi J, Mulligan TE, Hauptman JB. Long-term systemic exposure of orlistat, a lipase inhibitor, and its metabolites in obese patients. J Clin Pharmacol. 1999;39:41–6.
Part XIV
Pathology and Abnormal Aspects: Neurological
wwwwwwwwwwwwwww
Chapter 87
Gastrointestinal Disorders in Neurologically Impaired Children Alja Gössler and Karel Krafka
Abbreviations GER GERD NIP PEG
Gastroesophageal reflux Gastroesophageal reflux disease Neurologically impaired patients Percutaneus endoscopic gastrostomy
87.1 Introduction There is a saying “Liebe geht durch den Magen” – “The way to a man’s heart is through his stomach.” In neurologically impaired patients (NIPs), often eating or being fed something gustatorily pleasant is one of the highlights of their life. Despite this need for autonomic or assisted nutrition, problems and disorders of the gastrointestinal tract exist in a high percentage in this special group of patients, reducing their and their caregivers’ quality of life. Motility of the gut is controlled both by extrinsic inputs from the dorsal motor nucleus of the vagus and paravertebral sympathetic ganglia and by local reflexes mediated by intrinsic neurons of the enteric nervous system. Motility of the gastrointestinum results from interplay of a few fundamental mechanisms, including myogenic mechanisms, neurogenic propulsive mechanisms, and migrating neurogenic motor activity. This complex interplay can be interrupted or disturbed by abnormalities of one or more of these mechanisms. Severe neurologic impairment affects both children and adults considerably. In many cases, severe gastrointestinal problems accompany these children throughout their life (Moore 2008). Failure to thrive, difficulties in oral nutrition, recurrent vomiting, and respiratory tract infections caused by aspiration as well as loss of appetite and abdominal pain are results of frequent associated conditions. Dysphagia is present in 60% of NIPs, gastroesophageal reflux (GER) in 77%, resulting in chronic pulmonary aspiration in 41%; 74% of NIPs suffer from chronic obstipation (Del Giudice et al. 1999).
A. Gössler (*) Department of Pediatric and Adolescent Surgery, General Hospital Klagenfurt, St.Veiter Str. 47, 9020 Klagenfurt, Austria e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_87, © Springer Science+Business Media, LLC 2011
1353
1354
A. Gössler and K. Krafka
It is difficult for NIPs to express the pain and discomfort caused by gastrointestinal disorders to their caregivers. Thus, the interpretation of behavior indicating pain experience is a challenge for parents and health care professionals, resulting in late diagnosis. This chapter attends to the pathogenesis, diagnosis, and treatment of dysphagia, GER, and constipation as three of the main gastrointestinal problems in severe NIPs. Knowledge and understanding of these disorders will lead the way to earlier, timely diagnosis and treatment, thus preventing pain and further damage to the gastrointestinum as well as to the pulmonary system and improving the quality of life in this special group of patients.
87.2 Dysphagia 87.2.1 Background 87.2.1.1 Physiology of Normal Swallowing The act of swallowing is a complex process and requires the coordination of cranial nerves, the brain stem, cerebral cortex, and 26 muscles of the mouth, pharynx, and esophagus. The main cranial nerves that influence swallowing include the trigeminal (V), the facial (VII), glossopharyngeal (IX), vagus (X), and hypoglossal (XII). These nerves mediate the sensation and movement related to swallowing. Any abnormalities affecting these nerves, the cerebral cortex, mid brain, or cerebellum may have a negative impact on the individual’s ability to swallow. Four phases of swallowing can be distinguished: the oral-preparatory, the oral, the pharyngeal, and the esophageal phases. Dysphagia is characterized by a dysfunction in the sequential oral, pharyngeal, and esophageal phases of the swallowing process. The presence of abnormal movement patterns, for example tongue thrust in children with cerebral palsy, disrupts the normal movement of food from the anterior to the posterior regions of the mouth. Children with dysphagia have trouble in tongue control and bolus manipulation, problems with movement of food from the mouth to the pharynx as well as delayed pharyngeal swallow. Delayed or lack of initiation of the swallowing reflex results in increased risk of aspiration from an unprotected airway.
87.2.1.2 Causes of Dysphagia Dysphagia can be caused by acute onset of intracranial hemorrhage, cerebral infarction, or traumatic injuries (Schaller et al. 2006). Alternatively, it can be the result of congenital and chronic disorders resp. diseases: intracranial tumors, cerebral palsy, genetic disorders, encephalopathy, or neuropathy. As can be observed, dysphagia from chronic causes may worsen progressively or remain static. Worsening dysphagia will lead to deterioration of feeding and swallowing skills. In static dysphagia, swallowing skills will remain stable, or there may be a slow improvement.
87.2.2 Signs and Symptoms Difficulties in swallowing or transporting a food bolus may result in frequent aspirations, especially in fluid intake. Thus, it may eventually lead to pulmonary infections. Refusal of food, pain in swallowing, and inpatient or aggressive behavior of the children can be observed frequently. In some
87 Gastrointestinal Disorders in Neurologically Impaired Children
1355
children, dystrophy might be present. Food intake or feeding can show enormous duration. Excessive drooling, signs of fatigue, and respiratory distress during feeding are often found. Notable oral hypersensitivity to touch with spoon, food, or finger is sometimes present and should lead to further investigation considering dysphagia.
87.2.3 Diagnostics An understanding of normal and abnormal swallowing patterns as well as other developmental characteristics unique to children is essential for assessment. Identification and assessment of dysphagia is complex and requires the expertise of a multidisciplinary team. A multidisciplinary team can use various assessment methods including the following.
87.2.3.1 History of Feeding A feeding history obtained from parents or caregivers helps assessing severity of dysphagia considering preferred texture of meals, duration of feeding, and eventually occurring signs of aspiration.
87.2.3.2 Physical Assessment Physical examination should include clinical bedside evaluation by the speech pathologist and oral motor examination. Look for structural abnormalities of the tongue, palate, and jaw, difficulties in any of the four phases of swallowing or abnormalities in oral, laryngeal, or pharyngeal movement. A neurological examination for presence of dystonia, which might affect the ability to feed, as well as for any indices of other neurological or neuromuscular disorders should be performed. Upper motor impairments are common in children with neurogenic dysphagia and may affect the ability to control their head, neck, and trunk and thereby the subsequent ability to swallow and ability to self-feed. Presence of dystonia and dyskinesia will affect the patient’s ability to chew, manipulate the bolus in the mouth, and swallow. Those suffering from muscular hypotonus may experience poor coordination of posterior tongue resulting in difficulties with the pharyngeal phase of swallowing. The child’s hydration and nutritional status and its growth and development must be assessed. Since GER can be associated with dysphagia, the signs and symptoms of this disorder need to be investigated.
87.2.3.3 History of Medications Some neuroleptics and medications used to control seizures may reduce alertness and ability to swallow. Muscle relaxants administered to patients suffering from spasticity may affect their ability to swallow.
87.2.3.4 Radiological Examinations A videofluroscopic modified contrast swallow study may be helpful to assess the type of dysphagia and the individual risk of aspiration.
1356
A. Gössler and K. Krafka
87.2.4 Therapy Dysphagia often occurs parallel with many other streams of abnormal or delayed development. This may include impaired cognitive, oral motor, and fine and gross motor skills. A child’s developmental age and current level of functional swallowing skills, for example the ability to chew, and/or control and manipulate a bolus, should be considered in any management program. Management of dysphagia requires the expertise and cooperation of a multidisciplinary team. Members of this team include a medical practitioner, a speech therapist, a physiotherapist, an occupational therapist, a dietitian, and nurses as well as pediatric surgeons.
87.2.4.1 Monitoring Nutrition and Hydration In oral motor dysfunction there may exist an inadequate nutritional intake. This sometimes is even worsened due to difficulties in communicating desire for food and food preferences, inability to selffeed, coexisting GER, and aspiration. Assessment of the diet may help in maintenance of nutrition. This includes fluid and food intake and loss with consideration of supplemental nonoral feeding. To get a better understanding of this disorder and to find possible ways to improve symptoms, duration of meal times has to be observed.
87.2.4.2 Positioning the Patient Children with poor head control and poor trunk stability will require appropriate and individualized positioning techniques. The aim of positioning is to maintain a central body alignment. In children with severe cerebral palsy and feeding problems, feeding position can be dependent on degree of dysphagia and on its occurrence in the oral or pharyngeal phase. The chin tuck and 30º reclining position and flexed hips may be effective in eliminating aspiration in children with major oral phase swallowing problems. In children with minor oral phase but greater pharyngeal phase swallowing difficulties, the erect position with flexed neck and hips was recommended.
87.2.4.3 Diet Children with dysphagia may have difficulties managing different bolus sizes, flavors, and textures. Modifications will vary according to the needs of each child. Barium swallow studies may be used to determine the safest textures for each child with dysphagia. Children with neuromuscular disorders and weakened or uncoordinated swallowing may swallow a semi-solid consistency more easily as a single bolus. Thickened fluids are recommended as they assist in reducing the risk of aspiration.
87.2.4.4 Prevention of Complications Patients must be observed for signs of aspiration (coughing, choking, and respiratory distress). If aspiration is suspected, oral feeding should be stopped until the cause is investigated.
87 Gastrointestinal Disorders in Neurologically Impaired Children
1357
PIC 87.1 Laparoscopic view of a percutaneus endoscopic gastrostomy probe (PEG). The PEG penetrates the skin and the gastric wall to ensure endogastrical feeding
87.2.4.5 Nonoral Feeding In severe dysphagia, oral feeding can not only become extremely exhausting both for patients as well as for caregivers but can endanger the health of the patients by recurrent pulmonary aspirations. In these cases, enteral feeding through a percutaneous endoscopic gastrostomy (PEG) probe or a button, to which the gastrostomy feeding probe can be fixed is preferred. In severe cachexia, continuous feeding through jejunal tube can be a good answer to this problem (Valletta and Angelini 2004) (PIC 87.1).
87.3 Summary Points • Dysphagia can be caused by acute onset of intracranial hemorrhage, cerebral infarction, or traumatic injuries. Alternatively, it can be the result of congenital and chronic disorders and diseases: intracranial tumors, cerebral palsy, genetic disorders, encephalopathy, or neuropathy. • Caring for children with dysphagia is stressful and to some degree exhausting for patients as well as for caregivers or parents. • Information should be provided to assist caregivers to manage the child’s swallowing and feeding difficulties. These include strategies for oral feeding, preparation of nutritious meals, adaptive equipment, positioning techniques, positive interactive behaviors, and child’s progression in regaining swallowing skills. • A multidisciplinary approach with parental involvement in assessment and management of their child’s dysphagia is important. In severe dysphagia, enteral feeding through a PEG can be an extremely helpful approach. (Adapted with the permission of the publisher from The Joanna Briggs Institute. Identification and management of dysphagia in children with neurological impairment. Best Practice: Evidence Based Practice Information Sheets for Health Professionals. 2000;4(3):1–6.)
1358
A. Gössler and K. Krafka
87.4 Gastroesophageal Reflux 87.4.1 Background Gastroesophageal reflux is defined as the return of gastric content into the esophagus. Up to some amount, this condition is normal and is found in each healthy individual. In this healthy context, there are mechanisms preventing an acid or pepsin refluat to overwhelm the intact epithelium: the antireflux barrier and functioning clearance mechanisms of the esophagus. The antireflux barrier as the first line of mucosal defense consists of the lower esophageal sphincter and the crural diaphragm as well as of the angle of HIS. The latter is formed between the cardia at the entrance to the stomach and the gastric fundus, thus providing a valve which prevents reflux of gastric content from entering into the esophagus. The angle of His is created by the collar sling fibres and the circular muscles around this gastro-oesophagal junction. The second line of defense consists of the esophageal clearance function, which limits the time of contact between refluate and esophageal mucosa by eliminating esophageal content into the stomach. Salivary secretions from the esophageal glands as well as gravitation in an upright position support this clearing function. If these mechanisms are not sufficient, cellular defense, or mucosal resistance is discussed as the last line of defense against aggressive refluat (Boix-Ochoa and Ashcraft 2005). A malfunctioning of these barrier or clearance mechanisms as well as a delayed gastric emptying leads the way to pathologic GER. When the lower esophageal sphincter shows disturbed function, transient relaxations occur. These relaxations are not associated with swallows, a clearing wave of peristalsis, thus leading to regurgitation of gastric contents (Kawahara et al. 1997). In NIPs, the dysfunction of the esophageal motility causes longer contact of the acid gastric content with the esophageal mucosa leading to esophagitis and gastroesophageal reflux disease (GERD). Additionally these patients often lack upright positioning. Thus, gravity fails to support esophageal clearance function. Another cause for the high occurrence of GER in NIPs is considered the high incidence of epilepsy or spastic disorders elevating the muscular tonus and thereby increasing the intra-abdominal pressure. To prevent further damage to the esophageal mucosa and the pulmonary system, to ease the pain and discomfort caused by pathologic GER, early diagnosis and treatment is mandatory. Therefore, health care professionals should pay attention to typical signs of GER as well as to behavioral alterations in children with neurological impairment and evaluate the possible causes including GER.
87.4.2 Signs and Symptoms The diagnosis of GER is based on typical symptoms such as vomiting and regurgitation but is often diagnosed late, especially when symptoms like chronic pulmonary disease, recurrent pulmonary infections, and changes in behavior mislead the diagnosis. Dental erosions, recurrent laryngitis, and esophagitis reveal the aggressive character of gastric content on extragastric tissue. Recurrent episodes of uncontrollable singultus, belching, or acid foetor ex ore represent frequently existing and often underestimated symptoms. In some cases a sudden rotation of the head and neck to one side and the legs to the opposite side in a spastic dystonic mode, typically appearing in the postprandial period may be present. This condition is called the Sandifer syndrome and is associated with GER.
87 Gastrointestinal Disorders in Neurologically Impaired Children
1359
Symptomatic GER presents as GERD and may be associated with recurring pain, thus impairing health-related quality of life (Madisch et al. 2003). GER and GER-related esophagitis is often detected late due to severe deficits in communication – thus challenging the parents and caregivers abilities to interpret behavior indicating pain experience. It is difficult for NIPs to express the pain caused by acid reflux and esophagitis to their caregivers. This reflux-related pain in neurologically impaired children has in many cases a significant influence on the behavior of these patients, as it may lead to increased agitation with increased undirected movements, moaning, crying, and difficulties to pacify as well as auto-aggressive behavior. Special attention should be given to changes in behavior indicating pain experience (Figs. 87.1 and 87.2). In neurologically normal children the prevalence of symptoms associated with GER has been shown to peak at 4 months of age (67%) but decreases to 5% at 12 months of age (Nelson et al. 1997). In contrast, in children with neurological impairment pathological GER is a often lifelong condition. pH-metric reflux index
Correlation between pathologic pH -metric time of reflux and changes in behavior
30.00 25.00 20.00 15.00 10.00 5.00 0.00
children with autoaggression
children with agitation
children without agitation or autoaggression
Fig. 87.1 Correlation of pH-metric time of reflux and changes in behavior. Children with behavioral changes, represented by increased agitation and auto-aggression, showed highly pathologic levels of gastroesophageal reflux here measured in RI with a sign difference to those without behavioral changes Correlation between histologic degree of inflammation and changes in behavior Degree of histologic 3 inflammation
2
1
0 children with autoaggression
children with agitation
children without agitation or autoaggression
Fig. 87.2 Correlation of degree of inflammation of the esophageal mucosa and changes in behavior. Children with agitation and auto-aggressive behavior showed higher degrees of inflammation (grading after Savary Miller) and thus more severe inflammation than those without behavioral changes
1360
A. Gössler and K. Krafka
The patients suffering from GER frequently haves difficulties with oral nutrition which may result in weight loss or failure to gain weight. Delayed diagnosis of reflux and esophagitis not only results in difficult nutrition and recurrent pulmonary infections through aspirations, but also may finally result in mucosal damage leading to esophageal carcinoma arising from Barrett’s epithelium. Symptoms corresponding with pain for patients with severe neurological impairment were used to develop pain evaluation scales (Giusiano et al. 1995; Mc Grath et al. 1998). Despite the attempts to standardize the documentation of pain by medical professionals it has been shown that parents are best qualified to assess a typical behavior indicating that their child is feeling pain (Callery 1997). Therefore, medical professionals should respond to the caregivers’ judgment of abnormalities in behavior even if otherwise typical symptoms are missing. The clear correlation between auto-aggressive and agitated behavior and the severity of GERD shows that pathological GER may induce chronic pain in children with mental disabilities. Auto-aggressive or agitated behavior can be an indicator for GERD in neurologically impaired children and can be considered as a marker for reoccurring or first time diagnosed pathologic GER (Hunt et al. 2003). Adapted with the permission from the publisher from. Gössler A, Schalamon J, Huber-Zeyringer A, Höllwarth ME, Gastroesophageal reflux and behavior in neurologically impaired children. J Pediatr Surg. 2007 Sep; 42(9):1486–90.
87.4.3 Diagnostics 87.4.3.1 Physical Assessment A thorough examination of the patient should be performed to assess signs of GER or related conditions. It is important to know if the patient is ambulatory or nonambulatory, if he prefers upright or prone position. Knowing this may lead to better understanding at which position the child suffers from more pathologic reflux and in which position each single child is feeling more comfortable. There is a healthy circulus, which is lacking in children who are bound to the wheel chair or the bed. The patient must be ambulatory, thus both increasing intestinal mobility and preventing obstipation. This is especially important, since the latter causes increased abdominal pressure which is associated with a higher incidence of GER. Movement patterns should be observed as well as markers of auto-aggressive behavior like scratches or hematomas or bite marks. Nutritional status as well as skin hydration should be examined; ideally, the development of weight and height should be stated. Examination of the teeth to detect possible erosions from acid refluat should be included in the examination. Auscultation, palpation, and percussion of the abdomen have to be performed to assess peristalsis, possible abdominal tumors, or hernias of the abdominal wall.
87.4.3.2 History In most cases, a typical history of symptoms can be found. A thorough interview with patients and/ or caregivers is mandatory.
87 Gastrointestinal Disorders in Neurologically Impaired Children
1361
87.4.3.3 Sonography Sonography is performed to investigate occurrence of GER, gastric emptying, and to exclude abdominal masses as the cause of increased abdominal pressure or barrier of intestinal movement. 87.4.3.4 Contrast Study A contrast study of the upper gastrointestinal tract is performed to assess anatomy, gastric emptying, possible hiatal hernias, or esophageal strictures. In some patients, gastroesophageal reflux can be observed during this study, showing the heights of return of gastric content into the esophagus (PIC 87.2). 87.4.3.5 24-h pH Monitoring pH is monitored to assess the total as well as the percentage time of acid refluat into the esophagus as well as the number and duration of occurring refluxes and the clearance function of the esophagus. Since this investigation monitors only acid refluat, a 24-h combined impedance and pH monitoring is now standard for investigation of acid and nonacid refluat (Del Buono et al. 2006) (PIC 87.3, PIC 97.4). 87.4.3.6 Endoscopy Esophagogastroduodenal endoscopy is performed to assess or rule out mucosal damages. Esophagitis is assessed according to Savary and Miller. Ulcers, strictures, or Barrett mucosa can be found in
PIC 87.2 Upper gastrointestinal contrast study. Contrast study showing gastroesophageal reflux up to the upper half of the esophagus
1362
A. Gössler and K. Krafka
PIC 87.3 Measurement of gastroesophageal reflux by impedance pH monitoring. Impedance pH monitoring shows reflux of acid gastric content in the esophagus. Note drop of pH (lowest line) during reflux, the beginning and ending of which is marked with vertical lines
endoscopic examinations. Biopsies have to be taken, especially since in up to 48% of inflammation of the esophageal mucosa this has not been seen under macroscopic examination. During endoscopy, possibly occurring hiatal hernias can be observed. 87.4.3.7 Esophageal Manometry Manometrically, the tonus, resting pressure, length, and transient as well as swallow-induced relaxations of the lower esophageal sphincter can be documented. The common cavity phenomenon with equalization of gastric and esophageal pressure is the manometric equivalence of reflux. Esophageal peristalsis and occurrence of GER can be investigated.
87.4.4 Therapy 87.4.4.1 Conservative Treatment A special diet which consists of less acid, low fat, and not too spicy food should be administered. Elevation of the head of the bed and, especially in children with frequent aspirations, thickened feedings will support gravity’s part of esophageal clearance.
87 Gastrointestinal Disorders in Neurologically Impaired Children
1363
PIC 87.4 Positioning control of the impedance probe. Radiographic control of the endoesophageal position of the impedance probe
Reduction of acid refluat may permit healing of the esophageal mucosa in cases of mild to moderate GER. Thus both a better distal esophageal peristalsis and fewer transient lower sphincter relaxations may result, disrupting the circulus vitiosus of GER leading to esophagitis enhancing GER. The superior efficacy and safety of proton pump inhibitors (PPIs) have changed the diagnostic and therapeutic recommendations. PPIs belong to a group of chemically related compounds whose primary function is the inhibition of acid production in the final common metabolic pathway of gastric parietal cells. The suppression of gastric acid secretion achieved with H2 receptor antagonists has proved to be suboptimal. PPIs have been shown to be more effective than them (Vandenplas, Current Concepts in the Medical Therapy of GER in Children). In combination with mucosa protecting drugs like sucralfat, these mediations effectively may lead to healing of reflux-associated esophagitis. In some cases motility enhancing, prokinetic drugs (cisapride, erythromycin) can be helpful. Cisapride is a prokinetic mainly acting via indirect release of acetylcholine from the myenteric plexus.
87.4.4.2 Surgery How can the surgeon help? GERD has been shown to occur with an increased incidence in neurologically impaired patients, representing a major factor in the long-term care and management and life in these patients.
1364
A. Gössler and K. Krafka
Reduction of gastric acid production can improve symptoms of GERD. But while some improvement after healing of esophagitis may be present after some weeks or months of medication in mild GER, conservative therapy does not help with the natural history of development of gastroesophageal reflux disease. Reflux-depending symptoms and pathological oesophageal findings show recurrence in up to 80% within the first year. Thus, severe cases might require a long-term or even lifelong administration of medications (Stanghellini 2003). In cases of severe GER, anatomic anomalies, failure to thrive, and recurrent pulmonary infections, fundoplication will lead to an impressive overall improvement of symptoms, health, and quality of life (Ceriati et al. 2006; Spitz and McLeod 2003).
87.5 Summary Points • Pathologic GER is the result of malfunction of the barrier function and of the esophageal clearance. • If left untreated or unrecognized, GERD can lead to damage of the esophageal mucosa or the pulmo by recurrent infections, failure to thrive, and changes in behavior like auto-aggression and agitation. • Treatment includes special diet, acid suppressing medications, and surgery in severe cases.
87.6 Obstipation 87.6.1 Background 87.6.1.1 Physiology The function of the colon is modulated through several separate systems, including neural, endocrine, and luminal factors. There are two kinds of nervous systems controlling the colon – the intrinsic colonic nervous system and the extrinsic colonic nervous system. While the intrinsic nervous system consists of nerve cell bodies and endings that are located between the circular and the longitudinal muscle coats in the submucosal ganglia (Meissner’s plexus) and the myenteric ganglia (Auerbach’s plexus), the extrinsic acts through sympathetic and parasympathetic functions. Stimulation of parasympathetic fibers increases the overall activity of the gastrointestinal tract. It initializes peristalsis, increases local blood flow and intestinal secretion, and promotes the defecation reflex. The sympathetic innervations of the gastrointestinal tract show an inhibiting effect on noradrenalin on the enteric nerves. It is supposed to regulate the contraction of the internal anal sphincter. The external sphincter is innervated via branches of the pudendal nerves, as is sensation from the perianal area and perineum, whereas tension and stretch in the rectal wall and proximal part of the anal canal is carried in the pelvic nerves. Through mass movements of the distal colon, feces are pushed into the otherwise empty rectum prior to defecation. The resulting distension and filling of the rectum initiates the sensation of the need
87 Gastrointestinal Disorders in Neurologically Impaired Children
1365
to defecate. Now relaxation of the internal anal sphincter and voluntary relaxation of the external sphincter promote defecation. Defecation can be delayed by voluntary contraction of the external anal sphincter and the urge to defecate gradually decreases in intensity over a period of minutes. The central nervous system contributes to the regulation of bowel function, and is involved in the timing and initiation of defecation. Asides from neurologic function, three elements are important for normal bowel function: dietary bulk, fluid intake, and exercise (Winge et al. 2003). Any factor that influences neurologically induced peristalsis or these three elements may cause bowel dysfunction and chronic constipation. Chronic constipation is a common problem in children. And it is particularly common in children with disabilities. Other possible causes of constipation include the disregard of the impulse to defecate, emotional conflicts, overuse of laxatives, or prolonged dependence on enemas as well as it can result as a side effect of some commonly used medications, especially antiepileptic drugs or sedative. There is frequently a delay in the recognition and adequate treatment of constipation in children. Difficulties in communication make it nearly impossible for neurologically impaired children to express the extreme discomfort caused by constipation. This leads to existence of problems and symptoms for months or years before their importance is understood.
87.6.2 Signs and Symptoms Constipation is defined as small, infrequent, or difficult bowel movements. But what is infrequent? Constipation always must be regarded in relation to the single patient. When defecation costs extreme effort, hurts and produces hard stool, constipation is present. In most cases, constipation might just be a minor annoyance. But sometimes it presents as an acute abdominal condition with massive pain and respiratory distress caused by extreme metoristic abdomen and elevation of the diaphragm. Adequate treatment of constipation provides relief for the child and impressively leads to an improvement in appetite and sometimes behavior. If left untreated, chronic constipation may lead to painful fecal impaction. It can result in hemorrhoids, fissures, and abdominal pain. Additionally, elimination of hardened stool can become extremely difficult and painful. Pseudo incontinence can occur, when only fluid or softened stool can pass around hard and impacted stool without possibility to control this. Many patients loose their appetite and refuse their meal, caused by nausea, pain, or simply the feeling of metoristic and “full” abdomen and frequent episodes of abdominal cramping. Chronic obstipation can lead to an acquired afunctional distention of the colon with reduced or poor colonic peristalsis. As a secondary result of excessive use of laxatives, in some patients fluid and electrolyte depletions occur. Especially in the neurologically impaired, this discomfort without possibility to communicate their problems can lead to an increase in auto-aggressive behavior as well as in disquietness and aggression. Chronic obstipation is frequently associated with gastroesophageal reflux. A possible cause for this is found in the increased abdominal pressure as well as in the frequently additionally delayed gastric emptying.
1366
A. Gössler and K. Krafka
87.6.3 Diagnostics 87.6.3.1 Physical Assessment Especially in the neurologically impaired, an entire physical examination should be performed. The abdomen should be auscultated to determine absence or presence of peristalsis as well as to determine the kind of peristalsis. Check the patient’s abdomen for visible signs of distension or abdominal wall hernias, which may interfere with the necessary amount of abdominal pressure during defecation. Palpation of the abdomen can show masses of hardened stool as well as abdominal tumors. In many cases, left lower quadrant pain can be found in severe acute constipation. In many patients, signs of healed or yet active anal fissures or hemorrhoids can be found. Less frequently, anal polyps or rectal prolapse may be present. To some amount, sonography can assist in determining the rectal distension in patients in whom a rectal examination cannot or can only be performed with difficulty. A thorough neurologic examination concerning anal, cremasteric, and abdominal wall reflexes should be performed. Labor chemistry should be performed with special regard to the electrolyte status and thyroid status to rule out hypothyroidism in patients whose symptoms proved to be refractory to dietary management.
87.6.3.2 Radiology Always a screening sonography should be performed to evaluate peristalsis, exclude abdominal or retroperitoneal tumors, and show rectal distension (PIC 87.5). An abdominal X-ray can be useful in both indicating the nature of the problem and assessing the degree of constipation and, hence, the appropriate treatment.
PIC 87.5 Transabdominal sonography of the rectum. Sonography showing rectal distension by stool masses
87 Gastrointestinal Disorders in Neurologically Impaired Children
1367
87.6.4 Therapy The degree of neurologic involvement and abnormalities in defecation dynamics will guide in developing an appropriate management plan. A cleaning phase of the colon followed by medical and dietary management as well as changes in life style – if possible – is often successful. Since continence may not always be a realistic possibility in people with severe disabilities, the goal of therapy in these cases is to soften the stool so that it is painless to pass.
87.6.4.1 Treatment of Acute Constipation To prevent abdominal cramping pain and to lead the way to an effective use of laxatives, the entire colon and rectum should first be emptied. This is achieved by use of enemas. In less severe cases, some suppositories can be used alternatively. These normally consist of CO2-supps to achieve a sensation of filled rectum, which results in rectal peristalsis. Glycerin supp can soften the distal stool masses to assess easier passage. In severe cases, digital disimpaction under anesthesia by a physician may be necessary, including use of irrigation to clean the colon. It should not be performed ambulatory, since freedom of pain has to be assured during this otherwise hurting and unpleasant procedure. After this cleaning phase, oral medications are administered.
87.6.4.2 Management of Chronic Constipation It should be aimed for frequent, at least each other day, effortless and – and this should not be underestimated – painless defecation. Especially in children with many physical and sometimes psychological problems, alleviation if not freedom of pain has to be achieved to improve their and consequently their caregivers’ quality of life. The need for patience has to be explained to caregivers, since sometimes medication and treatment can be necessary for several months to years. Laxatives and fiber therapies may be effective in improving bowel movement frequency. In less severe cases, adequate fluid and fiber intake alone may lead to adequate improvement. If the physical condition of the neurologically impaired child allows, some physical training should be performed. Since some medications are known to worsen colonic peristalsis, anticholinergics, opiates, antacids, and antiepileptics as well as sedatives should be reduced as possible. If these modifications alone fail to be effective, medication to ensure soft and easy-to-pass stool should be administered. For a short period, lactulose can be useful. In cases of extreme dilatation of the colon caused by prolonged, sometimes yearlong delay in defecation and tailback of stool masses, we found dihydroergotamin to be a good choice. Often used and very helpful is macrogol, which ensures the transport of softer, more voluminous stool masses. Medications such as pyridostigmin are also known to be effective in chronic constipation. Before administering these drugs, it has to be checked for contraindications. Enemas and suppositories can relieve constipation temporarily. Still, they should be used only for a short period of time and not regularly. Inflammations, interfere with natural bowel muscle control and last, but not least, the discomfort caused for the patients should lead to an use only prior to oral antiobstipation medication.
1368
A. Gössler and K. Krafka
When basic understanding of this training can be achieved, biofeedback techniques may be of some help in patients who initiate contraction of the external anal sphincter during the attempt to defecate. Biofeedback shows these children how they can change voluntarily the striated muscle contractions. How can the surgeon help? If changes in lifestyle, diet, and medication treatments fail to be effective, surgical therapy can improve bowel emptying and soften stress caused by this problem: −− Rectoscopy with anal dilatation or myotomy. −− Appendicostomy (Malone procedure) (Malone et al. 1990; Roberts et al. 1995; Zerhau et al. 2008) with antegrade continence enema (ACE). −− Colostomy.
87.7 Summary Points • Obstipation is caused by immobility, inadequate fluid and fiber intake, and disturbed neuronal control of the colon as well as by some frequently used medications. • If left untreated, it can lead to severe abdominal pain, anal fissures and refusal of food. The discomfort sometimes is shown by agitation and disquietness. • Changes in diet, and physical training pave the way to improved colonic function. • Laxatives, enemas, and stool softening medications help improve the symptoms and maintain regular stool passes. • Surgical treatments like dilatation of the anal sphincter or, in severe cases, the Malone procedure may be very helpful treatments.
87.8 Surgical Therapy 87.8.1 PEG in Dysphagia How can the surgeon help these patients? Nutritional support is often an integral part of patient management in this group of children. In patients where tube feeding is supposed to be of short duration, a nasogastric tube can be considered. There are three principal access routes used for nutrient delivery: oral, enteral, and parenteral. If oral intake is inadequate or contraindicated, then enteral tube feeding is preferable. Still, there are a number of problems associated with the use of the nasogastric tubes over the longer-term. A gastrostomy or jejunostomy is often the only way to secure the daily caloric need of the body in patients with a failure to thrive. Gauderer and Ponsky introduced a PEG 30 years ago. It is a safe and effective method for adequate enteral nutrition. The main indication for this is the inability to swallow or recurrent aspirations during the process of swallowing in children and adult patients with dysphagia as well as in extremely dystrophic or cachectic patients with GER. Their nutritional status as well as their quality of life improves sometimes impressively from the benefits of enteral feedings via gastrostomy or PEG (PIC 87.1). Since neurologically impaired children pose the main indication and frequently suffer from additional GER, this condition has to be evaluated prior to gastrostomy. Button: The major advantage of the PEG is that it allows a great number of patients to be discharged into family care. The simple button-gastrostomy tube is a next benefit in these cases.
87 Gastrointestinal Disorders in Neurologically Impaired Children
1369
87.8.1.1 Laparoscopy Laparoscopy in pediatric surgical practice has become a reality. Laparoscopy in children requires a special care and provides a special caring. Generally, low pressures about 6 mmHg are used in children up to 8 years. We use instruments and ports no larger than 2 to 5 mm in diameter, thus providing very small wounds and minimal scarring. Advantages of the laparoscopic approach are easily understood: there is a cosmetically better outcome, less tissue dissection, and disruption of tissue planes. Postoperatively, younger and older patients show less pain and overall there is a low intraoperative and postoperative complication rate. Laparoscopic-assisted PEG: In patients with V-P shunt or generally in patients after another surgical procedure of the epigastric region a laparoscopy with 3 mm camera is performed for visualization of the intraperitoneal part of the V-P shunt or for visualization of the adhesions between the stomach and the peritoneum (PIC 87.1). Laparoscopic gastrostomy: Laparoscopic gastrostomy is indicated when a PEG cannot be performed or is contraindicated (obstruction of the esophagus, colon or omentum are overlaying the stomach). In these patients, the laparoscopic gastrostomy is the best opinion. Laparoscopic jejunostomy eliminates the risk of aspiration associated with gastrostomy feeding. It will help assessing a healthier nutritional status in patients prior the further abdominal surgery. In severely malnutrition, a jejunal tube can be inserted via the gastrostomy catheter for continuous enteral feeding in contrast to bolus feeding via the gastrostomy tube. It is posed under endoscopic surveillance into the jejunum. 87.8.1.2 Contraindications of PEG Absolute contraindications for this procedure are: Malignant obesity Acute inflammation process of the abdominal wall Peritoneal dialysis Malignant ascites Tumors of abdominal wall Tumors of the left lobe of the liver with stomach dislocation Peritonitis Pancreatitis Portal hypertension Coagulation disorders Relative contraindication: Displacement of the stomach (e.g., scoliosis). 87.8.1.3 Complications Seldom, but possible complications are gastric and bowel perforation, gastric bleeding, migration of the gastric tube or wound infection, and abscess of the abdominal wall.
87.8.2 Fundoplication and Pyloroplasty in GERD How can the surgeon help?
1370
A. Gössler and K. Krafka
87.8.2.1 Pyloroplasty In delayed gastric emptying, a pyloroplasty can be performed. In most cases, this surgery will be performed together with an antireflux procedure to ease emptying of the stomach. When conservative treatment fails, or age, existing anatomic anomaly or severe symptoms make it necessary, surgery can present help against symptoms and GER-associated problems. 87.8.2.2 Fundoplication In fundoplication, a complete or partial wrap of the fundus around the intra-abdominal esophagus is performed. Thus transient lower esophageal relaxations, which pose one major mechanism of pathologic GER, can be reduced successfully. Additionally the reconstruction of the angle of HIS prevents complete relaxation of the lower esophageal sphincter via elevation of the intragastric pressure. The diaphragmatic crura are narrowed to prevent the fundoplicate from slipping into the thorax as well as to reduce a hiatal hernia (PIC 87.6). This surgery can be performed laparoscopically or in open access surgery. The 360° wrap or Nissen fundoplication has been used much more frequently than other antireflux operations. For this wrap, the intra-abdominal esophagus is mobilized to assure an adequate length. Then a stomach cuff is created by passing the fundus dorsally around the distal esophagus. Anteriorly the left and right margins of this fundic wrap are sutured together in a “floppy” way. This wrap reacts as a valve, when gastric pressure is transmitted to the distal esophagus, raising the lower esophageal sphincter pressure (PIC 87.7). In dorsal or ventral semifundoplications, the alternative surgical aim is to correct anatomic anomaly while permitting a physiologic amount of GER. The wrap is created similarly. A partial – 180°–270° wrap is constructed and sutured to the esophageal wall. Fundoplication can be a life-saving procedure in those patients who have severe manifestations such as recurrent aspiration or extreme failure to thrive.
PIC 87.6 Laparoscopic view of the narrowing of the diaphragmatic crura. Note the sutures to the diaphragmatic crurae which strengthen the distal esophageal sphincter and prevent the yet-to-be prepared fundoplicate from slipping into the thorax
87 Gastrointestinal Disorders in Neurologically Impaired Children
1371
PIC 87.7 Laparoscopic view of the fundoplicate. Laparoscopic achieved fundoplication after Nissen. Note the 360° fundic wrap around the distal esophagus
Antireflux surgery in the neurologically impaired carries a higher risk for wrap failure than in the neurologically normal patients. The most significant association of recurrent reflux in our own study was found with the state of preoperative dystrophy as an indicator of a catabolic status (Goessler et al. 2007). An optimal, continuously anabolic preoperative nutritional status in NIP is important for the success of surgical treatment. This condition should be achieved by adequate enteral and/or parenteral therapy before any surgical procedure is planned (Falcão 2002; Kudsk et al. 2003). The results of our own analysis showed that the surgical procedure results in significantly reduced GER-related symptoms and, as reported by the caregivers, a striking improvement of the quality of life (Teixeira et al. 2009) considering the often most difficult situation of caring and adequate feeding. Furthermore, an impressive reduction of agitation and auto-aggressive behavior was reported by the caregivers.
87.8.2.3 Complications Seldom, but possible complications can be intraoperative injury to the gastric, intestine, or esophageal wall as well as lesions of liver or spleen. While performing the wrap, in cases with inadequate viability, the vagal nerve can be hurt, thus posing a risk for functioning gastric emptying. Other possible complications would be disruption of the fundoplicate, slipped fundoplicate, or to narrow wrap with postoperative need of bouginage or reoperation.
87.8.3 Anorectal Myectomy or Malone Procedure in Chronic Pbstipation How can the surgeon help?
1372
A. Gössler and K. Krafka
87.8.3.1 Anorectal Myectomy or Sphincter Dilatation In some patients failing to correspond to the aforementioned conservative treatment of obstipation, the removal of a strip of rectal muscular wall can ease the process of defecation. This removal starts right below the dentate line and extends upwards to the level of the puborectalis muscle. The lowest part of the internal sphincter is not to be divided. In less severe cases, anal sphincter dilatation during anesthesia can be a helpful procedure.
87.8.3.2 Malone Procedure The Malone Procedure is easily understood when considering that it is just another way to administer an enema. In patients with neurological disorder, this can be extremely helpful not only because the enema runs through more efficiently when it is administered to Malone’s site but also is easily accessible. It consists of a special appendicostomy. The appendix is open or laparoscopically connected to the umbilicus or to the skin in right hypogastrium. The appendix is invaginated into the coecum or embedded submucosally after an incision of taenia libera. Thus, a valve mechanism is created. If an appendectomy has been performed prior, a continent neo-appendicostomy can be created with a flap from the wall of the coecum. The antegrade continence enema has met wide acceptance in the treatment of intractable fecal incontinence or for the use of repeated enemas in chronic intractable obstipation.
87.9 Summary Points • In severe or distressing problems, surgery can be helpful, lead to less pain, and thus increase patients’ and caregivers’ quality of life. • In dysphagia, gastral feeding via a PEG or jejunal continuous feeding via a jejunal tube can lead to a shortened duration of feeding and improvement of the nutritional status. • In GER, a fundoplication or semi-fundoplication reduces the risk of damage to the esophagus and the pulmonary system and eases pain and reflux-associated symptoms. • In obstipation, the Malone procedure is an easy access to perform frequent enemas.
87.10 Applications to Other Areas of Health and Disease Dysphagia, GER, and obstipation present frequently existing conditions in neurologically impaired children. Understanding these conditions, knowledge of their signs and symptoms, especially of their tendency to lead to changes in behavior, will lead to earlier recognition and thus to earlier intervention and therapy. This poses extreme importance to caregivers, parents, nurses, doctors, and all therapists who perform training with this special group of patients.
87 Gastrointestinal Disorders in Neurologically Impaired Children
1373
Definitions Dysphagia: In dysphagia, the transport of the fluid or food bolus from mouth to pharynx to esophagus is disturbed. Gastroesophageal reflux: In GER, gastric content returns into the esophagus via a disturbed barrier function Gastroesophageal reflux disease: Pathologic gastroesophageal reflux with damage to the esophageal mucosa or severe symptoms Obstipation: Obstipation is characterized by chronic infrequent, difficult passage of hardened stool.
Key Facts of gastrointestinal problems in neurologically impaired patients (NIPs) In 60% of NIPs, dysphagia is present Up to 77% of NIPs suffer from pathologic gastroesophageal reflux, resulting in chronic pulmonary aspiration in 41%. 74% of NIPs show typical signs and problems of chronic constipation. Due to deficits in verbal communication, the pain and discomfort resulting from these conditions is difficult to be expressed to caregivers and parents, resulting in late diagnosis and treatment. Knowledge of signs and symptoms is mandatory to achieve earlier, timely diagnosis and therapy, thus improving quality of life of neurologically impaired children.
References Callery P. Int J Nurs Stud. 1997;34:27–34. Ceriati E, De Peppo F, Ciprandi G, Marchetti P, Silveri M, Rivosecchi M. Acta Paediatr Suppl. 2006;95:34–7. Del Buono R, Wenzl TG, Rawat D, Thomson MJ. Pediatr Gastroenterol Nutr. 2006;43:331–5. Del Giudice E, Staiano A, Capano G, Romano A, Florimonte L, Miele E, Ciarla C, Campanozzi A, Crisanti AF. Brain Dev. 1999;21:307–11. Falcao MC, Tannuri U. Rev Hosp Clin Fac Med Sao Paulo. 2002;57:299–308. Giusiano B, Jimeno MT, Collignon P, Chau Y. Methods Inf Med. 1995;34:498–502. Goessler A, Huber-Zeyringer A, Hoellwarth ME. Acta Paediatr. 2007;96:87–93. Hunt A, Mastroyannopoulou K, Goldman A, Seers K. Int J Nurs Stud. 2003;40:171–83. Boix-Ochoa J, Ashcraft K. In: Ashcraft KW, Holcomb GW, Murphy JP editors. Pediatric surgery. Philadelphia: Elsevier Saunders; 2005. p. 383–404. Kawahara H, Dent J, Davidson G. Gastroenterology. 1997;113:399–408. Kudsk KA, Tolley EA, DeWitt RC, Janu PG, Blackwell AP, Yeary S, King BK. JPEN J Parenter Enteral Nutr. 2003;27:1–9. Madisch A, Kulich KR, Malfertheiner P, Ziegler K, Bayerdörffer E, Miehlke S, Labenz J, Carlsson J, Wiklund IK. Z Gastroenterol. 2003;41:1137–43. Malone PS, Rensley PG, Kiely EM. Lancet. 1990; 336(8725):1217–8. McGrath PJ, Rosmus C, Canfield C, Campbell MA, Hennigar A. Dev Med Child Neurol. 1998;40:340–3. Moore SW. Pediatr Surg Int. 2008;24:873–83. Nelson SP, Chen EH, Syniar GM, Christoffel KK. Arch Pediatr Adolesc Med. 1997;151:569–72. Roberts JP, Moon S, Malone PS. Br J Urol 1995;75:386–9. Schaller BJ, Graf R, Jacobs AH. Am J Gastroenterol. 2006;101:1655–65. Spitz L, McLeod E. Semin Pediatr Surg. 2003;12:237–40. Stanghellini V. Drugs Today (Barc). 2003;39 Suppl A:15–20. Teixeira JP, Mosquera V, Flores A. Hepatogastroenterology. 2009;56:80–4. Valletta E, Angelini G. Pediatr Med Chir. 2004;26:112–8. K Winge K, Rasmussen D, Werdelin LM. J Neurol Neurosurg Psychiatry. 2003;74:13–9. Zerhau P, Husár M, T?ma J. Rozhl Chir. 2008;87:593–5.
wwwwwwwwwwwwwww
Chapter 88
Dysphagia: Neurological and Behavioral Aspects Dorianne Feldman and Marlís González-Fernández
Abbreviations VFSS FEES SLP ALS MS TBI PD PEG
Videofluroscopic swallow study Fiberoptic endoscopic evaluation of swallowing Speech-language pathologist Amyotrophic lateral sclerosis Multiple sclerosis Traumatic brain injury Parkinsons disease Percutaneous endoscopic gastrostomy tube
88.1 Introduction Swallowing dysfunction is a frequent complaint in older individuals (White et al. 2008), particularly if they have underlying dementia or a neurologic conditions (Easterling and Robbins 2008). Disordered swallowing or dysphagia can be classified as a primary (i.e., stroke) or secondary problem (i.e., radiation fibrosis) with many causes and complications. Age-related changes also accompany normal swallowing (Barczi et al. 2000; Humbert and Robbins 2008). Dysphagia can significantly affect the health, quality of life, and nutritional status of those affected. The ramifications are serious and extremely concerning, especially in the elderly population, often overwhelmed by other medical conditions and social isolation and in whom swallowing difficulty is usually neurologically mediated (Humbert and Robbins 2008). There are many challenges when caring for these individuals. Changes in mental status, deficiencies in immunologic reserve, poor nutrition, and end-of-life issues increase medical complexity (White et al. 2008). As the population over age 65 increases, swallowing disorders and their impact on quality of life are of great importance.
M. González-Fernández (*) Department of Physical Medicine and Rehabilitation, School of Medicine, Johns Hopkins University, Baltimore, MD e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_88, © Springer Science+Business Media, LLC 2011
1375
1376
D. Feldman and M. González-Fernández
88.2 Epidemiology In the United States, about 15% of community dwelling individuals over age 65 have some type of swallowing dysfunction; whereas in hospitals, the prevalence is around 13–14% (Humbert and Robbins 2008; Dray et al. 1998). In institutionalized settings, the rates rise; affecting up to half of all individuals (Easterling and Robbins 2008). Gilmore et al. (1995) evaluated 290 individuals residing in a nursing facility and demonstrated that of those eating less than half their meals, 30% had dysphagia and were unable to maintain adequate nutrition or an optimal weight with oral feeding. Groher and McKaig (1995) found that out of 740 individuals living in an institutionalized setting, 31% were receiving a modified meal. Of those studied, dementia was the most common underlying medical condition (53%) followed by stroke (25%).
88.3 Physiology of Swallowing Swallowing is a finely regulated, intricate process that requires cortical and subcortical input from various regions of the brain to move food and/or liquid from the mouth to the stomach without compromising the airway (White et al. 2008). The physical act of swallowing can be divided into four discrete stages: oral preparatory, oral propulsive, pharyngeal, and esophageal (Matsuo and Palmer 2008). A normal swallowing videofluorographic sequence is described in Fig. 88.1. In the oral phase, the bolus is prepared and then propelled under voluntary control with the assistance of the tongue to the back of the oral cavity and pharynx (Gonzalez-Fernandez and Daniels 2008). Once the bolus reaches the pharynx, the pharyngeal phase begins, consisting of a series of coordinated movements that drive the bolus to the upper esophageal sphincter. The esophageal stage starts as the bolus moves through the upper esophageal sphincter, esophagus, lower esophageal sphincter, and finally into the stomach. The pharyngeal and esophageal phases are not voluntarily controlled. Dysphagia can occur as a result of problems in one or more of these phases or locations (Matsuo and Palmer 2008). Key features of swallowing are detailed in Table 88.1.
88.4 Dysphagia Many conditions can lead and/or increase susceptibility to dysphagia; making the identification process challenging, particularly in older individuals (Schindler and Kelly 2002). Nonetheless, neurologic disorders are the most common cause of swallowing disorders. Problems with sensation, muscle function/coordination, and behavioral issues are paramount and are usually related to a multifactorial rather than an isolated trigger (Easterling and Robbins 2008). Oropharyngeal disorders are typically characterized by impairments in swallow initiation and food propulsion to the esophagus and are strongly linked to aging (Gleeson 1999; White et al. 2008). Associated findings include difficulty in chewing and manipulating the quantity of material consumed, sensation that the swallowed bolus is trapped (globus sensation), coughing, choking, dysphonia (wet-sounding vocalization), or passing of contents backwards through the nose or mouth (Schindler and Kelly 2002; White et al. 2008). Tongue pressure also declines with increased age contributing to disordered pharyngeal phase function (Schindler and Kelly 2002). In contrast, esophageal phase dysfunction is the result of esophageal pathologies which restrict or slow food entry or passage (i.e., esophageal strictures, carcinoma, and diverticulae; Dray et al. 1998).
Fig. 88.1 Normal swallowing sequence eating a soft food (banana with a barium coating) on lateral projection videofluoroscopy. (a) Food is placed in the mouth. (b) Chewing starts. (c) Food that is of the appropriate size and consistency is moved posteriorly while chewing continues. (d) Food reaches the valleculae while chewing continues. (e) The remaining food from the mouth is moved posteriorly. (f) Swallowing ensues. (g) Upper esophageal sphincter opens to allow food passage. (h) Upper esophageal sphincter closes as food continues down in the esophagus
1378
D. Feldman and M. González-Fernández
Table 88.1 Key features of swallowing 1. Swallowing is the basic process that allows passage of food from the mouth to the stomach. 2. Swallowing can be divided into four discrete stages: oral preparatory, oral propulsive, pharyngeal, and esophageal. 3. Normal swallowing requires cortical and subcortical input from various regions of the brain to move food and/or liquid from the mouth to the stomach. 4. Failure to protect the airway during swallowing may result in aspiration. 5. Gastroesophageal reflux can result in aspiration after the swallow. 6. Dysphagia rehabilitation can improve swallowing and may allow patients to continue oral feeding. This table lists the key factors of swallowing including swallowing phases, neural control, and major consequences of swallowing dysfunction
Fig. 88.2 Cervical spine osteophytes causing dysphagia. This lateral projection radiograph abstracted from a videofluorographic swallowing study illustrates a large cervical anterior osteophyte (bony prominence) causing mechanical obstruction of the foodway resulting in dysphagia. The dark black material is barium contrast passing by the obstruction
Orthopedic changes in the cervical spine associated with osteoarthritis (bony projections or osteophytes) cause decreased neck range of motion, thus hindering swallowing efficacy (Gleeson 1999; Achem and Devault 2005). Osteophytes in the cervical spine can push forward on the posterior pharyngeal wall creating a physical barrier to overcome in order to swallow (Fig. 88.2). Other causes include polypharmacy and generalized age-related changes with breathing, gustation, sensation, and motor function of the tongue, face, neck, pharynx, and larynx (Gleeson 1999; Schindler and Kelly 2002; Leslie et al. 2005; Humbert and Robbins 2008). In general, with aging, all processes operate at a reduced pace. In fact, studies have demonstrated a reduction in oropharyngeal and hypopharyngeal peristalsis along with a decline in the range of laryngeal movements (Leslie et al. 2005). Most importantly, and physiologically, vocal fold adduction secures the airway and must be coordinated with the speed of content passage in order to prevent aspiration (Fig. 88.3). If the sequencing of these processes is not timely, the airway can be compromised (Ren et al. 1993; Leslie et al. 2005).
88 Dysphagia: Neurological and Behavioral Aspects
1379
Fig. 88.3 Aspiration of thin liquids. Lateral projection videofluoroscopy. Following administration of liquid barium contrast (dark gray) and a swallowing attempt, incomplete opening of the upper esophageal sphincter (UES) caused spillage of liquid into the pharyngeal vestibule (open arrow), through the vocal folds (solid arrow), and into the trachea (white arrow)
Swallowing safety is critical and is of utmost importance in preventing complications such as weight loss, fluid status deficits, malnutrition, and aspiration pneumonia (Easterling and Robbins 2008). Changes in mental capacity, diminished immune system function, and end-of-life issues amplify medical complexity and pose multiple care challenges in the elderly (White et al. 2008).
88.5 Neurologic Causes of Dysphagia Parkinson’s disease (PD), amyotrophic lateral sclerosis (ALS), multiple sclerosis (MS), dementia, and traumatic brain injury (TBI) are neurologic conditions associated with swallowing dysfunction of which stroke is the most common (Easterling and Robbins 2008; Gonzalez-Fernandez and Daniels 2008; Miller et al. 2008; Ward et al. 2007). In fact, neurologic disease accounts for three-quarters of dysphagia during the oropharyngeal phase (Ertekin and Aydogdu 2003). Incidence in stroke varies based on assessment methods, but can be as high as 90% (Gonzalez-Fernandez and Daniels 2008; Miller and Chang 1999). Stroke patients with dysphagia have longer duration of hospitalization and more difficulties with hydration and nutrition (Gordon et al. 1987; Axelsson et al. 1989; Kidd et al. 1995). A threefold increase in pneumonia has been documented in stroke patients with dysphagia (Martino et al. 2005); the risk is even higher in patients with confirmed aspiration (Teasell et al. 1996). Reports indicate that dysphagia in this population results from impairments in muscle control, tongue function, sensation (larynx and pharynx), and cough (Dray et al. 1998). Dysphagia is also an important consideration in TBI and is associated with the severity of brain injury, time span for cognitive recovery, tracheostomy placement, and duration of mechanical
1380
D. Feldman and M. González-Fernández
ventilation (Ward et al. 2007). Swallowing problems in this population are multifaceted and can be a result of impaired swallow response or lingual, pharyngeal, laryngeal, or cricopharyngeal dysfunction (Mayer 2004). Cognitive and executive dysfunctions, aberrant behavior, problems with speech production, and/or comprehension along with motor abnormalities are major constraints to intervention and airway protection (Mayer 2004). In TBI, these findings may be more problematic with regard to swallowing safety than the underlying pathologic swallow (Mayer 2004). As opposed to PD and ALS in which the disease course continues to advance, stroke and head injury-induced swallowing problems usually improve or resolve (Dray et al. 1998; Broadley et al. 2005; White et al. 2008). In stroke, dysphagia resolves in up to 90% of patients (Broadley et al. 2005; White et al. 2008). The rate of swallowing dysfunction in those with PD is variable (18–100%; Miller et al. 2008). Solid textures have been shown to be more challenging for these individuals (Edwards et al. 1992). Disordered swallowing usually occurs as a result of problems with mastication; most evident during the oral phase (Calcagno et al. 2002; Hartelius and Svensson 1994; Merson and Rolnick 1998; Thomas and Wiles 1999). In MS, dysphagia occurs less frequently (33–43%) and often remains unrecognized by the individual until later in the disease course, despite acknowledgment by others (Dray et al. 1998; Gonzalez-Fernandez and Daniels 2008; Pasquinelli and Solaro 2008). For this reason, identification tends to be delayed (Schindler and Kelly 2002). Features include decreased bolus movement through the pharynx and inability to trigger a pharyngeal swallow response (Dray et al. 1998). As well, dysphagia for liquids and solids may be exacerbated by low energy levels, especially in the beginning phases (Dray et al. 1998). ALS is a progressive neurodegenerative disease that destroys both upper and lower motor neurons specifically in the brainstem and spine along with the associated pathways. Dysphagia usually occurs at later stages of the disease but can be present at onset (Higo et al. 2004). Findings include protracted swallow responsiveness to the bolus and increased upper esophageal sphincter activity and pressure (Achem and Devault 2005). Oral phase difficulties, manifests soon after onset and in many cases, leads to increased time for food consumption and inability to tolerate textures that are not softened (Dray et al. 1998). Dementia is a common cause of dysphagia and is likely the result of disease effects superimposed on age-related alterations in sensation, muscle function, and coordination (Easterling and Robbins 2008). Horner et al showed disordered swallowing in approximately 28% of those examined with Alzheimer’s disease based on videofluoroscopic assessment (Horner et al. 1994). Another reason for disordered swallowing in this population is deterioration in olfaction (Easterling and Robbins 2008). Given the interdependence of smell and gustation, food tends to be less palatable and desirable; likewise nutritional status can become compromised (Easterling and Robbins 2008).
88.6 Swallowing in the Elderly Growing old, in general, causes a host of changes – thinning of muscles, weakness, decline in bone quality, slower nerve conduction (loss of myelin) – and can affect virtually all bodily functions (Gleeson 1999). Global features include performing activities at a reduced pace, decrement in body processes, weakness, decreased stamina, and impairments in dexterity (Gleeson 1999). Swallowing is not exempt. In fact, Robbins et al demonstrated that swallowing velocity is reduced as early as 45 years of age and will continue insidiously thereafter so that by 70 years, the changes are more pronounced than those seen at 45 years (Robbins et al. 1992).
88 Dysphagia: Neurological and Behavioral Aspects
1381
Adaptations in deglutition that are a result of the natural aging process are termed presbyphagia (Robbins et al. 1992; Leslie et al. 2005). While dysphagia refers to disordered swallowing, the alterations whether structural, instinctive, compensatory, and/or emotional are not considered to be pathologic (Humbert and Robbins 2008). For this reason, it is becoming increasingly more important to be able to make this distinction in order to manage these individuals appropriately (Humbert and Robbins 2008).
88.7 Nutritional Implications Malnutrition is one of the major sequelae of dysphagia. If not promptly recognized, the clinical outcome may be significantly compromised (Curran 1990). Malnutrition can occur either as an early complication of swallowing dysfunction or late in progressive cases (Pasquinelli and Solaro 2008). Despite the cause or time of onset of dysphagia, elderly individuals are at risk because of social isolation, medical comorbidities, decreased mobility, fatigue, poor cognition, diminished awareness, impaired vision, and inability to cook, clean, and/or self-feed (Gariballa and Sinclair 1998; Leslie et al. 2005). Side effects of medications and mood disorders can also adversely impact nutritional status by retarding appetite (Cole et al. 2000). In addition, foods may be selected based on underlying swallowing impairments. For example, in PD softer foods are preferred (Lorefalt et al. 2006). When the body is unable to maintain adequate nourishment and meet its inherent demands, the consequences can be devastating (Pasquinelli and Solaro 2008). Insufficient intake, malabsorption, and/or abnormal output are the usual culprits (Pasquinelli and Solaro 2008). Problems with olfaction (less palatable food) and memory (not remembering to eat) are major threats to sustaining sufficient intake (Easterling and Robbins 2008). Often, this is compounded by mealtime anxiety, general noncompliance, and an unwillingness to eat, especially in cases of dementia (Biernacki and Barratt 2001). More devastating are impairments in mastication and oral control, which necessitate adjustment in oral intake consistency, albeit not without drawbacks and/or nutritional risk (Easterling and Robbins 2008). In contrast, deglutition can also be worsened by malnutrition; inflicting further regression in an already impaired but safe swallow (Leslie et al. 2005). Nutrition is an important factor not only for preservation of overall health and vigor, but also for convalescence (Gariballa and Sinclair 1998). Managing dysphagia appropriately can facilitate feeding in a situation that can be extremely stressful and complicated for caregivers who feel compelled to keep the patient alive. Consumption can be improved by providing food that is safe (from a swallowing standpoint), pleasurable, and/or self-selected on a frequent, yet, desired basis in addition to already scheduled times (Biernacki and Barratt 2001). Supervision and/or help with feeding may also be of benefit (Pasquinelli and Solaro 2008).
88.8 Management of Swallowing Dysfunction A systematic and individualized approach is important in the management of swallowing dysfunction. In cases of dysphagia associated with neurological dysfunction it is important to consider the nature of the disease and the patient’s wishes (Fig. 88.4). The goal of dysphagia treatment is to ensure safe swallowing while concurrently avoiding complications; most worrisome being aspiration pneumonia. As a result, food and fluid consistencies
1382
D. Feldman and M. González-Fernández
Possible Dysphagia on Clinical Evaluation?
No
Yes
Monitor
Possible
Instrumental examination
No
Unsafe swallowing? Yes Corrected by consistency modification maneuvers or posture?
No
Yes
Stable disease or Improving?
Progressive disease but cognitively intact?
No
Yes
Yes Consider temporary alternative means of alimentation
Oral feeding with modifications
Consider:
Consider:
•
• •
• Swallowing
No
•
Swallowing rehabilitation to maintain function* Alternative means of alimentation No intervention
•
No intervention, or Hand feeding (+/-) consistency modifications Alternative means of alimentation
* Exercise might be contraindicated in certain diseases such as ALS. Reprinted from Physical medicine and Rehabilitation Clinics of North America, 19(4), Gonzalez-Fernandez M, Daniels SK, Dysphagia in Stroke and Neurologic Disease, Pages 881, Copyright (2008), with permission from Elsevier.
Fig. 88.4 Dysphagia evaluation and management of the neurologic disease patient. *Exercise might be contraindicated in certain diseases such as ALS. (From Gonzalez-Fernandez and Daniels 2008). This algorithm describes the decisionmaking process in evaluation, management, and treatment of dysphagia in patients with neurologic disease
may be adjusted (Vivanti et al. 2009). Although consistency modification can be important to prevent passage of contents into the airway, these dietary restrictions can be rather displeasing (Humbert et al. 2009). In fact, this can cause more difficulty at mealtimes as intake, manipulation, and movement of food and fluid becomes more time consuming; and may result in nutritional or hydration problems (Vivanti et al. 2009). It is extremely important that clinicians are able to identify swallowing dysfunction and coordinate appropriate treatment. Swallowing can be assessed clinically, with bedside screening tools such as the 3-oz water swallow test or the Mann assessment of swallowing ability, among others (Carnaby-Mann and Lenius 2008; DePippo et al. 1992; Easterling and Robbins 2008). Instrumental examination using videofluroscopy (VFSS), in which X-ray beams are used to detect the course of specialized, contrastlabeled food products during swallowing and/or fiberoptic endoscopy is important to determine the underlying cause of dysphagia (Schindler and Kelly 2002). Although fiberoptic endoscopic evaluation of swallowing (FEES) is superior for direct evaluation of the pharynx and larynx, it is invasive and does not allow for visualization of the oral and esophageal phases. VFSS remains the most used method (Schindler and Kelly 2002). The mainstay of dysphagia treatment includes compensatory and rehabilitative techniques (Table 88.2). Compensatory techniques (chin-up, chin tuck, head turn, head tilt, and reclining
88 Dysphagia: Neurological and Behavioral Aspects
1383
Table 88.2 Common interventions for dysphagia management Feeding modifications Postural maneuvers Rehabilitationa Thickened liquids Chin tuck Oral exercises Nectar thick liquids Head tilt Range of motion Honey thick liquids Head rotation Strength exercises Neck extension Lingual exercises Processed solids Reclining position Masako maneuver Pureed Shaker exercises Mechanical soft Effortful swallow Glottal attacks Volume control Pushing and pulling exercises Teaspoon feeding EMG biofeedback Special utensils a Lingual exercises – Aim is to augment multidirectional tongue movements, sucking and swallowing in order to improve oral control. Masako maneuver – Tongue-holding technique designed to facilitate tongue contact with the pharyngeal wall. Shaker exercises – Strengthening exercise for muscles located below the chin to facilitate upper esophageal sphincter opening and prevent pharyngeal food retention. Effortful swallow – A strategy in which a forceful swallow is encouraged increasing the oral/pharyngeal pressures to clear the bolus, from the vallecular space. Glottal attacks – Specialized adduction exercises in which the vocal folds are powerfully drawn toward midline prior to phonation increasing laryngeal muscular tension to enhance voice control. These can improve vocal cord adduction and closure during swallowing. EMG biofeedback – Adjunct therapy that allows visualization of muscle activity, in this case, visualization of submandibular or anterior neck muscles. It can allow greater control of those muscles and facilitate rehabilitation efforts
positioning, texture modification) serve to optimize bolus movement into the pharynx by altering head and/or chin positioning or improving bolus control (Logemann 2008). Rehabilitative efforts may include exercises to increase neck mobility and strengthen oral, lingual, pharyngeal, and laryngeal muscles and neuromuscular electrical stimulation (Logemann 2008). Postural maneuvers and texture modifications should be trialed during VFSS, FEES, and therapy sessions (Schindler and Kelly 2002). Other therapeutic strategies include sensory training, which targets gustation, and temperature accommodation (Logemann 2008). When eating is not an option, enteral feeding can be considered via percutaneous endoscopic gastrostomy tube (PEG) or, for short-term purposes, with use of oral or nasogastric feeding tubes (Easterling and Robbins 2008). The underlying disease-causing swallowing dysfunction is critical in considering nonoral feeding. In cases with dementia, feeding through PEG tubes has not been associated with improved survival, functional status, patient comfort, or reduced complications (Finucane et al. 1999). The decision to use non-oral feeding requires the input of the patient (when possible), family, treating physicians, and speech language pathologists to determine if the benefits of nonoral feeding outweigh possible complications.
88.9 Application to Other Areas of Health and Disease Eating is essential to the operation of virtually every body system; dysphagia can have a significant impact on an individual’s ability to eat and meet its nutritional needs. As previously discussed, nutritional status can deteriorate with senescence; dysphagia can have an additional detrimental effect on nutrition. Malnutrition can have a major impact on overall functioning and disease susceptibility and can
1384
D. Feldman and M. González-Fernández
Table 88.3 Key points of dysphagia 1. Neurologic disease is one of the most common causes of dysphagia. 2. Dysphagia associated with neurologic disease can affect any of the stages of swallowing. 3. Malnutrition is one of the most serious sequela of dysphagia. 4. Physical, cognitive, and behavioral changes can impact food preferences and contribute to swallowing difficulties in the elderly. 5. Age-related changes can increase the probability of developing dysphagia after neurologic events. 6. Early comprehensive dysphagia evaluation and treatment is essential for best outcomes. This table lists the key facts of dysphagia including cause, swallowing stages, sequela, and age-related changes
result in growth and gonadotropic hormone insufficiency, decreased insulin uptake, hypothyroidism, immune system deficits, impaired gastrointestinal absorption, osteoporosis, psychiatric disorders, altered sleep hygiene, age-related and/or nutritional anemia, muscle mass loss, and vitamin deficiencies (Lytras and Tolis 2007; Carmel 2008) Nutritional deficiencies can impair sensorimotor and cognitive functioning (Biernacki and Barratt 2001; Espeland and Henderson 2006). Endocrinologic and immune system alterations interfere with kidney output, nutrient uptake/processing, and generalized performance (Lytras and Tolis 2007). Physical impairments amplify associated medical risks leading to decubitus ulcers and pneumonia (Almirall et al. 2000; Gariballa and Sinclair 1998; Langmore et al. 2002). Thus, it is extremely important that clinicians understand the ramifications of malnutrition, its association with dysphagia, and other medical conditions/impairments in the aging population.
88.10 Final Points Dysphagia is common, associated with neurologic disease and senescence. Depending on the primary pathology, the course is rather variable. However, when present, the complications can be extremely devastating substantiating the need for aggressive diagnosis, management, and treatment. With a better knowledge and awareness of the swallowing process including its relation to the existing disease state, complication risk, and importance of rehabilitation services, care can be coordinated effectively to improve overall health, wellbeing, and quality of life. Key points of this chapter are described in Table 88.3. Summary Points • Dysphagia is a common problem, particularly in older individuals and is primarily associated with dementia and neurologic disease. • Depending on the primary pathology, the course of dysphagia is variable. • Malnutrition, dehydration, weight loss, and aspiration pneumonia are frequent, yet serious complications associated with disordered swallowing. • Older individuals are at greater risk for malnutrition due to medical comorbidities, decreased functional mobility, limited self-care capacity, and social isolation; which is exacerbated by underlying dysphagia. • Proper diagnosis, management, and treatment are essential to the care of patients with disordered swallowing. • Dysphagia therapy focuses on compensatory techniques and exercises for neck, facial, lingual, pharyngeal, and laryngeal muscles. • Enteral feeding should be considered when safe oral feeding is not possible.
88 Dysphagia: Neurological and Behavioral Aspects
1385
Key Terms Dysphagia: Disordered swallowing. Deglutition: The act of swallowing. Presbyphagia: Age-related, nonpathologic changes in swallowing function. Videofluroscopic swallow study (VFSS): X-ray beams are used to detect the course of specialized, contrast-labeled food products during swallowing. Fiberoptic endoscopic evaluation of swallowing (FEES): Passage of a small, flexible scope attached to a camera, through the nose and into the pharynx and larynx in order to visualize swallowing during the oral and esophageal phases. Traumatic brain injury: Intracranial injury due to traumatic forces causing injury to the head.
References Achem SR, Devault KR. Dysphagia in aging. J Clin Gastroenterol. 2005;39:357–71. Almirall J, Bolibar I, Vidal J, Sauca G, Coll P, Niklasson B, Bartolome M, Balanzo X. Epidemiology of communityacquired pneumonia in adults: a population-based study. Eur Respir J. 2000;15:757–63. Axelsson K, Asplund K, Norberg A, Eriksson S. Eating problems and nutritional status during hospital stay of patients with severe stroke. J Am Diet Assoc. 1989;89:1092–6. Barczi SR, Sullivan PA, Robbins J. How should dysphagia care of older adults differ? Establishing optimal practice patterns. 2000;Semin Speech Lang. 21:347–61. Biernacki C, Barratt J. Improving the nutritional status of people with dementia. Br J Nurs. 2001;10:1104–14. Broadley S, Cheek A, Salonikis S, Whitham E, Chong V, Cardone D, Alexander B, Taylor J, Thompson P. Predicting prolonged dysphagia in acute stroke: the Royal Adelaide Prognostic Index for Dysphagic Stroke (RAPIDS). Dysphagia. 2005;20:303–10. Calcagno P, Ruoppolo G, Grasso MG, De Vincentiis M, Paolucci S. Dysphagia in multiple sclerosis - prevalence and prognostic factors. Acta Neurol Scand. 2002;105:40–3. Carmel R. Nutritional anemias and the elderly. Semin Hematol. 2008;45:225–34. Carnaby-Mann G, Lenius K. The bedside examination in dysphagia. Phys Med Rehabil Clin N Am. 2008;19:747–68, viii. Cole C, Bigando K, DeSutter S. Is altered nutritional status the root cause of your clients’ negative outcomes? J Nurs Care Qual. 2000;14:41–56. Curran J. Overview of geriatric nutrition. Dysphagia. 1990;5:72–6. DePippo KL, Holas MA, Reding MJ. Validation of the 3-oz water swallow test for aspiration following stroke. Arch Neurol. 1992;49:1259–61. Dray TG, Hillel AD, Miller RM. Dysphagia caused by neurologic deficits. Otolaryngol Clin North Am. 1998;31:507–24. Easterling CS, Robbins E. Dementia and dysphagia. Geriatr Nurs. 2008;29:275–85. Edwards LL, Quigley EM, Pfeiffer RF. Gastrointestinal dysfunction in Parkinson’s disease: frequency and pathophysiology. Neurology. 1992;42:726–32. Ertekin C, Aydogdu I. Neurophysiology of swallowing. Clin Neurophysiol. 2003;114:2226–44. Espeland MA, Henderson VW. Preventing cognitive decline in usual aging. Arch Intern Med. 2006;166:2433–4. Finucane TE, Christmas C, Travis K. Tube feeding in patients with advanced dementia: a review of the evidence. JAMA. 1999;282:1365–70. Gariballa SE, Sinclair AJ. Assessment and treatment of nutritional status in stroke patients. Postgrad Med J. 1998;74:395–9. Gilmore SA, Robinson G, Posthauer ME, Raymond J. Clinical indicators associated with unintentional weight loss and pressure ulcers in elderly residents of nursing facilities. J Am Diet Assoc. 1995;95:984–92. Gleeson DC. Oropharyngeal swallowing and aging: a review. J Commun Disord. 1999;32:373–95; quiz 395–6. Gonzalez-Fernandez M, Daniels SK. Dysphagia in stroke and neurologic disease. Phys Med Rehabil Clin N Am. 2008;19:867–88. Gordon C, Hewer RL, Wade DT. Dysphagia in acute stroke. Br Med J (Clin Res Ed). 1987;295:411–4. Groher ME, McKaig TN. Dysphagia and dietary levels in skilled nursing facilities. J Am Geriatr Soc. 1995;43:528–32. Hartelius L, Svensson P. Speech and swallowing symptoms associated with Parkinson’s disease and multiple sclerosis: a survey. Folia Phoniatr Logop. 1994;46:9–17.
1386
D. Feldman and M. González-Fernández
Higo R, Tayama N, Nito T. Longitudinal analysis of progression of dysphagia in amyotrophic lateral sclerosis. Auris Nasus Larynx. 2004;31:247–54. Horner J, Alberts MJ, Dawson DV, Cook GM. Swallowing in Alzheimer’s disease. Alzheimer Dis Assoc Disord. 1994;8:177–89. Humbert IA, Fitzgerald ME, McLaren DG, Johnson S, Porcaro E, Kosmatka K, Hind J, Robbins J. Neurophysiology of swallowing: effects of age and bolus type. Neuroimage. 2009;44:982–91. Humbert IA, Robbins J. Dysphagia in the elderly. Phys Med Rehabil Clin N Am. 2008;19:853–66, ix-x. Kidd D, Lawson J, Nesbitt R, MacMahon J. The natural history and clinical consequences of aspiration in acute stroke. QJM. 1995;88:409–13. Langmore SE, Skarupski KA, Park PS, Fries BE. Predictors of aspiration pneumonia in nursing home residents. Dysphagia. 2002;17:298–307. Leslie P, Drinnan MJ, Ford GA, Wilson JA. Swallow respiratory patterns and aging: presbyphagia or dysphagia? J Gerontol A Biol Sci Med Sci. 2005;60:391–5. Logemann JA. Treatment of oral and pharyngeal dysphagia. Phys Med Rehabil Clin N Am. 2008;19:803–16, ix. Lorefalt B, Granerus AK, Unosson M. Avoidance of solid food in weight losing older patients with Parkinson’s disease. J Clin Nurs. 2006;15:1404–12. Lytras A, Tolis G. Assessment of endocrine and nutritional status in age-related catabolic states of muscle and bone. Curr Opin Clin Nutr Metab Care. 2007;10:604–10. Martino R, Foley N, Bhogal S, Diamant N, Speechley M, Teasell R. Dysphagia after stroke: incidence, diagnosis, and pulmonary complications. Stroke. 2005;36:2756–63. Matsuo K, Palmer JB. Anatomy and physiology of feeding and swallowing: normal and abnormal. Phys Med Rehabil Clin N Am. 2008;19:691–707, vii. Mayer V. The challenges of managing dysphagia in brain-injured patients. Br J Community Nurs. 2004;9:67–73. Merson RM, Rolnick MI. Speech-language pathology and dysphagia in multiple sclerosis. Phys Med Rehabil Clin N Am. 1998;9:631–41. Miller N, Allcock LM, Hildreth T, Jones D, Noble E, Burn D. Swallowing problems in Parkinson’s disease: Frequency and clinical correlates. J Neurol Neurosurg Psychiatry. 2008. Electronic-publication doi: 10.1136/ jnnp.2008.157701. Miller RM, Chang MW. Advances in the management of dysphagia caused by stroke. Phys Med Rehabil Clin N Am. 1999;10:925–41, x. Pasquinelli S, Solaro C. Nutritional assessment and malnutrition in multiple sclerosis. Neurol Sci. 2008;29 Suppl 4:S367–9. Ren J, Shaker R, Zamir Z, Dodds WJ, Hogan WJ, Hoffmann RG. Effect of age and bolus variables on the coordination of the glottis and upper esophageal sphincter during swallowing. Am J Gastroenterol. 1993;88:665–9. Robbins J, Hamilton JW, Lof GL, Kempster GB. Oropharyngeal swallowing in normal adults of different ages. Gastroenterology. 1992;103:823–9. Schindler JS, Kelly JH. Swallowing disorders in the elderly. Laryngoscope. 2002;112:589–602. Teasell RW, McRae M, Marchuk Y, Finestone HM. Pneumonia associated with aspiration following stroke. Arch Phys Med Rehabil. 1996;77:707–9. Thomas FJ, Wiles CM. Dysphagia and nutritional status in multiple sclerosis. J Neurol. 1999;246:677–82. Vivanti AP, Campbell KL, Suter MS, Hannan-Jones MT, Hulcombe JA. Contribution of thickened drinks, food and enteral and parenteral fluids to fluid intake in hospitalised patients with dysphagia. J Hum Nutr Diet. 2009;22:148–55. Ward EC, Green K, Morton AL. Patterns and predictors of swallowing resolution following adult traumatic brain injury. J Head Trauma Rehabil. 2007;22:184–91. White GN, O’Rourke F, Ong BS, Cordato DJ, Chan DK. Dysphagia: causes, assessment, treatment, and management. Geriatrics. 2008;63:15–20.
Chapter 89
Neuropsychological Aspects of Eating Disorders – A Focus on Diagnostic Criteria Jennie C. Ahrén
Abbreviations AN BN BED CSF ED EDNOS fMRI
Anorexia nervosa Bulimia nervosa Binge eating disorder Cerebrospinal fluid Eating disorders Eating disorders not otherwise specified Functional magnetic resonance imaging
89.1 Introduction Eating disorders (ED) are severe psychiatric disturbances with psychosomatic complications, mainly affecting young women (Treasure 2008; Palmer 2008). The long-term effects of starvation and deviations in eating behaviors are devastating, from a psychological as well as a physiological perspective. ED is stated as one of the leading causes of disease burden in terms of years lost through disability or death in industrialized countries and recent work shows that they are increasing (Mathers et al. 2000; Hay et al. 2008). The mortality rate is higher than any other psychiatric disorder (Kaye et al. 2009). The interest in neurological correlates of behavior has increased during the last decades. Modern neuropsychology assumes that intellectual capacities and cognitive processes are complex functions that depend on several interrelated systems in the brain. But even though our knowledge on neural networks and behavior has developed drastically, the classification in anatomical areas according to Brodmann in the early twentieth century is still frequently used and continually updated (Kandel 2000). The connection between brain and behavior as a key aspect of anorexia nervosa (AN) was introduced several years ago (Braun and Chouinard 1992) but new techniques have made it possible to evaluate and observe behaviors in ways that were not anticipated earlier. Functional magnetic resonance imaging (fMRI) is one of the latest methods in brain imaging, revealing patterns of neural activity and resulting in new insights on how the brain functions. fMRI is a specialized MRI scan
J.C. Ahrén Center for Health Equity Studies, CHESS, Karolinska Institutet/Stockholm University, Sveavagen 160, 106 91, Stockholm, Sweden e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_89, © Springer Science+Business Media, LLC 2011
1387
1388
J.C. Ahrén
where the neural activity in the brain or in the spinal cord is measured by the blood flow. It has come to dominate in the field of brain imaging, mainly due to the low radiation exposure. The relationship between behavioral changes in ED on the one hand and alterations of the brain on the other can be divided into two categories as suggested by a recent review on AN (Kaye et al. 2009). The first one is based on the assumption that there is a genetic component involved; that premorbid traits make an individual more vulnerable to develop AN. Secondly, it is possible that malnutrition causes alterations in the brain and thereby reinforces and helps to maintain the behavior. The overarching focus of this chapter was to describe subtypes of ED from a neuropsychological perspective. In order to establish how different functions are connected to behavior, definitions of ED will be presented rather detailed. The first section is a short overview of different subtypes of ED with a short description of the etiology of these conditions and comorbid symptoms. This is followed by a brief overview on neuropsychological functions in relation to ED. Furthermore, different parts of the brain involved in eating pathology are briefly described.
89.2 Etiology of ED The etiology of ED remains unclear, multifactorial models are best applied to increase our understanding on how different factors interact in the development of disorders such as AN and bulimia nervosa (BN). An ED does not only involve deviations in eating behavior, but typically include somatic complications due to starvation and decreased psychosocial functioning. ED often occur during adolescence, a developmentally critical period. The functions of the brain are important clues in understanding behavioral changes in puberty, the relationship between genetic vulnerability, endocrine changes, and environment are overlapping (McAnarney 2008). The rapid change in body composition during this period requires flexibility in regulating weight and hunger; this may be associated with an increased vulnerability in the appetite regulation systems, including hypothalamic structures (Connan et al. 2003). The interaction between genetic, environmental, and psychological factors is complex and demands integrated perspectives (see Fig. 89.1). Certain personality traits have been described as typical for different types of ED. AN is characte rized by high constraint, constriction of affect, perfectionism, conformity, and obsessive-compulsive behavior (Bulik et al. 2003; Kaye 2008). One of the central traits in patients with ED is the need for control. Controlling food intake is sometimes described as a way of controlling a life situation that is overwhelming (Fairburn et al. 1999). For BN impulsivity, sensation seeking and low self-esteem are often stated as predisposing traits (Fairburn et al. 1999; Jacobi et al. 2004). The comorbidity between ED and personality disorders is high. A recent methodological review states that obsessivecompulsive personality disorder shares many of the same features as AN (Lilenfeld et al. 2006). Anxiety disorders are frequent in ED patients and previous research indicates that those might be interpreted as predisposing traits (Swinbourne and Touyz 2007). Individual differences are likely to influence cognitive style and different aspects of problemsolving strategies. Certain traits that have been described as central for ED, such as obsessive or compulsive behavior, may actually be a result of neuropsychological impairments following the disorder. Starvation leads to a range of psychosocial consequences in the same way as physiological functions are affected. The well-known work by Keys et al. (1950) studied the effects of starvation in a group of healthy men during the mid-forties in Minnesota, USA. Thirty-sex men, described as intelligent, psychologically sound, and physically healthy had their food intake strictly limited during 6 months, resulting in significant weight loss. Interestingly, most of the men showed behavioral patterns similar to those described in patients with ED. Symptoms such as increased apathy, depression,
89 Neuropsychological Aspects of Eating Disorders – A Focus on Diagnostic Criteria
Childhood
Adolescence
1389
Traits Negative emotion Perfectionism Drive for thinness Increased interoceptive awareness Obsessive-Compulsive
Puberty Brain development Hormones Stress Dieting
Adulthood
Denial, rigidity anxiety, depression, obsessionality
Weight loss
Neurobiological changes
Chronic illness (30-50%)
Recovery (50-70%)
Fig. 89.1 The time course and phenomenology of anorexia nervosa (From Kaye et al. 2009)
and increased neuroticism were reported by most of the men. Furthermore, they showed an increased interest in food and were preoccupied with thoughts of food and became more socially withdrawn. They reported impaired concentration, comprehension, and bad judgment. Six months of semistarvation also brought on other physical and cognitive problems, such as decreased motor control, impaired sleep, and visual disturbances. The effects of starvation or binge-eating are obviously devastating from a physiological point of view, but neuropsychological aspects provide important clues in the development and maintenance of ED and need to be further explored.
89.3 Definitions of ED According to DSM-IV (American Psychiatric Association 1994) ED are divided into the diagnoses AN, BN, eating disorders not otherwise specified (EDNOS), and binge-eating disorder (BED). The life-time prevalence in women for full and partial AN range from 0.9% to 4.3% (Hudson et al. 2007; Wade et al. 2006). Prevalence for full and partial BN range from 4% to 7% in women (Favaro et al. 2003). For BED the lifetime prevalence is 3.5%. The corresponding figures for men are 0.9% for AN, 0.5% for BN, and 2.0% for BED (Hudson et al. 2007). Most of the studies cited in this text are exclusively based on women, and the main perspective will therefore be on women.
1390
J.C. Ahrén
It is commonly stated that AN and BN are modern phenomena, but a glance in historical descriptions shows that these conditions are not new. This particularly applies to AN, or “nervous consumption,” as described by the British physician Richard Morton as early as in 1689. In 1868 Sir William Gull, physician to Queen Victoria, presented his three case studies under the title “Anorexia Nervosa” and the diagnose was given its name (Pearce 2006). The essential feature of AN is low body weight and self-starvation. Diagnostic criteria according to the American Psychiatric Association (1994) are as follows (Table 89.1). BN mainly refers to recurrent binge-eating and compensatory actions such as vomiting or using laxatives. The diagnosis was first applied in 1979 and was included in the psychiatric classification in 1980. Diagnostic criteria for BN according to DSM-IV (American Psychiatric Association 1994) are listed in Table 89.2. The EDNOS category is for disorders of eating that do not meet the criteria for any specific ED. This does not imply that they are less severe; several features are the same as those stated in the diagnostic criteria for the other subtypes. Examples include the following (Table 89.3).
Table 89.1 Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria (APA 1994) for anorexia nervosa (AN) (A) Refusal to maintain body weight at or above a minimally normal weight for age and height (e.g., weight loss leading to maintenance of body weight less than 85% of that expected; or failure to make expected weight gain during period of growth, leading to body weight less than 85% of that expected) (B) Intense fear of gaining weight or becoming fat, even though underweight (C) Disturbance in the way in which one’s body weight or shape is experienced, undue influence of body weight or shape on self-evaluation, or denial of the seriousness of the current low body weight (D) In postmenarchal females, amenorrhea – the absence of at least three consecutive cycles (a woman is considered to have amenorrhea if her periods occur only following hormone (e.g. estrogen) administration) Types: • Restricting type: During the current episode of anorexia nervosa, the person has not regularly engaged in binge-eating or purging behavior (i.e., self-induced vomiting or the misuse of laxatives, diuretics, or enemas) • Binge-eating/purging type: During the current episode of anorexia nervosa, the person has regularly engaged in binge-eating or purging behavior (i.e., self-induced vomiting or the misuse of laxatives, diuretics, or enemas)
Table 89.2 Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria (APA 1994) for bulimia nervosa (BN) (A) Recurrent episodes of binge-eating. An episode of binge-eating is characterized by both of the following: • Eating, in a discrete period of time (e.g., within any 2-h period), an amount of food that is definitely larger than most people would eat during a similar period of time and under similar circumstances • A sense of lack of control over eating during the episode (e.g., a feeling that one cannot stop eating or control what or how much one is eating) (B) Recurrent inappropriate compensatory behavior in order to prevent weight gain, such as self-induced vomiting; misuse of laxatives, diuretics, enemas, or other medications; fasting; or excessive exercise (C) The binge-eating and inappropriate compensatory behaviors both occur, on average, at least twice a week for 3 months (D) Self-evaluation is unduly influenced by body shape and weight (E) The disturbance does not occur exclusively during episodes of anorexia nervosa Types: • Purging type: during the current episode of bulimia nervosa, the person has regularly engaged in self-induced vomiting or the misuse of laxatives, diuretics, or enemas • Nonpurging type: during the current episode of bulimia nervosa, the person has used other inappropriate compensatory behaviors, such as fasting or excessive exercise, but has not regularly engaged in self-induced vomiting or the misuse of laxatives, diuretics, or enemas
89 Neuropsychological Aspects of Eating Disorders – A Focus on Diagnostic Criteria
1391
Table 89.3 Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria for eating disorder not otherwise specified (EDNOS) (A) All diagnostic criteria for anorexia nervosa are met, except the menstrual cycle is normal (B) All diagnostic criteria for anorexia nervosa are met, except weight is normal for height and age even after considerable weight loss (C) All diagnostic criteria for bulimia nervosa are met, but the frequency of binges is less than twice weekly and for a duration of less than 3 months (D) There are recurring efforts to compensate (such as self-induced vomiting) for eating only small amounts of food, but body weight is normal for height and age (E) Regularly chewing and spitting out large quantities of food without swallowing (F) Binge-eating disorder – regular episodes of binge eating, but with no recurring efforts to compensate, such as purging or excessive exercise
Table 89.4 Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria for binge eating disorder (BED) (A) Recurrent episodes of binge-eating: an episode of binge-eating is characterized by both of the following: (1) Eating, in a discrete period of time (e.g., within any 2-h period), an amount of food that is definitely larger than most people would eat in a similar period of time under similar circumstances (2) The sense of lack of control over eating during the episode (e.g., a feeling that one cannot stop eating or control what or how much one is eating) (B) Binge-eating episodes are associated with three (or more) of the following: (1) Eating much more rapidly than normal (2) Eating until feeling uncomfortably full (3) Eating large amounts of food when not feeling physically hungry (4) Eating alone because of being embarrassed by how much one is eating (5) Feeling disgusted with oneself, depressed, or very guilty after overeating (C) Marked distress regarding binge-eating is present (D) The binge-eating occurs, on average, at least 2 days a week for 6 months (E) The binge-eating is not associated with the regular use of inappropriate compensatory behavior (e.g., purging, fasting, excessive exercise, etc.) and does not occur exclusively during the course of anorexia nervosa or bulimia nervosa
BED was first described by Stunkard in 1959 (Mathes et al. 2009). Binge-eating refers to overeating to the extent that you have none or little control over your food intake. This is followed by feelings of disgust and guilt. People suffering from BED often eat alone or late at night to hide their behavior. BED is distinguished from BN by the absence of purging or compensatory after bingeeating. BED is more common in obese or overweight individuals. The criteria for BED are included for clarification of eating pathology, but will not be developed further in the text (Table 89.4).
89.4 Neuropsychological Functioning in ED Earlier research proposes that neuropsychological deficits in various cognitive domains may serve as an underlying factor in the development of ED (Lena et al. 2004). A large study of women with AN showed that half of the patients studied had mild cognitive impairments and that more than one-third failed two or more neuropsychological tasks (Bayless et al. 2002). However, it is not clear whether malfunctions in cognitive processes are predisposing or stand as a consequence of the ED. In a review of cognitive functioning in patients with ED, Duchesne and co-workers (2004) reported that AN was associated with difficulties in executive functions, visuospatial abilities, and psychomotor
1392
J.C. Ahrén
speed, BN was more connected to impairments in selective attention and executive functions. They conclude that ED in general is associated with neuropsychological dysfunctions, although there is no consensus on exactly what functions are affected. Executive functioning was studied in AN showing that performance worsened at the presence of obsessive behavior, depression, and starvation (Wilsdon and Wade 2006). Neuropsychological assessment in a group of ED inpatients as compared with healthy controls showed that the performance level on tasks measuring attention demands and problem-solving abilities was impaired. Re-tests after recovery indicated that AN and BN were characterized by similar neuropsychological deficits, and that these malfunctions seemed to be reversible (Lauer et al. 1999). Gillbergs’ research group did a follow-up study of AN patients 10 years after onset of the ED. They found no major neuropsychological deficits in the group (Gillberg et al. 2007). It has been suggested that a weak central coherence is predisposing for AN. The focus on details and difficulties in assessing the “whole” are related to several core features of AN such as rigidity and impairments in cognitive flexibility (Gillberg et al. 1996; Tchanturia et al. 2004). Other studies have showed that people with ED have difficulties in global processing (Lopez et al. 2008a). The inconsistency in these findings points at the need for identification of neuropsychological deficits, their relation to clinical symptoms, and potential relationships with personality and biological measures (Tchanturia et al. 2005). It also addresses issues of state versus trait, and shows the need for future studies assessing whether neurological abnormalities are predisposing in ED or if these dysfunctions are secondary to the condition (Kaye et al. 2009; Schmidt 2003).
89.5 Neurological Correlates of Eating Pathology Recent neuro-imaging studies in AN have shown specific and persistent neuropsychological deficits associated with neurological abnormalities (Agrawal and Lask 2009). Earlier studies show that individuals with ED have reduced brain mass. Reductions in grey and white matter and increased levels of cerebrospinal fluid (CSF) has been found in both AN and BN subjects (Swayze et al. 2003; Giordano et al. 2001). A study on individuals who had recovered from their ED showed that both CSF and total volume of white and grey matter was restored (Wagner et al. 2006). This indicates that brain tissue abnormalities might be reversible after long-term recovery. It has recently been suggested that a disconnection between structures in the brain leads to features typical for AN (Agrawal and Lask 2009). Nunn et al. (2008) suggest certain cortical structures in connection with AN; frontal, somatosensory and parietal cortices and sub-cortical amygdala, hippocampus, hypothalamus, and the striatum. Another study demonstrated a bilateral reduction in hippocampal volume in women with AN, although these were not associated with any impairments in cognitive functions (Connan et al. 2006). A medial prefrontal response to symptom-provoking stimuli was identified as a common feature in both AN and BN. Neural correlates thus indicate that there might be transdiagnostic features for ED (Uher et al. 2004). Different parts of the brain have been put in connection with food intake. Two regions are located in the hypothalamus; one in the ventromedial part, and the other one in the lateral region. Further, the part of the cortex referred to as the insula, located in between the temporal lobe and the frontal lobe, is involved in regulation of appetite (Koh et al. 2003). Dieting is often stated as a major triggering factor in ED, even though the underlying reasons for this behavior can be multiple. The responses to taste stimuli in the insula and adjacent structures in the prefrontal cortex were stronger in the phase of fasting, especially in women as showed by Uher et al. (2006). Neurobiology of binge-eating has been compared to substance abuse behavior (Mathes et al. 2009). Dopaminergic neurons within the
89 Neuropsychological Aspects of Eating Disorders – A Focus on Diagnostic Criteria
1393
nucleus accumbens, often described as an important part in the “reward system” in the brain, are stimulated by food intake. Repeated activation of this center in an attempt to relieve stress has been linked to binge-eating (Koob and Le Moal 2008). Disgust and fear in response to images of food has been found in ED patients (Ellison et al. 1998). A study of changes in cerebral blood flow in BN patients showed that the right temporal lobe was activated after being exposed to images of the own body. This response was interpreted by the authors as caused by threat-related events (Beato-Fernandez et al. 2009). Similar results were found in a study on patients with AN. Body distortion was connected to activation of the amygdala, often stated as the brain’s center for fear (Seeger et al. 2002). Uher et al. (2005) showed that patients with ED had less activity in neural networks connected to processing of female body shapes. The patients consistently rated body shapes in all categories as more aversive than did controls. This suggests an inability to evaluate the body in a realistic way. The authors conclude that illness duration most likely has an impact on body image disturbances.
89.6 Applications to Other Areas of Health and Disease New insights into how the brain functions and develops have led to a better understanding on neurological aspects of eating pathology. This, in turn, renders new knowledge on how these conditions affect individuals, leading to better diagnoses and better opportunities for early treatment. Neuropsychological feedback has successfully been integrated in treatment suggesting that information-processing styles can be used for patients to develop a more balanced strategy in the relationship toward food and weight (Lopez et al. 2008b). For AN, impairments in cognitive flexibility have implications for rehabilitation, for example a focus on rigidity that goes beyond eating behaviors (Tchanturia et al. 2004). Older studies, pointing at deficiencies in visuospatial ability and difficulties in assessing the “whole” (Gillberg et al. 1996) are confirmed by recent work on information processing in AN strengthening the assumption that these patients often focus on details, indicating a weakness in central coherence (Southgate et al. 2008). Cognitive remediation therapy (CRT) which deals with processes rather than content of thoughts has been incorporated in the treatment of patients with AN (Agrawal and Lask 2009). Several studies on neuropsychological functions and neurological correlates of ED are currently ongoing. Insights into underlying mechanisms of disordered eating will evolve in the near future; this in turn will result in new perspectives on treatment and prevention of these severe disorders. Summary Points • ED are severe psychiatric disturbances resulting in impaired psychosocial functioning and physiological complications. • AN, BN, and BED are the main diagnoses. • The etiology of ED remains unclear; psychological, biological, and social perspectives need to be considered when assessing risk factors and clinical course. • Neuropsychological functioning is impaired in patients with ED; chronic states of the disorder further complicate cognitive malfunctions. • There are several inconsistencies in differences in cognitive ability between subtypes of ED and future studies need to assess how different functions in the brain are associated with deviations in eating behavior. • The limbic system, insular cortex, and hypothalamus are the neurological structures that seem to be of most relevance in eating pathology.
1394
J.C. Ahrén
Definitions White matter (substantia alba): Composed of myelinated axons; nerve fibers. Grey matter (substantia grisea): Tissue in the brain and the spinal cord that contains cell bodies. Hypothalamus: Located above the brain stem under the thalamus and controls the autonomic system, endocrine, and motor functions. It also regulates food and water intake, body temperature, the sleep and wake cycle, and emotions. Involved in controlling behaviors such as hunger, thirst, sleep, and sexual response. The insular cortex: Often referred to as the insula (the island in Latin), located between the temporal lobe and the parietal lobe. Involved in emotion, regulation of body homeostasis perception, motor control, and cognitive functioning. Limbic system: The limbic system is often described as the center of emotions, learning, and memory. Included in this system are the cingulate gyri, hypothalamus, amygdala (emotional reactions), and hippocampus (memory).
References Agrawal S, Lask B. Pediatric Health. 2009;3:209–121. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4th edition (DSM-IV). Washington: American Psychiatric Association; 1994. Bayless JD, Kanz JE, Moser DJ, McDowell BD, Bowers WA, Andersen AD, Paulsen JS. Ann Clin Psychiatry. 2002;14:203–7. Beato-Fernandez L, Rodriguez-Cano T, Garcia-Vilches I, Garcia-Vicente A, Poblete-Garcia V, Castrejon AS, Toro J. Psychiatry Res. 2009;171:129–37. Braun CMJ, Chouinard M-J. Neuropsychol Rev. 1992;3:171–212. Bulik CM, Tozzi F, Anderson C, Mazzeo SE, Aggen S, Sullivan PF. Am J Psychiatry. 2003;160:366–8. Connan F, Lightman SL, Landau S, Wheeler M, Treasure J. Physiol Behav. 2003;79:13–24. Connan F, Murphy F, Connor SE, Rich P, Murphy T, Bara-Carill N, Landau S, Krljes S, Ng V, Williams S, Morris RG, Campbell IC, Treasure J. Psychiatry Res. 2006;146:117–25. Duchesne M, Mattos P, Fontenelle LF, Veiga H, Rizo L, Appolinario JC. Rev Bras Psiquiatr. 2004;26:107–17. Ellison Z, Foong J, Howard R, Bullmore E, Williams S, Treasure J. Lancet. 1998;352:1192. Fairburn CG, Cooper Z, Doll HA, Welch SL. Arch Gen Psychiatry. 1999;56:468–74. Favaro A, Ferrara S, Santonastaso P. Psychosom Med. 2003;65:701–8. Gillberg IC, Gillberg C, Råstam M, Johansson, M. Compr Psychiatry. 1996;37:23–30. Gillberg IC, Råstam M, Wentz E, Gillberg C. J Clin Exp Neuropsychol. 2007;29:170–8. Giordano GD, Renzetti P, Parodi RC, Foppiani L, Zandrino F, Giordano G, Sardanelli F. J Endocrinol Invest. 2001;24:510–4. Hay PJ, Mond J, Buttner P, Darby A. PLoS ONE 2008;3 (2):e1541. Hudson JI, Hiripi E, Pope HG Jr, Kessler RC. Biol Psychiatry. 2007;61:348–58. Jacobi C, Hayward C, de Zwaan M, Kraemer HC, Agras WS. Psychol Bull. 2004;130:19–65. Kandel ER. The brain and behavior. In: Kandel ER, Schwartz JH, Jessell TM editors. Principles of neural science, 4th edition. New York: McGraw Hill; 2000. p. 5–18. Kaye W. Physiol Behav. 2008;94:121–35. Kaye WH, Fudge JL, Paulus M. Nature Rev. 2009;10:573–84. Keys A, Brozek J, Henschel A, Mickelsen O, Taylor HL. The biology of human starvation. Minneapolis: The University of Minnesota Press; 1950. Koh M, Wilkins E, Bernsteins I. Behav Neurosci. 2003;117:1416–22. Koob GF, Le Moal M. Annu Rev Psychol. 2008;59:29–53. Lauer CJ, Gorzewski B, Gerlinghoff M, Backmund H, Zihl J. J Psychiatr Res. 1999;33:129–38. Lena SM, Fiocco AJ, Leyenaar JK. Neuropsychol Rev. 2004;14:99–113. Lilenfeld LRR, Wonderlich S, Riso LP, Crosby R, Mitchell J. Clin Psychol Rev. 2006;26:299–320. Lopez C, Tchanturia K, Stahl D, Treasure J. Psychol Med. 2008a;38:1393–404.
89 Neuropsychological Aspects of Eating Disorders – A Focus on Diagnostic Criteria
1395
Lopez C, Roberts ME, Tchanturia K, Treasure, J. Eur Eat Disord Rev. 2008b;16:411–20. Mathes WF, Brownley KA, Mo X, Bulik CM. Appetite. 2009;52:5445–53. McAnarney ER. J Adolesc Health. 2008;42:321–3. Nunn K, Frampton I, Gordon I, Lask B. Eur Eat Disord Rev. 2008;16:355–60. Pearce JMS. Eur Neurol. 2006;55:53–6. Palmer B. Psychiatry. 2008;7:143–6. Seeger G, Braus DF, Ruf M, Goldberger U, Schmidt MH. Neurosci Lett. 2002;326:25–8. Schmidt U. Eur Child Adolesc Psychiatry. 2003;12:30–7. Southgate L, Tchanturia K, Treasure J. J Psychiatr Res. 2008;160:221–7. Swayze VW 2nd, Andersen AE, Andreasen NC, Arndt S, Sato Y, Ziebell S. Int J Eat Disord. 2003;33:33–44. Swinbourne JM, Touyz SW. Eur Eat Disord Rev. 2007;15:253–74. Tchanturia K, Anderluh MB, Morris RG, Rabe-Hesketh S, Collier DA, Sanchez P, Treasure, J. J Int Neuropsychol Soc. 2004;10:513–20. Tchanturia K, Campbell IC, Morris R, Treasure J. Int J Eat Disord. 2005;37:S72–6. Treasure J. Medicine. 2008;36:430–5. Uher R, Murphy T, Brammer MJ, Dalgleish T, Philips ML, Ng VW, Andrew CM, Williams SCR, Campbell IC, Treasure J. Am J Psychiatry. 2004;161:1238–46. Uher R, Murphy T, Friederich H-C, Dalgleish T, Brammer MJ, Giampietro V, Philips ML, Andrew CM, Ng VW, Williams SCR, Campbell EC, Treasure J. Biol Psychiatr. 2005;58:990–7. Uher R, Treasure J, Heining M, Brammer MJ, Campbell IC. Behav Brain Res. 2006;169:111–9. Wade TD, Bergin JL, Tiggemann M, Bulik CM, Fairburn CG. Austr NZ J Psychiatry. 2006;40:121–8. Wagner A, Greer P, Bailer UF, Frank GK, Henry SE, Putnam K, Meltzer CC, Ziolko SK., Hoge J, McConaha C, Kaye WH. Biol Psychiatry. 2006;59:291–3. Wilsdon A, Wade TD. J Psychiatr Res. 2006;40:746–54.
wwwwwwwwwwwwwww
Chapter 90
Lexical-gustatory Synesthesia and Food- and Diet-related Behavior Julia Simner
Abbreviations Hz Hertz fMRI Functional magnetic resonance imaging DTI Diffusion tensor imaging
90.1 Introduction Synesthesia is a multivariant condition that gives rise to fundamental differences in sensory and cognitive functions. For people with synesthesia – known as synesthetes – everyday activities such as reading or listening to music, give rise to extraordinary experiences such as colors, tastes, smells, and other sensations. The condition is characterized by the co-activation of two (or more) functions when only one is stimulated. In other words, synesthetes experience two (or more) phenomenological sensations when only one would be felt by the average person. For example, when synesthete JW hears a 370-Hz single piano note, she experiences it not only as a sound, but also as a darkish offyellow color (Ward et al. 2006). Synesthetic sensations often feel perceptually “real,” and can be phenomenologically almost identical to real-world experiences. However, they are not considered hallucinations because the synesthete is almost always aware that these perceptions are not part of the outside world. For example, synesthete JS experiences letters and digits in color (and this is known as grapheme-color synesthesia), and those colors are seen projected onto the written typeface when he reads. However, JS is fully aware that the print itself is black on white. In other words, synesthetes are aware that their sensations are produced by their own mind, although they often assume that these sensations are experienced by all people. Synesthesia has been a curiosity to psychologists for several hundred years, although the last decade has seen a considerable explosion of interest. Over 50% of all articles written on synesthesia during the last 100 years have been written within the last 5 years alone, and interest in the condition is higher now that at any previous time (see Fig. 90.1 for a statistical history of synesthetic reports over the last century). J. Simner (*) School of Philosophy, Psychology and Language Sciences, University of Edinburgh, 7 George Square, Edinburgh, EH8 9JZ, UK e-mail:
[email protected];
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_90, © Springer Science+Business Media, LLC 2011
1397
J. Simner Number of synaesthesia publications per year
1398
90 80 70 60 50 40 30 20 10 0 1909
1919
1929
1939
1949
1959
1969
1979
1989
1999
2009
Years (1909-2009)
Fig. 90.1 History of synesthesia research. The number of research articles on the topic of synesthesia over the last 100 years, 1909–2009 (Web of Science; www.wos.mimas.ac.uk)
Very early treatments of synesthesia in the last century were hampered by the behaviorist movement in the psychological sciences which emphasized behavior over mental experience, and effectively knocked synesthesia out of scientific consciousness for almost 80 years. Public knowledge remained low, and this was compounded by the fact synesthetes rarely mention their unusual experiences, believing that everybody shares them. At the end of the twentieth century, however, improved methodologies and the advent of brain imaging (see next) re-introduced synesthesia into empirical study, and it has since found its way into the scientific main-stream. As a result, synesthesia studies no longer focus on simple tests of genuineness, as they did in the nineteenth and twentieth centuries, but instead, use a range of methodologies to reveal the complex system of cognitive, neurological, and sensory processes that underlie the condition. In turn, improvements in the public understanding of science, and the usefulness of the World Wide Web in bringing together large populations of synesthetes, have further transformed the field. Next, I give details of what has been learned about variants of synesthesia that involve food, taste, flavor, and diet, focusing particularly on lexicalgustatory synesthesia, the most well-understood food-related variant, in which flavor sensations are triggered by words. First, however, I give a brief overview of synesthesia, describing its prevalence, inheritance, genetic roots, and neurological basis.
90.2 Synesthesia: A Background Synesthesia is an umbrella term that describes many different manifestations, according to the particular type of synesthetic inducer (i.e., triggering stimulus) and the particular type of synesthetic experience (or concurrent; Grossenbacher 1997). For synesthetes who experience synesthetic colors from sounds, for example, the inducer and concurrent are sound and color, respectively. However, at least 50 different variants have been attested to date (Day 2005). The most common type of synesthetic inducers are language units, such as letters, words, and digits, and these account for around 88% of all synesthesias identified in one recent large-scale assessment of prevalence (Simner et al. 2006).
90 Lexical-gustatory Synesthesia and Food- and Diet-related Behavior
1399
For example, synesthetes may experience the word Tuesday as a pale yellow color, or the name Phillip as tasting of oranges. The fact that many variants are triggered by language units such as these shows that synesthesia can involve higher level processing, and is more than simply a “merging of the senses.” Indeed, many studies now provide empirical evidence of a range of nonsensory synesthesias, involving personality constructs (e.g., Simner and Hubbard 2006), abstract word meanings (Simner and Ward 2006), and other language features (e.g., Simner et al. 2006a). Such nonsensory attributes are often acquired during childhood development (e.g., during literacy training), and this provides important evidence that synesthesia is sensitive to environmental influences. Next, we shall see that synesthesias involving taste are sensitive to the dietary environment of the synesthete. The most common synesthetic concurrent is color, and so the most prevalent type of synesthesia overall is the involuntary and automatic pairing of linguistic units (such as days of the week, letters, and digits) with colors. In contrast, the chemical senses are comparatively under-represented (see Fig. 90.2, which shows the proportion of synesthesias whose inducer or concurrent involves the chemical senses, compared with other variants; Simner et al. 2006b). Nonetheless, there have been a number of reports of synesthesias involving gustation, olfaction, and flavor. Day (2005) describes a variant in which the flavor of foods in the mouth triggers the visual perception of color. These colored photisms appear in the visual field in response to the flavors of food in the mouth. Flavors in the mouth can also give rise to haptic sensations of touch against the skin. This particular variant was first described in detail by Richard Cytowic’s The Man who Tasted Shapes (Cytowic 1993). His synesthete experiences geometric shapes triggered by flavors in the mouth, and these shapes can be felt against the skin, as well as seen in the mind’s eye. As the taste develops on the tongue, the form of these shapes apparently changes (e.g., from round to pointed), and the case was verified by consistency over time, and by reduced cortical blood flow in Xenon SPECT imaging (Cytowic 1993; Cytowic and Wood 1982). Taste and flavor can also arise as the synesthetic concurrent, and one recent example has been attested by Beeli et al. (2005). Beeli and colleagues report a synesthete who experiences both pure tastants, and complex flavors when hearing musical tone intervals. For example, hearing a minor sixth interval triggers the sensation of cream, while hearing a major third triggers the pure sensation of sweetness. However, perhaps the most well-understood variants involving the chemical senses is known as lexical-gustatory synesthesia in which words trigger sensations of flavor in the mouth. This variant is described in detail in the present article, after further information about the general characteristics of synesthesia.
chemical senses other senses (or non-sensory attributes)
Fig. 90.2 Synesthesia and the chemical senses. The proportion of synesthesias identified in random sampling (n = 500) which involve the chemical senses as either inducer or concurrent, compared with other senses (including nonsensory variants involving language) (Data adapted from Simner et al. 2006)
1400
J. Simner
Although synesthesia exists in many different forms, all variants have a number of characteristics in common. First, synesthesia appears to be typified by the consistency of the inducer-concurrent pairings over time. In other words, the type of synesthetic experience tied to any given trigger (e.g., the particular color of a letter) tends to remain constant throughout the synesthetes’ life-time. For example, if the letter a is deep carmine red for any given synesthete, it will have been red for as long as that synesthete can remember, and will continue to be red throughout her life-time. (Some letters can have two colors, although again, these two colors are consistent over time.) This same consistency is found across synesthesias involving taste and flavor. For example, if hearing the name Phillip triggers the flavor of unripe oranges, it will always trigger that exact same association. The assumption of consistency has been a fundamental tool for synesthesia researchers, who use this characteristic as a diagnostic for the condition. Specifically, all studies tend to include an obligatory description of how the synesthete has been verified as genuine: the synesthete is first asked to provide her sensations for a set of stimuli (e.g., to list her colors for the letters of the alphabet), and is then given a surprise retest many months, or even years later. Synesthetes tend to perform at least 90% consistent over considerable time intervals, while non-synesthete control participants perform very badly. Controls are asked to invent comparable associations (e.g., to invent colors for letters) and then to recall these associations in a retest, which they do very poorly, even after a far shorter time interval. For example, synesthetes tested across a year will significantly out-perform controls tested after only a week, and this pattern remains even when controls (but not synesthetes) are prewarned about the retest, and even if controls are given financial incentives to perform well (Ward and Simner 2003). Due to its ubiquitous use, the test of consistency is now considered the “behavioral gold standard” test of genuineness for synesthesia (e.g., Rich et al. 2005). It may yet be possible that a small number of synesthetes have associations that are not consistent over time, and this may be most likely where their experiences are mediated in some way by emotion (e.g., some synesthetes report colors influenced by how they feel). However, there has been virtually no study of such synesthetes in the literature, perhaps given the difficulties encountered in establishing genuineness in the absence of the behavioral gold standard test of consistency. Across all variants, synesthesia is found in at least 1 in 23 people, making a prevalence of at least 4%. Moreover, this figure is likely to be an underestimate, given that several novel variants of synesthesia have been identified since this seminal prevalence study took place (in 2005; reported in Simner et al. 2006b). For example, recently identified novel variants include mirror-touch synesthesia, in which tactile sensations are felt against the body in response to watching others being touched (Banissy and Ward 2007). The introduction of additional variants into the synesthesia literature means that a prevalence study carried out today would likely identify a yet higher proportion of synesthetes than was found previously. Another area that has seen development in recent studies is the estimate of female:male ratios among synesthetes. Older studies suggested that synesthesia was predominantly a female trait, because early methods for assessing prevalence found up to six times more female synesthetes than male synesthetes (e.g., Baron-Cohen et al. 1996). However, these studies relied on self-referral in their count of synesthetes, and it has since been shown that the selfreferral methodology may artificially inflate the count for women (Simner et al. 2006b). This is because, across a range of different psychological phenomena, women are more likely to report atypical behavior than are men. Hence, female synesthetes may simply be more likely to identify themselves in self-report compared with male synesthetes. When prevalence was subsequently assessed without reliance on the self-referral method, approximately the same number of female and male synesthetes was found (Simner et al. 2006b). As such, the condition is likely to affect women and men in equal numbers (or perhaps women in only very moderately greater numbers; see Simner et al. 2006b).
90 Lexical-gustatory Synesthesia and Food- and Diet-related Behavior
1401
Synesthesia is known to run in families (e.g., Ward and Simner 2005) and one recent whole genome scan of grapheme-color and music-color synesthetes found evidence of linkage to chromosomes 2q24, 5q33, 6p12, and 12p12 (Asher et al. 2009). Hence, the condition has a genetic basis which has now been identified, at least for some variants. This genetic inheritance is assumed to give rise to neuro-developmental differences in the brain maturation of synesthetes. The brains of adult synesthetes show increased structural connectivity, and this hyperconnectivity has been identified using a methodology known as diffusion tensor imaging (DTI; Rouw and Scholte 2007). The DTI technique identifies pockets of hyperconnectivity in the white matter of the brain, by tracking the movement of water molecules. Where white matter fiber bundles are more dense or more coherent, water moves in an anisotropic fashion (i.e., with restricted movement along some plane). In the brains of synesthetes, DTI imaging shows that pockets of hyperconnectivity are found in regions associated with the synesthesia. For example, synesthetes reporting colored letters showed increased connectivity (inter alia) in regions near to those implicated in the processing of letters and of colors (Rouw and Scholte 2007). Other areas implicated are those involved in the binding of features: i.e., areas responsible for the combing of featural information such as color and shape in the recognition of objects, for example. The role of these regions in synesthesia suggests that the condition may represent a type of hyperbinding – a propensity to combine features that are not present in the outside world. In addition to structural differences, synesthetes’ brains also show functional differences, using functional magnetic resonance imaging (fMRI). The fMRI technique tracks oxygenated blood flow and indicates which brain regions are likely to be active in certain tasks. Studies using fMRI to identify the regions implicated in synesthetic experiences show that synesthetic sensations activate the same regions as those that support veridical perception. For example, synesthetes reporting color sensations (e.g., from graphemes) show activation in color-selective regions (e.g., human V4) when hearing or reading graphemes (e.g., Nunn et al. 2002; for a review of functional imaging data in synesthetes, see Hubbard and Ramachandran 2005). Put differently, imaging patterns showed evidence of color processing, where no color was present in the outside environment. Similar patterns are found in synesthesias involving flavor perceptions. Parslow et al. (unpublished, see Ward and Simner 2003 for report) generated fMRI data from a single lexical-gustatory (word-to-flavor) synesthete, and showed activation of Brodmann’s area 43 when he listened to words (but not tones, while controls showed no such activity at all). This area is within the primary gustatory cortex, and provides strong evidence that the flavor experiences of lexical-gustatory synesthetes are, to some extent, perceptual in character. In all imaging studies, synesthetes’ patterns of neurological data are compared with those of the control participants, and these types of comparison show systematic differences in both structure and function across groups. But imaging studies have also shown differences within groups of synesthetes, and these differences correspond to the ways in which synesthetes experience their sensations. Dixon et al. (2004) identified a distinction between projector and associator synesthetes: projector synesthetes report that their synesthetic concurrents (e.g., colors) are experienced in a similar way to veridical perception, outside the body; for example, color may be seen superimposed onto the written type face. In contrast, associator synesthetes report that their sensations exist in “the mind’s eye” only; for example, that colors are “seen” only on an “internal screen.” This distinction is demonstrated not only in phenomenological reports and behavioral tasks, but is also seen in different patterns of fMRI activation. Hubbard et al. (2005) showed that both projector and associator synesthetes have significantly more activation in color-selective regions than control participants, but that projectors show most activation of all. These type of data provide neurological support for differences in the phenomenological reports of synesthetes, and validate the range of experiences they report. Next, we shall see that the same projector/associator distinction can be applied to synesthesias involving food, taste, flavor, and diet, and we turn to this subset of synesthesias now.
1402
J. Simner
90.3 Lexical-gustatory Synesthesia Previously we saw that each of the many different variants of synesthesia is characterized by the pairing of a particular type of synesthetic inducer with a particular type of synesthetic concurrent, and that taste/flavor can serve in either capacity. One of the best understood variants of the foodrelated synesthesias is lexical-gustatory synesthesia. In this variant, the inducer and concurrent are words and flavors, respectively. So for lexical-gustatory synesthetes, hearing, saying, or reading words gives rise to associated food experiences (Ward and Simner 2003; Ward et al. 2005; Simner and Ward 2006; Simner and Haywood 2009). For synesthete JIW, for example, the word this tastes of bread soaked in tomato soup, while the name Philip tastes of unripe oranges (Ward and Simner 2005). Equally, synesthete MM has sensations of flavor which flood her mouth whenever she hears proper names (e.g., John tastes of cornbread). Table 90.1 shows other examples of flavor sensations reported by lexical-gustatory synesthetes. As in other variants of synesthesia, the condition is characterized by the consistency of inducerconcurrent pairings over time. Hence, the particular flavor associated with any given word tends to remain highly constant across the synesthete’s life-time. In one extreme example, the consistency of word–food experiences of a lexical-gustatory synesthete was verified as being unchanged across almost 30 years. In this study, Simner and Logie (2007) presented a number of consistency tests to lexicalgustatory synesthete, JIW, including a retest of word–flavor associations first elicited in 1979. JIW was given a questionnaire in 2006 containing the target words from the original 1979 list in a randomized order, and was asked to write his synesthetic flavor associations, some 27 years after these had been initially elicited (by a third-party associate of JIW). JIW’s performance in this task was compared with a group of ten non-synesthete age-matched controls. Control participants were given the same list of words and were asked to invent analogous associations (i.e., they were asked to pair each word with any food name of their choice). Controls were forewarned that they would be immediately retested as soon as they had generated the last of their word–taste pairings, and that they should make their associations as easy to recall as possible (e.g., if the target word were Christmas, they should write turkey). Controls were around 48% consistent in their immediate retest after only 10 s. JIW scored 100% consistent in his associations over 27 years. In other words, JIW significantly out-performed controls, notwithstanding the considerable difference in retest intervals. Simner and Logie concluded that JIW’s performance was achieved by something other than the conscious, episodic retrieval of paired associations. Indeed, JIW reports that his score was simply the result of reporting exactly the flavors that flooded his mouth as he heard each word, and that these flavors simply do not change. Lexical-gustatory synesthesia is rare, even within the rarity of synesthesia overall, although 10–15 cases have now been reported in the contemporary literature (Cytowic 1989; Ward and Simner 2003; Table 90.1 Synesthetic flavors. Examples of flavor concurrents reported by lexical-gustatory synesthesia in a selection of historical and contemporary reports Source Inducer word Concurrent description Quality Simner and Haywood (2009) Safety “Lightly buttered toast” Food item Simner and Ward (2006) Tambourine “Crumbly biscuit” Food item including texture Ward and Simner (2003) Jail “Cold, hard bacon” Food item including texture and temperature Pierce (1907) Ethel “A thimble on the tongue” Texture only Ferrari (1907, 1910) Alessandro “Taste of fried potato and Food items such as flavor smell of burnt wool” and smell Simner and Haywood (2009) Spluk “Yoghurt” Food item from nonword Ward et al. (2005) Beef “Overcooked, dried out beef” Food item-matching inducer
90 Lexical-gustatory Synesthesia and Food- and Diet-related Behavior
1403
Ward et al. 2005; Simner and Ward 2006; Simner and Haywood 2009) as well as four cases from historical sources (Ferrari 1907, 1910; Pierce 1907). The rareness of the condition is attested by the fact that no cases at all were found in the n = 500 population recruited for one recent prevalence study based on random sampling (Simner et al. 2006b), which otherwise found 38 different instances of synesthesia. This suggests that the prevalence of lexical-gustatory synesthesia is less than 0.2% of the population, and may be considerably lower. In investigating this unusual condition, we can consider in turn the nature of the concurrent experience, the nature of the inducing stimulus, and the processes by which these become associated, and we address each in turn in the following.
90.3.1 The Synesthetic Concurrent: Issues of Flavor, Food, and Taste Previously we saw that synesthetes can be divided into projectors and associators, according to their phenomenological experience of the concurrent. We saw that projector grapheme-color synesthetes, for example, experience their colors as similar to veridical perceptions, subjectively located out in the word. In contrast, associator synesthetes see colors only in their “mind’s eye,” by association rather than with “real-world type” perception. The same projector/associator distinction has been made for lexical-gustatory synesthetes, who can be divided into the same two broad groups, this time according to the nature of their concurrent flavor experiences. For projector synesthetes, flavors are subjectively located in the mouth; synesthetic flavors are similar to veridical perceptual experiences and have the same reported phenomenology as those generated by food substances. For example, synesthete JIW reports that the only difference between his synesthetic flavors, and those he experiences while eating, comes from the fact that the former do not involve substances that can be rolled on the tongue. All other sensations, he reports, are phenomenologically identical (e.g., also see the subsequent part regarding texture and temperature). In a similar way, synesthete MM (Simner and Ward 2006) reports that her mouth is flooded with the taste of baked cornbread when she encounters the name John. In contrast, ‘associator’ lexical-gustatory synesthetes do not experience perceptual flavors in the mouth; instead, the food concurrent represents a type of cognitive association, a “mental link” to a food-type, which automatically enters into consciousness when the inducing word is encountered. Synesthete PS, for example, has the overwhelming notion of orange-flavored jelly when he hears, reads, or says the word shoulder. Although their subjective experiences differ, both associator and projector synesthetes represent the same, single type of synesthesia, with a shared cognitive basis, and with what is likely to be a shared neurological cause (see next). In all cases of (projected/associated) lexical-gustatory synesthesia, food experiences are complex sensations, rather than generic tastes of bitter/sweet, etc. The flavors themselves are highly specific and extremely rich in detail, and the synesthetes often go to considerable trouble when describing them (Ward and Simner 2003). For synesthete JG for example, the name Adrian tastes of dressed salad, but not just any type of dressed salad: specifically, it is lettuce coated with Caesar salad dressing, and cannot (to her mind) be anything else. Equally, synesthete CS tastes the word part as a type of soup: specifically, chicken noodle soup and no other (Ward et al. 2005). Associations can incorporate temperature and texture as well as taste, and so the variant might be more properly described as “lexical-flavor” synesthesia (L.E. Marks, personal communication). For synesthete JIW for example, the word jail generates the experience of bacon that is cold and hard (Ward and Simner 2003), while tambourine is a biscuit with a crumbling texture. Some lexical-gustatory synesthetes have also reported olfactory experiences, although it is difficult to draw strong conclusions from this (Ward et al. 2005). Taste and smell are difficult to separate subjectively, and the rated intensity of taste sensations in the mouth is increased in the presence of olfactory cues (Murphy et al. 1977).
1404
J. Simner
sweets/chocolate meat vegetables fruit bread/cereal dairy drinks synthetic inedibles bodily inedibles other
Fig. 90.3 Synesthetic flavor concurrents. Summary of synesthetic flavor associations of synesthete JIW, elicited by 1195 target words (Reproduced from Simner 2007)
However, in one historical case (Ferrari 1907) the lexical-gustatory synesthete reported that the same word triggered different food experiences in the gustatory versus olfactory domains (e.g., Alessandro gave the smell of burned wool, but tasted of fried potatoes). This historical report suggests that concurrents may indeed extend across both flavor and smell, and can do so independently. Studies have shown the influence of diet and eating behavior in the nature of lexical-gustatory experiences. Analyses of synesthetic concurrents have shown that certain flavors tend to dominate, while others are conspicuous in their absence. In particular, common synesthetic flavors are for sweets (candy) and chocolate, while alcohol and other “adult tastes” tend to be absent. The synesthetic flavor concurrents of synesthete JIW are shown in Fig. 90.3, which illustrates that some flavors are present in greater proportions than others. It is clear from Fig. 90.3 that certain food types dominate as synesthetic flavors, while others are less dominant, and studies have shown that these patterns are related to the dietary environment of the synesthete. The frequency of synesthetic flavors is correlated with the frequency with which the corresponding food is encountered in the synesthete’s diet: commonly eaten foods are significantly more likely to occur as synesthetic concurrents (Ward and Simner 2003). Moreover, the dietary environment is most influential during development: synesthetic tastes are statistically more likely to reflect the diet eaten during childhood compared with adulthood. For instance, although the adult JIW is now a heavy coffee drinker, he did not drink coffee as a child, and consequently, coffee is almost entirely absent from his synesthetic flavors. Indeed, the only time it occurs is in “childfriendly” form, as coffee-flavored chocolates which he may have encountered at a young age. Ward and Simner (2003) confirmed this link between environment, development, and synesthetic experience, by administering dietary questionnaires to both JIW and to his mother; the former was questioned about foods in his current diet, and the latter, about foods in JIW’s childhood diet. Both types of dietary information predicted his synesthetic flavors, although the most dominant influences came from his childhood eating patterns. Indeed, Simner (2007) points out that foods consumed during JIW’s childhood are 10 times more likely to occur as synesthetic concurrents, compared with those encountered only in later life. This dominance of early foods suggests that synesthetic associations are formed during childhood development, and that they are resistant to change after adolescence. Figure 90.3 shows that a small number of synesthetic flavors are nonfoods. These include bodily inedibles such as earwax, and synthetic inedibles such as plastics. It is possible that these tastes were experienced as part of early exploratory eating behaviors in childhood.
90 Lexical-gustatory Synesthesia and Food- and Diet-related Behavior
1405
90.3.2 The Synesthetic Inducer: Words, Phonology, and Meaning Synesthetic flavors are triggered by words, and are influenced by phonemes (i.e., speech sounds), since similar sounding words tend to taste alike (e.g. for JIW, the words message and college both taste of sausage meat). Linguistic and statistical analyses show that each synesthetic flavor can be traced to a particular phoneme (or phonemes). For example, words containing the sound /k/ tend to taste of egg for synesthete JIW, and this is true whether the phoneme /k/ is spelt with a k (e.g., York), c (e.g., accept), ck (e.g., chuck), or x (e.g., fax) (Ward and Simner 2003). Some phoneme “triggers” have been shown to derive from food names. For example, for JIW, words containing the phonemes /I/, /n/, and /s/ (e.g., prince, cinema) trigger the flavor taste of mince, and the name of this food itself also contains these same three phoneme (mince = /mIns/). However, other phoneme relationships have less obvious roots: words containing the phoneme /f/ are significantly likely to taste of sherbet, although there is nothing in the name of this food to suggest why this particular phoneme might be important. However, the semantic associations to the word fizz /fIz/ may play a role, given the effervescence experienced in the mouth from sherbet food stuff, and given too, other semantic associations which are known to dictate the mappings. For example, for JIW, the word blue tastes “inky,” the word bar tastes of “chocolate,” and the word newspaper tastes of chips (French fries, which are traditionally served wrapped in newspaper in the United Kingdom). In other words, both the meaning of a word, and its sounds, can determine the nature of the synesthetic mapping between word and flavor. It is known that synesthetic tastes can be triggered even when the synesthete is entirely unaware of how the word sounds. Simner and Ward (2006) placed lexical-gustatory synesthetes into tipof-tongue states. Tip-of-tongue is the familiar experience in which a word is known but temporarily cannot be recalled from memory. This arises when the word’s meaning, but not its spelling or phonemes, has been retrieved. Simner and Ward placed synesthetes in tip-of-tongue states by asking them to name pictures of uncommon objects (e.g., metronome, platypus). In tip of tongue, synesthetes began to taste the word before they could say it. One woman, for example, tasted Dutch chocolate while struggling to retrieve the word phonograph, which is her synesthetic flavor for that word. This study shows that synesthetic tastes can be triggered by word meaning alone, even when sound and spelling are temporarily inaccessible. Finally, synesthetes can taste foods from words they have never heard before, and from words that have no meaning at all. Simner and Haywood (2009; also Ward and Simner 2003; Ward et al. 2005) showed that synesthetes experiences tastes from nonsense words (e.g., noik), although these taste less intensely than real words. Figure 90.4 shows the selfrated intensity of synesthetic flavors from nonwords and real words, and illustrates the reduced strength of flavor in the former.
90.4 Applications to Other Areas of Health and Disease Synesthesia is not recognized as a disease in any of its manifestations, but rather, as an alternative form of perception. Indeed, the widely held view is that synesthesia is something of an asset, which is known to be linked to improved memory, for example (e.g., Yaro and Ward 2007; Smilek et al. 2002). On closer inspection, however, the condition has a rather complex profile of costs and benefits, depending on the variant. In sequence-space synesthesia, for example, in which sequenced units such as numbers are seen projected into particular spatial arrays, synesthetes are significantly slower at certain aspects of mental calculation (known elsewhere to involve spatial processes; Ward et al. 2009). Other costs come in collateral experiences: synesthetes report a sense of malaise and discomfort when viewing perceptual objects that conflict with their synesthetic sensations. For example, a synesthete with a red letter a might feel discomfort when viewing the letter a printed in green
1406
J. Simner 80 70 60 50 self reported intensity of 40 synaesthetic flavour
nonwords
30
real words
20 10 0 nonwords real words type of inducer word
Fig. 90.4 Real words and nonwords. Self-rated intensity of synesthetic sensations of flavor from nonwords (e.g., noik) and real words (Data adapted from Simner and Haywood 2009)
(Callejas et al. 2007). These are fairly mild afflictions however, and overall, most synesthetes report positive experiences and even pleasure from their synesthetic sensations. One very notable exception, however, are those synesthetes who experience flavor in the mouth. Lexical-gustatory synesthete JIW, for example, has described his synesthesia as causing problems at both work and home. He has trouble concentrating during meetings or on the phone, since over half of every word he encounters every minute of every day triggers sensations of strong flavors that flood the mouth. Sometimes these flavors are distracting in their intensity, and other times they are distracting in their unpleasantness (e.g., tastes of earwax, vomit, and other bodily inedibles). Some flavors are particularly strong and persistent, and can last until overridden by a subsequent taste. For these reasons, projector lexical-gustatory synesthetes often regard their synesthesia as a source of irritation, and one that makes it difficult to concentrate. In contrast, associator lexical-gustatory synesthetes are more likely to find their experiences a source of interest of pleasure. As in all variants of synesthesia, problems sometimes also arise in early childhood, when synesthetes first express their sensations to others, and are often met with disbelief and even ridicule (Day 2005). What is clear, however, is that synesthesias involving the chemical senses (especially in a projected form) appear in some way more noticeable to the synesthete, or more intrusive in daily life. First-person reports also suggest that taste-related synesthesias can have an impact on digestion and diet. Projector synesthetes have reported anecdotally that they encounter digestive discomfort, because their stomach is always anticipating foods that never arrive. Synesthetic tastes certainly appear to cause salivation, for example, and this may be tied to other physiological processes usually involved in food consumption. In addition, associator synesthetes have reported difficulties in maintaining a healthy weight. Associator synesthetes are constantly reminded of foods when speaking, hearing, or reading, and this, they suggest, triggers food-seeking behaviors. A common question from associator synesthetes is whether their condition has been linked to obesity, or weight-maintenance problems. However, no empirical work on this issue has yet been conducted. Lexical-gustatory synesthesia is a condition of unique interest to neuroscientists and psychologists, but is yet to be explored in any detail by those in the dietary sciences. The gustatory and olfactory “reality” of synesthetic tastes, textures, temperatures, and odors allow us to see, in a very direct
90 Lexical-gustatory Synesthesia and Food- and Diet-related Behavior
1407
Table 90.2 Lexical-gustatory synesthesia: key points 1 Inducer = heard, read, spoken, or subvocalized words 2 Concurrent = flavors, including taste, texture, temperature, and olfaction 3 Phenomenology = flavors are experienced either as veridical perceptions in the mouth (in projector synesthesia) or as an automatic cognitive association (in associator synesthesia). 4 Associations are automatic and cannot easily be repressed 5 Associations are highly consistent over time 6 Flavors are influenced by dietary experience, and frequently consumed foods generate the common synesthetic experiences. Childhood diet is particularly influential 7 Flavors can also include organic and synthetic inedibles, such as vomit, mucous, earwax, stones, plastics, and flowers. These were likely experienced at a young age through exploratory eating behaviors 8 Projector lexical-gustatory synesthetes report irritation, distraction, or sometimes discomfort from their experiences. Associator synesthetes report difficulties in weight maintenance
way, that the qualia of flavor perception are the result of internal processes, rather than external stimuli. Conditions such as lexical-gustatory synesthesias show us that flavor sensation is the product of a psychological reality which can be constructed in the absence of the usual external stimulation from food (Table 90.2).
Summary Points • Synesthesia is a condition that gives rise to a merging of sensory and/or cognitive functions. • Synesthetes experience two or more sensations when only one is stimulated. • Synesthesia has a range of different manifestations, depending on the particular triggering stimulus (inducer) and the particular type of synesthetic experience (concurrent). The most common inducer is language and the most common concurrent is color. • Taste and flavor can be either synesthetic inducers of synesthetic concurrents. • The best understood variant of synesthesia involving flavor/taste is lexical-gustatory synesthesia, in which words trigger complex sensations of flavor. • Lexical-gustatory synesthesia can produce flavor concurrents that are experienced as veridical sensations in the mouth (in projector synesthesia) or as automatic cognitive associations (in associator synesthesia). • Synesthesia has a complicated profile of costs and benefits, although synesthesias involving tastes are most often reported as problematic. Definitions and Explanations of Key Terms Synesthesia: An inherited neurological condition that gives rise to the merging of sensory and/ or cognitive functions. Synesthete: A person with developmental or inherited synesthesia. Phoneme: The minimal contrastive unit in the sound system of a language; substituting one phoneme for another changes the meaning of a word (e.g., /t/ vs /n/ because /bat/ differs in meaning to /ban/). Grapheme: The minimal contrastive unit in the writing system of a language (e.g., t vs n). Inducer: Stimulus that triggers the synesthetic sensation. Concurrent: The particular sensation triggered during synesthesia, in addition to usual perceptions (e.g., in music-color synesthesia; music triggers the synesthetic concurrent of color, in addition to auditory perception).
1408
J. Simner
References Asher JE, Lamb JA, Brocklebank D, Cazier JB, Maestrini E, Addis L, Sen M, Baron-Cohen S, Monaco AP. Am J Hum Genet. 2009;84:279–85. Banissy MJ, Ward J. Nat Neurosci. 2007;10:815–6. Baron-Cohen S, Burt L, Smith-Laittan F, Harrison J, Bolton P. Perception. 1996;25:1073–9. Beeli G, Esslen M, Jäncke L. Nature 2005;434:38. Callejas A, Acosta A, Lupianez J. Brain Res. 2007;1127:99–107. Cytowic RE, Wood FB. Brain Cognition. 1982;1:36–49. Cytowic RE. Synaesthesia: A union of the senses. Springer, New York; 1989. Cytowic RE. The man who tasted shapes. London: Abacus Books; 1993. Day S. In: Robertson LC, Sagiv N editors. Synesthesia: perspectives from cognitive neuroscience. New York: Oxford University Press; 2005. p. 11–33. Dixon MJ, Smilek D, Merikle PM. Cogn Affect Behav Ne. 2004;4:355–43. Ferrari GC. Riv Psicologia. 1907;3:297–317. Ferrari GC. Riv Psicologia. 1910;6:101–4. Grossenbacher PG. In: Baron-Cohen S, Harrison JE editors. Synaesthesia: classic and contemporary readings. Oxford: Blackwell; 1997. Hubbard EM, Ramachandran VS. Neuron. 2005;48:509–20. Hubbard EM, Annan AC, Ramachandran VS, Boynton GN. Neuron. 2005;45:975–85. Murphy C, Cain WS, Bartoshuk LM. Sens Process. 1977;1:204–11. Nunn JA, Gregory LJ, Brammer M, Williams SCR, Parslow DM, Morgan MJ, Morris RG, Bullmore ET, Baron-Cohen S, Gray JA. Nat Neurosci. 2002;5:371–5. Pierce AH. Am J Psychol. 1907;18:341–52. Rich AN, Bradshaw JL, Mattingley JB. Cognition. 2005;98:53–84. Rouw R, Scholte HS. Nat Neurosci. 2007;10:792–7. Simner J. Trends in cognitive Sciences. 2007;11:23–29. Simner J, Haywood SL. Cognition. 2009;110:171–81. Simner J, Hubbard EM. Neuroscience. 2006;143:805–14. Simner J, Ward J. Nature. 2006;444:438. Simner J, Logie RH. Neurocase. 2007;13:358–65. Simner J, Glover L, Mowat A. Cortex. 2006a;42:281–9. Simner J, Mulvenna C, Sagiv N, Tsakanikos E, Witherby SA, Fraser C, Scott K, Ward J. Perception. 2006b;35:1024–33. Simner J. In: Windhorst U, Binder M, Hirokawa N editors. Encyclopedia of neuroscience. GmbH, Heidelberg: Springer Verlag. Smilek D, Dixon MJ, Cudahy C, Merikle PM. Psychol Sci. 2002;13:548–52. Ward J, Simner J. Cognition. 2003;89:237–61. Ward J, Simner J. Perception. 2005;34:611–23. Ward J, Simner J, Auyeung V. Cogn Neuropsychol. 2005;22:28–41. Ward J, Huckstep B, Tsakanikos E. Cortex. 2006;42:264–80. Ward J, Sagiv N, Butterworth B. Cortex. 2009;45:1261–5. Yaro C, Ward J. Q J Exp Psychol. 2007;60:682–96.
Part XV
Pathology and Abnormal Aspects: Behavioral and Psychological
wwwwwwwwwwwwwww
Chapter 91
Rethinking the Eating Disorder Continuum: A Categorical Approach to Abnormal Eating Jessie L. Miller and Tracy Vaillancourt
Abbreviations DSM Project EAT EDI EAT EDNOS
Diagnostic and Statistical Manual of Mental Disorders Project eating among teens Eating disorder inventory Eating attitudes test Eating disorder not otherwise specified
91.1 Introduction Do eating disorders exist on a continuum? This question has been debated theoretically and empirically in both the clinical and scientific domains for over four decades (Gleaves et al. 2004; Polivy and Herman 1987; Williamson et al. 2005). This debate is not specific to eating disorders; rather it resonates throughout much of the past and current scientific inquiries into the nature of all psychopathology (Gangestad and Snyder 1985; Watson and Clark 2006; Widiger and Samuel 2005). Nearly 100 years ago German psychiatrist Emil Kraepelin (1913) argued that while personality dimensions were the basic underlying constructs necessary for development of any psychological disorder, mental illness itself was qualitatively distinct from ordinary human behavior. Kraepelin’s (1913) fundamental theories on the etiology and diagnosis of psychiatric disorders in the 1900s formed the foundation for the classification system later devised by the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM) first published in 1952. With the inclusion of eating disorders as a diagnostic category in the third edition of the DSM (American Psychiatric Association 1980), controversy sparked over whether eating disorders should be considered along a continuum. Many theorists argued that eating disorders were best conceptualized as existing along a continuum of severity, ranging from ‘healthy normals’ (no weight preoccupation) to mildly weight-preoccupied individuals (with or without disordered eating behaviors), to subthreshold eating syndromes, and finally to the more severe clinical manifestations evidenced by those meeting diagnostic criteria for anorexia nervosa and bulimia nervosa (Button and Whitehouse 1981; Fries 1977; Garner et al. 1993, 1983; Polivy and Herman 1987; Rodin et al. 1984; Striegel-Moore et al. 1986).
J.L. Miller (*) Department of Psychiatry and Behavioural Neurosciences, Offord Centre for Child Studies, McMaster University, Chedoke Site, Central Building, 3rd Floor, 1200 Main Street West, Hamilton, Ontario, Canada, L8N 3Z5 e-mails:
[email protected];
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_91, © Springer Science+Business Media, LLC 2011
1411
1412
J.L. Miller and T. Vaillancourt
A continuum model of eating disorders presumes, in part, that abnormal eating is linearly related to normal eating and that differences are a matter of degree or severity. It also presumes that the individual features or symptoms of an eating disorder can be found across all groups to varying degrees, and are not specific to clinical populations. Furthermore, a continuum model assumes a range of variability in symptoms from least pathological in nonclinical populations to most pathological in clinical populations. There are several problems with this logic however. First, weight preoccupied individuals may be on a continuum with individuals with an eating disorder with regards to the psychological symptoms, such as fear of fat, but they are not on a similar continuum with regards to the behavioral symptoms, namely, the bingeing, purging, and restricting behaviors of an eating disorder. In fact, most people who have some level of body shape/weight preoccupation are not engaged in any clinically relevant abnormal eating behaviors (Miller 2008). Second, not all abnormal eating behaviors found in clinical eating disorder populations are found among nonclinical populations. Purging behaviors are extremely rare in noneating disordered groups and found only infrequently among high risk or subclinical groups, as we will show later in this chapter. Third, not all eating disorder behaviors can be expressed on a continuum ranging from low pathology to extreme pathology. Extreme exercise, a compensatory behavior described in the DSM definition of anorexia nervosa restricting subtype, does not lend itself to a continuum very easily. Extreme exercise, a frequent and intense regimen of exercise, is not necessarily pathological. Consider elite athletes who would necessarily fall at the most extreme end of a continuum of exercise behaviors because of their profession, and yet for this very reason it would be difficult to consider this behavior to be pathological. Purging behaviors are another good example of an eating disorder behavior that does not conform to a continuum approach. At what point along a continuum would self-induced vomiting be considered normal and not pathological? Even at low frequencies, self-induced vomiting is abnormal and the distinction between normal and abnormal is very clearly drawn at the onset of vomiting behaviors. If purging behaviors can be classified as abnormal based on whether they are present or absent this is support for a categorical model of eating behaviors. In fact, some researchers and clinicians have disputed the notion of an eating disorder continuum all along and instead argued for fundamental differences in clinical eating disorders compared to the mild syndromes of weight preoccupation (Bruch 1973; Crisp 1965; Ruderman and Besbeas 1992). However over the last two decades, fewer research studies have been put forth in support of a discontinuous model of eating disorders. It seemed the field was making a shift toward a more continuous approach to understanding normal and abnormal eating behaviors. More recently there have been a number of research studies using advanced statistical methods (i.e., taxometrics) which have reintroduced the notion of a categorical model of eating disorders. These studies do not contradict the prevailing theoretical perspective of an eating disorder continuum, but rather, they provide evidence for simultaneously employing a dimensional and a categorical model (Lowe et al. 1996; Gleaves et al. 2000a, b) as the two are not mutually exclusive (Wilfley et al. 2007). Following the lead of recent taxometric studies, we present in this chapter an argument for a combined categorical-dimensional model of eating disorders. The focus of this model is on the relation between normal and abnormal populations for the symptoms of the eating disorder, but not the eating disorders as a whole. We propose a continuum of eating disorder thoughts (e.g., I am terrified of gaining weight, I think about vomiting after eating, I am dissatisfied with the shape of my body) and a discontinuum of eating disorder behaviors (severe and chronic restriction, vomiting, laxative abuse, bingeing, excessive exercise resulting in weight loss, etc.). Toward this aim, we discuss three areas of eating disorder research that will (a) provide evidence to support a dimensional view of eating disorder thoughts and a categorical view of eating disorder behaviors (see Fig. 91.1); and (b) demonstrate the importance of distinguishing between psychological and behavioral symptoms in the assessment and measurement of eating disorders.
91 Rethinking the Eating Disorder Continuum: A Categorical Approach to Abnormal Eating
1413
Fig. 91.1 A continuum/discontinuum model of eating disorders. What differentiates clinical from nonclinical groups is the presence of eating disordered behaviors and not extreme eating disordered thoughts. Once a person begins to engage in pathological compensatory behaviors (broken vertical line and solid vertical line) they are, in our opinion, at a subclinical or clinical level of an eating disorder
First, we discuss research indicating that many individuals (primarily females) in nonclinical populations experience the psychological symptoms of eating disorders and often at levels that closely approximate weight and shape concerns in clinical eating disorder groups. Yet the behavioral components of eating disorders are not necessarily present. In other words, it is possible to have elevated levels on the psychological criteria for an eating disorder with none of the behavioral features. Thus it seems that in nonclinical populations the psychological symptoms can present independent of an eating disorder. In contrast, while low levels of eating disorder behaviors may be present in nonclinical populations, they do not occur without psychological symptoms – suggesting that these groups may in fact be an undiagnosed subclinical or clinical population. Second, we review recent studies that have tested the latent constructs of eating disorders using taxometric methods and draw attention to the fact that both categorical and dimensional models of eating disorders seem to fit the data, but these results vary depending on whether the targeted indicators are behavioral symptoms or psychological symptoms and also whether the composition of the sample is clinical or nonclinical. Third, we examine the literature on treatment and recovery, and observe that behavioral recovery can often occur independent of psychological symptom recovery and recovery is most often defined by the absence of behavioral symptoms. The relevance of this literature is that it will highlight the fact that psychological symptoms of an eating disorder (i.e., the fear of fat and extreme body dissatisfaction) fail to clearly distinguish not only clinical and nonclinical populations, but also cannot clearly distinguish clinical and recovered populations. We conclude this chapter with a summary of the implications for theory and research in the prevention and detection of eating disorders and offer recommendations for improving detection and diagnosis in epidemiological studies of eating disorders (Table 91.1).
91.2 The “Normative Discontent” In 2004, Cash, Morrow, Hrabosky and Perry (2004), published a critical and systematic review of body image among college women and men using 22 studies – all published between 1983 and 2001. Using data from nonclinical samples they found 29% of white women, 17% of black women, and
1414
J.L. Miller and T. Vaillancourt
Table 91.1 Key points in the argument for a categorical model of eating disorder behaviors Main argument: The clinical behaviors of eating disorders are categorical (“I vomit after eating”) while the cognitive/psychological symptoms are continuous (“I am terrified of being fat”). Evidence presented in this chapter: 1. Psychological symptoms of an eating disorder (body dissatisfaction, fear of being or becoming fat, or gaining weight) are present at both low and high levels in nonclinical and in clinical populations. 2. Behavioral symptoms of eating disorders are present at low and high levels in clinical populations only; when they are recorded in nonclinical samples they are accompanied by psychological symptoms of eating disorders and most likely reflect an undiagnosed subclinical or clinical subgroup. 3. Empirical studies of the latent factor structure of eating disorders show dimensional models when using psychological indicators, but categorical models when using behavioral indicators. 4. Psychological symptoms poorly discriminate between recovered and nonrecovered patients and it is the cessation of behavioral symptoms which most often defines recovery.
16% of white men reported extreme body dissatisfaction. Body dissatisfaction, which entails an intense disparagement toward specific body regions, as well as a general dissatisfaction with one’s shape and weight (American Psychiatric Association 1994), is increasingly common among nonclinical populations and is often associated with significant psychopathology (Polivy and Herman 2002). In conjunction with the overvaluation of shape/weight and fear of fat, body dissatisfaction has been described as the most prominent risk and maintenance factor in pathological eating (StriegelMoore et al. 1986; Polivy and Herman 2002), the most robust predictor of bulimic symptoms (Stice 2002); and the core psychopathology of both bulimia and anorexia nervosa (Fairburn et al. 2003). Yet despite the prominent role of these cognitions in clinical eating disorders, their increasing prevalence in nonclinical populations has reduced the specificity of these symptoms in identifying eating disorders in epidemiological research studies. In the years following Cash et al.’s (2004) review of body dissatisfaction, a number of large epidemiological studies have demonstrated a similar level of discontent with weight and shape among females of varying ages. Approximately 46% of girls and 26% of boys are dissatisfied with their bodies according to recent statistics from Project EAT (Eating Among Teens), which to date is one of the largest (N = 4746), most comprehensive and ethnically diverse epidemiological studies of middle and high school students’ eating behaviors (Ackard et al. 2007). In a review of epidemiological studies of partial syndrome eating disorders from 1980 to 2003, Chamay-Weber, Narring, and Michaud (2005) found 46–80% of adolescent girls in the United States reported intense body dissatisfaction. McVey, Tweed, and Blackmore (2004) sampled 2279 females, 10–14 years of age, from 42 Canadian schools and found 31.3% of the sample felt “too fat” and 29.3% stated they were currently trying to lose weight. Two epidemiological studies on the prevalence of disordered eating using a national Canadian sample (N=36 984), found 26% of women aged 15–24 responded “yes” to having a strong fear of being too fat in the past 12 months (Piran and Gadalla 2006). Extending this age range to women aged 15–65 years, Park and Beaudet (2007) found one in five women (19%) responded “yes” to the same question of having a strong fear of being too fat in the past 12 months and this fear was associated with negative self-esteem, body image preoccupation, and food obsession. High weight preoccupation, however, is not necessarily associated with eating disorder behaviors. For example, Ackard et al. (2007) reported that 26.7% of girls and 20.4% of boys from Project EAT had severe body disparagement without any accompanying eating disorder behaviors (see Fig. 91.2). Bulik, Sullivan, and Kendler (2000) identified a subgroup of women who experienced preoccupied thoughts about weight and shape, but who had not engaged in behaviors associated with a clinical diagnosis of an eating disorder. Cooper and Goodyer (1997) found significant weight and shape concerns among 14.5% of 11–12 year olds, 14.9% of 13–14 year olds, and 18.9% of 15–16 year
91 Rethinking the Eating Disorder Continuum: A Categorical Approach to Abnormal Eating
1415
Fig. 91.2 Extreme body dissatisfaction among adolescent girls and boys in the USA. These statistics are based on Project EAT (Eating Among Teens), a large (N = 4746), and ethnically diverse epidemiological study of middle and high school students’ eating behaviors (Ackard et al. 2007)
olds. However, only the 15–16 year olds showed any significant presence of behavioral pathology. Miller, Vaillancourt and Hanna (2009) compared eating-disordered thoughts to eating-disordered behaviors using items from the Eating Disorder Inventory (EDI; Garner et al. 1983) in a large sample of 1816 female university students and found that several of the items that tapped extreme body dissatisfaction were endorsed by over 50% of the sample whereas items that tapped eating disorder behaviors were rarely endorsed. Further comparing thoughts to behaviors, Miller et al. found that five groups could be identified on the basis of how often women worried about their weight and body image and how often they engaged in pathological compensatory behaviors such as vomiting or extreme food restriction. The most important finding from this study was that in general, behaviors did not occur without psychological symptoms, but psychological symptoms did occur without behaviors. Figure 91.3 illustrates that a significant number of women endorsed moderate to high levels of body dissatisfaction and fears of gaining weight (41%) without endorsing any restricting, binge eating, or purging behaviors. In contrast there were no individuals who endorsed moderate or high levels of restricting, binge eating, or purging behaviors without also endorsing the psychological symptoms. And while there was a sizeable group of women who endorsed moderate to high levels of eating disorder behaviors, it was always in conjunction with a parallel endorsement of the psychological symptoms of an eating disorder. Thus, these results highlight that pathological compensatory behavior is inextricably linked to fears of weight gain and body dissatisfaction, as indicated by the absence of a group characterized by eating disorder behaviors without psychological symptoms of an eating disorder. Importantly, these results were replicated in a national epidemiological sample of Canadian females aged 15–34 years (N = 1,627; Miller 2008). Other studies have reported similar results. For example, Garner, Olmsted, Polivy, and Garfinkel (1984) compared patients with anorexia nervosa to weight-preoccupied and not-weight-preoccupied women drawn from a female college sample and a female sample of ballet students. The weightpreoccupied group displayed similar mean levels of body dissatisfaction and weight preoccupation as the patients with anorexia nervosa. However, only a portion of the weight-preoccupied women were engaged in clinical eating disorder behaviors. A cluster analysis of the weight-preoccupied group revealed one group of women who were elevated on all subscales of the EDI (Garner et al. 1983) and in fact scored as high as or higher than the anorexia nervosa group on dieting and weight preoccupation, but the second cluster was elevated only on scores of body dissatisfaction, drive for
1416
J.L. Miller and T. Vaillancourt
Fig. 91.3 Prevalence of eating disorder symptoms by type (behavioral or psychological). These data are based on results from Miller et al. (2009). No symptoms = 13.1%; Psychological symptoms only = 41%. This includes moderate to high levels of endorsement of body dissatisfaction and fears of gaining weight; Behavioral symptoms only = 0%; Psychological and Behavioral symptoms = 44%. This group includes individuals who endorsed moderate to high levels of psychological symptoms with moderate to high levels of behavioral symptoms. Note that only 2.2% of the sample endorsed both high levels of psychological and high levels of behavioral symptoms
thinness, and perfectionism. This weight-preoccupied-only group had low scores on all other EDI subscales including the bulimia subscale which contains items pertaining to binge eating and purging (core eating disorder behaviors). Although these results by Garner et al. (1984) suggest that eating disorder behaviors were low or absent in this second cluster of weight-preoccupied women, it is important to note that binge eating and purging do not represent all eating disorder behaviors and these women may have been elevated on other eating behaviors, such as caloric restriction leading to weight loss. Indeed, both clusters of weight-preoccupied women scored high on the drive for thinness subscale which includes eating behaviors related to dieting and/or restriction of food intake. However, it is important to keep in mind that measures of dieting do not necessarily represent or capture the caloric restriction that leads to weight loss in anorexia nervosa. The relation between dieting and restriction is addressed in more detail at the end of this section. Garner et al. (1984) concluded that their results supported the notion that some weight-preoccupied women resemble individuals with clinical eating disorders, and are likely a subclinical variant of a clinical syndrome, but that not all weight-preoccupied women resemble eating disorder populations. In fact, Garner and colleagues (1984) were the first to point out that the pursuit of thinness is not necessarily associated with psychopathology. The main difference between weight-preoccupied women who do and who do not resemble an eating disorder population is in whether or not they engage in clinical eating disorder behaviors. This is difficult to establish by examining existing research because measures of eating disorder symptoms have never clearly differentiated thoughts from behaviors. Even the studies by Garner et al. do not distinguish between behavioral and psychological factors since they measured symptoms using the EDI, where thought and behavior items are mixed within the same subscale. Eating disorder behaviors do occur in non-clinical populations, albeit at low frequencies, and in conjunction with the psychological symptoms of an eating disorder. For instance, quoting the same research studies as before, statistics from Project EAT note 9.4% of girls and 13.5% of boys engage
91 Rethinking the Eating Disorder Continuum: A Categorical Approach to Abnormal Eating
1417
Table 91.2 Key features of a continuous-discontinuous model of eating disorders 1. Subclinical and clinical eating disorders are on a continuum with one another with respect to the behavioral and psychological features of an eating disorder. In other words, they are different only in degree of symptoms but not in the kind of symptoms. 2. Subclinical and clinical eating disorders are discontinuous with nonclinical populations on the behavioral criteria of an eating disorder, but continuous with nonclinical populations with respect to the psychological symptoms of an eating disorder. 3. Subclinical and clinical eating disorders are discontinuous with partially recovered and recovered populations on the behavioral criteria of an eating disorder, but on a continuum with partially recovered and recovered populations with respect to the psychological symptoms of an eating disorder.
in recurrent purging behaviors such as vomiting, laxative abuse, or excessive exercise (Ackard et al. 2007); Chamay-Weber et al. (2005) report purging behaviors (vomiting, laxatives, and diuretics) between 5% and 16%; and McVey et al. (2004) found 3.9% of their sample endorsed binge eating and 1.5% endorsed self-induced vomiting. Importantly, while studies have reported high percentages of individuals who endorse the psychological symptoms of an eating disorder without the behavioral symptoms, there are no studies that have reported high percentages (or low percentages) of individuals who are engaged in eating disorder behaviors without the psychological symptoms. We would argue that the presence of these pathological eating disorder behaviors in the samples described by Ackard et al., Chamay-Weber et al., and McVey et al. are signaling an undiagnosed subclinical or clinical eating disorder. According to Stice, Ziemba, Margolis, and Flick (1996), the discontinuity perspective of eating disorders would be supported if research were to show that the same variables that distinguish nonclinical groups from subclinical groups fail to distinguish between subclinical and clinical populations. There is ample research evidence to support behaviors failing to distinguish subclinical and clinical populations. Subclinical and clinical groups both engage in eating disorder behaviors, and even though they may differ in the frequency with which they engage in these behaviors, current research does not support evidence of any meaningful distinction between those engaged in behaviors once a week versus twice a week (Crow et al. 2002; Garfinkel et al. 1995). In fact, the accompanying pathology of those exhibiting subclinical symptoms of eating disorders has been shown to resemble the pathology observed in individuals with a full-blown eating disorder (Garfinkel et al. 1995; Fairburn et al. 2007; Zaider et al. 2000). And, the preponderance of those meeting criteria for a diagnosis of “eating disorder not-otherwise-specified” (EDNOS; approximately 60% of cases) suggests that the frequency level of eating disorder behaviors is less critical than the actual presence of eating disorder behaviors (Table 91.2) (Fairburn and Bohn 2005; Fairburn and Cooper 2007; Wade et al. 2006).
91.3 The Dieting Conundrum We have argued that the behaviors outlined in the DSM criteria for eating disorders are the behaviors that are discrete – food restriction or food avoidance, binge eating, and compensatory or noncompensatory strategies. However, how these behaviors are defined and measured are critical aspects of understanding the continuous or discontinuous nature of these symptoms. Dieting and exercising are prime examples of the difficulty in applying a categorical framework to eating disorders. Dieting and exercising are both behaviors that are not necessarily pathological although they can be and the frequency of the behavior is not a good indicator of whether this behavior is pathological. Dieting, like exercise, does not necessarily increase in pathology the more you engage in the behavior and although
1418
J.L. Miller and T. Vaillancourt
it is unhealthy to be constantly going on and off of a diet, it is not pathological unless it occurs with or leads to other clinical behaviors such as binge eating, significant weight loss, or other psychological aspects such as distress, dysfunction, or disability. Dieting is considered a clinically relevant behavior for two reasons; its role as a risk factor and its potential role as a restricting feature of an eating disorder, and often the distinction between the two becomes blurred by inconsistent definitions of what constitutes a “diet.” Both prospective and cross-sectional research indicates eating disorders commonly begin with behaviors that “resemble” normal dieting (Fairburn and Harrison 2003; Jacobi et al. 2004b). In longitudinal studies, dieting is most clearly demonstrable as a variable risk factor and maintenance factor to the onset of bulimia nervosa and anorexia nervosa binge-purge subtype because of its purported relation to binge eating (Jacobi et al. 2004b; Polivy and Herman 1985). The temporal precedence between dietary restraint and the onset of binge eating supports this assumption (Patton et al. 1990, 1999). For the restricting type of anorexia, dieting is presumed to be a forerunner to the more severe caloric restriction and starvation that produces the abnormally low body weight exhibited by those with anorexia. But dieting is not synonymous with restriction nor is it sufficient in producing the abnormally low body weight typical of anorexia nervosa. If dieting were on a continuum with the restriction characteristic of an eating disorder, rates of anorexia nervosa would be higher, especially given the prevalence of self-reported dieting. In reality, anorexia restricting type is the rarest form of eating disorder (Hoek 2006). In addition, the majority of dieters do not go on to develop an eating disorder, offering support to the role of dieting as a risk factor, rather than a symptom/feature of an eating disorder (Patton et al. 1990, 1999). The relation between dieting and restriction in the eating disorders depends on how dieting is defined. Chronic dieting resulting in consistent caloric restriction accompanied by weight loss is consistent with clinical features of anorexia nervosa, and thus, in this sense, dieting appears to be interchangeable with restriction. Research by Lowe et al. (1996) indicates that dieting is related to eating disorder behaviors (e.g., binge eating) only when accompanied by food restriction and weight loss. In contrast, dieting practices accompanied by “normative” body dissatisfaction may lead to emotional distress, but are not indicative of eating pathology (Lowe et al. 1996). The DSM definition of anorexia nervosa does not describe which eating behaviors are considered abnormal or clinically relevant in terms of what leads to the “refusal to maintain minimally normal body weight.” Weight loss is referred to loosely as the result of a reduction in total food intake and this reduction is often preceded by the exclusion or restriction of certain foods from one’s diet (American Psychiatric Association 2000). Restriction in anorexia nervosa, by definition, always includes caloric deprivation and weight loss – in comparison, dieting can mean cognitive or behavioral deprivation, with or without weight loss. Dieting is most often defined and interpreted as a method of weight control, yet it is not always associated with weight loss (Brownell and Rodin 1994). In fact dieting is a notoriously ineffective means of achieving weight loss according to some researchers because 95% of those who lose weight will regain the weight within a few years and many will gain more than they originally lost (Grodstein et al. 1996; National Institutes of Health Technology Assessment Conference Panel 1993). The term dieting is also used in reference to cognitive restraint or the desire to lose weight and for this reason it seems clear why numerous research studies have found self-reported dieting to be negatively related to actual reductions in caloric intake or weight loss (Lowe 1993). Herman and Polivy (1984) have suggested that dieters may be best characterized by their attempts to lose weight than by actual weight loss. It is possible that subjective measures of dietary restraint do not correspond well with objective measures of weight loss because the dieting attempts have failed. For instance, an individual may engage in dieting behaviors by restricting food intake or lowering caloric intake during the first half of the day, and this we would consider “engaging in dieting behavior.” However, if in the latter half
91 Rethinking the Eating Disorder Continuum: A Categorical Approach to Abnormal Eating
1419
of the day, when the food restriction has left the dieter feeling hungry and they begin to consume more calories than if they had eaten normally throughout the entire day, this overeating in the second half of the day compensates for the lowered food intake earlier in the day and in the end, the balance of the calories taken-in is either the same or possibly more. In this case, objective measures of weight, such as daily or weekly weighing of participants or even more precise assessments of caloric intake recorded during a single meal would fail to correlate with subjective reports of dietary restraint because overall, there has not been any weight loss. This is not the same as saying the individual is not dieting. They are engaged in dieting behaviors but these behaviors have failed to result in weight loss because they were not maintained. Thus if asked to complete a measure of dietary restraint these individuals would likely score high because from their perspective they are in fact dieting. What is not being accounted (objectively) in self-report measures of dieting is whether the dieter has succeeded or failed in their dieting attempts. Restriction in anorexia nervosa is unhealthy and pathological – but dieting is not necessarily unhealthy or pathological. Dieting is a heterogeneous term that includes both healthy (eating less fat, eating more fruits and vegetables, exercising) and unhealthy behaviors (e.g. skipping meals, fasting). According to Lowe and Timko (2004) dieting is neither beneficial nor harmful, but simply ineffective for most individuals in the long-term. Dieting may have beneficial effects if the purpose of the diet is to counteract a predisposition to overeat or gain weight (Lowe and Timko 2004). Studies have found long-term low-calorie diets to be effective in decreasing binge eating in obese individuals with and without an eating disorder with length of follow-up varying from 3 to 24 months post-treatment (for a review, see Stice 2002). Even in a normal weight population, Presnell and Stice (2003) found that when college students lost a small amount of weight in a weight-loss program, their level of bulimic symptoms were reduced not increased after a 3 month follow-up. Lowe and Timko (2004) state that the merits of dieting are inconclusive without first clearly defining what is meant by the term, specifying the purpose of the diet, and toward which population it is intended. These differences between the interpretation of and consequences to dieting versus restriction suggest the two are not synonymous with one another and that despite any overlap in behaviors across those who engage in normative dieting and those who engage in clinical restriction, dieting should not be used as a proxy for restriction unless the instruments used to measure dieting take into account cognitive versus behavioral restraint in conjunction with abnormally low body mass index or measured weight loss. If the purpose is to detect restrictors rather than dieters, it is necessary to separate out the small percentage of successful dieters – 5%, from the unsuccessful dieters – 95% (Grodstein et al. 1996).
91.4 E mpirical Tests of the Latent Factor Structure of Eating Disorder Symptoms Most research on the continuous–discontinuous debate of eating disorders has come from crosssectional investigations using cluster analysis, latent class analysis, discriminant function analysis factor analysis, and most recently, a series of taxometric investigations (see Table 91.3). Taxometric analyses specifically test for dimensional versus categorical constructs in a dataset (Waller and Meehl 1998). A taxon is an underlying entity that drives the relations between common indicators of a disorder, and while the common indicators are dimensional, similar to a latent variable in factor analysis, a taxon indicates a discrete category. Thus, a taxon does not preclude dimensional aspects; however it does indicate a qualitative difference between its underlying construct and other constructs. The utility of this analytic technique has spurred a number of taxometric investigations in the field of eating disorders over the last 5 years.
1420
J.L. Miller and T. Vaillancourt
Table 91.3 Summary of empirical studies on the continuum of eating disorders Statistical test Description Published studies (peer reviewed) Garner et al. (1993), Hay et al. (1996), Cluster analysis • Assignment of a set of observations Stice and Agras (1999), Westen and or objects into clusters/subsets Harden-Fischer (2001), Williamson • Requires an a priori assumption et al. (1992) of the number of clusters/subsets in order to perform the statistical test • Distance between objects used to determine assignment to clusters Bulik et al. (2000), Duncan et al. Latent class analysis • Assignment of a case or subject to a (2007), Keel et al. (2004), group (i.e., latent class) Striegel-Moore et al. (2005), • There is an a priori assumption that Sullivan et al. (1998a, b), Sullivan the latent variables are discrete and Kendler (1998), Wade et al. • Conditional probabilities are used to (2006) determine the likelihood that the set of observed, discrete variables or “cases” will fall into a particular latent class • Observed variables within a class are statistically independent Discriminant • Purpose is to predict group Stice et al. (1996), Lowe et al. (1996), function analysis Franko and Omori (1999), Goldner membership or discriminate between et al. (1999), Stice et al. (1998), two or more mutually exclusive Tylka and Subich (1999) groups from a set of predictors • The dependent variable is group membership Factor analysis • Determine the number of subsets Gleaves and Eberenz (1993, 1995), Gleaves et al. (1993), Tobin et al. (latent factors) that a set of variables (1991), Vanderheyden et al. (1988), can form Williamson et al. (2002) • Subsets are largely independent of one another • Variables within a subset are related to one another Taxometric • Determine whether the latent Williamson et al. (2005), Gleaves et al. analysis (2000a, b), Tylka and Subich (2003) structure contains a taxon/a class in addition to its underlying dimension • There is no a priori assumption about the dimensional or categorical nature of the latent structure (see description in text)
91.4.1 Taxometric Analysis in Clinical Samples Williamson et al. (2002) examined three latent features of eating disorders derived through a series of exploratory and confirmatory factor analyses; binge eating, fear of fatness/compensatory behaviors, and drive for thinness. These latent scores were then used as indicators in a series of taxometric analyses comparing individuals with clinical eating disorders to obese persons without an eating disorder and to a normal weight comparison group. Binge eating disorder and bulimia nervosa were judged to be qualitatively distinct (taxonic) from both the normal-weight comparison group and the obese comparison group. In contrast, taxometric analyses did not reveal strong support for anorexia nervosa being qualitatively distinct from normalcy (i.e., not taxonic). Excluding noneating disorder groups, Williamson et al. went on to compare the categorical versus dimensional characteristics
91 Rethinking the Eating Disorder Continuum: A Categorical Approach to Abnormal Eating
1421
within the eating disorders (i.e., anorexia nervosa, bulimia nervosa, and binge eating disorder) and concluded that while there was mixed support for a categorical model of eating disorder subtypes, some of the features appeared categorical and some appeared dimensional. Taken together, the taxometric results reported by Williamson et al. support bingeing and purging disorders being taxonic (i.e., categorical), especially when using a mixed sample (clinical and non-clinical), while anorexia nervosa appears to be more dimensional. These findings, that certain features of eating disorders are discrete from nonclinical populations while others are continuous and certain features within the eating disorders are discrete while others are continuous, are consistent with the three existing taxometric studies conducted on eating disorder populations to date (Gleaves et al. 2000a, 2000b; Tylka and Subich 2003). How might we interpret these findings? If we examine these results from a framework of continuous eating disorder thoughts and discontinuous eating disorder behaviors we can make clearer the inconsistencies reported across taxometric studies. Bingeing and purging disorders (bulimia nervosa and anorexia nervosa binge-purge subtype) may appear taxonic especially in mixed samples because the central characteristic of these disorders are binge eating and purging which incidentally, occur at exceedingly low base rates in nonclinical populations. Bingeing and purging are extreme behaviors that are specific to eating disorder populations; thus when comparing clinical and nonclinical groups, rates of bingeing and purging will be very high in clinical populations and very low in normal populations, increasing the likelihood of detecting a taxon. In contrast, restricting behaviors (anorexia-restricting subtype) are often measured using dieting items (e.g., Drive for Thinness subscale of the EDI) which are common among nonclinical populations (Chamay-Weber et al. 2005). In general, when restriction is measured in terms of dieting there will be more people in both clinical and nonclinical populations who will endorse items such as, I eat diet foods, than items related to bingeing or purging, (i.e., I vomit after I have eaten). The range of restricting behaviors will be reflected by a more continuous process in taxometric analyses compared to bingeing and purging where the limited variability will result in stronger evidence for a taxon. Williamson et al. (2005) proposed a three-dimensional model that argued for binge eating being taxonic while fear of fatness/compensatory strategies, and drive for thinness were continuous. However, these results may be an artifact of including restricting items that range from somewhat normative behaviors (skipping breakfast, periods of fasting, eating diet foods) to more extreme behaviors (prolonged periods of starvation and chronic lowering of caloric intake) in taxometric studies. We suspect that if Williamson et al. were to re-specify their model by only including extreme restricting items that reflect weight loss, they may find restricting behaviors to appear taxonic. In addition, the continuous nature of this fear of fatness-compensatory strategies may be confounded by the combination of thought and behavioral items in these analyses. For example, latent factors for fear of fatness have typically combined items pertaining to thoughts (preoccupation with weight and shape) with items of behaviors (caloric limitation) within the same latent class (e.g. Keel et al. 2004). Therefore, conclusions regarding the continuous nature of restricting behaviors may be premature.
91.4.2 Taxometric Analysis in Nonclinical Samples Taxometric analyses on noneating disordered populations can also be understood within a thoughtbehavior framework. Tylka and Subich (2003) examined the latent structure of eating disorders by performing taxometric analyses on a sample of 532 college women. They included items representative of nonbehavioral features of eating disorders such as body dissatisfaction, and found five distinct factors, all indicative of a dimensional solution. Specifically, taxometric analyses revealed no presence of
1422
J.L. Miller and T. Vaillancourt
a taxon in the data, although importantly, Tylka and Subich did not include any behavioral indicators of eating disorders in their model. These authors argue that much of the opposing findings in the literature surrounding the existence of a continuous versus a discontinuous model of eating disorders results from the fact that different indicators are used by different researchers. While some researchers use behavioral indicators (binge eating, purging, restricting) with nonbehavioral indicators (interoceptive awareness, maturity fears, or body dissatisfaction), other researchers use one but not the other. When researchers utilize more behavioral indictors of eating disorders than nonbehavioral indicators, results support categorical models of eating disorders (Gleaves et al. 2000a, b; Tylka and Subich 2003; Lowe et al. 1996), whereas studies that use more nonbehavioral indicators find more evidence for dimensional models of eating disorders (Tylka and Subich 1999, 2002, 2003). Tylka and Subich (2003) argue that bingeing and purging should not be included in research aimed at uncovering the latent structure of eating disorders because these indicators may either inflate the notion of a categorical model or mask the presence of dimensional models of eating disorders. These authors suggest that one reason for this inflation may be that some researchers classify females in their sample as clinical or nonclinical based on bingeing and purging behaviors only to turn around and use these same behaviors as the criterion variable. Evidence in support of a taxon in the data may be artificial because of confounding indicators with the criterion (2003). However, while Tylka and Subich suggest excluding behavioral symptoms of eating disorders, behavioral indicators are fundamental attributes of individuals with eating disorders and must be considered in the continuum debate. Evidence for both categorical and dimensional models in taxometric studies are not conflicting findings; they are in fact telling us something critical about the nature of eating disorders.
91.5 P sychological Versus Behavioral Symptoms in Recovery from an Eating Disorder A number of researchers have found the rate and timing of eating disorder recovery to vary depending on the presence or absence of psychological criteria in the definition of recovery (Cogley and Keel 2003; Couturier and Lock 2006; Fennig et al. 2002; Saccoman et al. 1989; Strober et al. 1997). In general, psychological recovery occurs more slowly, extending years beyond the cessation of behavioral symptoms and as a result, rates of recovery vary depending on the length of remission studied and the variables included in definitions. Saccoman et al. (1989) found that when only the physical aspects of an eating disorder (restricting, bingeing, purging) were considered, 79% of those with anorexia in their study had recovered. Yet when the psychological criteria were examined, this recovery rate fell to 48%. Strober et al. (1997), report recovery rates based on normal weight and regular menstruation to be 86% occurring on average 57.4 months from illness onset. But in the same study, recovery based on psychological criteria was only 76%; occurring an average of 79.1 months from illness onset. Similarly, Fennig et al. (2002) found the psychosocial recovery to take on average 6 years longer than the physical (i.e., behavioral) recovery from an eating disorder. Bachner-Melman, Zohar, and Ebstein (2006) compared women behaviorally but not cognitively (i.e., lack of body image distortion or fear of weight gain) recovered from anorexia nervosa to a group of women recovered both cognitively and behaviorally and found that the symptoms and personality profile of the behaviorally recovered group showed residual features of anorexia whereas the cognitively and behaviorally recovered were indistinguishable from controls. It seems that while behavioral recovery is critical for initial symptom remission, psychological recovery is necessary for
91 Rethinking the Eating Disorder Continuum: A Categorical Approach to Abnormal Eating
1423
Table 91.4 Key differences in behavioral versus psychological recovery from an eating disorder 1. Behavioral recovery occurs earliest. 2. Recovery rates are highest when recovery is defined by the absence of the behavioral symptoms (i.e., amenorrhea, bingeing, purging, restricting). 3. Psychological recovery (the absence of extreme weight and shape overevaluation and body dissatisfaction, and fears of weight gain/being fat) occurs slowly, and takes an average of 2 years longer than behavioral recovery. 4. One-third of those with anorexia nervosa will continue to struggle with the psychological symptoms for many more years post-treatment. 5. The presence of psychological symptoms at post-treatment is the best predictor of relapse. 6. Approximately 30–50% of those with bulimia nervosa will continue to struggle with a high degree of body shape and weight concerns following treatment.
longer-term maintenance of recovery from eating disorders (Couturier and Lock 2006; see Table 91.4). In a similar study, Cogley and Keel (2003) evaluated the concurrent validity of requiring remission of ”undue influence of weight and shape on self-evaluation” in defining recovery from bulimia nervosa. Three groups were compared: 31 women fully recovered from bulimia, 28 women behaviorally recovered only, and 59 matched controls. Participants completed measures of mood, anxiety, psychosocial functioning, and body dissatisfaction. Results showed no differences between matched controls and the fully recovered individuals with bulimia on any measures, while the behaviorally recovered only group showed significant pathology across all reported measures. Of particular importance in the distinction between behavioral and psychological recovery are the findings by researchers that body dissatisfaction and overvaluation are the best predictors of relapse. For instance, Fairburn, Peveler, Jones, Hope, and Doll (1993) and Freeman, Beach, Davis, and Solyom (1985) found the residual level of shape concern that remained at the end of “successful” treatment was the strongest predictor of relapse among patients with bulimia nervosa. These findings are disconcerting given the high rate of shape concern remaining among individuals who have completed treatment successfully; as many as 30–50% according to some estimates (Farrell et al. 2005). Keel, Dorer, Franko, Jackson, and Herzog (2005) found similar results in their prospective study of predictors of relapse. Specifically, Keel et al. report that body image disturbance was the best predictor of relapse at 9 years post-treatment across both bulimia and anorexia, and this was across a range of relapse predictors including behavioral, psychological, and Axis I and II comorbid conditions. The overvaluation of shape and weight appears among individuals in nonclinical populations, clinical populations, and recovered or partially recovered populations. In contrast, behaviors are absent in nonclinical groups, acutely present in clinical groups, and absent or remitted in recovered or partially recovered groups. This ongoing battle for psychological recovery across the course of the eating disorder offers support for a continuum of psychological symptoms, while disordered behaviors are more specific to the active phase of the illness where individuals are still meeting diagnostic criteria for an eating disorder.
91.6 Summary The aim of this review was to offer support toward the notion of a combined categorical-dimensional model of eating disorders by demonstrating that it is the psychological symptoms of eating disorders that are dimensional and the behavioral symptoms that are categorical. Three areas of research were reviewed to offer support for this hypothesis: (1) eating disorder thoughts are so prevalent that they
1424
J.L. Miller and T. Vaillancourt
poorly discriminate between nonclinical and clinical groups; but behaviors are excellent markers of clinical groups because they are specific to clinical populations; (2) empirical studies of the latent factor structure of eating disorders show dimensional models when using psychological indicators, but categorical models when using behavioral indicators; (3) recovery rates for eating disorders are highest when defined by the absence of behavioral symptoms and lowest when defined by psychological symptoms, as the psychological symptoms fail to distinguish recovered individuals from ill or partially recovered individuals. Now we turn our attention to the potential implications of a categorical model of eating disorder behaviors.
91.7 Clinical and Research Implications Advancing our understanding of the continuum of eating pathology has both research and clinical implications. First, understanding the nature of the relationship between disordered eating and eating disorders advances the theoretical underpinning of eating disorder research, as classification of mental illness is at the heart of all psychiatric diagnoses and subsequent treatment protocols. Establishing valid categories for eating disorders is clinically important as it enables accurate diagnostic instruments to be used and creates standards for diagnosis which is an essential part of interpretation and communication among clinicians and researchers. Addressing the question of continuity versus discontinuity in eating disorders will inform our measurement of eating disorder symptoms in at-risk groups. Improving our measurement in epidemiological research will help reduce the number of false positives in two-stage screening studies, increasing efficiency in research protocols, reducing costs from unnecessary testing, lessen participant burden incurred by lengthy interviews, and free up valuable professional resources that can be directed toward prevention and treatment (Miller 2008).
91.8 Applications to Other Areas of Health and Disease 91.8.1 Personality and Individual Differences in Eating Disorders It is important to highlight that we are not suggesting eating disorder behaviors are the only difference between clinical and nonclinical populations, rather we would argue in a similar vein to Crisp (1965) and Bruch (1973), and as well Polivy and Herman (1987), that those with eating disorders have fundamental differences in personality, such as ego deficits or psychopathology such as comorbid mood disorder. Engaging in bingeing, purging, or restricting behaviors is both mentally and physically painful, and individuals who are able to engage in these aberrant behaviors likely have underlying personality features and/or brain neurochemistry that are very different from those who do not engage in these behaviors (e.g., Miller et al. 2006). For instance, impulsivity, characteristic of some individuals with bulimia (Bulik et al. 1995) may facilitate the disinhibited eating of a binge. Impulsivity is also associated with substance abuse, and the use of alcohol or other drugs among individuals with eating disorders has been linked to frequency of binges and purges (Davis and Claridge 1998). Concurrent mood disorder may facilitate restricting or bingeing behaviors as a result of changes to brain neurochemistry which can disrupt appetite and/or feelings of satiety (Halmi and Sunday 1991). Additionally, neuroticism has been linked to
91 Rethinking the Eating Disorder Continuum: A Categorical Approach to Abnormal Eating
1425
s ymptoms of eating disorders (Davis and Claridge 1998; Miller et al. 2006), with the suggestion that fear of weight gain may enable the highly neurotic individual to restrict food intake more easily than those low in neuroticism. Herman and Polivy’s (1984) boundary model of eating behavior postulates that individuals with anorexia and bulimia have a lowered hunger boundary that allows them to tolerate greater degrees of food deprivation than normal. This “hunger boundary” may explain the ability of individuals with anorexia (restricting type) to maintain consumption at such stringent levels (Herman and Polivy 1984). Although, the authors concede that the degree to which this hunger boundary is imposed by environmental factors, a genetic predisposition, or both, is uncertain. Individuals who purge may differ from those who restrict because of biological differences that may drive the choice of a particular coping mechanism (the behavior) in dealing with the shared psychopathology (the thoughts: overvaluation of shape and weight) of eating disorders. Thus, it may be beneficial to focus future research on understanding potential executive functioning deficits associated with the behaviors of an eating disorder. For example, reward deficiency syndrome has been put forth as an explanation for other compulsive, impulsive, and addictive disorders based on a common genetic deficiency in the dopamine D2 receptor (Blum et al. 1996). Dopamine is not the only neurotransmitter implicated in the reward system, serotonin is also known to be involved (Blum et al. 1996) and there is already evidence to support the role of serotonin imbalances in the eating disorders, bulimia nervosa in particular (Steiger et al. 2003).
91.8.2 Measurement of Eating Disorders Early identification of eating disorders is critical, as mortality rates increase linearly with duration of illness (Hoek 2006) and recovery rates are linked to symptom severity at the onset of treatment (Bulik et al. 1999). In general, recovery rates vary depending on the definition of recovery, but a substantial number of individuals, especially those with anorexia, will remain chronic, will relapse following treatment, or will continue to struggle with psychological symptoms (e.g., fear of weight gain, body dissatisfaction) despite achieving physical recovery (Strober et al. 1997). For these reasons, prevention and early detection is critical. Detection of undiagnosed cases of eating disorders in community settings requires epidemiological research; however the usefulness of such studies is highly dependent on valid measures and clearly defined indicators of risk. The most rigorous method for identifying cases of eating disorders in those who do not seek medical treatment voluntarily is to employ a two-stage screening study. A two-stage screening study is considered the gold standard in epidemiological research on eating disorders (Jacobi et al. 2004a). The two-stage screen typically involves administering a questionnaire, such as the Eating Attitudes Test (EAT-26) to a large sample of at-risk individuals (e.g., adolescent females) and identifying the subset of individuals who score above a predetermined threshold on the questionnaire (e.g., 20 or above on the EAT-26) to return for the second stage of the screen: the clinical interview. The EAT is the most widely used screen for detecting eating disorders in epidemiological studies, although its primary use has been for the detection of cases of anorexia nervosa, and not bulimia nervosa (Jacobi et al. 2004a). However, relying on screening measures such as the EAT for detecting undiagnosed eating disorders will be costly and time consuming because most existing screening tools, including the EAT, contain far more normative weight concern items and dieting items than clinically significant eating disorder behaviors and, importantly, it is the eating disorder behaviors which are pathognomonic. The presence of clinically significant eating
1426
J.L. Miller and T. Vaillancourt
disorder behaviors is the best indicator that the disorder is present. While the psychological symptoms of an eating disorder will always identify all true cases of eating disorders, as long as the criteria for an eating disorder continues to be partially defined by disturbances in body image and fears of weight gain, the psychological symptoms will also identify far too many noncases because these symptoms are normative. Given the level of normative discontent concerning weight and shape in our culture it would seem prudent to examine other measures of eating pathology that have higher specificity.
91.9 Conclusion The goal of this chapter was to introduce a new framework for understanding the inconsistent findings surrounding the continuous or discontinuous nature of eating disorders. If we scrutinize the many studies conducted over the last few decades that have found evidence in favor of either discrete or continuous models of eating disorders, we see that the competing results of most studies are accurate when we consider the sample composition and the features of eating disorders that were examined. Tylka and Subich (2003) argued that many studies find evidence for categorical models when using behavioral features of eating disorders and more dimensional models emerge when using nonbehavioral features; we would agree since it is precisely the behavioral features that appear categorical, not the disorder itself. When researchers examine clinical populations, or mixed samples, there will be more evidence for categorical models because the clinical groups will have more of the disordered behaviors, whereas a truly nonclinical population will mostly be comprised of a continuum of disordered thinkers who do not engage in disordered behaviors (Table 91.5). Regardless of whether future research supports or refutes the notion of taxonicity in disordered behaviors and/or continuity in disordered thoughts, it is important to highlight the significance and utility of examining the role of psychological symptoms in the eating disorders independently from the behavioral symptoms, as this distinction holds both practical and theoretical implications to the field as a whole.
Table 91.5 Population-based screening of eating disorders: best practice recommendations 1. Body dissatisfaction, fear of fat, or other items that tap body image concerns (psychological symptoms) should not be used to screen for the presence of anorexia nervosa or bulimia nervosa given the high level of endorsement made by women who do not have an eating disorder. Screening with these psychological symptoms will result in too many false positives (poor specificity). 2. If psychological symptoms are included in population-based screening for eating disorders, an extreme threshold should be used as a cut-off point and they should be measured and interpreted separate from the eating disorder behaviors (i.e., psychological symptoms should have a separate total score form behavioral symptoms). 3. Items that assess pathological eating behaviors such as purging (vomiting, laxative/diuretic abuse), restricting, and binge eating will have the highest diagnostic accuracy (specificity and sensitivity). 4. Dieting questions should not be used to assess restricting behavior (e.g., “I eat diet foods” from the EAT-26) because it will increase the number of false positives. 5. Binge eating questions should include a definition of a “binge” (e.g., see the Eating Disorder Examination Questionnaire; EDE-Q) and be assessed along with relevant compensatory behaviors, given the subjective interpretation of binge eating. In the absence of compensatory behavior, such as in binge eating disorder, screening for binge eating behavior will be very difficult and requires further research to understand how and what to ask when screening for binge eating behaviors.
91 Rethinking the Eating Disorder Continuum: A Categorical Approach to Abnormal Eating
1427
Summary Points • Addressing the issue of a continuum of eating pathology is necessary because understanding the nature of the relation between disordered eating and eating disorders advances the theoretical underpinning of eating disorder research (i.e., the etiology of eating disorders). • Understanding whether eating disorders exist on a continuum or are discontinuous informs our measurement and assessment of eating disorders. • It is important to distinguish between psychological and behavioral symptoms in the assessment and measurement of eating disorders. Screening instruments that rely heavily on psychological indicators of eating disorders will yield high rates of false positives and will suffer a trade-off of low specificity for high sensitivity. • The presence of clinically significant eating disorder behaviors is the best indicator that an eating disorder is present. It is the behavior which is pathognomonic. • Because weight discontent is so normative in western culture, it is problematic, from a measurement standpoint, to use these normal features to screen for an abnormal syndrome. • Screening for eating disorders using instruments such as the EDI or the Eating Attitudes Test (EAT26) will be a costly and time-consuming effort because both measures contain far more normative weight concern items and dieting items than clinically significant eating disorder behaviors.
Definitions and Explanations Behavioral symptoms: The behavioral symptoms of an eating disorder, as outlined in the DSM-IV include: a) binge eating and b) compensatory behaviors to prevent weight gain such as self-induced vomiting; misuse of laxatives, diuretics, enemas, or other medications; fasting; or excessive exercise. Psychological symptoms: The psychological symptoms of an eating disorder, as outlined in the DSM-IV include intense fear of gaining weight or becoming fat, even though underweight; disturbance in the way in which one’s body weight or shape is experienced, undue influence of body weight or shape on self-evaluation, denial of the seriousness of the current low body weight. Population-based screening: Population-based screening for eating disorders is a method used in epidemiological research where a large and ideally representative sample of people from a nonclinical population are administered a short survey or interview that identifies those who are at-risk of having an eating disorder. Those who are determined to be at-risk are followed up with a more in-depth interview to determine whether they meet diagnosis for a clinical eating disorder. Sensitivity: Sensitivity of an eating disorder measurement tool or diagnostic test refers to how good a test is at correctly identifying people who have an eating disorder. Specificity: Specificity of an eating disorder measurement tool or diagnostic test refers to how good a test is at correctly identifying people who do not have an eating disorder. Taxon: A taxon is like a category; it indicates a discrete group where in/out classifications can be made. However, the presence of a taxon does not mean there is no dimensional aspect. With eating disorders, if eating behaviors were taxonic, they could be measured on a continuum. The discrete part of the behavior comes when we compare clinical to nonclinical populations. A “dimension’ of eating behaviors exits in the clinical sample (dimensional aspect), but this dimension is only present in the clinical sample; it is unique or specific to the clinical population. This is the categorical aspect.
1428
J.L. Miller and T. Vaillancourt
References Ackard DM, Fulkerson JA, Neumark-Sztainer D. Int J Eat Disord. 2007;40:409–17. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 3rd edn. Washington, DC: APA; 1980. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4th edn. Washington, DC: APA; 1994. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4th edn., text revision. Washington, DC: APA; 2000. Bachner-Melman R, Zohar AH, Ebstein RP. J Nerv Ment Dis. 2006;194:697–703. Blum K, Sheridan PJ, Wood RC, Braverman ER, Chen TJH, Cull JG. J R Soc Med. 1996;89:396–400. Brownell KD, Rodin J. Am Psychol. 1994;49:781–91. Bruch H. Eating disorders. Basic Books, New York; 1973. Bulik CM, Sullivan PF, Kendler KS. Am J Psychiatry. 2000;157:886–95. Bulik CM, Sullivan PF, Joyce PR, Carter FA. J Nerv Ment Dis. 1995;183:593–8. Bulik CM, Sullivan PF, Carter FA, McIntosh VV, Joyce PR. Int J Eat Disord. 1999;26:137–44. Button EJ, Whitehouse A. Psychol Med. 1981;11:509–16. Cash TF, Morrow JA, Hrabosky JI, Perry AA. J Consult Clin Psychol. 2004;72:1081–9. Chamay-Weber C, Narring F, Michaud P. J Adolesc Health. 2005;37:417–27. Cogley CB, Keel PK. Int J Eat Disord. 2003;34:200–10. Cooper PJ, Goodyer I. Br J Psychiatry. 1997;171:42–4. Couturier J, Lock J. Int J Eat Disord. 2006;39:550–5. Crisp AH. Proc R Soc Med. 1965;11:509–16. Crow SJ, Agras WS, Halmi K, Mitchell JE, Kraemer HC. Int J Eat Disord. 2002;32:309–18. Davis C, Claridge G. Addict Behav. 1998;23:463–75. Duncan AE, Bucholz KK, Neuman RJ, Agrawal A, Madden PAF, Heath A. Psychol Med. 2007;37:1097–107. Fairburn CG, Bohn K. Behav Res Ther. 2005;43:691–701. Fairburn CG, Cooper Z. Int J Eat Disord. 2007;40:S107–10. Fairburn CG, Cooper Z, Bohn K, O’Connor ME, Doll HA, Palmer RL. Behav Res Ther. 2007;45:1705–15. Fairburn CG, Cooper Z, Shafran R. Behav Res Ther. 2003;41:509–28. Fairburn CG, Harrison PJ. Lancet. 2003;361:407–16. Fairburn CG, Peveler RC, Jones R, Hope RA, Doll HA. J Consult Clin Psychol. 1993;61:696–8. Farrell C, Shafran R, Lee M, Fairburn CG. Behav Cogn Psychother. 2005;33:189–200. Fennig S, Fennig S, Roe D. Gen Hosp Psychiatry. 2002;24:87–92. Franko DL, Omori M. J Adolesc. 1999;22:389–96. Freeman RJ, Beach B, Davis R, Solyom L. Psychiatry Res. 1985;19:349–53. Fries H. In: Vigersky R editor. Anorexia nervosa. New York: Raven; 1977. p. 163–176. Gangestad S, Snyder M. Psychol Rev. 1985;92:317–49. Garfinkel PE, Lin E, Goering P, Spegg C, Goldbloom DS, Kennedy S, Kaplan AS, Woodside DB. Am J Psychiatry. 1995;152:1052–8. Garner DM, Garner MV, Rosen LW. Int J Eat Disord. 1993;13:171–85. Garner DM, Olmsted MP, Garfinkel PE. Int J Eat Disord. 1983;2:11–20. Garner DM, Olmsted MP, Polivy J, Garfinkel PE. Psychosom Med. 1984;46:255–66. Gleaves DH, Brown JD, Warren CS. Behav Modif. 2004;28:739–62. Gleaves DH, Eberenz K. J Psychopathol Behav Assess. 1993;15:141–52. Gleaves DH, Eberenz KP. J Clin Psychol. 1995;51:181–9. Gleaves DH, Lowe MR, Green BA, Cororve MB, Williams TL. Behav Ther. 2000;31:195–219. Gleaves DH, Lowe MR, Snow AC, Green BA, Murphy-Eberenz KP. J Abnorm Psychol. 2000;109:56–68. Gleaves DH, Williamson DA, Barker SE. J Abnorm Psychol. 1993;102:173–6. Goldner EM, Srikameswaran S, Schroeder ML, Livesley WJ, Birmingham CL. Psychiatry Res. 1999;85:151–9. Grodstein F, Levine R, Spencer T, Colditz GA, Stampfer MJ. Arch Intern Med. 1996;156:1302–6. Halmi KA, Sunday SR. Appetite. 1991;16:219–37. Hay PJ, Fairburn CG, Doll HA. Psychol Med. 1996;26:801–12. Herman CP, Polivy J. In: Stunkard AJ, Stellar E editors. Eating and its disorders. New York: Raven Press; 1984. p. 141–156. Hoek HW. Curr Opin Psychiatry. 2006;19:389–94. Jacobi C, Abascal L, Taylor CB. Int J Eat Disord. 2004;36:280–95. Jacobi C, Hayward C, de Zwaan M, Kraemer HC, Agras WS. Psychol Bull. 2004;130:19–65. Keel PK, Dorer DJ, Franko DL, Jackson SC, Herzog DB. Am J Psychiatry. 2005;16:2263–8.
91 Rethinking the Eating Disorder Continuum: A Categorical Approach to Abnormal Eating
1429
Keel PK, Fichter M, Quadflieg N, Bulik CM, Baxter MG, Thornton L, Halmi KA, Kaplan AS, Strober M, Woodside DB, Crow SJ, Mitchell JE, Rotondo A, Mauri M, Cassano G, Treasure J, Goldman D, Berrettini WH, Kaye WH. Arch Gen Psychiatry. 2004;61:192–200. Kraepelin E. General Paresis. New York: Nervous and Mental Disease Publishing Company; 1913. Lowe MR. Psychol Bull. 1993;114:100–21. Lowe MR, Gleaves DH, DiSimone-Weiss RT, Furgueson C, Gayda CA, Kolsky PA, Neal-Walden T, Nelsen LA, McKinney S. J Abnorm Psychol. 1996;105:508–17. Lowe MR, Timko CA. Br J Nutr. 2004;92:S19–22. McVey G, Tweed S, Blackmore E. CMAJ. 2004;170:1559–61. Miller JL. Measurement of eating pathology: Distinct roles of thoughts and behaviours in the assessment of risk and detection of eating disorders. Unpublished doctoral dissertation, McMaster University, Canada; 2008. Miller JL, Schmidt LA, Vaillancourt T, McDougall P, Laliberté M. Eat Behav. 2006;7:69–78. Miller JL, Vaillancourt T, Hanna SE. Eat Behav. 2009;10:89–96. National Institutes of Health Technology Assessment Conference Panel. Ann Intern Med. 1993;199:764–70. Park J, Beaudet MP. Eur Eat Disord Rev. 2007;15:311–20. Patton GC, Johnson-Sabine E, Wood K, Mann AH, Wakeling A. Psychol Med. 1990;20:383–94. Patton GC, Selzer R, Coffey C, Carlin JB, Wolfe R. BMJ. 1999;318:765–8. Piran N, Gadalla T. Addiction. 2006;102:105–13. Polivy J, Herman CP. Am Psychol. 1985;40:193–210. Polivy J, Herman CP. J Consult Clin Psychol. 1987;55:635–44. Polivy J, Herman CP. Annu Rev Psychol. 2002;53:187–213. Presnell K, Stice E. J Abnorm Psychol. 2003;112:166–70. Rodin J, Silberstein L, Striegel-Moore R. Nebr Symp Motiv. 1984;32:267–307. Ruderman AJ, Besbeas M. J Abnorm Psychol. 1992;101:383–90. Saccoman L, Savoini M, Naselli A, Cirrincione M, Matricardi A. Minerva Pediatr. 1989;41:5–10. Steiger H, Israel M, Gauvin L, Ng YK, Young SN. Psychiatry Res. 2003;120:219–29. Stice E. Psychol Bull. 2002;128:825–48. Stice E, Agras WS. J Consult Clin Psychol. 1999;67:460–9. Stice E, Killen JD, Hayward C, Taylor CB. J Consult Clin Psychol. 1998;66:784–90. Stice E, Ziemba C, Margolis J, Flick P. Behav Ther. 1996;27:531–49. Striegel-Moore R, Franko DL, Thompson D, Barton B, Schreiber GB, Daniels SR. Psychol Med. 2005;35:1563–72. Striegel-Moore RH, Silberstein LR, Rodin J. Am Psychol. 1986;41:246–63. Strober M, Freeman R, Morrell W. Int J Eat Disord. 1997;22:339–60. Sullivan PF, Bulik CM, Kendler KS. Br J Psychiatry. 1998a;173:75–9. Sullivan PF, Bulik CM, Kendler KS. Psychol Med. 1998b;28:599–610. Sullivan PF, Kendler KS. Br J Psychiatry. 1998;173:312–9. Tobin DL, Johnson C, Steinberg S, Staats M. J Abnorm Psychol. 1991;100:14–21. Tylka TL, Subich LM. J Couns Psychol. 1999;46:268–76. Tylka TL, Subich LM. J Couns Psychol. 2002;49:273–9. Tylka TL, Subich LM. J Counsel Psychol. 2003;50:276–86. Vanderheyden DA, Fekken GC, Boland FJ. Int J Eat Disord. 1988;7:321–9. Wade TD, Crosby RD, Martin NG. Arch Gen Psychiatry. 2006;63:1377–84. Waller NG, Meehl PE. Multivariate taxometric procedures: Distinguishing types from continua. Thousand Oaks: Sage Publications; 1998. Watson D, Clark LA. Clin Psychol Sci Pract. 2006;13:210–5. Westen D, Harnden-Fischer J. Am J Psychiatry. 2001;158:547–62. Widiger TA, Samuel DB. J Abnorm Psychol. 2005;114:494–504. Wilfley DE, Bishop ME, Wilson GT, Agras WS. Int J Eat Disord. 2007;40:S123–9. Williamson DA, Gleaves DH, Savin SS. J Psychopathol Behav Assess. 1992;14:201–16. Williamson DA, Gleaves DH, Stewart TM. Int J Eat Disord. 2005;37:1–10. Williamson DA, Womble LG, Smeets MAM, Netemeyer RG, Thaw JM, Kutlesic V, Gleaves DH. Am J Psychiatry. 2002;159:412–8. Zaider TI, Johnson G, Cockell SJ. Int J Eat Disord. 2000;28:58–67.
wwwwwwwwwwwwwww
Chapter 92
Cued Overeating Anita Jansen, Remco C. Havermans, and Chantal Nederkoorn
Abbreviations WHO CS US UR CR CBT
World Health Organisation Conditioned stimulus (= cues) Unconditioned stimulus Unconditioned response Conditioned response Cognitive behaviour therapy
92.1 Introduction Obesity, defined as an unhealthy amount of body fat, is a major health problem that is increasing dramatically worldwide; figures show a current overweight and obesity prevalence of more than 60% in the USA and UK, and comparable statistics are documented for many other countries in the world (Wadden et al. 2002). In 2005, it was estimated by the World Health Organisation (WHO) that at least 400 million adults were obese and 1.6 billion people of over the age of 15 were overweight (World Health Organisation 2009). WHO further predicts that by 2015, about 2.3 billion adults will be overweight and more than 700 million will be obese. WHO refers to these figures as the obesity epidemic. The ultimate cause of obesity and overweight is an energy imbalance between calories consumed and calories expended. The main reasons for the energy imbalance are a shift in diet towards increased intake of energy-dense foods that are high in fat and sugars and decreased physical activity. Highly palatable energy-dense foods are widely available and difficult to resist. However, not everybody grows obese, so some people are better able to handle the temptations of the current environment than others. Understanding how people handle temptations is necessary to develop prevention strategies or to strengthen interventions for obesity. Why are some people better able to resist the “toxic” environment than others? A difference between normal eaters and overeaters might be related to the automatic responding of one’s body to appetitive cues as a consequence of learning. Overeating is associated with increased A. Jansen (*) Department of Clinical Psychological Science, Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_92, © Springer Science+Business Media, LLC 2011
1431
1432
A. Jansen et al.
cue reactivity, that is, increased appetitive responding to food cues (Jansen 1998; Jansen et al. 2003; Jansen et al. 2008). When overeaters are confronted with a diverse range of tasty food cues that are predictive of food intake – like the smell, taste, and sight of palatable high calorie foods – they show increased appetitive or cephalic phase responding, like salivation and the release of insulin, followed by increased food intake (Jansen 1998; Jansen et al. 2003; Wardle 1990; Woods 1991). Cephalic phase responses are responses that prepare for food intake: the greater the appetite, the more intense the response (Powley 1977). Signals that cause the cephalic phase response originate in the cerebral cortex, amygdala, and hypothalamus and are transmitted to the stomach. The cephalic phase or appetitive responding is believed to be experienced as an urge to eat, making it more difficult for people to refrain from eating. As a consequence, healthy eating in a “toxic” environment is much easier without than with these appetitive responses. Learning theory states that appetitive responses decrease and extinguish when exposure to foods remains systematically unreinforced, that is, when palatable high-calorie foods are seen or smelled but not eaten. In other words, being able to resist palatable high-fat food temptations and maintaining a regular and healthy diet even if exposed to the “toxic” environment, leads to a decrease and eventually extinction of the automatic urges to eat, which in turn makes it easier to refrain from eating high-fat foods for a longer period of time. In contrast, dieters who intermittently or always give in to the urge to eat, keep reinforcing the learned appetitive responding, ending up in increasingly stronger urges to eat the highly palatable foods (Jansen 1998). In the present chapter, it is argued and demonstrated that food cues might elicit almost reflex-like irresistible food cravings that could sabotage a healthy diet. It will also be shown that there are ways to decrease the abnormal food cue reactivity and overeating and it is suggested that cue exposure with response prevention should be used more often in the treatment of overeating.
92.2 Cue Elicited Eating The intake of food activates physiological responses. A large number of studies show that the physiological responses brought about by food intake, e.g. insulin release, blood sugar increase, and salivation, can be brought under the control of any stimulus predictive of food intake, like odors, time of the day, eating-related situations, seeing, smelling, tasting, and even thinking of food (see for an overview Jansen 1998; Siegel 1972; Woods 1991). Any time food is ingested, there is an opportunity to associate the food with cues that are present at the time (Bouton et al. 2006; Havermans et al. 2007). The place where the food is eaten, the people with whom it is eaten, the food preparing rituals, the smell and taste of the foods – they may all become signals for eating and when this happens, classical conditioning occurs (see Fig. 92.1). The conditioned responses (CRs) prepare the organism for food intake and contribute to the body’s internal homeostatic regulation. Food intake may, in terms of classical conditioning, be considered an unconditioned stimulus (US), whereas its metabolic effects are unconditioned responses (URs). Cues that reliably signal food intake, such as the sight, smell, and taste of food, or even the context in which one eats, may start to act as conditioned stimuli (CSs) that easily trigger cue reactivity (CRs). In short, it is theorized that cues that are (nearly) always and almost exclusively present at the time of eating will, in the long run, acquire the ability to predict the eating and its effects. One might think of both proximal cues, e.g., the sight, smell, and taste of the food, intake rituals such as the preparation of food, and interoceptive cues (e.g., affective states or typical palatable food-related cognitions), but also distal cues like context cues (e.g., the room where usually is eaten, time of eating). From the very moment that these cues reliably signal food intake and classical conditioning occurs, the cues or conditioned stimuli (CS) acquire the ability to elicit special food intake related
92 Cued Overeating
1433
Fig. 92.1 The classical conditioning model of (over)eating. Any time food is ingested, the food intake is associated with the cues that are present at that time. For example, when breakfast is always eaten at 07:00, this time will be associated with food intake. At the moment 07:00 is a reliable predictor of food intake, the time alone elicits responses that prepare for intake, like salivation
responses in the organism, like insulin release, blood sugar decrease, and the subjective experience of food cravings or appetite. These responses are calledCRs or cue-reactivity. It is assumed that food cravings or appetite reflect the subjective experience of these learned responses (Jansen 1998). The acquired responses (CRs) might have specific features. Deutsch (1974) showed that an initially neutral taste cue is able to elicit a glycometabolic effect in rats after pairing the taste with glucose administration in a classical conditioning paradigm; the rats learned that sugar would come and when they tasted the formerly neutral taste, they prepared for the glucose with a preparatory decline in blood sugar. In the same way, rats learned to respond with a decline in blood sugar to a placebo after they were repeatedly injected with glucose. Conditioned hypoglycemia was also demonstrated in dogs and humans after injections with intravenous glucose (Overduin and Jansen 1997; Overduin et al. 1997). Conversely, after repeated intravenous injections of insulin (which are usually followed by a decline in blood sugar level), rats learned to respond with a rise in blood sugar level to a placebo (Siegel 1972, 1983). Also eating behavior can be triggered by cues that were systematically associated with food consumption: rats that had already eaten till satiation were found to eat again when exposed to a tune that was previously associated with food consumption (Weingarten 1983). More recently, it was found that rats consumed significantly more (less palatable) chow when exposed to context cues that were earlier paired with the intake of highly palatable foods (Boggiano et al. 2009). Various context cues were used, e.g., different types of bedding or wallpaper. Boggiano et al. concluded that context-cues associated with palatable food intake might drive overeating in rats, even in sated rats and even when a less-preferred food (chow) was used in the test phase. The authors also report that the cue-conditioned overeating was quickly learned and appeared to be particularly strong when the taste of a palatable food was used to make cue-associations. In humans, it has been found that the mere anticipation of food and sham feeding (i.e. the sight, smell, taste, chewing, or swallowing of food without it entering the gastrointestinal tract) as well as cognitive processes (such as the thought of food or even hypnotic suggestion of food) elicit responses which prepare the organism for the digestion of food, such as insulin release and salivary responses (Mattes 1997; Nederkoorn and Jansen 2002; Nederkoorn et al. 2000; Power and Schulkin 2008; Powley 1977; Rodin 1985).
1434
A. Jansen et al.
In sum, it is well documented that exposure to food cues evokes cephalic phase responses. The cephalic phase responses gear up the body and serve the efficient use of nutrients. The classical conditioning model of food intake states that, after systematic association of cues (CS) and food intake (US), the cues will reliably signal the food intake effects. The moment the cues are good predictors of eating, they acquire the ability to elicit physiological responses that prepare for intake and are called cue reactivity. Appetite or craving is the subjective experience of cue reactivity. A main prediction flowing from this cued eating model is that the CRs or cue reactivity increase the likelihood of food intake and easily leads to overeating.
92.3 Cue Elicited Overeating Imagine that you are in a wedding party: in that situation you will definitely eat more than when you are at home on a weekly day. Presumably, you will also eat more than you need to: you overeat. Overeating is quite normal; many people overeat once in a while, depending on the context or situation. But there are also people who do overeat on a regular basis, almost habitually. They will gain weight and might become overweight or obese. Some other people, mostly young females, are frequent binge eaters. Binge eating refers to the eating of an objectively large amount of food in a discrete period of time, while experiencing a loss of control over intake (APA 1994). Mostly, palatable high-calorie foods are eaten during a binge and the food usually is stuffed into the mouth. Binge eating occurs in the (sub)clinical eating disorders Bulimia Nervosa, Anorexia Nervosa, and Binge Eating Disorder (APA 1994), and it is also been found to occur in about 12% of a normal female population sample (Bruce and Agras 1992) as well as 15–50% of the obese participating in weightcontrol programs (Marcus et al. 1985). The obese usually do not compensate for the extra calories that they eat during a binge, whereas the underweight and normal weight binge eaters do compensate by e.g. self-induced vomiting, use of laxatives, exercising, and dieting. The most frequently mentioned triggers of binge eating are negative emotions – like feeling depressed, hopeless, worried and dissatisfied – and appetitive cues that elicit craving – like the sight, smell, and taste of highly preferred foods (Jansen et al. 2008). Tasting or smelling palatable foods, being in a low mood, anxious or emotionally upset, thinking of eating; all these binge precursors could, hypothetically, be conditioned to the excessive intake of food. After systematic association of these cues with binge food intake, the cues reliably signal the food effects. The moment the cues are good predictors of intake, they will elicit physiological responses that are subjectively experienced as craving or appetite, which almost reflexively increase the likelihood of overeating. This model hypothesizes that classically conditioned associations between cues that predict food intake and actual eating behavior are stronger in overweight than in normal-weight children since parents of overweight children more frequently prompt their children to empty their plate and overweight children show higher external eating styles, meaning that their intake is more often triggered by food cues like seeing or smelling food. Both increase the probability that a food cue is followed by food intake, which strengthens the bond between cues and intake and makes smell and taste more predictive of intake in overweight than in normal-weight children. In line with the model it was found that overweight and normal weight 8–12 year old children ate comparable amounts when they were not tempted by food cues (Jansen et al. 2003). However, when they were tempted by cues that signal eating, like the smell and taste of highly palatable foods, the overweight children overate. They ate more in response to food cues than without, whereas the normal weight children did the opposite: they ate less in response to food cues than without these cues. Cue reactivity (salivation) was related to food intake but only in the overweight children: they showed a highly significant correlation between
92 Cued Overeating
1435
Table 92.1 Key points of cued overeating 1. The intake of food activates physiological responses 2. When food intake systematically is preceded by internal or external cues, classical conditioning occurs 3. During confrontation with the cue that predicts food intake, the body anticipates eating 4. Cues predictive of food intake are for example the smell, taste, and sight of palatable foods 5. These food cues bring about responses that prepare for intake, like salivation and the release of insulin 6. The anticipative appetitive responding increases appetite or the urge to eat 7. The anticipative appetitive responding increases intake This table lists the key points of cued overeating including the required systematic association between cue and food intake that leads to a process of classical conditioning. Being confronted with a cue that predicts eating prepares the body for food intake by anticipative appetitive responding
caloric intake and salivary flow after food cue exposure (r = 0.62), whereas this relation was almost absent in the normal-weight group (r = 0.05). The abovementioned study shows that overweight children eat normal amounts when tempting food cues are lacking. But when they are tempted by the taste or intense smell of palatable food, they fail to regulate their intake. This vulnerability to cued overeating in overweight children might follow from a learned association between the tempting cues and increased intake. The cue-elicited salivary response of the overweight sample was significantly related to their increased intake. Clearly, the cues elicited reactivity-related overeating in the overweight sample and not in the normal-weight sample, a finding that supports the idea that classically conditioned associations between cues that predict food intake (smell, taste) and actual eating behavior are stronger in overweight than in normal-weight children (Table 92.1).
92.4 The Role of Mood and Restraint Risk factor models for eating disorders have put forward that negative mood states are key triggers of overeating in samples with eating disorders (Stice et al. 2008). In some recent studies, eating disorders were subtyped along dimensions of negative affect and these studies showed that increased negative affect signaled a stronger vulnerability to disinhibition (Stice et al. 2008). In our lab, we showed that a specific state-trait interaction facilitated overeating: the high negative affect overweight/obese subtype was more vulnerable to overeating in the presence of a disinhibiting cue (negative mood induction or food exposure) than the overweight/obese subtype that was low in negative affect (Jansen et al. 2008). The data show that individual differences might play a critical role in the way overweight/obese people handle the temptations of the current “toxic” environment: negative affect makes it more difficult for the overweight/obese to resist modern temptations (see Fig. 92.2). Another trait that plays a role in the way people handle toxic temptations is restraint. Restrained eaters try to restrict their intake, mostly because they want to lose weight. But a substantial part of the restrained eaters usually alternates between restrained and overeating episodes: a very common eating pattern is that in the absence of tempting food cues, the so-called restrained eater succeeds in resisting highly palatable high calorie foods until a cued overeating episode announces itself. It should be noted that such an eating pattern facilitates classical conditioning: deprivation in the absence of cues and eating large amounts of high calorie palatable foods (strong USs) within a limited and specific range of cues (CSs) implies that the contingency between CS and US, and thus classical conditioning, will be strong (Bouton et al. 2006). Strong conditioning is reflected in strong CRs or cue reactivity including appetite or craving. Experimental studies indeed show that dieters overeat after exposure to cues that typically predict food intake, like tasting a priming dose (appetizer
1436
A. Jansen et al.
Fig. 92.2 The classical conditioning model of emotional eating. If snack foods are always eaten when emotionally upset, feeling upset will be associated with snack food intake. At the moment being emotionally upset is a reliable predictor of snack food intake, the emotion alone elicits responses that prepare for intake, like salivation
or preload; see for a classic experiment Herman and Mack 1975) and in other studies it was found that they overeat after mere exposure to the smell of binge food (e.g., Jansen and van den Hout 1991). The exposure to food cues (tasting, eating, or smelling the food) elicited a strong desire to eat in people that tend to alternate overeating and restrained eating.
92.5 Successful Dieting Some dieters never overeat and they might be called successful dieters. Although it is not precisely clear how many people are successful in dieting, it is estimated that about 20% of the obese are capable of reducing to a normal weight and to maintain this reduced weight for at least a year (Wing and Phelan 2005). It is however not at all clear why some people are better able to stick to their diets for a long period of time in the context of a “toxic” environment than others. Successful dieters report to especially refrain from eating palatable high calorie fat foods; they say they are strict dieters who eat little fat and show little variety in their diets (Gorin et al. 2004; Raynor et al. 2005; Wing and Hill 2001). The classical conditioning model of cued overeating would predict that if exposure to tasty high fat foods remains unreinforced in successful dieters, this will lead to the extinction of cue reactivity (including craving) during confrontation with the cues. The extinguished cue reactivity makes it easier to maintain one’s diet. In line with the cued overeating model, decreased salivary responding during exposure to palatable high fat food cues was found in postobese successful dieters, whereas unsuccessful obese dieters showed increased salivary responding during exposure to the same cues (Jansen et al. 2009). It was proposed that strict dieting extinguishes cue reactivity. In its turn, the extinction of cue reactivity will make dieting easier (see Fig. 92.3). Strict dieting is however quite difficult for many dieters, especially in the beginning of the dieting, and it might take much time before learned cue reactivity is extinguished. There are also dieters who do not expose themselves to high calorie foods but avoid them, for example because they only consume diet shakes for
92 Cued Overeating
1437
Fig. 92.3 The relation between the way of dieting, food cue exposure, food cue reactivity, and extinction or the success of dieting. If exposure to tasty high fat foods remains unreinforced this will lead to the extinction of cue reactivity and makes dieting easier
Fig. 92.4 The relation between the way of dieting, food cue exposure, food cue reactivity, and extinction or the success of dieting. If the dieter avoids the exposure to tasty high fat foods, this will not lead to an extinction of cue reactivity and sabotages dieting
periods of time. They are expected to remain cue reactive, which makes it more difficult for them to maintain the lost weight (see Fig. 92.4). It was further argued that the extinction of cue reactivity is necessary for successful dieting. Dieters that avoid highly palatable food cues, like the dieters that follow a specific shake-diet or another limited diet, do not enable their cue reactivity to extinguish. Likewise, in dieters who intermittently keep overeating high-calorie high-fat palatable foods, cue reactivity will not disappear because the CS is followed repeatedly by the US, which makes it difficult to learn that the CS does not predict the US anymore (see Fig. 92.5). For successful dieting and a reduction of overeating it seems necessary that dieters enter a vicious circle of strict dieting and decreased cue reactivity (see Fig. 92.3). Dieting appears to be difficult. Can we help overeaters to decrease cue reactivity, to facilitate dieting? (Table 92.2).
1438
A. Jansen et al.
Fig. 92.5 The relation between the way of dieting, food cue exposure, food cue reactivity, and extinction or the success of dieting. If the dieter intermittently overeats on tasty high fat foods, this will not lead to an extinction of cue reactivity and sabotages dieting
Table 92.2 Key points of successful dieting 1. When the body anticipates eating and prepares for consumption, appetite or the urge to eat is intense 2. When the cues are not followed by eating, the anticipative appetitive responding decreases and appetite or the urge to eat as well 3. After a series of nonreinforced exposures to the cues without eating, the cues do not predict eating anymore 4. When cues do not predict eating anymore, they will stop eliciting preparatory responses 5. Successful dieters confront themselves with the cues that predict food intake without eating 6. The preparatory responses in successful dieters are extinguished making it easier to refrain from foods This table lists the key points of successful dieting including the required extinction of the systematic association between cue and food intake. Being confronted with a cue without eating extinguishes the anticipative appetitive responding and makes dieting easier
92.6 The Reduction of Cue Reactivity The classical conditioning model of overeating states that cue reactivity follows from probabilistic CS (cues)–US (overeating) contingencies. Cues will elicit craving as long as they are reliable predictors of food intake or, to put it differently, as long as the CSs are systematically reinforced by the US. The model predicts that craving will extinguish when the CS–US bond is broken. This bond will be broken by prolonged and repeatedly nonreinforced exposure to the cues that predict overeating (CS). Strict dieting in the current “toxic” environment is a form of continuous nonreinforced exposure to palatable food cues. However, many dieters try to avoid the palatable food cues in order to make it easier to keep their diets. The toxic environment is however everywhere, so they will never succeed in always avoiding palatable food cues. The link between cues that indicate that overeating is forthcoming and the actual overeating might be eliminated by cue exposure with response prevention. During cue exposure with response prevention the subject is exposed to the craving-eliciting highly palatable food cues and eating is prevented. Individually customized sets of highly palatable food cues are the best to use; the more likeness the cues have to the favorite foods, the better the reactivity is. This means for the exposure to food cues, that all of the cues and contexts that play a part in the overeating should be included in the most perfect exposure: the exposure takes place at the customary overeating spot and with the most favorite foods for overeating. Special attention must be paid to
92 Cued Overeating
1439
Fig. 92.6 The extinction of overeating through cue exposure with response prevention. The emotional eater is exposed to the emotion (e.g., by inducing emotions) and responds with cue reactivity including a strong desire to eat the foods (or craving). But the emotional eater is prevented from eating. After several sessions of exposure with response prevention, the emotion is no longer predictive of overeating. The cue reactivity including the desire for eating or craving is extinguished
moods and thoughts that accompany the overeating. For many overeaters, negative moods are important cues that elicit craving; inducing a negative mood might magnify the urge to eat. The goal of the exposure is to elicit a strong craving for the palatable high-calorie food. The overeater touches the food, feels around in it, grabs it, holds it to the nose and smells it – but never eats; the CS–US bond will be broken. Craving will increase until it is experienced as almost irresistible. But if the exposure lasts long enough, the craving for food will slowly die down despite all attempts to keep it as strong as possible; CRs (cue reactivity and craving) will extinguish (see Fig. 92.6). Pilot studies show that cue exposure with response prevention elicits craving – the craving typically increases gradually until it is high, and then it extinguishes slowly in patients with bulimia nervosa (Jansen et al. 1989; Jansen et al. 1992; Toro et al. 2003; Martinez-Mallén et al. 2007). Extinction was found both within and between the exposure sessions (cue reactivity started lower each session). The exposure was not only highly effective in the reduction of cue reactivity and craving, also the binge frequency of the bulimia nervosa patients decreased significantly. Another study was, however, less positive (Bulik et al. 1998; Carter and Bulik 1994). It was examined whether the effectiveness of short cognitive behavior therapy (CBT) could be increased through adding food cue exposure with response prevention. All bulimia nervosa patients were first treated with eight sessions of CBT, and then eight exposure sessions followed. The exposure did not improve the results of the short cognitive behavior therapy. However, CBT was already extremely effective; an 80% reduction of binge eating was reached and 40% of the sample was abstinent, i.e. did not binge eat anymore. These excellent results touch the sore spot; the cognitive behavior therapy reached a ceiling effect – the design of the study did not permit the exposures to be successful. All in all, the positive findings from pilot studies suggest that cue exposure might be an effective treatment for binge eaters. Note, however, that although the data are promising, they were derived from pilot studies with small subject samples. The only large controlled clinical trial that was published is not positive about the contribution of cue exposure with response prevention to CBT in the treatment of binge eating. We pointed to several methodological shortcomings of that study. Further, as far as the authors know, the effects of cue exposure are yet only documented for bulimia nervosa patients. It remains to be seen whether nonbinging overeaters will also benefit from cue exposure with response prevention. Apart from the extinction of cue reactivity that is strived for by the cue exposure to highly palatable food cues, the olfactory stimulation during the exposure to palatable food cues might have induced
1440
A. Jansen et al.
olfactory sensory-specific satiety. A short period of olfactory stimulation (i.e., smelling) produced decreased pleasantness of the smell of the food, which is called olfactory sensory-specific satiety (Rolls and Rolls 1997). Thus, smelling foods may induce satiety. It was suggested by Jansen et al. (2003) that overeaters and people that show a tendency to overeat are characterized by a slowing down of the sensoryspecific decrease in neural activity. Indirect evidence for this idea is given in studies on the extinction of craving in binge eaters; binge eaters show increased craving after they started smelling binge foods (without eating it), and the highest of their craving is after about 20 min, after which craving gradually declines. In normal eaters the highest of craving is reached earlier and craving extinguishes more rapidly. Relating these findings to the cue reactivity model, it might be hypothesized that tasting or intense smelling of palatable food works as a prime that elicits cue reactivity especially in overeaters who are used to eat after being confronted with these cues. Normal eaters show sensory-specific satiety responses to the highly palatable foods. Note that the priming cues will elicit reactivity as long as they are reliable predictors of intake. The model predicts that the cue reactivity will lead to overeating until the cue–intake bond is broken by prolonged and repeated nonreinforced exposure to the cues. During cue exposure, the participant is exposed to the cues (smell, taste) and prevented from intake, leading to reduced reactivity and craving (Jansen 1998). Cue exposure with response prevention might be a promising new treatment intervention for overeaters. It might help to reduce cue reactivity and craving more quickly compared to when one is merely strict dieting. Thus, the first blow is half the battle; a successful start of the diet that frees the body of its cue reactivity will make dieting easier and more successful.
92.7 Applications to Other Areas of Health and Disease The learning model of excessive consumption was originally formulated for addictive behaviors; many studies showed that the craving and substance intake of addicts is cue-controlled. Cue exposure with response prevention seems to be a useful therapy for addictive behaviors as well (see e.g., Havermans et al. 2007). Cue-elicited overeating might also be present in some anorexia nervosa patients, suggesting that cue exposure with response prevention is indicated for anorexia nervosa as well. Therapists are however strongly advised against using cue exposure with response prevention in anorexia nervosa. A main characteristic of anorexia nervosa is the low body weight and cueinduced (over)eating might be a mean to consume at least some calories. As long as they are underweight, anorexia nervosa patients should not be treated with strategies that do decrease intake.
92.8 Conclusions The present model of cued overeating states that cues that reliably signal highly palatable food intake, such as the sight, smell, and taste of highly palatable foods, or even the context in which the overeating takes place, may start to act as conditioned stimuli that trigger cue reactivity and craving. It is assumed that learned cue reactivity increases the probability of overeating. Numerous experiments demonstrated that overeaters specifically overeat after confrontation with cues that predict the intake of highly palatable foods. It was further argued that the extinction of cue reactivity is necessary for successful dieting. Dieters that avoid highly palatable food cues, like the dieters that follow a specific shake-diet or another limited diet, do not enable their cue reactivity to extinguish. Likewise, dieters who intermittently keep eating high-calorie high-fat palatable foods will remain cue reactive. Only dieters who expose themselves to highly palatable food cues without eating them are expected to show the desired
92 Cued Overeating
1441
extinction of cue reactivity. This might however take a long time. The extinction of cue reactivity can be accelerated by cue exposure with response prevention. There have been a number of pilot studies suggesting that cue exposure with response prevention is an effective intervention to reduce binge eating. Whether cue exposure with response prevention is also suitable for overeaters and an effective intervention to reduce overeating remains to be studied. Studies on the effectiveness of food cue exposure should be encouraged. Is cue exposure with response prevention beneficial to help people reaching the extinction sooner? Will dieting become easier and more successful when cue exposure with response prevention sessions are introduced quickly after or even before the start of the dieting? It would be highly relevant to find out how long it takes for learned cue reactivity to extinguish, in dieters who expose themselves to highly palatable high calorie foods without eating them and in dieters who do not expose themselves or to a lesser degree. The cued overeating model predicts that the first blow is half the battle. Experimental studies are needed to find out whether this is true.
Summary Points • Obesity (partly) follows from overeating. • Overeating follows from classically conditioned appetitive responding to food cues. • The food cues elicit irresistible food cue reactivity including cravings that might sabotage a healthy diet. • If overeating systematically follows negative emotions, the negative emotions will become cues for overeating. • Cue exposure with response prevention decreases the abnormal food cue reactivity. • Successful dieting requires the extinction of food cue reactivity.
Key Terms Obesity: This refers to an unhealthy amount of body fat. Obesity usually is defined by a Body Mass Index (BMI = kg/m2) of 30 or more. Classical conditioning: It is a form of associative learning during which the organism learns that stimulus A predicts the occurrence of stimulus B, for example the smell of food (stimulus A) predicts eating (stimulus B). Food cues: Cues that are predictive of food intake like the smell, taste, and sight of food, but also place, time, or emotions and thoughts if they are predictive of food intake. Appetitive responding: Bodily responses that prepare for food intake, like salivation and insulin release, and the experience of appetite or urge to eat. Cue reactivity: Increased appetitive responding to cues that predict food intake. Extinction: The new learning that stimulus A does not predict stimulus B by presenting stimulus A without stimulus B. Cue exposure with response prevention: A treatment procedure in which the cues that predict eating (e.g., the smell and taste of foods) are presented without the actual eating until the urge to eat is decreased. After several exposure sessions without eating it is learned that the cues do not predict food intake anymore and at that very moment they will not induce any appetitive preparatory responses and appetite anymore. Successful dieting: The process by which dieters were successful in losing weight and in maintaining the weight loss.
1442
A. Jansen et al.
References American Psychiatric Association. Comittee on nomenclature and statistics: diagnostic and statistical manual of mental disorders. 4th ed. Washington, DC: American Psychiatric Association; 1994. Boggiano MM, Dorsey JR, Thomas JM, Murdaugh DL. The Pavlovian power of palatable food: lessons for weightloss adherence from a new rodent model of cue-induced overeating. Int J Obes. 2009;33:693–701. Bouton ME, Woods AM, Moody EW, Sunsay C, Garcia-Gutierrez A. Counteracting the context-dependence of extinction: relapse and tests of some relapse prevention methods. In: Craske M, Hermans D, Vansteenwegen D, editors. Lear and fearning. From basic processes to clinical implications. Washington, DC: APA; 2006. p. 175–96. Bruce B, Agras W. Binge eating in females: a population-based investigation. Int J Eat Disord. 1992;12:365–73. Bulik CM, Sullivan FA, Carter FA, McIntosh VV, Joyce PR. The role of exposure with response prevention in the cognitive-behavioural therapy for bulimia nervosa. Psychol Med. 1998;28:611–23. Carter FA, Bulik CM. Exposure treatments for bulimia nervosa: procedure, efficacy, and mechanisms. Adv Behav Res Ther. 1994;16:77–129. Deutsch R. Conditioned hypoglycemia: a mechanism for saccharin-induced sensitivity to insulin in the rat. J Comp Physiol Psychol. 1974;86:350–8. Gorin AA, Phelan S, Wing RR, Hill JO. Promoting long-term weight control: does dieting consistency matter? Int J Obes. 2004;28:278–81. Havermans RC, Mulkens S, Nederkoorn C, Jansen A. The efficacy of cue exposure with response prevention in extinguishing drug and alcohol cue reactivity. Behav Interv. 2007;22:121–35. Herman CP, Mack D. Restrained and unrestrained eating. J Pers. 1975;43:647–60. Jansen A. A Learning model of binge eating: cue reactivity and cue exposure. Behav Res Ther. 1998;36:257–72. Jansen A, van den Hout M. On being led into temptation: ‘counterregulation’ of dieters after smelling a ‘preload’. Addict Behav. 1991;5:247–53. Jansen A, van den Hout M, van Loof C, Zandbergen J, Griez E. A case of bulimia succesfully treated with cue exposure. Behav Ther Exp Psychiatry. 1989;20:327–32. Jansen A, Broekmate J, Heymans M. Cue exposure vs self-control in the treatment of binge eating: a pilot study. Behav Res Ther. 1992;30:235–41. Jansen A, Theunissen N, Slechten K, Nederkoorn C, Mulkens S, Roefs A. Overweight children overeat after exposure to food cues. Eat Behav. 2003;4:197–209. Jansen A, Vanreyten A, van Balveren T, Roefs A, Nederkoorn Ch, Havermans R. Negative affect and cue-induced overeating in non-eating disordered obesity. Appetite 2008;51:556–62. Jansen A, Stegerman S, Roefs A, Nederkoorn C, Havermans R. Decreased salivation to food cues in formerly obese successful dieters. Submitted for publication; 2009. Marcus MD, Wing RR, Lamparski D. Binge eating and dietary restraint in obese patients. Addict Behav. 1985;10:163–8. Martinez-Mallén E, Castro-Fornieles J, Lazaro L, Moreno E, Morer A, Font E, Julien J, Vila M, Toro J. Cue exposure in the treatment of resistant adolescent bulimia nervosa. Int J Eat Disord. 2007;40:596–601. Mattes RD. Physiological responses to sensory stimulation by food: nutritional implications. J Am Diet Assoc. 1997;97:406–12. Nederkoorn C, Jansen A. Cue reactivity and regulation of food intake. Eat Behav. 2002;3:61–72. Nederkoorn C, Smulders F, Jansen A. Cephalic phase responses, craving and food intake in normal subjects. Appetite 2000;35:45–55. Overduin J, Jansen A. Conditioned insulin and blood sugar responses in humans in relation to binge eating. Physiol Behav. 1997;61:569–75. Overduin J, Dworkin BR, Jansen A. Introduction and commentary to: Mityushov, M.I. (1954): conditioned reflex secretion of insulin. Integr Physiol Behav Sci. 1954;32:228–46. Power ML, Schulkin J. Anticipatory physiological regulation in feeding biology: cephalic phase responses. Appetite 2008;50:194–206. Powley T. The ventromedial hypothalamic syndrome, satiety and a cephalic phase hypothesis. Psychol Rev. 1977;84:89–126. Raynor HA, Jeffery RW, Phelan S, Hill JO, Wing RR. Amount of food group variety consumed in the diet and longterm weight loss maintenance. Obes Res. 2005;13:883–90. Rodin J. Insulin levels, hunger, and food intake: an example of feedback loops in body weight regulation. Health Psychol. 1985;4:1–24. Rolls ET, Rolls JH. Olfactory sensory-specific satiety in humans. Physiol Behav. 1997;61:461–73. Siegel S. Conditioning of insulin-induced glycemia. J Comp Physiol Psychol. 1972;78:233–41. Siegel S. Classical conditioning, drug tolerance and drug dependence. In: Israel Y, Smart FB, editors. Research advances in alcohol and drug problems, vol 7. New York: Plenum; 1983. p. 207–46.
92 Cued Overeating
1443
Stice E, Bohon C, Fischer K. Subtyping women with bulimia nervosa along dietary and negative affect dimensions: further evidence of reliability and validity. J Consult Clin Psychol. 2008;76:1022–33. Toro J, Cervera M, Feliu MH, Carriga N, Jou M, Martinez E, Toro E. Cue exposure in the treatment of resistant bulimia nervosa. Int J Eat Disord. 2003;34:227–34. Wadden TA, Brownell KD, Foster GD. Obesity: responding to the global epidemic. J Consult Clin Psychol. 2002;70:510–25. Wardle J. Conditioning processes and cue exposure in the modification of excessive eating. Addict Behav. 1990;15:387–93. Weingarten H. Conditioned cues elicit feeding in sated rats: a role for learning in meal initiation. Science 1983;220:431–2. Wing RR, Hill JO. Successful weight loss maintenance. Annu Rev Nutr. 2001;21:323–41. Wing RR, Phelan S. Long-term weight loss maintenance. Am J Clin Nutr. 2005;82(suppl):222S–5S. Woods SC. The eating paradox: how we tolerate food. Psychol Rev. 1991;4:488–505. World Health Organisation (2009). http://www.who.int/topics/obesity/en/
wwwwwwwwwwwwwww
Chapter 93
Frontal Behavioral Symptoms in Prader-Willi Syndrome Kaeko Ogura, Toshikatsu Fujii, and Etsuro Mori
Abbreviations PWS OFC MRI PET FTD
Prader-Willi syndrome Orbitofrontal cortex Magnetic resonance imaging Positron emission tomography Frontotemporal dementia
93.1 Introduction Prader-Willi syndrome (PWS) is a neurodevelopmental disorder occurring in approximately one out of 15,000 live births and is the most well-known genetic cause of marked obesity. PWS is characterized by a distinctive behavioral phenotype, including food-related and food-unrelated behaviors. The first report of PWS, published in 1956 by pediatricians Prader, Labhart and Willi, described a syndrome characterized by obesity, short stature, lack of muscle tone during infancy, impaired sexual development, and mental retardation (Prader et al. 1956). Important contributions to the understanding of the clinical features of PWS were made over the following 20 years, and the physical characteristics (Hall and Smith 1972), clinical traits (Holm et al. 1993; Cassidy 1997), and behavioral patterns (Dykens et al. 1996; Dimitropoulos et al. 2001) were characterized better. Figure 93.1a and b show typical individuals with PWS. As for details of general clinical features and genetic background of PWS, please refer to other chapters relevant to PWS. In this chapter, we review behavioral features of PWS and discuss the symptoms of PWS in relation to the dysfunction of the frontal lobe, especially the orbitofrontal cortex (OFC) (Fig. 93.2).
K. Ogura (*) Department of Pediatrics, National Rehabilitation Center for Persons with Disabilities, 1, Namiki 4-chome, Tokorozawa, Saitama 359-8555, Japan and Department of Behavioural Neurology and Cognitive Neuroscience, Tohoku University Graduate School of Medicine, Sendai, Japan e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_93, © Springer Science+Business Media, LLC 2011
1445
1446
K. Ogura et al.
Fig. 93.1 Individuals with PWS. (a) An 1-year-and-9-month old girl with PWS. She cannot stand up yet, because she has muscular hypotonia for her age. (b) Typical adults with PWS. Two individuals standing in a front row are Japanese adults with PWS (a male pictured in the left and a female in the right) and a person in a rear rank is a healthy adult male. Two individuals with PWS have typical features: short stature, central obesity, low muscular volume, and typical facial feature with narrow bifrontal diameter, small appearing mouth with thin upper lip, and down turned corners of mouth. In addition, their hands and feet are small (Published with permission)
93.2 Eating Behavior in PWS Abnormal eating behavior is the most striking feature of behavioral traits in PWS (Holm et al. 1993; Dimitropoulos et al. 2000). Persons with PWS show an intense preoccupation with food, hyperphagia, and incessant food seeking. Their fixation on foods sometimes strikes us as extraordinary. For instance, their families frequently report behavior such as food-stealing, scavenging, and eating raw or frozen food. The abnormal eating behavior is universal in children and adults with PWS, and the absence of it suggests an incorrect diagnosis (Holm et al. 1993; Dimitropoulos et al. 2000). The feature, however, is not present from birth. Individuals with PWS typically show failure to thrive in infancy, followed by the onset of obesity at the age of 18–36 months (Ehara et al. 1993). Infants with PWS are hypotonic, and have a poor suck, typically requiring tube feeding. During infancy they seem to have no interest in a feed. In early childhood, they develop an excessive interest in food and also in associated food seeking behavior, and they begin to gain weight excessively well above the 50th percentile for weight (Greenswag 1987). Hyperphagia and obesity are sustained into their adulthood. It is also shown that the abnormal body composition of those with PWS and their associated reduced energyuse contribute to their physical appearance and to the difficulty in preventing obesity they experience
93 Frontal Behavioral Symptoms in Prader-Willi Syndrome Fig. 93.2 Schematic drawings of the brain. (a) The left cerebral hemisphere viewed from the lateral aspect and (b) The right cerebral hemisphere viewed from the medial aspect
1447
a
Parietal lobe
Frontal lobe Occipital lobe
Temporal lobe
Orbitofrontal cortex (OFC)
b
Cingulate gyrus
Frontal lobe Occipital lobe
Orbitofrontal cortex (OFC)
Cerebellum
Hypothalamus yp Brain stem
(Theodoro et al. 2006). Because the resulting obesity is highly likely to lead to serious health problems and untimely death, it is a major problem to be concerned about (Greenswag 1987). It is generally stated that persons with PWS will continue to consume food as long as it is available and will search extensively for food when it is not available. In an experimental situation, individuals with PWS were shown to eat continuously, only a minority of them showing any reduction in their food intake over time (Zipf and Berntson 1987). Hyperphagia in PWS seems to be associated with an aberrant satiety response, particularly a delay in satiety (Holland et al. 1993). The foodrelated abnormal behavior have been attributed to a hormonal aberration caused by hypothalamic dysfunction, because those with PWS show other hypothalamic abnormalities, including growthhormone deficiency, hypogonadism, and an inability to regulate their body temperature (Swaab 1997). Decreased numbers of oxytocin-secreting neurons were found in autopsy materials from individuals with PWS (Swaab et al. 1995). In addition to hyperphagia, persons with PWS often show aberrant eating behaviors. Holm and Pipes (1976) reported that individuals with PWS have a propensity to consume products that are commonly considered as unappealing or unappetizing (e.g., dog food, garbage, sticks of butter, and decayed apples). Adults with PWS show a preference for sweet foods rather than salty, plain, or sour foods (Caldwell and Taylor 1983). Fieldstone et al. (1997) reported that individuals with PWS preferred high carbohydrate foods over high protein foods, and high protein foods over high fat foods. Anecdotal reports from parents and caregivers of persons with PWS suggest that those with PWS show a particular eating routine or ritualistic behavior during mealtime. These aberrant eating behaviors suggest that it is unlikely that eating behavior in PWS is simply attributable to the hypothalamic dysfunction.
1448
K. Ogura et al.
93.3 Other Behavioral Features in PWS Other behavioral characteristics in PWS are also distinctive, and they include temper tantrums, rigid thinking, uncontrollable impulse, aggressive behavior, lability of mood, and compulsive behavior (Holm et al. 1993; Dykens et al. 1996; Cassidy 1997; Dimitropoulos et al. 2001). Among them, the compulsive behavior has been of major interest to researchers investigating the PWS phenotype. The characteristic behaviors that are commonly seen in PWS include insistence on routines, repeating rituals, ordering and arranging objects, repetitive speech, and collecting behavior (Dykens et al. 1996). Repetitive behavior is common in early childhood among those with PWS. Children with PWS exhibit severer ritualistic behavior than typically developing children but not other children with developmental delays (Dimitropoulos et al. 2006). Older children and adults with PWS exhibit statistically significant elevations on the compulsivity subscale of the Yale-Broun Obsessive Compulsive Scale (Dykens et al. 1996). Dykens et al. (1996) found that individuals with PWS showed similar levels of symptom severity of compulsions when compared with those with obsessivecompulsive disorder who were matched for age and gender. They also found that the collecting behavior (see Table 93.1) was a salient symptom in PWS, and it was seen in more than half of the subjects. Recently, it is mentioned that compulsive manifestations is related to the severity of hyperphagia among individuals with PWS (Dimitropoulos et al. 2006).
93.3.1 Pathological and Neuroimaging Studies of the Brain in PWS Despite the observation of serious neurobehavioral symptoms associated with PWS, little is known about the neurobiology and brain development of individuals with PWS. At present, there are only several case studies examining brain structures in individuals with PWS. Postmortem anatomical and histopathological studies have revealed enlarged lateral ventricles, cortical atrophy, abnormalities in cerebellar dentate nucleus, disorganization of neuronal cell layer, neuronal cellar loss, and neuronal heterotopia in the cerebral white matter and brain stem (Hattori et al. 1985; Hayashi et al. 1992; Stevenson et al. 2004). Structural magnetic resonance imaging (MRI) studies have also shown nonspecific morphological abnormalities including ventriculomegaly, cortical atrophy, asymmetry of the Sylvian fissures, a small brainstem, and an abnormal folding pattern of cerebral gyri (Leonard et al. 1993; Hashimoto et al. 1998; Yoshii et al. 2002; Titomanlio et al. 2006; Miller et al. 2007a). Table 93.1 Key concepts of collecting behavior • Collecting is the tendency to obtain and retain objects, and is commonly seen among both children and adults, in modern as well as primitive societies • Abnormal collecting behavior is a symptom that involves acquiring things extensively, retaining objects even when they are of no immediate use, and resisting discarding the collected items even if they are excessive relative to the individual’s circumstances or they are of little or no value • The abnormal collecting behavior can appear as a manifestation of obsessive-compulsive disorder, autism, schizophrenia, anorexia, Tourette’s syndrome and various types of dementia • Little is known regarding its neurobiological mechanisms. It is deduced that the drive to collect would be initiated from subcortical and cortical mesolimbic structures involved in homeostatic regulation, and modulated by prefrontal and mesial temporal cortices • Damage to prefrontal cortex has been associated with an inability to organize and carry out goal-directed behavior, decision-making, planning and anticipating future consequences, and this may contribute to the failure in normal discard behavior This table lists the key concepts of abnormal collecting behavior and its neuronal substrates
93 Frontal Behavioral Symptoms in Prader-Willi Syndrome
1449
Recently, some neuroimaging studies have reported that individuals with PWS had changes in regional activity in some brain regions. Shapira et al. (2005) reported a functional MRI study showing that, after glucose administration, three individuals with PWS showed delayed signal reduction in the hypothalamus, the OFC, and nucleus accumbens, with signal increase in the dorsolateral prefrontal cortex and insula. Other two functional MRI studies comparing individuals with PWS with healthy controls have also demonstrated increased activity in the medial prefrontal cortex, which involves the OFC in response to food stimuli (Holsen et al. 2006; Miller et al. 2007b). A positron emission tomography (PET) study showed that neural response to food intake in the OFC differed between a PWS group and a non-PWS group (Hinton et al. 2006). Kim et al. (2006) reported that cerebral glucose metabolism at rest differed in the OFC, middle and inferior frontal gyri, anterior cingulate gyrus, temporal pole, and uncus between PWS individuals and controls. Lucignani et al. (2004), using [11C] flumazenil and PET, found possible functional changes of cerebral gammaaminobutyric acid receptor in the insula, cingulate gyrus, and frontal and temporal neocortices in six adults with PWS. These findings suggest that there may be a specific relationship between neurobehavioral symptoms and functional abnormality in the OFC in persons with PWS.
93.3.2 P ossible Relationship Between Behavioral Features in PWS and the Dysfunction of Orbitofrontal Regions From a viewpoint of symptomatology, the behavioral features in PWS are strikingly similar to behavioral symptoms of patients with frontal pathologies affecting the OFC. For instance, patients with frontotemporal dementia (FTD), a progressive neurodegenerative disorder affecting the frontal and anterior temporal lobes, often show early alterations in behavior and personality (Neary et al. 1998). Behavioral symptoms in FTD include mental rigidity, impulsivity, mood changes, abnormal eating behavior, and compulsive behavior. The abnormal eating behavior and compulsive behavior are the factors that best distinguish FTD from other etiologies of dementia (Bozeat et al. 2000; Ikeda et al. 2002). Recent studies have highlighted the high prevalence of alterations in food preference (e.g., escalating desire for sweet food), appetite, and eating behaviors in FTD (Neary et al. 1998; Bozeat et al. 2000; Ikeda et al. 2002). A significant weight gain has been reported to occur in from 30% (Ikeda et al. 2002) to 70% (Bozeat et al. 2000) of patients with FTD. Stereotypic eating behaviors, such as wants to cook or eat exactly the same foods each day and tends to eat foods in the same order, were often observed in FTD (Ikeda et al. 2002). The compulsive behavior and collecting behavior are also often emerged in FTD. The compulsive behavior in FTD shows a spectrum of complexity, with simple motor mannerism and verbal repetitions at one end and complex behavioral routines and repetitive conversational themes at the other (Bozeat et al. 2000). For instance, Mendez et al. (1997) reported that patients with FTD showed repetitive checking, recurrent cleaning rituals, wearing clothes of specific colors only, compulsive selfinjurious biting and hair-pulling, and collecting coupons entering multiple contests.
93.3.3 Frontal Behavioral Symptoms in PWS We took particular note of behavioral similarities between PWS and FTD, which may be caused by shared neural dysfunction. To investigate how frequently individuals with PWS manifest frontal behavioral symptoms, we surveyed the prevalence of eating and noneating behavioral features in PWS by using assessment tools developed originally for patients with FTD and with frontal lobe
1450
K. Ogura et al.
injury (Ogura et al. 2008). The questionnaire consisted of 35 questions related to three categories of behavior: eating behavior (with four domains: appetite, food preference, eating habits, and other oral behaivior), stereotyped behavior (with four domains: roaming, speaking, movements, and daily rhythm), and collecting behavior. It was administered in Japan to the parents of 250 individuals aged between 1–42 years with a clinical diagnosis of PWS. A parent was asked whether her/his child had the symptoms or not. If caregiver indicated that abnormal behavior was present, she/he was asked to rate its frequency and severity on a scale of one-to-three (where 1 is occasionally and 3 is frequently) and that of one-to-five (where 1 is slight and 5 is very marked), respectively. Then, a severity score was derived as the product of its frequency and severity. The results revealed that prevalence rates of all these behavioral categories were high for the individuals with PWS (Fig. 93.3 and Table 93.2). The overall prevalence rate was in good agreement with those in the previous reports for PWS (Caldwell and Taylor 1983; Holm et al. 1993; Dykens et al. 1996; Dimitropoulos et al. 2000), and it was comparable to those reported for FTD (Neary et al. 1998; Bozeat et al. 2000; Ikeda et al. 2002). Furthermore, the severity score of each domain in the eating behavior was significantly correlated with those in the stereotypy and the collecting behavior (Fig. 93.4). The findings mentioned above suggest that the behavioral features in PWS are associated with a dysfunction of the frontal lobe, especially that of the OFC. The OFC is thought to play a crucial role
early childhood late childhood adolescence adulthood
a
b
c
100
80
percent
60
40
20
0. Appetite
Food Eating Other oral preference habits behavior
Roaming Speaking Movements Daily rhythm
Collecting behavior
Fig. 93.3 The prevalence of symptom domains in each age group. The participating individuals were divided into four groups based on age: early childhood (aged 0–5 years), late childhood (aged 6–11 years), adolescence (aged 12–17 years), and adulthood (aged over 18 years). The overall prevalence of abnormal behaviors and the prevalence rates of each domain and of each item among the age bands were analyzed using the Chi-square test. (a) The prevalence of abnormal eating behaviors was extremely high in all age bands. (b) Stereotyped behaviors were also common, particularly stereotyped speaking and daily rhythm stereotypy. (c) The greatest group differences in prevalence were noted for collecting behaviors. Symbols indicate the prevalence and bars indicate 95% confidence intervals (Ogura et al. 2008: 472, Elsevier)
93 Frontal Behavioral Symptoms in Prader-Willi Syndrome
1451
Table 93.2 Prevalence rates of items in the eating behavior Prevalence rate (95% confidence intervals) (%) Early childhood Late childhood Adolescence 17 years
p value
Increase in appetite Seeks out food between meals Overeat at meal times Reports hunger Reports being overfull Food preference Prefers sweet foods or drinks more than before
53 (38–67) 49 (34–64) 53 (38–67) 71 (57–83) 53 (38–67)
60 (49–70) 66 (56–76) 74 (64–83) 83 (73–90) 55 (45–66)
71 (54–84) 71 (54–84) 78 (62–89) 80 (65–91) 32 (18–48)
3 (0–10) 72 (60–82) 87 (76–94) 84 (73–92) 87 (76–94) 41 (29–54)
NS NS 0.10 mg/l was 2.14 (95% CI: 1.21– 3.80, P = 0.007). The risk of AD was reduced in the presence of high concentrations of silica (RR: 0.73; 95% CI: 0.55–0.99, P = 0.04), and the authors concluded that a concentration of Al in drinking water above 0.1 mg/l may be a risk factor for dementia and AD although no dose-response effect was found (Rondeau et al. 2000) (Table 178.5). Recently, the same group revisited this topic with more precise data on daily Al and silica intake in a larger cohort followed for 15 years adding 400 subjects from the AluminumMaladie d’Alzheimer (ALMA+) cohort to the 3,777 elderly subjects from the PAQUID study (Rondeau et al. 2009) (Table 178.5). In this recent study, two measures of exposure to Al and silica were taken into account: geographic and individual exposure from daily consumption of tap water (including water used in making tea, coffee, soup or alcoholic drinks) and bottled water (spring or mineral). Of the whole sample, only 1,925 subjects were considered because they were free of cognitive impairment at baseline and had reliable water consumption data. Whereas geographic exposure
Table 178.5 Principal longitudinal studies on the relationships between dietary aluminium (Al) and dementia (i.e., Alzheimer’s disease, AD and vascular dementia, VaD), or predementia syndromes (i.e., age-related cognitive decline, ARCD) in older people References Cognitive functions Longitudinal studies Setting and study design Subjects Aluminium assessment assessment Results and conclusions The risk of dementia was higher MMSE and other In each residential area 3,777 subjects aged Longitudinal study Rondeau for individuals who lived in cognitive tests; the range and mean 65 years and (8 years of follow-up); et al. (2000) areas with high levels of Al in standardised exposure to Al older subjects from France water (>0.1 mg/l) compared questionnaire for (0.001–0.459 mg/l, PAQUID study with people residing in areas dementia’s median 0.009 mg/l) with Al levels less than diagnosis (DSM III and silica (4.2–22.4 0.1 mg/l (RR = 1.99, 95% revised criteria); mg/l) from drinking CI = 1.20–3.28). Higher silica NINCDS/ADRDA water were concentrations (>11.25 mg/l), criteria for AD; recorded were associated with a Hachinski score reduced risk of dementia for VaD (RR = 0.71, 95% CI 0.56–0.91). The adjusted RR of AD for individuals exposed to Al concentration >0.10 mg/l was 2.14 (95% CI 1.21–3.80); the risk of AD was reduced in the presence of high concentrations of silica (RR = 0.73, 95% CI 0.55–0.99). No doseresponse effect was found An inverse association between SPMSQ considering Questionnaire relative 7,598 women aged Cross-sectional and Gillette-Guyonnet silica intake from drinking women cognitively >75 years to the daily longitudinal study et al. (2005) water and AD was found. normal with a score consumption of tap (7 years of follow-up); France Women with AD were >8; MMSE and water and mineral multicentre study 2.7 times as likely to have a Grober and water and the brand (EPIDOS) low daily silica intake (0.1 mg/day; standardised RR = 2.26; P = 0.049). questionnaire for Conversely, an increase of diagnosis of 10 mg/day in silica intake dementia (DSM III reduced the risk of dementia revised criteria); (adjusted RR = 0.89; NINCDS/ADRDA P = 0.036). Using the criteria for AD; geographic measure of Hachinski score for tap-water exposure, Al or VaD; cognitive silica concentrations were decline was no more associated with the analyzed in both the risk of AD, although the PAQUID and the tendencies were similar ALMA+ cohorts; dementia and AD were investigated only in the PAQUID cohort This table lists the principal findings of longitudinal clinical and epidemiological studies on the relationships between dietary Al (from drinking water or foods) and dementia (i.e., AD and VaD) or predementia syndromes (i.e., ARCD) in older people, including setting, study design, and Al and cognitive assessment used PAQUID Personnes agées Quid, MMSE Mini Mental State Examination, DSM III Diagnostic and Statistical Manual of Mental Disorders, Third Edition, DSM III Diagnostic and Statistical Manual of Mental Disorders, Third Edition, NINCDS/ADRDA National Institute of Neurological and Communicative Disorders and Stroke/Alzheimer’s Disease and Related Disorders Association, DSM IV Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, RR relative risk, EPIDOS Study Epidemiology of Osteoporosis Study, SPMSQ Short Portable Mental Status Questionnaire, ALMA+ Aluminium-Maladie Alzheimer
Longitudinal studies
References
178 Aluminium in the Diet, Cognitive Decline and Dementia 2837
2838
V. Frisardi et al.
to Al or silica from tap water was not associated strongly with dementia, given that other territorial factors could influence cognitive decline, the conclusions of the study were that cognitive decline became greater over time in subjects with a higher daily intake of Al from drinking water (>0.1 mg/ day; adjusted RR =2.26; P = 0.005). Moreover, about silica intake, an increase of 10 mg/day was associated with a reduced risk of AD (adjusted RR = 0.89, P = 0,036) (Rondeau et al. 2009) (Table 178.5). Recently, new evidence has come to the fore in support of the Al-AD hypothesis. First, a case of a 58-year-old woman with a rapidly progressive, fatal neurological illness, who, at autopsy, showed dramatic Ab deposition of cerebral cortical and leptomeningeal blood vessels, modest numbers of NFTs and Lewy bodies, and evidence of very high Al content in affected brain regions. This neurological injury is potentially linked to Al exposure from drinking water. In fact, this woman was exposed, along with other 20,000 people, to high concentrations of Al in their water supplies, in excess of 500–3,000 times the limit of 0.2 mg/l. (Exley and Esiri 2006). Second, a very recent report that compared baseline and follow-up composition of drinking water and the level of cognitive function and possible risk of AD in women taking part in the Epidemiology of Osteoporosis Study (EPIDOS) found that a low silica concentration was associated with low cognitive performance at baseline. Further multivariate analysis including potential confounding factors showed that women with AD appeared to have been exposed to lower amounts of silica at baseline, suggesting a protective role against AD of silica in drinking water (Gillette-Guyonnet et al. 2005) (Table 178.5). On balance, all epidemiological studies of Al in drinking water are more or less open to critique, particularly considering the difficulty in producing high-quality data for exposure and especially for the disease (Lovell et al. 1996). A fundamental difficulty in the interpretation of the epidemiological studies indicating increased risk for AD with increased Al concentrations in drinking water is that even at high concentrations (0.1–0.4 mg/l), drinking water accounts for less than 5% of total daily Al intake. Moreover, in contrast with the findings of these epidemiological studies and evidence against the Al-AD hypothesis is the fact that many studies examining antacid exposure that involves 1000-fold higher exposure to Al compared to drinking water or diet and AD, have been largely negative (Lione 1985). Similarly, other studies did not support any association between Al in drinking water and AD. Wettstein and colleagues (1991) compared the cognitive functions of two groups of elderly, long-term residents of Zurich, who lived in two different areas: one, with high Al concentration in drinking water (> 0.10 mg/l), and the other with relatively low Al levels (65 years
260 subjects aged 65–90 years
Cross-sectional, population based
Pradignac et al. (1995)
Inverse relationship between MUFA intake and cognitive decline. Significant inverse association between MUFA intakes and selective attention. No association was found between nutritional variables and episodic memory Cognitive functioning are affected mainly by age and education, not by dietary fatty acids MMSE: subjects with a score between 10 and 17 versus subjects with a score between 28 and 30 Fatty fish and marine n-3 PUFA consumption Concurrent to the dietary was associated with a reduced risk and intake assessment, the VVLT, the of cholesterol and saturated fatty acids with CST, an abbreviated an increased risk of impaired cognitive SCWT, the LDST, a CFT function in this middle-aged population were administrated
MMSE, DCT, and BSRT
MMSE, Geronte scale for the In men, alcohol intake was associated with improved functional and cognitive paramassessment of daily living eters, while PUFA intake only with functional activities status. In women, lipid intakes were related to better cognitive performance. Overweight in both sexes was associated with an improvement in functional status MMSE, PMSQ A diet poor in fatty acids, saturated fatty acids, and cholesterol, but rich in carbohydrates, fibers, vitamins (folates, vitamins C and E, and b-carotene, and minerals [zinc and iron) seems to improve cognitive skills
Table 179.6 Principal cross-sectional studies on the relationships between dietary fatty acids and predementia syndromes in older people Setting and study Reference design Subjects Dietary assessment Cognitive outcomes Results and conclusions Cross-sectional studies
2856 V. Solfrizzi et al.
Setting and study design Subjects Dietary assessment
Cognitive outcomes
Results and conclusions
Nurk et al. (2007)
Cross-sectional, population based
2,031 subjects, 70–74 years old
Evaluation of dietary intakes with a 169-item FFQ
Six cognitive tests were administered: m-MMSE, m-DST, m-BD; KOLT; TMT-A, and the S-task of the COWAT
Consumers of fish and fish products had better cognitive function than did non-consumers. The associations between fish and fish product intake and cognition were dosedependent. The effect of fish on cognition differed according to the type of fish and fish product consumed High levels of fish consumption are associated A standardized battery of Evaluation of fish 867 subjects, Dangour Randomized with better cognitive function in later life. cognitive tests: CVLT; consumption 70–79 years et al. (2009) clinical trial Furthermore, there was an apparent linear subjective memory variable that took old from 20 (24 months) trend for increased cognitive function across assessment; 3 tests of into account both general the five-item fish consumption variable, with prospective memory; story frequency and type practices in highest cognitive function levels found in recall (immediate and of fish England and those individuals who report eating the largest delayed); verbal fluency; consumption Wales. amount of fatty, as opposed to white fish letter cancellation; location memory (immediate and delayed); symbol-letter substitution; digit span forwards and backwards; simple and choice reaction time This table lists the principal findings of cross-sectional clinical and epidemiological studies on the relationships between dietary fatty acids and predementia syndromes (i.e., agerelated cognitive decline, ARCD, and mild cognitive impairment, MCI) in older people, including the setting and study design, and the dietary and cognitive assessment used. FFQ food frequency questionnaire, MMSE Mini-Mental State Examination (global cognitive functioning), PUFA = polyunsaturated fatty acids, PMSQ Pfeiffer’s Mental State Questionnaire (global cognitive functioning), DCT Digit Cancellation Test (selective attention), BSRT Babcock Recall Story Test (episodic memory); MUFA = monounsaturated fatty acids; SFA = saturated fatty acids; VVLT = Visual Verbal Learning Test (verbal memory), CST Concept Shifting Task (mental processing speed), SCWT Stroop Color Word Test (selective attention), LDST Letter Digit Substitution Test (perceptual-motor speed), CFT Category Fluency Test (semantic memory), m-MMSE modified Mini-Mental State Examination (global cognitive functioning), m-DST modified Digit Symbol Test (perceptual speed), m-BD modified Block Design (visuo-spatial and motor skills), KOLT Kendrick Object Learning Test (episodic memory), TMT-A Trail Making Test, part A (executive function), S-task of the COWAT the abridged version of the Controlled Oral Word Association Test (access to semantic memory), CVLT Californian Verbal Learning Test (verbal memory)
Reference 179 Dietary Fatty Acids, Cognitive Decline, and Dementia 2857
2858
V. Solfrizzi et al. 100 MEAN ENERGY INTAKE (kJ%)
10
80
60
17
8 47
7 20
18
8
9
9
49
48
50
16
Macronutrients ALCOOL
40
PUFA MUFA
20
SFA 12
13
14
13
0
CARBOHYDRATE PROTEIN
65–74 yrs Men
75–84 yrs Men
65–74 yrs Women
75–84 yrs Women
SUBJECTS STRATIFIED FOR AGE AND SEX
Fig. 179.1 Mean energy intakes from the Italian Longitudinal Study on Aging (ILSA). This figure shows the distribution of mean energy intakes (%), stratified for age and gender of the ILSA (Casamassima, Bari, Italy), first prevalence Survey, 1992–1993
Vercambre et al. 2009) (Table 179.7), indicating a crucial need for prospective studies that could confirm initial observations. In particular, one of these prospective studies, the Zutphen Study of 476 men aged 69–89 years, found that high linoleic acid intake was positively associated with cognitive impairment in elderly subjects only in one cross-sectional study after an adjustment for age, education, cigarette smoking, alcohol consumption, and energy intake. High fish consumption, tended to be inversely associated with cognitive impairment and cognitive decline at 3-year follow-up, but not significantly (Kalmijn et al. 1997a) (Table 179.7). Furthermore, in the cohort of the Etude du Viellissement Arteriel (EVA) Study, moderate cognitive decline (a > 2-point of MMSE decrease) and erythrocyte membrane fatty acid composition were evaluated in 264 elderly subjects aged 63–74 years, during a 4-year follow-up. In this study, a lower content of n-3 PUFA was significantly associated with a higher risk of cognitive decline. After adjusting for age, gender, educational level and initial MMSE score, stearic acid and total n-6 PUFA were consistently associated with an increased risk of cognitive decline. Moreover, a lower content of n-3 PUFA was significantly associated with cognitive decline, but after adjustment, this association remained significant only for DHA, and not for EPA (Hende et al. 2003) (Table 179.7). Findings from the CHAP on 2,560 persons aged 65 years and older, showed that in a large population-based sample, a high intake of saturated and trans-unsaturated fat was associated with a greater cognitive decline over a 6-year follow-up. Intake of MUFA was inversely associated with cognitive change among persons with good cognitive function at baseline and among those with stable long-term consumption of margarine, a major food source. Slower decline in cognitive function was associated with higher intake of PUFA, but the association appeared to be due largely to its high content of vitamin E, which shares vegetable oil as a primary food source and which is inversely related to cognitive decline. Finally, cognitive change was not associated with intakes of total fat, animal fat, vegetable fat, or cholesterol (Morris et al. 2004) (Table 179.7). In the same CHAP sample on 3,718 persons aged 65 years and older, high copper intake was associated with a significantly faster rate of cognitive decline but only among persons who also consumed a diet that was high in saturated and trans fats in a 6-year follow-up (Morris et al. 2005a) (Table 179.7). Moreover, in a total of 732 men and women aged 60 years or older, participating
Longitudinal, population based (4 years) Longitudinal, population based (6 years)
Hende et al. (2003)
Longitudinal, population based (6 years)
Longitudinal, population based (6 years)
Morris et al. (2005a)
Morris et al. (2005b)
Morris et al. (2004)
Longitudinal, population based (3 years)
Kalmijn et al. (1997a)
Evaluation of dietary intake with the cross-check dietary history method
3,718 subjects, aged Evaluation of dietary intake with a 139-items 65 years and FFQ older
3,718 subjects, aged Evaluation of dietary intake with a 139-items 65 years and FFQ older
Evaluation of fatty acid composition in erythrocyte membranes 2,560 subjects, aged Evaluation of dietary intake with a 139-items 65 years and FFQ older
246 subjects aged 63–74 years
476 subjects, aged 69–89 years
Cognitive impairment defined as a MMSE score 2 points of MMSE over a 3-year period MMSE score with a >2-point of decrease in a 4-year follow-up Cognitive change at 3-year and 6-year follow-ups measured with the EBT of Immediate and Delayed Recall, the MMSE, and the SDMT Cognitive change at 3-year and 6-year follow-ups measured with the EBT of Immediate and Delayed Recall, the MMSE, and the SDMT Cognitive change at 3-year and 6-year follow-ups measured with the EBT of Immediate and Delayed Recall, the MMSE, and the SDMT
(continued)
Dietary intake of fish was inversely associated with cognitive decline over 6 years. There were no consistent associations with the n-3 fatty acids, although the effect estimates were in the direction of slower decline
High copper intake was associated with a significantly faster rate of cognitive decline, but only among persons who also consumed a diet that was high in saturated and trans fats
Inverse association between cognitive decline and the ratio of n-3 to n-6 PUFA in erythrocyte membranes A diet high in saturated and trans-unsaturated fat or low in non-hydrogenated unsaturated fats may be associated with cognitive decline among older people
High linoleic acid intake (PUFA) was positively associated with cognitive impairment. High fish consumption was inversely associated with cognitive impairment
Table 179.7 Principal longitudinal studies on the relationships between dietary fatty acids and predementia syndromes in older people Setting and study Reference design Subjects Dietary assessment Cognitive outcomes Results and conclusions Longitudinal studies
179 Dietary Fatty Acids, Cognitive Decline, and Dementia 2859
Longitudinal, population based (8.5 years)
Longitudinal, population based (2.6 years)
Longitudinal, population based (5 years)
Longitudinal, population based (median 8 years)
Solfrizzi et al. (2006a)
Solfrizzi et al. (2006b)
van Gelder et al. (2007)
Psaltopoulou et al. (2008)
Table 179.7 (continued) Setting and study Reference design
Incident MCI. Diagnostic criteria for MCI: 1.5 SDs below mean age and education adjusted on the MMSE and 10th percentile below age and education adjusted on memory test, without SMC and intact ADL/IADL MMSE
Evaluation of MUFA and PUFA dietary intakes with a 77-item FFQ
Information about habitual food consumption was collected using the cross-check dietary history method
Evaluation of dietary intakes with a 150-item FFQ. A dietary composite score (MeDi score) evaluated adherence to Mediterranean diet
278 subjects, 65–84 years old from a cohort of 5,632 subjects
210 subjects,70–89 years old
732 subjects, 60 years or older
MMSE
High MUFA, PUFA, and total energy intake were significantly associated with a better cognitive performance in time. The association between high MUFA, PUFA intakes and cognitive performance remained robust even after adjustment for potential confounding variables such as age, sex, educational level, CCI, BMI, and total energy intakes Dietary fatty acids intakes were not associated with incident MCI. However, high PUFA intake appeared to have borderline nonsignificant trend for a protective effect against the development of MCI that may be important.
MMSE
Evaluation of MUFA and PUFA dietary intakes with a 77-item FFQ
278 subjects, 65–84 years old from a cohort of 5,632 subjects
Fish consumption was associated with less subsequent 5-y cognitive decline than was no fish consumption. Furthermore, a doseresponse relation was noted between the combined intake of EPA and DHA and cognitive decline, which suggests that a higher intake of EPA plus DHA was associated with less cognitive decline No significant association between MeDi score and MMSE scores, whereas a statistically significant inverse association was found between MMSE performance and some individual dietary components, such as seed oil or PUFA intakes
Results and conclusions
Cognitive outcomes
Dietary assessment
Subjects
2860 V. Solfrizzi et al.
Longitudinal, population based (13 years)
Setting and study design Results and conclusions Elderly women that were reported by informants to have undergone recent cognitive decline had, 13 years previously, lower intakes of poultry, fish, and animal fats, as well as higher intakes of dairy desserts and ice-cream. They had lower habitual intakes of dietary fiber and n-3 PUFA, but a higher intake of retinol. Elderly women that were reported by informants to be functionally impaired had, in the past, lower intakes of vegetables and vitamins B2, B6 and B12 Elevated SFA intake at midlife was associated with poorer global cognitive function and prospective memory and with an increased risk of MCI. High intake of PUFA was associated with better semantic memory. Frequent fish consumption was associated with better global cognitive function and semantic memory. Higher PUFA/SFA ratio was associated with better psychomotor speed and executive function
Cognitive outcomes DECO and IADL
Dietary assessment Evaluation of dietary intakes with a 208-item FFQ
Subjects
4,809 elderly women, 76–82 years old
Eskilinen et al. (2008)
Longitudinal, population based (21 years)
1,449 subjects aged 65–80 years
Evaluation of dietary intakes with a 208-item FFQ
The Mayo Clinic AD Research Center criteria were applied for diagnosing MCI; MMSE, CFT, PPBt, LDST, episodic memory with immediate word recall tests; executive function with the Stroop test, and prospective memory with a task by Einstein This table lists the principal findings of longitudinal clinical and epidemiological studies on the relationships between dietary fatty acids and predementia syndromes (i.e., agerelated cognitive decline, ARCD, and mild cognitive impairment, MCI) in older people, including the setting and study design, and the dietary and cognitive assessment used MMSE Mini-Mental State Examination (global cognitive functioning), PUFA polyunsaturated fatty acids, FFQ food frequency questionnaire, EBT East Boston Memory test (immediate and delayed episodic memory), SDMT Symbol Digit Modalities Test (perceptual-motor speed), MUFA monounsaturated fatty acids, CCI Charlson comorbidity index, BMI body mass index, SMC subjective memory complaint, ADL activities of daily living, IADL instrumental activities of daily living, EPA eicosapentaenoic acid, DHA docosahexaenoic acid, DECO DEtérioration Cognitive Observéè scale (observed cognitive deterioration), CFT Category Fluency Test (semantic memory), PPBt Purdue Peg Board task (psychomotor speed), LDST Letter Digit Substitution Test (perceptual-motor speed)
Vercambre et al. (2008)
Reference 179 Dietary Fatty Acids, Cognitive Decline, and Dementia 2861
2862
V. Solfrizzi et al.
in the European Prospective Investigation into Cancer and Nutrition (EPIC), Greece cohort, and residing in the Attica region, 6–13 years follow up showed that seed-oil consumption may adversely affect cognition, whereas adherence to the Mediterranean diet, as well as intake of olive oil, MUFA, and SFA exhibited weakly positive but not significant associations (Psaltopoulou et al. 2008) (Table 179.7). Finally, 4.809 elderly women (born between 1925 and 1930) were studied in a French longitudinal cohort, the Etude Epidémiologique de Femmes de la Mutuelle Générale de l’Education Nationale (E3N) study. Elderly women participating in the E3N cohort who were reported by informants to have undergone recent cognitive decline had, 13 years previously, lower intakes of poultry, fish, and animal fats, as well as higher intakes of dairy desserts and ice-cream. They had lower habitual intakes of dietary fiber and n-3 PUFA, but a higher intake of retinol. Furthermore, elderly women who were reported by informants to be functionally impaired had, in the past, lower intakes of vegetables and vitamins B2, B6, and B12 (Vercambre et al. 2009) (Table 179.7). More recently, in the Cardiovascular Risk Factors, Aging and Dementia (CAIDE) Study from eastern Finland, abundant SFA intake from milk products and spreads at midlife was associated with poorer global cognitive function and prospective memory and with an increased risk of MCI in average follow-up period of 21 years after adjusting for demographic and vascular factors, other fats, and apolipoprotein E (APOE). On the contrary, high PUFA intake was associated with better semantic memory. Also frequent fish consumption was associated with better global cognitive function and semantic memory. Further, higher PUFA/SFA ratio was associated with better psychomotor speed and executive function (Eskelinen et al. 2008) (Table 179.7). Therefore, on the basis of the previous significant suggestions (Solfrizzi et al. 2005), we tested further the hypothesis that high MUFA and PUFA intakes may protect against the development of cognitive impairment over time in a median follow-up of 8.5 years of the ILSA. The major finding of this study was that high MUFA, PUFA, and total energy intake were significantly associated with a better cognitive performance in time (Figs. 179.2 and 179.3). Total energy intake should be considered
28
26
MMSE Score
Fig. 179.2 Cognitive profile across time for total energy intake from the Italian Longitudinal Study on Aging (ILSA). This figure shows the mean observed Mini-Mental State Examination (MMSE) score profile across time for total energy intake (= 11330
Total Energy Intake (kJ/day)
179 Dietary Fatty Acids, Cognitive Decline, and Dementia 28
26
MMSE Score
Fig. 179.3 Cognitive profile across time for monounsaturated fatty acids (MUFA) intake from the Italian Longitudinal Study on Aging (ILSA). This table Mean observed MMSE score profile across time for daily MUFA intake (= 53.1
Monounsaturated fatty acid intake (g/day)
an important confounder of diet-ARCD relationships and, as we proposed in our methodological approach, suggesting that association between macronutrient intake and cognitive decline should be adjusted by total energy intake. No other individual dietary component of our study population was significantly associated with cognitive impairment in time (Solfrizzi et al. 2006a) (Table 179.7). The association between high MUFA, PUFA intakes, and cognitive performance remained robust even after adjustment for potential confounding variables such as age, sex, educational level, Charlson comorbidity index, body mass index, and total energy intakes (Solfrizzi et al. 2006a). Finally, recent findings from the ILSA demonstrated that while dietary fatty acids intakes were not associated with incident MCI, high PUFA intake appeared to have a borderline nonsignificant trend for a protective effect against the development of MCI (Solfrizzi et al. 2006b) (Table 179.7).
179.2.3 Fish Consumption and Cognitive Decline Epidemiological observational studies reporting associations of fish consumption with cognitive function have shown mixed results; some cross-sectional and longitudinal studies have reported a positive association with higher fish consumption (Morris et al. 2005b; van Gelder et al. 2007; Nurk et al. 2007; Dangour et al. 2009), while others have found no association (Kalmijn et al. 1997a) (Tables 179.6 and 179.7). Fish, particularly fatty fish (e.g., herring, mackerel, salmon, or trout), is the principal source of n-3 PUFA in the Mediterranean diet. Very recently, the baseline data from the Older People And n-3 Long-chain polyunsaturated fatty acid (OPAL) study, a double-blind randomized placebo-controlled trial examining the effect of daily supplementation with 700 mg n-3 PUFA on cognitive performance in healthy older persons aged 70–79, suggested that higher fish consumption is associated with better cognitive function in later life (Dangour et al. 2009) (Table 179.6).
2864
V. Solfrizzi et al.
In the Chicago Health and Aging Project (CHAP), dietary intake of fish was inversely associated with cognitive decline over 6 years of follow-up. In this cohort, there was little evidence that the n-3 PUFA were associated with cognitive change (Morris et al. 2005b) (Table 179.7). Furthermore, in the Zutphen Elderly Study, fish consumers had significantly less 5-year subsequent cognitive decline than did non-consumers. A linear trend was observed for the relation between the intake of EPA + DHA and cognitive decline, and an average difference of 380 mg/day in EPA plus DHA intake was associated with a 1.1-point difference in cognitive decline (van Gelder et al. 2007) (Table 179.7). Finally, findings from the Hordaland Health Study suggested that subjects whose mean daily intake of fish and fish products was >10 g/day had significantly better mean test scores and a lower prevalence of poor cognitive performance than did those whose intake was 15 points observed in a small group of patients months open for both with very mild AD (MMSE >27 points) groups The MCI group with supplementation The cognitive functions were A single dose of 240 mg/day of Kotani et al. (2006) 21 patients with mild showed a significant improvement of evaluated using Japanese version ARA and DHA, or 240 mg/ cognitive dysfunction the immediate memory and attention of RBANS at two time points: day of olive oil (placebo). (12 MCI patients with score. Organic group showed a before and 90 days after the The duration of exposure supplementation and significant improvement of immediate supplementation was 3 months 9 MCI patients with and delayed memory. However, there placebo), 10 patients were no significant improvements of with organic brain each score in AD and MCI placebo lesions, and 8 patients groups with AD No significant overall treatment effects on Neuropsychiatric symptoms were A single dose of 1.7 g of DHA Freund-Levi 204 patients with mild to neuropsychiatric symptoms, on measured with NPI and MADRS. plus 0.6 g of EPA was et al. (2008) moderate AD and with activities of daily living or on Care givers’ burden and activities administered. The duration acetylcholine esterase caregiver’s burden were found. of daily living (DAD) were also of exposure was 6 months inhibitor treatment and However, significant positive treatment assessed. placebo-controlled and a MMSE >15 points effects on the scores in the NPI 6 months open for both agitation domain in APOE e4 carriers groups and in MADRS scores in non-APOE e4 carriers were found.
179 Dietary Fatty Acids, Cognitive Decline, and Dementia 2871
23 patients with mild to moderate AD and 23 patients with MCI
Independently living individuals (n = 302) aged ³ 65 years CES-D score < 16 MMSE score > 21
Chiu et al. (2008)
Van de Rest et al. (2008a)
Effects of interventions This supplementation may improve global clinical function (CIBIC-plus) in MCI patients relative to placebo. No associations were found between randomization group and ADAS-cog, MMSE or HDRS scores Treatment with neither 1800 mg nor 400 mg EPA+DHA differentially affected any of the measures of mental well-being after 13 or 26 weeks of intervention compared with placebo
Outcome measures Global clinical function measured with CIBIC-plus, cognitive function with ADAS-cog and MMSE, depressive symptoms with HDRS Changes in mental well-being were assessed as the primary outcome with the CES-D, MADRS, GDS-15, and HADS-A
n-3 PUFA 1.8 g/day in monotherapy or placebo (olive oil). The duration of exposure was 24 weeks
A single dose of 1,800 mg/day EPA+DHA (n = 96), 400 mg/day EPA+DHA (n = 100), or placebo capsules (n = 106); the duration of exposure was 26 weeks A single dose of 1,800 mg/day EPA+DHA (n = 96), 400 mg/ day EPA+DHA (n = 100), or placebo capsules (n = 106); the duration of exposure was 26 weeks
Intervention and duration of exposure
Van de Rest et al. (2008b)
Independently living individuals (n = 302) aged ³ 65 years CES-D score < 16 MMSE score > 21
There were no significant differential Cognitive performance was assessed changes in any of the cognitive using an extensive neuropsychodomains for either low-dose or logical test battery that included high-dose fish oil supplementation the cognitive domains of compared with placebo; an effect of attention (SC-WT; fWDST), EPA–DHA supplementation in sensorimotor speed (TMT-A), subjects who carried the APOE e4 memory (WLT; bWDST), allele was also found, but only on and executive function the cognitive domain of attention (TMT-B; VFT) This table lists the principal findings of clinical trials on PUFA supplementation in patients with MCI, VaD, AD, and ARCD, including the intervention and duration of exposure, and the outcome measures used DHA docosahexaenoic acid, HDS-R Hasegawa’s Dementia rating scale, MMSE Mini-mental State Examination, EPA eicosapentaenoic acid, ADAS-cog cognitive portion of the Alzheimer Disease Assessment Scale, CDR Clinical Dementia Rating Scale, ARA arachidonic acid, RBANS Repeatable Battery for Assessment of Neuropsychological Status, NPI Neuropsychiatric Inventory, MADRS Montgomery Asberg Depression Scale, DAD Disability Assessment for Dementia, APOE apolipoprotein E, CIBIC-plus Clinician’s Interview-Based Impression of Change Scale, HDRS Hamilton Depression rating Scale, DRS-2 Dementia Rating Scale 2, CDT Clock Drawing Tests, ADCS-ADL Alzheimer’s Disease Cooperative Study–Activities of Daily Living, CES-D Center for Epidemiologic Studies Depression Scale, GDS-15 15-item Geriatric Depression Scale, HADS-A Hospital Anxiety and Depression Scale, SC-WT Stroop Color–Word Test, fWDST forward test of the Wechsler Digit Span Task, TMT-A Trail Making Test version A, WLT Word Learning Test, bWDST backward test of the Wechsler Digit Span Task, TMT-B Trail Making Test version B, VFT Verbal Fluency Test
Participants
Reference
Table 179.10 (continued)
2872 V. Solfrizzi et al.
179 Dietary Fatty Acids, Cognitive Decline, and Dementia
2873
memory and attention score for MCI patients, and a significant improvement of immediate and delayed memories for patients with organic brain damages (Kotani et al. 2006) (Table 179.10). Finally, the preliminary results from a 24-week, randomized, double-blind placebo-controlled study on 23 participants with mild or moderate AD and 23 with MCI randomized to receive n-3 PUFA 1.8 g/day or placebo (olive oil), suggested that his supplementation may improve global clinical function, as measured by Clinician’s Interview-Based Impression of Change scale which included caregiver-supplied information (CIBIC-plus), relative to placebo. No associations were found between the randomization group and ADAS-cog, MMSE, or Hamilton Depression rating Scale scores. Levels of EPA on erythrocyte membrane, were associated with cognitive function, measured by ADAS-cog, in these patients (Chiu et al. 2008) (Table 179.10). However, in a secondary analysis, participants with MCI showed more improvement of ADAS-cog than those with AD associated with n-3 PUFA administration (Chiu et al. 2008), which support recent reports showing that PUFA supplementation could be more beneficial on cognition in people with very mild AD (Freund-Levi et al. 2006) or MCI Kotani et al. 2006) than in AD patients (Table 179.10). A few years ago, a Cochrane review concluded that there was a growing body of evidence from biological, observational, and epidemiological studies that suggested a protective effect of n-3 PUFA against dementia. However, the Cochrane review team was unable to locate a single published randomized controlled trial on which to base recommendations for the use of dietary or supplemental n-3 PUFA for the prevention of cognitive impairment or dementia (Lim et al. 2006). However, very recently, a randomized, double-blind, placebo-controlled trial on 302 cognitively healthy (MMSE score > 21) individuals aged 65 years or older investigated the possible impact of n-3 PUFA on the mental well-being and cognitive performance of nondepressed (CES-D score < 16), older individuals (van de Rest et al. 2008a, b) (Table 179.10). In this RCT, participants were randomly assigned to 1.800 mg/d EPA–DHA, 400 mg/d EPA–DHA, or placebo capsules for 26 weeks. In older Dutch subjects, no effect of daily supplementation with low or high doses of EPA-DHA on mental wellbeing as assessed by depression and anxiety questionnaires was found (van de Rest et al. 2008a). Furthermore, there were no significant differential changes in any of the cognitive domains (attention, sensorimotor speed, memory, and executive function) for either low-dose or high-dose fish oil supplementation compared with placebo (van de Rest et al. 2008b). However, an effect of EPA–DHA supplementation in subjects who carried the APOE-e4 allele was found, but only on the cognitive domain of attention (van de Rest et al. 2008b). Fish oil may be beneficial in these subjects who are most sensitive to developing dementia. These two substantially negative studies on ARCD may be explained by the samples investigated (nondepressed and noncognitively impaired older subjects). Further trials in depressed patients or e4-carriers with MCI are needed. Finally, there is another ongoing RCT with cognitive endpoints of n-3 PUFA supplementation in healthy cognitively intact older persons. The OPAL study is a double-blind randomized placebo-controlled trial examining the effect of daily supplementation with 700 mg n-3 PUFA (500 mg DHA and 200 mg EPA) for 24 months on cognitive performance in healthy older persons aged 70–79 with good cognitive function (MMSE equals or greater than 24 out of 30 points at baseline), who are recruited from 20 primary care practices (Dangour et al. 2006). The OPAL study was completed at the end of 2007 and its findings will be published shortly. Thus, epidemiological evidence has suggested a possible association between PUFA (particularly, n-3 PUFA) and reduced risk of cognitive decline and dementia. However, due to the small number of studies that inform this topic, further research is necessary before a strong conclusion can be drawn. Some recent RCTs assessed the cognitive or functional effect of n-3 PUFA supplementation on patients with VaD, AD, MCI, or ARCD in cognitively unimpaired older subjects. These RCTs suggested a positive effect of this intervention only in very mild AD or MCI patients, or in subgroups (e.g., APOE-e4 carriers) for cognitive performance in nondemented subjects or for neuropsychiatric symptoms in mild to moderate AD patients. On the basis
2874
V. Solfrizzi et al.
of these evidences, we strongly suggest also for predementia syndromes, a high-risk condition for progression to dementia of vascular and degenerative origin, intervention trials using measures of dietary supplementation similar to the OmegAD Study to determine if such supplements will slow cognitive decline.
179.5 D ietary Unsaturated Fatty Acids and Cognitive Decline: Possible Mechanisms 179.5.1 Monounsaturated Fatty Acids and Cognitive Decline Different pathways could contribute to the neuroprotective as well as the neurotrophic properties of UFA. As seen above, in the older subjects of the ILSA, there was a high consumption of olive oil, with an MUFA energy intake of 17.6%, 85% of which was derived from olive oil (Solfrizzi et al. 1999) (Fig. 179.1). In our population, the prolonged protection of MUFA intake against age-related changes in cognitive functions may be linked to the relevant quota of antioxidant compounds in olive oil, including low-molecular-weight phenols (Solfrizzi et al. 2005). In fact, animal studies suggested that diets high in antioxidant-rich foods, such as spinach, strawberries, and blueberries, rich in anthocyanins and other flavonoids may be beneficial in slowing age-related cognitive decline (Solfrizzi et al. 2005). The possible role of antioxidant compounds from olive oil do not diminish or otherwise alter the argument concerning the fatty acids, because this is only a possible explanation of the role of MUFA on age-related cognitive changes in our population, in which MUFA intake derived for a large part from olive oil. The neuroprotective effects of dietary UFA could rely on their impact on membrane architecture. In fact, UFA have an important role in maintaining the structural integrity of neuronal membranes, determining the fluidity of synaptosomal membranes and thereby regulating neuronal transmission. Furthermore, essential fatty acids can modify the activity of certain membrane-bound enzymes (phospholipase A2, protein kinase C, and acetyltranferase), and the function of the neurotransmitters’ receptors. Finally, free fatty acids, lipid metabolites, and phospholipids modify the function of membrane proteins including ion channels (Solfrizzi et al. 2005). Moreover, fatty acid composition of neuronal membranes in advancing age demonstrated an increase in MUFA content and a decrease in PUFA content (Solfrizzi et al. 2005). n-3 PUFA increase membrane fluidity by replacing n-6 PUFA and cholesterol from the membrane (Solfrizzi et al. 2005) maintaining an optimal membrane fluidity as obligatory for neurotransmitter binding and signaling within the cell (Solfrizzi et al. 2005). There is also evidence associating a dietary deficiency of n-3 PUFA with changes in cortical dopoaminergic function (Solfrizzi et al. 2005) (Fig. 179.5).
179.5.2 Polyunsaturated Fatty Acids and Cognitive Decline In adult rats, learning and cognitive behavior are related to brain DHA status, which, in turn, is related to the levels of the dietary n-3 PUFA (Moriguchi et al. 2000). In fact, administration of DHA seems to improve learning ability in b-amyloid (Ab)-infused rats (Hashimoto et al. 2005) and inhibit decline in avoidance learning ability in the AD model rats, associated with an increase in the corticohippocampal n-3/n-6 ratio, and a decrease in neuronal apoptotic products (Hashimoto et al. 2002).
179 Dietary Fatty Acids, Cognitive Decline, and Dementia
2875
Modulation of membrane properties - physicochemical properties - structural properties Modulation of signaling cascades and lipid raft remodelling due to Modulation of serotonin and structural and physicochemical mesocorticolimbic dopamine changes in the membranes neurotransmission
n-3 Polyunsaturated fatty acids Docosahexaenoic acid (DHA)
Modulation of gene expression - in various tissues - mainly through interaction with risponde factors (eg RXR, PPAR)
Modulation of immunity and inflammation through -anti-inflammatory metabolites (NPD1) -- inflammatory gene regulation
Fig. 179.5 Synopsis of the neuroprotective properties of n-3 polyunsaturated fatty acids (PUFA) and docosahexaenoic acid (DHA). This figure lists various possible molecular mechanisms of n-3 PUFA and DHA linked to their neuroprotective properties including modulation of membrane properties, serotonin, and dopamine neurotransmission, signaling cascades and lipid raft remodeling, gene expression, and immunity and inflammation (Modified from Florent-Béchard et al. 2007. With permission)
Similarly, recent studies showed that dietary DHA in an aged AD mouse model could be protective against Ab production, deposition in plaques and cerebral amyloid angiopathy (Lim et al. 2005; Hooijmans et al. 2007), and increases cerebral blood volume (Hooijmans et al. 2007). In other transgenic AD mouse models, DHA also protects against dendritic pathology (Calon et al. 2004) and prevents neuronal apoptosis induced by soluble Ab peptides (Florent et al. 2006), increases synaptic protein and phospholipid densities (Cole et al. 2005; Wurtman et al. 2006), and inhibits degradative endopeptidase activities (Park et al. 2003). Some experimental evidence has suggested that essential n-3 PUFA effectively lower Ab production in transgenic mice, as reported in studies from several laboratories (Lim et al. 2005; Cole et al. 2005; Oksman et al. 2006). Yet, plaque burden was reduced in only one study using aged transgenic mice, following 3 months DHA enriched diet (Lim et al. 2005), but not in several other studies that started dietary intervention at a much younger age (Oksman et al. 2006; Hooijmans et al. 2007). Furthermore, DHA, given its unique structural properties, could allow of the modification of the architectural properties of the membrane, especially the distribution and the abundance of lipid raft microdomains (Florent-Béchard et al. 2007). Lipid rafts are liquid ordered sphingomyelin-rich cholesterol-rich PUFA-poor microdomains. DHA and PUFA enrichment is known to be accompanied by lateral phase separation and local lipid redistribution, subtle membrane remodeling and selective displacement of proteins (Florent-Béchard et al. 2007) (Fig. 179.5). By changing the organization and/or composition of the lipid rafts, DHA could directly modify many signaling pathways such as those initiated at the plasma membrane (Florent-Béchard et al. 2007) (Fig. 179.5). It is well known that PUFA of both n-3 and n-6 families control gene expression in a variety of tissues including liver and adipose tissue. However, the underlying mechanisms of the direct effects of dietary PUFA-induced differential gene expression pattern in the brain have been addressed by few studies (Florent-Béchard et al. 2007). Gene regulation by PUFA can occur through interactions with specific or nonspecific ligands such as transcription factors like peroxysome proliferator-activated
2876
V. Solfrizzi et al.
receptors (PPAR) or retinoid X receptor (RXR) that directly modulate the expression of target genes (Florent-Béchard et al. 2007) (Fig. 179.5). The direct effects of PUFA on gene regulation might be one of the clues to understand the beneficial effects of the n-3 PUFA on the nervous system. Several clinical and epidemiological studies have identified beneficial effects of PUFA for a variety of inflammatory diseases, yet without mechanistic explanations for these beneficial effects. Resolvins and protectins are recently identified molecules that are generated from n-3 PUFA precursors and can orchestrate the timely resolution of inflammation in model systems. In fact, DHA also serves as a precursor for the biosynthesis of additional bioactive counter-regulatory lipid mediators. For (neuro)protectin D1 (N)PD1 formation, DHA is rapidly released for conversion to 17S-hydroxyDHA that serves as a biosynthetic precursor (Serhan et al. 2006). In AD, NPD1 biosynthesis is activated by soluble APP-a (Lukiw et al. 2005). In this disorder, levels of DHA, NPD1, and 15-lipoxygenase (15-LOX) are selectively decreased in the hippocampus, providing a plausible mechanism for decreased neuroprotection in AD: less inhibition of apoptosis and subsequently, increased neuronal cell death (Lukiw et al. 2005) (Fig. 179.5). In a placebo-controlled randomized trial, the OmegAD study, AD patients treated with DHA-rich dietary supplements had reduced release of interleukin (IL)-1b, IL-6 and granulocyte colony-stimulating factor from peripheral blood mononuclear cells (Vedin et al. 2008). The n-3 PUFA from fish may be inversely associated with dementia because it lowers the risk of thrombosis, stroke, cardiovascular disease, and cardiac arrhythmia, reducing the risk of thromboembolism in the brain and consequently of lacunar and large infarcts that can lead to VaD and AD (Solfrizzi et al. 2005). Furthermore, the n-3 PUFA may be important as lipids in the brain, particularly for the possible influence of DHA on the physical properties of the brain that are essential for its function (Solfrizzi et al. 2005). Furthermore, fish oil was a better source than a-linolenic acid for the incorporation of n-3 PUFA into rat brain phospholipid subclasses (Solfrizzi et al. 2005). On the contrary, high linoleic acid intake (n-6 PUFA) may increase the susceptibility of LDL cholesterol to oxidation, which makes it more atherogenic, even if the association between linoleic acid and atherosclerosis is controversial (Solfrizzi et al. 2005). Therefore the ratio of dietary n-3/n-6 PUFA intake may influence the potential role of PUFA on cognitive decline and dementia, the optimal ratio of n-6/n-3 should be 20 on the EAT instrument, participants who were identified as alcohol dependent and participants whose drinking interfered with their daily activities. Table 185.2 Rates of alcohol use disordered eating behaviors (EAT-26 With permission) Alcohol dependence – past 12 months Alcohol interference – past 12 months Sample size
in Canadian men and women in association with score £ 20 vs >20) (From Gadalla and Piran 2007b.
EAT-26 Men Women Score £ 20 3.8%* 1.3%*** Score > 20 8.3% 3.8% Score £ 20 2.2%*** 0.6%*** Score > 20 9.1% 3.0% Score £ 20 16,685 19,638 Score > 20 88 573 This table shows the percentages of survey participants who scored > 20 versus £ 20 on the EAT instrument by within those who were identified as alcohol dependent Significance levels refer to the associations between measures of alcohol use and disordered eating behavior within gender using chi-square test p < 0.05, *** p < 0.0005 Data presented in the table indicate a strong association between disordered eating behavior and each of alcohol dependence and interference in women. For men, disordered eating behavior was strongly associated with alcohol interference and moderately associated with alcohol dependence
alone would be 0.012% (0.5% × 2.2%). However, the observed probability of the co-occurrence of eating disorders and alcohol interference was 0.048%, four times the chance probability, indicating an association between the two conditions. Similarly, the observed probability of co-occurrence of these two conditions for women was 0.085%, which was 4.4 times the probability of occurring by chance alone. All measures of alcohol use were consistently higher for women with disordered eating behavior, which was not the case for men (Table 185.2). While disordered eating behavior in women was strongly associated with alcohol dependence and alcohol interference, disordered eating behavior in men was strongly associated with alcohol interference and marginally associated with alcohol dependence. By contrast, some studies did not find an association between eating disorders and alcohol use disorders (e.g., Dunn et al. 2002; Matthews 2004; Stock et al. 2002; Welch and Fairburn 1996). Dunn and colleagues (2002) noted that their nonclinical sample of college students who met the criteria for bulimia nervosa did not drink significantly more alcohol than non-eating disordered students. Furthermore, Stock and colleagues (2002) reported that female adolescents with restrictive eating disorders drank significantly less alcohol than matched controls. Additionally, Welch and Fairburn (1996) compared alcohol consumption in three groups of women recruited from the community,
185 Gender Aspects in the Comorbidity of Eating Disorders and Alcohol Use Disorders
2967
women with bulimia nervosa, controls with other psychiatric disorders, and controls without such disorders. The authors found no difference in current alcohol consumption between bulimia nervosa cases and either control group. More recently, Matthews (2004) found no evidence of a concurrent relationship between eating disorders and alcohol use disorders in college students.
185.2.4 Review Studies Holderness and colleagues (1994) reviewed 51 studies on eating disorders and substance use and abuse that were published between 1977 and 1991. In 24 of the studies reviewed, the percentage of individuals with bulimia nervosa who reported alcohol abuse and/or dependence ranged between 2.9% and 48.6% with a median of 22.9%. These authors also noted that among eight studies that included women with both bulimia nervosa and anorexia nervosa, the rates of alcohol abuse and/or dependence ranged between 12% and 39% with a median of 26%. More recently, Gadalla and Piran (2007b) conducted a meta-analysis of 41 studies on the co-occurrence of eating disorders and alcohol use disorders in women that were published between 1985 and 2006. Over one third of studies reviewed (15 studies) recruited their participants from the community, approximately one third (14 studies) recruited participants from educational institutions, and 29% (12 studies) recruited their participants from clinical settings (Table 185.3). The type of participants recruited for the comparison group in clinical studies also varied, with four studies recruiting controls from the community and three studies recruiting controls from general psychiatric current or previous patients. As would be expected, criteria used for eating disorder assessments were significantly associated with the type of sample used. Clinical measures were used mostly with clinical samples and behavioral measures with student samples (Fisher’s exact test = 16.17, p-value = 0.001). Only four out of the 41 studies in the meta-analysis reported negative associations between eating disorders and alcohol use disorders. Two of the studies reported a negative relationship between anorexia nervosa and alcohol use disorders, with effect sizes of -0.85 and -0.49, while the remaining two studies reported negative relationships between bulimia nervosa or bulimic behavior and alcohol use disorders, with effect sizes of -0.07 and -0.92. As shown in Table 185.4, the mean effect size of overall eating disorders was 0.38 (se = 0.07) and was significantly different from zero (p < 0.001). A significant mean effect size was found for the relationship between alcohol use disorders and each eating disorder pattern except for anorexia nervosa. Based on Cohen’s (1988) categorization of an effect size of 0.2 as small, 0.5 as moderate and 0.8 as large, the average effect size found in this meta analysis ranged from small to moderate and was largest for bulimia nervosa/bulimic behavior and alcohol use disorders. Table 185.3 Types of samples and eating disorders assessment criteria used in the reviewed articles (From Gadalla and Piran 2007b. With permission) Type of sample Assessment of eating disorders Clinical Community Students Total Clinical 10 (83.3%) 8 (53.3%) 1 (7.1%) 19 Psychometric 1 (8.3%) 2 (13.3%) 5 (35.7%) 8 Behavioral 1 (8.3%) 5 (33.3%) 8 (57.1%) 14 Total 12 15 14 41 This table includes a count of the number of studies used in the meta-analysis by type of sample participants and the criteria used to assess eating disorders Fisher’s exact test was used to test the association of type of participants and criteria used to assess eating disorders. Fisher’s test = 16.17, p-value = 0.001, i.e. there was a significant association between the two characteristics
2968
T.M. Gadalla
Table 185.4 Means, standard errors and homogeneity indices of effect size of the relationship between eating disorders and alcohol use disorders in women, by type of eating disorder (From Gadalla and Piran 2007b. With permission) Homogeneity Between-study Type of eating disorder Number of studies index, Q variability, I2 Mean of ES (st error) 75.15 0.38 (0.07)*** Any eating disorder 41 160.94*** Bulimia nervosa/Bulimic 29 132.13*** 78.81 0.46 (0.10)*** ** behavior 8 22.59 69.01 0.09 (0.14) Anorexia nervosa 6 13.51* 62.98 0.41 (0.12)** Purging 6 1.06 0.00 0.13 (0.03)*** Restriction/dietary restraint 5 1.04 0.00 0.39 (0.14)** Binge eating disorder Eating disorders not 9 12.09 33.82 0.41 (0.12)*** otherwise specified For articles that reported more than one measure of alcohol use disorders or eating disorders, effect sizes were derived for all measures and their mean calculated such that only one effect size per study per pattern of eating disorder was used. The overall mean (and its standard error) of all studies per type of eating disorder is shown in the above table Mean effect sizes were transformed to z-values to test whether they were significantly different from zero. Except for anorexia nervosa, the mean effects for all types of disordered eating were significantly different from zero * p < 0.05, **p < 0.01, ***p < 0.001 The Q statistic measures the homogeneity of effect sizes across studies. Higher values of Q indicate lack of homogeneity among study results The I2 index is an estimate of the percentage of total variability in effect sizes that is due to between-studies variability rather than sampling errors within studies. The table shows significant heterogeneity among studies in the relationships between AUD and each of any ED, BN/bulimic behavior, AN, and purging reported by different studies. On the other hand, the relationship between AUD and each of binge eating disorder and dietary restraint were the most consistent with minimal heterogeneity across studies
The authors also noted that the co-occurrence rates of different patterns of disordered eating with alcohol use disorders were generally weakest and most divergent when participants were recruited from clinical settings and strongest and most homogeneous when participants were recruited from student populations.
185.3 P roposed Mechanisms of the Relationships Between Eating Disorders and Alcohol Use Disorders While high rates of comorbidity of eating disorders and alcohol use disorders have been well documented and observed in samples drawn from clinical settings, educational institutions and the community, the nature of this relationship is still unclear. Several hypotheses have been proposed, each with supportive as well as contradictory research findings.
185.3.1 The Addictive Personality Hypothesis The addictive personality hypothesis is based on the belief that certain individuals are predisposed to become addicted to one or more substances (Brisman and Siegel 1984). Supporters of this hypothesis stress the clinical and behavioral similarities between eating disorders, especially bulimia nervosa and binge eating, and alcohol use disorders. Two suppositions are at the heart of this hypothesis: (1) eating disorders can be characterized as an addiction and (2) individuals who are vulnerable to these two addictions share certain personality traits that can be used to identify them. Neither assumption is universally accepted (Wolfe and Maisto 2000).
185 Gender Aspects in the Comorbidity of Eating Disorders and Alcohol Use Disorders
2969
It has been suggested that both bulimia nervosa and alcohol use disorders involve similar brain systems and share elements of preoccupation, loss of control (Gold et al. 2003), impulsivity (Fichter et al. 1994; Wiederman and Pryor 1996), and compulsivity (e.g., Franko et al. 2005). For example, bulimic women have been described as having symptoms of impulse control, which may explain the elevated alcohol use in this population (e.g., Fichter et al. 1994; Wiederman and Pryor 1996). It is not clear, however, whether binge eating carries other addiction criteria, such as physical dependence and tolerance. Research by Keel et al. (2001) found that purging in the absence of binging in normal-weight individuals may be related more to substance use than previously believed. Findings presented by Franko and colleagues (2005) found alcohol use to be more closely related to purging than binging and concluded that both vomiting and alcohol use represent a compulsive dimension of psychopathology. Research studies documenting higher prevalence of borderline personality disorders in individuals with both eating disorders and substance use disorders compared with individuals with eating disorders alone (e.g., Grilo et al. 1995) provide supporting evidence to the addictive personality hypothesis. However, with the overlap between diagnostic criteria for borderline personality disorder and those for bulimia nervosa and substance abuse, these high prevalence rates could be the result of the diagnostic instruments utilized (Wolfe and Maisto 2000).
185.3.2 The Developmental Hypothesis Vulnerability to adolescent stressors has also been used to explain the observed high co-occurrence of eating disorders and alcohol use disorders. Some researchers have emphasized the role of dysregulation in the co-occurrence of disordered eating patterns and substance use (e.g., Stewart et al. 2000). This view proposes that regulatory challenges may be expressed through engagement in different disordered eating and drug consumption behaviors. For example, some adolescents may engage in experimenting in drugs as well as disordered eating behavior to conform to social and culture pressures (Wolfe and Maisto 2000). Several studies have reported associations between the use of problem weight loss methods, such as diet pills and vomiting and alcohol use in adolescents (e.g., French et al. 1997; Neumark-Sztainer et al. 1998; Keel et al. 1998). For example, French and colleagues (1997) found alcohol use to predict purging and binge eating among white adolescent females but not black females. In another study of adolescent boys, Keel and colleagues (1998) did not find a difference in alcohol use between boys with disordered eating and those without disordered eating. More recently, alcohol use was found to predict the use of laxatives and vomiting in white females and black males in a large sample of middle school students (Garry et al. 2003). Garry and colleagues (2003) also found that female students consistently reported use of all weight loss methods more frequently than male students. The dysregulation hypothesis could be evaluated further by examining associations between disordered eating behavior and other forms of behavioral dysregulation, such as risky sexual behavior, in addition to drinking behavior (Stewart et al. 2000).
185.3.3 The Family History Hypothesis A familial relationship has been suggested as a contributor to the observed high co-occurrence rates of eating disorders and alcohol use disorders. This hypothesis is based on evidence of genetic links to both types of disorders. However, the mechanisms underlying this hypothesis have not been adequately described (Wolfe and Maisto 2000) and the findings of a number of studies have disputed it.
2970
T.M. Gadalla
Research evidence in support of a common genetic predisposition to both eating disorders and alcohol use disorders include higher rates of eating disorders in adult children of alcoholics (e.g., Jonas and Gold 1988) and high rates of family history of alcohol abuse in bulimic patients (Bulik 1987). On the other hand, no relationship was found between parental alcohol use disorders and eating disorder symptomatology (Mintz et al. 1995), nor between prevalence of eating disorders and having an alcohol-dependent relative (Meyer 1997; Schuckit et al. 1996). Finally, in a large scale female twin study, Kendler and colleagues (1995) found the genetic factors for alcohol use disorders to be unrelated to the genetic factors of alcohol use disorders.
185.3.4 The Food Deprivation Hypothesis Dietary restraint has been suggested as a mediating mechanism in explaining the relationship between body weight and shape preoccupation and adverse patterns of alcohol drinking, especially patterns of binge drinking (Stewart et al. 2000). Dietary restraint may result in intake disinhibition leading to both eating and drinking binges. Similarly, alcohol drinking restraint, motivated by the need to avoid high caloric intake, may also result in drinking binges (Stewart et al. 2000). Stewart and colleagues (2000) found dietary restraint to be strongly associated with measures of binge drinking, quantity of alcohol consumed and excessive drinking in female university undergraduates. The authors concluded that chronic dieting was related to heavy drinking patterns. Kleiner and colleagues (2004) examined the relationship between body mass index and alcohol use in a sample of weight management female patients and found obese patients to have lower rates of alcohol use than women in the general population. Additionally, as body mass index increased, alcohol consumption decreased and it was found that body mass index increased during supervised abstinence. The authors concluded that overeating may compete with alcohol for brain reward sites and result in reduced alcohol intake and dependence rates and that alcohol may replace existing food reward pathways. However, some controversial evidence counters the argument for the food deprivation hypothesis. Research by Bulik and Brinded (1993) reported that bulimics and controls failed to increase their alcohol consumption following a 19-h food deprivation period. It is not known, however, whether a longer food deprivation period would result in an increase in alcohol consumption.
185.3.5 The Negative Affective Hypothesis Negative feelings and emotional instability have also been offered as explanations of the relationship between eating and substance use disorders, including alcohol (Benjamin and Wulfert 2005). This hypothesis stipulates that some individuals use both food and alcohol to self medicate when they feel distressed or anxious. In a longitudinal study of patients with anorexia nervosa or bulimia nervosa, poor psychosocial functioning and a history of substance use were found to predict prospective onset of alcohol use disorders (Franko et al. 2005). Franko and colleagues (2005) also found depression to predict alcohol use disorders in women with anorexia nervosa. Additionally, in a sample of female relatives of alcoholics, women with lifetime bulimia nervosa and alcohol dependence were more likely than women with either disorder alone to have major depression (Duncan et al. 2006). In a community-based study of correlates of binge eating disorders in men and women (Grucza et al. 2007), the authors reported a significantly higher prevalence of depression and probable alcohol
185 Gender Aspects in the Comorbidity of Eating Disorders and Alcohol Use Disorders
2971
use disorder in persons with binge eating disorder compared with those without binge eating disorders. Further, in a nationally representative sample of Canadians, having disordered eating behavior was associated with significantly higher odds of major depression for both genders (Gadalla and Piran, 2009). Gadalla and Piran (2009) found that the initially significant association between disordered eating behavior and alcohol dependence became nonsignificant when depression was controlled for, while the significant association between disordered eating behavior and alcohol interference was reduced when depression was controlled for. Based on the findings of the later study, the authors suggested that depression may fully mediate the relationship between eating disorders and alcohol dependence and partially mediate their relationship with alcohol interference for both men and women. It should be mentioned, however, that both behavioral and affective dysregulation can be related to life experiences, such as trauma (van der Kolk et al. 1996), as well as to inborn temperamental tendencies (Martin et al. 2000). Indeed trauma has been found to explain the observed relationship between binge eating and alcohol abuse (Dansky et al. 2000).
185.4 Discussion and Recommendations Although research has generally reported high rates of co-occurrence between eating disorders and alcohol use disorders, there is great variability in the reported co-occurrence rates across studies. Methodological differences among studies make it difficult to compare findings across them. For example, studies investigating the comorbidity of eating disorders and alcohol use disorders have recruited participants from heterogeneous settings, such as, schools, clinics, and communities with large variability in the size of samples. Another factor that contributes to the variability in the reported rates is the use of a variety of instruments to measure eating disorders and a wide range of criteria to assess alcohol use disorders. Alcohol consumption is a complex behavior and measures of total number of drinks consumed may not yield the same pattern of association with eating disorders as measures of binge drinking or other measures that take into account adverse consequences to drinkers’ lives (Stewart et al. 2000). In addition, women on average have lower body weight relative to men, a fact which should be taken into account when comparing patterns of alcohol consumption (Matthews 2004). Another factor that complicates the comparison of findings across studies is the use of different reference times for the diagnosis of both disorders which ranged from point-prevalence rates to lifetime occurrences. Only when researchers use common criteria for assessing both eating disorders and alcohol use disorders can meaningful comparisons between studies be made. Many studies collapse all eating disorders into one category and likewise for drug and alcohol abuse. This practice adds to the difficulty of making general inferences about the comorbidity of these disorders. In addition, since clinical based studies are affected by sampling biases, researchers have repeatedly highlighted the role of population-based studies in assessing the extent and nature of comorbidity (Dansky et al. 2000; Welch and Fairburn 1996) as well as the importance of selecting appropriate controls in clinical studies. Clinical-based studies are affected by sampling biases especially when studying patients with multiple problems since comorbidity itself may influence individuals to seek treatment (Welch and Fairburn 1996). Moreover, these patients do not represent the entire range of symptoms present in the population. Researchers are encouraged to continue to explore the mechanisms that could mediate this consistent pattern of co-occurrence of eating and alcohol use disorders, beyond specific mechanisms related to dietary restraint and body shape preoccupation that may lead both men and women to engage in disruptive alcohol binges. Well-designed empirical research regarding the etiological
2972
T.M. Gadalla
relationship between eating disorders and alcohol use disorders is lacking (Wolfe and Maisto 2000). Most studies, to date, have used cross-sectional data, which precludes any definitive conclusions regarding the temporal sequence. Longitudinal, controlled studies are needed in order to establish the temporal sequence of the occurrence of eating disorders, alcohol use disorders, depression, anxiety, and family history of substance dependence. Very few studies have focused on eating disorders in men and even fewer have attempted to compare men with and without eating disorders or men with eating disorders versus women with eating disorders. Evidence exists to suggest that eating disorders in men are underdiagnosed and undertreated (Weltzin et al. 2005; Woodside et al. 2004). Men may not seek treatment due to their experience of less severe symptoms or because they may not consider themselves at risk for eating disorders (Woodside et al. 2004). Other barriers to seeking treatment may include cultural biases and lack of treatment settings that are dedicated for men with eating disorders (Weltzin et al. 2005). Consequently, very little is known about eating disorders in men and its relationship with other forms of psychopathology. However, with the increasing pressure on men to be fit and to look muscular, there is supportive evidence that suggests body dissatisfaction and eating disorders in men are increasing (O’Dea and Abraham 2002). More research efforts should be directed to the study of all aspects of eating disorders in men.
185.5 Applications to Other Areas of Health and Disease The focus of this chapter has been the comorbidity of eating disorders and alcohol use disorders. However, the empirical observations and discussion of underlying mechanisms presented here can be easily extended to the comorbidity of eating disorders and drug use disorders. Elevated rates of co-occurrence of eating disorders and drug abuse have been reported by many researchers (e.g., Carlat et al. 1997; Jackson and Grilo 2002; Piran and Robinson 2006; Welch and Fairburn 1996; Wiederman and Pryor 1996). Findings of these studies suggest that, in contrast with the consistent pattern regarding the association of eating disorders and alcohol use disorders in women and men, the findings regarding illicit drug use are markedly different between genders. For example, Gadalla and Piran (2007a) found disordered eating behavior to be associated with current illicit drug use and lifetime use of cocaine/crack, amphetamines (speed), MDMA (ecstasy) and hallucinogens, PCP, or LSD, and cannabis in a community sample of women, whereas in men, disordered eating behavior was only associated with amphetamine use. Most studies, however, tended to examine the associations between eating disorders and a restricted range of substance classes (Piran and Robinson 2006) or combine all substances in one group. Some substances may be used as appetite suppressants for the purpose of weight loss (Nappo et al. 2002) and others used for their stimulant and euphoric effects (Feldman et al. 1997). Thus, it is important that future studies examine the relationship between eating disorder patterns and each class of substances separately.
Summary Points • Empirical research provides evidence of high comorbidity between alcohol use and all disordered eating patterns, except anorexia nervosa. This comorbidity was observed for both genders and is highest with binge eating and bulimic behavior. • More research is needed to determine the temporal sequence of the occurrence of eating disorders and alcohol use disorders and to explain the mechanisms underlying their co-occurrence.
185 Gender Aspects in the Comorbidity of Eating Disorders and Alcohol Use Disorders
2973
• Emerging research indicates that disordered eating in men is a growing concern. However, few studies included male participants and even fewer examined gender differences. • Future research is encouraged to use longitudinal data collected from population-based samples. • The importance of using appropriate control groups in clinical studies is highlighted.
Definitions and Explanations of Key Terms Anorexia nervosa: Anorexia nervosa is a psychiatric diagnosis that describes an eating disorder characterized by low body weight and body image distortion with an obsessive fear of gaining weight. Bulimia nervosa: Bulimia nervosa is an eating disorder characterized by recurrent binge eating, followed by compensatory behaviors, such as self-induced vomiting and the use of laxatives, enemas, and diuretics. Purging: Purging refers to the act of using self-induced vomiting and the use of laxatives, enemas, or diuretics to get rid of ingested food. Binge eating disorder: Binge eating consists of episodes of uncontrollable overeating, during which a person rapidly consumes a large amount of food. Dietary restraint: Dietary restraint refers to behaviors that are intended to limit food intake, in an effort to manage weight. Such attitudes may include cutting food into small pieces, obsession with counting calories, etc. Eating disorders not otherwise specified: Eating disorder not otherwise specified is described in the DSM-IV-TR as a “category [of] disorders of eating that do not meet the criteria for any specific eating disorder”. Alcohol dependence: Alcohol dependence describes the use of alcohol despite significant detrimental consequences. For a person to meet criteria for Alcohol Dependence, the person must meet three of a total of seven possible DSM-IV criteria within a 12 month period. The criteria include tolerance and withdrawal, losing control of drinking, a progression of addiction, and continuing to drink despite being aware that it is causing or psychological or physiological problem(s). Alcohol interference: Alcohol interference refers to the interference of drinking alcohol with the person’s work, study, social relationships, etc.
Key Points • Empirical research provides evidence of high comorbidity between alcohol use and all disordered eating patterns, except anorexia nervosa. This comorbidity was observed for both genders and is highest with binge eating and bulimic behavior. • Several hypotheses have been proposed to explain the mediating mechanisms between the two psychopathologies. By and large, these hypotheses have not been empirically tested. In addition, the temporal sequence of the occurrence of eating disorders and alcohol use disorders is not yet determined. • Although there is evidence to suggest that disordered eating in men is a growing concern, few studies included male participants and even fewer examined gender differences.
2974
T.M. Gadalla
References Adams TB, Araas TE. Int J Eat Disord. 2006;39:240–4. American Psychological Association Research agenda for psychosocial and behavioral factors in women’s health: Health behaviours.. Recommendations from the Advisory Committee of the Psychosocial and Behavioral Factors in Women’s Health: Creating an Agenda for the 21st Century Conference. 2005. http://www.apa.org/pi/wpo/ behavior.html#alcohol. Accessed 15 May, 2005. Asci FH, Tuzun M, Koca C. Eat Behav. 2006;7:362–7. Barry DT, Grilo CM, Masheb RM. Int J Eat Disord. 2002;31:63–70. Benjamin L, Wulfert E. Eat Behav. 2005;6:197–209. Brisman J, Siegel M. J Subst Abuse Treat. 1984;1:113–8. Bulik CM. Am J Psychiatry. 1987;144:1604. Bulik CM, Brinded EC. Addiction. 1993;88:1545–51. Carlat DJ, Camargo CA, Herzog DB. Am J Psychiatry. 1997;154:1127–32. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale: Lawrence Earlbaum Associates; 1988. Corcos M, Nezelof S, Speranza M, Topa S, Girardon N, Guilbaud O, Taïeb O, Bizouard P, Halfon O, Venisse JL, Perez-Diaz F, Flament M, Jeammet Ph. Eat Behav. 2001;2:27–38. Dansky BS, Brewerton TD, Kilpatrick DG. Int J Eat Disord. 2000;27:180–90. Duncan AE, Neuman RJ, Kramer JR, Kuperman S, Hesselbrock VM, Bucholz KK. Drug Alcohol Depend. 2006;84:122–32. Dunn EC, Larimer ME, Neighbors C. Int J Eat Disord. 2002;32:171–8. Elgin J, Pritchard M. Eat Weight Disord EWD. 2006;11:e96–101. Feldman RS, Meyer JS, Quenzer LF. Principles of neuropsychopharmacology. Massachusetts: Sinauer Associates, Inc. 1997. Feldman MB, Meyer IH. Int J Eat Disord. 2007;40:218–26. Fichter MM, Quadflieg N, Rief W. Psychol Med. 1994;24:591–604. Field AE, Austin SB, Frazier AL, Gillman MW, Camargo CA, Jr Colditz GA. J Am Acad Child Adolesc Psych. 2002;41:846–53. Franko DL, Dorer DJ, Jackson S, Manzo MP, Herzog DB. Int J Eat Disord. 2005;38:200–7. French SA, Story M, Neumark-Sztainer D, Downes B, Resnick M, Blum R. Int J Eat Disord. 1997;22:315–22. Gadalla TM. Int J Mens Health. 2009;8(1):72–81. Gadalla TM, Piran.N. J Ment Health, 2009;18(6):486–94. Gadalla TM, Piran N. Eat Disord. 2007a;15:189–203. Gadalla TM, Piran N. Arch Womens Ment Health. 2007b;10:133–40. Garfinkel PE, Lin E, Goering P, Spegg C, Goldbloom DS, Kennedy S, Kaplan AS, Woodside DB. Am J Psychiatry. 1995;152:1052–8. Garner DM, Olmsted MP, Bohr Y, Garfinkel PE. Psychol Med. 1982;12:871–8. Garry JP, Morrissey SL, Whetstone LM. Int J Eat Disord. 2003;33:55–63. Gold MS, Frost-Pineda K, Jacobs WS. Psychiatr Ann. 2003;33:117–22. Grilo CM, Levy KN, Becker DF, Edell WS, McGlashan TH. Addict Behav. 1995;20:255–60. Gruber AJ, Pope HG, Borowiecki JJ, Cohane G. J Clin Psychol. 2001;62:609–11. Grucza RA, Przybeck TR, Cloninger CR. Compr Psychol. 2007;48:124–31. Holderness C, Brooks-Gunn J, Warren M. Int J Eat Disord. 1994;16:1–34. Jackson TD, Grilo CM. Eat Weight Disord. 2002;7:276–83. Jonas JM, Gold MS. Psychiatry Res. 1988;24(2):195–99. Keel PK, Mayer SA, Harnden-Fischer JH. Int J Eat Disord. 2001;29:294–301. Keel PK, Klump KL, Leon GR, Fulkerson JA. Int J Eat Disord. 1998;23:125–32. Kendler KS, Walters EE, Neale MC, Kessler RC, Heath A, Eaves LJ. Arch Gen Psychiatry. 1995;52:374–83. Kessler RC, Andrews G, Mroczek D, Ustun B, Wittchen H. Int J Methods Psychiatr Res. 1998;7:171–85. Kleiner KD, Gold MS, Frost-Pineda K, Lenz-Brunsman B, Perri MG, Jacobs WS. J Addict Dis. 2004;232:105–18. Krahn DD, Kurth CL, Gomberg E, Drewnowski A. Eat Behav. 2005;6:43–52. Lindblad F, Lindberg L, Hjern A. Int J Eat Disord. 2006;39:662–6. Mangweth B, Hausmann A, Walch T, Hotter A, Rupp CI, Biebl W, Hudson JI, Pope HG. Int J Eat Disord. 2004;35:102–8. Martin GC, Wertheim EH, Prior M, Smart D, Sanson A, Oberklaid F. Int J Eat Disord. 2000;27:150–62. Matthews CR. J Addict Dis. 2004;23:67–80. Meyer DF. J Clin Psychol. 1997;53:107–16. Mintz LB, Kashubek S Tracy LS. J Couns Psychol. 1995;42:65–70. Nappo SA, Tabach R, Noto AR, Galduroz J.C. & Carlini E.A. Eating Behaviors. 2002;3:153–65.
185 Gender Aspects in the Comorbidity of Eating Disorders and Alcohol Use Disorders
2975
National Center on Addiction and Substance Use at Columbia University. Food for Thought: Substance Abuse and Eating Disorders Conference. 2001;25(6). http://www.casacolumbia.org/absolutenm/templates/article.asp?articleid= 296&zoneid=40. Accessed June 25, 2005. Neumark-Sztrainer D, Story M, Dixon LB, Murray DM. Am J Public Health. 1998;88:952–5. O’Dea JA, Abraham S. J Am Coll Health. 2002;50:273–8. Piran N, Gadalla TM. Addict. 2006;102:105–13. Piran N, Robinson S. Women Health. 2006;44:1–20. Schuckit MA, Tipp JE, Anthenelli RM, Bucholz KK, Hesselbrock VM, Nurnberger JI, Jr. Am J Psychiatry. 1996;153:74–82. Sinha R, O’Malley SS. Alcohol Clin Exp Res. 2000;24:1312–9. Stock SL, Goldberg E, Corbett S, Katzman DK. J Adolesc Health. 2002;31:176–82. Stewart SH, Angelopoulos M, Baker JM, Boland FJ. Psychol Addict Behav. 2000;14:77–82. van der Kolk BA, McFarlane AC, Weisaeth L. Traumatic stress: the effects of overwhelming experience on mind, body, and society. New York: Guilford Press; 1996. Welch SL, Fairburn CG. Br J Psychiatry. 1996;169:451–8. Weltzin TE, Weisensel N, Franczyk D, Burnett K, Klitz C, Bean P. J Mens Health Gend. 2005;2:186–93. Wiederman, MW, Pryor T. Int J Eat Disord. 1996;20:163–8. Wolfe WL, Maisto SA. Clin Psychol Rev. 2000;20:617–31. Woodside DB. Healthy Weight J. 2002;16(4):52–5. Woodside DB, Bulik CM, Thornton L, Klump KL, Tozzi F, Ficher MM, Halmi KA, Kaplan AS, Strober M, Delvin B, Bacanu SA et al. J Psychosom Res. 2004;57:273–8. Woodside DB, Garfinkel PE, Lin E, Goering P, Kaplan AS, Goldbloom DS, Kennedy SH. Am J Psychiatry. 2001;158:570–4.
wwwwwwwwwwwwwww
Chapter 186
The Great Disinhibitor: Alcohol, Food Cues, and Eating Behavior Wilhelm Hofmann, Georg Förster, Wolfgang Stroebe, and Reinout W. Wiers
Abbreviation BAC Blood alcohol concentration
186.1 The Great Disinhibitor: Alcohol, Food Cues, and Eating Behavior Lay theories of the association of alcohol and overweight focus mainly on beer-drinking men, who are believed to be at risk of developing beer bellies, defined as “a man’s fat stomach, caused by too excessive consumption of beer” (Oxford American Dictionary, 2nd ed., 2005, p. 146). If one inferred from this that female beer-drinkers or wine drinkers of both genders were not at risk of gaining weight from excessive alcohol consumption, one would be wrong. The active ingredient in beer which is responsible for weight gain (or any other problem) is alcohol (ethanol) and beer shares this ingredient with all other alcoholic beverages. Thus, there can be wine bellies as well as beer bellies and while women are likely to store their weight in other places, both genders are equally at risk of weight gain. In this chapter, we focus on two different ways by which alcohol can facilitate weight gain. First, alcohol does have a direct effect on caloric intake through the calories contained in alcohol. With 7 kilocalories per gram (kcal/g), alcohol has the second highest calorie content of any macronutrient, and people rarely compensate by eating less for the extra calories they take in by drinking beer or wine with their meals. Second, alcohol impairs people’s ability to regulate or control their food intake. We will present a model illustrating the various influences by which alcohol can disrupt the self-regulation of eating by boosting the impulsive processing of tempting food cues in the environment and by disrupting executive control in the short and long run. Furthermore, we will discuss the modulating role of alcohol expectancies, possible spiraling patterns with respect to the interplay of alcohol and eating, and applications to other health domains.
W. Hofmann (*) University of Chicago, Booth School of Business, 5807 South Woodlawn Avenue, Chicago, IL 60615 e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_186, © Springer Science+Business Media, LLC 2011
2977
2978
W. Hofmann et al.
186.2 The Direct Route: Alcohol as Non-compensated Caloric Intake Drinking two glasses of dry white wine (200 mL) adds 150 kcal to one’s meal. This might not seem much, but the daily excess of 150 kcal would result in a weight gain of 3.5 kilogram per year (Stroebe 2008). To work off an extra 150 kcal, an adult weighing 70 kg would have to walk for 37 min or jog for 18 min. Ironically, many people seem to reverse this logic and drink a lot of alcohol and/or high caloric “energy drinks” after a sporting event. Adding alcohol to a meal would not pose a risk of weight gain, if people stayed within their daily calorie limit by proportionally reducing their calorie intake from food. However, people do not (or only very imperfectly) compensate for the additional caloric burden contained in high-caloric beverages of all sort, including alcohol (Mattes 2006; Yeomans et al. 2003). One cognitive explanation for this regulatory failure is that even individuals who closely monitor their calorie intake often do not take account of the caloric content of beverages to the same degree that they monitor the caloric content of food. Hence, they do not adjust their selfregulatory standards appropriately. A second more physiological explanation is that beverages, due to their low viscosity, are far less satiating than solid foods (Hulshof et al. 1993). Taken together, alcohol’s direct influence on caloric intake is typically ignored and this effect alone can pose a serious threat to a person’s weight goals.
186.3 The Indirect Route: Undermining Self-regulation As Herman and Polivy (2004) point out, viewing eating behavior as simply regulated by bodily hunger and satiety signals is a great oversimplification. Eating behavior, at least in modern societies, is to a large extent also influenced by social norms (e.g., adherence to mealtimes, the presence of other people), the food environment (e.g., the presence of tempting food cues), and, most centrally to the present purpose, by internalized self-regulatory goals people harbor with regard to their food intake. Such self-regulatory goals are often related to matters of appearance (e.g., “I want to keep a slim figure because thin is attractive”) or health (“I want to avoid eating too high-cholesterol food”). However, most psychological theories of eating regulation assume that people differ in the extent to which they regulate their weight by cognitively controlling their calorie intake. For example, Schachter’s (1971) Externality Theory postulated that in contrast to normal weight individuals, who regulate their eating in response to internal cues of hunger or satiation, these cues are irrelevant to eating by obese. Their eating is triggered by external food-relevant cues such as the palatability of food or external cues in the environment, which signaled palatable food (e.g., sight or smell of food; salience of food cues). Schachter’s theory was later integrated by Herman and Polivy (1984) into their “Boundary Model of Eating,” which has dominated eating research for decades. Herman and Polivy (1984) introduced the individual difference variable of “Restrained Eating” and developed the “Restraint Scale” to place people on this dimension. People vary on a continuum from unrestrained to restrained eating. At the unrestrained eating end of the continuum are normal eaters, individuals who have no weight problems and regulate their calorie intake in response to internal cues. At the restrained endpoint are restrained eaters (i.e., chronic dieters), who continuously monitor their calorie intake, trying to keep it below some regulatory standard or “diet boundary” which defines the maximum amount of calories they allow themselves to consume. As the cognitive control of calorie intake requires more cognitive resources than the automatic regulation in response to internal hunger and satiety cues, regulation can easily be disrupted by external factors, which reduce the individual’s ability or motivation
186 The Great Disinhibitor: Alcohol, Food Cues, and Eating Behavior
2979
to control their calorie intake. It is these restrained eaters whose eating control is likely to be deregulated by alcohol consumption. Both externality theory and boundary model agree on the basic assumption that (a) weight is homeostatically regulated through bodily signals of hunger and satiety and that (b) for various reasons this homeostatic regulation malfunctions in individuals with weight problems or in restrained eaters. Although there can be no question that food intake and body weight are homeostatically controlled, the importance of this type of regulation for the development of overweight and obesity has increasingly been questioned (e.g., Lowe and Butryn 2007; Pinel et al. 2000). For example, Pinel and colleagues (2000) argued that people living in food-replete environments rarely experience energy deficits, but eat because of the pleasure that can be derived from food. Lowe and Butryn (2007) even suggested a distinction between homeostatic hunger, due to the prolonged absence of energy intake, and hedonic hunger, which is strongly influence by the availability and palatability of food. Stroebe and colleagues (e.g., Stroebe 2008; Stroebe et al. 2008) incorporated the idea of hedonic eating into a goal conflict model of eating, which assumes that eating regulation of restrained eaters is dominated by a conflict between two incompatible goals, namely, the goal of maintaining or reducing their weight and the goal of enjoying palatable food. These individuals try to shield their weight control goal by suppressing thoughts about palatable food. Unfortunately (at least from the point of eating control), most of us live in food-rich environments, where palatable food is widely available and where we are surrounded by cues signaling palatable food. Such cues are likely to prime the eating enjoyment goal and to increase its cognitive accessibility. Once the goal of eating enjoyment is instigated by such palatable food cues, the goal of eating control will be inhibited, leading the individual to overeat. The idea that eating involves relatively “primitive” hedonic influences as well as opposing higherorder processes of self-regulatory goal pursuit can also be derived from dual-process or dual-system conceptions of the mind (e.g., Hofmann et al. 2009; Strack and Deutsch 2004; Wiers et al. 2010). Specifically, these models generally assume an impulsive, association-based route to behavior determination whereby stimuli in the environment can automatically and effortlessly trigger motivational states such as cravings and desires and corresponding behavioral approach tendencies that have proven useful in the past to satisfy these cravings and desires. In contrast, long-term goal pursuit involves a series of coordinated higher-order processes of reasoning, behavioral decision making and behavior regulation (for more details, see Strack and Deutsch 2004). Importantly, the latter type of processing is typically more effortful in nature, and hence more easily disrupted than impulsive processing. Thus, according to both the goal conflict and the dual-systems framework, the eating regulation of restrained eaters is dominated by a conflict between the reflective process of maintaining and shielding the weight control goal in the service of long-term goal pursuit and the more impulsive process that capture the hedonic short-term consequences of yielding to the temptation at hand. Whether one or the other side of this conflict wins, depends on the relative strength with which regulatory and impulsive influences impinge on final behavior determination. Figure 186.1 illustrates the competing influence of impulsive and reflective processing on caloric intake. In the following sections of this chapter, we will introduce the basic elements of the model and finally discuss the ways by which alcohol may modulate the effects of impulsive and reflective processing on eating behavior. Focusing on reflective processes, Baumeister and Heatherton (1996) have identified three necessary factors for successful self-regulation: standards, self-monitoring, and the execution of control. First, self-regulatory goal standards have to be mentally represented in an accessible and consistent state during self-regulation (see Fig. 186.1). Active maintenance and shielding of self-regulatory goals from interference appears to build on working memory resources (e.g., Hofmann et al. 2008). Self-monitoring, in turn, pertains to the updated comparison of ongoing behavior with one’s standards, similar to the “test” phase in feedback-loop models of self-regulation (e.g., Carver and Scheier 1981).
2980
W. Hofmann et al.
Internal Need States Appetite/ Deprivation
×
Attention to Tempting Food Cues
Myopia
IMPULSIVE PROCESSE S
Acute Alcohol Intake Emergence of Desires and Cravings for Food Strength and Salience Self-Monitoring
of Self-Regulatory Goals and Standards
Activation of Approach Tendencies to Consume Food
REFLECTIVE PROCESSE S Executive Control: Inhibition/Overriding
Caloric Intake
Fig. 186.1 A framework summarizing the key issues for possible acute alcohol effects on eating behavior as discussed in the present chapter. Straight lines indicate positive (facilitating) pathways. Dotted lines indicate negative (inhibitory) pathways. The framework holds that alcohol increases hunger and attention to salient food cues, thereby increasing the strength of cravings and desires for food. At the same time, alcohol appears to hamper reflective processing (underlined) in the service of self-regulatory goals. The figure does not show the direct link between alcohol consumption and caloric intake, indicating that alcohol in itself is a source of caloric intake that is typically not compensated for (see text)
In case discrepancies between standards and ongoing mental events are detected, an operative phase of executive control has to be set in motion if self-regulation is to be successful. Most forms of executive control can be described as either (a) the inhibition of a prepotent response (e.g., stopping an impulse to put a spoonful of mousse o chocolat into one’s mouth) or (b) the overriding of a prepotent response, that is, substituting the undesired behavioral tendency with an alternative behavior (e.g., eating some healthy celery sticks instead). Self-regulation can go awry with regard to all three factors: First, people may fail to keep a clear representation of their self-regulatory goal in working memory in the first place. Second, even with intact goal-representations, they may fail to notice discrepancies between their standards and their actual thoughts or behavior. And third, even with goal representation and self-monitoring intact, people may lack sufficient self-regulatory resources (Baumeister and Heatherton 1996) or fail to recruit available resources (Muraven and Slessareva 2003) in the service of self-regulatory goal pursuit.1 As outlined in more detail elsewhere (Hofmann et al. 2009; Hofmann et al. 2008a), situational and dispositional boundary conditions may strongly influence people’s self-regulatory success. Research on the self-regulation of eating has accumulated knowledge about conditions that disrupt the normal self-control of eating, especially for people who generally have high standards to restrain their caloric food intake (restrained eaters). Violation of dietary standards (Herman and Mack 1975) 1 Note that we sketch self-regulation here as a resourceful, intentional, and conscious process of goal-pursuit. It is possible, however, that self-regulation can become habitual and automatized via repeated training (e.g., Fishbach and Shah 2006). Even though not in the focus of this chapter, such automatic forms of self-regulation may be far less susceptible to alcohol’s detrimental influences on self-regulatory success described below. Common violations of self-regulatory standards suggest, however, that many individuals have not sufficiently transformed self-regulatory strategies into fully automatized routines.
186 The Great Disinhibitor: Alcohol, Food Cues, and Eating Behavior
2981
temporary depletion of self-regulatory resources (Vohs and Heatherton 2000), cognitive load (Ward and Mann 2000), or tempting food stimuli (Fedoroff et al. 1997; Stroebe et al. 2008) have all been demonstrated as disinhibiting factors. As will be discussed in the next section, acute alcohol consumption is a very powerful situational disinhibitor, which appears to have a detrimental impact with regard to all the three factors of successful self-regulation identified above. The present tug-of-war metaphor implies that the outcome of a self-regulatory conflict should not only hinge on the quality of self-regulatory goal pursuit, but also on the power of the temptation at hand. In the present framework we argue that the power of the temptation is jointly determined by the motivational need state of the organism (i.e., appetite) and the presence and features of food cues in the environment (see Fig. 186.1).2 Exposure to palatable food stimuli triggers hedonic thoughts about the palatability of this food in restrained but not normal eaters (Papies et al. 2007) and the attention of restrained eaters becomes glued to those food stimuli (Papies et al. 2008). The hedonic quality associated with the food cue can then lead to the intrusion of cravings and desires into consciousness (Kavanagh et al. 2005). Such cravings and desires occupy attentional and working memory resources and therefore directly compete with the representation of self-regulatory goals and standards in working memory (see Fig. 186.1). Once in the center of attention, the suppression of cravings and desires becomes difficult and can produce ironic effects such that the very thought that is supposed to be suppressed becomes even more strongly activated (Boon et al. 2002). In addition, cravings and desires are assumed to trigger the very behavioral schemas in memory that have proven to be useful in the past to fulfill the need (e.g., Strack and Deutsch 2004). Hence, tempting food cues in the environment may lead to the activation of behavioral tendencies to approach and consume the tempting object at hand (see Fig. 186.1). Whether these impulsive behavioral tendencies are enacted or not depends on whether sufficient executive control capacity can be mustered in order to inhibit or override the impulsive action tendencies. One straightforward implication of the described role of food cues is that, all else being equal, more vivid and attractive food cues should lead to stronger desires and hence to more disinhibited behavior. For instance, work on delay of gratification suggests that resistance to temptation is more difficult when the tempting object is literally present than when it is not visible (Mischel et al. 1996). More direct evidence for this conjecture stems form a series of studies by Fedoroff, Polivy, and Herman (Fedoroff et al. 1997). Participants in the experimental group were exposed over an extended period to the sight and smell of highly attractive food. When given access to palatable food afterward, restrained eaters more likely show increased disinhibited eating. Thus far, we have outlined a self-regulation framework that tries to understand and specify how impulsive, stimulus-triggered and reflective, goal-dependent forces clash in determining eating behavior. In the following section, we will illuminate potential mechanisms by which alcohol may affect the outcome of this struggle.
186.3.1 A lcohol’s Acute Effects on Impulsive Processing Appetite and Attentional Myopia to Food Cues There is reason to believe that alcohol may have a facilitating influence on the impulsive processing of tempting food stimuli. Specifically, alcohol may promote the strength of food cravings and desires by bolstering appetite and by enhancing attention to salient food cues (see Fig. 186.1). Indeed, one Because one focus of the present chapter is on external influences provided by present food cues, a second possibility not discussed further is that strong need states can also trigger from memory visual images of temptations that are not currently present, leading to similar cravings and desires as those emanating from present cues (for more details, see Kavanagh et al. 2005). 2
2982
W. Hofmann et al.
of alcohol’s psychoactive effects appears to consist in a general stimulating effect on appetite (Westerterp-Platenga and Verwegen 1999). This effect is supported both by research in animals and humans although the exact physiological mechanisms by which alcohol exerts this effect are not perfectly well understood (for a review, see Yeomans et al. 2003). Accordingly, it has been found both in experimental and field studies in humans that alcohol – at least above a certain threshold – increases subjective ratings of hunger (e.g., Caton et al. 2004). As a consequence, craving and desires for tempting food cues in the environment should increase and overeating (and drinking) should become more likely. A second acute effect by which alcohol may boost impulsive processing is attentional myopia (see Table 186.1 for key features). Originally proposed by Steele and Josephs (1990), the attentional myopia model holds that alcohol narrows the focus of attention and thoughts to the most salient stimuli in the environment. As a consequence, behavior will be disproportionally influenced by these salient cues rather to the exclusion of more distal cues. If salient cues favor the emergence of desires and cravings (e.g., rich food cues in one’s environment), more disinhibited eating behavior is expected. In the foodrich environments people often find themselves in or even create themselves, this is arguably often the case. However, if the salient cues attended to favor self-controlled behavior (e.g., a dieting plan pinned on one’s kitchen wall), alcohol myopia theory would even predict more self-regulated behavior. The predictions of the attentional myopia model for alcohol consumption have been confirmed in other domains of self-regulation such as sexual risk taking (e.g., MacDonald et al. 2000). Mann and Ward (2004) applied the attentional myopia model in a study on food consumption in chronic dieters employing cognitive load (rather than alcohol) as a method to manipulate attentional capacity. Supporting the model, dieters in the low attentional capacity condition showed relatively more disinhibited eating behavior when available cues (i.e., a milkshake) promoted consumption than dieters in the high attentional capacity condition. Conversely, dieters in the low attentional capacity condition showed more restrained behavior when high-fat food contents and dieting were made salient prior to testing the milkshake as compared to the high attentional capacity group (Mann and Ward 2004). Additional measures of thought content supported the idea that attentional load induced more hedonic thoughts in the milkshake-salient condition and more dieting-related thoughts in the diet-salient condition whereas participants in the no load condition exhibited a broader range of thoughts. In sum, alcohol’s acute effects on the stimulation of appetite as well as on the narrowing of the perceptive field to the most salient cues – often consummatory in nature – may jointly contribute to an increased intrusion of food desires and cravings into consciousness. These in turn may compete with self-regulatory goals in working memory for scarce representational resources (see the negative link between cravings and self-regulatory goals in Fig. 186.1) and lead to an activation of consumption-oriented behavioral action schemas. If not held in check properly by executive control functions, overeating will become more likely.
Table 186.1 Key features of alcohol myopia 1. Alcohol myopia (Steele and Josephs 1990) refers to the tendency of alcohol to focus an individual’s attention on proximal, highly salient stimuli or events in the environment to the disadvantage of more distant stimuli or events (hence the reference to myopia which is shortsightedness) 2. As a consequence, tempting stimuli (such as highly palatable food) in the environment may have a disproportionally strong influence on thought and emotion, leading to stronger cravings and desires and, ultimately, more impulsive behavior 3. However, to the degree that cues signalizing control (rather than temptations) are salient, alcohol may even foster control (e.g., MacDonald et al. 2000; Mann and Ward 2004)
186 The Great Disinhibitor: Alcohol, Food Cues, and Eating Behavior
2983
186.3.2 A lcohol’s Acute Effects on Reflective Processing: Goal Standards, Self-Monitoring, and Executive Control Next to alcohol’s boosting influence on impulsive processes in eating (and indirect influences on the representation of self-regulatory goal standards), alcohol also appears to directly hamper reflective processing in a multitude of ways (Hull and Slone 2004). In our framework, we will focus on three main factors: representation of goal-standards, self-monitoring, and executive control capacity. First, alcohol may impair the processes necessary to retrieve self-regulatory goal-standards into working memory and then in an active state necessary for conscious goal-pursuit. These two processes may together lead to weaker and less consistent representations of self-regulatory standards under alcohol intoxication (Baumeister et al. 1994). Some evidence indicates that alcohol, at least in higher doses, may hinder recall from long-term memory (Nelson et al. 1986). Hence, alcohol may hinder people from retrieving the very self-regulatory plans and intentions they had formed earlier. For instance, otherwise strongly represented intentions to diet may become temporarily suppressed. Also, it has been argued that successful self-regulation requires the maintenance of self-regulatory goals in working memory (Hofmann et al. 2008). Some research indicates that alcohol interferes with the maintenance function of working memory (Saults et al. 2007). Hence, self-regulatory goals may fade out of consciousness more easily under alcohol and cannot serve as well as comparison standards for self-monitoring and for mentally simulating the long-term consequences of one’s conduct. Second, there is strong evidence that alcohol negatively affects self-awareness (e.g., Hull and Bond 1986). Hence, intoxicated people may lose the ability to successfully attend to and monitor their behavior. With self-monitoring gone awry, discrepancies between the implications of an impulse to consume tempting food at hand and relevant goal standards to diet may simply go unnoticed. Third and perhaps most central from a self-regulatory perspective, alcohol may seriously reduce people’s capacity to inhibit or override prepotent impulses directed at the consumption of tempting food. That is, even though intoxicated persons may still be aware of existing conflicts between their impulses and their self-regulatory goals to some degree, they may nevertheless lack the cognitive resources for the kind of control – behavioral inhibition – that is necessary in order to stop impulsive action tendencies from becoming transformed into action or in order to resolve a conflict between two incompatible response tendencies (e.g., to eat chocolate versus celery sticks) in favor of one. This hypothesis is in line with a range of experimental studies demonstrating reduced inhibitory control after alcohol intake (e.g., Fillmore and Vogel-Sprott 1999). Most of these studies have employed the stop-signal paradigm (Logan et al. 1984) to examine the effects of alcohol on behavioral control. In a typical setup of such a task, participants have to react with quick, accurate choice responses to go-signal trials (e.g., categorizing letters via keypress). On a fraction of trials, however, a stop-signal (e.g., a brief auditory tone) accompanies a typical go-trial. In these trials, subjects are required to inhibit (i.e., suppress) the typical go-response. Inhibitory control is measured by the ability of the individual to inhibit prepotent go-responses at varying delay times with which the stop signal appears (with longer delays rendering inhibition of the go-response more difficult). Using such a paradigm, Fillmore and Vogel-Sprott (1999) demonstrated that a moderate dose of alcohol resulting in an average BAC (blood alcohol concentration) of 73 mg/100 mL reduced the drinker’s ability to inhibit behavior in response to stop-signals as compared to a placebo control group, whereas responses to go-signals were left unaffected. In other words, response inhibition seems to be much more vulnerable to the physiological effects of alcohol than response execution. Subsequent studies have shown that this effect depends on the presence of conflict between two competing action tendencies (Fillmore and Vogel-Sprott 2000). The above effect occurs when rewards are given both for acting (go-response) and for inhibition (stop-response) but disappears
2984
W. Hofmann et al.
when either one response is rewarded but the other is not. Taken together, these studies indicate (a) that alcohol’s effects on response inhibition may be most decisive in mental tug-of-war situations in which an approach-avoidance conflict exists and (b) that alcohol is likely to tip the scales in favor of the approach reaction in such cases. The cognitive capacities (representing standards, self-monitoring, inhibition) discussed in this section can all be subsumed among other functions under the umbrella term of executive cognitive functioning. A large body of neuropsychological data indicates that the neural substrate of executive cognitive functioning is provided by the prefrontal cortex, particularly its dorsolateral region (e.g., Giancola 2000). Hence, alcohol’s acute effects on a whole range of executive cognitive functioning abilities may be explained by its detrimental effects on this brain region in particular. In fact, alcohol seems to predominantly affect the glucose metabolism in the prefrontal cortex as corroborated by neuroimaging research (e.g., Volkow et al. 1995). The prefrontal cortex is the central part of a system of neural networks that are assumed to be involved in working memory functions and in the topdown control (i.e., inhibition and overriding) of subcortical structures such as the amygdala, insulae, mesolimbic cortex and the striatum (Bechara 2005), which provide the neuroanatomical basis for stimulus-triggered affective states and impulsive action tendencies. Hence, the observation that alcohol leads to less inhibited behavior accords well with the notion that it selectively impairs the very structure that is “in a logical neuroanatomical position to intercept and inhibit these lower brain impulses” (Giancola 2000, p. 585).
186.3.3 Supporting Empirical Evidence Taken together, the present framework predicts that alcohol may boost the strength of impulses (via food cravings and the activation of impulsive action tendencies) while at the same time reducing the capacity for reflective processing. In other words, alcohol should lead to an increased influence of impulsive processing and to a reduced influence of self-regulatory goal standards on eating behavior. These predictions were directly tested in a study by Hofmann and Friese (2008). At the beginning of the study, female participants completed a number of screening questionnaires including a measure of dietary restraint standards (Stunkard and Messick 1985) to assess individual differences in self-regulatory goal standards. We also assessed individual differences in impulsive affective reactions toward candy with a version of the Implicit Association Test (Greenwald et al. 1998). This task measures how quickly participants associate a given object (M&M’s candies in this case) with positive versus negative affect (for more details, see Hofmann and Friese 2008). High scores on the measure indicate more positive automatic affective reactions toward the food and hence can be taken as an indicator of impulsive processing. After the assessment phase, participants engaged in two different product tests. In the first product test, they consumed a drink that either consisted of orange juice with vodka (alcohol condition) or solely orange juice (control condition). An intermediate filler task gave the alcohol dose time to unfold its impact before participants tasted and rated candy in a second product test. A first finding was that participants in the alcohol condition on average consumed significantly more candy than participants in the control condition (Hofmann and Friese 2008). More importantly, however, we also investigated the degree to which candy consumption could be predicted by impulsive influences (i.e., automatic affect) versus reflective influences (i.e., dietary restraint standards) depending on whether participants had consumed alcohol or not. As expected, candy consumption was reliably predicted by automatic affect for participants in the alcohol condition as indicated by the positive slope between automatic affect and candy consumption in Fig. 186.2 (left panel). However, candy consumption was not predicted by automatic affect in the control condition (Fig. 186.2, left panel).
186 The Great Disinhibitor: Alcohol, Food Cues, and Eating Behavior
Candy Consumption
1.5
2985
Control
Alcohol
1 0.5 0 −0.5 −1
low (-1 SD)
high (+1 SD)
low (-1 SD)
high (+1 SD)
−1.5 Automatic Affect
Restraint Standards
Fig. 186.2 Candy consumption as a function of alcohol consumption, automatic affect toward candies, and dietary restraint standards. The y-axis represents z-transformed candy (m&m’s) intake during an unobtrusive taste and rate task with higher numbers indicating more candy consumption. Automatic affect toward the candy was assessed with an Implicit Association Test (Greenwald et al. 1998) as a measure of hedonic, impulsive reactions toward the candy. Restraint standards with regard to eating were assessed with a self-report scale. As can be seen from the two graphs, the relationship between automatic affect and candy consumption was stronger (as indicated by the steeper regression slope) for participants who had consumed alcohol before the taste and rate task (such that participants with high automatic affective reactions consumed most candy). In contrast, the expected negative relationship between restraint standards and candy consumption was more pronounced in sober participants (such that participants with high restraint standards consumed least candy) (Adapted from Hofmann and Friese 2008. With permission)
Conversely, dietary restraint standards were quite ineffective in participants who had consumed alcohol as indicated by the flat line in the right panel of Fig. 186.2. Dietary restraint, however, regulated candy consumption effectively for sober participants as indicated by the negative slope in Fig. 186.2 (right panel). Viewed in combination, these results yield strong evidence for the hypothesis that alcohol consumption fosters the influence of impulses on eating behavior while at the same time reducing the behavioral impact of self-regulatory goal standards as a controlling influence. Similar findings on the relative impact of impulsive versus reflective influences on consummatory behavior has been obtained in a number of studies using other moderators such as cognitive load (Friese et al. 2008), ego depletion (Hofmann et al. 2007), or individual differences in working memory capacity (Hofmann et al. 2008; Thush et al. 2008). The strong convergence across the alcoholrelated findings and the other moderators raises the question whether there may be a common element among these moderators that is responsible for producing functionally equivalent results. We think that the underlying connecting element may lie in the impairment of two central executive cognitive functions involved in impulse control: working memory capacity and inhibitory control (Giancola 2000; Hull and Slone 2004). Without support from these executive functions, bringing behavior in line with self-regulatory goals becomes more difficult. As reflective processing wanes, impulsive processes gain weight in determining the final outcome of a self-regulatory conflict.
186.3.4 Physiological Versus Alcohol Expectancy Effects We have argued that alcohol impairs self-regulation by boosting impulsive influences and interfering with reflective processing. One remaining issue is whether all of these effects are purely due to the physiological effects of alcohol, or whether at least parts of these effects are the result of cognitive
2986
W. Hofmann et al.
Table 186.2 Key features of alcohol expectancies 1. Alcohol expectancies refer to people’s beliefs about the effects of alcohol on thought, feeling, and action 2. Alcohol expectancies can be positive or negative 3. Typical positive alcohol expectancies comprise the beliefs that alcohol improves mood, reduces tension, increases sociability, and boosts self-esteem and assertiveness (e.g., Hull and Slone 2004) 4. Typical negative alcohol expectancies encompass the beliefs that alcohol decreases self-control, impairs intellectual performance, increases aggressiveness, and, at least in higher doses, creates unpleasant feelings of nausea. 5. It is generally assumed that positive alcohol expectancies increase the likelihood of alcohol consumption whereas negative expectancies decrease it. 6. Next to influencing alcohol consumption, alcohol expectancies have also been shown to mediate the effects of consuming alcohol. Thus, like a self-fulfilling prophecy, individuals may tend to behave in accordance with their alcohol expectations. Interestingly, it seems already sufficient for some of these effects to occur that individuals believe that they have consumed alcohol regardless of whether their actual drink really contained alcohol or not (Hull and Bond 1986).
mechanisms. More specifically, it has been argued that alcohol effects may be modulated by alcohol expectancies (see Table 186.2 for key features). Alcohol expectancies refer to the cognitive expectations people hold with regard to the consequences of alcohol consumption. A first assumption in the alcohol literature is that alcohol expectancies will influence the degree of alcohol consumption in a given situation of interest (Hull and Slone 2004). That is, people who harbor certain positive alcohol expectancies (e.g., “Alcohol makes me more relaxed when I am stressed”) will consume more alcohol if they are in a situation where these expectations are relevant (e.g., having to deliver a stressful speech) than people who do not hold these expectancies or people who hold negative expectancies (e.g., “Alcohol will make me lose my concentration”). Applied to the relationship between alcohol and eating, the degree of alcohol consumption in the presence of eating may be influenced by the type of expectancies people hold. For instance, dieters holding the negative expectancy that alcohol will impair their self-regulation should be more cautious in consuming alcohol than dieters who do not hold that expectation and these kinds of negative expectancies may aid the self-regulation of eating as far as the consumption of alcohol is concerned. On the other hand, some individuals may primarily harbor positive alcohol expectancies and these may then have detrimental effects on dietary control. For instance, individuals may believe that alcohol helps them to deal with negative affect such as the frustration of having to refrain from so many tasty foods they would otherwise like to eat or feelings of guilt upon violation of dietary schedules. As discussed in more detail below, such positive expectancies may stimulate alcohol consumption in the presence of tempting food cues and ultimately lead to violations of standards and even to patterns of spiraling distress (Baumeister et al. 1994). A second assumption concerning the role of alcohol expectancies is that alcohol expectancies can modulate the very effects of alcohol on behavior (Hull and Slone 2004). These expectancy effects may even be independent of the pharmacological effects of alcohol. That is, individuals may behave in accordance with their expectations whenever they believe that they have consumed alcohol, regardless of whether they actually received an alcoholic drink or a placebo (Hull and Bond 1986). For instance, it is plausible that individuals who expect alcohol to disinhibit their behavior in turn act more disinhibited than individuals who do not hold these expectations, all else being equal. This is because alcohol expectations may bias ongoing information processing systematically such that the expected behavioral outcome becomes more likely. Whether physiological and expectancy effects are additive or whether they interact is still an issue of ongoing debate (Hull and Bond 1986). There is, however, some indication that alcohol’s physiological effects on the impairment of control tend to be stronger among those expecting detrimental effects on control (Fillmore et al. 1998), although negative expectancies may also sometimes lead to increased efforts at compensating the expected
186 The Great Disinhibitor: Alcohol, Food Cues, and Eating Behavior
2987
impairments (Fillmore and Blackburn 2002). Possibly, whether such negative expectancies lead to decreased versus increased efforts at self-regulation following alcohol consumption may depend on how strongly people are motivated to summon an extra amount of willpower to counteract the expected effects. If control motivation is low, negative alcohol expectancies may lead to a pattern of acquiescence (Baumeister and Heatherton 1996) by which people actively give in to the temptation at hand (in the sense of the what-the-hell effect). If control motivation is high, however, negative expectancies may trigger increased efforts at self-regulation, similar to compensatory effects obtained with regard to other risk-situations such as ego depletion (Muraven and Slessareva 2003). Interestingly then, whereas negative alcohol expectancies about alcohol’s disinhibiting influence may help people to resist consuming alcohol in the first place, once people are led to consume alcohol (though a lapse of self-control or social pressure) the same expectancies may be hard to counteract and amplify the physiological effect.
186.4 Alcohol’s Long-term Effects on Executive Functioning Next to alcohol’s acute effects, there is mounting evidence that chronic alcohol consumption may have negative long-term effects on self-regulatory capacity (Volkow et al. 2004; Wiers et al. 2007). Specifically, chronic alcohol use may lead to frontal lobe dysfunctions as indicated by neurophysiological studies documenting decreased frontal lobe glucose utilization and blood flow in alcoholism (for a review, see Moselhy et al. 2001). These neuropathological changes may underlie the performance deficits in alcoholics as compared to nonalcoholics on a large range of executive cognitive functioning (Moselhy et al. 2001). Alcohol’s long-term deleterious effects may be particularly consequential during adolescence, a sensitive period of brain maturation that is fraught with both opportunities as well as risks for the development of self-regulatory capacities (Steinberg 2005). In other words, chronic alcohol use during adolescence may slow down or even disrupt the development of the very functions needed for adult self-regulatory competencies. Early alcohol use and its negative effects on self-regulatory ability may be one reason underlying the co-occurrence of alcohol abuse and eating disorders, characterized by impulsive eating such as bulimia nervosa, binge eating dis order, and obesity (e.g., Grilo et al. 2002).
186.5 T he Dynamic Interplay of Alcohol Consumption and Eating: Patterns of Snowballing Thus far, we have examined the impact of alcohol as a predictor/moderator on eating behavior as an outcome in a linear fashion. As typical for the psychology of self-regulation, reality is often more complex. Specifically, adopting a feedback-loop perspective on self-regulation (Carver and Scheier 1981), self-regulatory outcomes may often serve as inputs for new self-regulatory loops and so forth. In this section, we want to specifically highlight one such dynamical pattern that may help to explain why the combination of alcohol use and eating behavior may sometimes “snowball,” leading to a vicious circle of self-regulatory failure (Baumeister et al. 1994). An example of a potential snowballing pattern is illustrated in Fig. 186.3. As outlined above, acute alcohol intoxication may lead to disinhibited eating by boosting cravings and desires with regard to tempting food as well as by weakening resistance to these urges (Hofmann and Friese 2008). Failure to meet one’s dietary standards may in turn trigger emotional distress (especially feelings of guilt or remorse) as soon as the person
2988
W. Hofmann et al.
Alcohol Consumption
Overeating: Violation of SelfRegulatory Goals
Negative Feelings of Guilt and Remorse
Attempt to Regulate Negative Emotions and Thoughts
Positive AlcoholExpectancies
Fig. 186.3 A vicious cycle of alcohol consumption and overeating. The figure illustrates a possible “snow-balling” pattern of self-regulatory failure whereby alcohol consumption is both a cause of overeating (i.e., alcohol leading to disinhibited eating) as well as a consequence of overeating (i.e., alcohol as a means to cope with negative emotions due to overeating)
starts to self-monitor and reflect upon these failures. These negative emotional states, coupled with positive alcohol expectancies about its soothing effects, may lead the person into a new drinking episode in an attempt to cope with the negative feelings and thoughts (Baumeister et al. 1994), whereby the whole cycle may start all over again. Although clearly speculative, such dysfunctional patterns of misregulation may underlie the development of pathological forms of eating and account in part for the above-mentioned comorbidity between alcohol abuse and eating disorders. Once alcohol use enters the scene of eating disorders, it may be best understood as an integral part (i.e., both cause and effect) in an often vicious cycle of self-regulatory failure.
186.6 Conclusion: Alcohol and Eating As this chapter has shown, alcohol can be considered a great threat to the self-regulation of eating behavior. What makes alcohol a particularly dangerous threat is the fact that, as we have tried to show, alcohol is akin to an enemy that uses several different weapons at once. To recapitulate, these include the direct caloric impact of alcohol as a beverage not typically compensated for, boosting effects on impulsive processing which may lead to increased craving for tempting food, and detrimental effects on the self-regulatory mechanisms involved in inhibiting and overriding prepotent but unwanted action impulses (see also summary points). Even though the attack is a broad one, all of these direct and multiple indirect effects ultimately appear to converge toward the same target: enhanced caloric intake. Alcohol thus really seems to earn the title chosen for the present chapter. However, at least two qualifications need to be emphasized. First, alcohol’s effects seem to emerge mainly if there is a cognitive conflict in the first place (Fillmore and Vogel-Sprott 2000). Only if the hedonic, short-term prospects clash with highly valued long-term goals such as weight control or health concerns is alcohol likely to enter the ring as a decisive force. For this reason, restrained eaters are particularly at risk to experience alcohol’s negative consequences on self-regulatory goal pursuit. Second, these effects may be modulated by alcohol expectancies such that people who expect alcohol to have a disinhibiting effect on their eating behavior may have a good reason for abstaining from alcohol consumption, but, once alcohol is consumed may be at special risk for what-the-hell types of effects (unless an extra surplus of willpower is summoned to counteract these expectancy effects).
186 The Great Disinhibitor: Alcohol, Food Cues, and Eating Behavior
2989
186.7 Application to Other Areas of Health The present framework can also be applied to other areas dealing with the self-regulation of health behavior (Hofmann, Friese et al. 2008; Wiers et al. in press). Most centrally, alcohol may act as a disinhibiting influence on other appetitive domains such as sexual behavior and drinking itself by boosting impulsive processes and impairing reflective processing. In other domains of health behavior where impulsive influences may be less salient (e.g., the use of sunscreen, for instance), alcohol may still impact reflective processing and lead to more careless behavior because people may fail to represent and act in accordance with their long-term health goals and standards. However, to the degree that attention can be directed toward salient cues signaling control (Mann and Ward 2004) or to the degree that self-regulatory behavior can be delegated to more automatic and habitual routines, the great disinhibitor may loose some of its disruptive power on the self-regulation of human behavior.
Summary Points • Eating in humans can be regarded as self-regulatory behavior by which individuals often strive to attain certain self-regulatory goals such as weight control or healthy food intake • Eating as self-regulation typically involves a fragile conflict between self-regulatory goals and impulsive, appetitive influences on eating behavior the latter of which are driven by the interplay of internal need states and external food cues in the environment • The outcome of such self-regulatory conflicts can be strongly influenced by situational boundary conditions such as alcohol consumption. • Alcohol has multiple short-term effects on eating behavior. These comprise direct and indirect effects. • Direct short-term effects are given by the caloric load of alcohol itself are typically not compensated for. • Indirect short-term effects involve (a) an increase in appetite and (b) detrimental short-term effects on self-regulation as given by a weaker representation of goal standards, reduced self-monitoring, and reduced executive control. • These effects have a physiological basis but are also modulated by alcohol expectancy effects. • Alcohol also has long-term effects on cognitive functioning which may further undermine the self-control of eating behavior. These chronic effects may partly account for the observed comorbidity between alcohol abuse and impulsive eating disorders. • Alcohol use may be an integral part of dysfunctional self-regulatory loops whereby alcohol promotes overeating and overeating in turn promotes alcohol consumption. Such “snowballing” patterns may be a second factor accounting for the comorbidity between alcohol abuse and eating disorders.
Definition of Key Terms Self-regulation: An individual’s attempt to behave in accordance with his or her self-regulatory goal standards (e.g., dietary standards). Self-monitoring: A comparison of actual behavior with one’s self-regulatory goal standards. Discrepancies between the two indicate that further self-regulation is necessary.
2990
W. Hofmann et al.
Behavioral inhibition: Higher-order mental control process by which a prepotent action tendency is stopped. Alcohol myopia: Narrowing of attention to most salient cues in the stimulus environment due to alcohol consumption. Alcohol expectancies: People’s beliefs about the positive or negative effects of alcohol on thought, feeling, and action. Acknowledgments Preparation of this chapter was supported by grant HO 4175/3–1 from the German Science Foundation (DFG) and by a grant from the German Academic Exchange Service (DAAD) to Wilhelm Hofmann. Portions of this chapter were completed while the first author was a visiting researcher at Utrecht University. We thank Katie Lancaster for her valuable comments on a previous draft.
References Baumeister RF, Heatherton TF. Psycho Inq. 1996;7:1–15. Baumeister RF, Heatherton TF, Tice DM. Losing control: how and why people fail at self-regulation. San Diego: Academic Press; 1994. Bechara A. Nat Neurosci. 2005;8:1458–63. Boon B, Stroebe W, Shut H, Ijntema R. Brit J Health Psych. 2002;7:1–10. Carver CS, Scheier MF. Attention and self-regulation: a control-theory approach to human behavior. New York: Springer; 1981. Caton SJ, Ball M, Ahern A, Hetherington MM. Physiol Behav. 2004;81:51–8. Fedoroff IC, Polivy J, Herman CP. Appetite. 1997;28:33–47. Fillmore MT, Blackburn J. J Stud Alcohol. 2002;63:237–46. Fillmore MT, Carscadden J, Vogel-Sprott M. J Stud Alcohol. 1998;59:174–9. Fillmore MT, Vogel-Sprott M. Exp Clin Psychopharm. 1999;7:49–55. Fillmore MT, Vogel-Sprott M. J Stud Alcohol. 2000;61:239–46. Fishbach A, Shah JY. J Pers Soc Psychol. 2006;90:820–32. Friese M, Hofmann W, Wänke M. Br J Soc Psychol. 2008;47:397–419. Giancola PR. Exp Clin Psychopharm. 2000;8:576–97. Greenwald AG, McGhee DE, Schwartz JLK. J Pers Soc Psychol. 1998;74:1464–80. Grilo CM, Sinha R, O’Malley SS. Alcohol Res Health. 2002;26:151–60. Herman CP, Mack D. J Pers. 1975;43:647–60. Herman CP, Polivy J. A boundary model for the regulation of eating. In JA Stunkard, E Stellar, editors. Eating and its disorders. New York: Raven Press; 1984. p. 141–156. Herman CP, Polivy J. The self-regulation of eating. In RF Baumeister, KD Vohs, editors. The handbook of self- regulation: research, theory, and applications. New York: Guilford; 2004. p. 492–508. Hofmann W, Friese M. J Abnorm Psychol. 2008;117:420–7. Hofmann W, Friese M, Strack F. Pers Psychol Sci. 2009;4:162–76. Hofmann W, Friese M, Wiers RW. Health Psychol Rev. 2008a;2:111–37. Hofmann W, Gschwendner T, Friese M, Wiers RW, Schmitt M. J Pers Soc Psychol. 2008;95:962–77. Hofmann W, Rauch W, Gawronski B. J Exp Soc Psychol. 2007;43:497–504. Hull JG, Bond CF. Psychol Bull. 1986;99:347–60. Hull JG, Slone LB. Alcohol and self-regulation. In RF Baumeister, KD Vohs, editors. The handbook of self-regulation: research, theory, and applications. New York: Guilford; 2004. p. 466–491. Hulshof T, de Graaf C, Weststrate JA. Appetite. 1993;21:273–86. Kavanagh DJ, Andrade J, May J. Psychol Rev. 2005;112:446–67. Logan GD, Cowan WB, Davis KA. J Exp Psychol Hum Percept Perform. 1984;10:276–91. Lowe MR, Butryn ML. Physiol Behav. 2007;91:432–9. MacDonald TK, Fong GT, Zanna MP, Martineau AM. J Pers Soc Psychol. 2000;78:605–19. Mann T, Ward A. J Abnorm Psychol. 2004;113:90–8. Mattes RD. Physiol Behav. 2006;89:66–70. Mischel W, Cantor N, Feldman S. Principles of self-regulation: The nature of will-power and self-control. In ET Higgins, AW Kruglanski, editors. Social psychology: handbook of basic principles. New York: Guilford Press; 1996. p. 329–360.
186 The Great Disinhibitor: Alcohol, Food Cues, and Eating Behavior
2991
Moselhy HF, Georgiou G, Kahn A. Alcohol Alcohol. 2001;36:357–68. Muraven M, Slessareva E. Pers Soc Psychol B. 2003;29:894–906. Nelson TO, McSpadden M, Fromme K, Marlatt GA. J Exp Psychol Gen. 1986;115:247–54. Papies E, Stroebe W, Aarts H. J Exp Soc Psychol. 2007;43:810–7. Papies EK, Stroebe W, Aarts H. J Exp Soc Psychol. 2008;44:1283–92. Pinel JPJ, Assanand S, Lehman DR. Am Psychol. 2000;55:1105–16. Saults JS, Cowan N, Sher KJ, Moreno MV. Exp Clin Psychopharm. 2007;15:576–87. Schachter S. Emotion, obesity, and crime. New York: Academic; 1971. Steele CM, Josephs RA. Am Psychol. 1990;45:921–33. Steinberg L. Trends Cogn Sci. 2005;9:69–74. Strack F, Deutsch R. Pers Soc Psychol Rev. 2004;8:220–47. Stroebe W. Dieting, overweight and obesity: self-regulation in a food-rich environment. Washington, DC: American Psychological Association; 2008. Stroebe W, Mensink W, Aarts H, Schut H, Kruglanski A. J Exp Soc Psychol. 2008;44:26–36. Stunkard AJ, Messick S. J Psychosom Res. 1985;29:71–83. Thush C, Wiers RW, Ames SL, Grenard JL, Sussman S, Stacy AW. Drug Alcohol Depend. 2008;94:116–24. Vohs KD, Heatherton TF. Psychol Sci. 2000;11:249–54. Volkow N, Wang G, Doria JJ. Alcohol Health Res World. 1995;19:296–9. Volkow ND, Fowler JS, Wang GJ. Neuropharmacology. 200447 Suppl 1:3–13. Ward A, Mann T. J Pers Soc Psychol. 2000;78:753–63. Westerterp-Platenga MS, Verwegen C. Am J Clin Nutr. 1999;69:205–12. Wiers RW, Bartholow BD, van den Wildenberg E, Thush C, Engels RCME, Sher KJ, et al. Pharmacol Biochem Be. 2007;86:263–83. Wiers RW, Houben K, Roefs A, De Jong P, Hofmann W, Stacy AW. Implicit cognition in health psychology: Why common sense goes out of the window. In B Gawronski, BK Payne, editors. Handbook of implicit social cognition. New York: Guilford Press; 2010;463–488. Yeomans MR, Caton S, Hetherington MM. Curr Opin Clin Nutr Metab Care. 2003;6:639–44.
wwwwwwwwwwwwwww
Chapter 187
Brain Atrophy in Alcoholics E. González-Reimers and F. Santolaria-Fernández
Abbreviations AMPA BMI cAMP CREB CT DNA DEXA EEG GABA HT MDA MRI NF-kB NADP NADH NMDA TNF
Alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid Body mass index Cyclic adenosine monophosphate cAMP responsive element binding-protein Computerized tomography Desoxiribonucleic acid Dual-energy x-ray absorptiometry Electroencephalogram Gamma-amino-butyric acid Hydroxytryptamine Malondialdehyde Magnetic resonance Nuclear factor kappa B Nicotinamide adenine dinucleotide phosphate Reduced nicotinamide adenine dinucleotide N-methyl-d-aspartate Tumor necrosis factor
187.1 Introduction Ethanol exerts deleterious effects on the brain, both acutely and chronically. Acute alcohol intake affects brain function in a dose-dependent manner, so that at lower doses of ethanol induces euphoria and altered mood, in greater amounts it impairs speech and motor function, whereas higher doses affect respiratory and cardiovascular functions and may result in death. These risks are also present in binge drinkers; in addition, these alcoholics also develop organic brain damage and cognitive dysfunction. In fact, with time, chronic alcohol consumption leads to dependence and frank addiction,
E. González-Reimers (*) Servicio de Medicina Interna, Hospital Universitario de Canarias, Ofra s/n 38320, La Laguna, Tenerife, Canary Islands, Spain e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_187, © Springer Science+Business Media, LLC 2011
2993
2994
E. González-Reimers and F. Santolaria-Fernández
whereas progressive cognitive dysfunction and brain atrophy also ensue, even with moderate amounts of ethanol consumption. Although not the only neurological alteration suffered by the chronic alcoholic patient, brain atrophy is frequently observed; it often affects relatively young people, and severely impairs mood, memory, judgement and other cognitive skills, so that it constitutes a major challenge for patients, physicians and society. This chapter focuses on some of the factors involved in the pathogenesis of brain atrophy, especially those related with altered nutrition in the wide sense of the term.
187.2 The Clinical Problem Drinking patterns differ across countries and also within countries, according to age, sex, social class and lifestyle. In rural areas, excessive drinkers, mainly consumers of wine, develop cirrhosis after many years of regular consumption. At the other end of the spectrum, socially ostracised, homeless, undernourished alcoholics, usually living in an urban environment, are more prone to develop vitamin deficiency and Wernicke–Korsakoff syndrome. Another group of drinkers is constituted by binge drinkers, more frequently found among young people living in urban areas; sometimes binge periods coexist with heavy smoking and consumption of illicit drugs. Therefore, the neurocognitive alterations usually found in alcoholics are heavily influenced by lifestyle, drinking pattern, and coexistence of liver disease and/or other neurotoxic factors, making it difficult to discern if the observed effects are due to ethanol itself, to liver dysfunction, coexisting malnutrition, or depend more heavily on the effects of other drugs, such as tobacco or illicit drugs. In addition, different educational levels – roughly associated with drinking patterns – and previous behavioural disorders, which in many alcoholics act as contributing factors to alcohol dependence, complicate the scenario. It has been reported that more than 80% of those adolescents with alcohol abuse or dependence also showed another psychopathology (Thatcher and Clark 2005). Moreover, genetic predisposition, and the ability of ethanol to modulate the activity of at least 58 genes involved in neuronal activities (38 downregulated, 15 up-regulated, 5 up-/down-regulated), potentially subjected to polymorphism (Wang et al. 2009), further complicates our understanding and knowledge of the mechanisms, which underlie brain alterations in alcoholics (Table 187.1). Central nervous system affectation in chronic alcoholics includes brain atrophy, with variable degrees of cognitive dysfunction; the Wernicke–Korsakoff syndrome, related to thiamine deficiency, and cerebellar cortical degeneration, leading to dysmetria and ataxia, but also playing a role in cognitive impairment. Although pellagroid dermatitis is sometimes observed in malnourished alcoholics, ataxia and dementia due to vitamin B3 deficiency is uncommon. Other syndromes, such as Marchiafava–Bignami syndrome, or acute pontine myelinolysis, are less frequently encountered, but should be considered in the clinical evaluation of alcoholic patients with impaired mood and/or judgement, obtundation, stupor or frank coma, with or without variable symptoms related to cerebellar dysfunction. In addition, alcoholics show reduced cerebral blood flow (Christie et al. 2008), and are at increased risk of stroke and cerebral trauma. Furthermore, neurological effects of acute ethanol intoxication and of ethanol withdrawal, and the less frequently described ethanol- and tobaccorelated amblyopia, complete the clinical syndromes derived from affectation of the central nervous system in the alcoholics. Thus, the spectrum of ethanol-induced brain alterations is wide, and the underlying mechanisms are incompletely understood, especially taking into account that the pathogenesis of central nervous system dysfunction may differ according to the nature and intensity of exposure, i.e., acute alcohol intoxication, chronic alcohol consumption and binge-drinking. Other factors, such as malnutrition or liver failure or, possibly, even the type of alcoholic beverage, may
187 Brain Atrophy in Alcoholics
2995
Table 187.1 Key points of brain atrophy Anatomoclinical forms of chronic Brain cortical and subcortical atrophy brain damage Cerebellar degeneration Toxic amblyopia Pellagra Wernicke–Korsakoff encephalopathy Marchiafava–Bignami disease Central pontine myelinolysis Increased prevalence of stroke and cerebral trauma Clinical features of brain atrophy May vary from subtle alterations (Impaired verbal problem solving, abstracting abilities, visual–spatial performance, verbal memory and the ability to adapt problem-solving strategies to changing requirements) to frank dementia Mostly reversible with prolonged ethanol withdrawal Underlying anatomic lesions Neuronal death White matter alterations (demyelination and axonal death) Proposed mechanisms Direct effect of ethanol Cytokine (especially TNF-a) mediated neuroinflammation. Oxidative damage (disbalance between antioxidants and prooxidants) Thiamine deficiency Protein deficiency and malnutrition Excitotoxicity Coexisting liver disease
contribute. Importantly, the risk of dementia is increased in binge-drinkers (Jarvenpaa et al. 2005), as well as the risk of developing chronic alcoholism. In any case, brain dysfunction is frequently observed in alcoholics admitted to general care services due to organic problems. Despite some discrepancies regarding its true prevalence, cognitive dysfunction, at least in its mildest forms, has been documented in more than 50% of alcoholic patients. Although verbal reasoning and verbal learning skills are relatively preserved, verbal problem solving, abstracting abilities, visual–spatial performance, verbal memory and the ability to adapt problem-solving strategies to changing requirements become more or less severely impaired. In a study of 100 alcoholics who had been sober for at least 21 days before inclusion in the analysis, cognitive efficiency (assessed by a battery of tests including perceptual motor skills, visuospatial processing, learning and verbal and non-verbal problem solving) was significantly impaired compared with 80 controls (Nixon et al. 1995). In that study, alcoholics also showed more depression, anxiety and symptoms related to childhood behavioural disorders, attention deficit disorders and conduct disorders, than controls, but the cognitive impairment could not be attributed to the coexisting behavioural dysfunction. Thus, although some conditions during childhood, such as conduct disorders, stress, anxiety and depression, or attention-deficit/hyperactivity disorders may predispose to ethanol abuse in later years, cognitive impairment takes place without the necessity of any other predisposing situation. Classically, cognitive impairment was attributed to frontal lobe lesions and their associated connections with other cortical and subcortical regions, but in recent years it has been shown that alcohol-mediated brain damage affects many other parts of the central nervous system, including hippocampus, corpus callosum (alterations in attention and executive functions have been related to subtle alterations in corpus callosum, Pfefferbaum et al. 2006), and pathways connecting the cerebellum and other parts of the central nervous system, such as cerebellothalamocortical and cerebelloponto cerebellar pathways.
2996
E. González-Reimers and F. Santolaria-Fernández
187.3 Neuropathology Ethanol provokes generalized brain shrinkage (Fig. 187.1), largely due to loss of cerebral white matter, but also to a reduction in the cortical grey matter, especially observable with magnetic resonance (MRI) or computerized tomography (CT). Regarding white matter, the maximum atrophy has been observed in the prefrontal white matter (Kril et al. 1997), corpus callosum (Pfefferbaum et al. 2006) and cerebellum (Fitzpatrick et al. 2008). In general, the degree of brain atrophy has been related with the intensity of ethanol consumption. Ding et al. (2004) observed a relation between the amount of ethanol ingested and the area of the fluid-filled spaces in the brain. Harper showed an increase in the pericerebral space in men drinking more than eight drinks per day (Harper 2009). In the study by Kubota et al. (2001) of 1,432 non-alcoholic individuals, moderate alcohol consumption (less than 50 g/day) did not seem to affect brain volume; however, in the Framingham study Paul et al. (2008) did find a relation between any amount of ethanol consumption and brain atrophy. White matter regions in the frontal lobe were those mostly affected in alcoholics, especially among patients with Wernicke’s encephalopathy, but also showing a relation with ethanol intake, estimated as maximum daily ethanol consumption (Kril et al. 1997). In a study on 43 alcoholics we found a significant (p = 0.007) relationship between the duration of alcohol intake and the bifrontal index, suggesting a relationship between ventricular dilatation and ethanol consumption (Fig. 187.2). Also, a relation was found between cortical atrophy assessed by CT and duration of ethanol consumption (rho = 0.37, p = 0.014). However, no relation was found between the total life-long amount consumed (expressed as kilogram ethanol/kilogram body weight) and the presence of brain atrophy. Changes in white matter probably include both alterations in myelination and axonal integrity (Harper 2009), and it is possible that a synergist effect between ethanol consumption and thiamine deficiency also exists. These changes are reversible, at least in part, after ethanol withdrawal. Therefore, in alcoholic brain damage, there may be two components, one reversible and one permanent. In the latter, neuronal death would provoke axonal degeneration and white matter shrinkage, whereas structural alterations of myelin would be reversible after alcohol abstinence. Bartsch et al. (2007), in a study of 15 patients who completed abstinence, showed a significant brain volumetric gain, whereas no changes were observed among 10 controls. Changes in abstinent alcoholics were more marked in the superior vermis, perimesencephalic infratentorial and supratentorial periventricular borders, and
Fig. 187.1 Computerized tomography (CT) of the brain, showing some parameters relevant for calculation of indices of brain atrophy and/or ventricular dilatation. a–c CT of the brain, showing the diverse parameters utilised in this study. Panel A: BF/BF1= bifrontal index; B/B1= bicaudate index; BF/E=Evans index. B+ BF=Huckmann’s digit. Panel B: H/H1=Cella media index. Panel C: Cortical atrophy is defined as the sum of the width of the four widest sulci at the two highest scanning levels, two of which are shown in the figure
187 Brain Atrophy in Alcoholics
Relationship between bifrontal index and duration of ethanol consumption �
�
r=0.40; p=0.007
0,50
Bifrontal index
Fig. 187.2 Relationship between ethanol intake and bifrontal index. Relationship between ethanol intake and bifrontal index, showing that the longer the duration of ethanol intake, the more intense the ventricular dilatation
2997
�
��
�
0,40
� �� � � �
0,30 �
0,20
�
�
� �� � �
� �
�
�� �
�
�� � ��
�
�
�
�
� �
�
�
�
10,00 20,00 30,00 40,00 50,00 60,00
Months
frontomesial and frontoorbital edges, and are not explainable only on the basis of simple rehydratation. These changes were accompanied by regain in cerebral choline, suggesting recovery from white matter damage potentially consistent with astrocyte regrowth and remyelination. Also, an increase in frontomesial N-acetylaspartate was observed. Cerebellar choline was significantly related with brain volume recovery, and frontomesial N-acetyl aspartate increase showed a significant correlation with improvement in attention and concentration, although not with other neuropsychological tests (auditory verbal learning test, standard progressive matrices, vocabulary test). In another study also performed with MRI and an automated three-dimensional method (boundary shift integral), the most rapid volume recovery was observed among those with the greatest baseline brain shrinkage and drinking severity. The study was performed with 18 alcoholics, and, in the first month of abstinence, tissue gain was apparent around the third, fourth and lateral ventricles, cerebellum, pons, hippocampi and the boundaries of frontal, parietal and superior temporal lobes. This study also showed a rapid reversal of volume gain to baseline levels even with consumption of relatively small amounts of ethanol (five drinks per day during 9 days) in reincident patients (Gazdzinski et al. 2005). These data strongly support the direct effect of ethanol on brain alterations and stresses the benefits of ethanol withdrawal, confirming early reports, which showed that even short-term (1 month) abstinence increases cortical grey matter volume, whereas long-term abstinence (1 year) leads to shrinkage of the enlarged ventricles. Working memory, visual spatial abilities and motor abilities also improve (Crews and Nixon 2009). Frontal cortical grey matter is also affected, with neuronal loss and dendritic shrinkage, which are reversible after abstinence (Harper 2009). Indeed, as early as 2 days after intragastric administration of large amounts of ethanol to rats, neuronal death is already observed in several areas of the brain, accompanied by reduced hippocampal neurogenesis. Changes are more pronounced in adolescent rats and in those genetically predisposed (Crews and Nixon 2009). Detailed autopsy studies have revealed that neuronal loss was found in the superior frontal association cortex, while no loss was observed in the motor cortex. Loss of neurons was also observed in the hypothalamus, affecting the supraoptic and paraventricular nuclei, especially among those who consumed more than 100 g ethanol/day (Harding et al. 1996). Together with increased neuronal death, it is possible that ethanol also inhibits neuronal regeneration, reducing the survival of progenitor neuronal cells and also blunting the maturation and growth of the dendritic arbor of these cells (Crews and Nixon 2009).
2998
E. González-Reimers and F. Santolaria-Fernández
Cerebellum alterations are also widespread among the alcoholics, especially among heavy drinkers during 10 or more years (Andersen 2004). Prevalence studies of cerebellar degeneration among alcoholics yield variable results depending upon the criteria employed (autopsy, radiological (MRI/ CT scan) or clinical data), but may be estimated to be around 25–30%. In a series of 36 alcoholics with brain atrophy attended at our hospitalization unit, cerebellar atrophy affected 33% (GarcíaValdecasas-Campelo et al. 2007). Cerebellar affectation is characterized by shrinkage of the anterior superior cerebellar vermis and loss of Purkinje cells in the vermis and lateral lobes of cerebellum. These alterations cause ataxia and instability. Although some data suggest that ethanol alone does not cause cerebellar damage but only when coexisting with thiamine deficiency, experimental evidence does support that ethanol treatment and withdrawal reduces the number of Purkinje cells (Andersen 2004). In addition, white matter atrophy, particularly affecting the vermis, has also been described in chronic alcoholics. Cerebellar alterations in alcoholics may not only affect motor functions, but may also be related to cognitive and emotional disturbances – the so-called cerebellar cognitive affective syndrome (Fitzpatrick et al. 2008). These manifestations depend on the disruption of the cerebellocerebral circuitry, which connects the cerebellum and cerebral associative and paralimbic areas. The disruption of this circuitry provokes deficits in executive functioning and visuospatial skills, language alterations and disturbed personality and affective behaviour, all of which may add to the alterations due to direct frontal lobe affectation observed in the alcoholics. Cerebellar atrophy may also recover with abstinence (Cardenas et al. 2007).
187.4 Pathogenesis 187.4.1 Inflammation and Oxidative Damage Detailed studies using a binge drinker rat model suggest that neuronal death occurs during ethanol intoxication, and not during withdrawal periods (Crews and Nixon 2009). Ethanol-induced brain damage may be related to oxidative stress derived from proinflammatory enzymes activated during ethanol intoxication. Inflammation not only causes neuronal death, but also inhibits regeneration. Transcription factors, such as cyclic adenosine monophosphate (cAMP) responsive element bindingprotein (CREB), regulate the transcription of pro-survival factors, protecting neurons from excitotoxicity and apoptosis (Mantamadiotis et al. 2002).On the other hand, nuclear factor kappa B (NF-kB) is involved in proinflammatory and immune responses, is activated by cytokines, oxidative stress and glutamate, and promotes transcription of genes involved in the synthesis of proinflammatory cytokines such as tumor necrosis factor (TNF)-a. Ethanol has been associated with an increase in DNA binding to NFkB and a decrease in DNA binding to CREB, at least in hippocampal entorhinal cortex slice cultures (Zou and Crews 2005). By this way it is possible to explain the increase of reduced nicotinamide adenine dinucleotide phosphate oxidase (NADPH-oxidase) and other enzymes involved in the production of reactive oxygen species and the increase of oxidative damage mediated by ethanol. In this regard, there is the possibility that oxidative damage mediate brain atrophy (Nordmann 1994). In addition, several studies point out that flavanoids and other antioxidants exert protective effects in patients with Alzheimer’s disease and there are even some reports suggesting that ingestion of wine in moderate amounts may protect against dementia, an effect that could be related to antioxidants present in wine. Neuroinflammation may be related to neuronal degeneration. Microglial and asctrocyte responses to various brain insults include release of highly toxic products, such as reactive oxygen species, nitric
187 Brain Atrophy in Alcoholics
2999
oxide, excitatory amino acids, cytokines and complement components, which cause neurodegeneration. Increased proinflammatory cytokines and chemokines expression suggest the presence of a brain proinflammatory status in several models of chronic brain damage. Experimental data have shown that an acute ethanol load cause lipid peroxidation in several areas of the brain (Uysal et al. 1989). Several studies suggest that antioxidants protect the brain from binge ethanol-induced damage and also tend to reverse ethanol-induced inhibition of neurogenesis (Crews et al. 2006). All these data, together with the well-known effect of TNF-a enhancing oxidative stress, strongly suggest a major role of this cytokine in neurodegeneration. Indeed, it is possible that part of increased lipid peroxidation is mediated by pro-inflammatory cytokines (Crews et al. 2006). In alcoholic subjects and experimental animals, increased intestinal permeability leads to a continuous exposure of Kupffer cells to intestinal antigens; these cells become activated and secrete pro-inflammatory cytokines, TNF-a being the most abundant of these. Increased TNF-a enters the brain; there, its increase lasts longer than in other tissues and exerts neurotoxic effects, including demyelination and neurodegeneration. The different behaviour of TNF-a after alcohol exposure, with rapid decrease in liver and serum, but persistence in brain, could be important in binge drinkers, with progressive neurodegeneration (Crews et al. 2006). TNF-a potentiates glutamate excitoxity, linked to excessive glutamate activation of N-methyl-d-aspartate (NMDA) receptor, since it reduces glial glutamate transporter activity and thus may also play a role in neurodegeneration (Zou and Crews 2005). Increased glutamate is related to an increase in the desire to consume ethanol. Therefore, increased TNF-a would be not only related to brain damage, but also to alcohol dependence. In accordance with these statements, in a study on 34 alcoholics we found significant relationships between TNF-a values (in percentiles) and several indices of cortical atrophy, such as bifrontal index, Evan’s index and Huckmann’s digit (Fig. 187.3). In acutely intoxicated rats (5 g/kg ethanol, intraperitoneally), raised liver and brain malondialdehyde (MDA) levels were observed, reaching maximum differences with respect to controls 4 h after ethanol injection (Uysal et al. 1989), a result which lends support to the hypothesis of oxidative damage as a major pathogenic mechanism in ethanol-related brain damage. However, rats subjected to chronic ethanol ingestion (5 g/kg in saline for 6 weeks, although details on how rats ingested ethanol and the amount consumed are not provided) did not show changes in brain MDA or glutathione, despite an increase in both liver MDA and glutathione levels.
Relationship between TNF-alpha and Evans´ index
b
Relationship between TNF-alpha (percentiles) and bifrontal index
�
0,60
0,30
� � � � �
0,20
� � � � � � � �
� �
� � � �
0,70
� �
Bifrontal
Evans
�
0,40
0,30
� � � � �
� � � � � �
� �
� �
� �
� �
� � � � � �
�
�
1,00
2,00
3,00
TNF percentiles r=0.37; p=0.036
4,00
�
� �
0,60 � �
0,50 0,40 0,30
�
0,20
� �
� �
� �
0,40
Relationship between TNF-alpha and Huckmann´s index 0,80
� �
0,50 0,50
c
Huckmann
a
� � � �
� �
�
� � � �
�
� �
�
� �
�
1,00
� � �
� �
2,00
3,00
TNF-a Percentiles rho=0.37; p=0.03
4,00
1,00
2,00
3,00
4,00
TNF percentiles r=0.35; p=0.041
Fig. 187.3 Relationships between ventricular dilatation and Tumor necrosis factor (TNF)-alpha. Relationship between TNF-alpha (in percentiles) and several parameters related with ventricular dilatation (a = Evans’ index; b = bifrontal index; c = Huckmann’s digit). In the three panels it is evident that more intense degrees of ventricular dilatation are observed among those patients with higher TNF-alpha percentiles
3000
E. González-Reimers and F. Santolaria-Fernández
187.4.2 Trace Elements In close relation with oxidative damage, some trace elements could play a role in brain atrophy of alcoholics. Theoretically, some of them, such as zinc, copper, manganese and selenium, are involved in the antioxidant systems, so that their deficiency could explain some of the alterations observed. Chronic alcoholics show low serum zinc and selenium levels, and some controversy exists regarding copper. However, the relation of changes in trace elements with brain damage has been scarcely explored. There are several studies dealing with the relation between brain atrophy and copper levels. Squitti et al. (2002) found a relation between serum copper and brain atrophy in patients with Alzheimer’s disease, so that levels over 1.02 mg/dl discriminated individuals with Alzheimer’s disease from controls, with a sensitivity of 60% and a specificity of 95%. Toxicity of copper was probably related to the induction of peroxidation, but treatment with d-penicilamine, despite showing a lowering effect on lipid peroxidation, did not alter cognitive decline. On the other hand, in patients with Wilson disease, midbrain atrophy shows a correlation with neurological symptoms, both in probable relation with copper toxicity (Strecker et al. 2006). We also analysed the relationship between serum copper and brain atrophy, but our (preliminary) results do not support the existence of a relationship between copper and brain atrophy in alcoholics. Increased iron has been related with early edema and, later, brain atrophy, after intracerebral bleeding, a process which can be reversed by deferoxamine. Indeed, after bleeding, once haemoglobin is degraded, iron concentration increases severalfold (Hua et al. 2007), and is capable of causing edema and brain damage by oxidative stress. Maschke et al. (2005) failed to find any relation between vermal atrophy and dentate iron concentrations – assessed by MRI – in alcoholics, but in a study of 29 alcoholics we found that serum ferritin is significantly higher among those with cerebellar atrophy (Fig. 187.4). No differences, however, were observed with serum iron. Thus, ferritin being an acute phase reactant, the possibility exists that the difference is more in relation with inflammation than with a direct effect of iron storage. In Alzheimer’s disease, as well as in alcoholic brain damage, controversy also exists regarding the role of zinc (Menzano and Carlen 1994). Low zinc may be involved in neuronal apoptosis, and in an increase in NMDA-gated neurotoxicity (Peters et al. 1987). It is hypothesized that low zinc levels may Cerebellar atrophy/ferritin
Fig. 187.4 Relationship between serum ferritin and cerebellar atrophy. Relationship between serum ferritin and cerebellar atrophy. Patients (12) with cerebellar atrophy (1, left bar) showed higher ferritin levels (in ng/ml) than those (17 patients) without cerebellar atrophy (2, right bar)
ferritin
750,00
500,00
250,00
0,00
1,00
2,00
t=2.31, p=0.029
187 Brain Atrophy in Alcoholics
3001
Fig. 187.5 Relationship between serum zinc and Evans’index. Direct relationship between serum zinc and Evans’index. Although the number of cases is small, this result does not support a role of low zinc in brain atrophy
Relationship between serum zinc and Evan´s index 0,60
Evans
0,50
0,40
0,30
0,20
0,00
1000,00
2000,00
3000,00
4000,00
Serum zinc (mg/kg) rho=0.64; p=0.002
account for chronic alcoholic brain damage as well as alcohol withdrawal seizures, and a possible link between low zinc levels and hypercortisolism, as well as between hypercortisolism and brain dysfunction has been postulated (Menzano and Carlen 1994). However, this link is still merely hypothetical, and some results are inconsistent. For instance, we found a direct relationship between Evan’s index and serum zinc (Fig. 187.5), a result that directly contrasts with the hypothesis linking brain atrophy to low zinc levels. We also failed to find any relation between cortisol levels and brain atrophy. Selenium is another important element in the antioxidant system, since it is a cofactor of the glutathione-peroxidase system. It has been also shown that selenium is a necessary cofactor for normal brain development. Moreover, selenium depletion is associated with decreased activity of selenium-dependent enzymes and enhanced neuronal loss in animal models (Schweizer et al. 2004). Although low serum selenium levels have been repeatedly reported in alcoholics, the role of selenium deficiency in alcohol-related brain atrophy is largely speculative and results are not consistent. For instance, in a study of 21 alcoholics, serum selenium was directly related with bifrontal index (i.e., the more selenium, the more intense atrophy, Fig. 187.6), a result which does not support any role of low selenium on brain atrophy.
187.4.3 Vitamins Deficiency of several vitamins, such as ascorbic acid, vitamin A, and vitamin E (due to antioxidant capabilities); cyanocobalamin, vitamin B6 and folic acid (by their relation with hyperhomocysteinaemia), and thiamine (as an etiologic factor of the Wernicke–Korsakoff encephalopathy) may be involved in the pathogenesis of brain atrophy in alcoholics. As the above-mentioned trace elements, vitamin E may also protect against lipid peroxidation. Chronic ethanol feeding results in increased vitamin E demand by the liver (Nordmann 1994). Although the role of vitamin E in the central nervous system is uncertain, it has been shown that
3002
E. González-Reimers and F. Santolaria-Fernández
Fig. 187.6 Relationship between serum selenium and bifrontal index. Direct relationship between serum selenium and bifrontal index
Relationship between serum selenium and bifrontal index
Bifrontal index
0,50
0,40
0,30
0,20 50,00
100,00
150,00
200,00
Serum selenium (mg/kg) rho = 0.44; p = 0.048
vitamin E supplementation may decrease ethanol-induced lipid peroxidation in rat cerebellum (Nordmann 1994). A condition that must be considered in the study of brain disturbance in alcoholics is the Wernicke– Korsakoff syndrome, due to thiamine deficiency, which may also appear in nonalcoholics with protracted vomiting and/or different degrees of malnutrition. Typically, the clinical picture improves with thiamine supplementation. However, in many alcoholics – reaching as much as 80% of those affected by Wernicke’s encephalopathy – Korsakoff psychosis ensues (Fitzpatrick et al. 2008). This is characterized by anterograde amnesia, cognitive impairment and confabulation, reduced affect, visuospatial alteration and also altered problem-solving capacity. Mammilary bodies and the dorsomedial nucleus of the thalamus are well-described anatomic targets of this entity, although others also describe neuronal loss in the anterior thalamic nuclei (Harding et al. 2000). In alcoholics, thiamine deficiency may result from inadequate intake, impaired absorption, reduced liver storage, and decreased transformation of thiamine in its active form. Thiamine deficiency may act synergistically with ethanol in the pathophysiology of cognitive impairment in alcoholics (Sechi and Serra 2007). Several mechanisms may be involved, such as reduction of thiamine-dependent enzyme activity, alteration of mitochondrial function and impaired oxidative metabolism, all leading to neuronal death. Decreased transketolase, alphaketoglutarate dehydrogenase and pyruvate dehydrogenase are observed in autopsy samples of cerebellar vermis of alcoholic patients with Wernicke–Korsakoff. Although a genetic predisposition has been sought for Wernicke–Korsakoff enecephalopathy, results are not conclusive (Guerrini et al. 2009). On the other hand, many alcoholics with thiamine deficiency show brain atrophy, but the degree of atrophy is similar to that of patients with normal thiamine levels. As in alcoholics without any superimposed disease, atrophy of corpus callosum is also observed in relation with Wernicke’s encephalopathy, but only in those cases related to ethanol abuse (Lee et al. 2005). Therefore, probably, there is a synergistic effect of thiamin deficiency and ethanol intake on brain alterations, at least regarding white matter shrinkage. In addition, several observations support a role of thiamine deficiency in cerebellar shrinkage, although the precise mechanism remains elusive. In this sense, the supplementation of staple foods like bread (and, eventually, beverages such as beer) with thiamine is a recommended practice (Harper 2009).
187 Brain Atrophy in Alcoholics
3003
Brain atrophy has been put in relation with hiperhomocysteinaemia, especially at the hippocampus, in a study of 52 alcoholics who showed high levels of homocysteine (reaching 27 µmol/l in female alcoholics), low levels of folate and B6, but normal B12 levels (Bleich et al. 2003). This result is consistent with observations performed on otherwise healthy elderly subjects and in cases of dementia. Possibly, oxidative damage is the underlying mechanism, although low vitamin B12 levels are associated with brain atrophy (Vogiatzoglou et al. 2008) and more rapid cognitive decline. However, the precise pathogenic mechanisms, as well as the role of vitamin B12 in brain atrophy of alcoholics are largely unknown.
187.4.4 Nutritional Status: Protein–Calorie Malnutrition and Obesity Protein–calorie malnutrition is a well-recognized alteration in alcoholics: indeed, not only muscle wasting, but frank malnutrition is a widespread finding in several studies dealing with alcohol-related pathologies. Although commonly associated to other nutritional deficiencies, several studies performed in growing children and individuals with eating disorders (Nogal et al. 2008) have described alterations in the myelinization process, ventricular dilatation and increased width of cortical sulci with cerebral atrophy in protein calorie malnutrition. In addition, prolonged malnutrition frequently precedes cerebellar alterations. In a study of 36 chronic alcoholics not affected by Wernicke– Korsakoff encephalopathy and 12 patients with Wernicke’s encephalopathy it was found that malnutrition was associated with a sixfold increase in the risk of cerebellar atrophy. Cerebellar atrophy was present in 10 out of 12 undernourished alcoholics not affected by Wernicke’s encephalopathy, but also in 11 out of 24 well-nourished alcoholic patients (Nicolas et al. 2000). These results also suggest that factors other than malnutrition are involved in cerebellar shrinkage in alcoholics. Indeed, ethanol consumption was also directly related to cerebellar shrinkage using stepwise regression analysis, in accordance with a previous study of the same group, performed on 40 well-nourished alcoholics, who showed a greater degree of brain shrinkage with age than controls, accompanied by functional derangement. Both morphological and functional alterations were correlated with the lifetime amount of ethanol consumed (Nicolas et al. 1997). However, in 42 non-selected alcoholics, several nutritional parameters, especially those related to muscular assessment, such as brachial perimeter and handgrip strength, were also related to brain atrophy (Fig. 187.7a–c). Moreover, total lean mass was lower among alcoholics with frontal atrophy (not quantified, but assessed by a neuroradiologist), as well as trunk lean mass and limb lean mass. Interestingly, no relation was found between body mass index (BMI) and brain atrophy, or between fat parameters and brain atrophy. In this regard, obesity may also lead to brain atrophy and altered cognitive function. In a cross-sectional study on 114 individuals aged 40–66 years, brain atrophy was independently related both with age (b = −0.39) and BMI (b = −0.22; Ward et al. 2005). The association between increased BMI, brain atrophy and deranged cognitive performance has been confirmed in further studies, even in healthy adults, although the possibility exists than brain atrophy in these individuals is related to other risk factors, such as hyperhomocysteinaemia, dyslipemia, hypertension or diabetes. Malnutrition in alcoholics has many causes, but, undoubtedly, poor nutritional intake is an important one (Santolaria et al. 2000). Irregular feeding is frequently observed in heavy alcoholics who have lost familial and wider social links. Classifying a series of patients in three groups according to eating habits, we found that cortical atrophy, estimated by measuring the width of cortical sulci, was more intense among those with poor eating habits (Fig. 187.8). Although it is likely that those with poor nutritional intake not only had protein calorie malnutrition, but also vitamin and micronutrient deficiency, this result stresses the importance of adequate nutrition in the alcoholic population at risk of brain atrophy (García-Valdecasas-Campelo et al. 2007).
3004
a
E. González-Reimers and F. Santolaria-Fernández Relationship between bifrontal index and handgrip strength
b
c
Relationship between brachial perimeter and cella index
Frontal atrophy/lean mass
6,00 50000,00
total lean mass (g)
r=0.34; p=0.048 0,40
Cella index
Bifrontal index
0,50 5,00
4,00
0,30 3,00
40000,00 30000,00
20000,00 10000,00
0,20 0,00
100,00
200,00
300,00
400,00
handgrip strength (kgx10)
20,00
30,00
40,00
1,00
2,00
frontal atrophy t=2.30, p=0.029
brachial perimeter (cm) r=0.32; p=0.035
Fig. 187.7 Relationships between nutritional parameters and brain atrophy. Relationships between nutritional parameters and brain atrophy (panel a = relationship between handgrip strength and bifrontal index; panel b = relationship between cella index and brachial perimeter; panel c = relationship between the presence (1) or not (2) of frontal atrophy assessed by a neuroradiologist and total lean mass (assessed by dual energy x-ray absorptiometry)
Cortical atrophy/eating habits
0,10
Fig. 187.8 Relationship between eating habits and cortical atrophy. Relationship between eating habits and cortical atrophy in 35 patients, who reported normal food consumption (left bar, n = 11), irregular meals (central bar, n = 20) or very irregular eating habits (right bar, n = 4), consuming only snacks or sandwiches, always with large amounts of ethanol
0,05
0,00 1,00
2,00
3,00
1=adequate; 2=irregular; 3=poor F=5.25, p=0.011
187.4.5 Liver Disease Impaired liver function may also be associated with brain atrophy. In a classic neuropathological study on autopsies of 25 alcoholics and 44 controls, Harper and Krill (1985) found that the pericerebral space was 8.3% ± 3.3% in controls, but 16.2% ± 4.4% in alcoholics with liver disease, even greater than in alcoholics with Wernicke’s encephalopathy. Amodio et al. (2003) found that several indices of brain atrophy were altered in 68 alcoholics, some of them of non-alcoholic etiology; indeed, atrophy was independent of alcoholic etiology, but was related only to age and to psychometric
187 Brain Atrophy in Alcoholics
Ventricular index / cirrhosis
0,60
Ventricular index
Fig. 187.9 Relationships between ventricular dilatation and liver cirrhosis. Cirrhotics (left bar, 17 cases) show less ventricular dilatation than non-cirrhotics (right bar, 30 cases). This result speak against a possible relation between liver failure and brain atrophy
3005
0,40
0,20
0,00 1,00
2,00
cirrhosis t=2.30p=0.029
impairment and EEG changes, confirming previous studies. In this regard, Bernthal et al. (1987), studying 49 non-alcoholic cirrhotics, found that some of them showed brain atrophy, but others had brain edema, and that these changes were related to altered psychometric tests; Tarter et al. (1986) found a significant inverse relationship between serum albumin and bifrontal index (r = −0.32, p < 0.05), and serum albumin and third ventricle (r = −0.30, p < 0.05) in 42 patients affected by nonalcoholic chronic liver diseases. Although high ammonia has been hypothesized to play a role in these alterations, no relation was found between brain atrophy indices and blood ammonia. It is possible that other factors, such as nutritional status, vitamin deficiency, or oxidative damage, may also account for these results. In this regard, in 17 alcoholic cirrhotics and 30 non-cirrhotic alcoholics (all heavy drinkers), we found that non-cirrhotic alcoholics showed more dilated ventricles than cirrhotics (Fig. 187.9) despite similar age. Moreover, a direct relationship was observed between cella index and bilirubin (r = 0.39, p = 0.008), and an inverse one between cella index and prothrombin activity (r = −0.32, p = 0.03), suggesting again that ventricular dilatation is more intense among those with preserved liver function, so that, in our experience, liver dysfunction is not related to brain atrophy.
187.4.6 Biochemical Changes and Altered Neurotransmission It is hypothesized that alcohol-induced brain damage is attributable to two main mechanisms. The first is direct neurotoxicity mediated by ethanol, possibly by disturbances in the excitatory neurotransmission system. In addition, as commented before, thiamine deficiency may play an additive role, and this would be the second mechanim. There are marked differences in the alterations observed in neurotransmitter systems in binge drinkers and in chronic alcoholics (Ward et al. 2009). Glutamate, an excitatory neurotransmitter involved in learning and its modulation of consolidation and recall, is found to be raised in the nucleus accumbens in a binge-drinking experimental model in rats, but not in chronic alcoholics during alcohol consumption. Moreover, the amount of released glutamate depends on rat strain. Meléndez et al. (2005) showed an increase in extracellular glutamate in the nucleus accumbens, which was related to decreased uptake 24 h after a seven-day period
3006
E. González-Reimers and F. Santolaria-Fernández
of intraperitoneal administration of 1 g /kg ethanol. Glutamate acts by binding to several receptors, including NMDA, alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and metabotropic glutamate receptor. It seems that in chronic alcohol exposure NMDA receptors are inhibited. However, as early as 3–5 h after detoxification there are changes in NMDA receptor sensitivity and an increase in glutamate release, which may be related to behavioural disturbances and alcohol craving (Ward et al. 2009). Dopamine is involved in the mesolimbic reward pathway, and ethanol ingestion increases dopaminergic transmission and the firing rate of dopaminergic neurons. Chronic alcoholics need larger amounts of ethanol to evoke dopamine release to maintain the pleasurable effects of ethanol intake. These mechanisms are dramatically altered during ethanol withdrawal, leading to dysphoria and malaise. In addition, activation of D1 dopamine receptors, through phosphorilation of intermediate elements of the intracellular signal cascade, may affect glutamate activation of NMDA receptor, overriding the alcohol-induced inhibition of NMDA. Importantly, excessive activation of NMDA receptors is a major cause of neuronal cell death (Ward et al. 2009), probably by oxidative damage. Ethanol modulates serotonin release in various areas of the central nervous system, especially through receptor 5HT-3, with different effects if exposure is acute or chronic. Serotonin release is involved in several aspects of alcohol-seeking, alcohol addiction and alcohol intoxication (Rodd et al. 2007), since it affects mood and cognitive performance. Serotonin levels depend on the levels of its precursor tryptophan, which competes with other large neutral amino acids for transport into the brain. For this reason, mental performance may be related to the ratio tryptophan/large neutral amino acids, which decreases after ethanol consumption. However, the possible effects of the altered metabolism of serotonin on brain shrinkage and neuronal death, as well as those of g-aminobutyric acid (GABA), the main inhibitory neurotransmitter, also modulated by ethanol, are not known. The opioid system is also influenced by ethanol. In a binge-drinking model in newborn rats (2.5 g/kg ethanol by nasogastric tube during days 2–6 of life) increased apoptosis of beta endorphinsecreting neurons (Sarkar et al. 2007) was observed, via activation of transforming growth factor beta-1 linked apoptotic signalling. It is possible that neurotransmitters are also involved in the recovery of brain damage after abstinence. Alcohol withdrawal may lead to regeneration of the atrophic brain, with a nearly total functional recovery after prolonged abstinence. This is associated with ultrastructural changes including growth of both white and grey matter. As early as one day after stopping a binge-drinking period, proliferating microglia is observed in cortical and non-cortical brain regions (Crews and Nixon 2009). This first burst of cell proliferation is followed by the identification of immature neurons, peaking around 14 days of abstinence, located in certain areas, such as dentate gyrus. Although precise mechanisms are unknown, it is possible that glutamate-activated trophic NMDA receptors lead to increased CREB transcription, resulting in cell growth.
187.5 Conclusions In summary, brain atrophy is a well-described entity in the alcoholic population, which may severely impair cognitive functions. Ethanol itself may cause brain atrophy through several mechanisms involving both accelerated neuronal death and blunted regeneration; probably, the principal mediator of these alterations is ethanol-induced neuroinflammation, which leads to cytokine activation and oxidative damage. Undoubtedly, coexisting alteration in vitamins, hormones micronutrients, as well as poor nutritional status, may all contribute to diffuse brain damage. It is also likely that thiamine deficiency, common in the alcoholics, also plays a synergistic role. Finally, some data, obtained by
187 Brain Atrophy in Alcoholics
3007
other authors, also suggest a role of ethanol-associated liver disease on brain damage. In any case, our knowledge about this entity is still fragmentary and incomplete, and considerable research is needed for the efficient control of this complication.
187.6 Applications to Other Areas of Health and Disease The abuse of ethanol, either in the form of excessive chronic consumption, or as binge drinking, is widespread, especially among adolescents and young adults. Even with appropriate nourishment, both forms of alcoholism may cause neurocognitive impairment and organic brain damage. Therefore, this problem is of paramount importance for the entire society, since it confers a risk of dementia to a significant proportion of young people, in addition to the many other disturbances caused by ethanol excess. Since withdrawal is associated with a marked improvement of neurocognitive damage, urgent social measures to limit excessive ethanol consumption are needed, especially directed at young people. Equally important for the entire society is the Wernicke–Korsakoff syndrome, due to thiamine deficiency. Supplementation of basic foods, such as bread, or even alcoholic beverages, such as beer, with thiamine is a safe practice, already implemented in several countries. Brain alterations in alcoholics also constitute another example of the rapidly expanding field of cytokine-related organ damage. This is linked to oxidative damage, and considerable research is still needed to precisely define the mechanisms involved, the potential therapeutic role of antioxidants, and the relation between altered cytokine release and neurotransmitter modulation and excitotoxicity. The bulk of evidence indicate that severe protein deficiency may alter maturation and central nervous system function. As commented before, some alcoholics with intense addiction usually suffer loss of social and family links, and, literally forget about eating, a fact which strongly contributes to impaired nutrition. Brain atrophy in alcoholics is related with protein undernutrition. It is important to deepen our knowledge about the mechanism connecting cognitive impairment and protein deficiency, since this may help a wide proportion of people threatened by hunger in the developing world.
Summary Points • Chronic brain damage associated to excessive ethanol consumption include cortical and subcortical atrophy, cerebellar degeneration, toxic amblyopia, pellagra, Wernicke–Korsakoff encephalopathy, Marchiafava–Bignami disease, central pontine myelinolysis and increased prevalence of stroke and cerebral trauma. • Clinical features of brain atrophy may vary from subtle alterations (impaired verbal problem solving, abstracting abilities, visual-spatial performance, verbal memory, and the ability to adapt problem-solving strategies to changing requirements) to frank dementia. They are mostly reversible with prolonged ethanol withdrawal. • Underlying anatomic lesions include neuronal death and white matter alterations (demyelination and axonal death) • Pathogenesis of brain atrophy is complex and partially understood, but several mechanisms undoubtedly play a role, including: direct effect of ethanol metabolism, cytokine (especially TNF-a)-mediated neuroinflammation, oxidative damage (disbalance between antioxidants and prooxidants), thiamine deficiency, protein deficiency, excitotoxicity, and, perhaps, hyperammoniaemia related to coexisting liver disease. • Recovery may be achieved with prolonged ethanol withdrawal and adequate nutrition
3008
E. González-Reimers and F. Santolaria-Fernández
Definitions and Explanations Wernicke–Korsakoff encephalopathy: Common complication observed in alcoholics, related to thiamine deficiency. Lethargy, disorientation and stupor may be accompanied by ophthalmoplegia and ataxia. Organic symptoms may improve with thiamine supplementation, but in some cases, confabulation, hallucinations, anterograde amnesia and intellectual impairment may ensue (Korsakoff’s dementia). Central pontine myelinolysis: Uncommon, usually fatal, complication of alcoholics, of unknown pathogenesis, characterized by demyelination and necrosis of pons and other areas of the brain leading to spastic quadriplegia, pseudobulbar palsy, and coma. Lesions and clinical picture are similar to those observed after rapid correction of profound hyponatremia. Marchiafava–Bignami disease: An uncommon disorder which affects corpus callosum, leading to demyelination and necrosis. Initially described in Italian consumers of red wine, it also affects people worldwide, including malnourished nonalcoholics. There are three clinical forms (acute, subacute and chronic). Clinical picture ranges from convulsions and coma to mild dementia and progressive neurological deterioration. Tobacco- and Alcohol-related amblyopia: Scotomas (blind spots) and decreased visual acuity within the central portion of the visual field, which affects undernourished alcoholic smokers, presumably caused by alcohol-mediated optic neuropathy, possibly associated with deficiency of thiamine and other micronutrients. It improves with proper diet, but may lead to permanent loss of vision if untreated. Proinflammatory Cytokines: Molecules with pleiotropic effects, mainly involved in the inflammatory response, produced by many cells but especially by macrophages and cells involved in the immune response. Macrophages and immune cells secrete cytokines usually in response to antigens, and they generally act as signaling molecules, inducing synthesis and/or secretion of other inflammatory mediators. Oxidative damage: Alteration of different molecules (especially DNA, lipids) of the cells by reaction with highly reactive compounds, mainly related to oxygen metabolism, including superoxide anion, hydroxyl anion, and hydrogen peroxide (powerful triggers of cytokine secretion). DNA alterations disrupt cellular metabolism and may be carcinogenic, and membrane lipid peroxidation causes cell death. Antioxidants: Substances and/or enzymatic pathways that transform highly reactive oxidant species into less dangerous ones. Clinically important antioxidant systems which become altered in the alcoholic include glutathione peroxidase, superoxide dismutase and catalase. Several vitamins (A, C, E), and micronutrients also exert antioxidant activity. Excitotoxicity: Mechanism of neuronal cell death, due to intense and prolonged activation of excitatory neurotransmitter (glutamate) receptors which generate increased influx of cations such as calcium and sodium, which in turn alter intracellular homeostasis and enzymatic pathways, ultimately leading to oxidative damage and cell death. Tolerance: It is said that a subject is tolerant to alcohol when he needs progressively larger doses to achieve the same effects Tolerance leads to physical dependence, which is a state in which the patient needs ethanol to feel well, so that negative physical symptoms of withdrawal result from abrupt discontinuation or dosage reduction. Physical dependence is different from addiction, which refers only to the compulsive psychological necessity to consume a certain drug. Dopamine neurotransmitter: Neurotransmitters are substances released into the synaptic cleft during neuronal stimulation. Dopamine acts in the mesolimbic system and plays a role in ethanol reward. During chronic consumption larger amounts of alcohol are required to evoke dopamine response (Tolerance), whereas in abstinence, rapid fall in dopamine release causes dysphoria (dependence).
187 Brain Atrophy in Alcoholics
3009
References Amodio P, Pellegrini A, Amista P, Luise S, Del Piccolo F, Mapelli D, Montagnese S, Musto C, Valenti P, Gatta A. Metab Brain Dis. 2003;28:63–78. Andersen BB. Brain Res. 2004;1007:10–18. Bartsch AJ, Homola G, Biller A, Smith SM, Weijers HG, Wiesbeck GA, Jenkinson M, De Stefano N, Solymosi L, Bendszus M. Brain. 2007;130:36–47. Bernthal P, Hays A, Tarter RE, Van Thiel DH, Lecky J, Hegedus A. Hepatology. 1987;7:107–14. Bleich S, Bandelow B, Javaheripour K, Müller A, Degner D, Wilhelm J, Havemann-Reinecke U, Sperliong W, Rüther E, Kornhuber J. Neurosci Lett. 2003;335:179–82. Cardenas VA, Studholme C, Gazdzinski S, Durazzo TC, Meyerhoff DJ. Neuroimage. 2007;34:879–87. Christie IC, Price J, Edwards L, Muldoon M, Meltzer CC, Jenningd JR. Alcohol. 2008;42:269–75. Crews FT, Nixon K. Alcohol Alcohol. 2009;44:115–27. Crews FT, Bechara R, Brown LA, Guidot DM, Mandrecker P, Oak S, Qin L, Szabo G, Wheeler M, Zou J. Alcohol Clin Exp Res. 2006;30:720–30. Ding J, Eigenbrodt ML, Mosley TH, Hutchinson RG, Folsom AR, Harris TB, Nieto FJ. Stroke. 2004;35:16–21. Fitzpatrick LE, Jackson M., Crowe SF. Neuriosci Behav Rev. 2008;32:466–85. Gazdzinski S, Durazzo TC, Meyerhoff D. Drug Alcohol Depend. 2005;78:263–73. García-Valdecasas-Campelo E, González-Reimers E, Santolaria-Fernández F, de la Vega-Prieto MJ, Milena-Abril A, Sánchez-Pérez MJ, Martínez-Riera A, Rodríguez-Rodríguez E. Alcohol Alcohol. 2007;42:533–8. Guerrini I, Thomson AD, Gurling HM. Alcohol Alcohol. 2009;44:166–70. Harding A, Halliday G, Ng JLF, Harper CG, Kril JJ. Neuroscience. 1996;72:699–708. Harding A, Halliday G, Caine D, Kril J. Brain. 2000;123:141–54. Harper C. Alcohol Alcohol. 2009;44:136–40. Harper C, Kril J. J Neurol Neurosurg Psych. 1985;48:211–7. Hua Y, Keep RF, Hoff JT, Xi G. Stroke. 2007;38:759–76. Jarvenpaa T, Rinne JO, Koskenvuo M, Räihä I, Kaprio J. Epidemiology. 2005;16:766–71. Kril JJ, Halliday GM, Svoboda MD, Cartwright H. Neuroscience. 1997;79:983–98. Kubota M, Nakazaki S, Hirai S, Saeki N, Yamaura A, Kusaka T. J Neurol Neurosurg Psych. 2001;71:104–6. Lee ST, Jung YM, Na DL, Park SH, Kim M. J Neuroimaging. 2005;15:367–72. Mantamadiotis T, Lemberger T, Bleckmann SC, Kern H, Kretz O, Martín Villalba A., Tronche F, Kellendonk C, Gau D, Kapfhammer J, Otto C, Schmid W, Schütz G. Nature Gen. 2002;31:47–54. Maschke M, Weber J, Bonnet U, Dimitrova A, Böhrenkamper J, Sturm S, Müller BW, Gastpar M, Diener HC, Forsting M, Timmann D. J Neurol. 2005;252:704–11. Meléndez RI, Hicks MP, Cagle SS, Kalivas PW. Alcohol Clin Exp Res. 2005;29:326–33. Menzano E, Carlen LP. Alcohol Clin Exp Res. 1994;18:895–901. Nicolas JM, Estruch R, Salamero M, Orteu N, Fernández-Sola J, Sacanella E, Urbano-Marquez A. Ann Neurol. 1997;41:590–8. Nicolas JM, Fernández-Sola J, Robert J, Antúnez E, Cofán M, Cardenal C, Sacanella E, Estruch R, Urbano-Marquez A. QJM. 2000;93:449–56. Nixon SJ, Tivis R, Parsons OA. Alcohol Clin Exp Res. 1995;19:577–81. Nogal P, Pniewska-Siark B, Lewinski A. Neuro Endocrinol Lett. 2008;29:879–83. Nordmann R. Alcohol Alcohol. 1994;29:513–22. Paul CA, Au R, Fredman L, Massaro JM, Seshadri S, Decarli C, Wolf PA. Arch Neurol. 2008;65:1363–7. Peters S, Koh J, Choi DW. Science. 1987;236:589–93. Pfefferbaum A, Adalsteinsson E, Sullivan EV. Aging. 2006;27:994–1009. Rodd ZA, Gryszowka VE, Toalston JE, Oster SM, Ji D, Bell RL, McBride WJ. J Pharmacol Exp Ther. 2007;321: 1003–12. Santolaria F, González-Reimers E, Pérez-Manzano JL, Milena A, Gómez-Rodríguez MA, González-Díaz A, de la Vega MJ, Martínez-Riera A. Alcohol. 2000;22:47–157. Sarkar DK, Kuhn P, Marano J, Chen C, Boyadijeva N. Endocrinology. 2007;148:2828–34. Schweizer U, Bräuer AU, Köhrle J, Nitsch R, Savaskan NE. Brain Res Rev. 2004;45:164–78. Sechi G, Serra A. Lancet Neurol. 2007;6:442–55. Squitti R, Lupoi D, Pasqualetti P, Dal Forno G, Vernieri F, Chiovenda P, Rossi L, Cortesi M, Cassetta E, Rossini PM. Neurology. 2002;59:1153–61. Strecker K, Schneider JP, Barthel H, Herrmann W, Wegner F, Wagner A, Schwarz J, Sabri O, Zimmer C. J Neurol. 2006;253:1024–9. Tarter RE, Hays A, Sandford SS, Van Thiel DH. Lancet 1986;II:893–5.
3010
E. González-Reimers and F. Santolaria-Fernández
Thatcher DL, Clark DB. In: Preedy VR, Watson RR, editors. Comprehensive handbook of alcohol related pathology. Amsterdam: Elsevier; 2005. pp. 181–93. Uysal M, Kutalp G, Odzemirler G, Aykac G. Drug Alcohol Depend. 1989;23:227–30. Vogiatzoglou A, Refsum H, Johnston C, Smith SM, Bradley KM, de Jager C, Budge MM, Smith AD. Neurology. 2008;71:826–32. Wang LL, Li Y, Zhou SF. Toxicol Lett. 2009;187:99–114. Ward MA, Carlsson CM, Trivedi MA, Sager MA, Johnson SC. BMC Neurology. 2005;5:23 (doi.10.186/1471-2377-5-23). Ward RJ, Lamelland F, de Witte P. Alcohol Alcohol. 2009;44:128–35. Zou JY, Crews FT. Brain Res. 2009;1034:11–24.
Chapter 188
Alcohol Consumption in Predementia and Dementia Syndromes Francesco Panza, Vincenza Frisardi, Patrick G. Kehoe, Cristiano Capurso, Alessia D’Introno, Anna M. Colacicco, Gianluigi Vendemiale, Antonio Capurso, and Vincenzo Solfrizzi
Abbreviations AD APOE ARCD bA CAD CASI CVD MCI MUFA PUFA SFA VaD WML
Alzheimer’s disease Apolipoprotein E Age-related cognitive decline Amyloid beta Coronary artery disease Cognitive abilities screening instrument Cerebrovascular disease Mild cognitive impairment Monounsaturated fatty acids Polyunsaturated fatty acids Saturated fatty acids Vascular dementia White matter lesions
188.1 Introduction 188.1.1 Current Epidemiology of Dementia and Predementia Syndromes Since population aging has become a worldwide phenomenon, the burden of age-related neurodegenerative diseases, particularly dementia, is expected to increase dramatically in both developed and developing nations. The deterministic boundaries of perceived normal cognitive aging are not clearly defined while the clinical categorisation of predementia and dementia syndromes remain, at present, a work in progress. Dementia is a syndrome defined by impairments in memory and other cognitive functions that are severe enough to cause significantly reduced performance from a previous level of social and occupational functioning. Dementia is estimated to affect approximately 6% of the people aged 65 years and older, with the prevalence increasing exponentially with age, rising F. Panza (*) Department of Geriatrics, Center for Aging Brain, Memory Unit, University of Bari, Policlinico, Piazza Giulio Cesare, 11, 70124 Bari, Italy e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_188, © Springer Science+Business Media, LLC 2011
3011
3012
F. Panza et al.
Table 188.1 Key features of dementia syndromes 1. Dementia is a syndrome definite by impairments in memory and other cognitive functions that are severe enough to cause significant decline from a previous level of social and occupational functioning 2. AD is the most common dementia and primary neurodegenerative disorder in the elderly, gradually leading to a complete psychological and physical dependency and finally to death within 1–2 decades 3. The diagnosis of AD is essentially a clinical one, and it is based on a typical clinical picture and findings, with a set of clinical criteria often used in research 4. Cognitive function declines over time, and the diagnosis of AD can be considered when the patient has impairments in memory and at least in one other cognitive function (executive dysfunction, agnosia, aphasia, apraxia), severe enough to cause impairment in social or occupational functioning 5. In advanced AD, common symptoms include also confusion, behavioural and gait disturbances, and the patients are increasingly dependent on others in activities of daily living 6. Another common form of dementia is VaD, its clinical presentation varies greatly depending on the causes and location of cerebral damage 7. Large-vessel disease leads commonly to multiple cortical infarcts and a multifocal cortical dementia syndrome 8. Small-vessel disease, usually resulting from hypertension and diabetes, causes periventricular white matter ischemia and lacunar strokes characterized clinically by subcortical dementia with frontal lobe deficits, executive dysfunction, slow information processing, impaired memory, inattention, depressive mood changes, slowing of motor function, Parkinsonian features, small-step gait, urinary disturbances and pseudobulbar palsy 9. The characteristic neuropsychological profile of VaD is believed to include frequently early impairment of attention and executive control function, with slowing of motor performance and information processing, while episodic memory is relatively spared compared to that in AD This table lists the key facts of dementia syndromes including clinical pictures and criteria of Alzheimer’s disease (AD) and vascular dementia (VaD), the most common dementia syndromes
to between 40% and 70% at the age of 95 years and above (Qiu et al. 2007). In Westernised countries, the most common forms of dementia are Alzheimer’s disease (AD) and vascular dementia (VaD), with respective frequencies of 70% and 15% of all dementias (Whitehouse et al. 1997) (Table 188.1). Therefore, AD is the most common dementia and primary neurodegenerative disorder in the elderly that gradually leads to a complete psychological and physical dependency on others and finally to death within 1–2 decades. In the present chapter, we will use the term “predementia syndrome” to identify all conditions with age-related (non-pathological) deficits in cognitive function reported in the literature, including a mild stage of cognitive impairment based on a model of normality and the presence of pathological conditions considered predictive or early stages of dementia (Panza et al. 2005) (Table 188.2). Such predementia syndromes have been defined for AD and VaD, but have not yet been operationalized for other specific forms of dementia. Therefore, the term “predementia syndromes” includes different conditions and among these predementia syndromes, mild cognitive impairment (MCI) is, at present, the most widely used term to indicate nondemented aged persons with no significant disability and a mild memory or cognitive impairment that cannot be explained by any recognisable medical or psychiatric condition (Petersen et al. 1999; Winblad et al. 2004) (Table 188.2). There is now ample evidence that MCI is often a pathology-based condition with a high rate of progression to AD (Petersen et al. 1999). Therefore, MCI has been classed as the predementia syndrome for AD (Panza et al. 2005). Recently, a number of subtypes of MCI have been proposed, intended to reflect the heterogeneity of different types of dementia (Table 188.2). In fact, the recent subclassification of MCI according to its cognitive features (including dysexecutive MCI and amnestic-MCI (aMCI) or aMCI and non-amnestic MCI (naMCI); single or multiple domain aMCI or naMCI) and clinical presentation (MCI with parkinsonism or cerebrovascular disease), or likely aetiology (MCI-AD, vascular MCI, or MCI-Lewy Body Disease) all represent an attempt to exert some control over this heterogeneity (Panza et al. 2010). Recently, a critical review was undertaken
188 Alcohol Consumption in Predementia and Dementia Syndromes
3013
Table 188.2 Key features of predementia syndromes 1. The term “predementia syndromes” identifies all conditions with age-related deficits in cognitive function reported in the literature, including a mild stage of cognitive impairment based on a normality model and pathological conditions considered predictive or early stages of dementia 2. Among predementia syndromes, MCI is, at present, the most widely used term to indicate non-demented aged persons with a mild memory or cognitive impairment that cannot be accounted for any recognized medical or psychiatric condition 3. The general criteria for MCI include (a) memory complaint, (b) objective memory disorder, (c) absence of other cognitive disorders or repercussions on daily life, (d) normal general cognitive function, (e) absence of dementia 4. MCI definitions can be broadly classified as amnestic (aMCI) and nonamnestic (naMCI) 5. There is now ample evidence that MCI is often a pathology-based condition with a high rate of progression to AD, and aMCI, with a central role for memory disorder and with relative preservation of other cognitive domains, was identified as the predementia syndrome for AD 6. aMCI can be subdivided into a single-domain subtype with a pronounced memory deficit or a multiple-domain subtype that includes memory impairment along with some impairment in other cognitive domains such as language, executive function, and visuospatial skills 7. The other major MCI subtype is naMCI, which similarly can be subdivided into single and multiple domain subtypes This table lists the key facts of predementia syndromes including diagnostic criteria and clinical classification of mild cognitive impairment (MCI), the most common predementia syndrome
to try to define a new consensus on MCI, and a modification of Petersen’s original criteria (Petersen et al. 1999) was proposed during the conference in Montreal (Winblad et al. 2004). Subsequently, the European Consortium on Alzheimer’s Disease (EADC) (http://eadc.alzheimer-europe.org/introduction.html) working group on MCI proposed a novel diagnostic procedure with different stages, combining neuropsychological evaluation and family interview to detect MCI at the earliest possible stage (Portet et al. 2006). The clinical presentation of VaD varies greatly depending on the causes and location of cerebral damage (Roman 2002). Large-vessel disease leads commonly to multiple cortical infarcts and a multifocal cortical dementia syndrome, whereas small-vessel disease, usually resulting from hypertension and diabetes, causes periventricular white matter ischemia and lacunar strokes characterized clinically by subcortical dementia with frontal lobe deficits, executive dysfunction, slow information processing, impaired memory, inattention, depressive mood changes, slowing of motor function, Parkinsonian features, small-step gait, urinary disturbances, and pseudobulbar palsy (Roman and Royall 1999). Recently, the term vascular cognitive disorder (VCD) has been proposed by Sachdev (Sachdev 1999) and it would become the global diagnostic category for cognitive impairment of vascular origin (Roman et al. 2004). VCD would include the group of syndromes and diseases characterized by cognitive impairment resulting from a cerebrovascular etiology. The main categories of VCD are vascular cognitive impairment (VCI) [i.e., vascular cognitive impairment no dementia (vascular CIND), and vascular MCI], VaD, and mixed AD plus cerebrovascular disease (CVD) previously termed “mixed dementia” (Sachdev 1999; Roman et al. 2004). Dementia is defined as executive control deficit producing loss of function for instrumental activities of daily living, while mixed AD plus CVD is defined as preexisting AD worsened by stroke (equivalent to prestroke dementia). Finally, VCI is a term referred to all forms of mild to severe cognitive impairment associated with CVD, including vascular CIND and vascular MCI, e.g., predementia syndromes with a presumed primary vascular basis. VCI is considered a premonitory phase of VaD, although VCI not always proceeds to VaD (Sachdev 1999; Roman et al. 2004). The characteristic neuropsychological profile of VCI is believed to include frequently early impairment of attention and executive control function, with slowing of motor performance and information processing, while episodic memory is relatively spared compared to that in AD (Erkinjuntti et al. 2000) (Table 188.1).
3014
F. Panza et al.
188.1.2 L ifestyle Changes and Prevention of Dementia and Predementia Syndromes At present, there is no curative treatment for dementia and AD, nor is there a therapeutical approach to prevent the conversion of MCI to dementia. In previous years, extensive research has increased our knowledge of the aetiology of AD, other dementing disorders, and predementia syndromes, and several hypotheses have emerged from both laboratory and epidemiological research. Epidemiological evidence in particular supported the hypothesis that modifiable vascular and lifestyle-related factors are associated with the development of dementia and predementia syndromes in later life, opening new potential avenues for the prevention of these diseases (Panza et al. 2009a). Given the lack of effective pharmacotherapies to treat dementia, lifestyle changes may offer alternative or supplementary treatment options for predementia syndromes, i.e., MCI or age-related cognitive decline (ARCD). Evidence from population-based longitudinal epidemiologic studies suggests that moderate exercise and physical activity are associated with a lower risk of dementia (Panza et al. 2009a). Among lifestyle- and vascular-related factors, the impact of diet, Mediterranean diet in particular, has also been the subject of recent interest (Panza et al. 2009a) while regular moderate consumption of alcohol (£24 g/day) have been reported to prevent coronary artery disease (CAD) and to reduce the risk of ischaemic and haemorrhagic stroke (Panza et al. 2009a). There is also much evidence existing and emerging, from population-based longitudinal cohort studies predominantly but also from some case-control studies, which suggest that alcohol consumption, particularly red wine within limits and/or of certain types, decreases risk of cognitive impairment or decline, predementia, and dementia syndromes. This is despite chronic alcohol abuse causing progressive neurodegenerative disease (Panza et al. 2009a) (Table 188.3). However, some of the variability in these studies may be due to cross-sectional designs used, restrictions by age or sex, or incomplete ascertainment (Panza et al. 2009a). It is especially important to examine data for men and women separately when alcohol consumption is a predictor variable, because gender-based consumption levels
Table 188.3 Key features of alcohol 1. An alcohol is any organic compound in which a hydroxyl group (–OH) is bound to a carbon atom of an alkyl or substituted alkyl group 2. Generally, the word alcohol refers to ethanol, the type of alcohol found in alcoholic beverages 3. Ethanol is a colourless, volatile liquid with a mild odour, which can be obtained by the fermentation of sugars 4. Ethanol in alcoholic beverages has been consumed by humans since prehistoric times for a variety of hygienic, dietary, medicinal, religious, and recreational reasons 5. The short-term effects of alcohol on the human body can take several forms. Alcohol, specifically ethanol, is a potent CNS depressant, with a range of side effects. The consumption of large doses of ethanol causes drunkenness (intoxication), which may lead to a hangover as its effects wear off. The amount and circumstances of consumption play a large part in determining the extent of intoxication 6. The long-term effects of alcohol in excessive quantities are capable of damaging nearly every organ and system in the body. Regularly consumption of alcohol is correlated with an increased risk of developing alcoholism, cardiovascular disease, malabsorption, chronic pancreatitis, alcoholic liver disease, and cancer. Damage to the CNS and peripheral nervous system can occur from sustained alcohol consumption 7. Research has found a correlation between light to moderate consumption of alcohol, one to two alcoholic beverages per day, and reduced risk of heart disease as well as other health benefits, including reduction in all-cause mortality and decreasing risk of dementia, including AD. However, at present, due to poor study design and methodology, the literature is inconclusive on whether moderate alcohol consumptions increases the risk of dementia or decreases it This table lists the key features of alcohol including definition, short- and long-term health effects, and current research findings on the health effects of light to moderate consumption of alcohol CNS central nervous system, AD Alzheimer’s disease
188 Alcohol Consumption in Predementia and Dementia Syndromes
3015
are very different. In virtually every study which included both sexes, women consumed alcohol less frequently and in smaller amounts than men. Moreover, education, smoking, or the widely recognized AD genetic risk factor apolipoprotein E (APOE) e4 allele often modified the association between alcohol drinking and cognitive impairment or decline. Indeed reported association between less years of education and predementia and dementia syndromes is supported by the majority of studies, although few studies have investigated lifestyle factors as a possible covariant with education (Panza et al. 2009a). Socioeconomic and educational factors, which contribute to drinking behaviour in different populations and countries, might influence the strength of association of alcohol and cognitive impairment or decline. Furthermore, the APOE e4 allele, which has variable prevalence in different geographic locations (Panza et al. 1999) could be a possible effect modifier for the associations between alcohol/vascular risk factors and dementia syndromes (Panza et al. 2009a). An alternative growing body of evidence suggests that chronic cigarette smoking is associated with abnormalities in brain morphology, cerebral blood flow, increased oxidative stress, and with increased risk of stroke, and these factors could confound and counteract possible cognitive benefits afforded by low to moderate alcohol consumption among smokers (Panza et al. 2009a). However, smoking and alcohol use are related; there are more heavy drinkers who smoke, which may produce a possible multiplicative risk effect for cognitive impairment or decline. Of note, drinking inversely correlates with increased age while risk for cognitive impairment occurs most often at extremely old age long after the peak drinking years of youth (Panza et al. 2009a). It is possible that over time, possibly influenced by cognitive decline, alcohol consumption will change, and therefore long-term follow-up studies are needed to clarify the outcomes of these changes. Furthermore, aspects of when and how drinking is measured and how these measurements relate to when and how cognitive decline is measured are important sources of variability that need to be borne in mind in these studies. In this chapter, we have summarized the findings of the studies of alcohol consumption in cognitive impairment or decline, predementia, and dementia syndromes. We have reviewed clinical and epidemiological studies from the international literature, including both cross-sectional and longitudinal studies that involved subjects aged 60 years and above and where description of the diagnostic criteria for predementia or dementia syndromes has been attempted. Special attention was paid to the possible mechanisms behind reported associations of alcohol drinking with cognitive impairment or decline, predementia, and dementia syndromes.
188.2 Alcohol Consumption and Cognitive Functions in Older Age 188.2.1 Regular Alcohol Consumption and Cognitive Functions in Older Age Several studies have assessed regular alcohol consumption and cognitive function among older adults, but with inconsistent results (Goodwin et al. 1987; Herbert et al. 1993; Christian et al 1995; Hendrie et al. 1996; Launer et al. 1996; Dent et al. 1997; Dufouil et al. 1997; Broe et al. 1998; Edelstein et al. 1998; Elias et al. 1999; Carmelli et al. 1999; Elwood et al. 1999; Leibovici et al. 1999; Dufouil et al. 2000; Cervilla et al. 2000a,b; Galanis et al. 2000; Bond et al. 2001, 2005; Zuccalà et al. 2001; Schinka et al. 2002; Verhaegen et al. 2003; Zhou et al. 2003; Stampfer et al. 2005; Lindeman et al. 2005; Ganguli et al. 2005; Reid et al. 2006; Ngandu et al. 2007; McGuire et al. 2007) (Tables 188.4 and 188.5). Early studies involving relatively small samples of young to middle-aged male social drinkers supported the notion that drinking at any level was associated with poorer performance on cognitive tests (Panza et al. 2009a). However, other studies corroborated these results
Cross-sectional, population-based
Cross-sectional, population-based
Cross-sectional, population-based
Cross-sectional, population-based
Goodwin et al. (1987)
Hendrie et al. (1996)
Dufouil et al. (1997)
Carmelli et al. (1997)
589 male participants aged 59–69 years
2,040 participants aged 65 and older from a community-dwelling sample of black Americans 574 men and 815 women, aged 59–71 years
270 men and women aged 65–89 years
MMSE, TMT-B, WAIS -Revised, BVRT, Benton Facial Recognition Test, Paced Auditory Serial Addition Test, Auditory Verbal Learning Test, RPM, Word Fluency Test, and Finger Tapping Test; evaluation of alcohol intake and smoking habits MMSE, DSST, and BVRT. Evaluation of alcohol intake, smoking habits, and APOE genotyping
Cognitive abilities assessed with a 30-item mental status questionnaire, abstract thinking measured with the HCT, and WMS. Emotional status measured by a 92-item self-rating checklist, and social interaction evaluated with a revised form of the Interview Schedule for Social Interaction. Alcohol consumption assessed during a 3-day period and questionnaires about present and past alcohol intake Community Screening Interview for Dementia, delayed recall of the EBMT, and ADL; evaluation of alcohol consumption
Also after potential confounders were included, there was a small but significant dose effect of drinking for the drinkers, with subjects in the heaviest drinking category scoring poorest in cognitive tests and daily functioning. The scores of abstainers were worse than those of subjects in the lightest drinking category In men, no significant relation between alcohol consumption and cognitive scores was found. In contrast, among women, with a range of daily alcohol consumption between zero and approximately four drinks, an overall positive linear association between cognitive scores and alcohol consumption was found, also after adjustment for age, income, education, and depressive symptomatology After adjustment for age, education and CVD, smoking was significantly associated with poor cognitive function in current smokers compared with never smokers, whereas light drinking (one or fewer drinks per day) showed a protective effect compared with abstainers. Stratification by APOE e4 indicated that the protective effect of light drinking was stronger and the harmful effect of smoking was weaker among APOE e4 carriers than among noncarriers
Present or past alcohol intake was not associated with decreased cognitive, psychological, or social status
Table 188.4 Principal cross-sectional studies on the relationship between alcohol consumption and cognitive impairment in older subjects Setting and study References design Subjects Methods Results and conclusions Cross-sectional studies
3016 F. Panza et al.
15,807 hospitalized patients, mean age 70.9 years
395 participants aged 60–84 years
Cross-sectional, multicenter pharmacoepidemiology survey
Cross-sectional, population-based
Cross-sectional, population-based
Cross-sectional, population-based
Zuccalà et al. (2001)
Schinka et al. (2002)
Zhou et al. (2003)
Lindeman et al. (2005)
883 participants aged 65 years old and older
3,012 participants aged 60 years old and older
Hodkinson Abbreviated Mental Test score, and evaluation of alcohol intake
1,836 participants aged 65 and older
Cross-sectional and longitudinal, population-based (8 years)
Bond et al. (2001)
MMSE, WAIS -Revised, Digits Forward, Fuld Object-Memory Evaluation, CDT, and two Color TMT; evaluation of alcohol intake
MMSE and ADL, evaluation of alcohol intake, and smoking habits
Neuropsychological battery that provided measures of general cognitive ability, executive function, memory, evaluation of alcohol intake and cigarette smoking habits
CAMCOG and MMSE, AH4 evaluating verbal and mathematical reasoning, and CRT. Evaluation of alcohol intake and smoking habits. CASI, CRT, and NART. Evaluation of alcohol intake and cigarette smoking habits
1,870 men aged 55–69 years
Cross-sectional, population-based
Elwood et al. (1999)
Methods
Subjects
Setting and study References design Cross-sectional studies
(continued)
Light and moderate drinking showed no association with cognitive functions. Cigarette smoking showed no association, but there is evidence that the more able smokers quitted and become ex-smokers Lower cognitive test scores were observed for men who were either abstainers or in the heavy drinking group. For women, a linear relationship between alcohol consumption and cognitive performance was seen on two of the four measures of cognitive functioning, suggesting a possible positive relationship between light to moderate drinking and cognitive performance Adjusting for potential confounders, alcohol consumption was associated with decreased probability of cognitive impairment, a daily alcohol consumption of less than 40 g for women and 80 g or less for men might be associated with a decreased probability of cognitive impairment No evidence for a beneficial J-curve or threshold effect for drinking was found, but did not reveal any detrimental effect. No detrimental effect of smoking was found in any analysis; nor was there any evidence of an interaction between alcohol and cigarette use on any cognitive measure Alcohol drinking was associated with cognitive impairment, and in all people who drink every day, there was a significantly increased risk of cognitive impairment. Smoking was also related to cognitive impairment, and current smoking was associated with a significantly increased risk of cognitive impairment Older participants who ingested alcohol had significantly better test scores than did the abstainers on seven of nine cognitive function tests after adjusting for differences in sex, ethnicity, age, and years of education
Results and conclusions 188 Alcohol Consumption in Predementia and Dementia Syndromes 3017
Subjects
Methods
Results and conclusions
Reid et al. (2006)
Cross-sectional, population-based
760 men aged 65–89 years
TMT-B, DSST, FAS Test, and Hopkins Verbal Learning test, and evaluation of alcohol use
Current light to moderate drinking (i.e., 7 or fewer drinks per week), as compared to never and former drinkers, and the number of years drinking at this level are both associated with better cognitive performance in older males This table lists the principal findings of cross-sectional clinical and epidemiological studies on the relationship between alcohol consumption and cognitive impairment in older subjects, including setting, study design, and cognitive assessment used HCT Halstead Category Test, WMS Wechsler Memory Scale, EBMT East Boston Memory Test, ADL Activities of Daily Living, MMSE Mini-mental State Examination, TMT Trail Making Test, WAIS Wechsler Adult Intelligence Scale, BVRT Benton Visual Retention Test, RPM Raven Progressive Matrices, DSST Digit Symbol Substitution Test, APOE Apolipoprotein E, CAMCOG Cambridge Cognitive Examination, CRT Choice Reaction Time, CASI Cognitive Abilities Screening Instrument, NART National Adult Reading Test, CVD Cerebrovascular Disease, CDT Clock Drawing Test
Table 188.4 (continued) Setting and study References design Cross-sectional studies
3018 F. Panza et al.
Longitudinal, populationbased (3 years)
Retrospective cohort and co-twin-control study
Cross-sectional and longitudinal, populationbased (3 years)
Longitudinal, random sample of veterans of the second world war (9 years)
Longitudinal, populationbased (3 years)
Herbert et al. (1993)
Christian et al. (1995)
Launer et al. (1996)
Dent et al. (1997)
Broe et al. (1998)
MMSE, evaluation of alcohol intake and smoking habits
WAIS, WMS, BVRT, RPM, Rey Auditory Verbal Learning Test, Controlled Oral Word Association Test, Boston Naming Test, NART, CRT, and HCT evaluation of alcohol intake and noncontrast computed tomography MMSE, Reid Memory Test, tests of verbal fluency, subsets of the Boston Naming Test and similarities; CDT, and copied drawings of a cube, coils, and interlocking infinity loops; diagnosis of dementia and AD; evaluation of physical exercise, alcohol and smoking use
489 men aged 69–89 years
209 men, mean age: 64.3 years
327 subjects, aged 75 years and older
4,739 twins born between 1917 and 1927
Structured performance tests of immediate memory, digit span (from WAIS), and orientation; diagnosis of AD; evaluation of alcohol intake and smoking habits Two self-reported drinking histories (1970s and 1980s) and a telephone mental status interview (1990 and 1991)
1,201 subjects aged 65 years and older
(continued)
There were few significant associations between health habits and cognitive performance and these were not found consistently across cognitive measures
No evidence was found to indicate an association between moderate long-term alcohol intake and lower cognitive scores in aging individuals. There was a suggestion of a small protective effect of past moderate alcohol intake on cognitive function with aging After adjustment for age, education, and smoking status, men with CVD/diabetes and low-to-moderate alcohol intake had a significantly lower risk for poor cognitive function than abstainers. Alcohol intake was not associated with cognitive decline Persistent lifelong consumption of alcohol and the level of intake seemed not to have any impact on cognitive performance among men in old age
Change in only one of the 3 cognitive tests (digit span) was significantly associated with one alcohol category (15 ml per day), and no dose-response relation was observed
Table 188.5 Principal longitudinal studies on the relationship between alcohol consumption and cognitive decline in older subjects Setting and study References design Subjects Methods Results and conclusions Longitudinal studies
188 Alcohol Consumption in Predementia and Dementia Syndromes 3019
Longitudinal, populationbased (3 years)
Longitudinal, populationbased (4 years)
Longitudinal, populationbased (1 year)
Leibovici et al. (1999)
Dufouil et al. (2000)
Cervilla et al. (2000a)
Elias et al. (1999)
Longitudinal, populationbased (13–18 years) Cross-sectional and longitudinal, populationbased (24 years)
Edelstein et al. (1998)
Table 188.5 (continued) Setting and study References design Longitudinal studies Methods
889 subjects, aged 65 or older
Cognitive impairment assessed at baseline and 1 year later using the organic brain syndrome (OBS) cognitive impairment scale from the short CARE structured assessment., evaluation of alcohol intake and smoking habits
511 men and women MMSE, TMT-B, Category Fluency, aged 40–80 years BFSRT, and BVRT, evaluation of alcohol intake and smoking habits 1,786 subjects, aged Eight cognitive tests of verbal 55–88 years memory, learning, visual organization and memory, attention, abstract reasoning, and concept formation; evaluation of weekly alcohol intake 833 subjects over A computerized neuropsychometric 60 years examination assessed attention, primary and secondary memory, implicit memory, visuospatial ability, and language. Diagnosis of AD and evaluation of alcohol and tobacco consumption were also performed 1,389 subjects, aged MMSE, evaluation of alcohol 59–71 years consumption, APOE geno typing, and smoking habits
Subjects
Moderate alcohol consumption and cigarette smoking patterns, reported 13–18 years and 3–7 years previously were weakly and inconsistently associated with subsequent cognitive function Women who drank moderately (2–4 drinks/day) showed superior performance in many cognitive domains relative to abstainers. For men, superior performance was found within the range of 4–8 drinks/day, although fewer significant relations were observed. These results were confirmed by prospective analyses of 24-year drinking history Wine consumption was associated with an increased risk of decline over time in attention and in secondary memory. Smoking was associated with a decreased risk for decline over time in attentional and visuospatial functioning. No clear combined effect of smoking and drinking was found, even though smoking was found to increase the risk of decline in language performance when adjusted on wine consumption Alcohol consumption was associated with a decreased risk of cognitive decline in individuals without the APOE e4 allele, whereas moderate drinking increased the risk of decline in APOE e4 allele carriers. Also, lifetime smoking was a risk factor for cognitive decline in individuals without the APOE e4 allele. The data also suggested a slight protective effect of smoking in APOE e4 allele carriers No association between alcohol consumption and onset of cognitive impairment was found. Persistent cigarette smoking into late-life increased the risk for cognitive impairment
Results and conclusions
3020 F. Panza et al.
Cross-sectional and longitudinal, populationbased (4 years)
Cross-sectional and longitudinal, populationbased (2 years)
Cross-sectional and longitudinal, populationbased (8 years)
Longitudinal, populationbased (7 years)
Verhaegen et al. (2003)
Stampfer et al. (2005)
Bond et al. (2005)
Ganguli et al. (2005)
Galanis et al. (2000)
Longitudinal, populationbased (9–12 years) Longitudinal, populationbased (18 years)
Setting and study design
Cervilla et al. (2000b)
References Longitudinal studies
Cognitive by unit-weighted consisted of four intellectual abilities (perceptual speed, episodic memory, fluency, and knowledge), each assessed by composites of two tests, evaluation of alcohol consumption and smoking behaviour Telephone Interview for Cognitive Status, EBMT, TICS 10-word list, a test of verbal fluency, and the digit span backward test. Evaluation of alcohol consumption and APOE genotype
516 subjects aged 70 years and older
CASI, evaluation of alcohol intake and cigarette smoking habits
Neuropsychological test panel of the CERAD, and among these tests: MMSE and TMT. Evaluation of alcohol intake and smoking habits
1,836 participants aged 65 and older
1,681 individuals aged 65 years or older
(continued)
After multivariate adjustment, moderate drinkers had better mean cognitive scores than nondrinkers. For cognitive decline, on test of general cognition, the relative risk of a substantial decline in performance over a 2-year period was 0.85 among moderate drinkers, as compared with nondrinkers. There were no significant differences in risks according to the beverage and no interaction with the APOE genotype Alcohol consumers had higher scores (less cognitive decline) on cognition, measured by the CASI over an 8-year follow-up period, than abstainers. There were no significant gender differences in the absolute scores on CASI, and the rate of change over time did not vary This cohort showed a consistent pattern of better baseline scores and lesser decline over time in individuals who consumed alcohol minimally or moderately, compared to those who reported no drinking at baseline
Positive association between a history of moderate alcohol consumption and cognitive performance in the elderly, as men who had consumed up to one drink a day during middle age were later found to have significantly better cognitive test results than nondrinkers Cross-sectionally, better cognitive performances have been observed with higher levels of alcohol drinking, while alcohol was not associated with 4-year declines in cognition
CASI, and evaluation of alcohol intake
3,556 men of Japanese ancestry, aged 71–93 years
12,480 subjects aged 70–81 years
Older subjects who were abstinent before the age of 60 had poorer cognitive outcome than did those who drank mildly or moderately
MMSE, evaluation of alcohol intake and smoking habits
1,083 subjects, aged 65–74 years
Results and conclusions
Methods
Subjects 188 Alcohol Consumption in Predementia and Dementia Syndromes 3021
Longitudinal, populationbased (21 years)
Methods
1,341 participants MMSE, and neuropsychological aged 65–79 years tests evaluating episodic memory, semantic memory, psychomotor speed, executive function, prospective memory, and subjective memory. Evaluation of alcohol intake, smoking habits, and APOE genotyping 2,716 subjects, aged Adapted Telephone Interview for Cognitive Status, and evaluation 70 years and of alcohol intake older
Subjects
The nondrinkers both at midlife and later had a poorer cognitive performance than drinkers, especially in the domains related to fluid intelligence, i.e., executive function, psychomotor speed, as well as episodic memory, whereas the other cognitive functions showed little association with alcohol drinking. No interactions between APOE e4 and alcohol or sex and alcohol were found.
Results and conclusions
McGuire et al. (2007)
Cross-sectional, populationbased
For older adults with a level of cognitive functioning within normal ranges, moderate amounts of alcohol, an average of one drink or less daily, was protective for women, but not men This table lists the principal findings of longitudinal clinical and epidemiological studies on the relationship between alcohol consumption and cognitive impairment in older subjects, including setting. study design, and cognitive assessment used. WAIS Wechsler Adult Intelligence Scale, AD Alzheimer’s disease, MMSE Mini-Mental State Examination, WMS Wechsler Memory Scale, BVRT Benton Visual Retention Test, RPM Raven Progressive Matrices, CRT Choice Reaction Time, HCT Halstead Category Test, CDT Clock Drawing Test, BFSRT Buschke-Fuld Selective Reminding Test, APOE Apolipoprotein E, CASI Cognitive Abilities Screening Instrument, CERAD Consortium to Establish a Registry for Alzheimer Disease, TMT Trail Making Test, EBMT East Boston Memory Test
Ngandu et al. (2007)
Table 188.5 (continued) Setting and study References design Longitudinal studies
3022 F. Panza et al.
188 Alcohol Consumption in Predementia and Dementia Syndromes
3023
for a subsample of women whose drinking patterns were similar to their male counterparts (Panza et al. 2009a). Subsequent research with male and female college students and elderly men led to the conclusion that no significant negative relation exists between social drinking and level of cognitive functioning, although elevated alcohol consumption among young female social drinkers were related to better performance on many cognitive tests (Panza et al. 2009a). More recent studies involving older subjects have indicated that a U- or J-shaped curve may best describe the relation between the level of alcohol consumption and cognitive performance (Goodwin et al. 1987; Christian et al. 1995; Hendrie et al. 1996; Launer et al. 1996). In general, light to moderate drinkers of alcohol performed at a higher cognitive level than either abstainers or heavy drinkers (Goodwin et al. 1987; Hendrie et al. 1996), although there were some notable exceptions (Goodwin et al. 1987; Launer et al. 1996). One study in particular, that involved older African American adults found that consumption of up to 5–6 g of alcohol per day (1/2 drinks) was correlated with better cognitive performance than abstainers (Hendrie et al. 1996). On the other hand, in one report, statistical adjustment for age, income, education, and gender rendered the findings non-significant (Goodwin et al. 1987). In another study, the protective effects of moderate alcohol consumption (odds ratio OR (95% confidence interval (CI)) = 0.3 (0.2–0.7) for less than one drink and 0.2 (0.1– 0.4) for 1–2 drinks per day) were limited only to those participants who exhibited clinical conditions associated with atherosclerosis (Launer et al. 1996). However, as discussed above, significant variability in different countries on the level of alcohol content in drinks as well as in the criteria used in different articles to define terms such as light, moderate, and heavy drinking make the interpretation of findings difficult (Panza et al. 2009a). Lower levels of alcohol intake have proportionally greater effects in the elderly, due to their reduced lean body mass and lower percentage of body weight made up of water. In fact, in several cross-sectional studies, moderate drinking, from up to 1 drink per day (up to 14 g of alcohol) to 4 drinks per day (52 g of alcohol), as compared with nondrinking has been associated with a better performance in many cognitive domains (Elias et al. 1999; Carmelli et al. 1999; Bond et al. 2001, 2003). In fact, in the Framingham Heart Study, the association between alcohol consumption and cognitive performance was analyzed separately for men and women, since the researchers anticipated a different gender-based alcohol-cognition relationship (Elias et al. 1999). Test performance of moderate male drinkers (>2 and 4 and 2 drinks/day
In patients with MCI up to 1 drink/day of alcohol or wine may decrease the rate of progression to dementia. No significant associations were found between any levels of drinking and the incidence of MCI in noncognitively impaired individuals vs. abstainers 1,462 women aged Diagnosis of dementia; evaluation of Wine was protective for dementia, and the association was Mehlig Longitudinal, 38–60 years alcohol intake and smoking habits strongest among women who consumed wine only. In et al. (2008) populationcontrast, consumption of spirits at baseline was associated based study with slightly increased risk of dementia (34 years) This table lists the principal findings of longitudinal clinical and epidemiological studies on the relationships among alcohol consumption and dementia, VaD, AD, and MCI, including setting, study design, and cognitive assessment used WAIS Wechsler Adult Intelligence Scale, AD Alzheimer’s Disease, OR Odds Ratio, 95% CI: 95% Confidence Interval, MMSE Mini-Mental State Examination, CDT Clock Drawing Test, VaD Vascular Dementia, MCI Mild Cognitive Impairment, APOE Apolipoprotein E, RR Relative Risk
Deng et al. (2006)
References 188 Alcohol Consumption in Predementia and Dementia Syndromes 3029
3030
F. Panza et al.
glasses of wine each day compared to the abstainers the relative risk for AD was reduced significantly (0.55) (Orgogozo et al. 1997). This protective effect remained significant after further post hoc analyses were conducted (Lemeshow et al. 1998). Furthermore, these cross-sectional findings were confirmed in the subsequent 8-year follow-up of the PAQUID study, where among 2,950 initially nondemented subjects, moderate drinkers had a decreased relative risk of developing dementia (relative risk = 0.56) compared to nondrinkers (Lemeshow et al. 1998). These results were consistent with the findings from some cohort studies (Ruitenberg et al. 2002; Lindsay et al. 2002; Huang et al. 2002; Deng et al. 2006, 98–100, 107), but not others (Hebert et al. 1992; Yoshitake et al. 1995; Broe et al. 1998; Fujishima and Kiyohara 2002; Kivipelto et al. 2001). The relationship between alcohol consumption and risk of dementia (AD, VaD, or other dementia) was also examined in the Rotterdam Study. The findings of this study, with an average follow-up of 6 years, suggested that light to moderate alcohol consumption (1–3 drinks per day) is associated with a reduced risk of dementia in individuals aged 55 years or older (hazard ratio (HR) (95% confidence interval) (CI) = 0.58 (0.38–0.90)); this effect seems to be unchanged by the source of alcohol (Ruitenberg et al. 2002). No association was found in women, whereas a lower risk was found for men drinking 1–3 drinks per day (HR (95% CI) = 0.58 (0.40–0.74)). A modification effect was found when the APOE e4 allele was taken into consideration; the risk was lower among drinkers with an APOE e4 allele, whereas it was less clear for drinkers without the APOE e4 allele (Ruitenberg et al. 2002). This topic was also investigated in the Kungsholmen Project, a community-based dementiafree cohort (n = 402) followed for almost 6 years. Light to moderate alcohol consumption (as defined by them to be 1–21 drinks per week in men, 1–14 drinks per week in women) was associated with a decreased incidence of dementia (relative risk (95% CI) = 0.5 (0.3–0.7)) and AD (relative risk (RR) (95% CI) = 0.5 (0.3–0.7)) (Huang et al. 2002). In the Canadian Study of Health and Aging (CSHA), on 4,615 cognitively normal older subjects reassessed 5 years later, wine consumption (at least weekly consumption) was associated among other factors with a reduced risk of AD (OR (95% CI) = 0.49 (0.28–0.88)) (Lindsay et al. 2002). Finally, in China, light to moderate drinkers (1–21 drinks per week in men, 1–14 drinks per week in women) had a lower risk of dementia (OR (95% CI) = 0.52 (0.32–0.85)) than nondrinkers, but non-significant evidence towards elevated risk in heavy drinkers (OR (95% CI) = 1.45 (0.43–4.89)). A greater reduction of risk was observed for men (OR = 0.37) than for women (OR = 0.76) (Deng et al. 2006). Moreover, the effect of light to moderate drinking seemed most prominent for AD (OR (95% CI) = 0.63 (0.55–0.72)) than for VaD (OR (95% CI) = 0.31 (0.19–0.51)) or other dementia (OR (95% CI) = 0.45 (0.12–1.69)) (Deng et al. 2006). In a nested case–control study on 373 cases with incident dementia and 373 controls who were selected from 5,888 adults aged 65 years and older, and participated in the CHS, the adjusted OR for dementia among whose weekly alcohol consumption was less than 1 drink was 0.65, compared with abstention; was 0.46 and 0.69 compared with 1–6 drinks and with 7–13 drinks, respectively, and was 1.22 when compared with 14 or more drinks (Mukamal et al. 2003). A trend toward greater odds for dementia associated with heavier alcohol consumption was most apparent among men and participants bearing an APOE e4 allele, with similar relationships of alcohol use with AD and VaD (Mukamal et al. 2003). In the Copenhagen City Heart Study which rated alcohol consumption in a different manner to many other studies, the risk of developing dementia was significantly lower among monthly wine drinkers (HR (95% CI) = 0.43 (0.23–0.82)), in weekly wine drinkers (HR (95% CI) = 0.33 (0.23–0.82)) and, but not significantly, in daily drinkers. An increased risk for beer and for spirits was found in monthly, weekly, and daily drinkers, but not significantly. No difference was found between men and women. No association was found between any number of drinks (less than 1, 1–7, 8–14, 15–21, 22, or more) of alcohol consumed per week and the risk of dementia (Truelsen et al. 2002). The findings from the CHS were consistent with the PAQUID Study (Orgogozo et al. 1997) and the Rotterdam Study (Ruitenberg et al. 2002), but suggested a higher risk of dementia
188 Alcohol Consumption in Predementia and Dementia Syndromes
3031
with consumption greater than 2 drinks per day. Surprisingly, the Rotterdam Study found that the lower risk of dementia associated with alcohol use was more consistent among individuals with an APOE e4 allele (Ruitenberg et al. 2002), but no significant interaction was detected. In the Washington Heights Inwood-Columbia Aging Project, with 908 subjects aged 65 years and older, the number of drinks per week was collected at baseline and subjects were classified as nondrinkers, light drinkers (less than 1 drink per month to 6 drinks a week), moderate drinkers (1–3 drinks a day), and heavy drinkers (more than 3 drinks a day). The light and moderate drinking categories were combined because of the low number of moderate drinkers. A significantly lower risk of AD was found in light to moderate wine drinkers in elderly individuals without the APOE e4 allele (HR = 0.44, p = 0.004). No modification effect by sex was found (Luchsinger et al. 2004). Some other population-based, prospective studies, with longer follow-up periods studied the effects of different patterns of alcohol intake on dementia (Järvenpää et al. 2005; Simons et al. 2006; Mehlig et al. 2008). In fact, in the Finnish Twin Cohort Study with a follow-up period of 25 years, reports of midlife binge drinking (i.e., alcohol exceeding the amount of five bottles of beer or a bottle of wine on one occasion at least once per month) or losing consciousness due to excessive alcohol intake at least twice during the previous year were risk factors for dementia later in life (RR (95% CI) = 3.2 (1.2–8.6)) (Järvenpää et al. 2005). A longitudinal cohort study conducted in Dubbo (Australia) on 2,805 subjects aged 60 years and older, initially free of cognitive impairment and followed for 16 years, confirmed as a modest intake of alcohol seemed to offer substantial protection against the onset of dementia, showing larger effect for 15–28 units per week (HR (95% CI); 0.40 (0.21–0.79) (Simons et al. 2006; McCallum et al. 2007). Finally, in the Prospective Population Study of Women in Goteborg, Sweden, in a 34-year follow-up, wine was protective for dementia (current drinking vs former or never drinking) (HR (95% CI) = 0.6 (0.4–0.8)), and the association was strongest among women who consumed wine only (HR (95% CI) = 0.3 (0.1–0.8)). In contrast, consumption of spirits at baseline was associated with slightly increased risk of dementia (HR (95% CI) = 1.5 (1.0–2.2)) (Mehlig et al. 2008). For predementia syndromes, in a group of 369 nondemented, community-dwelling older men who participated in the National Heart, Lung, and Blood Institute (NHLBI) Twin Study, alcohol consumption was found to be slightly protective (relative risk (95% CI) = 0.93 (0.88–0.99)), but if individuals with CVD were excluded from the analysis this association disappeared (DeCarli et al. 2001). Recently, the impact of alcohol consumption on the incidence of MCI was evaluated in 1,445 cognitively normal individuals and on its progression to dementia in 121 patients with MCI, aged 65–84 years, participating in the Italian Longitudinal Study on Aging (ILSA), a large, populationbased, prospective study with a sample of 5,632 subjects 65–84 years old with a 3.5-year follow-up. Patients with MCI who consumed up to 1 drink per day of alcohol had a reduction in the rate of progression to dementia in comparison with patients with MCI who never consumed alcohol (HR (95% CI) = 0.15 (0.03–0.78)). Overall, patients with MCI who consumed less than 1 drink per day of wine compared to nondrinkers had a decrease in the rate of progression to dementia of about 85% (HR (95% CI) = 0.15 (0.03–0.77)). Moderate intake of alcohol deriving from wine, in drinks controlled for the intake of alcohol deriving from other sources across a number of levels, was also associated with a significantly lower rate of progression to dementia. No significant associations were found between any levels of drinking and the incidence of MCI in non-cognitively impaired individuals versus abstainers (Solfrizzi et al. 2007) (Fig. 188.1). To the best of our knowledge, only two other studies have examined the effect of alcohol consumption on risk for the incidence of MCI (Anttila et al. 2004; Espeland et al. 2005). After an average follow-up of 23 years, nondrinkers (OR (95% CI) = 2.15 (1.01–4.59)) and frequent drinkers (OR (95% CI) = 2.57 (1.19–5.52)) were both more than twice as likely to have MCI in old age as occasional drinkers (Anttila et al. 2004). However, the APOE genotype seemed to modify the relationship, such
3032
F. Panza et al.
Fig. 188.1 Multivariate hazard ratios of incident mild cognitive impairment (MCI) among noncognitively impaired subjects and progression to dementia among patients with MCI. The Italian Longitudinal Study on Aging (ILSA). This figure shows the multivariate hazard ratios of incident MCI among noncognitively impaired subjects (Model 1) and progression to dementia among patients with MCI (Model 2) who drank 2 alcoholic drinks/day (rhomb with solid line) versus abstainers (from Solfrizzi et al. Neurology. 2007;68:1790–89, with permission)
that the risk of old age dementia increased with increasing midlife alcohol consumption only among carriers of the APOE e4 allele (Anttila et al. 2004). In the ILSA sample, we failed to confirm these findings, but we note that the alcohol consumption reported was a midlife determination (Solfrizzi et al. 2007). Probably, a follow-up period longer than 3.5 years would have revealed that moderate alcohol consumption might influence the incidence of MCI. On the other hand, our findings are consistent with those obtained in the Women’s Health Initiative Memory Study with a 4.2 year-followup, which found that moderate alcohol intake (³1 drink per day) was associated with an approximately 50% reduced risk of combined probable dementia and MCI (OR (95% CI) = 0.4 (0.28–0.99)) (Espeland et al. 2005). However, after adjusting for demographic and socioeconomic factors and baseline Modified Mini Mental State Examination (3MSE), the significance disappeared (Espeland et al. 2005). Currently, ours is the first study in which alcohol consumption was associated with the rate of progression of MCI to dementia, in fact, up to 1 drink/day of alcohol or wine may decrease the rate of progression to dementia in patients with MCI (Solfrizzi et al. 2007). Very recently, a systematic review with meta-analysis was carried out to investigate any relationship between incident cognitive decline or dementia in the elderly and alcohol consumption between 1995 and March 2006 that only included data on subjects aged ³65 from longitudinal studies (Peters et al. 2008). The quality of the studies included in this meta-analysis was assessed using standard criteria that measured key factors including appropriate design, recruitment, analysis, and provision of suitable information relating to key aspects of the study (Peters et al. 2008). Outcomes measured were very variably defined ranging from dementia alone or as well as various subtypes (mainly AD and VaD) or of subtypes of dementia (AD and VaD) alone or in combination (Peters et al. 2008). Eleven studies assessed cognitive decline or predementia syndromes (Peters et al. 2008). Studies with inadequate definition of the outcomes of interest were also excluded. Meta-analyses suggested that, at least in epidemiological studies, light to moderate alcohol use was associated with a 38% reduced risk of unspecified incident dementia (Fig. 188.2). Also in the studies included in this metaanalysis, there was no close agreement as to what might be considered an “appropriate” level of
188 Alcohol Consumption in Predementia and Dementia Syndromes
3033
Fig. 188.2 Dementia and alcohol consumption. Forest plot generated using a random effects model in order to allow graphical representations of the findings of various studies with dementia as an outcome (from Peters et al. Age Ageing. 2008;37:505–12, with permission)
alcohol consumption particularly since the classification of light to moderate drinking varied widely. Possible benefit against dementia was shown for a variety of definitions including more than 1 drink per day, for weekly or monthly wine consumption; for 250–500 ml (usually wine); for more than 3 drinks per day and from 1 to 28 units per week (Peters et al. 2008). For effects on AD risk, light to moderate alcohol was associated with a significantly reduced risk of 32%, defining optimal amounts as weekly consumption of wine, 1–6 or more than 2 drinks per week, or more than 3 drinks/250–500 ml per day (usually wine), or where studied by gender, 1–3 per day in males (Peters et al. 2008). Although the point estimates were also in a similar direction for VaD and cognitive decline (0.82 and 0.89, respectively), the results were not statistically significant (Peters et al. 2008) (Fig. 188.3). With regard to effects on cognitive function, results for what could be considered “optimal” or non-deleterious consumption were mixed regarding the amount consumed per month or per day, in subjects with cardiovascular disease or diabetes where results ranged from 1–2 drinks per week, while for VaD patients 1–3 drinks per day in males appeared to be beneficial (Peters et al. 2008). Finally, another very recent meta-analysis included 15 prospective studies (follow-ups ranged from 2 to 8 years), with samples including 14,646 participants evaluated for AD, 10,225 participants evaluated
3034
F. Panza et al.
Fig. 188.3 Alzheimer’s disease (AD) and alcohol consumption. Forest plot generated using a random effects model in order to allow graphical representations of the findings of various studies with AD as an outcome (from Peters et al. Age Ageing. 2008;37:505–12, with permission)
for VaD, and 11,875 followed for any type of dementia (Anstey et al. 2009). This meta-analysis indicated that light to moderate alcohol intake was associated with a 25–28% reduction in risk of AD, VaD, and any dementia compared with alcohol abstinence in older adults. Heavy drinking was not associated with increased risk of dementia in the present study (Anstey et al. 2009).
188.4 Alcohol and Cognitive Disorders: Possible Mechanisms The mechanism by which light alcohol intake could be protective against cognitive decline impairment or decline in older age or against predementia and dementia syndromes is, at present, unclear (Fig. 188.4). The association between alcohol drinking and cognitive function could have different explanations in relation to cognitive domains explored. White matter lesions (WMLs), for example, would play a neuropathological role, given that alcohol drinking influenced measures of psychomotor speed, episodic memory, and executive function (Panza et al. 2009a). In fact, alcohol drinking has been associated with fewer brain infarcts and was shown to have a U-shape relationship with the prevalence of WMLs (Panza et al. 2009a). Furthermore, WMLs and infarcts, in turn, may reflect a vascular mechanism responsible for the observed association between alcohol and cognitive functions (Panza et al. 2009a). Different mechanisms may underlie the observed adverse effects of heavy
188 Alcohol Consumption in Predementia and Dementia Syndromes
3035
Association with fewer brain infarcts and U shape relationship with white matter lesions Stimulation the release of hippocampal acetylcholine in animal models
Association in APOE ε4 carriers with hippocampal and amygdalar atrophy
Light to moderate alcohol consumption
Protection against dementia via a reduction in vascular risk factors - increased prostacyclin concentrations -- reduced generation of thromboxane A2 - inhibition of platelet function -increased levels of HDL cholesterol
Wine consumption may exert a protective effect also through the antioxidant effects of polyphenols richly represented in red wine
Fig. 188.4 Synopsis of the mechanism by which light to moderate alcohol intake could be protective against cognitive impairment or decline in older age or against predementia and dementia syndromes. This figure lists various possible mechanisms of light to moderate alcohol consumption linked to their neuroprotective properties against cognitive decline including modulation of cerebrovascular disease (e.g., white matter lesions), stimulation of the release of hippocampal acetylcholine in animal models, association with hippocampal and amygdalar atrophy, reduction in vascular risk factors or through the antioxidant effects of polyphenols richly represented in red wine
drinking and the potential beneficial effects of light to moderate drinking, and may also partly explain why deficits are reported in certain functions (e.g., delayed recall), whereas benefits are seen in others (e.g., learning) (Panza et al. 2009a). Indeed, higher doses of alcohol may affect cognitive functioning through increased release of acetylcholine from the hippocampus (Fadda and Rossetti 1998), which is an important neurotransmitter in memory and attention (Panza et al. 2009a). In contrast, evidence from animal studies indicated that low doses of alcohol may stimulate the release of hippocampal acetylcholine (Stancampiano et al. 2004). Finally, moderate alcohol consumption was associated with lower rates of cardiovascular disease and reduced cardiovascular risk factors (Naimi et al. 2005), which may serve to protect brain vasculature and prevent subclinical strokes, resulting in better preservation of cognitive performance (Panza et al. 2009a). Alcohol consumption might protect from unspecified dementia by effects on the cerebral vasculature, supporting the observation that light to moderate alcohol intake might be protective against ischemic stroke (Palomaki and Kaste 1993). Moreover, light to moderate alcohol use is associated with a lower prevalence of MRI-defined WMLs and subclinical infarcts (den Heijer et al. 2004), although MRI abnormalities, HDL cholesterol levels, and fibrinogen levels only marginally influenced the association of alcohol consumption and dementia in the CHS (Panza et al. 2009a). The suggested protection against CVD resulting from light to moderate alcohol consumption may explain the perceived protection alcohol offers against VaD. In the Rotterdam study, the protective effect of alcohol consumption was found mainly for VaD, and the authors suggested that moderate alcohol intake might protect against dementia via a reduction in vascular risk factors (Panza et al. 2009a). In fact, moderate doses of alcohol may alter blood clotting mechanisms though increase of prostacyclin concentrations and reduction of thromboxane A2 generation and thus may inhibit platelet function (Panza et al. 2009a). Moderate alcohol consumption may also contribute to increased plasma levels of endogenous tissue-type plasminogen activator (tPA), a serine protease that regulates intravascular
3036
F. Panza et al.
fibrinolysis, and fibrinolytic activity while decreasing plasma fibrinogen levels (Panza et al. 2009a). It is also known that alcohol is associated with increased levels of HDL cholesterol, its subfractions HDL2 and HDL3, and its associated apolipoproteins A-I and A-II (Panza et al. 2009a). The association with HDL cholesterol is deemed to account for up to a half of the reduction in coronary events associated with moderate alcohol consumption (Panza et al. 2009a). While the aforementioned factors could affect the risk of unspecified dementia and, probably, of VaD, other experimental and clinical findings may partly explain the suggested protection against AD provided by light to moderate alcohol consumption. Small amounts of alcohol have been reported to be associated with a lower prevalence of vascular brain findings and, in APOE e4 carriers, with hippocampal and amygdalar atrophy, as assessed by MRI (den Heijer et al. 2004). Experimental studies found that ethanol initially increases hippocampal acetylcholine release, which could conceivably improve memory performance (Panza et al. 2009a). Processes that originate, modulate, or precipitate the deposition of amyloid beta (bA) in the brain, such as oxidative stress, rather than vascular processes, may better explain the development of AD, and the vascular effects of the alcohol in alcoholic beverages may not be enough to explain the protective effects of the moderate intake of alcohol from dementia. The previously mentioned contribution of alcohol to increased plasma levels of endogenous tPA (Panza et al. 2009a) could promote Ab removal since tPA is responsible for the formation of plasmin which is suggested to be an Ab degrading enzyme (Miners et al. 2008). Wine consumption may exert a protective effect, either through alcohol intake itself, or through the antioxidant effects of polyphenols richly represented in red wine, or through both (Panza et al. 2009a). Although there have been some effects associated with alcohol-free red wine (Panza et al. 2009a), the constituents of red wine also have potentially beneficial vascular effects (139, 140) including, enhanced endothelial nitric oxide release (141), and reduced atherosclerosis in APOE-deficient mice (Waddington et al. 2004; Panza et al. 2009a). Red wine polyphenols are a complex mixture of flavonoids (mainly anthocyanins and flavan-3-ols) and non-flavonoids (such as resveratrol and gallic acid). Flavan-3-ols are the most abundant, with oligomeric and polymeric procyanidins (condensed tannins) often representing 25–50% of the total phenolic constituents (Waterhouse 2002). A recent study identified procyanidins as the principal vasoactive polyphenols in red wine and showed that they are present at higher concentrations in wines from areas of Southwestern France and Sardinia, where traditional production methods ensure that these compounds are efficiently extracted during vinification (Corder et al. 2006). Given the link between VaD, vascular risk, and the increasing body of evidence suggesting that AD may be influenced by vascular factors (Panza et al. 2009a), it may be concluded that the vascular protection associated to wine consumption decreases the risk of incident dementia/AD. In fact, in the 5-year follow-up PAQUID cohort, a significant inverse association between flavonoid intake and the risk of dementia was found (Commenges et al. 2000). It has been also suggested that the antioxidant properties of the flavonoids in wine may help prevent the oxidative damage implicated in dementia. Furthermore, oxidative stress is also likely to develop in the brain, contributing to neuronal death by various mechanisms such as formation of bA peptide, DNA damage, and abnormal tau protein (Panza et al. 2009a). The presence in wine of non-alcoholic components, such as particular antioxidants, could explain a differential effect of wine on dementia since liquor which does not seem to have as strong an effect has been shown to have less antioxidant activity than wine (Panza et al. 2009a). Nonetheless, in some studies on the neuroprotective role of moderate alcohol consumption, the most typically consumed alcohol types were beer and spirits (Panza et al. 2009a). It is also possible that moderate lifestyles in general, which obviously vary according to different cultural environments, protect from cognitive impairment. Thus, it may not be the direct effect of alcohol or specific substances in alcoholic drinks that provide the protection, but moderate alcohol drinking may be an indicator of a complex set of favourable social and lifestyle factors. A protective
188 Alcohol Consumption in Predementia and Dementia Syndromes
3037
effect of alcohol on cognitive function in moderate drinkers may reflect a relatively poor health status among abstainers or because cognitive status influences alcohol consumption and overall health status (Panza et al. 2009a). The Mediterranean diet may serve as an interesting model for further studies of the association between alcohol and cognitive functioning, given the suggested role of many components of this diet (monounsaturated fatty acids (MUFA), polyunsaturated fatty acids (PUFA), and red wine) (Panza et al. 2009a) in different degrees of cognitive impairment. Indeed, the typical dietary pattern of Mediterranean diet is characterized by high intake of vegetables, fruits and nuts, legumes, cereals, fish, and MUFA with relatively low intake of meat, and dairy products, and moderate consumption of alcohol. Recently, in a community-based study involving 2,258 nondemented individuals in New York, adherence to a traditional Mediterranean diet was associated with a significant reduction in risk for AD (Scarmeas et al. 2006). However, in this study, the median daily intake of MUFA to saturated fatty acids (SFA) ratio for individual food categories by Mediterranean diet score tertiles was 4 drinks on a single occasion for men or >3 drinks for women within 2 h (Panza et al. 2009a). However, definitions are not standardized in the literature and are not agreed on by all researchers (Panza et al. 2009a). In the United States, a standard alcoholic drink is more specifically defined as 12 ounces of regular beer, five ounces of wine, 1.5 ounces of 80 proof distilled spirits or liquor (gin, rum, vodka, whiskey), 17.74 ml of alcohol or 14 g of alcohol (Panza et al. 2009a). Thus, at the extreme, 2 drinks a day would be up to approximately 30 g of alcohol.
3038
F. Panza et al.
In particular, for cardiovascular conditions, reduction in both incidence of clinical CAD and severity of angiographically documented CAD has been documented with moderate alcohol drinking (Panza et al. 2009a). The benefit was consistently shown with various types of alcoholic drinks and drinking patterns (Panza et al. 2009a). Until large, prospective, randomized trials are available, which may be very difficult to undertake in practical terms, the preponderance of data that suggest that consumption of 1–2 drinks in men and 1 in women will be beneficial to the cardiovascular system (Panza et al. 2009a). Furthermore, heavy drinking is associated with increased hemorrhagic stroke risk, and there is some consensus about this relation in relevant reports (Klatsky 2005). The antithrombotic actions of alcohol might also result in increased hemorrhagic stroke risk at moderate drinking levels, but reports differ about whether lighter drinking increases hemorrhagic stroke risk or is unrelated (Panza et al. 2009a). The alcohol–hemorrhagic stroke relation seems similar for subarachnoid and intracerebral hemorrhage (Klatsky 2005). Reports are less concordant about alcohol– ischemic stroke interactions, but several analyses suggest a U-shaped or J-shaped distributions when the amount of drinking is plotted against ischemic stroke risk (Klatsky 2005). A meta-analysis performed on 35 studies on this topic revealed that heavy alcohol consumption increases the relative risk of stroke while light to moderate alcohol consumption may be protective against total and ischemic stroke (Reynolds et al. 2003). Nonetheless, the Cardiovascular Health Study (CHS) confirmed that association of alcohol use and risk of ischemic stroke was a U-shaped distribution, with modestly lower risk among consumers of 1–6 drinks per week, but even moderate alcohol intake may be associated with an increased risk of ischemic stroke among APOE e4-carrier older adults (Mukamal et al. 2005a). Finally, in the Health Professionals Follow-up Study, a prospective analysis of more than 38,000 men who were free of known cardiovascular disease, alcohol consumption appeared to be associated with a higher risk for ischemic stroke among those who consumed more than 2 drinks per day (Mukamal et al. 2005b). However, alcohol was apparently not associated with risk at lower levels of intake. A pattern of intake characterized by moderate use 3–4 days per week appeared to be associated with the lowest risk. Red wine consumption was inversely associated with ischemic stroke risk, an association that was significantly different from the comparable associations of other beverages. These findings may also offer support for stroke reduction where to gain cardiovascular benefits people should avoid consuming more than 2 drinks daily (Mukamal et al. 2005b). Regular, moderate wine consumption is often associated with reduced morbidity and mortality from and to a variety of chronic diseases in which inflammation is a root cause (Walzem 2008). Wine comes in a wide variety of styles that contain different ethanol and polyphenol contents. Controversy remains as to whether the alcohol or polyphenols contribute more to the health benefits of regular moderate wine consumption. The overall effect of wine consumption on health depends upon the total amounts consumed, the style and possibly the pattern of consumption (Walzem 2008). The apparent effect of wine consumption may be modified by other aspects of diet, for example, those consuming various levels of alcohol may also consume differential volumes of fruit, vegetable, and whole grain, and as such phytochemical intake and benefit may vary, particularly if wine may serve as a primary dietary source of phytochemical (Walzem 2008).
188.6 Conclusions Drugs currently used in the symptomatic treatment of cognitive impairment and dementia unfortunately have a very limited therapeutic value, particularly on the management of psychiatric and behavioural symptoms. This is a constant reminder of the necessity to seek new therapeutic options to slow down the progression of predementia and dementia syndromes.
188 Alcohol Consumption in Predementia and Dementia Syndromes
3039
In the past few years, vascular and lifestyle-related factors for predementia and dementia syndromes have been an area of intensive research. At present, cumulative evidence has suggested that vascular risk factors may be important in the development of MCI, dementia, and AD (Panza et al. 2009a). Light to moderate alcohol drinking has been proposed as a protective factor against MCI and dementia in several longitudinal studies, but contrasting findings also exist and deriving overall conclusions from these studies has many limitations. Many of these studies were limited to cross-sectional design, restriction by age or sex, or incomplete ascertainment. Moreover, outcomes measured, the range of beverages that are available, drinking patterns and how they are categorised as well as follow-up periods studied and possible interactions with other lifestyle-related (i.e., smoking) or genetic-related (i.e., APOE genotype) factors are all sources of great variability. Indeed, the body of evidence examined in this chapter and the recent meta-analyses cited (Peters et al. 2008; Anstey et al. 2009) of all research published within the last 10 years suggested that light to moderate alcohol use may be associated with a reduced risk of unspecified incident dementia and AD, while for VaD, cognitive decline, and predementia syndromes the current evidence is only suggestive of a protective effect. Thus, at present, the most commonly used outcome measures for these kinds of studies appear to be unspecified dementia and AD, where prevalence and diagnostic classifications make the studies more practical compared with the small number of studies reporting VaD and other subtypes of cognitive decline. These are more difficult to classify, which is often further complicated by the relatively high vascular factor contribution to AD cases. On the other hand, the cardiovascular mechanisms that related to the suggested protective effects of alcohol may have an even greater effect on VaD (Table 188.7). For the different types of beverages that exist, several studies have suggested that light to moderate wine consumption may be protective against dementia and cognitive decline but not total alcohol intake, beer, or spirits, although this is not conclusive (Panza et al. 2009a). In this context, it would be desirable to differentiate types of wine (e.g., red, white, rosé), but this information is not currently available. The failure to detect differences between types of beverages in some studies might be a consequence of differential gender interactions, although these analyses were adjusted for gender. Other issues are that some geographical regions typically consume specific types of alcoholic beverage (e.g., beer and spirits in Northern Europe), although observed associations and benefits offered by specific types of beverages were inconclusive. In general, wine-only drinkers may tend to consume wine less often than those who also drink other kinds of alcohol. In this way, wine-only drinkers could be viewed as the ones who consume the moderate quantities of ethanol beneficial for health, while the others may consume larger quantities; negative consequences of ethanol may outweigh the positive effects of healthy ingredients. Of course, measurement of alcohol consumption
Table 188.7 Key points of clinical and epidemiological studies on the relationships among alcohol consumption and dementia and predementia syndromes A. Among lifestyle-related factors, low to moderate alcohol drinking has been proposed as a protective factor against the development of age-related changes of cognitive function, predementia syndromes, and cognitive decline of degenerative (Alzheimer’s disease, AD) or vascular origin (vascular dementia, VaD) in several longitudinal studies, but contrasting findings also exist B. Light to moderate alcohol use may be associated with a reduced risk of unspecified incident dementia and AD, while for VaD, cognitive decline, and predementia syndromes the current evidence is only suggestive of a protective effect C. Epidemiological evidence suggested that the protective effects of alcohol are more likely with wine consumption and the absence of an apolipoprotein E (APOE) e4 allele. At present, there is no indication that light to moderate alcohol drinking would be harmful to cognition and dementia, and it is not possible to define a specific beneficial level of alcohol intake This table lists the principal features of clinical and epidemiological studies on the relationships among alcohol consumption and dementia (AD and VaD) and predementia syndromes
3040
F. Panza et al.
must consider both volume consumed and the alcohol content of drinks, which is very variable in everyday living and in previous research, and this will remain an important consideration in considering previous research and planning future studies. Whether the present divergent findings can be explained by the drinking patterns has not been extensively investigated, and studies with long follow-up periods are few (Panza et al. 2009a). It is possible that over time, and perhaps in association with cognitive decline, alcohol consumption patterns will change, and, therefore, long-term follow-up studies are needed to clarify the issue. Furthermore, with respect to the measurement of alcohol intake, in addition to the volume and alcohol content issue mentioned, the amount of detail collected on patterns of consumption varies widely in some studies which often range from data on recent consumption to some assessing historial patterns of consumption or lifetime abstinence/change in consumption. This may be particularly pertinent in the elderly as people may reduce consumption as they age, although not all studies agree with this (Panza et al. 2009a), however, it also contains hazards associated with recall bias. It may be that the prevalence of binge drinking has only increased recently, especially for women, but it is surprising that so few studies assessed this in their participants (Panza et al. 2009a). Therefore, the impact of alcohol drinking on cognitive functions might be somewhat different at different time points in individual’s lives. Nonetheless, the current balance of evidence suggests that alcohol consumption in old age substantially agrees with that of midlife consumption in relation to cognitive decline (Panza et al. 2009a). At present, there is no evidence indicating whether starting to drink at a later age would be beneficial against predementia or dementia syndromes. Furthermore, smoking may be a significant confounder of alcohol effects and thus both may not only individually affect dementia and cognitive decline, but each may also modify the effects of the other. In fact, there were several studies focused on alcohol and cognitive impairment or decline, in which there was no evidence of an interaction between alcohol and smoking on cognitive measures, with a few notable exceptions (Panza et al. 2009a). The overall findings from three Canadian data sets suggested that smoking may reduce the risk of dementia among drinkers (Tyas et al. 2000) while another recent study reported that these interactions were significant only for wine and smoking, suggesting that ingredients other than ethanol in wine may be protective against the adverse effects of smoking or of course it could suggest that there is higher consumption of wine amongst smokers compared with nonsmokers (Mehlig et al. 2008). Further analysis of this interaction, particularly in longitudinal studies and within genetic risk groups, is needed to determine whether this interaction can be replicated. Finally, genetic susceptibility seems to modify the effect of alcohol on risk of dementia and predementia syndromes, with some, but not all studies suggesting that the APOE e4 carrier status acting as a possible effect modifier for these associations (Panza et al. 2009a). Nonetheless, the protective effects of light to moderate alcohol consumption against dementia and cognitive decline are more likely in the absence of an APOE e4 allele (Panza et al. 2009a). The effect of the APOE e4 allele on dementia is suggested to attenuate with increasing age, which might partly explain the conflicting results concerning effect modification in studies on elderly cohorts (65 years and older) (Panza et al. 2009a). Very recently, in the CAIDE study, after an average follow-up of 21 years, there was a statistically significant multiplicative (p-values: infrequent drinking = 0.04, frequent drinking = 0.03) and additive interaction (infrequent drinking: relative excess risk from interaction (RERI) = 1.97, p < 0.001, frequent drinking: RERI = 2.95, p < 0.001) between the APOE e4 carrier status and alcohol drinking for dementia risk (Kivipelto et al. 2008). One possible explanation could be that people with the e4 allele have less effective neural repair mechanisms (Panza et al. 2009a), and thus they would be more susceptible to the deleterious effects of alcohol. It is also possible that it is a particular drinking pattern (i.e., binge drinking) which together with APOE e4 carrier status forms a hazardous combination in some populations (Panza et al. 2009a). However, what has yet to be investigated as a possible additional variable is whether any additional genetic variants, such as genes encoding enzymes involved in alcohol metabolism, may also be involved.
188 Alcohol Consumption in Predementia and Dementia Syndromes
3041
Ultimately, since intervention studies are unlikely to be performed in this area, perhaps the most useful evidence comes from overview of epidemiological studies which still have a number of limitations. At present, there doesn’t appear to indication that light to moderate alcohol drinking would be harmful to cognition and dementia. However, it is not possible to define a specific beneficial level of alcohol intake in relation to what might be beneficial to cognitive function and dementia. Acknowledging the well-known harmful effects related to heavy consumption of alcohol and the lack of long-term follow-up studies or randomized controlled trials, it would be premature to recommend whether some alcohol consumption, even to abstainers would be helpful in preventing cognitive decline or predementia syndromes or even delaying the onset of dementia. Possibly, at present, in moderate to heavy alcohol drinkers, experiencing memory difficulty, or with suggested diagnoses of MCI or early AD, among management options, given the challenge in achieving total abstinence, allowing some light to moderate drinking (rather than encouraging which may convey confusing health messages), provided there are no other contraindications to suggest otherwise may at the very least not contribute significant additional harm (Panza et al. 2009a).
Summary Points • An increasing body of epidemiological evidence suggests that light to moderate alcohol consumption could offer some benefit for some health outcomes including cognitive function and heart disease. • In particular, light to moderate alcohol drinking has been proposed, although not universally, as a protective factor against the development of age-related changes of cognitive function including predementia syndromes, and cognitive decline of degenerative (Alzheimer’s disease, AD) or vascular origin (vascular dementia, VaD) in several longitudinal studies. • Different outcome measures, types of beverages, measurement of drinking patterns, or follow-up periods, or possible interactions with other lifestyle-related (e.g., smoking) or genetic factors (e.g., APOE or other genes) are current sources of great variability where there need to be efforts to refine. • Light to moderate alcohol consumption may be associated with a reduced risk of unspecified incident dementia and AD, while it may also be a benefit for less studied VaD, cognitive decline, and predementia syndromes. • Reported protective effects from alcohol consumption seem to be most likely attributed to the consumption of wine and in people not having APOE e4 alleles. However, at present, there is no indication whether light to moderate alcohol drinking would be harmful to cognition and dementia, and it is not possible to define a specific beneficial level of alcohol intake.
Key Terms and Definitions Alcohol: Any organic compound where a hydroxyl group (−OH) is bound to a carbon atom of an alkyl or substituted alkyl group. Generally, the word alcohol refers to ethanol, the type of alcohol found in alcoholic beverages. Ethanol is a colourless, volatile liquid with a mild odour which can be obtained by the fermentation of sugars. Dementia: It is a syndrome defined by impairments in memory and other cognitive functions that are severe enough to cause significant impairment and decline from a previous level of social and occupational functioning.
3042
F. Panza et al.
Predementia syndrome: This term identifies all conditions with age-related deficits in cognitive function, including a mild stage of cognitive impairment based on a model of normality or also due to pathological conditions considered predictive or early stages of dementia. Mild cognitive impairment (MCI): It is a clinical label that includes non-demented aged persons with memory impairment that is more pronounced than what would be expected normal for that age but not severe enough as to cause significant disability. Age-related cognitive decline (ARCD): It is defined as an objectively identified decline in cognitive functioning consequent to the aging process that is within normal limits given the person’s age, not as pronounced as for MCI, but for which there are no defined diagnostic criteria, and scant usage in epidemiological studies.
Acknowledgments This work was supported by the Italian Longitudinal Study on Aging (ILSA) (Italian National Research Council - CNR-Targeted Project on Ageing - Grants 9400419PF40 and 95973PF40).
References Anstey KJ, Mack HA, Cherbuin N. Am J Geriatr Psychiatry. 2009;17:542–55. Anttila T, Helkala EL, Viitanen M, Kåreholt I, Fratiglioni L, Winblad B, Soininen H, Tuomilehto J, Nissinen A, Kivipelto M. B.M.J. 2004;329:539. Bond GE, Burr R, McCurry SM, Graves AB, Larson EB. Int Psychogeriatr. 2001;13:207–23. Bond GE, Burr R., Rice MM, McCurry SM, Graves AB, Teri L, Bowen JD, McCormick WC, Larson EB. J Aging Health. 2003;15:371–90. Bond GE, Burr R, McCurry SM, Rice MM, Borenstein AR, Kukull WA, Teri L, Bowen JD, McCormick WC, Larson EB. J Aging Health. 2004;16:615–40. Bond GE, Burr R, McCurry SM, Rice MM, Borenstein AR, Larson EB. Int Psychogeriatr. 2005;17:653–68. Broe GA, Creasey H, Jorm AF, Bennett HP, Casey B, Waite LM, Grayson DA, Cullen J. Aust N. Z. J. Public Health. 1998;22:621–3. Carmelli D, Swan GE, Reed T, Schellenberg GD, Christian JC. Neuroepidemiology. 1999;18:125–33. Cervilla JA, Prince M, Mann A. J Neurol Neurosurg Psychiatry. 2000a;68:622–6. Cervilla JA, Prince M, Joels S, Lovestone S, Mann A. Long-term predictors of cognitive outcome in a cohort of older people with hypertension. Br J Psychiatry. 2000b;177:66–71. Commenges D, Scotet V, Renaud S, Jacqmin-Gadda H, Barberger-Gateau P, Dartigues J-F. Eur J Epidemiol. 2000;16:357–63. Corder R, Mullen W, Khan NQ, Marks SC, Wood EG, Carrier MJ, Crozier A. Nature. 2006;444:566. Christian JC, Reed T, Carmelli D, Page WF, Norton JA Jr, Breitner JC. J Stud Alcohol. 1995;56:414–6. DeCarli C, Miller BL, Swan GE, Reed T, Wolf PA, Carmelli D. Arch Neurol. 2001;58:643–7. Deng J, Zhou DH, Li J, Wang YJ, Gao C, Chen M. Clin Neurol Neurosurg. 2006;108:378–83. Dent OF, Sulway MR, Broe GA, Creasey H, Kos SC, Jorm AF, Tennant C, Fairley MJ. B.M.J. 1997;314:l655–7. den Heijer T, Vermeer SE, van Dijk EJ, Prins ND, Koudstaal PJ, van Duijn CM, Hofman A, Breteler MM. Am J Clin Nutr. 2004;80:992–7. Dufouil C, Ducimetière P, Alperovitch A. Am J Epidemiol. 1997;146:405–12. Dufouil C, Tzourio C, Brayne C, Berr C, Amouyel P, Alperovitch A. Epidemiology. 2000;11:280–4. Edelstein SL, Kritz-Silverstein D, Barrett-Connor E. J Women’s Health. 1998;7:1271–81. Elias PK, Elias MF, D’Agostino RB, Silbershatz H, Wolf PA. Am J Epidemiol. 1999;150:580–9. Elwood PC, Gallacher JE, Hopkinson CA, Pickering J, Rabbitt P, Stollery B, Brayne C, Huppert FA, Bayer A. J Epidemiol Community Health. 1999;53:9–14. Erkinjuntti T, Inzitari D, Pantoni L, Wallin A, Scheltens P, Rockwood K, Desmond DW. Ann N. Y. Acad Sci. 903: 262–272. Espeland MA, Gu L, Masaki KH, Langer RD, Coker LH, Stefanick ML, Ockene J, Rapp SR. Am J Epidemiol. 2005;161:228–38. Fadda F, Rossetti ZL. Prog Neurobiol. 1998;56:385–431.
188 Alcohol Consumption in Predementia and Dementia Syndromes
3043
Fujishima M, Kiyohara Y. Ann N. Y. Acad Sci. 2002;977:1–8. Galanis DJ, Joseph C, Masaki KH, Petrovich H, Ross GW, White L. Am J Public Health. 2000;90:1254–9. Ganguli M, Vander Bilt J, Saxton JA, Shen C, Dodge HH. Neurology. 2005;65:1210–7. Goodwin JS, Sanchez CJ, Thomas P, Hunt C, Garry PJ, Goodwin JM. Am J Public Health. 1987;77:173–7. Hebert LE, Scherr PA, Beckett LA, Funkenstein HH, Albert MS, Chown MJ, Evans DA. Am J Epidemiol. 1992;135:347–55. Herbert LE, Scherr PA, Beckett LA, Albert MS, Rosner B, Taylor JO, Evans DA. Am J Epidemiol. 1993;137:881–91. Hendrie HC, Gao S, Hall KS, Hui SL, Unverzagt FW. J Am Geriatr Soc. 1996;44:1158–65. Huang W, Qiu C, Winblad B, Fratiglioni L. J Clin Epidemiol. 2002;55:959–64. Järvenpää T, Rinne JO, Koskenvuo M, Räihä I, Kaprio J. Epidemiology. 2005;16:766–71. Klatsky AL. Stroke. 2005;36:1835–6. Kivipelto M, Helkala EL, Hänninen T, Laakso MP, Hallikainen M, Alhainen K, Soininen H, Tuomilehto J, Nissinen A. B.M.J. 2001;322:1447–51. Kivipelto M, Rovio S, Ngandu T, Kåreholt I, Eskelinen M, Winblad B, Hachinski V, Cedazo-Minguez A, Soininen H, Tuomilehto J, Nissinen A. J Cell Mol Med. 2008;12:2762–71. Larrieu S, Letenneur L, Helmer C, Dartigues JF, Barberger-Gateau P. J Nutr Health Aging. 2004;8:150–4. Launer LJ, Feskens EJ, Kalmijn S, Kromhout D. Am J Epidemiol. 1996;143:219–27. Leibovici D, Ritchie K, Ledesert B, Touchon J. Int J Epidemiol. 1999;28:77–81. Lemeshow S, Letenneur L, Dartigues JF, Lafont S, Orgogozo JM, Commenges D. Am J Epidemiol. 1998;148:298–306. Lindeman RD, Wayne SJ, Baumgartner RN, Garry PJ. J Gerontol A Biol Sci Med Sci. 2005;60:1065–70. Lindsay J, Laurin D, Verreault R, Hébert R, Helliwell B, Hill GB, McDowell I. Am J Epidemiol. 2002;156:445–53. Luchsinger JA, Tang MX, Siddiqui M, Shea S, Mayeux R. J Am Geriatr Soc. 2004;52:540–6. McCallum J, Simons LA, Simons J, Friedlander Y. Ann N. Y. Acad Sci. 2007;1114:121–9. McGuire LC, Ajani UA, Ford ES. Ann Epidemiol. 2007;17:93–9. Mehlig K, Skoog I, Guo X, Schütze M, Gustafson D, Waern M, Ostling S, Björkelund C, Lissner L. Am J Epidemiol. 2008;167:684–91. Miners JS, Baig S, Palmer J, Palmer LE, Kehoe PG, Love S. Brain Pathol. 2008;18:240–52. Mukamal KJ, Kuller LH, Fitzpatrick AL, Longstreth WT Jr, Mittleman MA, Siscovick DS. J.A.M.A. 2003;289:1405–13. Mukamal KJ, Chung H, Jenny NS, Kuller LH, Longstreth WT Jr, Mittleman MA, Burke GL, Cushman M, Beauchamp NJ Jr, Siscovick DS. Stroke. 2005a;36:1830–4. Mukamal KJ, Ascherio A, Mittleman MA, Conigrave KM, Camargo CA Jr, Kawachi I, Stampfer MJ, Willett WC, Rimm EB. Ann Intern Med. 2005b;142:11–19. Naimi TS, Brown DW, Brewer RD, Giles WH, Mensah G, Serdula MK, Mokdad AH, Hungerford DW, Lando J, Naimi S, Stroup DF. Am J Prev Med. 2005;28:369–73. Ngandu T, Helkala EL, Soininen H, Winblad B, Tuomilehto J, Nissinen A, Kivipelto M. Dement Geriatr Cogn Disord. 2007;23:140–9. Orgogozo JM, Dartigues JF, Lafont S, Letenneur L, Commenges D, Salamon R, Renaud S, Breteler MB. (Paris) 1997;153:185–92. Palomaki H, Kaste M. Stroke. 1993;24:1828–32. Panza F, Solfrizzi V, Torres F, Mastroianni F, Del Parigi A, Colacicco AM, Basile AM, Capurso C, Noya R, Capurso A. Neurosci Lett. 1999;277:53–6. Panza F, D’Introno A, Colacicco AM, Capurso C, Del Parigi A, Caselli RJ, Pilotto A, Argentieri G, Scapicchio PL, Scafato E, Capurso A, Solfrizzi V. Am J Geriatr Psychiatry. 2005;13:633–44. Panza F, Capurso C, D’Introno A, Colacicco AM, Frisardi V, Lorusso M, Santamato A, Seripa D, Pilotto A, Scafato E, Vendemiale G, Capurso A, Solfrizzi V. J Alzheimers Dis. 2009a;17:7–31. Panza F, Capurso C, Solfrizzi V. J.A.M.A. 2009b;299:2853–4. Panza F, Frisardi V, Capurso C, D’Introno A, Colacicco AM, Imbimbo BP, Santamato A, Vendemiale G, Seripa D, Pilotto A, Capurso A, Solfrizzi V. Am J Geriatr Psychiatry. 2010;18:98–116. Peters R, Peters J, Warner J, Beckett N, Bulpitt C. Age Ageing. 2008;37:505–12. Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Arch Neurol. 1999;56:303–8. Portet F, Ousset PJ, Visser PJ, Frisoni GB, Nobili F, Scheltens P, Vellas B, Touchon J; MCI Working Group of the European Consortium on Alzheimer’s Disease (EADC). J Neurol Neurosurg Psychiatry. 2006;77:714–8. Qiu C, De Ronchi D, Fratiglioni L. Curr Opin Psychiatry. 2007;20:380–5. Reid MC, Van Ness PH, Hawkins KA, Towle V, Concato J, Guo Z. J Geriatr Psychiatry Neurol. 2006;19:98–105. Reynolds K, Lewis LB, Nolen JD, Kinney GL, Sathya B, He J. J.A.M.A. 2003;289:579–88. Roman GC. Acta Neurol Scand. 2002;Suppl. 178:6–9. Roman GC, Royall DR. Alzheimer Dis Assoc Disord. 1999;13 (Suppl. 3):S69–80.
3044
F. Panza et al.
Román GC, Sachdev P, Royall DR, Bullock RA, Orgogozo JM, López-Pousa S, Arizaga R, Wallin A. J Neurol Sci. 2004;226:81–7. Ruitenberg A, van Swieten JC, Witteman JC, Mehta KM, van Duijn CM, Hofman A, Breteler MM. Lancet. 2002;359:281–6. Sachdev P. Int J Geriatr Psychiatry. 1999;14:402–3. Scarmeas N, Stern Y, Tang MX, Mayeux R, Luchsinger JA. Ann Neurol. 2006;59:912–21. Schinka JA, Vanderploeg RD, Rogish M, Graves AB, Mortimer JA, Ordoric PI. J Int Neuropsychol Soc. 2002;8:811–8. Simons LA, Simons J, McCallum J, Friedlander Y. Med J Aust. 2006;184:68–70. Solfrizzi V, D’Introno A, Colacicco AM, Capurso C, Del Parigi A, Baldassarre G, Scapicchio P, Scafato E, Amodio M, Capurso A, Panza F, Italian Longitudinal Study on Aging Working Group. Neurology. 2007;68:1790–89. Stampfer MJ, Kang JH, Chen J, Cherry R, Grodstein F. N Engl J Med. 2005;352:245–53. Stancampiano R, Carta M, Cocco S, Curreli R, Rossetti ZL, Fadda F. Brain Res. 2004;997:128–32. Tyas SL, Koval JJ, Pederson LL. Stat Med. 2000;19:1685–96. Truelsen T, Thudium D, Grønbæk M. Neurology. 2002;59: 1313–9. U.S. Department of Health and Human Services and U.S. Department of Agriculture. Dietary guidelines for Americans, 6th ed. Washington, DC: U.S. Government Printing Office; 2005. Verhaegen P, Borchelt M, Smith J. Health Psychol. 2003;22:559–69. Waddington E, Puddey IB, Croft KD. Am J Clin Nutr. 2004;79:54–61. Walzem RL. Inflammopharmacology. 2008;16:265–71. Waterhouse AL. Ann N. Y. Acad Sci. 2002;957:21–36. Whitehouse PJ, Sciulli CG, Mason RM. Psychopharmacol Bull. 1997;33:129–33. Winblad B, Palmer K, Kivipelto M, Jelic V, Fratiglioni L, Wahlund LO, Nordberg A, Bäckman L, Albert M, Almkvist O, Arai H, Basun H, Blennow K, de Leon M, DeCarli C, Erkinjuntti T, Giacobini E, Graff C, Hardy J, Jack C, Jorm A, Ritchie K, van Duijn C, Visser P, Petersen RC. J Intern Med. 2004;256:240–6. Yoshitake T, Kiyohara Y, Kato I, Ohmura T, Iwamoto H, Nakayama K, Ohmori S, Nomiyama K, Kawano H, Ueda K, Sueishi K, Tsuneyoshi M, Fujishima M. Neurology. 1995;45:1161–8. Zuccalà G, Onder G, Pedone C, Cesari M, Landi F, Bernabei R, Cocchi A; Gruppo Italiano di Farmacoepidemiologia nell’Anziano Investigators. Alcohol Clin Exp Res. 2001;25:1743–8. Zhou H, Deng J, Li J, Wang Y, Zhang M, He H. Age Ageing. 2003;32:205–10.
Chapter 189
Sweet Preference and Mood: Implications for the Risk of Alcoholism Alexei B. Kampov-Polevoy
Abbreviations SD SL GABA FH+ FH– PMS
Sweet disliking Sweet liking g-Aminobutyric acid Positive family history of alcoholism Negative family history of alcoholism Premenstrual syndrome
189.1 Introduction We all like sweets from the moment we are born. Mother Nature designed us this way because sweet taste is an indicator of highly caloric carbohydrate foods that are necessary for us to survive at early age. Mother’s milk is very sweet, if anyone remembers. We perceive sweet taste as calming, soothing, and comforting. Reaction to the sweet taste is our basic pleasure, the yardstick by which we measure all our positive experiences. That is why even in our language the word “sweet” is synonymous with anything good. We call our loved ones sweethearts and sweeties, our favorite music is sweet music, and, even in business, we are trying to make sweet deals with our partners. Our response to the sweet taste is closely associated with our overall ability to experience all natural pleasures. Therefore, any changes in our hedonic response to sweets may reflect changes in functional activity of the brain reward system. For example, preference for stronger sweet taste and craving for sweets was consistently reported in individuals suffering from anxiety, depression, and dysphoria. Furthermore, our knowledge of the soothing effect associated with eating sweet foods that we acquired in infancy can provide us a serious disservice in our adult life: some of us continue to use sweet foods as a tranquilizer which, in some cases, leads to the development of binge eating behavior and obesity. The key role in regulation of our pleasurable response to sweet taste is played by endogenous opioid peptides (e.g., endorphins and enkephalins) that have similar physiological effects to exogenous opiate drugs such as morphine and heroine. These endogenous peptides as well as their multiple
A.B. Kampov-Polevoy (*) Department of Psychiatry, University of North Carolina at Chapel Hill, 237 Medical School Wing B, Campus Box 7160, Chapel Hill, NC 27599-7160 e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_189, © Springer Science+Business Media, LLC 2011
3045
3046
A.B. Kampov-Polevoy
receptor subtypes (e.g., m, b, k) are found in various networks throughout the brain, but particularly within the regions involved in emotional regulation, responses to pain and stress, endocrine regulation, and food intake. Despite the importance of sweet taste and its association with emotional health, the literature about this subject is scarce. In this chapter I will attempt to systematize the existing knowledge regarding physiological mechanisms underpinning hedonic response to sweets and the links between distortions of sweet taste and various psychiatric conditions with an emphasis on alcoholism.
189.2 Hedonic Response to Sweet Taste The gustatory system is primarily devoted to a quality check of food, while simultaneously detecting nutrients and avoiding toxic substances. Sweet is one of the five primary taste qualities (along with bitter, sour, salty, and umami). Sweet-tasting foods and beverages are highly preferred by plant-eating animals including humans, presumably because sweetness reflects the presence of caloric sugars. Liking sweet tastes is strongly influenced by inborn (innate) factors and positive hedonic (pleasurable) responses to sweet taste can be detected in human infants during the first minutes after birth (Steiner et al. 2001). Hedonic response to sweet taste is believed to be a stable heritable trait that reflects qualitative differences in taste perception rather than more cognitive reactions based on associations between the taste and attitudes toward, or social acceptability of, liking sweet foods (Looy and Weingarten 1991). The role of genetic factors in perception of sweet taste and preference for sweet foods can be illustrated by the results of a recent study of 473 twin pairs, which showed that approximately half of the variation in liking for sweet solution and liking and use-frequency of sweet foods (49–53%) was explained by genetic factors, whereas the rest of the variation was due to the environmental factors unique to each twin individual (Keskitalo et al. 2007).
189.2.1 Sweet-liking/Sweet-disliking Phenotypes It is important to note that although all humans have an innate preference for the sweet taste, the magnitude of hedonic response to sweet taste as well as avidity to consume sweet foods significantly varies across individuals. Such differences may be noted in infancy (Steiner et al. 2001) and persist as children become young adults (Desor and Beauchamp 1987). Two major stable patterns of hedonic response to sweet taste have been described in the literature (Thompson et al. 1976). With Type I response (often referred to as sweet disliking), subjects prefer progressively higher concentrations of sucrose solution up to the middle range of concentration, followed by a breakpoint after which preference declines with increased concentration. The Type II response (also referred to as sweet liking) is characterized by continuous increase with eventual leveling off in liking of progressively more concentrated sucrose solutions (Fig. 189.1). Similar patterns have been described in animals as well (Sinclair et al. 1992), which may indicate that these patterns can be a characteristic of all mammals.
189.2.2 Factors Influencing Hedonic Response to Sweet Taste Although perception of sweet taste appears to be a fairly stable trait, there are some factors that are known to be associated with variability of hedonic response to the sweet taste.
189 Sweet Preference and Mood: Implications for the Risk of Alcoholism
3047
Fig. 189.1 Types of hedonic response to sweet taste in animals and humans. Although both animals and humans have an innate preference for sweet-tasting substances, the magnitude of hedonic response to sweet taste as well as avidity to consume sweet foods significantly varies across individuals. Two major stable patterns of hedonic response to sweet taste have been described in the literature. With Type I response (often referred to as sweet disliking, SD), subjects prefer progressively higher concentrations of sucrose solution up to the middle range of concentration, followed by a breakpoint after which preference declines with increased concentration. The Type II response (also referred to as sweet liking, SL) is characterized by continuous increase with eventual leveling off in liking of progressively more concentrated sucrose solutions. These two patterns of hedonic response to sweet taste are believed to be associated with activity of the brain opiate system that to a large extent is regulated by genetic factors. Sweet liking is believed to be associated with suppression of the opioidergic transmission in the brain
Ethnic background: Ethnic origins have been shown to have an impact on the hedonic response to sweet taste. For example, Americans of African descent were shown to prefer stronger sweet taste than Americans of European descent (Bacon et al. 1994). Age: The degree of preferred sweetness also may vary with age. Children tend to prefer sweeter solutions compared to adults (De Graaf and Zandstra 1999). Younger people also consume more sugar than do older people (Drewnowski 1997). Gender-related factors: Sex differences may also influence hedonic response to sweet taste, with women being more sensitive to the mood-altering effect associated with eating sweet-tasting foods and reporting stronger cravings for sweet-tasting foods (Kampov-Polevoy et al. 2006). It is of interest that the sex differences in hedonic response to and craving for sweets can be seen only in adult individuals, while no such difference was noted in infants (Beauchamp and Moran 1982). These findings may indicate that sex hormones may play a significant role in the development of sweet preferences as was initially suggested by Wade and Zucker (1969). The other indications that sex hormones can modulate a perception of and craving for the sweet taste come from studies that showed a fluctuation of sweet cravings and consumption of sweet foods (but not other types of foods) throughout the menstrual cycle (Bowen and Grunberg 1990). In summary, a pleasurable response to the sweet taste is an innate reaction and the level of positive emotional response to sweet taste, as well as consumption of sweets, are influenced to a large degree by genetic factors. There are two major patterns of the hedonic response to sweet taste – sweet disliking and sweet liking. Of the factors identified to date that influence hedonic response to sweet taste, gender seems the most influential with women being more sensitive to moodaltering effect of sweet-tasting foods and more likely to crave and eat sweets in response to emotional stress.
3048
A.B. Kampov-Polevoy
189.3 1 89.3 Physiological Mechanisms Underpinning Hedonic Response to Sweet Taste The mechanisms regulating hedonic response to sweet taste are complex, and detailed discussions of these complexities are beyond the scope of this chapter. Here, a brief overview of the two major categories of mechanisms of response to sweet taste is presented: the peripheral mechanisms, responsible for the perception of sweet taste, and the central mechanisms, responsible for emotional response for sweets.
189.3.1 Peripheral Mechanisms/Role of the Taste Receptors The initial event of perception of sweet taste occurs in taste receptor cells, which are clustered in the taste buds in taste papillae on the tongue surface. The perception of sweetness intensity is related to the number of papillae that varies widely among people. These variations may be due to alleles in genes that develop and maintain sensory cells. To date, three sweet-taste receptors have been identified in both mice and humans, which are regulated by polymorphisms of sweet-receptor genes (TAS1R1, TAS1R2, TAS1R3). Polymorphisms of these genes have a large effect on the intake of sweet-tasting substances in both animals and humans (Liao and Schultz 2003).
189.3.2 Central Mechanisms/Role of the Brain Reward System Hedonic responses to sweet taste may be modulated not only by peripheral mechanisms, but also by the brain reward system that regulates perception of all naturally occurring pleasures. This system was first described by Olds and Milner (1954) who observed that when electrodes were placed in certain areas of the brain, rats would actively self-stimulate these areas often to the exclusion of other activities including eating. The brain reward system is located in the mesolimbic part of the brain and involves complex interrelationships of at least four important neurochemical pathways (i.e., serotonergic, opioidergic, GABAergic, and dopaminergic), with a pleasurable response associated with increased dopamine accumulation in the nucleus accumbens. Without the normal functionality of the brain reward system, an individual will experience a so called reward-deficiency syndrome that includes hypohedonia (diminished ability to experience pleasure) and inability to cope with stress. Individuals with the reward-deficiency syndrome are likely to seek substances and/or behaviors that will overcome this hypohedonic state by activating the brain reward system. These substances and behaviors include alcohol, opiates, psychostimulants, nicotine, carbohydrates, cannabinoids, gambling, sex, and indulgence in any excessive pleasure or thrill-seeking behaviors, such as video gaming, etc. (for review, see Comings and Blum 2000). Sweet-tasting substances (both caloric and noncaloric sweeteners) stimulate sweet-taste receptors on the surface of the tongue, which, in turn, results in activation of the µ-opioid receptors on GABA interneurons (projection neurons that interact with dopamine neurons) in the ventral tegmental area, to cause a disinhibition of dopamine cell firing in the nucleus accumbens that is subjectively perceived as pleasure (Lemon et al. 2004). Microdialysis studies show a robust rise in extracellular accumbens dopamine levels following exposure to sweet tastes. Recently, Pecina and Berridge (2005) have shown that the perception of pleasantness of sweet taste in rats is strongly influenced by
189 Sweet Preference and Mood: Implications for the Risk of Alcoholism
3049
µ-opioid receptor activation in a single cubic millimeter site localized in the rostrodorsal quarter of the medial shell of the nucleus accumbens, an observation indicating very tight neuroanatomical localization of the hedonic response to sweet taste. According to DiChiara and North (1992), opioid pathways in the brain reward system are associated with feelings of satiety, sedation, rest, and “bliss.” In this regard, it is important to note that sweet preference and hedonic response to sweet taste seem to be associated with activity of the brain opioid pathways. Animal studies showed that stimulation of these pathways with morphine shifts the sweet preference/aversion curve left toward preference for weaker sweet taste (Calcagnetti and Reid 1983), while its blockade shifts sweet preference toward more concentrated sweet solutions (Leventhal et al. 1995). Therefore, the sweet-liking phenotype, described earlier in this chapter may be associated with inhibition of the opioidergic pathways in the brain reward system and can be a potential marker of the reward-deficiency syndrome. Knowledge of physiological mechanisms underpinning hedonic response to sweet taste leads to several hypotheses that may have important practical implications: 1. Considering that eating of sweet-tasting foods is associated with stimulation of the opioidergic pathways in the brain, one would expect that consumption of sweets will exert an effect similar to those of opiate agonists (e.g., morphine) – produce a positive affective state, reduce emotional stress, and cause analgesia. On the other hand, chronic consumption of sweet-tasting foods may lead to the development of elements of physical dependence such as increased consumption of sweets, loss of control over eating sweets, as well as development of symptoms similar to symptoms of opiate withdrawal when consumption of sweet-tasting foods is discontinued. 2. Considering the ability of sweet-tasting foods to produce positive emotional states and reduce emotional stress, one would also expect that people will eat sweets to self-medicate the negative mood states such as anxiety, fatigue, and/or depression. 3. Considering that preference for stronger sweet taste (sweet liking) is associated with blockade of opioidergic pathways that play an important role in regulation of the activity of the brain reward system, it is logical to hypothesize that the sweet-liking phenotype may serve as a marker for the reward-deficiency syndrome. If this hypothesis is correct, we would expect the sweet-liking phenotype to be associated with the negative affect and with the inclination to self-medicate this affective state with sweet-tasting foods, alcohol, and drugs of addiction. The following sections seek support for the above-mentioned hypotheses in the available literature.
189.4 Effects of Ingestion of Sweet Foods The effects of ingestion of sweet tasting food are summarized in Table 189.1.
Table 189.1 Effects associated with eating of sweet-tasting foods Effects of acute intake Effects of chronic intake Feeling of well-being, sedation, and relaxation Development of symptoms that are similar to symptoms of physical dependence: Reduction of emotional stress and alleviation of negative mood states 1. Enhanced intake of sweet-tasting foods 2. Abrupt discontinuation of eating sweets after a period of daily intake of sweet-tasting foods may lead to symptoms similar to symptoms of opiate withdrawal Analgesia Development of tolerance to analgesic effect of opiate drugs
3050
A.B. Kampov-Polevoy
189.4.1 Effect of Acute Ingestion of Sweet-Tasting Substances 189.4.1.1 Effect of Sweet Taste on Mood As stated earlier, positive emotional reactions to sweet taste can be detected in humans within the first few minutes after birth. Intraoral sucrose administration has been shown to induce rapid and sustained calming effects in crying newborns (for review, see Bhattachrjee and Mathur 2005). Similar mood-altering effects of sweet-tasting foods have been noted in adults as well. For example, eating of sweet food improves an experimentally induced negative mood state in healthy adults (Macht and Mueller 2007), with the effect being more pronounced in women than in men. Furthermore, college students reported that the magnitude of their craving for sweet carbohydrate foods correlates with the severity of the negative mood states such as anxiety, fatigue, and/or depression (Christensen and Pettijohn 2001). Therefore, positive mood effects of sweet-tasting foods may contribute to the habit of eating sweets as a coping mechanism against stress.
189.4.1.2 Effect of Sweet Taste on Perception of Pain There is a large body of evidence indicating that consumption of sweet-tasting foods may produce significant analgesia in both animals and humans. For example, acute exposure to sweet (e.g., sucrose) solutions was shown to produce analgesia sufficient to conduct minor surgical procedures in human infants. A similar effect was demonstrated in prepubertal children as well as in adult females but not in adult males. The brain opioidergic pathways seem to play an important role in sweet-induced analgesia because such analgesia may be reversed by the opiate antagonist naloxone (for review, see Bhattachrjee and Mathur 2005).
189.4.2 E ffect of Chronic Ingestion of Sweet-tasting Foods: Do We Become Dependent on Sweetness? Although acute consumption of sweet foods may cause marked opiate-mediated analgesia, chronic exposure to sweets attenuates morphine-induced analgesia, which indicates the development of tolerance to opiates. Furthermore, animal studies demonstrate that chronic exposure to sweet-tasting foods may lead to the development of symptoms similar to those seen in individuals with substance dependence. For example, rats maintained on a diet of chronic intermittent access to a sucrose solution and chow tend to increase their sugar intake (Avena et al. 2006) and show behavioral and neurochemical changes similar, albeit smaller in magnitude, to rats dependent on opiates (Rada et al. 2005; Colantuoni et al. 2002). There is emerging evidence showing that chronic exposure to sweet foods may cause long-term changes in preference for and consumption of sweet foods in humans, at least in children. For example, infants fed with sugar solutions prefer stronger sweet taste and like sweet foods later in life more than children who have not been fed with sweet water as infants (Beauchamp and Moran 1982). Children who were routinely fed sugar water during infancy preferred significantly higher levels of sucrose when compared to those who were rarely or minimally exposed (Pepino and Mennella 2005). In adults, immediate but short-lived reduction in sweet preference can be noticed after ingestion of the sweet solution (Cabanac and Duclaux 1970). However, this effect was noted only in individuals with sweet-disliking phenotype but not in sweet-liking individuals (Looy et al. 1992; Looy and Weingarten 1992).
189 Sweet Preference and Mood: Implications for the Risk of Alcoholism
3051
189.4.3 Acute Effect of Noncaloric Sweeteners It is important to emphasize that the effects sweet foods have on the brain reward system are associated with the oral stimulations of sweet-taste receptors rather than with the chemical composition or caloric value of ingested food. This hypothesis is supported by the elegant animal experiments using a sham-feeding procedure, when the food, after being eaten, did not enter the stomach but rather was removed from the gastrointestinal tract, thus eliminating any postabsorptive effect of consumed foods (Hajnal et al. 2004). This study provided a clear indication that concentration-dependent increase in dopamine levels in the rat nucleus accumbens depends on the taste but not on the composition of the studied foods. These findings may also explain why the noncaloric sweetener saccharine increases beta-endorphin levels in rat cerebrospinal fluid and plasma of the same magnitude as natural sugars (Yamamoto et al. 2000). In summary, oral sucrose administration has been shown to produce positive emotional responses, to alleviate negative mood states, and to produce marked analgesia. These phenomena have been reported in both animals and humans, with women being more sensitive to these effects than men. Such effects are more likely to be associated with the sweet taste than with the chemical composition or caloric value of ingested food. Chronic consumption of sweet-tasting substances may lead to a condition similar to physical dependence to opiates with symptoms including elevated sugar consumption, increased tolerance to morphine, and signs of opiate withdrawal.
189.5 Sweet Taste and Mood Disorders The literature provides consistent evidence showing that negative mood states are associated with the preference for stronger sweet taste, craving for sweet foods, and excessive consumption of sweettasting foods. For example, a study of healthy children shows significant correlation between sweet preference and depressive symptoms (Mennella et al. 2008). It is also believed that some people use sweet-tasting foods to self-medicate their negative emotional state. Healthy individuals: In the study of healthy college students, Christensen and Pettijohn (2001) have shown that 90% of females and 55% of males identified themselves as cravers for sweet carbohydrate foods. In this sample, the magnitude of carbohydrate cravings significantly correlated with the experience of negative mood states such as anxiety, fatigue, and/or depression. Patients with depression: In a clinical population, depressed individuals were shown to have a greater craving for and consumption of sweet carbohydrate foods than non-depressed controls (Christensen and Somers 1996; Wurtman 1988). Furthermore, the magnitude of the preference for sweet foods in depressed patients positively correlated with the level of depression (Fernstrom et al. 1987). The association between depressed mood and sweet craving/consumption is most evident in patients with seasonal affective disorder, a condition characterized by recurrent depressive episodes that typically begin to appear in the fall and remit in the spring. These patients report strong cravings for sweet foods that are consistently associated with the depressive episodes (Rosenthal et al. 1984). Moreover, patients with seasonal affective disorder report that indulging in their craving significantly reduces negative mood (Leibenluft et al. 1993). Patients with premenstrual syndrome (PMS): The other illustration of a link between negative affect and craving for sweets is PMS. It is commonly known that food cravings and mood tend to fluctuate throughout the menstrual cycle. In comparison to other phases of the menstrual cycle, during the luteal phase (10 days preceding menses), women report increased craving and consumption of sweet carbohydrate foods, which can double in comparison with postmenstrual period (Bowen
3052
A.B. Kampov-Polevoy
and Grunberg 1990). This phenomenon is especially pronounced in women with PMS when the depressed mood positively correlates with craving for sweet foods (Both-Orthman et al. 1988). Women also reported an improvement in mood following consumption of the carbohydrate-rich meal (Wurtman et al. 1989). In summary, sweet preference and craving for sweet-tasting foods positively correlates with severity of the negative mood states (e.g., anxiety, depression) in both healthy individuals and patients with clinical depression. Consumption of sweets alleviates negative mood states and can be used for self-medication.
189.6 Hedonic Response to Sweet Taste and Excessive Alcohol Intake 189.6.1 Animal Studies 189.6.1.1 A ssociation Between Consumption of Sweet-tasting Solutions and Alcohol Intake During the last three decades, evidence indicating a close positive association between hedonic response to sweet taste, avidity to consume sweet-tasting foods, and excessive alcohol intake in both animals and humans has accumulated. The first evidence connecting consumption of sweets with alcohol intake came from animal studies. In 1978, Ramirez and Sprott reported that C57BL mice, known for their high voluntary alcohol intake, consume much larger quantities of saccharin solution than do DBA/2J mice that are known for their relatively low alcohol intake. A similar association (correlation coefficient = 0.7) between voluntary consumption of a 0.1% saccharin solution and subsequent voluntary consumption of 15% alcohol solution was demonstrated in randomly bred Wistar rats (for review, see Kampov-Polevoy et al. 1999). Later studies showed that association between consumption of sweet solutions and alcohol intake may have a genetic origin. For example, Belknap et al. (1993) showed a high genetic correlation (correlation coefficient = 0.77) between saccharin and alcohol intake in 15 inbred mouse strains. A similar correlation between sucrose consumption and alcohol intake can be seen in the F2 generation of crosses between alcohol-preferring C57BL/6ByJ and alcohol-avoiding 129/J strains of mice. Further analysis showed that intakes of sucrose and ethanol are influenced by a few genes and that the “genetically determined component of these correlations was stronger than the component related to environmental factors” (Bachmanov et al. 1996). Similar results have been demonstrated in rats genetically selected for their high/low alcohol intake (Overstreet et al. 1997).
189.6.1.2 Saccharin-induced Polydipsia and Excessive Alcohol Intake One of the interesting characteristics of intake of sweetened solutions by alcohol-preferring rats is their tendency to consume these solutions far beyond the limit of normal daily fluid intake. This tendency was initially reported in randomly bred rats (Kampov-Polevoy et al. 1990), when it was noted that, although all studied rats had a high preference for 0.1% saccharin solution, some of them (40%) consumed saccharin solution on average 32% over the limit of their normal daily fluid intake, a condition we described as saccharin-induced polydipsia. Subsequently, these polydipsic rats consumed almost 10 times as much alcohol during the first week of an alcohol/water choice experiment as rats that consumed saccharin solution within the limits of their normal daily fluid intake.
189 Sweet Preference and Mood: Implications for the Risk of Alcoholism
3053
Fig. 189.2 Association between consumption of sweet-tasting substances and alcohol intake in rats. The figure illustrates the results of the study of consumption of sweet 0.1% saccharine solution in a free choice with water by rats genetically selected for excessive voluntary alcohol consumption and rats genetically selected to reject alcohol. The diagram shows that both strains of rats have similar high preference for saccharine solution that exceeds 90%. However, alcohol-nonpreferring rats, that are known to be sweet dislikers, consume sweet solution within the limits of normal daily fluid intake, when alcohol preferring rats who are known to be sweet likers, may double and in some cases quadruple their DFI when sweet solution is available. These findings are consistent with the results of clinical studies showing that sweet-liking individuals have an elevated risk to both alcohol-use disorders and bulimia nervosa that manifests itself by loss of control over food intake. Furthermore, comorbidity between substance-use disorders and bulimia is reported to be as high as 50%
Saccharine-induced polydipsia is especially apparent in rats genetically selected for alcohol preference. For example, high alcohol drinking rats exhibited a 370% increase in daily fluid intake when 0.1% saccharin solution was available along with water, whereas low-alcohol-drinking animals consumed saccharin within their normal daily fluid intake (Overstreet et al. 1997) (Fig. 189.2). Saccharin-induced polydipsia was shown to be a reliable predictor of subsequent alcohol intake, with a correlation coefficient of up to 0.9 between these two variables in P rats (Kampov-Polevoy et al. 1995), which makes this trait one of the best predictors of alcohol intake in rodents. The dramatic increase in daily fluid intake in the presence of saccharin exhibited by alcoholpreferring rats may be an animal analogue of the clinical phenomenon known as loss of control. In the clinical situation, loss of control refers to the behavior in which a rewarding substance is taken in larger amounts or over longer periods of time than is intended. Interestingly, the link between loss of control over the consumption of sweets and the excessive alcohol intake was noted in humans as well. For example, overall comorbidity between eating disorders (e.g., bulimia nervosa that manifests itself by loss of control over food intake in general and sweet foods in particular) and alcohol abuse/dependence was reported to be as high as 50% (Dansky et al. 2000).
189.6.1.3 Preference for the Strong Sweet Taste and Alcohol Intake In 1992, Sinclair et al. reported that rats with a genetically determined predisposition to high alcohol consumption preferred more concentrated sweet solutions compared to alcohol-avoiding rats. In this study, alcohol-preferring rats, as well as alcohol-nonpreferring rats, were tested by being given a free choice between tap water and an ascending series of saccharin concentrations, starting at 2 mg/L and doubling the concentration every day until a final level of 4096 mg/L was reached.
3054
A.B. Kampov-Polevoy
This experiment showed that, when rats were exposed to up to the 64 mg/L saccharin solution, their preference for it was generally the same in all animal groups. However, when exposed to more concentrated saccharin solutions, alcohol-nonpreferring rats generally had a lower preference ratio, compared to alcohol-preferring rats. The relevance of these findings to human alcoholism was later tested by evaluating the sweet preferences of alcoholics and nonalcoholic control subjects, as described in the following section.
189.6.2 Human Studies 189.6.2.1 Association Between the Sweet-liking Phenotype and Alcohol Intake There is growing evidence linking the sweet-liking phenotype (preference for stronger sweet taste) and genetic risk for alcoholism. Our own study of healthy college students (Kampov-Polevoy et al. 2003a) indicates that sweet liking is more prevalent in children of alcoholic parents who are known to have four to nine times greater risk of becoming an alcoholic than children of nonalcoholic parents. Similar findings have been reported in the samples of alcoholics, addicts, and control (nonaddicted) subjects (Kampov-Polevoy et al. 2003b). Thus, these and similar studies support the hypothesis that the sweet-liking phenotype is associated with the genetic risk of alcoholism. However, unlike animal studies, in humans, the sweet-liking phenotype by itself seems to be insufficient to predict a diagnosis of alcohol dependence in clinical samples. As can be seen in the
Fig. 189.3 Preference for five sucrose concentrations in male alcoholics with (FH + ) and without (FH-) family history of alcoholism. This diagram presents a distribution of preferred sugar solutions in the sample of male patients with alcohol dependence (n = 48). The diagram shows approximately equal representation of sweet likers (patients preferring highest offered [0.83 M] sugar solution) and sweet dislikers (patients preferring lower sucrose solutions) in the sample. Sweet liking seems to be associated with the genetic risk of alcoholism and, therefore, is more prevalent among the children of alcoholic parents (n = 20) than among children of nonalcoholics (n = 28). Therefore, the sweetliking phenotype may be considered as a potential marker for the familial form of alcoholism
189 Sweet Preference and Mood: Implications for the Risk of Alcoholism
3055
Fig. 189.3, in a sample of alcoholic patients there is a similar representation of sweet-liking and sweet-disliking individuals, which indicates that sweet-liking status is linked not to diagnosis of alcoholism but to genetic risk for this disease evaluated on the basis of family history. So now, the question is whether the sweet-liking phenotype has any practical value for prediction of the actual risk for having alcohol-use disorders in any given individual. The studies conducted during the last decade show that such prediction can be made if sweet liking is combined with another factor – personality trait called novelty seeking. Novelty seeking has been consistently associated with deviance proneness, disregard of social norms as well as with early onset of alcohol drinking and excessive alcohol drinking (Finn et al. 2000). In their recent report, Grucza et al. (2006) demonstrated that novelty seeking modulates the level of the alcoholism risk in high-risk families with high novelty seeking magnifying this risk and low novelty seeking acting as a protective factor. On the other hand, in the low-risk families, there was no effect of novelty seeking on the prevalence of alcohol-use disorders. These findings are consistent with our own data indicating that, in clinical populations, the estimated odds of being an alcoholic, on average, increase by 11% as NS score increases by 1 point in sweet-liking but not in sweet-disliking subjects (Kampov-Polevoy et al. 2004) (Fig. 189.4). Combination of preferred sucrose concentration and novelty-seeking score was shown to predict alcoholic vs. nonalcoholic group status at 65% sensitivity and 94% specificity, with correct classification in 85% of subjects (Kampov-Polevoy et al. 1998). Our most recent study (unpublished data) of 158 healthy college students demonstrated that individuals with a sweet-liking phenotype and high novelty seeking had a 27-fold increase in likelihood for having alcohol-related problems compared to sweet-disliking individuals with low novelty seeking. In summary, there is growing evidence showing the hedonic response to sweet taste and the avidity to consume sweet-tasting foods is closely associated with the level of alcohol consumption in both animals and humans. This phenomenon, which, to a large extent, is based on genetic mechanisms,
Fig. 189.4 Interaction between sweet-liking/sweet-disliking phenotypes and novelty seeking in prediction of alcoholic status of an individual. Both the sweet-liking and high novelty-seeking phenotypes are frequent findings in patients with alcohol-use disorders. However, neither of these traits separately can accurately predict an individual’s risk for alcoholism. Sweet liking is associated with elevated sensitivity to the rewarding effect of alcohol and family history of alcoholism, but sweet liking alone has not been linked to excessive drinking. On the other hand, novelty seeking was shown to be closely associated with early onset of drinking, frequency and severity of drinking episodes, and disregard of societal norms. The diagram illustrates interaction of two independent factors, novelty seeking and sweet-liking phenotype, in the prediction of the alcoholism risk. One can see a dramatic increase of the risk of being an alcoholic in sweet likers with high novelty-seeking scores. At the same time in sweet-disliking individuals who are less sensitive to the rewarding effect of alcohol, excessive alcohol drinking associated with high novelty seeking does not lead to the development of alcohol-use disorders
3056
A.B. Kampov-Polevoy
was first described in laboratory animals where consumption of sweet saccharin solution can be considered as the most accurate predictor of the voluntary alcohol consumption of any given animal. In humans, the relation between response to sweet taste and alcohol intake is more complex. While genetic risk of alcoholism estimated on the basis of family history of alcoholism was shown to be associated with hedonic response to sweet taste, sweet response by itself is not sufficient to predict the risk of alcohol-related problems. Such prediction can only be made using a combination of hedonic response to sweet taste that reflects activity of the brain reward system and novelty seeking – a personality trait that is consistently associated with deviance proneness and excessive drinking.
189.7 Application to Other Areas of Health and Disease As discussed above, the pleasure associated with consumption of sweet-tasting foods is one of our fundamental pleasures. Stimulation of the sweet-taste receptors in the oral cavity leads to the release of b-endorphins in the brain that have an effect similar to those of opiate agonists, such as morphine, that cause analgesia as well as feelings of well-being, sedation, and “bliss.” Individuals with hypofunction of the brain opiate system have a diminished ability to experience pleasure, including pleasure associated with eating sweets. That leads to the preference for stronger sweet taste (i.e., sweet liking), which may be considered to be a marker of the activity of the brain opiate system. The link between hypofunction of the brain opiate system, diminished ability to experience pleasure, and perception of sweet taste may explain the positive association between distortions of the sweet taste and negative mood states such as depression and anxiety that has been noted in both clinical populations and healthy individuals. Considering that a mentioned dysfunction of the brain opiate system may at least partially be determined by genetic mechanisms (for review see Gianoulakis 2004), testing hedonic response to the sweet taste may be an important instrument for evaluation of genetic predisposition to these disorders although further investigations are needed. Furthermore, individuals with the hypofunction of the brain opiate system and reward-deficiency syndrome are likely to seek substances and/or behaviors that will overcome this hypohedonic state by activating the brain reward system. The most natural choice in this situation is the one we all learned in infancy – when in distress, we need to eat something sweet. At first it was sweet mother’s milk or sugar water that calmed the baby down. Later in life, a candy served the same purpose. Eating sweets leads to the instant release of b-endorphins in the brain that has an effect of the internal tranquilizer. As adults, women, who are more sensitive to the mood-altering effect of sweets (Kampov-Polevoy et al. 2006), are more likely to eat sweets to cope with negative emotions and stress, which may explain a higher prevalence of bulimia nervosa noted in women compared to men (Antczak and Brininger 2008). It should be also mentioned that women diagnosed with bulimia nervosa are likely to be sweet likers (Franko et al. 1994), which may indicate that sweet-liking status and the associated dysfunction of the brain opiate system may predispose women to this disorder. On the other hand, men, who are less sensitive to the positive mood-altering effect of sweets, may choose alcohol to cope with negative emotional states. This behavior is more likely to be observed in sweet-liking men who are more sensitive to the rewarding effect of alcohol. Furthermore, combination of sweet liking, indicative of sensitivity to the positive emotional effect of alcohol, with high novelty seeking that facilitates drinking behavior and disregard of societal norms, produces high risk for developing alcohol-related problems and alcohol-use disorders. A conceptual model that illustrates a link between hedonic response to sweet taste, affective disorders, alcohol-use disorders, and bulimia are presented in Fig. 189.5.
189 Sweet Preference and Mood: Implications for the Risk of Alcoholism
3057
Fig. 189.5 The following diagram may help to conceptualize material presented in this chapter. The brain opioidergic pathways play an important role in mediation of hedonic responses to sweet taste. Inhibition of these pathways caused by either genetic factors (e.g., in children of alcoholics) or by opioid receptor blockers (e.g., naloxone) shifts the preference toward stronger sweet taste. Therefore, preference for stronger sweet taste (sweet liking) may be an indicator of the dysfunction of the brain opioidergic pathways. These pathways play an important role in functioning of the brain reward system and their inhibition may lead to diminished ability to experience pleasures and elevated sensitivity to rewarding effects of sweet-tasting foods and alcohol. As a result, sweet-liking individuals, both animals and humans, are prone to use and abuse sweet foods and alcohol to self-medicate their negative emotional state and to cope with emotional stress. Sweet-liking women seem to be more sensitive to moodaltering effects of sweet foods and use these foods for coping with negative affect. This elevated sensitivity in combination with impaired control over eating sweets may explain higher prevalence of bulimia nervosa in women compared to men. Sweet-liking men, who are less sensitive to the mood-altering effect of sweet-tasting foods, are more likely to use alcohol for self-medication of negative affect. High novelty seeking that facilitates frequent episodes of excessive drinking significantly increases the risk of alcohol-use disorders in sweet-liking men. The high comorbidity between alcohol-use disorders and bulimia nervosa also supports the hypothesis that these disorders may have similar mechanisms
Summary • A pleasurable response to sweet taste is an innate reaction. The level of positive emotional response to sweet taste, as well as consumption of sweets, is influenced to a large degree by genetic factors. • There are two major patterns of the hedonic response to sweet taste – sweet disliking and sweet liking. • Oral sucrose administration produces positive emotional responses, alleviates negative mood states, and produces marked analgesia in both animals and humans, with women being more sensitive to these effects than men. Such effects are more likely to be associated with sweet taste than with chemical composition or caloric value of ingested food. • Chronic consumption of sweet-tasting substances may lead to a condition similar to physical dependence to opiates with symptoms including elevated sugar consumption, increased tolerance to morphine, and signs of opiate withdrawal when access to sweets is discontinued. • Sweet preference and craving for sweet-tasting foods positively correlates with severity of the negative mood states (e.g., anxiety, depression) in both healthy individuals and patients with clinical
3058
A.B. Kampov-Polevoy
depression. Consumption of sweets leads to alleviation of the negative mood states and can be used for self-medication. • Hedonic response to sweet taste and the avidity to consume sweet-tasting foods are closely associated with the level of alcohol consumption in both animals and humans. In animal experiments, consumption of sweet saccharin solution can be considered as the most accurate predictor of the voluntary alcohol consumption of any given animal. • In humans, the association between response to the sweet taste and alcohol intake is more complex than in animals. While the genetic risk of alcoholism estimated on the basis of family history of alcoholism was shown to be linked with hedonic response to sweet taste, sweet response by itself is not sufficient to predict the risk of alcohol-related problems. Such prediction can be made only using a combination of hedonic response to sweet taste that reflects activity of the brain reward system and novelty seeking – a personality trait that is consistently associated with excessive drinking. It is important to keep in mind that: • Sweet likers are more sensitive to mood-altering and analgesic effects of sweets than sweet dislikers. • Women are more sensitive to these effects than men. • Children are more sensitive to these effects than adults.
Terminology List Sweet disliking: Describes a pattern of sweet preference that is characterized by continuous increase, with eventual leveling off in liking of progressively more concentrated sweet solutions. Sweet liking: Describes a pattern of sweet preference that is characterized by preference to progressively higher concentrations of sucrose solution up to the middle range of concentration, followed by a breakpoint after which preference declines with increased concentration. GABA: g-Aminobutyric acid is the main inhibitory neurotransmitter in the mammalian central nervous system. Polydipsia: Excessive or abnormal fluid consumption. Hypohedonia: Diminished ability to experience pleasure.
References Antczak AJ, Brininger TL. (2008) Eat Disord. 2008;16:363–377. Avena NM, Rada P, Moise N, Hoebel BG. (2006) Neuroscience. 139:813–820. Bachmanov AA, Reed DR, Tordoff MG, Price RA, Beauchamp GK. (1996) Behav Genet. 26:563–573. Bacon A, Miles J, Schiffman SS. Physiol Behav. 1994;55:603–606. Beauchamp GK, Moran M. (1982). Appetite. 3:139–152. Belknap JK, Crabbe JC, Young ER. (1993) Psychopharmacology. 112:503–510. Bhattachrjee M, Mathur R. (2005) J Physiol Pharmacol. 49:383–394. Both-Orthman B, Rubinow DR, Hoban MC, Malley J, Grover GN. (1988) Am J Psychiatry. 145:628–631. Bowen D, Grunberg N. (1990) Physiol Behav. 47:287–291. Calcagnetti D, Reid L. (1983) Pharmacol Biochem Behav. 18:567–569. Colantuoni CRP, McCarthy J, Patten C, Avena NM, Chadeayne A, Hoebel BG. (2002) Obes Res. 10:478–488. Comings DE, Blum K. (2000) Prog Brain Res. 126:325–341. Review. Christensen L, Pettijohn L. (2001) Appetite. 36:137–145. Christensen L, Somers S. (1996) Int J Eat Disord. 20:105–109.
189 Sweet Preference and Mood: Implications for the Risk of Alcoholism
3059
Dansky BS, Brewerton TD, Kilpatrick DG. (2000) Int J Eat Disord. 27:180–190. De Graaf C, Zandstra EH. (1999) Physiol Behav. 67:513–520. Desor JA, Beauchamp GK. (1987) Physiol Behav. 39:639–641. DiChiara G, North RA. (1992) Trends Pharmacol Sci. 13:185–193. Drewnowski A. (1997) Annu Rev Nutr. 17:237–253. Review. Cabanac M, Duclaux R. (1970) Science. 1970;168:496–497. Fernstrom MH, Krowinski RL, Kupfer DJ. (1987) Biol Psychiatry. 22:529–539. Finn PR, Sharkansky EJ, Brands KM, Turcotte N. (2000) J Abnorm Psychol. 109:122–133. Franko DL, Wolfe BE, Jimerson DC. (1994) Physiol Behav. 56:969–973 Gianoulakis C. (2004) Curr Top Med Chem. 4:39–50. Review. Grucza RA, Robert Cloninger C, Bucholz KK, Constantino JN, Schuckit MI, Dick DM, Bierut LJ. (2006) Alcohol Clin Exp Res. 30:1176–1183. Hajnal A, Smith GP, Norgren R. (2004) Am J Physiol Regul Integr Comp Physiol. 286:R31–37. Kampov-Polevoy AB, Kasheffskaya OP, Sinclair JD. (1990) Alcohol. 1990;7:83–85. Kampov-Polevoy AB, Overstreet DH, Rezvani AH, Janowsky DS. (1995) Pharmacol Biochem Behav. 57:791–795. Kampov-Polevoy AB, Garbutt JC, Janowsky DS. (1999) Alcohol Alcohol. 34:386–95. Kampov-Polevoy AB, Alterman A, et al. (2006) Eat Behav. 7:181–187. Kampov-Polevoy AB, Eick C, Boland G, Khalitov E, Crews FT. (2004) Alcohol Clin Exp Res. 28:1291–1298. Kampov-Polevoy AB, Garbutt JC, Janowsky DS (1998) Alcohol Clin Exp Res. 22:610–614. Kampov-Polevoy AB, Garbutt JC, (2003a) Alcohol Clin Exp Res. 27:1743–1749. Kampov-Polevoy AB, Ziedonis D, (2003b) Alcohol Clin Exp Res. 27:1929–1936. Keskitalo K, Tuorila H, Spector TD, Cherkas LF, Knaapila A, Silventoinen K, Perola M. (2007) Am J Clin Nutr. 86:1663–1669. Leibenluft E, Fiero PL, Bartko JJ, Moul DE, Rosenthal NE. (1993) Am J Psychiatry. 150:294–301. Lemon CH, Brasser SM, Smith DV. (2004) J Neurophysiol. 92:536–544. Leventhal L, Kirkham T, (1995) Brain Res. 685:205–210. Liao J, Schultz P. (2003) Mamm Genome. 14:291–301. Looy H, Callaghan S, Weingarten HP. (1992) Physiol Behav. 52:219–225. Looy H, Weingarten H. (1991) Chem Senses. 16:123–130. Looy H, Weingarten HP. (1992) Physiol Behav. 52:75–82. Macht M, Mueller J. (2007) Appetite. 49:667–674. Mennella JA, Pepino MY, Yourshow LM, Lehman SM (2008) Int Alcohol Clin Exp Res. 32:239A, Abstract 916. Olds J, Milner P. (1954) J Comp Physiol Psychol. 47:419–427. Overstreet DH, Halikas JA, Seredenin SB, Kampov-Polevoy AB, Viglinskaya IV, Kashevskaya O, Badishtov BA, Knapp DJ, Mormede P, Kiianmaa K, Li T.-K., Rezvani A. (1997) Alcohol Clin Exp Res. 21:840–8. Pecina S, Berridge KC. (2005) J Neurosci. 25:11777–11786. Pepino MY, Mennella JA. (2005) Chem Senses. 30 Suppl 1:i319–20. Rada P, Avena NM, et al. (2005) Neuroscience. 134:737–744. Rosenthal NE, Sack DA, (1984) Arch Gen Psychiatry. 41:72–80. Sinclair JD, Kampov-Polevoy A, Stewart R, Li TK. (1992) Alcohol. 9:155–160. Steiner J, Glaser D, Hawilo ME, Berridge KC. (2001) Neurosci Biobehav Rev. 25:53–74. Thompson D, Moskowitz H, Campbell RG. (1976). J Appl Physiol. 41:77–83. Wade GN, Zucker I. (1969). J Comp Physiol Psychol. 69:291–300. Wurtman JJ. (1988) J Clin Psychiatry. 49 Suppl:37–39. Wurtman RJ, O’Rourke D, Wurtman JJ. (1989) Ann N Y Acad Sci. 575:75–82; discussion 82–5. Yamamoto T, Sako N. (2000) Physiol Behav. 69:345–350.
wwwwwwwwwwwwwww
Chapter 190
Anxiety and Self Medication with Alcohol Carmen C. Moran and Anthony J. Saliba
Abbreviations GAD Generalised anxiety disorder PANAS Positive affect negative affect schedule
190.1 Introduction People drink alcohol for many reasons and in many contexts. The reasons for drinking can be categorised in several ways, including physiological and familial perspectives, and can be influenced by various factors, including social pressure and marketing of the product. Estimates of drinkers in the population are extremely variable, in one study alone ranging from 92% to 6%, but the variation included differences across countries, gender, and other sample characteristics (Gmel et al. 2006). The variability associated with estimates may be at least somewhat related to the difficulty in defining a ‘drinker’; is someone who consumes a few glasses per year at social functions such as weddings to be included for instance? In many countries over half the population are drinkers, at least some of the time (Breslow and Graubard 2008; Gmel et al. 2006). In a US-wide study of over 20,000 drinkers, 26% of men and 20% of women admitted to drinking daily (Breslow and Graubard 2008). Early research sought to establish a relationship between alcohol consumption and the reduction of negative emotions. However, not all drinking is problematic nor related to psychopathology. There is a vast literature on problem drinking, addiction, and the threats to well-being that overconsumption poses for the drinker and for others. Several journals are dedicated to this problem. This chapter does not review that literature; rather, this chapter examines the research on drinking motives, which focuses on nonclinical samples and reasons for drinking potentially independent of alcohol addiction or dependence. The presumption behind this approach is that ‘drinking motives are the most proximate factor that precedes alcohol use’ (Kuntsche et al. 2007, p. 76). Such information can, in turn, help inform health providers about circumstances when nonclinical aspects of alcohol consumption may or may not be problematic, such as moderate consumption to self-manage sub-clinical levels of anxiety.
A.J. Saliba (*) National Wine and Grape Industry Centre, Charles Sturt University, Locked Bag 588, Wagga Wagga, NSW, 2678, Australia e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_190, © Springer Science+Business Media, LLC 2011
3061
3062
C.C. Moran and A.J. Saliba
190.2 Anxiety One of the major factors that affects drinking motives is anxiety. While there is concern about giving too much emphasis to mood reduction as a prime reason for drinking (Hussong et al. 2008; Wilson et al. 1989), current research shows that anxiety level is an important influence when included with other variables in research on consumption. Table 190.1 presents the key features of anxiety and its relationship to drinking motives. Anxiety has many lay and formal definitions. Even in the research literature it is not consistently defined or measured. In part this is due to the different ways people experience anxiety and in part due to the different words used to describe the feelings associated with anxiety. Various terms are used apparently interchangeably with anxiety, including some that are associated with specific disorders, such as ‘phobia’ and ‘panic’, with passing states, such as ‘apprehensiveness’ and ‘fear’, or with longer term emotional states such as ‘angst’. In some cases anxiety is seen as part of a temperamental disposition, in others as a reaction to events. In the latter case, the term anxiety may be used in the same way as ‘stress’. Symptoms of anxiety vary across individuals, but there are some commonly reported ones: high levels of apprehension and worry, restlessness, muscle tension, and sometimes other physical symptoms such as accelerated heart rate and difficulty getting to sleep (Andrews et al. 2003; Menzies and Moran 1994). When symptoms become excessive, prolonged, and very distressing, they may result in a clinical diagnosis such as generalised anxiety disorder (GAD). When accompanied by specific beliefs, fears, or behavioural avoidance, then the conditions of agoraphobia or social phobia may be diagnosed. There are many other anxiety disorders with some overlap in symptomatology and diagnostic criteria (American Psychiatric Association 1994; Andrews and Slade 2002). The prevalence of anxiety and anxiety disorders is high, especially when compared with mood and behavioural disorders (e.g., see Fig. 190.1). Despite the variety of anxiety and related disorders, a common picture of anxiety is one of worry, tension, and internal physical discomfort, with an overarching negative emotional state. The precise nature of anxiety is important in the study of motives for drinking. It is generally accepted that anxiety is accompanied by negative affect but unlike depressive states, anxiety is not associated with an absence of positive affect. Thus, an anxious person may have a range of emotions, Table 190.1 Key features of anxiety 1. Anxiety is an unpleasant subjective state, which can also be related to a personality trait, that is, be part of an enduring characteristic of a person 2. Anxiety can be characterised by cognitive, emotional, physiological, and behavioural symptoms. People vary in the extent to which they experience anxiety in these domains 3. Anxiety is related to reactions associated with the fight-or-flight (and freeze) response. Current research examines anxiety reactions across numerous biopsychosocial systems 4. Anxiety is sometimes differentiated from fear, with fear having a more objective cause than anxiety. That is, fear is considered to be about a threat that most people would recognise. Anxiety is more subjective. 5. Estimates on the prevalence of anxiety is difficult, because people can use different terms to explain the feelings associated with anxiety, such as ‘stressed’, ‘fearful’, ‘jittery’. Anxiety can also be passing. Surveys usually focus on the prevalence of anxiety disorders 6. Anxiety disorders are characterised by very high levels of anxiety, usually prolonged, and accompanied by varying physical and behavioural patterns depending on the disorder. For example, these patterns can include avoidance, compulsive repetitive behaviours, and hyperactivity, which in turn relate to disorders such as phobia, obsessive–compulsive disorder, and post-stress disorders 7. Survey estimates put the 12 month prevalence of anxiety disorders at 14% (Australian Bureau of Statistics 2008). This means that within a 1-year period, 14 people in 100 reported having anxiety disorder This table lists the key features of anxiety, and the basic distinction between the feelings of being anxious and an anxiety disorder
190 Anxiety and Self Medication with Alcohol
Prevalence of anxiety, affective and substance disorders in Australia 20 18 Percent prevalence
Fig. 190.1 Prevalence of anxiety, affective, and substance use disorders in Australia. Prevalence data show the percentage of people affected by anxiety, affective (mood) and substance use disorders in a 12-month period. These data were collected as part of large nationwide survey on health (Data from Australian Bureau of Statistics 2008. With permission)
3063
16 14 12 total
10
men
8
women
6 4 2 0 anxiety
affective
substance use
both positive and negative, and alcohol consumption may vary depending on the prevailing emotional state, that is, whether it is positive or negative. Consumption will also vary according to whether an individual has a tendency to use alcohol to cope. In recent times individual differences in anxiety and using alcohol to cope have been researched using one of two main frameworks: (a) the Self Medication Model and (b) the Drinking Motives Model. The Self-Medication Model focuses mainly on negative affective states, whereas the Drinking Motives Model considers the reasons for drinking in terms of both positive and negative effects.
190.3 Self-medication Model The Self-medication model (also known as the Self-Medication Hypothesis) relates to the consumption of alcohol in order to deal with negative emotions that cannot be addressed in other ways or through other resources (Khantzian 2003). The Self-Medication Model overlaps somewhat with the Drinking Motives Model described below. However, research using the Self-Medication Model is more focused on clinical disorders and is less likely to look at positive emotions as a reason for drinking. Individuals high in arousability have been shown to exhibit higher levels of anxiety (Hicks et al. 1992). Self medication with alcohol may be more frequent in those who are physiologically reactive under stress. Using ‘normal’ college students, Colder examined reasons for drinking, alcohol use, concurrent or recent stressors and physiological reactivity in an experimental context (Colder 2001). The coping motive (as a measure of self-medication) was related to stressful life events and to emotional reactivity measured physiologically through respiration bellows, electrocardiograph, and skin conductance. In this study, the coping motive was not related to trait1 negative affect (measured with PANAS items such as ‘scared’ and ‘distressed’). These results suggest that the construct drinking to cope is more likely to be associated with being physiologically sensitive to stress. It could be proposed that those whose physiological reactions are less elevated by stressful events are less likely to regard drinking as a way of coping with their reactions, and thus less likely to use alcohol for self-medication. According to trait theory in Psychology, traits are relatively stable over time and although differ amongst individuals, are not influenced by external factors nearly as much as are ‘states’.
1
3064
C.C. Moran and A.J. Saliba
Much of the work on the Self-Medication Model looks at depression rather than anxiety. For example, Suh et al. (2008) report that the tendency to self medicate with alcohol was associated with lower depression scores. The low depression scores were not a consequence of drinking at the time of measurement. Because the low depression was also associated with high repression scores, Suh suggested that participants were denying their depression (i.e., they were repressing their depression). It is not clear from other aspects of the results whether this was in fact the case. Indeed, it is possible that the self-medication worked, at least in terms of the target symptoms. Because self-medication research overlaps with that of the drinking motives research, further relevant research is discussed below under the motives model, and occasionally the terms coping motive and self-medication are used interchangeably.
190.4 The Drinking Motives Model In the Drinking Motives Model three motives are frequently discussed: the enhancement motive, the social motive, and the coping motive (Cooper et al. 1992). Enhancement motives are measured by items such as ‘because it [drinking alcohol] makes you feel good’. Social motives are measured by items such as ‘it makes social gatherings more enjoyable’. Coping motives are measured by items such as ‘to forget your worries’. A fourth motive, the conformity motive, is sometimes included. The relationship between drinking motives and type of affect is summarised in Fig. 190.2. The underlying assumption is that people can drink to increase positive affect or drink to reduce negative affect. The focus of the motive may be internal or external. The internal–positive affect combination provides the enhancement motive, which is drinking to feel good about oneself. Increasing positive affect can also be external in focus, such as drinking to enjoy a party more, and this is the social motive. Decreasing negative affect can be internal in focus, and this is captured in the coping motive and the desire to be rid of anxiety and other negative feelings. Finally decreasing negative affect can be external in focus and this relates to the conformity motive and the desire to reduce isolation and rejection.
Benefit gained: Internally Increase positive affect
Enhancement motive
Externally Social motive
Direction of Change Decrease negative affect
Coping motive
Conformity motive
Fig. 190.2 Representation of the Drinking Motives Model. The Drinking Motives Model as represented in this figure is adapted by the authors from Cooper’s motives questionnaire (Cooper et al. ; Cooper 1994), which in turn has been examined in other research. This figure shows how the motives for drinking are differentially focussed on either increasing or decreasing negative affect (or feelings) and on gains that are internal or external
190 Anxiety and Self Medication with Alcohol
3065
190.5 Drinking Motives and Consumption Patterns: Initial Results In early research, each of the three main drinking motives was associated with alcohol consumption frequency, but in different ways. The coping motive was associated with more frequent drinking and potential abuse and dysfunction, whereas the enhancement motive was associated with frequent and heavy drinking but not alcohol abuse. The social motive was mainly associated with increased drinking in social situations (Cooper et al. 1992). In a later study on adolescent drinking and overuse, a conformity motive was added to the questionnaire to capture issues related to adolescent peer pressure to drink (Cooper 1994). The results resembled those of the 1992 study in that the coping motive was associated with high use and ‘drinking problems’, whereas the enhancement motive was associated with high frequency and quantity but not ‘drinking problems’. The social and conformity motives were not strong predictors of quantity consumed or problem drinking, but subsequent research has examined social motives as they relate to social anxiety and extended information on this aspect of drinking. This research is discussed under the heading of social anxiety below.
190.6 Anxiety and the Coping Motive The coping motive is especially relevant in the context of anxiety and can be defined as drinking to cope with emotional discomfort or distress, in other words to deal with anxiety. The coping motive is usually hypothesised to be the best predictor (within the motives model) of problem drinking. While taste is a strong motive for drinking alcohol, in data from our ongoing research we note that drinking to reduce anxiety is also very frequently reported as a reason for drinking, for different types of alcohol beverages (see Fig. 190.3). High anxiety on its own, however, is not a predictor of
Fig. 190.3 Motives for drinking by type of beverage. In terms of the Drinking Motives Model discussed in this chapter, ‘feeling good’ parallels enhancement motive, ‘more outgoing’ parallels the social motive, ‘reducing anxiety’ parallels the coping motive, and ‘peer pressure’ parallels the conformity motive. In this data, ‘taste’ is the most strongly cited reason for drinking alcohol, but other motives are also salient, particularly the desire to relax, which could be related to the other motives, for example to the coping, enhancement and social motives. The patterns of motives are relatively consistent across different beverage types, except for the premixed drinks (Data are authors’ own)
3066
C.C. Moran and A.J. Saliba
increased alcohol consumption. Research has shown that motives and drinking behaviours interact with type of mood or emotional state. In order to evaluate this claim, it is necessary to differentiate between coping motives for alleviating depressed mood and those for alleviating anxious mood, rather than just looking at drinking ‘to feel better’. Participants high on anxiety coping motives are more likely to drink when anxious, whereas those high on depression coping motives are more likely to drink when depressed (Grant et al. 2009). This pattern suggests that high anxiety alone is not responsible for increased alcohol consumption, but does predict an increase in consumption for those already with a tendency to drink to cope with anxiety. Thus, ‘anxiety-drinkers’ do not necessarily drink when depressed, but do when anxious. Other people high in anxiety do not drink to cope, and thus do not necessarily increase their drinking when anxious. In a similar vein, Martens et al. reported that individuals high in negative affect (measured on the PANAS, which includes items such as afraid, nervous) and coping drinking motives were at greater risk for alcohol-related problems (Martens et al. 2008) Personality interacts with both anxiety and having a high coping motive. Individuals drinking because of a coping motive are also higher on measures of neuroticism and anxiety sensitivity (perceiving anxiety symptoms as highly aversive; Goldstein and Flett 2009). High generalised anxiety is more likely to be associated with the coping motive when mediated by a high expectation that alcohol reduces anxiety. These variables, high anxiety, coping motive, and high expectancy have been shown to combine to predict high consumption. Goldsmith and colleagues sampled undergraduate students, and those who formed a hazardous drinking subgroup that were more likely to have positive expectations about alcohol’s beneficial effects on tension and worry (Goldsmith et al. 2009). The coping motive is defined as drinking in order to reduce anxiety; research supports the pattern that those high on coping motive are also likely to be higher on anxiety, perceive anxiety symptoms as more aversive, and be more likely to expect alcohol to relieve their anxiety.
190.7 Social Anxiety and the Coping Motive Social anxiety tends to be related more to the coping motive than the social motive for drinking. Social anxiety refers to anxiety that occurs in social situations, usually where a person feels they are being evaluated. It can be related to specific situations, such as parties or public speaking, or it may be more generalised. It could be expected that the social motive would be especially relevant to this group. The general rubric of social anxiety is different to the personality trait of introversion (the two are sometimes considered one and the same); however, individuals with social anxiety are often higher in introversion (Janowsky et al. 2000). The current authors have several unpublished datasets that suggest a link between introversion and alcohol consumption, presumably driven by a coping motive – allowing them to overcome the anxiety associated with tasks they find difficult under normal circumstances yet are expected to do as part of modern-day life. Drinking is also mediated by self-image, a construct which includes assessing one’s self-worth according to others’ approval (Moeller and Crocker 2009). Being highly concerned with self-image, however, also generates high levels of anxiety. Self-image factors may indirectly determine drinking levels when those who score high on self-image goals see high alcohol use as an expected behaviour within their peer group and their high anxiety about self image then makes them more likely to drink to reduce that anxiety. Self-image goals have been shown to be associated with both coping motives and with higher heavy episodic drinking in University undergraduates (Moeller and Crocker 2009). Moeller and Crocker did not measure social motives, as they were particularly interested in drinking to cope with anxiety.
190 Anxiety and Self Medication with Alcohol
3067
The impact of others’ opinions on one’s behaviour does not always indicate problems of self-image or anxiety, as other people sometimes provide useful information about social norms (the descriptive social norm, as distinct from the injunctive social norm). Lee and colleagues found that when combined with a high social motive for drinking, friends’ approval (the injunctive norm) could lead to risky drinking in social situations (Lee et al. 2007). They did not measure anxiety and its relationship to need for approval. Social anxiety is an inconsistent predictor of drinking in the absence of other mediating variables and has been related to both increased and decreased consumption (Ham and Hope 2005; Ham et al. 2005). Lewis et al. (2008) offer a complex mediation model that suggests this may be the case because social anxiety is not directly related to alcohol problems, except when socially anxious subjects are also high on the coping or conformity motives. High socially anxious subjects have been shown to have both higher consumption in social situations and greater expectancies that the alcohol will reduce their social anxiety (Tran et al. 2004). On the other hand, socially anxious subjects did not differ from others when drinking in positive affect situations (Tran et al. 2004). Social anxiety, therefore, relates more strongly to the coping motive, that is, reducing anxiety, than to the social motive which is more about increasing enjoyment. Nevertheless, social motives have been reported as more relevant predictors of alcohol misuse than coping motives in some social contexts, for example in a sample of younger adolescents aged 13–16 (Bradiza et al. 1999). Misuse in this context was assessed by a mix of frequency and amount consumed, which is suggestive of frequent binge drinking. Frequent binge drinking, although intermittent and thus ‘passing’, can be associated with other behaviours that have more enduring negative consequences (e.g., as a result of inappropriate sexual activity and drink-driving). Some researchers combine motives, for example, the desire to feel good in social situations is interpreted as a coping motive as well as a social motive. Some questionnaires may specifically address this, such as The Alcohol Expectancies for Social Evaluative Situations Scale (Bruch 1992), which measures coping motives in terms of social motives. Socially anxious people have an attentional bias to social threat that may prompt increased consumption to cope with that threat. That is, self-medication with alcohol may be also prompted by perceptual factors in social contexts (Carrigan et al. 2004). Generalising this to the treatment of problem drinking, therapists need to consider not just social anxiety levels but perceptual factors operating in social contexts.
190.8 In Vivo Research Most of the work on drinking motives is done in the context of the three or four motives proposed by Cooper and relies on survey methodologies. As an alternative to the survey method, Swendsen and colleagues asked participants to respond to questions generated on a handheld computer three times a day across 30 days (Swendsen et al. 2000). The questions assessed current mood: active, peppy, happy, relaxed, quiet, bored, sad, or nervous. The participants also entered information on their drinking either as they drank or just prior to drinking. Information from the Beck Depression Inventory, State-Trait Anxiety Inventory and demographic questions were also collected. Although the study was framed in terms of the Self-Medication Model, the results resemble those framed in the coping model framework because they addressed interactions between positive and negative affective states and drinking. Nervousness was the only negative state to be associated with increased alcohol consumption. Being happy was associated with increased alcohol consumption as well, which supports the drinking model’s assumption that drinking is not always motivated by a need to manage negative affect.
3068
C.C. Moran and A.J. Saliba
Swendsen’s study suggests using alcohol to cope worked because it was associated with a subsequent reduction in anxiety. As already noted in other research, in this study the effect was more marked for those higher in anxiety than depression: People high on anxiety were more ‘rewarded’ by drinking than those high on depression. People seem to be well aware of this differential effect themselves and are more likely to drink to cope when anxious but not when depressed (Hussong et al. 2005). Whether such rewards are beneficial or harmful to the individual over time will depend on other factors including the presence of specific anxiety disorders and comorbidity factors (Robinson et al. 2009). The physiological and behavioural consequences of alcohol consumption are also important considerations.
190.9 Measurement of Alcohol Intake Within research using the survey method, consumption is most often measured in terms of selfreported frequency, such as number of days per month, and quantity, such as number of drinks in a sitting. The amount of alcohol is not always clearly specified in terms of standard measures of alcohol or ‘standard drinks’. Survey participants often estimate the amount in terms of number of drinks. Consumption may be reported as separate constructs of frequency, quantity, and problem-related drinking behaviours, but treated as a global measure such as ‘increased consumption’ or ‘problem drinking’. Self-reported drinking may be influenced by socially desirable response patterns; these can be powerful enough in some individuals to cause an underestimation of consumption by as much as 33% (Davis et al. 2010). It is not possible to make a simple linear adjustment to responses since individual differences impact on response bias in a complex way. Differences in patterns of drinking may be more informative than individual analyses on frequency, quantity and problems. For example, low frequency but high consumption is indicative of binge drinking, which may then be associated with problems as a result of drinking. O’Connor and Colder (2005) demonstrated five classifications of alcohol usage patterns in a sample of college students, which they then differentially related to the personality variables ‘sensitivity to reward’ and ‘sensitivity to punishment’. Sensitivity to reward was a significant predictor of the usage patterns of binge drinking and problem drinking. Behavioural problems were more associated with binge drinking for female subjects. That is, if they engaged in binge drinking they were more likely to have subsequent problems, which include social consequences. Relating the personality variables to the Drinking Motives Model, O’Connor and Colder found that sensitivity to reward was significantly correlated with all drinking motives, whereas sensitivity to punishment was more strongly correlated to the coping motive and the conformity motive. Thus sensitivity to reward was associated with binge and problematic drinking, and furthermore, the relationship was mediated by the coping, social and enhancement motives for drinking but not the conformity motive. This study suggests that consumption is related more to a desire to feel better in oneself, rather than appear better to others.
190.10 Anxiety Disorders and Alcohol Consumption The relationship between alcohol use and anxiety is bi-directional: people drink when anxious (selfmedication, high coping motive) but drinking can lead to higher anxiety and in some cases generalised anxiety or panic disorder (Brady et al. 2007). With other anxiety disorders, especially social phobia and agoraphobia, the symptoms cause the person to self-medicate with alcohol which may or may not prove problematic over time (Brady et al. 2007; Kushner et al. 1990; Page and Andrews
190 Anxiety and Self Medication with Alcohol
3069
1996). Alcohol increases anxiety in some disorders, but decreases it in others. In cases where there is one broad diagnosis such as agoraphobia, self-medication for anxiety levels may depend on other comorbidity factors such as panic attacks. As Brady and colleagues point out, unravelling the relationship is complicated by co-existing diagnoses and alcohol reactions (Brady et al. 2007). From a health perspective, the individual’s anxiety as well as any problematic drinking would need to be addressed in terms of bi-directional effects. The literature shows a long history of self-medication with alcohol (thus supporting the relevance of the coping motive) although other terms were used in the very early literature. Conger (1956) for example, who highlighted the impact of the anxiolytic properties of alcohol, framed alcohol use in the context of learning theory rather than a motives model. The moderate consumption of alcohol to control subclinical levels of anxiety may be an acceptable practice (though further research is required). What is clearer is that alcohol consumption to control clinical levels of anxiety that remain otherwise untreated is unwise.
190.11 Gender Differences Gender is a potentially important individual difference in coping with alcohol because of the way alcohol is differentially used and metabolised by males and females. However, it is not possible to provide a consistent picture of gender differences in the context of drinking motives. Whereas there are some known patterns of gender differences in alcohol consumption, for example, males drink more, on average (Gmel et al. 2006), the results as they relate to anxiety, self-medication, and the coping motive are not clear. In specific anxiety disorders, such as agoraphobia or social phobia, there is a long-held clinical impression that males are more likely to self-medicate with alcohol (Moran and Andrews 1985). On the other hand, women report higher levels of arousability and are more likely to present with anxiety disorders (Saliba et al. 1998). Whereas there is a demonstrated relationship between high alcohol consumption and anxiety reduction for males, alcohol may increase anxiety in women in social contexts (Wilson et al. 1989). However, when Wilson and colleagues tested a female sample they found alcohol elevated mood, but the effect was found in the placebo group as well suggesting a strong expectancy effect operating in their study. They did not report on alcohol use outside the laboratory setting other than to state that preexisting expectations about alcohol were not related to their results (Wilson et al. 1989). In our dataset (N = 1,229) we have found some small but statistically significant gender differences. This information is presented in Table 190.2, which shows that men score higher on items related to social and enhancement motives. Of special relevance to this chapter, we found no significant gender differences for drinking to cope with anxiety. Men utilise health services much less frequently than women, are less likely to arrange regular checkups and are also less likely to have symptoms checked for fear of wasting the health professional’s time (Australian Institute for Health and Welfare 2001; O’Brien et al. 2005). Unsurprisingly therefore, the incidence of a range of diseases is higher in males than females (Saliba 2008). Although speculative, it may be that males are more in need of self-medication practices due to an unwillingness to seek help. Using an adolescent sample, Hussong and colleagues were unable to find any consistent patterns of self-medication with alcohol and variables such as mood and conduct problems, and decided that the gender differences did not deserve further examination in their results (Hussong et al. 2008). This seems to be the pattern in several studies, where the differences are not sufficiently consistent, if they exist at all, to be able to draw broad conclusions. A whole of modelling approach, where multivariate analyses incorporates the influence of age, situation, personality, arousability and other variables, as well as gender, is required to determine the influence of gender on self-medication with alcohol.
3070
C.C. Moran and A.J. Saliba Table 190.2 Motives for drinking alcohol by gender Means (s.e.m.)b Female Male p Motivesa Enhancement (makes me 3.19 3.34 .036 feel good) (.048) (.056) Social (makes me more 2.49 2.69 .004 outgoing) (.046) (.055) Coping (reduces my level 2.83 2.80 .681 of anxiety) (.049) (.056) Conformity (most of my 2.36 2.47 .129 friends drink it) (.051) (.059) a Motives are based on ‘reasons for drinking’ rather than the Cooper (1992) scale referred to in text b Standard error of the mean Gender differences are inconsistent in the literature on drinking motives. In our large Australian sample (N = 1,229) men scored significantly higher on a five-point scale for social and enhancement motives than women (albeit with small differences) but there were no significant gender differences for drinking to cope with anxiety or conforming to friends’ behaviour (Data are the authors’ own)
In the motives and self-medication literature, therefore, there is no consistent evidence that males are more likely to score higher on self-medication or on the coping motive, or show a differential relationship between motive and consumption when compared with females. This lack of consistent gender difference may relate to the age group of several samples, the nonclinical nature of the samples used, or changes to gender differences in drinking behaviours in recent years.
190.12 Other Factors Related to Anxiety, Motives and Self-medication One of the major limitations in the motives and self-medication literature is the reliance on either student or clinical samples. While the importance of problem drinking in these groups is unequivocal, there is also a need for information on the everyday drinker who is older and not presenting at clinics, and the extent to which drinking to alleviate ‘everyday’ anxiety poses a health risk. In addition, there is little or no emphasis on the type of beverage consumed. For example, there is no information in this body of research on motives for drinking wine and whether the effects differ from those for drinking spirits. Our currently unpublished data suggest there are differences across beverages, particularly with consumers of mixed drinks (see Fig. 190.3). In light of this and recent research demonstrating a health benefit for wine, over and above alcohol (Lindberg and Ezra 2008), it would seem to be important to examine motives for drinking across individual beverages. The recent media discussion on the putative health benefits of wine has added to the list of possible motives for drinking that are not covered in the ‘Drinking Motives Model’. The literature is sparse on information about how many people believe wine is healthy. Figure 190.4 offers some data in this regard, showing that approximately 26% believe wine to be healthy, with the remainder undecided or disagreeing. Saliba and Moran (2010) found that those who perceive wine as healthy consume more frequently in terms of daily patterns, without consuming more volume of wine across time (see Fig. 190.5). There is also some initial evidence in this data to support the idea that those who drink wine to reduce anxiety are also more likely to believe it is healthy for them (r = .237, p < .0001).
190 Anxiety and Self Medication with Alcohol Fig. 190.4 Percentage of people who consider wine to be healthy. Over 26% of a large Australian telephone survey (N = 1229) agree or strongly agree that wine is healthy A large proportion did not agree. The question did not ask whether wine was unhealthy, so disagreement does not indicate a belief that wine is specifically unhealthy (Data are authors’ own)
3071
100% 90% 80% 70% Strongly Agree 60%
Agree Undecided
50%
Disagree 40%
Strongly Disagree
30% 20% 10% 0%
Health Belief
Fig. 190.5 Frequency and Volume of Consumption by level of belief that wine is healthy. This figure shows that those who strongly agree that wine is healthy drink more frequently (bottom line) but tend to drink the same or less overall (top line). Thus a belief that wine is healthy does not appear to predispose to overuse (Data are authors’ own)
3072
C.C. Moran and A.J. Saliba
This effect is more evident in older adults. In light of the widespread information about the putative health affordance of wine, it is timely to extend the research on drinking motives to include other information on beliefs and motives for drinking specific types of alcohol.
190.13 Applications to Other Areas of Health and Disease In Western counties, large percentages of people drink alcohol at least some of the time and anxiety reduction can be a proximal motive for drinking. For the health provider, the challenge is to determine when this is problematic. The coping motive framework helps predict when well-being or health is threatened rather than enhanced, but a multivariate approach is needed, for example incorporating personality and situational variables. Research suggests that the coping motive is likely to be associated with higher consumption and potential problematic drinking, in high anxious individuals. The social motive is more likely to be associated with binge-drinking, which may be age related, but there are no clear predictions for longer term problems due to drinking per se, although there can be major negative consequences as a result of alcohol-induced risk behaviours. The enhancement motive has been shown to have inconsistent relationships with level of alcohol use. These patterns can vary depending on samples. In most cases, current work is based on adolescent or young college students. The motives model has application to other areas of consumption and well-being. It provides a useful pathway to consider the broader set of variables that impinge on consumption and over-consumption of food, alcohol, and other drugs. The motives model is now being applied to marijuana use (Lee et al. 2009; Zvolensky et al. 2009). There is a clear need to reduce smoking, and psychological therapies examine patients’ motives for smoking across a broad biopsychosocial perspective (Lujic et al. 2005). Using the motives framework can provide additional information on reasons for smoking, without being in competition with well-established theoretical perspectives (Piko et al. 2007). However, as the motives model suggests, behaviour is not driven only by a desire to reduce negative affect, whether anxiety or depression, but is also driven by a desire to increase positive affect. Personality traits and situational factors influence the import of particular motives. As a result, health and disease behaviours can be driven by motives as outlined in the motives model above, but a multivariate approach will be most fruitful in understating the drivers of healthy and unhealthy behaviours.
Summary Points • Large proportions of people drink alcohol and some drink often. Not all alcohol consumption is problematic. A broad base of information on motives for drinking contributes to the differentiation between problem and nonproblem drinking. • Anxiety is a commonly cited reason for drinking, with an underlying assumption that alcohol will help alleviate anxiety symptoms. • The relationship between anxiety and alcohol consumption can be researched using either the Self-Medication Model or the Drinking Motives Model. The Self-Medication Model focuses mainly on negative affective states, whereas the Drinking Motives Model considers the reasons for drinking in terms of both positive and negative affect. • The Self-Medication Model treats consumption of alcohol as a means to deal with negative emotions that cannot be addressed in other ways or through other resources (Khantzian 2003).
190 Anxiety and Self Medication with Alcohol
3073
• In the Drinking Motives Model three to four motives are frequently discussed: the enhancement motive, the social motive, the coping motive and the conformity motive (Cooper et al. 1992; Cooper 1994). The enhancement motive is measured by items such as ‘because it [drinking alcohol] makes you feel good’. The social motive is measured by items such as ‘it makes social gatherings more enjoyable’. The coping motive is measured by items such as ‘to forget your worries’. The conformity motive is measured by a desire to fit in with others expectations. • The coping motive is the best predictor (within the motives model) of problem drinking. Research suggests that the coping motive is likely to be associated with higher consumption and potential problematic drinking, in high anxious individuals. High anxiety on its own, however, is not a predictor of increased alcohol consumption. Motives and drinking behaviours interact with type of mood or emotional state, as well as expectations about the effects of alcohol. • Social anxiety refers to anxiety that occurs in social situations, usually where a person feels they are being evaluated. Social anxiety is associated with an increase in the coping motive rather than the social motive for drinking. • The social motive is more likely to be situationally determined and associated with binge drinking, which may be age related. The enhancement motive has been shown to have inconsistent relationships with level of alcohol use. The conformity motive is not a strong predictor in general circumstances. These patterns can vary depending on samples. • The relationship between alcohol use and anxiety is bi-directional: people drink when anxious (self-medication, high coping motive) but drinking can lead to higher anxiety and in some cases can be associated with the onset of an anxiety disorder. • Clinical interventions for problem drinking cannot target anxiety alone because the state itself does not predict problem drinking; rather incorporating a motives model with knowledge of anxiety level provides a better predictor. • The recent media discussion on the putative health benefits of wine has added to the list of possible motives for drinking that are not covered in the Drinking Motives Model. Other motives may also be relevant to drinking behaviours. • The motives model has application to other areas of consumption and well-being. It provides a useful pathway to consider the broader set of variables that impinge on consumption and overconsumption of food, alcohol and drugs. Definitions and Explanations of Key Terms Anxiety: A state characterised by high levels of apprehension and worry, restlessness, muscle tension, and sometimes other physical symptoms such as accelerated heart rate and difficulty getting to sleep. When it becomes chronic or is associated with extreme distress or behavioural problems it becomes an anxiety disorder. Social anxiety: Anxiety that occurs in social situations, usually where a person feels they are being evaluated. Positive affect: A pleasant emotional state or feeling, characterised by positive labels such as happiness, calm, or a sense of well-being. Negative affect: An unpleasant emotional state or feeling, characterised by negative labels such as anxiety, depression, or distress. Self-medication: Ingestion of substances based on experience rather than formal professional advice, in order to achieve a beneficial effect, usually a change in sense of well-being. The Drinking Motives Model: Model to explain drinking based on special goals related to management of positive and negative affect.
3074
C.C. Moran and A.J. Saliba
References American Psychiatric Association. Diagnostic and statistical manual of mental health disorders. 4th ed. Washington: APA; 1994. Andrews G, Slade T. Psychopathology. 2002;35:100–6. Andrews JG, Creamer M, Crino R, Hunt C, Lampe L, Page A. The treatment of anxiety disorders. 2nd ed. New York: Cambridge University Press; 2003. Australian Bureau of Statistics (ABS). National survey of mental health and wellbeing: summary of results, 2007 (cat. no. 4326.0). Canberra: ABS; 2008. Australian Institute for Health and Welfare. Australian health trends 2001. Canberra Australia: Australian Institute of Health and Welfare; 2001. Bradiza CM, Reifman A, Barnes GM. J Stud Alcohol. 1999;60:491–9. Brady KT, Tolliver BK, Verduin ML. Am J Psychiat. 2007;164:217–21. Breslow RA, Graubard BI. Alcohol Clin Exp Res. 2008;32:513–21. Bruch M. J Res Pers. 1992;26:137–49. Carrigan MH, Drobes DJ, Randall CL. Psychol Addict Behav. 2004;18:374–80. Colder CR. Psych Addict Behav. 2001;15:237–45. Conger JJ. QJ Stud Alcohol. 1956;17:296–305. Cooper ML, Russell M, Skinner JB, Windle M. Psychol Assessment. 1992;4:123–32. Cooper ML. Psychol Assessment. 1994;6:117–28. Davis CG, Thake J, Vilhena N. Addict Behav. 2010;35:302–11. Gmel G, Graham K, Kuendig H, Kuntsche S. Addiction. 2006;101:16–30. Goldsmith AA, Tran GQ, Smith JP, Howe SR. Addict Behav. 2009;34:505–13. Goldstein AL, Flett GL. Behav Modif. 2009;33:182–98. Grant VV, Stewart SH, Mohr CD. Psych Addict Behav. 2009;23:226–37. Ham LS, Hope DA. Addict Behav. 2005;30:127–50. Ham LS, Carrigan MH, Moak DH, Randall CL. J Psychopathol Behav. 2005;27:115–21. Hicks RA, Conti PA, Nellis T. Percept Motor Skill. 1992;74:659–62. Hussong AM, Gould LF, Hersh MA. J Stud Alcohol Drugs. 2008;69:296–307. Hussong AM, Galloway CA, Feagans LA. J Stud Alcohol. 2005;66:344–53. Janowsky DS, Morter S, Tancer M. Depress Anxiety. 2000;11:121–5. Khantzian EJ. Prim Psychiat. 2003;10:47–54. Kuntsche E, Knibbe R, Engels, R, Gmel G. J Stud Alcohol Drugs. 2007;68:76–85. Kushner MG, Sher KJ, Beitman BD. Am J Psychiat. 1990;147:685–95. Lee CM, Geisner IM, Lewis MA, Neighbors C, Larimer ME. J Stud Alcohol Drugs. 2007;68:714–21. Lee CM, Neighbors C, Hendershot CS, Grossbard JR. J Stud Alcohol Drugs. 2009;70:279–87. Lewis MA, Hove C, Whiteside U, Lee CM, Kirkeby BS, Oster-Aaland L, Neighbors C, Larimer ME. Psych Addict Behav. 2008;22:58–67. Lindberg ML, Ezra AA. Clin Cardiol. 2008;31:347–51. Lujic C, Reuter M, Netter P. Eur Psychol. 2005;10:1–24. Martens MP, Neighbors C, Lewis MA, Lee CM, Oster-Aaland L, Larimr ME. J Stud Alcohol Drugs. 2008;69:412–9. Menzies RG, Moran CC. The nature and treatment of anxiety. Sydney: Anxiety Disorders Clinic University of Sydney; 1994. Moeller S, Crocker C. Psych Addict Behav. 2009;23:334–40. Moran CC, Andrews G. Brit J Psychiat. 1985;146:262–7. O’Brien RK, Hunt K, Hart G. Soc Sci Med. 2005;61:503–16. O’Connor RM, Colder CR. Psych Addict Behav. 2005;19:10–20. Page AC, Andrews G. Aust NZ J Psychiat. 1996;30:410–4. Piko BF, Wills TA, Walker C. Addict Behav. 2007;32:2087–98. Robinson J, Jitender S, Cox BJ, Bolton J. J Anxiety Disord. 2009;23:38–45. Saliba AJ (2008). Impact of rurality on optical health: review of the literature and relevant Australian Bureau of Statistics data. Rural and Remote Health, 8 (online), 1056. http://www.rrh.org.au Saliba AJ, Henderson RD, Deane FP, Mahar D. J Gen Psychol. 1998;125:263–9. Saliba AJ, Moran CC. Food Quality and Preference. 2010;21(7):692–96. Suh JJ, Ruffins S, Robins EC, Albanese MJ, Khantzian EJ. Psychoanal Psychol. 2008;25:518–32. Swendsen JD, Tennen H, Carney MA, Affleck G, Willard A, Hromi A. J Abnorm Psychol. 2000;109:198–204. Tran GQ, Anthenelli RM, Smith JP, Corcoran KJ, Rofey DL. J Stud Alcohol. 2004;65:715–24. Wilson GT, Brick J, Adler J, Cocco K, Breslin C. J Stud Alcohol. 1989;50:226–35. Zvolensky MJ, Marshall EC, Johnson K, Hogan J. Exp Clin. 2009;17:31–42.
Part XXXIII
Quality of Life
wwwwwwwwwwwwwww
Chapter 191
Developmental Aspects of Health Related Quality of Life (HRQL) in Food Related Chronic Disease: The Example of Food Allergy Audrey DunnGalvin and Jonathan O’B. Hourihane
Abbreviations HRQL FAQLQ PF FAIM OIT
Health Related Quality of Life Food Allergy Quality of Life Questionnaire Parent Form Food Allergy Independent Measure Oral Immunotherapy
191.1 Introduction In recent decades our understanding of human health has changed. In the 1970s, older biomedical definitions of health, based on ‘an absence of disease’, were justly criticized as reductionist and limited in scope and clinical researchers began defining health as a dynamic, multifactor, biopsychosocial phenomenon that influences physical, psychological and social functioning (Engle 1977). Recognition of the importance of these influences on health and disease is consistent with evolving conceptions of mind and body and represents a significant change in medicine and the life sciences. Recent developments include the idea that emotional processes such as stress moderate activity in nearly all systems of the body and can directly influence the pathophysiology of disease. Discovery of these and other relationships between behaviour and health has changed the way health and disease are understood. The biopsychosocial perspective has also changed how health and disease are measured. It is now recognised that it is essential to include outcome measures that reflect the patient perspective for evidence-based decision making in clinical practice. Outcomes research has been key in altering the culture of clinical practice and health care research by changing how we assess the end results of health-care services, including clinical and therapeutic interventions, evaluation and health policy. Further, the promotion of evidence-based practice has increased the demand for outcome data. Health related quality of life measures provide a powerful means of measuring outcomes, enabling service providers in the clinical field to audit ‘outcome’ information for particular populations, thereby altering and improving resources and programmes, and prioritizing needs.
A. DunnGalvin (*) Clinical Investigations Unit, Department of Paediatrics and Child Health, Cork University Hospital, University College Cork (UCC), Wilton, Cork, Ireland e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_191, © Springer Science+Business Media, LLC 2011
3077
3078
A. DunnGalvin and J.O’B. Hourihane
Health Related Quality of Life (HRQL) is a multidimensional construct, which evaluates physical, psychological and social components that may be impacted by a disease or medical condition, from the patient perspective. There are two major types of HRQL instruments, generic and disease- specific. Generic HRQL instruments are not specific to any particular disease and are therefore useful for comparing HRQL across different conditions, whereas disease-specific questionnaires focus on issues pertinent to one disease. However, generic instruments are necessarily more ‘general’ and therefore less sensitive to the particular problems associated with a particular condition. Diseasespecific HRQL questionnaires which provide an in-depth picture of the day to day concerns of patients are an increasingly important outcome measure particularly in the context of chronic diseases. They are also able to capture small changes in HRQL that may occur as a result of clinical or therapeutic treatment. However, there are important questions that must be answered before HRQL measures can reach their full potential in research, practice and policy. These questions include: what are the correlates of HRQL (e.g. anxiety, compliance, risk perception, coping behaviours) and how may they impact on perception of health and health reporting? Which variables are causally related to HRQL status, and which variables are the effects of HRQL status? Since HRQL depends on the subjective perception of a disease, what are the underlying neurobiological mechanisms, and are these unidirectional or bidirectional? Such questions have relevance for the interpretation and usefulness of HRQL measures in clinical practice (e.g. treatment choices for certain patient groups and individual patients), health policy (e.g. the allocation of health-care funds), the development of psycho-educational interventions (the precise targeting of information) and research (the causal direction of factors related to HRQL). We know that physiological measures often correlate poorly with functional capacity and well-being (Guyatt 1985) and patients with the same clinical criteria often have dramatically different responses, which depend on the subjective perception of disease impact. Such perception may be impacted by gender, for example, resulting in different health-reporting rates between sexes (DunnGalvin et al. 2006, 2008). Some researchers in the field have attempted to develop causal pathway models to explain the direction of factors related to health related quality of life. Wilson and Cleary (1995) proposed that there is an unidirectional relationship between several kinds of outcomes, for example, biological and physiologic phenomena give rise to symptoms (and treatment side effects), which in turn have effects on functioning domains (such as physical, social and role functioning). The constellation of these effects leads to general health perceptions and, ultimately, an individual’s concept of his/her overall HRQL. All of the above are also influenced by innate characteristics and environmental factors. Ferrans et al. (2005) modified the Wilson and Cleary model to make it simpler and have added more complete explanations for the components of the model. Sousa and Kwok (2005) used structural equation modelling to examine the relationship between the components of the original model. They conclude that the model fits the data (derived from patients with HIV) reasonably well but suggested that links be added between symptoms and general health perceptions and symptoms and HRQOL. The correlations for these links, however, were modest to low. The previous models briefly allude to the possibility that the flow of the model may not be strictly unidirectional but do not provide data on this important consideration. If the flow is bidirectional for some of the components, this has profound implications in terms of interpretation and application of HRQL results. Furthermore the models have been developed from adult patients, and developmental considerations have not been taken into account (DunnGalvin et al. 2009a–d). The evolution of the biopsychosocial perspective on health and health related quality of life has also coincided with a growing recognition of the multidimensionality and complexity of causation, including how environmental, social, psychological and biological systems interact to influence
191 Developmental Aspects of Health Related Quality of Life (HRQL)
3079
health and developmental outcomes. Thelen and Smith (1994:127) suggest that ‘the boundaries between what is innate and what is acquired become so blurred as to be at the very least uninteresting compared to the powerful questions of developmental process’. A developmental trajectory or pathway may be understood as a lifelong process of developmental integration that involves complex interactions between biological and environmental factors that influence the phenotypic expression of physiology, psychology and behaviour (Halfon and Hochstein 2002). They may also delineate critical or sensitive transition points in development when physiological or environmental variables associated with a particular disease may have a relatively greater impact and/or interact with already existing normative demands and changes in socialization (Halfon and Hochstein 2002). Because children are rapidly changing and developing in response to these biopsychosocial influences, the developmental process plays an important role in shaping and determining their health and health related quality of life. Research on the perception of health related quality of life – and its impact in terms of behaviour – may be particularly relevant in the context of chronic disease in childhood, as children not only have to meet their age-related developmental tasks, but they also have to manage their disease, which leads to a heightened risk of maladaptation (DunnGalvin et al. 2009a–d). A chronic condition may affect and/or interact with already existing normative demands and changes in socialization (Schmidt. 2003). Thus, although most children follow normative developmental pathways and encounter predictable transition points, disease-specific pathways may be embedded within these trajectories and influence the phenotypic expression of physiology, psychology and behaviour (Halfon and Hochstein 2002). Adaptational processes of children and adolescents with chronic conditions are of utmost importance because of their long-term consequences. Children with any chronic condition have twice the risk of developing mental health disorders of healthy children, even without an accompanying physical disability (Schmidt 2003). In this chapter we review literature on the impact of food allergy on HRQL of children, teens and their parents. Biological hypersensitivity to environmental stimuli is a central feature of food allergy entailing a need for constant vigilance about, and avoidance of, certain foods (Table 191.1). We begin with a brief overview of prevalence, mechanisms and clinical symptoms of food allergy. We then examine literature on the impact of food allergy on the perceived health related quality of life of children, teens and parents. Next, we present research on developmental differences in perception of the impact of food allergy and the behaviours or coping strategies children evolve in order to deal with this impact. We then draw on selected neurobiological literature in allergic diseases, in addition to some of the key psychobiological theories in current work on threat perception in health, to argue for a broader understanding of HRQL. This review also aims to provide a scientific basis for
Table 191.1 Key facts of food allergies • A food allergy is an adverse immune response to a food protein and the food protein triggering the allergic response is termed a food allergen. • Food allergy is distinct from other adverse responses to food, such as food intolerance, pharmacologic reactions and toxin-mediated reactions. • Six to eight percent of children under the age of three, and nearly 4% of adults, have food allergies and prevalence is rising. • Food allergies cause roughly 30,000 emergency room visits and 100–200 deaths per year in the USA. • The most common food allergies in adults are shellfish, peanuts, tree nuts, fish and eggs, and the most common food allergies in children are milk, eggs, peanuts and tree nuts. • Treatment consists of avoidance diets, in which the allergic person avoids all forms of the food to which they are allergic. For people who are very sensitive, this may include touching or inhaling the problematic food. • Those diagnosed with a food allergy may carry an autoinjector of epinephrine such as an EpiPen or Twinject.
3080
A. DunnGalvin and J.O’B. Hourihane
the development of appropriate models linking symptoms, functioning, development, underlying physiological mechanisms and HRQOL.
191.2 Prevalence, Mechanisms and Clinical Manifestations of Food Allergy Atopy may be defined as a genetically and environmentally determined predisposition to clinically expressed disorders, including allergic rhinitis, atopic dermatitis or eczema, food allergy and allergic asthma, regulated through immune phenomena in which many cells (i.e. mast cells, eosinophils and T-lymphoctytes) and associated cytokines, chemokines and neuropeptides play a role. Food allergy is growing in prevalence, and increasing rates of diagnosis means that many more parents and children must learn to live and cope with food allergy. Food allergy affects approximately 6–8% of young children and 3–4% of young adults in the UK, USA and Europe (Sampson 2005). Allergy, particularly to peanuts, is the most common cause of anaphylaxis outside hospital yet there are other common food causes such as shellfish, fish, milk, soy, wheat and eggs (Sampson 2005). These foods may not only cause fatal or near-fatal reactions, but also tend to induce persistent sensitivity in most patients, in contrast to other foods such as milk, eggs and soybeans, which are frequently associated with milder reactions and are usually outgrown. The life-threatening nature of anaphylaxis makes prevention the cornerstone of therapy (Hourihane et al. 1998a). Avoidance of the responsible food allergen and emergency management in the form of injectable epinephrine (Epipen or Anapen), in case food allergen is accidentally ingested, is the only reliable therapy offered to those living with food allergy. Anticipatory guidance measures form the cornerstone of advice, including reading food ingredient labels, concern for cross-contamination, vigilance in a variety of social activities and immediate access to the Epipen. However, avoidance is complicated by the fact that peanuts, nuts and soy can be found in many foods (e.g. breads, muffins, pastries, biscuits, cereals, soups, ice creams, seasoning, sauces) and in different forms as an emulsifier or thickening agent. Food allergy occurs when the body’s immune system mounts an exaggerated response against the offending food, which acts as an allergen. It is a type of hypersensitivity reaction. It can be either: • A type I, IgE-mediated reaction: this is the usual cause of food allergy. After initial sensitisation, the release of mediators such as histamine are triggered each time a person is exposed to the food. It is these mediators that cause symptoms. • A delayed, type IV-mediated reaction: these reactions are mediated mainly by T-cells. They typically affect the gastrointestinal tract or skin, for example, exacerbation of eczema in children after milk ingestion. The European Academy of Allergy and Clinical Immunology has proposed a revised nomenclature for allergic and related reactions (Johansson et al. 2004). According to this proposal, adverse reactions to food should be termed ‘food hypersensitivity’. The term food allergy should be used when immunological mechanisms have been demonstrated, and includes both IgE- and non-IgEmediated reactions. All other reactions, which have sometimes been referred to as ‘food intolerance’, should be termed non-allergic food hypersensitivity (Fig. 191.1). In an IgE-mediated reaction, symptoms involving the oropharynx and gastrointestinal tract may occur within minutes of ingesting a food allergen. Itching and swelling of the lips, tongue and soft palate as well as nausea, abdominal pain, vomiting and diarrhoea have all been demonstrated secondary to food allergy (Sicherer 2002). Anaphylaxis refers to a sudden, severe, potentially fatal, systemic allergic reaction that can involve skin, respiratory tract, gastrointestinal tract and cardiovascular system. The most dangerous symptoms include breathing difficulties and a drop in blood
191 Developmental Aspects of Health Related Quality of Life (HRQL)
3081
Food Hypersensitivity
Food Allergy
Non-Allergic Hypersensitivity
IgE-Mediated Food Allergy
Non IgE-Mediated Food Allergy
Fig. 191.1 Nomenclature proposed by the European Academy of Allergology and Clinical Immunology. Flow chart with terms describing adverse reactions to foods (Adapted from Johansson et al. 2004)
p ressure, or shock, which are potentially fatal. Symptoms of anaphylaxis may develop within seconds or a few hours after ingestion of a food allergen, with the vast majority of reactions developing in the first hour. Symptoms can include swelling (especially lips, tongue or throat), difficulty in breathing, abdominal cramps, vomiting, diarrhoea, circulatory collapse, coma and death. Typical allergy medications such as antihistamines work too slowly and cannot reverse the effects of chemical mediators. Adrenaline or epinephrine, therefore is the treatment of choice and must be administered by injection promptly (Hourihane 1998b; Sicherer 2002). A growing number of families must live and cope with food allergy on a day-to-day basis, however, the socio-emotional impact of food allergy on children and families has been little researched until recently (DeBlok 2009; DunnGalvin et al. 2007, 2008). The majority of research in food allergy has been bio-medical in orientation, focusing on issues such as the molecular structure of allergens, or methods of diagnosis. In the last 5 years, there has been a growing interest in the development of questionnaires to measure the impact of food allergy on health related quality of life (DeBlok et al. 2007). These studies have provided an insight into the everyday burden of living with food allergy.
191.3 The Impact of Food Allergy on Health Related Quality of Life (HRQL) Early studies used generic HRQL questionnaires, in particular the Child Health Questionnaire (Landgraf et al. 1999) to investigate the impact of food allergy on perceived health related quality of life (DunnGalvin et al. 2008, 2010).
191.3.1 Research Using Generic HRQL Measures The first study on HRQL was carried out by Primeau and colleagues (2000), who studied a sample of 301 patients and evaluated the quality of life and family relations of children and adults with
3082
A. DunnGalvin and J.O’B. Hourihane
p eanut allergy, and compared the results to that of children and adults with rheumatological disease (CRD). It was shown that the parents with food allergic children had difficulties in many areas. Remarkably, the authors found that families with peanut allergic children experience significantly more disruption in their familial and social interactions/activities than families with a child with CRD and suggested that this may be due to the constant risk of sudden death in the peanut allergy group leading to greater parent restriction of activities. There was also evidence that the educational and emotional support needs of families with food allergy are not being met (Sicherer 2002). The authors found that the effect of manufacturers labelling their product with ‘May contain traces of nut’ gave rise to parents’ frustration and limited food choices. In addition, parents have the responsibility of ensuring their children do not ingest peanuts or food containing traces of peanuts which may be difficult in the context of modern food preparative techniques. Sicherer et al. (2001) measured parental perception of physical and psychological functioning of families living with food allergy. The authors randomly selected 400 members of the food and anaphylaxis network, with families of children age 5–18 years old and had 253 responses. Results indicated that peanut allergy impacted significantly on general health, parental distress and family activities. Those with two or more food allergies scored significantly lower, depending on how many foods they were avoiding. There was also evidence to suggest that the educational and emotional support needs of these families are not being met. Bollinger et al. (2006), in a survey of 87 parents in the USA, found that over half had made significant changes to their social activities to accommodate their child’s food allergy, avoiding birthday parties, soccer games and school field trips. Forty-one percent of the parents surveyed said their child’s allergy had a significant impact on their own stress levels. The authors conclude that studies are needed about how stress and avoiding activities might affect the psychological and social development of children with food allergies. Avery et al. (2003) assessed the effect of peanut allergy on the quality of life in children aged 7–12 years and contrasted this with experiences of children with insulin-dependant diabetes mellitus (IDDM). Their results indicated that children with IDDM have similar problems as children with peanut allergy. These include food choices, social restriction, issues relating to school, the carrying and use of a syringe and the chronic nature of the condition. Results showed that children with peanut allergy had poorer quality of life and are more anxious concerning accidental ingestion of peanut than children with diabetes are of having a hypoglycaemic reaction. Gender differences have also been noted (Marklund et al. 2004), with girls reporting lower HRQL.
191.3.2 HRQL Research Using Disease-Specific Measures The first validated HRQL food allergy specific measure, the Food Allergy Quality of Life–Parental Burden (FAQL-PB) questionnaire (Cohen et al. 2004) measures the parental burden associated with having a child with food allergy. Scores in the food-allergic cohort were significantly lower for general health perception, parental distress and worry, and interruptions and limitations in usual family activities, than in healthy controls. Scales were also lower in subjects with multiple food allergies. More recently, several measures have been developed to assess quality of life in children and teens, under the aegis of EuroPrevall, an EU project which aims to improve quality of life for parents, children, teenagers and adults with food allergy (DeBlok et al. 2007; DunnGalvin et al. 2008, 2010). EuroPrevall is a multidisciplinary integrated project (IP) involving 17 European member-states. Of the 63 partners, there are 15 clinical organisations and six small–medium sized enterprises (SMEs) as well as the leading allergy research organisations in Europe. Since the project began in 2005, new partners have also joined from New Zealand, Australia, Russia, India, Ghana and China. Three food
191 Developmental Aspects of Health Related Quality of Life (HRQL)
3083
allergy specific questionnaires have been developed and published under the auspices of EuroPrevall (DunnGalvin 2008, 2010; Flokstra DeBlok et al. 2008, 2009a, b). These are Food Allergy Quality of Life Questionnaire – Parent Form (parent-administered for children aged 0–12 years), and the Food Allergy Quality of Life Questionnaire – Child/Teen/Adult Form (self-administered for children and teens aged 8–17 years and adults aged 18 +) The FAQLQ-PF, -CTF and AF were developed and validated in five stages: (1) item generation using focus groups with children, teens, and parents’ expert opinion and literature review; (2) item reduction, using clinical impact and factor analysis; (3) the evaluation of internal and test-retest reliability and construct validity; (4) cross-cultural and content validity examined by administering the questionnaire in a US sample (FAQLQ-PF, only); and (5) longitudinal validity examined by administering the questionnaire over three time points pre/post food challenge.
Key Points of the Food Allergy Quality of Life Questionnaire: Parent Form (FAQLQ-PF) Developed using gold standard quantitative and qualitative methodology Three age groups: • 0–3 years – 14 items • 4–6 years – 26 items 1 Form • 7–12 years – 30 items Three subscale scores (emotional impact; food anxiety; social and dietary limitations) calculated as the mean of each scale. The total score is calculated as the mean of the three subscales. Supplementary sections: • • • • • •
clinical child variables parental concern for their child’s emotional and physical health stress levels experienced by parents and family impact on child and family activities expectation of outcome following accidental ingestion of allergen satisfaction with clinical therapy, intervention, information, etc.
Very high reliability and validity (cross-sectional, cross-cultural, longitudinal). Validated in seven languages, to date. The development and validation studies found a severe impact of food allergy on HRQL in relation to psychosocial aspects of children’s and teen’s everyday lives. For example, in the initial focus groups put in place to generate items for the FAQLQ-PF, parents suggested that the anxiety associated with the risk of a potential reaction has more profound effects on emotional and social aspects of a child’s everyday life, than clinical reactivity induced by food intake. The importance of a subscale assessing this aspect of anxiety was subsequently confirmed using clinical impact and factor analytic methodologies (DunnGalvin et al. 2008–2010). Children were also found to be ‘generally anxious’, that is, the anxiety associated with food often ‘generalised’ to non-food situations (DunnGalvin & Hourihane, 2009,a,b,c). In addition, multivariate analysis showed an interaction between sex and age group for impact of general emotional impact on HRQL scale, in effect, parents of boys reported higher mean total scores up to the age of 6 years; parents of girls reported higher mean scores in the 6–12 years age group, particularly in the subscales ‘general emotional impact’ and ‘food anxiety’; whereas boys had higher scores in the ‘social and dietary limitations’ subscale at all ages. Example of items associated with each subscale may be seen in Table 191.2.
3084
A. DunnGalvin and J.O’B. Hourihane
Table 191.2 Examples of items and content in the three subscales of the Food Allergy Quality of Life Questionnaire – Parent Form (FAQLQ-PF) (Reprinted from DunnGalvin et al. 2008) Emotional Impact: Items concern psychological phenomena such as feeling different from other children, frustration, control and generalised anxiety stemming from food allergy: My child feels different from other children because of food allergy. Food Anxiety: Items concern anxiety relating to food: My child is afraid to try new foods because of food allergy. Social and Dietary Limitations: Items concern everyday dietary and social restrictions: My child’s ability to take part in pre-school events involving food (class parties, treats, lunchtime) is limited by food allergy. Three factors (emotional impact, food anxiety, social and dietary limitations) emerged following exploratory and confirmatory factor analysis in the development and validation of the Food Allergy Quality of Life Questionnaire – Parent Form (FAQLQ-PF)
Fig. 191.2 Flowchart showing study design and FAQLQ-PF scores at three time points (baseline, 2 months, 6 months) for children aged 0–12 years undergoing diagnostic food challenges. Although significant differences were found between positive and negative groups on all subscales and total score at 6 months (F (2,59) = 6.221, p < 0.003), HRQL improved significantly post challenge time points (all p < .05) for both positive and negative groups. Higher scores indicate greater impact of food allergy on HRQL (Reprinted from Hourihane and DunnGalvin 2009)
We evaluated longitudinal validity by administering the FAQLQ-PF to parents of children 0–12 years before the child underwent a clinically indicated food challenge, and at 2 months and 6 months post food challenge (DunnGalvin et al. 2010; Hourihane et al. 2009). In total, 82 children underwent a challenge (42 positive, 40 negative). Although significant differences were found between positive and negative groups on all subscales and total score at 6 months (F (2,59) = 6.221, p < 0.003), interestingly we found that HRQL improved significantly post challenge time points (all p < .05) for both positive and negative groups (Fig. 191.2). A possible explanation for improvement in the ‘positive’ groups (long suspected but never documented) concerns the impact of uncertainty on perception of HRQL. This also suggests that a food challenge (open or double blind) may be valuable, not only as an essential diagnostic tool, but as a therapeutic one. In effect, by providing a sense of certainty, a food challenge may have a positive impact on HRQL, irrespective of outcome. Other published research on uncertainty in chronic disease strengthens this argument (e.g. Mullins et al. 2007).
191 Developmental Aspects of Health Related Quality of Life (HRQL)
3085
Table 191.3 Odds ratios for likelihood of volunteering to take part in IT study calculated according to level of expectation of adverse outcomes following accidental ingestion by their child (FAIM): 1 = Group B, consent to immunotherapy (DunnGalvin et al. in press) FAIM: Expectation of Adverse Outcomes (1) Odds Ratio (95% CI) p-Value Adjusted Effectsa What chance do you think your child has of accidentally ingesting the food 3.421 (1.811–6.267) .03 to which they are allergic? ( >3.5) What chance do you think your child has of having a severe reaction if food 3.945 (1.965–5.105) .001 is accidentally ingested? ( >3.5) What chance do you think your child has of dying from their food allergy 4.263 (2.351–7.613) .005 following ingestion in the future? ( >3.5) 3.267 (1.905–5.344) .01 What chance do you think your child has of effectively treating them or receiving effective treatment from others (including Epipen administration) if they accidentally ingest a food to which they are allergic? (>3.5) FAIM total score (>3.5) 6.753 (3.451–9.728) .002 a Adjusted for age, sex, experience of anaphylaxis, experienced symptoms, socio-economic variables Parents who perceive that their child is at high risk of dying from food allergy are more likely to enrol their child in an investigational trial in which the child will be given peanut immunotherapy (OR 6.75; CI 3.45–9.73). This perceived level of threat may be an important factor motivating parents to consent to their children taking part in investigational therapies in a ‘controlled’ environment
A recent study (DunnGalvin et al. 2009d) examined specific psychological factors, related to HRQL, that may impact on parents’ decisions to take part in clinical studies. Parents of food allergic children were offered investigational oral immunotherapy (OIT) in the regular outpatient clinic. Forty parents (Group A) declined, and 25 parents (Group B) agreed to take part. Both groups completed the Food Allergy Quality of Life – Parent Form (FAQLQ-PF). Our results show that parents who perceive that their child is at high risk of dying from food allergy (Table 191.3) are more likely to enrol their child in an investigational trial in which the child will be given peanut immunotherapy (OR 6.75; CI 3.45–9.73). This is in spite of the fact that the experimental therapy is intensive and has attendant adverse risks including induction of anaphylaxis, compared to the routine clinical practice. The association was independent of the severity of symptoms, experience of anaphylaxis and perception of the impact of food allergy on HRQL. Socioeconomic status was not a significant factor. These findings may be explained, in part, by parental concern to avoid potentially life-threatening consequences of accidental ingestion in the often ‘uncontrolled’ environment of their child’s everyday life. This perceived level of threat may be an important factor motivating parents to consent to their children taking part in investigational therapies in a ‘controlled’ environment, even though this involves a protocol in which reactions are more likely than if not in the trial. These findings concur with Zupanic et al. (1997), who found that determinants of parental authorisation for involvement of newborn infants in clinical trials included perceptions of risk and benefit to the child.
191.4 Living and Coping with the Impact of Food Allergy The way in which children and adolescents perceive and cope with chronic health conditions is considered as an increasingly important predictor of health and psychological well-being in clinical and psychosocial research (Schmidt et al. 2003). In medical and health psychology, efforts have increasingly been made to assess coping of children and adolescents with chronic conditions. Coping has
3086
A. DunnGalvin and J.O’B. Hourihane
not only been shown to be related to patient HRQL, but mediates health behaviour as well as healthcare utilisation (Feeney 2000). Although quantitative studies provide consistent, replicable results that can be compared across populations, there are also inherent complexities in studying the experience of living with a disease in the dynamic, interactional process within the family. Life transitions provide a naturalistic research opportunity to investigate adaptability to stress and the link to health outcomes and health related quality of life. Thus, there is an increasing recognition of the need to qualitatively explore patients’ experiences, of what it is like to live with a chronic condition in order to better understand the decisions people make about coping and managing their condition. The personal meaning of having a particular disease ‘is strongly related to the patient’s self-care and to the degree of psychological and social adaptation to the disease’ (Kyngas and Hentinen 1995:734) and compliance with medical direction is thus determined in part by the person’s individual perception of the condition and its management. A recent study (DunnGalvin et al. 2008, 2009) represented a first attempt to provide an integrated developmental framework to explain the onset, development and maintenance of food allergy related cognitions, emotions and behaviour. Sixty-two children/teenagers aged 6–15 years took part in 15 age appropriate focus groups, 52% of whom were female. Parents were also interviewed. All children were physician diagnosed with IgE-mediated food allergy and had been issued with an anapen/epipen. Through qualitative enquiry, a framework for evaluating children with food allergy was developed. Developmentally appropriate techniques such as vignettes (where children could comment on characters in the third person) and activity books were designed to stimulate discussion, maintain interest and minimize threat to the child’s self-esteem. Analyses of the data encompassed precipitating events (stressful events in children’s lives caused by food allergy related factors), psychological impact (cognitive appraisal and emotional effects) and behavioural consequences or coping strategies. Open coding (the first step in analysis) in qualitative grounded theory (Charmaz 2000) may be seen as a descriptive ‘still’ of all the meaning in the data. For example, children discussed growing up and living with food allergy, about feeling different, about low awareness in their social worlds, about their fears and uncertainties in relation to experiencing an allergic reaction, food safety and socialising in many different contexts (eating out, going to the cinema, being with friends, meeting new people). The transcript data was organised into multiple codes and then into categories. Axial coding (the next step in analysis) links codes with distinct
Open Coding
Structure
Axial Coding
Selective Coding
Process
Fig. 191.3 Flowchart showing analytic method in grounded theory. Open coding in grounded theory attempts to fracture the data in order to extract as much information as possible. The transcript data was organised into multiple codes and then into categories. Axial coding links codes with distinct categories structure (experience) with process (what happens as a result) while, at the same time, attempting to retain as much meaningful information as possible. Selective coding results in a theoretical model (DunnGalvin and Hourihane 2009)
191 Developmental Aspects of Health Related Quality of Life (HRQL)
3087
Growing up with the rules Maximisation
Growing up with Difference
The Search for Normality : Driver to Coping
Balanced Adaptation
Minimisation
Growing up with uncertainty
Fig. 191.4 The basic model of structure and process: experience and behaviour. The experience of living with food allergy, ‘living with the rules’, ‘living with difference’ and ‘living with uncertainty’ leads children to cope by means of minimisation, maximisation and balanced adaptation, in the search for normality (DunnGalvin and Hourihane 2009)
categories to form a substantive theory of action, while, at the same time, attempting to faithfully represent the dynamic interrelationships of the children’s experience. The analytic method is illustrated in Fig. 191.3. Our findings indicated that experience and coping in food allergy is more than simply a strategy, it is a cumulative history of interactive processes (age, gender and disease specific) that are embedded in a child’s developmental organization. We can conceptualise food allergy as a central ‘lens’ in children’s lives through which they interpret experiences. When children and teens are confronted with a stressful event, such as a birthday party, a novel situation, an allergic reaction or making new friends, the way in which they appraise the event and its attendant emotional impact are viewed through this lens. The basic model is illustrated in Fig. 191.4. How this lens is constructed and its psychological impact (uncertainty, anxiety, confusion, difference) on individual children is modified by age, gender, context, prior experience, attitudes of parents, attitudes of peers and levels of general awareness. In most children under the age of 8 years, there is a certainty of parental and adult knowledge and a consequent sense of control of events relating to food allergy. However, an important transition point occurs when children learn or feel that parents (and therefore children themselves) cannot conclusively prevent an allergic reaction, after which we see a change in cognitions, emotions and behaviour, and coping strategies become more differentiated.
191.5 L iving with Uncertainty Is a Central Theme in the Experience of Food Allergy To provide an example, living with uncertainty (DunnGalvin et al. 2008a, b; 2009a–d) is an important concept that affects children’s sense of control, beliefs about risk, level of vigilance and confidence in safety. Young children have an illusory perception of control because of parent protection. However, we see the roots of uncertainty in even very young children who are aware of parent anxiety and speak about
3088
A. DunnGalvin and J.O’B. Hourihane
the possibility of a reaction occurring at any time, ‘because you never know what might happen’. However, always being aware and alert to the possibility of danger is a heavy burden for children in their everyday lives. Becky (age 10) explains ‘because food is always around it is hard to forget about it’ and Matt (age 11) says ‘…I can’t just eat something like my friends … or be with people without thinking about what they are eating’. Being constantly vigilant also affects children’s enjoyment of social events ‘well … it means you can never relax at a party and just enjoy it’ (Kevin, age 11). Anxiety appears to be particularly strong in children aged 9–12 years. Even when carefully following the rules, children often cannot pinpoint why a reaction occurred. Older children and teens emphasise the uncertainty of living with food allergy and the consequent feeling of a loss of control: ‘sometimes you can’t find the cause [of a reaction] … it just happens, you know … not knowing makes you worried and unsure of yourself … what can you do’ (Fran, age 15); ‘people have died … and sometimes they don’t even know why … even toothpaste has been blamed’ (Patrick, age 14). Low general awareness also contributes to uncertainty. Adolescents have a full understanding and realisation of uncertainty in their everyday lives. For example, Grace (age 13) captured the feelings of many teens when she describes why she feels anxious: ‘when I get up in the morning I can’t be sure I won’t have a reaction that day’. A growing awareness of uncertainty impinges on children’s beliefs and subsequent coping strategies. Although being vigilant allows children to feel some form of control over uncertainty, this is undermined because of low understanding and awareness in restaurants, shops, activity camps, schools, peers, etc., in the general population and difficulties in the interpretation of labels on foods. Although their roots may be discerned in children in the youngest age group, by adolescence, children’s coping strategies become more defined, in some cases more rigid, and an expanding social world gives further impetus to the search for normality. Normality has multiple meanings depending on the particular developmental trajectory of the child in question. The search for normality becomes clearer when we turn to the coping strategies children use to manage food allergy. For some, normality may mean assurance that they are safe at all times and are accepted and understood by particular friends, for others it
Reliance on parents/growing awareness of rulels/living with the rules
Specialness/growing awareness of difference/living with difference
Roots of: Maximisation/balanced adaptation/minimisation
The Search for Normality : Driver to Coping
6-8 yrs
9-11 yrs Development of: Maximisation/balanced adaptation/minimisation
12-15 yrs Certainty/growing awareness of uncertainty/living with uncertainty
Reinforcement of: Maximisation, balanced adaptiation/minimisation
Fig. 191.5 The developmental pathway model. Following axial coding, developmental differences were examined based on the Fig. 191.4 above, in order to describe a conceptual theory of the studied phenomenon: growing up, living and coping with food allergy (DunnGalvin et al. 2009)
191 Developmental Aspects of Health Related Quality of Life (HRQL)
3089
means being able to interact freely and being accepted as normal ‘in the real world’, and for some it means finding a balance between the two. The developmental model is shown in Fig. 191.5. Coping strategies were found to lie on a maximisation/avoidance to minimisation/risk continuum. They may be emotion focused or problem focused, often they are both. Some are actions, interactions or cognitions. Their defining quality is that they are used in clusters by particular children. Maximisation involves placing food allergy at the centre of your life, minimisation involves rejecting food allergy as an important part of your life and balanced adaptation involves balancing safety with integration. Emotional, cognitive and behavioural strategies are associated with each of these axial categories. Anxious children tended to use avoidant strategies to cope with living with food allergy. Many clinicians assume that these strategies are necessary and adaptive, if they are proportionate. However, we found that high levels of anxiety, vigilance and generalised avoidance of situations and people not directly related to food consumption are associated with maladaptive avoidant strategies. A surprising finding was that anxious children and teens are not necessarily those who experience the most or recent reactions. For example, many of the children who described themselves as anxious or worried about food allergy could not remember ever having had a serious reaction. Minimising strategies are also maladaptive in that children who use them also engage in risky behaviour, such as deliberately eating an allergic food. Children are told by clinicians and parents that being allergic to a certain food means that they can never have ‘even a taste’ of that allergen, any food containing the allergen or any food that may have come into contact with the allergen. Because of the difficulty for clinicians and scientists in determining risk thresholds, children must live by this rule. There were, however, important developmental differences. For example, it was noticeable that younger children described deliberately eating an allergic food in their homes or when parents were present at a social occasion, out of curiosity to experience what peers can eat without difficulties. By ‘eating just a little bit’ and seeing how they react, children in the middle age group appear to be trying to determine their own risk thresholds: ‘you’d have a small bit now and then and see what happens’ (Johnny, age 11). It may also be a way for children to exert control over uncertain conditions. Older children, and particularly teens, appeared deliberately eating an allergic food as a means of coping in a social situation in order to cope with feelings of difference. This risk behaviour developmental process is illustrated in Fig. 191.6. Parents and children share many of the same broad-based experiences, concerns and anxieties, and use many of the same coping strategies as emerged from focus groups with children and teens. Parents struggle with how to support children’s independence while controlling their own anxiety and genuine fears of increased risk. Parents respond to conditions in individual ways in a search for normality by maximization, minimization and balanced adaptation, clusters of coping strategies that were also found in children. Our findings of high levels of anxiety in a food allergic sample of children and parents are supported by epidemiological research. In child/adolescent populations with allergic diseases in general, up to one third may meet criteria for co-morbid anxiety disorders (Bender Berz et al. 2005). In adult populations with asthma, the estimated rate of panic disorder ranges from 6.5% to 24% (Katon 2004). Studies have also documented associations between anxiety disorders and allergy (Kovalenko et al. 2001). Patients attending allergy clinics reported higher levels of depression compared to the general population. A birth cohort study in Finland (Timonen 2003) revealed that at epidemiological levels, skin prick test positive females exhibited up to a 1.8-fold greater risk of developing lifetime depression when compared with skin prick test negative subjects. In addition, the corresponding risk increased up to 2.7-fold among females, who had a positive skin prick test together with self-reported allergic symptoms. Maternal atopy alone almost doubled the risk of lifetime depression in female probands when compared with families in which no maternal atopy existed. In contrast, parental atopy did not predict any type of depression in male probands.
3090
A. DunnGalvin and J.O’B. Hourihane
•Parent Protection •Specialness •Certainty
Growing awarenss of : •Rules as restrictive •Difference •Uncertainty
Adapting : •To the rules •To difference •To uncertainty
Roots of minimisation ‘I wish’
The Search for Normality
Developing minimisation ‘try it and see’
Minimisation applied to everyday events ‘having a life’
6-8 yrs
9-11 yrs
12-15 yrs
Fig. 191.6 The developmental pathway model applied to the evolution of risk behaviour. Because they were diagnosed when infants, young children feel that they are ‘the same’ as other children, and parents help them to feel normal and protected in their everyday lives. They therefore have an illusory perception of control and certainty. As children become more aware of the rules as restrictive, together with a growing awareness of difference and uncertainty, the search for normality becomes stronger and children evolve strategies in order to cope. Although their roots may be discerned in children in the youngest age group, by adolescence, children coping strategies become more defined, and in some cases more rigid, and an expanding social world gives further impetus to the search for normality (DunnGalvin and Hourihane 2009)
191.6 U nderlying Mechanisms Related to Perception of HRQL: Physiological and Psychological Responsiveness Research (animal and human) shows how the environment (including the prenatal environment) can change our anatomy, brain and central nervous system. Neuroendocrine sensitisation effects following exposure to maternal stress during the first year of life have been reported in 4.5-year-olds (e.g. Essex et al. 2002). A recent study undertaken in a sample of 10-year-old children from the Avon Longitudinal Study of Parents and Children (ALSPAC) has demonstrated for the first time a significant link between prenatal anxiety, particularly in late pregnancy, and individual differences in salivary cortisol (O’Connor et al. 2003). A relatively small study has also identified a link between maternal anxiety and salivary cortisol in children at 5 years of age. Children whose mothers exhibited higher levels of morning cortisol during pregnancy, and more fear of bearing a disabled child, showed higher levels of salivary cortisol (Gutteling et al. 2005). The same group showed similar associations in another sample of children at 4–6 years of age (Gutteling et al. 2004). Clearly, further studies are required to fully understand the relationship between PS/anxiety during pregnancy and HPA function in human children and adults. Neuroendocrine changes have also been associated with social adaptation in pre-school children (Gunnar et al. 2003). In understanding childhood influences on health and health perception, this neuroendocrine window provides an opportunity to examine these person-environment responses. Assessing cortisol activity in response to starting pre-school, Gunnar and colleagues emphasise that it is neuroendocrine adaptability that is important rather than simply the level of response. Providing evidence for the context-specific HPA activation during childhood experiences, it appears that it is the repeated triggering of the stress response rather than neuroendocrine activation itself which may be problematic for shy children, as they perceive threat to a greater number of everyday events (Watamura et al. 2004).
191 Developmental Aspects of Health Related Quality of Life (HRQL)
3091
For most children who become allergic or asthmatic, the polarization of their immune system into an atopic phenotype probably occurs during early childhood. The establishment of the Th1/Th2 balance during early childhood and the final tuning at the important middle childhood years (Sampson 2005) implies that children may be especially vulnerable to environmental and lifestyle stressors affecting this balance. This period coincides with the transition point discussed earlier, when children begin to experience uncertainty. Chronic diseases that are characterised by dysregulation of inflammation, such as food allergy, are particularly susceptible to modulation by stress and emotion. Prenatal, perinatal and early life events (e.g. stress in mothers) appear to be crucial in programming the infant’s immune system, independent of genetic susceptibility. Maternal genetics related to oxidative stress genes may influence the child’s atopic risk beginning in utero. Sustained cortisol secretion can affect Th1/Th2 differentiation both in the foetus and in the newborn infant, and is able to increase susceptibility to allergic diseases (von Hertzen 2002). Wright and colleagues (2004a) found that higher caregiver stress in the first 6 months after birth was associated with increases in the children’s allergen-specific proliferative response (a marker of the allergic immune response), higher total IgE levels and increased production of TNF-a and reduced IFN-g in a birth cohort of children predisposed to allergic disease. This risk is thought to be mediated by the effects of stress on neuroimmunoregulation, which in turn modulates hypersensitivity responses. A dysfunctional neuroendocrine-immune interface associated with abnormalities of the ‘systemic anti-inflammatory feedback’ and/or ‘hyperactivity’ of the local pro-inflammatory factors may play a role in the pathogenesis of atopic/allergic diseases and later co-morbid anxiety disorders. Chronic stress may induce a state of hyporesponsiveness of the HPA axis, whereby cortisol secretion is attenuated leading to increased secretion of inflammatory cytokines typically counterregulated by cortisol. While the ability to activate an increase in cortisol in response to some stimuli in early life may be adaptive, prolonged exposure to stress may change the cortisol response if examined at a later developmental stage (Wright et al. 2004b). Sex differences have also been noted. Different patterns of cytokine responses between males and females have been suggested as contributing in part to gender-specific differences in the self-reporting rates and perception of HRQL in food allergy, discussed earlier. We already know that components of stress and the stress response differ between men and women. The tend-and-befriend response, mediated by oxytocin and endogenous opioids, may be more applicable to women than the fight-or-flight response, which was based largely on studies of men. Even within the flight-or-flight response pattern there are sex-based differences. The HPA axis interacts with the reproductive function, such as menstruation. Further, the nature of stressors may also influence sex differences in immune reactivity to stress (Kang et al. 1997) involving a complex interaction between biology and environment. For example, there are gender differences in the types of stressors to which an individual is likely to be exposed. The complexity of these sex- and gender-based interactions may explain the more adverse effects of food allergy on female over male general emotional well-being, discussed earlier. As discussed, a number of studies have examined the impact that the activity of the stress system may have on immune activation and symptoms in humans. A state of stress-induced HPA hyporesponsiveness has also been demonstrated in research participants with chronic inflammatory disorders. Wamboldt et al. (2003) found an attenuated cortisol response among adolescents with positive skin test reactivity to an allergic disease compared with those with skin test positivity alone or nonatopic individuals. It appears that chronic stress may induce a state of hyporesponsiveness of the HPA axis, whereby cortisol secretion is attenuated leading to increased secretion of inflammatory cytokines typically counter-regulated by cortisol. Some studies have considered whether immune activation and the experience of having an atopic disease, particularly during childhood, influence the responsiveness of the HPA axis. The extant literature indicates that both physiological and psychological stressors activate similar neural circuitry, acting as two different routes to a bidirectional communication network between the brain
3092
A. DunnGalvin and J.O’B. Hourihane
and the immune system (Maier and Watkins 1998). During an immune response, the brain and the immune system communicate with each other in order to maintain homeostasis. Two major pathways, the HPA-axis and the SNS are involved in this bidirectional interaction (Elenkov et al. 2000). Consistent with this model, neural circuitry underlying stress and emotions can regulate inflammation (Black 2002) and peripheral inflammatory mediators can influence mood and cognitive function (Wichers and Maes 2003). The advent of cognitive neuroscience and functional neuroimaging has brought unprecedented new opportunities to study the neurobiology of these processes. Rosencrantz and colleagues (2005) examined the neural circuitry underlying the interaction between emotion and asthma symptom exacerbation, using fMRI during antigen challenge to examine regional brain activation in adults with mild allergic asthma. They demonstrated an association between activity in the anterior cingulated cortex (ACC) and insula to asthma-relevant emotional stimuli (e.g. wheeze) compared with valence-neutral stimuli and markers of inflammation in participants exposed to an antigen. The activation accounted for > 40% of the variance in peripheral markers. These brain regions may be hyperresponsive to disease-specific emotional and afferent physiological signals, which may contribute to the dysregulation of peripheral processes, such as inflammation (McAfoose and Baune 2009). The authors suggest that reciprocal modulation may occur between peripheral factors regulating inflammation and central neural circuitry underlying emotion and stress reactivity. This is an area of research which has the potential to provide answers to questions relating to the complex interaction between physiology and psychology in the developmental pathways of allergic diseases in males and females (DunnGalvin et al. 2006), and may have preventative and therapeutic implications in terms of both immune and anxiety disorders (DunnGalvin et al. 2009c).
191.7 Cognitive Emotional Sensitisation The theory of cognitive emotional sensitisation (Brosschot 2002) may further elucidate explanatory mechanisms for findings in the biopsychosocial impact of food allergy. This theory is based on the notion of cognitive bias, one of the best-documented cognitive phenomena in experimental psychopathology. The simplest form of plasticity in nervous systems is that repeated stimulation may lead to habituation (decreased response) or sensitization (increased response). Sensitization has been widely observed across the phylogenetic scale, and may be present at multiple levels in the organism: at the cellular, psychological and interpersonal level (see Brosschot 2002). Generally, sensitisation is caused by an increased efficiency in the synapse due to repeated use, in particular following irregular and extreme stimulation. It constitutes a feed-forward mechanism, helping the individual to react more efficiently in situations with increased probability of harm (Ursin and Eriksen 2001; Overmier 2002). Brosschot (2002) used the metaphor of an unfolding scanner or antenna, directing its dish towards the source of potentially threatening information and amplifying it. The level of background arousal is very important: high levels of arousal result in the stimulus repetition inducing sensitization, even though repetition of that same stimulus under conditions of low arousal would lead to habituation (see Overmier 2002). Cognitive emotional sensitisation is therefore a higher form of sensitisation, involving cognitive bias. Much research has shown that anxious persons have a cognitive processing priority for information that is related to their fears. The emotional Stroop results show that this multilevel sensitization could develop for many different types of concern-related information. For example, in relation to pain, Crombez (1998) emphasises two important determinants that pertain to our findings: novel pain produces a large disruption in a primary task, as does the temporal unpredictability of an aversive
191 Developmental Aspects of Health Related Quality of Life (HRQL)
3093
noxious stimulus. In an attempt to find an explanation for the lack of consistency across some attentional bias studies, Crombez also emphasizes the importance of anxiety in the manifestation of cognitive bias, and suggests analysing more precisely what type of information is really threatening for what type of patient, whether, as discussed earlier, this is a threat to safety or to self-concept. Emotional arousal tends to bring to awareness of one’s predominant cognitive bias and potentially triggers one’s most salient behavioural programmes. Anxiety in itself, unrelated to relevant health threat, may activate neural networks in children with food allergy because these networks are easily activated by any stimulus that induces limbic activity (Ursin 2001). Thus, children need not have actually, or recently, experienced an allergic reaction in order to feel anxious. The type of threat, in addition to emotional arousal, is also very important; for example, if a child perceives that revealing that he/she is food allergic will have an impact on his/her ability to fit in with peers, this may result in risky behaviour rather than avoidant behaviour. Finally, each time anxiety is felt and the networks are triggered, arousal probably helps to further strengthen the associations in this network. Emotional arousal at the time of experiencing a stimulus (internal as well as interaction with environment) is thought to be critical in influencing memory strength for this stimulus and therefore consolidation of the strong fear network and its sensitization (Brosschot 2002). The amygdala modulates the establishment of memory traces in other brain regions (e.g. the hippocampus; see McCaugh 2004). Although much research attributes a specialised role for harm avoidance to the amygdala circuits (see review LeDoux and Phelps 2000), and for reward processing to the nucleus accumbens (DiChiara et al. 2004), these structures support a number of additional functions, such as associative learning and attention filtering which cut across both appetitive and aversive processing. Associative learning is affected by the way in which feedback is processed, that is, the representation of the value of the outcome that becomes linked through learning to the stimuli options (Ernst et al. 2003). The learning itself may be fully developed by middle to late childhood and behaviour expressed as highly trained (and probably neurally pre-activated), easily triggered behavioural patterns or, at the neural level, sensitized motor programs, coupled to the corresponding highly activated cognitive (and emotional) network (Brosschot 2002; Ursin and Eriksen 2001).Finally, increasing coordination and integration of the mostly physiological regulatory systems (e.g. information processing) over the course of development means that by the time children reach their teens, self-perception, emotional reactions and cognitive appraisal mechanisms have become relatively stable and consistent. Therefore, perception of HRQL and attendant behaviours may develop and depend on many factors, such as the amount and type of subjective threat, parental risk perception, peer and general attitudes to the disease, social environment, sex, age of development and culture (including gender).
191.8 Applications to Other Areas of Health and Disease In this chapter, using food allergy as an example, we reviewed relevant literature to argue for a broader evaluation of HRQL in order to provide a framework for the construction of appropriate models linking symptoms, functioning, development, underlying physiological mechanisms and HRQOL in chronic disease involving food in children. The research presented here implies that studies in health related quality of life that take account of biobehavioural developmental subsystems are likely to be more informative than those restricted to physiological or psychological domains alone. Developing functional systems interact, react with and program one another. For example, early life origins research has delineated mechanisms linking psychological stress, personality and emotion to neuro-immunoregulation as well as
3094
A. DunnGalvin and J.O’B. Hourihane
increased risk of atopy. Furthermore, it appears that immune activation and the experience of having an allergic disease, particularly during childhood, influences the responsiveness of the HPA axis. The implications of this developing field of research are enormous with applications extending to include the effect of social experience on learning and on health outcomes. Social and physiological adaptation is required in order to cope with transitional life experiences, whether these are due to early maternal separation in the form of childcare, starting school, the experience of competition or a particular disease. A chronic condition may affect and/or interact with normative developmental pathways and risk of maladaptive outcomes may follow transition points that are particular to a specific disease. Disease-specific pathways may be embedded within developmental trajectories and predict phenotypic expression of physiology, psychology and behaviour. A simple illustration of this dual pathway is shown in Fig. 191.7. Findings may also be applicable to other chronic diseases involving diet such as diabetes and celiac. Diabetes, like food allergy, is a ‘hidden’ disease, largely unsymptomatic but characterised by sudden and unpredictable symptomatic events. In a series of focus groups with children aged 6–12 with diabetes (submitted), we found many of the same transition points (e.g. balancing safety with the need to ‘belong’); however, there were also differences. In addition to a growing awareness of ‘uncertainty’ in Typical developmental pathway Pre-and Ante-natal Impact (interaction between parent and child physiology)
Disease developmental pathway
Transition Point
Infancy and early childhood
Entry to formal schooling
Diagnosis
Parental protection
Transition Point
Precipitating events (stressful events in the children’s lives caused by developmental & disease related factors)
Psychological/Physiological impact (e.g.cognitive appraisal and emotional effects) Environment: sensitisation & elicitation.
Biology : sensitisation & elicitation
(e.g. nutrition,. stress, gender, parent/peer attitudes)
Integration of disease identity
Anxiety/Avoidance
(e.g. immune mechanisms, neurochemistry, sex )
Adolescence
Transition Point
Rejection of disease identity
Minimisation/Risky behaviour
Precipitating events (stressful events in the teens lives caused by developmental & disease related factors)
Habitual Cognitive/Emotional appraisal mechanisms
Reinforcement of Disease Identity
Habitual response patterns (physiological, psychological, behavioural)
Fig. 191.7 Developmental framework model illustrating the integration of food allergy-specific pathway within the normal trajectory. A chronic condition may affect and/or interact with normative developmental pathways and risk of maladaptive outcomes may follow transition points that are particular to a specific disease. Disease-specific pathways may be embedded within developmental trajectories and predict phenotypic expression of physiology, psychology and behaviour (Dunn Galvin and Hourihane in press)
191 Developmental Aspects of Health Related Quality of Life (HRQL)
3095
managing diabetes, there was a growing awareness of ‘embarrassment’ and a public attitude of ‘blame’ toward the diabetic adolescent. This awareness resulted in concealment and risky behaviour in some cases. Research in diabetes shows that adolescence, when responsibility for self-care is largely transferred from parent to child, is consistently associated with a decline in metabolic control (e.g. Hoey et al. 1999). While the decline is partly attributed to physiological changes the main cause is a substantial decline in levels of self-care (e.g. Morris et al. 1997). Anxiety and depression associated with uncertainty has also been noted in childhood diabetes (Mullins et al. 2007). Further, the need to address uncertainty in the context of its differing gender impact may also be discerned from research into other ‘hidden’ diseases. Austin et al. (2000), in a 4-year study on adjustment in epilepsy, found that girls evolved more behaviour problems from pre- to post-adolescence. In addition, they found that children were very worried about the possibility and unpredictable nature of seizures.
191.9 Conclusion An integrated developmental perspective provides a powerful place to illuminate our understanding of individual differences in the expression and impact of chronic diseases. For example, research into the behavioural significance of the different trajectories of biopsychosocial maturation can aid in the development of psychoneurobiological models that may ultimately predict health-related quality of life outcomes. A HRQL research infrastructure that has sufficient scope (both breadth and depth) will still fail in its goal if the knowledge is not linked across levels and domains. Four elements are essential to success in integration. The first is the development of a coherent conceptual framework within which the connections make sense. The second involves a cross-disciplinary research project in order to address key issues of linkage between developmental processes and outcomes at different levels. The third element relates to the need for innovative quantitative and qualitative methodologies which can be used to tease out the relative contribution of biopsychosocial factors to HRQL. The last relates to the need for longitudinal birth cohort studies. A broader understanding of HRQL will ultimately lead to the promotion of earlier, more effective preventive strategies and interventions focused on maximizing optimal health development and quality of life. Summary Points • Outcomes research has been key in altering the culture of clinical practice and health care research by changing how we assess the end results of health care services. • Health Related Quality of Life (HRQL) is a multi-dimensional construct, which evaluates physical, psychological, and social components, from the patient perspective. • There are important questions that must be answered before HRQL measures can reach their full potential in research, practice and policy. • Disease- specific pathways may be embedded within normal developmental trajectories and predict phenotypic expression of physiology, psychology and behaviour. • The immune system in concert with psychological factors, such as stress, may play a role in the development of abnormal immune and allergic responses. In turn, immune activation and the experience of having an atopic disease, particularly during childhood, may influence the responsiveness of the HPA axis.
3096
A. DunnGalvin and J.O’B. Hourihane
• Appropriate pathway models must be developed to incorporate broader aspects linked to HRQL and to incorporate developmental biopsychosocial trajectories. • An integrated developmental perspective may promote increased understanding of individual differences in the expression and impact of many chronic diseases, with applications extending to include the effect of social experience on learning and on health outcomes. • A broader understanding of HRQL may lead to the promotion of earlier, more effective preventive strategies and interventions focused on maximizing optimal health development and quality of life. Key Terms Health related quality of life: A multidimensional construct, which evaluates physical, psychological and social components that may be impacted by a disease or medical condition, from the patient perspective. Reliability: The consistency of a measurement, or the degree to which an instrument measures the same way each time it is used under the same condition with the same subjects. Validity: There are several forms of validity, but in essence, validity refers to the ability of a measure to capture the construct of interest (e.g. health related quality of life) in a meaningful and effective manner. Developmental trajectory: A lifelong process of developmental integration that involves complex interactions between biological and environmental factors that influence the phenotypic expression of physiology, psychology and behaviour. Sensitive transition point: Period in development when physiological or environmental variables associated with a particular disease may have a relatively greater impact and/or interact with already existing normative demands and changes in socialization. Coping/adaptation: How individuals mobilize, guide, manage, energise and direct behaviour, emotion or orientation, or how they fail to do so, under stressful conditions, such those associated with a chronic disease. Sensitisation/cognitive emotional sensitisation: Sensitisation is caused by an increased efficiency in the synapse due to repeated use, in particular following irregular and extreme stimulation. Cognitive emotional sensitisation is a higher form of sensitisation, involving cognitive bias.
Acknowledgement This work was funded by the EU through the EuroPrevall project (FOOD-CT-2005-514000).
References Austin JK, Dunn DW, Huster GA. Childhood epilepsy and asthma: changes in behavior problems related to gender and change in condition severity. Epilepsia. 2000;41:615–23. Avery NJ, King RM, Knight S, Hourihane JOB. Assessment of quality of life in children with peanut allergy. Pediatr Allergy Immunol. 2003;14:378–82. Black PH. Brain Behav Immun. 2002;16:622–53. Brosschot JF. Cognitive–emotional sensitization and somatic health complaints. Scand J Psychol. 2002;43 (2):113–21. Bollinger ME, Dahlquist LM, Mudd K, Sonntag C, Dillinger L, McKenna K. The impact of food allergy on the daily activities of children and their families. Ann Allergy Asthma Immunol. 2006;96:415–21 Charmaz K. Grounded theory: objectivist and constructivist methods. In Denzin N, Lincoln Y, editors. Handbook of qualitative research. 2nd edn. Thousand Oaks: Sage; 2000. p. 509–535. Cohen BL, Noone NS, Munoz-Furlong A, et al. Parental burden in food allergy. Journal of Allergy Clin Immunol. 2004;114 5:1159–63.
191 Developmental Aspects of Health Related Quality of Life (HRQL)
3097
DiChiara G, Bassareo V, Fenu S, DeLuca MA, Spina L, Cadoni C, Acquas E, Carboni E, Valentini V, Lecca D. Dopamine and drug addiction: the nucleus accumbens shellconnection. Neuropharmacology. 2004;47:227–41. DunnGalvin A, Hourihane J O’B, Frewer L, Knibb RC, Oude Elberink JNG, Klinge, I. Incorporating a gender dimension in food allergy research: a review. Allergy. 2006;61:1336–43. DunnGalvin A, de Blok BMJ. Dubois A, Hourihane JO’B. Development and Validation of the Food Allergy Quality of Life – Parent Administered Questionnaire (FAQLQ-PF) for food allergic children aged 0-12 years. Clinical and Experimental Allergy. 2008;38:977–86. DunnGalvin A, Dubois AEJ, Hourihane J. Child vs maternal perception of HRQoL in food allergy: developmental trajectories and evolution of risk behaviour. Allergy. 2007;62 (Suppl. 83):70–166. DunnGalvin A, Roberts S, Hourihane JO’B. Causal pathway analysis of mothers’ responses to government advice on peanut avoidance. Oxford Journal of Public Health. 2008a;30(1):114–5. DunnGalvin A, Gaffney A, Hourihane JO’B. Developmental pathways in food allergy: a new theoretical model. Allergy. 2009a;64:560–8. DunnGalvin A, Hourihane JO’B. Self-assessment of reaction thresholds in food allergy: a new theory of risk taking which changes over time. J Allergy Clin Immunol. 2009b;123:S142. DunnGalvin A, Hourihane JO’B. Developmental trajectories in allergic diseases: A review. Advances in Food and Nutrition Research, Elsevier Inc. 2009c. Volume 56 (pp. 65–102). DunnGalvin A, Burks WJ, Dubois AEJ, Chang WC, Hourihane JO’B. Profiling Families Enrolled in Food Allergy Immunotherapy Studies DunnGalvin A, Chang WC, Laubach S, Steele PH, Dubois AEJ, Wesley Burks AW, Hourihane JO’B, Pediatrics 2009d;124:e503–e509. DunnGalvin A, Cullinane C, Daly D, Flokstra-de Blok BMJDubois AEJ, Hourihane JO’B..Longitudinal validity and responsiveness of the Food Allergy Quality of Life Questionnaire – Parent Form (FAQLQ-PF) in children 0-12 years following positive and negative food challenges. Clin Exp Allergy. 2010;40(3):476–85 DunnGalvin A, Hourihane JO’B. (submitted) A Chronic Disease Impacts Directly on Child’s Normal Trajectory of Psychological Development in Both a Generic and Disease-Specific Manner: Similarities and Differences in Food Allergy and Diabetes. Elenkov IJ, Wilder RL, Chrousos GP, Vizi ES. The sympathetic nerve – an integrative interface between two supersystems: The brain and the immune system. Pharmacol Rev. 2000;52:595–638. Engle G. The need for a new medical model: a challenge for biomedicine. Science 1977;197:129–96. Feeney JA. Implications of attachment style for patterns of health and illness. Child Care Health Dev. 2000;26: 277–88. Ferrans CE, Zerwic JJ, Wilbur JE, Larson JL. Conceptual model of health related quality of life. J Nursing Scholarship. 2005;27:336–42. Flokstra-de Blok BMJ, DunnGalvin A, Vlieg-Boersta BJ, Oude Elberink JNG, Duiverman EJ, Hourihane JO, Dubois AEJ. Development and validation of a self-administered Food Allergy Quality of Life Questionnaire for children. Clin Exp Allergy. 2009;39:127–37. Flokstra-de Blok BMJ, DunnGalvin A, Vlieg-Boerstra BJ, Oude Elberink JNG, Duiverman EJ, Hourihane JO, Dubois AEJ. Development and validation of the self-administered Food Allergy Quality of Life Questionnaire for adolescents. J Allergy Clin Immunol. 2008;122:139–44. Flokstra-de Blok BMJ, van der Meulen GN, DunnGalvin A, Vlieg-Boerstra BJ, Oude Elberink JNG, Duiverman EJ, Hourihane JO, Dubois AEJ. Development and validation of the first disease-specific quality of life questionnaire for adults; The Food Allergy Quality of Life Questionnaire-Adult Form (FAQLQ-AF). Allergy 2009;64(8):1209–17. Gutteling BM, de Weerth C, Buitelaar JK Maternal prenatal stress and 4–6 year old children’s salivary cortisol concentrations pre- and post-vaccination. Stress. 2004;7:257–60. Gutteling BM, de Weerth C, Buitelaar JK. Prenatal stress and children’s cortisol reaction to the first day of school. Psychoneuroendocrinology. 2005;30:541–9. Gunnar M, Sebanc A, Tout K, Donzella MA, van Dulman M. Temperament, peer relationships, and cortisol activity in preschoolers. Dev Psychbiol. 2003;43:346–68. Halfon N, Hochstein M. Life course health development: An integrated framework for developing health, policy and research. In: Milbank Q.80:433–79; 2002. Hoey, H, Mortensen, H, McGee, H, & Fitzgerald, M, for the Hvidore Study Group. Is metabolic control related to quality of life? A study of 2103 children and adolescents with IDDM from 17 countries. Diab Res Clin Pract. 1999;44 (Suppl):S3. Hourihane JO, Roberts SA, Warner JO. Resolution of peanut allergy: case-control study BMJ. 25 Apr 1998a;316 (7140):1271–5. Hourihane JO. Prevalence and severity of food allergy – need for control. Allergy. 1998b;53 (46 Suppl):84–8. Hourihane JOB, Flokstra-de Blok BMJ, Dubois AEJ, Kahlon R, Dunn Galvin A. Food challenge has a rapidly established and persistent positive effect on quality of life of children 0–12 years irrespective of the clinical outcome of the challenge. J Allergy Clin Immunol. 2009;123:S143. Johansson SGO, Bieber T, Dahl R. Revised nomenclature for allergy for global use: report of the Nomenclature Review Committee of the World Allergy Organisation. J Allergy Clin Immunol. 2004;5:832–6.
3098
A. DunnGalvin and J.O’B. Hourihane
Katon W. Clinical and health services relationships between major depression, depressive symptoms, and general medical illness. Biol Psychiatry. 2004;54:216–26. Kang DH, Coe CL, McCarthy DO, Ershler WB. Immune responses to final exams in healthy and asthmatic adolescents. Nurs Res. 1997;46:12–9. Kovalenko PA, Hoven CW, Wu P, Wicks J, Mandell DJ, Tiet Q. Association between allergy and anxiety disorders in youth. Aust New Zealand J Psychiatry. 2001;35: 815–21. Kyngas H, Hentinen M. Meaning attached to compliance with self-care, and conditions for compliance among young diabetics. J Adv Nursing. 1995;21:729–36. LeDoux JE, Phelps EA. Emotional networks in the brain. In Lewis, M, Haviland J, editors. Handbook of emotions. New York: Guilford Press; 2000. pp. 134–145. Maier SF, Watkins LR. Stressor controllability, anxiety, and serotonin. Cognit Ther Res 1998;22:595–613. Marklund B, Ahlstead, Nordstrom G. Health-related quality of life among adolescents with allergy-like conditions – with emphasis on food hypersensitivity. Health and Quality of Life Outcomes 2004;2:65–9. McAfoose J, Baune BT. Evidence for a cytokine model of cognitive function. Neurosci Biobehav Rev. 2009;33:355–66. McCaugh JL. The amygdala modulates the consolidation of memories of emotionally arousing experiences. Annu Rev Neurosci. 2004;27:1–28. Mullins LL, Wolfe-Christensen C, Hoff Pai AL, Carpentier MY, Gillaspy S, Cheek J, Page M. The relationship of parental overprotection, perceived child vulnerability, and parenting stress to uncertainty in youth with chronic illness. J Paediatric Psychol 2007;32:973–82 Morris AD, Boyle DI, McMahon AD, Greene SA, MacDonald TM, Newton RW. Adherence to insulin treatment, glycaemic control and ketoacidosis in insulin-dependent diabetes mellitus. The DARTS/MEMO Collaboration. Diabetes Audit and Research in Tayside Scotland. Medicines Monitoring Unit. Lancet. 1997;350:1505–10. O’Connor TG, Heron J, Golding J, Glover V, the ALSPAC study team. maternal antenatal anxiety and behavioural/ emotional problems in children: a test of a programming hypothesis. J Child Psychol Psychiat 2003;44:1025–36. Overmier JB. Sensitization, conditioning, and learning: can they help us understand somatization and disability? Scand J Psychol. 2002;43 (2):105–12. Primeau MN, Kagan R, Joseph L, Lim H, Dufresne C, Duffy C, Prhcal D, Clarke A. The psychological burden of peanut allergy as perceived by adults with peanut allergy and the parents of peanut-allergic children. Clin Exp Allergy. 2000;30:113. Rosenkranz MA, Busse WW, Johnstone T, Swenson CA, Crisalfi GM, Jackson MM et al. Neural circuitry underlying the interaction between emotion and asthma symptom exacerbation. Proc Natl Acad Sci. 2005;102:13319–24. Sampson HA. Food allergies. In M Feldman et al., editors. Sleisenger and Fordtran’s gastrointestinal and liver disease. 8th ed., vol. 1. Philadelphia: Saunders; 2005. pp. 427–439. Schmidt S. Coping with chronic disease from the perspective of children and adolescents – a conceptual framework and its implications for participation. Child Care Health Dev. 2003;29 1: 63–75. Sicherer SH. Food allergy. Lancet. 2002;360:701–10. Timonen M, Jokelainen J, Hakko H, Silvennoinen-Kassinen S, Meyer-Rochow VB, Herva A Räsänen P .Atopy and depression: results from the Northern Finland 1966 Birth Cohort Study.Mol Psychiatry. 2003;8(8):738–44. Ursin H, Eriksen HR. Sensitization, subjective health complaints, and sustained arousal. In: Sorg B, Bell I, editors. The role of neural plasticity in chemical intolerance. Ann N Y Acad Science 933:119–29; 2001. von Hertzen LC. Maternal stress and T-cell differentiation of the developing immune system: possible implications for the development of asthma and atopy. J Allergy Clin Immunol. 2002;109 (6):923–8. Wamboldt MZ, Laudenslager M, Wamboldt FS, et al. Adolescents with atopic disorders have an attenuated cortisol response to laboratory stress. J Allergy Clin Immunol. 2003;111:509–14. Watamura SE, Donzella B, Kertes DA, Gunnar MR. Developmental changes in baseline cortisol activity in early childhood: relations with napping and effortful control. Dev Psychobiol. 2004;45:125–133. Wilson IB, Cleary PD. Linking clinical variables with health-related quality of life. A conceptual model of patient outcomes. JAMA. 1995;273:59–65. Wright RJ, Finn P, Contreras JP, Cohen S, Wright RO, Staudenmayer J, et al. Chronic caregiver stress and IgE expression, allergen-induced proliferation, and cytokine profiles in a birth cohort predisposed to atopy. J Allergy Clin Immunol. 2004a;113 (6):1051–7. Wright RJ, Cohen RT, Cohen S. The impact of stress on the development and expression of atopy. Curr Opin Allergy Clin Immunol 2004b;5:23–9. Zupanic JAF, Gillie P, Streiner DL, Watts JL, Schmidt B. Determinants of parental authorisation for involvement of newborn infants in clinical trials. Pediatrics. 1997;99:213–6.
Chapter 192
Nutrition and Quality of Life in Older People Salah Gariballa
Abbreviations BMI BMR CCK DHSS EAR FAO HRQol PEU RDA tHcy WHO
Body mass index Basal Metabolic Rate Cholecystokinin Department of Health and Social Security Estimated Average Daily Requirements Food and Agriculture Organisation Health-related quality of life Protein-Energy Undernutrition Recommended dietary allowances Total plasma homocysteine World Health Organisation
192.1 Introduction The number of older people is growing rapidly worldwide and looks set to continue to increase further in the future. For example by 2025, one-tenth of the world’s population will be aged 65 or older and Asia will see the proportion of its elderly population almost double, from about 6% in 2000 to 10% in 2025. In absolute terms, this represents a stark increase in just 25 years from about 216 millions to about 480 million older people. This has created a need for additional knowledge of agerelated changes relevant to nutrition, which has importance in the treatment and prevention of disease, and in maintaining good health and quality of life (QoL) in an ageing population. It is well recognized that with advancing age, there is a high incidence of chronic diseases, and evidence points to the importance of nutrition in the development, susceptibility and outcome of these diseases. There is no doubt that good nutrition contributes to the health and well-being of elderly people and to their ability to recover from illness (Fig. 192.1).
S. Gariballa (*) Department of Internal Medicine, Faculty of Medicine and Health Sciences, UAE University, Al-Ain, United Arab Emirates e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_192, © Springer Science+Business Media, LLC 2011
3099
3100
S. Gariballa
Trends in Aging, by World Region Population Ages 65 and Older Percent
21
14 11
10
7
6 3
World
6
4
Africa
10
Asia
2000
Latin America and the Caribbean
More Developed Regions
2025
Source: United Nations, World Population Prospects: The 2004 Revision (medium scenario), 2005. © 2006 Population Reference Bureau
Fig. 192.1 Trends in aging by world region (www.prb.org/presentations/j_trends-aging)
192.2 A ge-Related Physiological and Pathological Changes Relevant to Nutrition Ageing in man may be accompanied by changes, which may impair the search for food and its subsequent intake, but such changes are complex and difficult to document. Anorexia and weight loss are common and important clinical problems in the elderly; and the causes are multifactorial. There is a growing recognition that age-related physiological anorexia may predispose to protein-energy undernutrition (PEU), in the elderly particularly in the presence of other ‘pathological’ factors associated with ageing; such as social, psychological, physical and medical factors, the majority of which are responsive to treatment.
192.2.1 Physiological Changes 192.2.1.1 Hormonal Changes The potential mechanisms of physiological anorexia of ageing are however, poorly understood but they are the focus of recent research. Current evidence suggests that a combination of reduced sensory perception within the gastrointestinal tract, a decline in opioid modulation of feeding, particularly in older women, and an increase in the satiating effects of Cholecystokinin (CCK) contribute to this anorexia. For example, CCK, the best characterized of the gastrointestinal hormones is known
192 Nutrition and Quality of Life in Older People
3101
to play a role in the control of food intake. There is evidence that sensitivity to the satiating effects of CCK increases with age. The combination of increased circulating CCK concentrations and enhanced sensitivity to the satiating effects of CCK in older people suggests that CCK may be a significant contributor to the anorexia of ageing. With age, the time taken for the emptying of the stomach after large volumes of food is altered, this affects satiation. This may explain why older adults feel a greater satiating effect after an average meal compared to younger ones. Other hormones (leptin), neurotransmitters (opioids & nitric oxide) and protein (cytokines) may also have a role to play in anorexia and weight loss of ageing.
192.2.1.2 Gastrointestinal Tract Objective changes in smell and taste have been observed, which may directly decrease food intake or alter the type of foods, which are selected. With ageing, there may be a progressive loss in a number of taste buds per papilla on the tongue. The remaining taste buds, which detect primarily bitter or sour tastes, show a relative increase with ageing. In addition, the ability to identify foods while blindfolded decreases with advancing age. This is a common perceived problem among elderly individuals who complain of loss of both taste and smell. Impaired appetite is often associated with reduction in taste and smell, which occur in up to 50% of elderly people. Taste thresholds are higher among institutionalized than in healthy elderly men and the use of drugs, particularly antihypertensive medication, appears to be a contributing factor. There are some documented gastrointestinal changes in the elderly, which could affect their food intake. For example changes in peristaltic activity of the oesophagus, which may result in delay of oesophageal emptying. Absorption of some nutrients, in particular vitamin B12 may be impaired because of mild ageing-related achlorhyria, but the evidence here is incomplete. Some researchers have reported widespread nutritional deficiencies associated with bacterial contamination of the small bowel. Others have reported a significant improvement in nutritional status in elderly patients after treatment of bacterial contamination with antibiotics. Stronger evidence point to no association between bacterial colonization of the bowel and nutritional status. The most likely interpretation of these apparently conflicting reports is that bacterial contamination of an anatomically normal small bowel in the elderly is the result rather than the cause of malnutrition. The mechanisms through which malnutrition might cause bacterial growth are not fully understood but there is evidence that the activity of several enzyme systems involved in bactericidal processes may be reduced in malnutrition.
192.2.1.3 Body Mass and Composition Changes in body composition seen with ageing includes a decrease in lean body mass and an increase in body fat. Decreased physical activity accounts for the increased body fat and this may lead to decreased energy or calorie intake with ageing. These changes in body composition including those in fat distribution may be associated with changes in various physiological functions that affect metabolism, nutrient intake, physical activity and risk for chronic disease.
192.2.1.4 Energy Requirement To date, the scientific evidence about energy requirement in the elderly is often incomplete and highly variable. The reasons for this include: paucity and variability of data on energy intake and
3102
S. Gariballa
requirements; and most important of all, diversity of physical activity patterns in the elderly population. In a series of studies, elderly subjects from the USA consumed on average more energy than subjects in European studies, however the USA trials included less people compared to the European studies. The DHSS longitudinal study, which examined energy intake in 365 elderly people in 1967/1968 and 5 years later found that the average energy intake had fallen from 2,235 to 2,151 kcal per day for men and from 1,711 to 1,636 kcal per day for women. A similar trend for energy intakes to fall with age over 5 years was observed in a study of 269 elderly people in Gothenberg, Sweden.
192.2.2 Energy Expenditure 192.2.2.1 Basal Metabolic Rate (BMR) BMR reflects the energy requirements for maintenance of the intracellular environment and the mechanical processes such as respiration and cardiac function that sustain the body at rest. It usually accounts for between 60% and 75% of total energy expenditure. The FAO/WHO/UNU Expert Consultation (1985), used equations to predict BMR. These equations may be less appropriate for the elderly populations, especially older men because of small numbers in the study, since more data have been collected which allowed a more precise estimate of current energy requirement in the elderly. BMR increases with body size, particularly with lean body mass and this explains why it is higher in men than women; and 10–20% less in old people because of reduced muscle mass and increased fat mass with ageing.
192.2.2.2 Physical Activity In most working populations, physical activity accounts for 10–35% of total energy expenditure. The energy expenditure of different activities depend on the amount of work being carried out, the weight of the individual and the efficiency with which that work is carried out. In general, ageing is associated with a reduction in efficiency, which may make standard tasks like walking expend up to 20% more energy in older people. This reduced efficiency may be one reason why older individuals slow down. It may be contributing to negative energy balance, weight loss and undernutrition in some settings.
192.2.2.3 Thermogenesis The term Thermogenesis encompasses a wide variety of phenomena which include energy expenditure and heat generation associated with feeding, body temperature maintenance and thermogenic response to various specific stimuli such as smoking, caffeine, and drugs. thermogenesis has also been postulated to play a part in the regulation of body weight. This field of research is complex in humans, and the theory is derived mainly from animal models. In the elderly, resting circulating catecholamine concentrations are elevated, and the responsiveness to catecholamines may decline with age, as is the case in experimental animals. Thermic response to meal ingestion in human appears to be influenced by age, physical activity and body composition. It is possible that the fall in the capacity for thermogenesis with age may explain the increased risk of hypothermia in the elderly. However, in most cases of hypothermia there is a precipitating physical cause such as stroke, which may or may not have a direct effect on thermogenesis.
192 Nutrition and Quality of Life in Older People
3103
192.2.2.4 Protein Requirement There is almost a consensus regarding the current recommendation for protein intake of free living healthy elderly, which is between 0.75–0.8 g/kg. Total protein contained in lean body mass falls with age and protein synthesis, turnover, and breakdown all decrease with advancing age. Based on series of studies and literature review in 1980, Munro and Young stated that progressive loss of protein is a major feature of ageing throughout adult life. This appears to affect some tissues, notably muscle, more than others. There is no direct evidence to suggest that this erosion of tissue protein is due to lack of adequate amounts of protein in the average diet. Ill health, trauma, sepsis and immobilization may upset the equilibrium between protein synthesis and degradation. A group of researchers studied the dietary protein requirements of elderly men and women aged 56–80 years using short-term nitrogen balance techniques and calculations recommended by the 1985 Joint FAO/WHO/UNU Expert consultation. They have also recalculated nitrogen-balance data from three previous protein requirement studies in elderly people. From the current and retrospective data they reported that a safe protein intake for elderly adults would be 1.0–1.25 g/kg/day.
192.2.2.5 Vitamins Because of low food intake and increased incidence of physical diseases, which may interfere with intake, absorption, metabolism and utilization, vitamin deficiency is more likely in the elderly compared to the young. Intake of most vitamins is reduced in smokers and alcoholics are more likely to suffer from folate and thiamine deficiency. Up to 50% of elderly in the surveyed populations ingests vitamin supplements even though there is no documented benefit from this practise when the diet is adequate. Some studies showed that multivitamin supplement on elderly patients produced significant clinical benefits. However, most other studies, which examined vitamin supplementation often, showed no consistent statistically significant difference between supplement and placebo administration. Two large surveys of vitamin status in elderly people within the past 12 years have improved knowledge of this subject: The Boston Nutritional Status Survey and the Survey in Europe on Nutrition and the Elderly. Current evidence including the SENECA and the Boston surveys on vitamin requirements of elderly people with reference to the National Research Council recommended dietary allowances (RDA) is that “there are data to indicate that the 1989 RDAs are too low for the elderly population (i.e., ³ 51 years) for riboflavin, vitamin B6, vitamin D and vitamin B12 – at least for certain groups of elderly people. The present RDAs for elderly people appear to be appropriate for thiamin, vitamin C and folate, but are probably too high for vitamin A. There are not enough data to make judgement on the appropriateness of the RDAs, or safe and adequate intakes for elderly people for vitamin K, niacin, biotin and pantothenic acid”.
192.2.2.6 B-group Vitamins and Homocysteine B vitamins including folate and vitamins B2, B6, and B12, are major determinants of homocysteine metabolism and plasma tHcy concentration and hyperhomocysteinaemia is associated with cardiovascular, mental and bone health. The National Diet and Nutrition Survey of older people in the UK reported low biochemical status of one or more micronutrients in 40% of the older population, including B vitamins. Although several case-control and prospective cohort studies showed associations between modest elevation of plasma total homocysteine and cardiovascular disease, including stroke, only recently that sufficient evidence has mounted to suggest that the association is independent
3104
S. Gariballa
and may be dose related. Several observations suggest that homocysteine have a role in the development of vascular diseases. For example, homocysteine may promote the oxidation of low-density lipoprotein cholesterol, vascular smooth muscle cell proliferation, platelet and coagulation factors activation and endothelial dysfunction. The prevalence of hyperhomocysteinaemia in the general population is between 5% and 10%. However, rates may be as high as 30% to 40% in the elderly population. Homocysteine may therefore, represent an important and potentially modifiable risk factor for cardiovascular diseases in the elderly. Causes of moderate hyperhomocysteinaemia which are all more common in the elderly include nutritional deficiencies of folic and B-group vitamins (B12, B6 and B2), renal impairment, hypothyroidism, malignancies (acute lymphoblastic leukaemia, carcinoma of the breast, ovary and pancreas), medications/toxins (folate antagonists such as methotrexate, phenytoin, carbamazepine and vitamin B6 antagonists such as theophylline, azarabine, oestrogen-containing oral contraceptives, cigarette smoking), severe psoriasis, genetic defects in homocysteine metabolism and ageing per se. In a prospective study of 2,127 men and 2,639 women aged 65–67 years in 1992–1993 from Hordaland County, Norway, 162 men and 97 women died during a median 4.1 years of follow-up. The association between mortality and plasma total homocysteine (tHcy), distributed by quintiles with use of those with a concentration < 9.0 µmol/L as the referent group, was highly significant for both nonvascular and cardiovascular causes of death. An increase in tHcy of 5 µmol/L was associated with a 49% increase in all-cause mortality, a 50% increase in cardiovascular mortality, a 26% increase in cancer mortality and a 104% increase in noncancer, noncardiovascular mortality. Thus, plasma tHcy was a strong predictor of both cardiovascular and noncardiovascular mortality. A number of recently completed randomized trials on B vitamin homocysteine lowering and risk of stroke do not provide clear evidence of any beneficial effect, although in one trial fewer patients assigned to active treatment than to placebo had a stroke. There are many ongoing prospective, controlled intervention trials using folate, vitamin B12 and vitamin B6 as homocysteine-lowering agents, the results of which (plus future metanalyses) may provide important information as to whether these vitamins can be protective against cardiovascular diseases including stroke. However, even if homocysteine-lowering therapies prove to be effective it still does not clear up whether the beneficial effect can be ascribed to a reduction in homocysteine or to an independent effect of the B-vitamins themselves. A recent cross-sectional study from Scotland in the UK tested the association between cognitive performance and plasma vitamin B12, folate and homocysteine in community-dwelling elderly. The authors used several cognitive tests in 2 cohorts, one aged 63 years and the other aged 78 years. They reported that concentrations of folate and B12 were positively associated with cognitive ability after controlling for childhood IQ. Cognitive function was inversely related to plasma homocysteine concentrations. A new hypothesis of the link between high levels of homocysteine and depressed mood is emerging. A plausible explanation for the association is that high homocysteine levels cause cerebrovascular disease and neurotransmitter deficiency, which cause depression of mood. Interventional studies would be needed to test this hypothesis.
192.2.2.7 Antioxidants There are many theories on the ageing process. One important theory is that accumulation of oxygen-free radicals over the years leads to cumulative damage to cellular structure and function and consequently physical changes of ageing. Since then, there has been much interest in the role of antioxidants on the ageing processes. There are several reasons why consumption of fruit and vegetables merits special attention. Besides contributing to non-starch polysaccharides, they are rich sources of vitamins and minerals
192 Nutrition and Quality of Life in Older People
3105
such as carotene, vitamins A, E, C and potassium. Several of these micronutrients have antioxidant properties and they may have a role in protecting against oxidative free radicals, which may be involved in the mechanism of atherosclerotic injury and ageing. Evidence is also accumulating to show that free radical damage may be important in other diseases such Parkinson’s disease, Alzheimer’s disease, chronic inflammatory disease and cancer and that some of diseases (cardiovascular and cancer) may be prevented or delayed to some extent by dietary changes such as reduction in fat intake and increased consumption of fruits, grains and vegetables.
192.2.2.8 Trace Elements Knowledge of the exact role and dietary requirements for some of the following minerals (Cobalt, Copper, Chromium, Flouride, Iodine, Manganese, Molybdenum and Selenium) is incomplete for three reasons: they have only recently been found to be essential; dietary deficiencies of many are unknown; and the utilization of one may be affected by the amount of other elements present. However, for some there are recommended dietary intakes, which may be adequate and safe, but their optimum intakes are unknown.
192.3 Pathological Changes 192.3.1 Medical and Social Factors Risk factors for undernutrition amongst elderly people in the community includes: isolation with an inability to go out shopping, loss of spouse, depression and bereavement, decreased mobility, dementia, anorexia due to disease especially cancer, medications, poor dentition, alcoholism and most important of all acute illness. In institutions, lack of supervision and assistance at mealtimes may be an important factor resulting in poor food intake (Table 192.1). Because old people are disproportionately isolated, on low income or disabled, socio-economic factors and disease are likely to have more influence on their nutritional status than age alone.
Table 192.1 Factors associated with poor nutritional status in older people in community and home care settings Poor eye sight and hearing problems Joint problems and hand tremors Isolation Inability to go out shopping and poor income Depression and bereavement Poor cognitive and physical functioning Nausea, and vomiting Poor appetite Anorexia due to disease especially cancer, medications Poor dentition and chewing problems Acute illness In institutions, lack of supervision and assistance at mealtimes may be an important factor resulting in poor food intake
3106
S. Gariballa
192.3.2 Cognitive Function Cognitive decline and dementia are common in old age. For example, one in five of hospitalized older people has cognitive impairment at any one time, whereas dementia affects one in 20 people over the age of 65 and one in five over the age of 80. The central nervous system requires constant supply of glucose, and adequate brain function and maintenance depends on almost all essential nutrients. For example, B-group vitamins (folate and vitamin B6 and B12) deficiency or congenital defects in the enzymes associated with these pathways has been found to be associated with severe impairment of brain function. Although severe deficiencies and congenital defects are rare, milder subclinical vitamin deficiencies are not uncommon in the elderly. Many recent experimental and epidemiological studies have shown associations between loss of cognitive function or dementia and inadequate B-group vitamin status in older people. A recent study has found graded association between elevated plasma homocysteine, an indicator of inadequate B-group vitamins and cognitive impairment in healthy elderly community dwellers. A recent Cochrane review examined the effects of folic acid supplementation, with or without vitamin B12, on elderly healthy or demented people, in preventing cognitive impairment or retarding its progress. The small number of studies which have been reviewed provide no consistent evidence either way that folic acid, with or without vitamin B12, has a beneficial effect on cognitive function of unselected healthy or cognitively impaired older people. In a preliminary study, folic acid was associated with improvement in the response of people with Alzheimer’s disease to cholinesterase inhibitors. In another, long-term use appeared to improve the cognitive function of healthy older people with high homocysteine levels. More studies are needed on this important issue. A growing body of evidence also supports the notion that oxidation and inflammation are part of the mechanisms responsible for cognitive decline and progression to dementia in older people. Recent research found that the use of vitamin E supplements, but not vitamin C supplements, may be related to modest cognitive benefits in older women. Supplementation with Vitamin E has also been found to lead to significant delay in the progression of dementia. Macronutrients such as carbohydrates may have a role on cognitive function. A study of 11 men and 11 women aged 61–79 years consumed either 300-ml drink containing 774 KJ as pure protein, carbohydrate, or fat or a nonenergy placebo on four separate mornings after an overnight fast. Energy intake from protein, carbohydrate, or fat was found to enhance memory independently of elevations in blood glucose. The extent of the relationship between inadequate nutrients status and loss of cognitive function in older people remains unclear. Advances in the understanding of this complex relationship may depend on the outcome of longitudinal prospective nutritional intervention studies starting prior to the onset of neurocognitive decline.
192.3.3 Depression and Well-being The relationship between nutrition and older people psychiatry has received little attention. Recent research on the role of micronutrients in psychiatric disorders in older adults has revealed that low folic acid/vitamin B12 has been found to be associated with depression in older persons. For example, the Rotterdam study has reported a relationship between hyperhomocyteinaemia, vitamin B12 and folate deficiency and depressive disorders. Quality of life has also recently become a clinically relevant outcome measure when evaluating new treatment strategies in patient’s population, particularly an older one. Studies have shown a close relationship between undernutrition and poor quality of life in some populations such as institutionalized older people and cancer patients. A study
192 Nutrition and Quality of Life in Older People
3107
80% 70%
76 (76%) 134 (67%) 111(58%)
63(62%)
56 (63%)
Placebo
64(60%)
60% 50%
Supplement
40%
20%
21 (21%)
25 (12.50%) 16 (8%)
10%
33 (31%)
31 (30%) 24 (27%)
56 (29%) 50 (25%)
30%
8 (8%)
9(10%)
11(10%) 3 (3%)
0% No
M Baseline GDS
S
no
m 6 weeks GDS
s
no
m
s
6 month GDS
p=0.007 for between group difference; GDS=Geraitric Depression Score
Fig. 192.2 The effect of supplements on the number (%) of patients with no depression (No), mild depression (M), and severe depression (S), over 6 months compared to placebo
published recently has shown that oral nutritional supplementation of hospitalized acutely ill older patients led to a statistically significant benefit on depressive symptoms. Improvement in depressive symptoms also coincided with increase in red-cell folate and plasma vitamin B12 concentrations (Fig. 192.2). There is also some evidence, which suggests that a low intake of fish and/or n-3 PUFA is associated with depressed mood. Recently completed randomized controlled intervention trials found no effect of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) supplementati on mental well-being in the general older population studied.
192.4 Nutritional Status and Quality of Life in Older People With advancing age both undernutrition and chronic diseases become more common 1. There is evidence linking protein-energy undernutrition or its markers with clinical outcomes in acute and nonacute care settings. Poor nutrition leads to ill health and ill health to poor nutrition, so identifying priorities for managements still remain a challenge. A recent review of the literature on nutrition and older people psychiatry reported that although this issue has received little attention, there has been recent research on the role of micronutrients in psychiatric disorders in older adults. Studies have also shown a close relationship between undernutrition and poor quality of life in some populations such as institutionalized older people and cancer patients. Quality of life is a subjective multidimensional measure reflecting functional status, emotional and social well-being as well as general health. It has recently become a clinically relevant outcome measure when evaluating new treatment strategies in patient’s population particularly older ones (Table 192.2). Studies on assessment of the efficacy of nutritional support on older patient’s outcomes hitherto have been limited to open ones. A metaanalysis of trials of protein and energy supplementation in older people reported that data were limited by the poor quality of most included trials. Future trials should have sufficient power, proper concealed allocation and blinding of treatment and should focus on outcome measures of relevance to patients such as improvement in function and quality of life.
3108
S. Gariballa
Table 192.2 Key features of quality of life • Health is defined by a state of complete physical, mental and social well-being and not only absence of disease • Health-related quality of life scores often attempt to measure the psychological, physical and social effects of an illness and it is response to treatment • Quality of life is therefore a subjective multidimensional measure reflecting functional status, emotional and social well-being as well as general health • Quality of life has recently become a clinically relevant outcome measure when evaluating newtreatment strategies in patient’s population particularly older ones
Despite growing evidence that nutritional support improves outcome in older people, there is however lack of good-quality data of the effects of nutritional support on quality of life measures in older people.
192.4.1 Assessment of Quality of Life Health is defined by a state of complete physical, mental and social well-being and not only absence of disease. Health-related quality of life scores often attempt to measure the psychological, physical and social effects of an illness and it is response to treatment. This latter point is important because in this day and age, efficacy of treatment is not only judged on safety and ability to improve clinical outcomes but also on acceptability to patients and cost-effectiveness. Outcome measures which encompass quality of life measure such as the SF 36 are therefore very important especially in older people (Table 192.3). A large number of generic and disease-specific quality of life measures have been developed and some have been validated mainly for cancer in palliative care.
192.4.2 Undernutrition and Quality of Life A prospective study of 579 randomly selected home-living older people has found that lower selfperceived health had the highest power to predict risk of malnutrition, with increased number of depression symptoms and higher age as second and third predictors. A very recent systematic review examined the relationship between health-related quality of life and nutritional status of the patient. The authors reviewed studies that relate health-related quality of life to nutritional status and examined the tools (questionnaires) that they were to use to investigate this relationship. A critical review of published studies was carried out via an investigation of the following databases: MEDLINE (via PubMed), EMBASE, The Cochrane Library, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Institute for Scientific Information (ISI) Web of Science, Latin American and Caribbean Health Sciences Literature (LILACS), Spanish Health Sciences Bibliographic Index (IBECS). The search was carried out from the earliest date possible until July 2007.The medical subject heading terms used were ‘quality of life’, ‘nutritional status’ and ‘questionnaires’. The articles had to contain at least one questionnaire that evaluated quality of life. Twentyeight documents fulfilling the inclusion criteria were accepted, although none of them used a specific questionnaire to evaluate HRQoL related to nutritional status. However, some of them used a combination of generic questionnaires with the intention of evaluating the same. Only three studies selectively addressed the relationship between nutritional status and quality of life, this evaluation being performed not by means of specific questionnaires but by statistical analysis of data obtained via validated
192 Nutrition and Quality of Life in Older People
3109
Table 192.3 Key features of SF-36 quality of life assessment questionnaire • The SF36 is a validated Quality of life Medical Outcomes Study 36-items General Health Survey questionnaire • The questionnaire consists of 36 questions forming 8 multi-item scale including physical functioning, role limitations- physical, role limitation- emotional, bodily pain, general health, vitality, social functioning and mental health • The SF 36 validity is now well established and it has been used in several large studies • The SF 36 has been adapted for use with older adults • The questionnaire is measured on a 0 to 100 (good health) scale, self-administered with help provided when needed and takes about 10 minutes to complete
questionnaires. The first study by Eriksson et al. investigated measurements of nutritional status and health-related quality of life. The author’s objective was to relate a well-established questionnaire of nutritional status (MNA) to a likewise well-established questionnaire of health-related quality of life (SF-36) in community dwelling, free-living and, healthy 70–75-year-old persons. Before an interview, the MNA and SF-36 questionnaires were filled in by 128 participants from a sample of 262 subjects. Their results showed that the MNA worked well as a measurement in this sample. Many MNA aspects correlated with the SF-36 scales. The correlations between MNA total score and the eight SF-36 scales varied from.27 to.62. This correlation was partly due to the fact that MNA has questions of health but also to the fact that there is an empirical relation between nutrition and health. The authors concluded by saying “The MNA measurement is applicable to a healthy, free-living elderly population and parts of the MNA can be interpreted as measurements of health-related quality of life. Low values of SF-36 could also be used as predictors of risk of malnutrition, although further studies are required to confirm this result”. The second study was a cross-sectional survey to determine the independent association of nutritional risk with HR-QOL in frail older adults. Data were collected by intervieweradministered questionnaire. Nutritional risk was measured by SCREEN (Seniors in the Community: Risk Evaluation for Eating and Nutrition) and HR-QOL by perceived health status and report of number of days in the past month where physical or mental health was not good, or where activities were limited. Frail (n = 367) seniors were recruited from 23 community service providers. A wide variety of covariates were also measured. Multivariate modelling based on a conceptual model was used to identify factors associated with HR-QOL. The results showed that nutritional risk appears to be a significant and important factor associated with HR-QOL. Other significant covariates were: falls, social supports, social activity, health behaviours, pain and medication use. Nutritional risk as measured by SCREEN appears to be a significant covariate in explaining differences in HRQOL among frail older adults. Further work should determine if nutritional risk predicts changes in HR-QOL over time. The third study evaluated the scored Patient-generated Subjective Global Assessment (PG-SGA) tool as an outcome measure in clinical nutrition practise and determine its association with quality of life (QoL). This was a prospective 4-week study assessing the nutritional status and QoL of ambulatory sixty cancer patients aged 24–85 years receiving radiation therapy to the head, neck, rectal or abdominal area. Outcome measures were scored PG-SGA questionnaire, subjective global assessment (SGA), QoL (EORTC QLQ-C30 version 3). According to SGA, 65.0% (39) of subjects were found to be well-nourished, 28.3% (17) moderately or suspected of being malnourished and 6.7% (4) severely malnourished. PG-SGA score and global QoL were correlated (r = -0.66, P < 0.001) at baseline. There was a decrease in nutritional status according to PG-SGA score and SGA; and a decrease in global QoL after 4 weeks of radiotherapy. There was a linear trend for change in PG-SGA score and change in global QoL between those patients who improved (5%) maintained (56.7%) or deteriorated (33.3%) in nutritional status according to SGA. There was a correlation between change in PG-SGA score and change in QoL after 4 weeks of radiotherapy. Regression analysis determined that 26% of the variation of change in QoL was explained by change in PG-SGA.
3110
S. Gariballa
192.4.3 Effects of Nutritional Support on Quality of Life Poor nutrition may lead to ill health and ill health to poor nutrition. For example, many studies have found that lower self-perceived heath/quality of life and/or symptoms of depression increased risk of malnutrition and that malnutrition leads to poor quality of life and increased depression symptoms. So identifying priorities for management remains a challenge. Prior to coming into hospital elderly people in the community are more likely however, to have premorbid decrease in energy or calorie intake, less lean body mass and impaired immune response, which may be associated with poor nutritional. Their nutritional status is likely to deteriorate further as the result of the catabolism associated with the acute illness. This is compounded further by the demands of the sometimes-prolonged period of rehabilitation. Nutritional depletion during rehabilitation, however, may be more serious than during acute illness, since rehabilitation periods may extend over weeks and months, and weight loss, although less marked than in the early catabolic phase may be greater overall. In 1971, a study compared nutrient intakes of long stay, acutely ill and rehabilitation patients, and showed that energy and protein intakes were lowest in the latter group. A clinical review by Caro and colleagues on the effects of nutritional intervention on quality in adult oncology patients reported that nutritional intervention should be integrated into oncology care because it increases tolerance and response to treatment, decrease complications and improve quality of life by controlling symptoms such as nausea, vomiting, pain. Quality of life has also been found to be related to nutritional status in dialysis patients. Providing individualized nutritional counselling improves many components of quality of life, compared with standard nutritional care, in the stage prior to dialysis treatment. A recent randomized placebo-controlled study has reported that nutritional supplementation of older people during acute illness and convalescence/rehabilitation period significantly improves quality of life. The improvements in quality of life indices was accompanied by significant improvements in biomarkers of nutritional status in the supplement group, which were evident at 6 weeks and sustained at 6 months (Table 192.4 and Fig. 192.3). Improvement of micronutrients status such as vitamin B12 and red-cell folate would be the most plausible explanation for the improvement in well-being of our study population. Although there are well-known changes in micronutrients status in old age that significantly correlate with adverse physical outcomes such as cardiovascular, disease, cerebrovascular disease, impaired immune function and bone health all contributing to the development of frailty less is known of a relationship between B12, folate and quality of life. There is however evidence of a strong link between B12/folate status and depression in older adults. The lack of statistically significant differences in anthropometeric measures between the supplement and placebo group could be due to the short time frame of the supplementation and to inherent difficulties in measuring these nutritional indices in ageing patients. This is especially true for studies in the elderly, being affected by age-related changes, disability, illness and injury. Another plausible explanation for our results would be that mild subclinical nutritional deficiencies, which are known to be common even in relatively healthy persons, which otherwise would have gone unnoticed in our supplement group, have been corrected, hence the clinical benefit. This trial has demonstrated that nutritional supplementation of hospitalized older people does lead to a clinically important benefit. In conclusion, the number of older people is growing rapidly worldwide and looks set to continue to increase further in the future act on quality of life for older people. This has created a need for additional knowledge of age-related changes relevant to nutrition which has importance in the treatment and prevention of diseases and in maintaining good health and quality of life in an ageing population. Evidence is emerging of a link between undernutrition and poor quality of life in older people and that improvement in nutrition status leads to improvement in quality of life but there is an urgent need for more research in this field. The prospect of the effects of improved nutritional status of older people quality of life could have an important and a substantial health and economic impact.
Table 192.4 Effect of supplements on quality of life (SF-36 domains) compared to placebo Mean Difference Group N Baseline (SD) Mean 6 months (SD) at 6 months SF-36 Physical Placebo 107 34.7 (30.5) 32.5 (27.3) 6.6 Function Supplements 95 33.5 (29.0) 39.1 (30.5) SF-36 Role Placebo 106 17.7 (32.2) 28.8 (37.1) 10.3 Physical Supplements 94 17.3 (33.4) 39.1 (38.7) SF-36 Bodily Placebo 107 41.4 (29.7) 55.1 (28.9) 2.5 Pain Supplements 95 42.0 (31.4) 57.7 (30.2) SF-36 General Placebo 106 46.3 (22.2) 49.4 (23.0) 1.3 health Supplements 96 48.5 (22.4) 50.7 (23.3) SF-36 Vitality Placebo 106 39.6 (23.3) 42.3 (21.4) 8.1 Supplements 95 45.8 (24.7) 50.4 (24.3) SF-36 Social Placebo 107 51.0 (32.0) 58.7 (30.0) 7.4 Function Supplements 95 48.5 (33.5) 66.1 (29.9) SF-36 Role Placebo 105 42.9 (45.0) 52.1 (44.3) 5.2 Emotional Supplements 96 43.1 (45.6) 57.3 (42.9) SF-36 Mental Placebo 105 65.1 (21.3) 66.9 (21.8) 5.7 Health Supplements 95 69.4 (19.3) 72.6 (18.4) Barthel Placebo 119 16.5 (4.4) 18.6 (2.6) –0.4 Supplements 106 16.3 (4.6) 18.2 (3.1) Treatment effect: difference in quality of life scores at 6 months, after adjustment for baseline quality of life scores, age, and gender Outcomes are based on patients who completed both the baseline and 6-month assessments Treatment effect (95% CI) 7.0 (0.5 to 13.6) 10.2 (0.1 to 20.2) 1.9 (–5.7 to 9.5) –0.1 (5.4 to 5.1) 4.7 (–0.7 to 10.1) 7.8 (0 to 15.5) 5.0 (–6.3 to 16.2) 3.3 (–1.5 to 8.1) –0.3 (–0.8 to 0.3)
P-value 0.035 0.047 0.619 0.956 0.088 0.050 0.384 0.181 0.369
192 Nutrition and Quality of Life in Older People 3111
3112
S. Gariballa
Fig. 192.3 Overall SF-36 scores of placebo and supplement group over 6 months mean (SD)
192.5 Application to Other Areas of Health and Disease Undernutrition can adversely affect physical, psychological and behavioural function and this can have major social and economic implications. Evidence is emerging of a link between undernutrition and poor quality of life in older people and that improvement in nutrition status leads to improvement in quality of life. Older people should therefore be advised to eat a balanced diet containing a variety of nutrient-dense foods; more fruits, vegetables and grains; foods containing adequate amounts of calcium and vitamin D. Widespread implementation of this strategy could have a substantial economic impact and improve mental health for older people.
Summary Points • Many societies worldwide have experienced a considerable increase in the number of elderly people. • Ageing, disease, life style, and environmental factors account for many of the changes observed in older people. • There is a growing recognition that age-related physiological changes may predispose to proteinenergy undernutrition, in the elderly particularly in the presence of other ‘pathological’ factors associated with ageing; such as social, psychological, physical, and medical factors, the majority of which are responsive to treatment. • This has created a need for additional knowledge of age-related changes relevant to nutrition, which has importance in the treatment and prevention of diseases and in maintaining good health and quality of life in an ageing population. • Evidence is emerging of a link between undernutrition and poor quality of life in older people and that improvement in nutrition status leads to improvement in quality of life but there is an urgent need for more research in this field.
192 Nutrition and Quality of Life in Older People
3113
Key Points of Diet, Nutrition and Quality of Life in Older People 1. A healthy older person’s dietary patterns and the food eaten are not likely to be that much different from what is known about those of younger age. 2. The majority of ‘pathological’ factors associated with ageing; such as social, psychological, physical and medical factors, which may predispose to undernutrition, are responsive to treatment. 3. New evidence is emerging of a link between undernutrition and poor quality of life in older people and that improvement in nutrition status leads to improvement in quality of life. 4. Future research should focus on the role of adequate nutrition and active life in prevention and treatment of disease and improving quality of life in the ageing population. 5. Older people should be advised to eat a balanced diet containing a variety of nutrient-dense foods; more fruits, vegetables and grains; foods containing adequate amounts of calcium and vitamin D and this may need to be monitored in certain individuals.
References Bhat RS, Chiu E, Jeste DV. Curr Opin Psychiatry. 2005;18:609–14. Campbell KL, Ash S, Bauer JD. Clin Nutr. 2008;27:537–44. Eriksson BG, Dey DK, Hessler RM, Steen G, Steen B. Nutr Health Aging. 2005;9:212–20. Gariballa S, Forster S. Clin Nutr. 2007;26:545–51. Gariballa S, Forster S. J Am Geriatr Soc. 2007;55:2030–4. Gariballa SE, Sinclair AJ. Ageing & the elderly. In Geissler C, Powers H, editors. International textbook of human nutrition, 11th ed. New York: Elsevier Science; 16; 2005. p. 321–336. Gariballa SE, Sinclair AJ. Br J Nutr. 1998;80:7–23. Johansson Y, Bachrach-Lindström M, Carstensen J, Ek AC. J Clin Nurs. 2009;18:1354–64. Isenring E, Bauer J, Capra S. Eur J Clin Nutr. 2003;57:305–9. Keller HH. J Nutr Health Aging. 2004;8:245–52. MacIntosh CG, Morley JE, Horowitz M, Chapman IM. Nutrition. 2000;16:983–95. Milne AC, Potter J, Avenell A. Cochrane Library. 2002; issue 3, Oxford. United Nations, World Population Prospects: The 2004 Revision (medium scenario) 2005. www.prb.org/presentations/ j_trends-aging Wanden-Berghe C, Sanz-Valero J, Escribà-Agüir V, Castelló-Botia I, Guardiola-Wanden-Berghe R. Red de malnutrición en iberoamérica – ciencia y tecnología para desarrollo (Red MeI – CYTED. Br J Nutr. 2009;101:950–60.
wwwwwwwwwwwwwww
Chapter 193
Biopsychosocial, Behavioural Aspects and Quality of Life with Home Enteral Nutrition Agostino Paccagnella, Alessandra Mauri, Gessica Schiavo
Abbreviations AN EN HEN HRQoL PN QoL WHO
Artificial nutrition Enteral nutrition Home enteral nutrition Health-related quality of life Parenteral nutrition Quality of life World Health Organization
193.1 Introduction Many diseases that are curable from a medical standpoint cannot be healed by medicine as yet. This channels patients presenting such disorders into a condition that is defined as chronic. A chronic disease is characterised by the need to manage it in the long term to undergo complex, long-term, therapeutic interventions from the onset of complications subsequent to the treatment itself. Chronicity is almost always accompanied by a worsened health condition, following the progress of the basic disease. The ability to diagnose a disease or a condition of deficiency (e.g. malnutrition) is considered a positive factor – from a biomedical viewpoint – as it enables to plan a series of interventions focused on managing the physical health of the subject. When an acute disease is diagnosed, the process is scheduled, medical treatments are decided in advance and healing can be achieved during a certain time interval. This encourages patient compliance. The acute disease alters the life of the subject for a defined period of time. Moreover, social, physical changes and alterations in the framework and habits of the patient are limited in time and can therefore often be managed. The situation concerning the diagnosis and treatment of chronic diseases from which the patient will not heal is quite different. The consequences of the chronic disease influence all aspects of the life of a subject, affecting him psychologically, socially and contextually. Disease characteristics can change in time, producing several types of complications, new symptoms and new therapeutic needs. In this case, the biomedical approach alone for scheduling and evaluating interventions will not suffice
A. Paccagnella (*) Nutrition, Metabolism and Diabetes Unit, Ospedale Ca’ Foncello, 31100 Treviso, Italy e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_193, © Springer Science+Business Media, LLC 2011
3115
3116
A. Paccagnella et al.
anymore because the aspects involved are many. The needs, feelings and expectations of both the patient and his relatives become key factors in treatment, which entails taking care of the whole patient and not only of his body. A typical example of treatment, combined with a chronic disease is HEN, which is administered to subjects who cannot be fed by mouth due to an underlying disease and who are sent home on artificial nutrition (AN) after a certain period of hospital treatment. The possibility of sending home patients with chronic diseases that prevent food intake by mouth has basically changed home health care. It has changed the economic conditions of health-care institutions, and also raised ethical issues on the risk of therapeutic obstinacy. Home care is highly beneficial for those who were either not receiving appropriate treatment or were forced to stay long in hospital. HEN has become a therapeutic option that considerably lengthens the life of patients who were previously doomed to malnutrition, in some cases even significantly postponing the time of death. But this very outcome and its clinical (patients whose basal disease deteriorates slowly and progressively), ethical (should therapy be terminated? when?) and social implications have an impact on the patient as a whole (Buchman 2005). Taking for granted that therapeutic results and actions must be evaluated with the appropriate clinical, anthropometric, blood-chemical monitoring systems and equipment to provide the patient with the best physical conditions possible given the basal disease, it would be naive to theorise that only physiological parameters are indicators of the state of well-being of a patient undergoing HEN, since this treatment becomes part of the life of the patient in various moments of the day, and for very long periods, if not forever. Hence the idea of shifting focus from a biochemical or biomedical model, which considers the illness from a purely objective standpoint, to a model that can deal with the patient in all his facets, considering his broader biopsychosocial framework. This chapter describes the reference theoretical framework, treatment characteristics and impact on both subjects and their lives of a transition to artificial nutrition from oral feeding. It defines quality of life and health-related ‘quality of life’ (QoL), evaluating the importance of measuring the ‘health-related quality of life” (HRQoL) in patients who are sent home on enteral artificial nutrition. The chapter briefly describes standardised tools that are most used according to literature for the evaluation of HRQoL and the possible problems found in measuring this parameter. AN comprises artificial parenteral nutrition (PN) and artificial enteral nutrition (EN). PN consists in using venous catheters to administer nutrients directly into the bloodstream through a peripheral or central venous access. EN allows nutrients to be administered directly into the gastrointestinal system with special probes that are inserted either through the nose or stomies created through the abdominal wall. This chapter will specifically enlarge on EN by analysing aspects of the impact of treatment in the daily life of subjects.
193.2 The Theoretical Framework: A Biopsychosocial Approach The World Health Organization (WHO) has long invited medicine to adopt a new approach to the evaluation and management of chronic diseases (WHO 1986). According to this theory, the correct approach and management of a chronic disease demands a biopsychosocial approach that centres treatment on the patient and his relatives who become the lead actors in the management of the disease. In medical literature, the ‘biochemical model’ (that evaluates the objective alterations of the patient) is increasingly integrated into the holistic model that can consider the viewpoint of the patient (e.g. how he experiences his disease and subsequent treatment and effects on daily life) (Aspinwall and Staudinger 2003). This approach shifts focus from the disease ‘of the body’ to the
193 Biopsychosocial, Behavioural Aspects and Quality of Life with Home Enteral Nutrition
3117
disease ‘of the patient’ who is deemed a person with an inner world of his own and lives in a specific physical and social framework. Adopting this overall standpoint, the therapeutic intervention is managed along with the patient who becomes the active protagonist of his decisions, and even the idea of wellness is considered from a subjective viewpoint, conversely to the directive and prescriptive intervention of the biomedical approach. The ‘biopsychosocial’ model combines the medical and psychosocial models of health to implement an intervention that issues from a ‘reasonable dialogue’ with the subject under treatment and his relatives. Focus is not on the disease anymore but on health, which is deemed as an attempt to create new balances to reach physical, psychological and subjective wellness, starting from the sociocultural framework the individual belongs to. Even treatment-based relations change in this context. While in the acute patient, especially at the time of diagnosis, the biomedical approach defines the operator as the competent person who, in a directive and prescriptive framework, is entrusted with the most appropriate decisions required to define the disease, in chronic patients treatment-based relations develop over a prolonged period of dialogue with the patient, assuming therapeutic features in a broad sense. The health care operator and the user are the two key actors of the relationship. The operator is trained on diagnostic, prognostic, treatment and prevention techniques, but only the user has direct experience of the disease, of the social framework in which he lives and of his values and preferences. The operator must know where the patient lives, his attitude towards health and treatment and his expectations to directly establish relations with the patient and work on the treatment. This is implemented in a new perspective in which the organic disease and subjective experience of the user concerning the disease have the same importance.
193.3 Eating and Feeding Sharing food is an essential factor in social relations and friendship. The intake of food, the foodrelated attitude and the act of eating itself have complex and varied features that continuously intersect the biological framework with the affective and social life of the individual. Therefore, the type of food shared, meal characteristics and rate of occurrence are very strong indicators and components of affective bonds. They are directly related with the construction and reproduction of emotional relations, converging to form a basic element in the life of a subject. In various societies food is a way of establishing bonds, managing power and defining roles. Several messages that do not merely stop at the level of nutritional function but can be inserted into the complex and circular biopsychosocial interpretative model move through the symbology of food that is prepared, offered and accepted. When a disease makes it impossible for a patient to eat again, causing him to resort to tube feeding it is easy to imagine the deep lifestyle changes subsequent to the diagnosis of a chronic disease and the intervention of AN as treatment and prevention of malnutrition. This chapter defines the concept of the intrinsic artificiality of EN, considering the attitude of the patient towards the mode of administration and not towards the type of ‘food’ used, which can be natural like home-made blended foods, or an industrial item that is produced and packaged in a totally artificial framework. According to this definition, home-made blended foods infused through a nasogastric tube possess the therapeutic traits of artificiality, while an artificial substance taken by mouth (e.g. an industrial product for patients with Crohn’s disease) is, by the same definition, a natural food that has no therapeutic valency. As a matter of fact, the person who ‘eats by mouth’ can manage food both in a specific (e.g. choose the food from a menu) and social sense (e.g. organise a dinner to close a business), as he can share it with others in the most diverse frameworks. This can paradoxically occur with an industrial product for oral use that is administered to a patient who
3118
A. Paccagnella et al.
p resents a specific disease, which requires him to take special oral food. He can, however, sit down at table, experience lunch as a social moment and on some occasions even let another person taste the artificial food. Conversely, the need to infuse either blended or industrial food through a tube, which deprives the patient of the sense of taste and smell, modifies the feeling of satiety and, especially, refrains the patient from attending social food-sharing moments – if only for the embarrassment of having to expose the terminal part of the PEG to others – expresses in itself the concept of artificiality of enteral nutrition. Therefore, this chapter will hereinafter designate as ‘eating’ everything that is taken by mouth and has implicit repercussions on the affective, social, cultural and biological framework. It will designate that which is commonly and historically called ‘food’ and which envisages the act of eating. AN and, especially, EN (including HEN), instead, designate all calories and macro- and micronutrients that are infused through a tube to ensure nourishment to the body, but which deprive the person of the psychosocial implications of “eaten food”. Hence, AN envisages important clinical monitoring, requires informed consent to position the tubes, the continuous use of nutripumps to infuse food, specific result indicators and blood chemistry monitoring. These concepts make it easy to understand that artificial nutrition is an example of therapeutic intervention in which the integrated and holistic dimension that creates wellness is damaged, and subjective balances are modified. EN shifted to the domestic framework (HEN) requires specialist health-care skills (e.g. evaluation of the need, choice of nutritional mixture and route of administration), presupposes continuous contact with health-care operators and the acquisition of some specific management skills on the part of both the patient and his family (caregivers). The therapeutic intervention that modifies ‘eating’ and turns it into ‘nutrition’ (namely EN) does not necessarily prevent the patient from performing routine daily tasks. Some patients go on working and travelling. But to maintain their normal productive activities through HEN, it certainly generates new biological, emotional and relational variables. Moreover, the social and affective characteristics of food are not the only factors that must be considered in the transition from eating to AN, as there are other forms of change associated with sensory deprivations such as, for instance, the loss of the taste and smell of food, altered body image (e.g. stepping into the swimming pool with the PEG) or interference in certain frameworks of life of the subject (e.g. sexuality). The life characteristics of these subjects, their emotional and psychosocial difficulties, the physical and practical obstacles they face in carrying out their tasks, the complications and the reactions of relatives are all factors that must be considered and evaluated in planning home-based interventions and treatment. A special subgroup of patients is the one formed by subjects in a permanent vegetative condition that was mainly caused by cerebrovascular accidents or degenerative diseases, such as Parkinson’s syndrome or Alzheimer’s disease. In such subjects (who were defined as ‘non-competent’), the concept of quality of life evaluation necessarily shifts to the relational and environmental framework, such as the impact of treatment and home care, on their caregivers (Hebert et al. 2007), and the situation is evaluated by operators on the basis of clinical scales (Fayers and Machin 2001).
193.4 Biomedical Aspects of Enteral Nutrition As already mentioned, enteral nutrition entails introducing either industry-produced nutrients (calories, macro- and micronutrients) or processed foods (e.g. blended foods and chopped foods) with the assistance of a ‘tube’ (nasogastric tube, NGT; percutaneous gastrostomy, PEG; surgical jejunostomy, SJ; nasojejunal tube, NJT; percutaneous jejunostomy, PEJ; surgical gastrostomy SGS). It therefore either improves or maintains an adequate nutritional condition by treating or preventing malnutrition
193 Biopsychosocial, Behavioural Aspects and Quality of Life with Home Enteral Nutrition
3119
and its complications in single organs and systems, influencing the risk of complications and altering the risk of death for patients (NICE 2005; Lochs et al. 2006; Bankhead 2007). HEN is a set of organisational modes for EN administration at home when the clinical condition of the patient is stable, and his social and family framework can guarantee the safety and efficacy of treatment outside the hospital on a home care regime. HEN has the same indications as EN administered in the hospital but it allows treatment to be administered at home, thus avoiding long stays in hospital and reintegrating the patient into his family, social and, at times, work-related framework (Lochs et al. 2006; Bankhead 2007).
193.4.1 Epidemiology Epidemiological studies on enteral nutrition at home are still few due to the few registers or databases appointed to collect such data. This data estimates a progressive increase in HEN with a growth trend of over 20% in the past. A multicentre, retrospective study promoted by the European Society of Parenteral and Enteral Nutrition to evaluate indications and practice of HEN in 23 centres in eight European countries in 1998 highlighted an incidence of 163 cases/106 inhabitants/year (range: 62–457 cases/106 inhabitants/year). The paper reported that the main indication for HEN was the presence of dysphagia (84.6%), and the main diseases associated with the request were neurological (49.1%) and neoplastic (head and neck region: 26.5%) (Hebuterne et al. 2003). Despite this important rise in HEN, its incidence and prevalence is extremely varied and diversified, and depends on the clinical condition and organisation of the reference territory (Jones et al. 2007; Paccagnella et al. 2008). Compared with the epidemiological data of HEN, the method was applied to patients hospitalised as nursing home residents (NHRs) in few cases. A recent survey carried out in the USA found that about 34% of the 1.4 million NHRs require some assistance to eat (Centers for Medicare and Medicaid Services 2005). It is currently estimated that about 2–34% of NHRs are nourished by tube feeding (Mitchell et al. 2003). Our data on NHRs who underwent enteral nutrition (period: 2000–2005) revealed that 6.6% of all subjects were fed by means of tubes, with an incidence of 223.4 cases/106 inhabitants/year and a prevalence of 279.4 cases/106 inhabitants (Morello et al. 2009). If the study considers the entire population of patients aged >65 years who undergoes HEN (including 664 subjects at home in the period 2001–2006) (Fig. 193.1), a mean incidence of 269.9 cases/106 inhabitants/year can be noticed, with a prevalence of 313.5 cases/106 inhabitants. If the prevalence study is based on patient age in
5,0
Other
1,5
Amyotrophic Lateral Sclerosis
2,9
Other Cancer
6,9
Abdominal Cancer
9,2
Head-Neck Cancer
40,1
Neurodegenerative
34,5
Neurovascular
0,0
10,0
20,0
30,0 %
Fig. 193.1 Incidence of diseases leading to HEN (664 subjects > 65 year at home)
40,0
50,0
3120
A. Paccagnella et al.
this sample group, there would be 95.8 cases/106 inhabitants for patients aged 66–75 years, 151.3 for those aged 75–85, 116.0 for those aged 86–95 and 17.7 for patients aged >95 years.
193.4.2 Medical Indications for Treatment National and international guidelines agree that EN is the first choice method in all conditions, which recommend AN as they have adequate anatomofunctional gastrointestinal integrity. This is based on the fact that EN is more physiological, is worsened by fewer complications (especially infections), costs less and can be managed with greater ease. The transit of nutrients along the intestine safeguards its trophism, either conserving or restoring the integrity of the absorbing surface. Table 193.1 specifies when a patient should undergo EN. EN (just as HEN) must basically either prevent or reduce the risk of malnutrition (or undernutrition) that can act synergically with the basic disease, increasing the risk of complications and of the decease of patients. Table 193.2 reports the physical and psychosocial consequences of hyponutrition. Contraindications to EN are intestinal occlusion or subocclusion, chronic intestinal ischaemia, high flow jejunal or ileal fistulae, untreatTable 193.1 Indications for EN (Amended by the National Institute for Health and Clinical Excellence, NICE 2005) EN should be considered for patients when • The patient has eaten very small amounts for the last 5 days or more, or • The patient is very unlikely to eat more than very small amounts for the next 5 days or more (whatever current BMI or history or weight loss), or • The patient has a BMI 10% body weight within the previous 3–6 months, or • The patient has a BMI 5% within the previous 3–6 months, or • The patient has poor absorptive capacity, is catabolic and/or has high nutrient losses and/or has a condition that increases their nutritional needs (i.e. hypermobility) This table specifies when a patient should undergo EN The criterion, amended by NICE 2005, includes reduction of food intake, fasting, weight loss, low BMI, alteration of absorptive capacity and condition of increasing nutritional needs Table 193.2 Same physical and psychosocial effects of undernutrition in chronic patients (Amended by the National Institute for Health and Clinical Excellence, NICE 2005) Adverse effect Consequence Impaired immune responses Risk of infections Impaired wound healing Surgical wound dehiscence, development of post-surgical fistulae, anastomotic failures, risk of wound infections and ununited fractures Inactivity, inability to work, poor self-care; abnormal muscle (or neuromuscular) Reduced muscle and function; poor cough pressure, risk of chest infections; difficulty weaning respiratory muscle malnourished patients from ventilators strength, and fatigue Inactivity Predisposition for pressure sores and thromboembolism especially in bed-bound patients Water and electrolyte Cardiac arrhythmias, muscle pain, increased vulnerability to refeeding syndrome, disorders and iatrogenic sodium and water overload Vitamin deficiencies Specific vitamin deficiency states (e.g. Wernike–Korsakoff syndrome), osteoporosis Impaired thermoregulation Hypothermia and falls, especially in the elderly Impaired psychosocial Apathy, depression, self-neglect, hypochondriasis lack of self-esteem, poor function body image, lack of interest in food, deterioration in social interactions Undernutrition causes many consequences for human organs and systems with many physical and psychosocial adverse consequences
193 Biopsychosocial, Behavioural Aspects and Quality of Life with Home Enteral Nutrition
3121
able vomiting, paralytic ileum and/or intense diarrhoea and severe alterations in intestinal function subsequent to enteropathies or deficiencies in the absorbing surface (DiBaise and Scolapio 2007). The same indications reported in Table 193.2 apply to the home-based patient who undergoes HEN. However, this requires concurrent suitable clinical and environmental conditions, a patient with a stable clinical condition, the possibility of managing the basic disease at home, suitability of the house to ensure correct management and the presence, in patients who are not self sufficient, of a caregiver or of nursing care at home. Lastly, as discussed below, since HEN interferes considerably in the life of the patient and his family, it requires the informed consent of the patient and a correct evaluation of the economic and psychological cost for the patient and his relatives, considering the prognosis of the basic disease and/or the results that can really be obtained with HEN.
193.4.3 Enteral Feeding Routes of Access Once HEN is recommended, the clinical condition of the patient has been evaluated, the anatomy and functionality of the digestive system and envisaged duration of treatment have been defined and the ideal access route for tube feeding of nutrient mixtures must be established. Many types of enteral feeding tubes can be used to deliver nutrition to the stomach or upper gastroenteric tract. In practice, enteral feeding tubes can be divided into two main categories: (a) those positioned through the nose (nasogastric, nasoduodenal and nasojejunal tubes), which are generally recommended for EN that is scheduled for at most 30 days; (b) those requiring minor surgery (percutaneous endoscopic gastrostomy, PEG; percutaneous endoscpic jejunostomy, PEJ), which are recommended for EN >30 days (Bankhead 2007). An epidemiological study conducted in Europe estimated that 813 on 1,3972003 patients undergoing HEN (58.2%) wore a PEG, 410 (29.3%) an NGT, 76 (5.4%) underwent surgical jejunostomy and 3.4% other means of access (Hebuterne et al. ). In our experience, PEG and SNG were the main access routes chosen by patients undergoing HEN either at home or in a nursing home, but the NGT is used more than the PEG (Paccagnella et al. 2008; Morello et al. 2009). Figure 193.2 reports the incidence of use for the main types of enteral feeding tubes in 664 patients aged >65 years treated by our unit in the period 2001–2006.
100%
100%
97,9%
97,1% NGT
80%
70,5%
PEG
NJN
70,9%
PEJ
SGS
69,4%
%
60% 40%
28,7% 1,1%
20%
1,1%
28,2%
29,9%
2,1% 0,7%
0,8%
0,9%
0,7%
0% 2001
2002
2003
2004
2005
2006
years Fig. 193.2 Enteral access device from 2001 to 2006 (664 subjects > 65 year at home). NGT nasogastric tube, PEG percutaneous gastrostomy, SJ surgical jejunostomy, NJT nasojejunal tube, PEJ percutaneous jejunostomy, SGS surgical gastrostomy
3122
A. Paccagnella et al.
In some clinical situations as, for instance, tumours in the head and neck region under chemoradiotherapy, PEG is to be preferred to NGT to enhance patient mobility because it is less visible and has a lower impact on the quality of life (Arends et al. 2006). Placing of the PEG for prophylactic purposes prior to the implementation of radiochemotherapy treatment proved effective, in these cases, in reducing the risk of malnutrition (Cady 2007). However, in our experience, similar results have been obtained with the NGT as long as the patient follows a treatment and monitoring track that is monitored by specialists and is proposed on the first medical examination of the patient (Paccagnella et al. 2009).
193.4.4 Complications of HEN The progress of placement techniques and improvements in management have achieved success percentages of 90–95% and 95–98%, respectively, for SNG and PEG. But management of these medical aids is not free of complications (Iyer and Crawley 2007). Data published in literature reports several mechanical complications related to SNG and PEG placement, such as mechanical traumas (haemorrhages, perforations) and inhalation of food into airways (1.8%). Migration (16–41%) or dislocation of the tube and mechanical malfunction such as kinking, cracking and breaking are more common and can occur in 6–20% of cases. A frequent complication is an occluded tube (20%) (Iyer and Crawley 2007). Placement of a PEG can cause endoscopic complications such as bleeding (0.02–0.06%), oesophageal perforation (0.008–0.04%) and inhalation (0.3–1.0%). Unintentional dislocations can occur in 1.6–0.4% of cases, and a clogged PEG tube is often reported (45%). A serious but not common complication is the buried bumper syndrome (1.5–1.9%). Gastrocolocutaneous fistula, small bowel, liver and splanchnic injury occur rarely (Schrag et al. 2007). A retrospective study conducted on a cohort of 55 patients wearing a PEG highlighted that the most common complications recorded during a period of 25.9 months were granulation tissue (67%), broken or leaky tube (56%), leakage around the tube site (6%) and stomal infection requiring antibiotics (45%) (Crosby and Duerksen 2005). Infections at the site of the stomy are a frequent complication, whose incidence varies from 5% to 30% and can be reduced by antibiotic prophylaxis. Necrotising fascitis is rare (Iyer and Crawley 2007). EN often causes gastrointestinal complications, such as nausea (10–20%), diarrhoea (30%), abdominal distension and pain, constipation, gastroesophageal reflux and vomiting. Mechanical and gastrointestinal complications can be associated with metabolic complications, such as the refeeding syndrome (RS), alterations in the electrolyte and water balance and hypo/hyperglycaemias (DiBaise and Scolapio 2007). A very rare complication described in literature in patients wearing a PEG is tumour implantation or metastasis at the site of stomy (Iyer and Crawley 2007). Risk factors that increase mortality in patients subject to PEG placement after the age of 75 have been defined in recent years. Urinary infections and past episodes of pneumonia (ab ingestis) have recently been associated with diabetes mellitus, chronic obstructive bronchopneumopathy and low blood levels of albumin.
193.4.5 Enteral Formulas Enteral tube feeding formulas are designated by the US Food and Drug Administration (FDA) as medical foods: ‘a food which is formulated to be consumed or administered enterally under the
193 Biopsychosocial, Behavioural Aspects and Quality of Life with Home Enteral Nutrition
3123
Table 193.3 Main types of formulas for EN Polymeric Elemental or semi-elemental Proteins Whole or hydrolysed proteins L-aminoacids, tripeptides Carbohydrates Maltodextrins, polysaccharides, oligosaccharides oligosaccharides (70 years). 5.5 per cent patients in the population analysed remained dependent on HEN, 32.6% resumed full oral nutrition, 20.2% died during the first month on HEN and 35% died after
3124
A. Paccagnella et al.
more than 1 month on HEN (219 ± 257 days). This poor outcome was also influenced by the severity and nature of the underlying disease. In the USA, 59% of cancer patients and 48% with neurological swallowing disorders died 1 year after commencing HEN (Hebuterne and Schneider 2005).
193.5 Psychosocial Aspects of Home Enteral Nutrition 193.5.1 Evaluating Quality of Life and Health-Related Quality of Life Aspects related to the cost of treatment, complications, mortality and prognosis cannot be considered as the only indicators to evaluate the efficacy of a treatment. Chronic patients should achieve psychological wellness that is commonly defined as ‘quality of life’ (QoL). In this sense QoL can be deemed as the modern operative term that defines the concept of ‘happiness’. When QoL is mentioned without further specifications, it will embrace all factors that contribute towards it, but direct actions on the quality of life – even to improve it – must refer to a series of ‘indicators’. The wellness study, namely QoL evaluation, can also be applied to a condition of imbalance and discomfort to explore the daily impact of a disease or treatment on the life of everybody, on relations, mood and self-awareness. Health is deemed as the subjective, overall well-being of the patient, making room for his daily experiences and subjective viewpoint. During a disease, especially a chronic condition, reference is made to a concept that is more wide ranging than quality of life, namely health-related quality of life (HRQoL). It is ‘the subjective assessment of the impact of disease and treatment across physical, psychological, social and somatic domains of functioning and wellbeing’ (Reviky et al. 2000). In our case, for instance, the impact of HEN on subjects presenting a chronic disease and on their caregivers must be evaluated. In these special conditions, an individual develops a personal evaluation of what good quality of life is on the basis of his physical conditions, and depending on his personality profile, mode of coping and attitude towards health. HRQoL can be distinguished from QoL in that it concerns itself primarily with those factors that fall under the purview of health-care providers and health-care systems.
193.5.2 Measuring HRQoL The HRQoL evaluation has several therapeutic goals and advantages, and implies the use of measurement scales and statistical correlations that embrace several domains. The most widespread compound measurements are HRQoL profiles generated by self-administered scales, which can be used to measure cross-sectional differences in quality of life between patients at a point in time (discriminative instruments) or longitudinal HRQoL changes in the same patient during a period of time (evaluative instruments). Both discriminative and evaluative instruments must be validated. There are two basic approaches to quality-of-life measurement available, namely generic instruments that provide a summary of well-being and HRQoL, and specific instruments that focus on problems associated with single disease conditions, patient groups, functional areas, or specific treatments applied. Approaches are not mutually exclusive. Of the many questionnaires available, the SF-36 (Ware et al. 1993) and Psychological Wellbeing Questionnaire (Ryff and Keyes 1995) stand out for the considerable availability of data on their validity and reliability. Especially, the SF-36 questionnaire is short and easy to complete is validated in many countries, and is therefore often used for comparative
193 Biopsychosocial, Behavioural Aspects and Quality of Life with Home Enteral Nutrition
3125
s tudies of data. Other self-administered scales, like the Satisfaction profile (SAT-P; Majani and Callegari 1998), have a ‘user-friendly’ structure and only require brief administration and scoring times. It has an analogical, non-verbal response and is ideal for administration to both the elderly and subjects with writing problems. The use of standardised instruments saves time and efforts required to perfect new custom-made instruments. There are no specific scales to evaluate the impact of HEN on HRQoL. Loeser et al. (2003) constructed an ‘ad hoc’ scale created by crossing a specific PEG impact evaluation module in conjunction with the QLQ-C30 scale for evaluations carried out during their study. But this scale has not been validated as yet. However, it confirms the need for specific instruments for subjects treated with home enteral nutrition.
193.5.3 N on-Standardised and/or Qualitative Instruments: Semistructured Interviews and Subjective Evaluation scales Authors are increasingly integrating these standardised scales with semistructured interviews (for qualitative evaluations) or non-standardised scales to extend the information collected. Semistructured interviews were, for instance, used to analyse the impact on QoL in patients wearing PEGs (Brotherton et al. 2006). Other cases required a list of the advantages and disadvantages perceived during HEN treatment (Paccagnella et al. 2007) or scales to evaluate subjective tolerance to nutritional techniques (Roberge et al. 2000). Aspects of environmental impact have also been evaluated, since the therapy is a treatment that must be administered in the home of the patient. In our recent paper (Paccagnella et al. 2007) we used the Environmental Impact Exam (EIE), a questionnaire that was specially designed to study the perception of the living environment in patients and relatives during the illness, on their return home.
193.5.4 Why Measure the Quality of Life in HEN The concepts of QoL and of HRQoL are central in the contemporary bioethical debate. Several controversies are in progress in medical and bioethical literature concerning placement of feeding tubes in patients presenting chronic diseases. These controversies centre on the decision concerning therapy, and also on the subjects involved in the same decisions, on ethical and legal aspects and on the role of various members of the multidisciplinary team in appropriately defining the characteristics of patients who need treatment with feeding tubes. The organisation that HEN requires can be naturally modulated or evaluated in response to the insurance and/or financial system provided by the country that envisaged it. Some questions should be asked in this framework. Is there a way of offering patients a quality intervention that can reduce risks associated with HEN? Is it possible to find a model that marries the culture of medicine with the current patient-centred approach? Which parameters, irrespective of the clinic, could show appropriate relations between the efficacy of the intervention, its cost and the well-being of the single patient? Feeding tubes are unattractive and low-profile ‘button’ type tubes are the choice for gastric or jejunal feeding in some, especially the young. For feeding, low-profile tubes require brand-specific adapters or extension tubes that can wear out over time; hence, patients will need a supply of these. Regular gastric tubes may occasionally be too long and cumbersome, interfering with activity,
3126
A. Paccagnella et al.
c lothing, and moving. AN-related changes are many, and the subject must be actively involved in preparing his return to his life and home. The therapeutic nutrition programme can be adequate for hospital treatment, but it must be adapted and specially designed for the patient when he is sent home to ensure integration – as far as possible – with the characteristics and conditions of his life and work (see Box 193.1). The use of syringes, pumps and other medical aids must be discussed and evaluated with the patient to suit his preferences (for instance, many patients who use the pump prefer night feeding). All these aspects lead to evaluate, along with the patient, the impact of treatment on his quality of life in order to implement all improvements required.
Box 193.1 Two Stories of Hen Mrs. Piera is 75 years. She cannot eat due to a tumour in the larynx that caused her to undergo several laryngectomies with subsequent progressive dysphagia. Every morning Mrs. Piera goes out to buy bread and the newspaper. She is autonomous and keeps her house and self in order. She regularly meets her friends and often plays with her grandchildren. She infuses nutrients and water into her PEG four times a day. For Mrs. Piera, HEN is a byword for autonomy and an acceptable quality of life. She has a positive attitude towards the PEG. Carlo is 45 years. He suffers from a significant gastroparesis with chronic abdominal pain. He has undergone multiple surgeries for duodenal ulcers and obstructions. He is unmarried and lives alone. He has only one PEG through which he is slowly infused a nutrient and water almost the whole day. He must do the same at night. His abdominal pains often force him to stay in bed and prevent him from working for days, from going out, from meeting friends or a woman to form a family. His attitude towards HEN is ambivalent.
193.6 HEN and Its Impact on HRQoL As already mentioned, HRQoL measures the combination between clinical, objective and measurable data, and subjective evaluations. Several studies (Terrel et al. 2004; Brotherton et al. 2006; Paccagnella et al. 2007; Bozzetti 2008) report the significant impact on QoL that, especially, surfaces concerning psychological and emotional factors associated with the inability to eat, taste and share food with others, the subsequent exclusion from family and social life, the inability to share the non-nutritional aspects of food with the social functions that eating food with others implies. Such inconvenience must certainly be associated with the specific characteristics of the treatment, which removes all eating-related factors and requires acceptance of treatment centred solely on feeding, something patients consider as one of the major disadvantages of the treatment. First difficulties arising from PEG feeding include symptoms such as vomiting, diarrhoea, infection of the PEG site and leakage that are reported as problems impacting on QoL by patients who wear a PEG (Table 193.4). Patients also report several psychosocial inconveniences. In fact, if on the one hand they are relieved of the pressure to consume an oral diet, and appreciate recovery of physical wellbeing and are less concerned over the risks related to their health (e.g. they are less stressed by problems related to dysphagia or risks related to malnutrition), on the other hand they are disturbed by several psychosocial effects related to PEG that have an impact on their QoL (Table 193.5). Three review articles evaluate the latest studies in the field of HRQoL and HEN (Brotherton and Judd 2007; Bozzetti 2008; Sampson et al. 2009) concluding that clearly tube feeding has an impact
193 Biopsychosocial, Behavioural Aspects and Quality of Life with Home Enteral Nutrition
3127
Table 193.4 Symptoms associated with enteral feeding subjectively perceived as impacting on QoL (Roberge et al. 2000) Low-impact symptoms Intermediate-impact symptoms High-impact symptoms Nausea Pain Fatigue Vomiting Dyspnoea Constipation Insomnia Diarrhoea Appetite Enteral feeding can cause symptoms of different kind that can be differently tolerated. This table indicates the main symptoms experienced by patient and the degree of impact as subjectively experienced divided by low-, intermediateand high-impact symptoms
Table 193.5 Psychosocial symptoms, subjectively perceived, as PEG-related problems impacting on QoL (Bannermann et al. 2000; Roberge et al. 2000; Verhoef and Van Rosendaal 2001; Malone 2002; Gee et al. 2005; Brotherton et al. 2006; Paccagnella et al. 2007; Rogers et al. 2007) Problems with sleep Difficulty in finding a place to feed Missing being able to eat and drink Problems in social occasions when food is shared Perception of negative attitudes of others towards feeding them Feeling a burden to family members Poor body image Fear about feeding tubes Sensory deprivation Grief Anger Depression Frustration Limitations concerning social life and travel Interference with family life Interference with intimate relationships Interference with social activities Interference with hobbies Discomfort experienced while dressing and washing and restricted choice of clothing In the table are summarized the main inconvenience perceived by patients wearing a PEG. Data are collected with quantitative and qualitative measures and describe all the daily discomforts expressed by patient in managing artificial nutrition
on the patient’s quality of life. But these reviews underscore a varied and heterogeneous picture of the HRQoL impact of HEN. The differences depend on several factors, such as variable factors related with the underlying primary diagnoses, access routes, and measuring instruments that produce results that are not always consistent and comparable. Underlying diseases significantly, but not exclusively, include acute (e.g. stroke and head injury) and degenerative (e.g. dementia, Parkinson’s disease and lateral sclerosis) neurological disorders. However, subjects undergoing PEG also count patients with cancer in the head, neck or oesophagus, and patients with cystic fibrosis. Hence, literature describes a very heterogeneous population – in terms of basic diagnosis – of subjects undergoing HEN. Even access routes can be of various types. In the studies mentioned, most patients were fed via PEG, but some studies recruited patients with NGT or NJT. All these differences are an obstacle to defining a coordinated, comparable picture of results, and we can only describe some common emerging themes. But not all results were consistent. In fact, some studies back the theory that HEN is deemed to help maintain QoL although it may be at a reduced level, especially when it is measured in time (Klose et al. 2003) and found PEG subjectively perceived as
3128
A. Paccagnella et al.
useful in maintaining QoL over time, especially in the younger subjects (Weaver et al. 1993). This study found an interesting discrepancy between evaluations made by family members concerning the benefit of the procedure for the patient, and their refusal of applying the procedure to themselves if they were in the same situation. Even subjective opinions of patients on the impact of HEN present some ambivalences that reveal both perception of the utility (or inevitability?) of treatment (‘I cannot do without it’), and an evaluation of its cost in terms of changes brought in their daily lives. Such ambivalences also surfaced in the assessment of other physical features and of the impact of the treatment and, especially, of inconveniences related to the use of stomies and tubes on the physical image. An interesting point surfaces from a longitudinal study, whose authors report that, though most subjects and caregivers have no second thoughts over the decision to have a PEG, this does not necessarily indicate better QoL (Verhoef and Van Rosendaal 2001). In the study conducted by Sampson et al. (2009) that analysed the impact of HEN in subjects presenting dementia, the authors concluded that despite the very large number of patients receiving this intervention, there was insufficient evidence to suggest that enteral tube feeding was beneficial in patients with advanced dementia in influencing survival, quality of life, prevention of bed sores, nutritional and functional condition and behavioural and psychiatric symptoms. All reviews of latest contributions to relations between HEN and HRQoL lead to disappointing conclusions. In short, the data reveals that patients acknowledge the life-saving function of the therapeutic intervention of HEN, but they also declare that tube feeding finally came to dominate their lives and was associated with an appreciable burden of treatment (Jordan et al. 2006). We think that this depends on several factors, especially the fact that eating does not only involve the intake of food, but is associated with complex factors – as mentioned – the lack of which causes psychosocial deprivation in the life of subjects. This is added to the emotional impact (anticipatory grief) generated by basic diseases.
193.6.1 Problems and Limits of the Evaluation of HRQoL Considering the discussion so far, it is hard to evaluate the effect of HEN on the QoL of a subject without taking into account the emotional impact of the basic disease, which is often chronic and has a fatal prognosis. Another problem that should be considered when analysing the aforementioned reports concerns the use of different measurements and scales, which makes it rather difficult to conduct a comparative study. Even the problem of access routes that are neither consistent nor identical in terms of management in these studies is not negligible. Different tubes can produce different reactions both in terms of symptoms and of emotional experience related to the tolerance of treatment and the image of self. Moreover, the number of subjects enrolled in many studies is less than the representative sample. Again, despite the hope of evaluating the progress of the long-term impact on QoL, longitudinal studies have struggled with loss of patient numbers at the time of follow-up data collection. A recent review (Marin Caro et al. 2007) centred on nutritional intervention and QoL in adult oncology patients found that no study is reported to prospectively investigate the effects of EN on QoL.
193.6.2 Caregivers The term caregiver indicates the person who attends to either disabled or sick people, namely professional caregivers and family caregivers are those who look after a person in need. The difference between the two roles concerns their theoretical and practical background, but above all it relates to their motivational features. The professional caregiver’s motivation is occupational, whereas a family
193 Biopsychosocial, Behavioural Aspects and Quality of Life with Home Enteral Nutrition
3129
caregiver is in a position he or she has not freely chosen, and which constitutes an additional role arising from the illness of a loved one. Therefore, family caregivers have to cope with the challenge of a role they were not prepared to take on. Taking care of a sick family member requires a strong commitment that can change in time. Since care recipients are mainly attended to by family caregivers, irrespective of the presence of an external support, focus must be on them, also because the sick person’s overall well-being highly depends on the psychophysical status of the family member who is taking care of him. The term ‘burden’ indicates the overall impact of caring responsibilities on the caregiver, who might feel demoralised, isolated and trapped in the role. Therefore, while assessing care needs, it is also essential to consider the demands of informal caregivers and anything else that might support them in this delicate community care task, both in ‘material’ and ‘moral’ terms.
193.6.3 The Viewpoint of Caregivers Some studies analysed the effect on the subjective well-being of caregivers in managing HEN in various chronic patient types. The scope of these studies was to evaluate the impact of home care that significantly involved caregivers. Caregivers manage all treatment prescribed for non-competent subjects; hence, the importance of evaluating the impact of home care on them and of assessing whether social support, such as help from a family member, hospital assistance and professional help, could have some effect in reducing the caregiver’s burden. Although caregivers are constantly under a lot of stress, there seems to be a strong link between the physical and cognitive disabilities of the patient and the extent of his daily needs, and the well-being of caregivers (Douglas et al. 2005). Studies compared HRQoL in patients and their caregivers. SAT-P test results reveal that the group of caregivers does not differ from the group of patients concerning perceived satisfaction in QoL. In particular, the scores of males fall within the normal range in the considered functions, while females – the majority of caregivers – seem to suffer a greater perceived cost in the management of patients. Perceived satisfaction on sleep-eating-leisure functions and at physical and psychological levels are below the average, when compared to the control group. Caregivers underlined different advantage and disadvantage concerned HEN (Table 193.6) even if management services are always deemed adequate and competent and all caregivers have been pleased by the service received (Silver and Wellman 2002; Brotherton et al. 2006; Paccagnella et al. 2007). Although more attention has been paid to the viewpoint of caregivers, and there is greater awareness of the risk of the burden of managing a chronically ill person at home, less attention is paid to understanding the kind of support required to reduce the stress of caregivers. When the location of therapy is shifted from the hospital to the home of patients, the assessment of the impact of managing therapy should take into account all facilitating features that can be activated.
193.6.4 Health-Care Operators There are no specific comparative studies on the viewpoint of health-care operators and those of patients or caregivers. Evaluations are often based either on clinical data or on instruments such as, for instance, the Karnofky index that does not cover all areas of autonomy and quality of life evaluation of the subject. The important factor in evaluating the quality of life of the patient after the therapeutic intervention concerns the need to train operators to make a biopsychosocial evaluation of subjects in order to provide an intervention designed to suit both the effects on the body of the person and on the person as a whole.
3130 Table 193.6 Main disadvantages and advantages of HEN pointed Paccagnella et al. 2007) Main disadvantages Seeing the missing sensory aspects of nutrition in the patient Problems in physical/sexual relations between adults Problems due to emotional relations between the patient and his caregiver Remarks on the limited freedom of the caregiver himself
A. Paccagnella et al. out by caregivers (Brothertorn et al. 2006; Main advantages Improved physical conditions General comfort/practicality of HEN Hope in the patient’s survival Feel less anxious or fearful of complications of eating by mouth To have the relative back home
Lack of autonomy of the patient Technical difficulties associated with HEN management QoL of patients considered boring due to the loss of pleasure generated by food and the act of eating Lack of quality of life Problems caused by reduced patient mobility Strong commitment to treatment management, which diminishes their own autonomy In the table are listed the main advantages and disadvantages pointed out by caregivers of HEN patients as subjectively perceived. To each subject was asked to point out the main disadvantages and the advantages, as they perceived it, in managing the home enteral nutrition of their relative. Results are then categorised into main classes of disadvantages and advantages by different judges. It is a qualitative method to evaluate the impact of home enteral nutrition at home and to have more information about quality of life of caregivers. In the table are pointed out main results
193.7 The Link Between HEN, QoL and ethics Hebuterne (2005) presents us with questions on HEN that force us to consider its use: Is it well tolerated by the patient? Does it improve the clinical condition of the patient or does it only maintain it unchanged? Does it improve the quality of life of the patient or does it leave it unaltered? Does it improve the outcome of the patient and his survival? These questions make room for many considerations, and do not make it easy to formulate univocal answers. National and international scientific societies have long wondered about the ethical aspects of artificial nutrition. To date, it is a matter for discussion whether AN must be deemed as a medical act or a compulsory caregiving intervention. In 2005,2005 the case of Terry Schiavo fuelled an international debate. The patient was in a permanent vegetative condition, and died after suspension of PEGbased HEN. The suspension, which was authorised by judiciary authorities on request of the husband, supported the former wish of the patient (Charatan ). Even our clinical experience counts several cases of patients in whom the decision to commence HEN raised more than a few questions and doubts (Box 193.2). It is interesting to notice that a study conducted in the early 1990s by Solomon et al. (1993) and based on interviews with 1,146 doctors and nurses, found that while the feasibility of using a ventilator or any other end of life support was questioned, there was extensive consent in not discontinuing nutrition and hydration. On the other hand further studies (Callahan et al. 1999) report that home care with HEN is not preceded by appropriate information for the patient or his relatives, and that relatives and other care givers request doctors to especially use the PEG (Table 193.7). Over the past 20 years, the debate has been enriched by extensive literature, and every scientific society specialised in these features of home care enlarge on the ethical aspect of HEN in their guidelines. Basically, all agree on the advisability of suspending HEN when it becomes ‘futile treatment’. Unfortunately, the concept of ‘futile treatment’ lacks a consistent definition, though some of them consider it as either ineffective or incapable of achieving a desired goal or result, despite heroic efforts (Andrews and Geppert 2007). On the other hand, it is very hard to establish ‘if’ and ‘when’
193 Biopsychosocial, Behavioural Aspects and Quality of Life with Home Enteral Nutrition
3131
Box 193.2 The case of Paola Mrs. Paola, aged 70 years was diagnosed with left parietal meningioma: a benign tumour for which she underwent surgery. At 74, she underwent surgery again for a recurrence. The next year, following a left sylvian infarction she became completely hemiparetic and aphasic. During the next months she became totally dependent on the care of her relatives. At 77, she developed a third recurrence of the tumour that the neurosurgeons of her city deem inoperable. Therefore, the relatives decided that Paola should undergo surgery again in the neurosurgery department of another country. During the postoperative phase a complication forced Paola to undergo further resection surgery of the tumour. At that point Paola was totally unconscious, bed-bound and fed with an NG tube. When she returned to her city, a neurologist wrote: ‘The patient is drowsy. She opens her eyes only under intensive stimulation. Pain stimuli evoke movement in the left lower limb. She does not answer verbally, nor can she understand when spoken to.’ The picture is compatible with a permanent vegetative condition. At the age of 78 a PEG was positioned for HEN, as requested by her relatives. One year elapsed from the placement of the PEG to her death. The survival of Paola was influenced by the HEN, but did this prolong her life or her agony? Photo: Paola at the time of PEG placement.
Table 193.7 Key features of home enteral nutrition 1. HEN is a set of organisational modes for EN administration at home and has the same indications as PN. 2. HEN is administered to subjects who cannot be fed by mouth due to an underlying disease and who are sent home on artificial nutrition (AN) after a certain period of hospital treatment. 3. HEN must basically either prevent or reduce the risk of malnutrition (or undernutrition) that can act synergically with the basic disease, increasing the risk of complications and of patient decease. 4. Enteral feeding tubes can be divided into two main categories: (a) those positioned through the nose (nasogastric, nasoduodenal and nasojejunal tubes), recommended for at most 30 days; (b) those requiring minor surgery (percutaneous endoscopic gastrostomy, PEG; percutaneous endoscopic jejunostomy, PEJ), which are recommended for EN > 30 days. 5. Enteral tube feeding formulas are divided into two categories: polymeric or elemental and semi-elemental are industrially prepared mixtures that fully meet the need for macro- and micronutrients and water in patients who cannot meet daily requirements via oral intake but possess a functional digestive system. 6. SNG and PEG placement causes several mechanical complications such as mechanical traumas (haemorrhages, perforations) and inhalation of food into airways. 7. Literature reported a 1-month mortality rate of 8–41%, a 1year mortality rate of 40–65% after feeding tube placement, and a median survival rate that was well under a year.
3132
A. Paccagnella et al.
HEN should be implemented, depending on the HRQoL evaluation of a patient. This seems absolutely unacceptable when the patient lacks self-awareness. A paradox that we experience daily, and which is already documented in the USA (Hoefler 2000), is that many people are inclined to make their relatives (e.g. elderly parents and children in a permanent vegetative condition) undergo HEN even when some cases risk the administration of futile treatment, but the same people usually say that they would not like to undergo HEN or other forms of hydration, when they are questioned about what they would choose for themselves. The fact remains that in most European countries, in America and in Asia, AN (to which HEN belongs) is not deemed as a basic duty of care – such as ‘eating’, considering the theories explained so far – but as a ‘medical treatment’, whose implementation always requires the consent of the patient or of his relatives (NICE 2005), and which can be discontinued on the basis of clinical reasons or administered for a palliative purpose, depending on the clinical condition of the patient.
193.8 Future Developments Many quality of life measurements that are widely used are limited in their ability to record the quality of life of individual patients. These limitations depend on the structure and content of the measurements, how they were developed and their weighting systems. Some of these problems can be overcome by using individualised measurements, but these have their own problems which need further attention. A compromise could lie in using recently developed standardised measurements that are sufficiently wide ranging to include most facets of life that are important to any patient, but which also use direct weighting systems. This should lead to an individualised assessment of patient quality of life. The extent to which such measurements mirror the quality of life of an individual requires further assessment, and the clinical utility and interpretability of these measures must also be established. Concluding, the themes that surfaced from the studies analysed assuredly require further study, but they provide important descriptions of the psychological and social life of the patient, clarifying the key factors that must be considered by all operators in clinical practice.
193.9 Applications to Other Areas of Health and Disease The recent biopsychosocial approach promoted by the World Health Organization wants us to consider the person from a ‘global’ perspective, taking into account the physiological, psychological and social implications of any treatment administered. In special conditions as chronic treatment as HEN (Home Enteral Nutrition) an individual develops a personal evaluation of what good quality of life is on the basis of his physical conditions, and depending on his personality profile, mode of coping, and attitude towards health. For chronic diseases patients should achieve psychological wellness that is commonly defined as quality of life” (QoL) or health-related quality of life (HRQoL). Therefore, all chronic diseases that need a prolonged treatment such as artificial nutrition should be evaluated on health-related quality of life. It is important to choose measures that are diffusely shared to be able to compare results of different studies and to make a follow-up of the evaluation whenever possible. Using measures of health-related quality of life studies represents a new promising field to improve the impact of the therapy in chronic patients and to understand the different subjective outcomes.
193 Biopsychosocial, Behavioural Aspects and Quality of Life with Home Enteral Nutrition
3133
Summary Points • In the chapter we point out the difference between eating and feeding and we designate as ‘eating’ everything that is taken by mouth and has implicit repercussions on the affective, social, cultural and biological framework. It designates that which is commonly and historically called ‘food’ and which envisages the act of eating. • Enteral nutrition is an example of therapeutic intervention (feeding) in which the integrated and holistic dimension that creates wellness is damaged, and subjective balances are modified. AN designates all calories and macro- and micronutrients that are infused through a tube to ensure nourishment to the body, but which deprive the person of the psychosocial implications of ‘eaten food’. • Patients undergoing home enteral nutrition should achieve psychological wellness that is commonly defined as ‘quality of life’ (QoL). In these cases of chronic condition, reference is made to a concept that is more wide ranging than quality of life, namely health-related quality of life (HRQoL). • The data reveals that patients acknowledge the life-saving function of the therapeutic intervention of home enteral nutrition, but they also declare that tube feeding finally came to dominate their lives and was associated with an appreciable burden of treatment. • To date, it is a matter for discussion whether artificial nutrition must be deemed as a medical act or a compulsory caregiver intervention. Key Terms Biopsychosocial approach: An approach promoted by the World Health Organization that centres treatment on the patient and his relatives who become the lead actors in the management of the disease. This approach shifts focus from the disease ‘of the body’ to the disease ‘of the patient’ who is deemed a person with an inner world of his own and lives in a specific physical and social framework. Quality of life quality of life (QoL): is used in health care to refer to an individual’s emotional, social and physical well-being, including their ability to function in the ordinary tasks of living. Health-related quality of life: The wellness study can also be applied to a condition of imbalance and discomfort to explore the daily impact of a disease or treatment on the life of everybody. HRQoL refers to patient outcome measures that extend beyond traditional measures of mortality and morbidity to include such dimensions as physiology, function, social activity, cognition, emotion, sleep and rest, energy and vitality, health perception and general life satisfaction. Home enteral nutrition: AN comprises artificial parenteral nutrition (PN) and artificial enteral nutrition (EN). EN allows nutrients to be administered directly into the gastrointestinal system with special probes that are inserted either through the nose or stomies created through the abdominal wall. Home enteral nutrition is when the treatment is moved to the home of the patients and it is often administered by a caregiver. Caregivers: The term caregiver indicates the person who attends to either disabled or sick people, namely professional caregivers and family caregivers are those who look after a person in need.
References Andrews MR, Geppert CMA. In: DeLegge MH, Mattox T, Mueller C, Worthington P, editors. The A.S.P.E.N. nutrition support core curriculum: a case-based approach-the adult patients. Silver Spring: ASPEN; 2007. p. 740–60. Arends J, Bodoky G, Bozzeti F, Fearon K, Muscaritoli M, Selga G, Van Bokhorst-de van der Schueren MA, von Meyenfeldt M, DGEM (German Society for Nutritional Medicine), Zurcher G, Fietkau R, Aulbert E, Frick B,
3134
A. Paccagnella et al.
Holm M, Kneba M, Mestrom HJ, Zander A; ESPEN (European Society for Parenteral and Enteral Nutrition). Clin Nutr. 2006;25: 245–59. Aspinwall L, Staudinger U. A psychology of human strengths. Washington: American Psychological Association; 2003. Bankhead R, Fang, JC. In: DeLegge MH, Mattox T, Mueller C, Worthington P, editors. The A.S.P.E.N. nutrition support core curriculum: a case-based approach-the adult patients. Silver Spring: ASPEN; 2007. p. 233–45. Bankhead R, Boullata J, Brantley S, Corkins M, Guenter P, Krenitsky J, Lyman B, Metheny NA, Mueller C, Robbins S, Wessel J; A.S.P.E.N. Board of Directors. J Parenter Enteral Nutr. 2009;33:122–66. Bannerman E, Pendlebury J, Phillips F, Ghosh S. Eur J Gastroenterol Hepatol. 2000;12:1101–109. Bozzetti F. Curr Opin Clin Nutr Metab Care. 2008;11:661–5. Brotherton A, Abbott J, Aggett P. J Hum Nutr Diet. 2006;19:355–67. Brothertorn AM, Judd PA. J Hum Nutr Diet. 2007;20: 513–22. Buchman, AL. In: Lochs H, Thomas DR, editors. Home care enteral feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, Vol 10. Basel: Karger; 2005. p. 143–66. Cady J. Clin J Oncol Nurs. 2007;11:875–80. Callahan CM, Haag KM, Buchanan NN, Nisi R. J Am Geriatr Soc. 1999;47:1105–109. Centers for Medicare and Medicaid Services. MDS active resident information report: third quarter, physical functioning and structural problems eating ADL support. Baltimore: Centers for Medicare and Medicaid Services; 2005. Charatan F. BMJ. 2005;330:1467. Chen Y, Peterson SJ. Nutr Clin Pract. 2009;24:344–55. Crosby J, Duerksen D. Dig Dis Sci. 2005;50: 1712–17. DiBaise JK, Scolapio JS. Gastroenterol Clin North Am. 2007;36:123–44. Douglas SL, Daly BJ, Kelley CG, O’Toole E, Montenegro H. Chest. 2005;128:3925–36. Fayers PM, Machin D. Quality of life. Assessment, analysis and interpretation. Chichester: Wiley; 2001. Gee L, Abbott J, Hart A, Conway SP, Etherington C, Webb AK. J Cyst Fibros. 2005;4:59–66. Hebert RS, Arnold RM, Schulz R. J Pain Symptom Manage. 2007;34:539–46. Hebuterne X, Bozzetti F, Moreno Villares JM, Pertkiewicz M, Shaffer J, Staun M, Thul P, Van Gossum A; ESPENHome Artificial Nutrition Working Group. Clin Nutr. 2003;22:261–6. Hebuterne, X, Schneider, SM. In: Lochs H, Thomas DR, editors. Home care enteral feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, Vol 10. Basel: Karger AG; 2005. p. 89–102. Hoefler JM. Death Stud. 2000;24:233–54. Iyer KR, Crawley TC. Gastrointest Endosc Clin N Am. 2007;17:717–29. Jones B, Holden C, Stratton R, Micklewright A, Dalzell M. Annual BANS report 2007. Artificial nutrition support in UK 2000–2006. A report by the British Artificial Nutrition Survey (BANS), a committee of BAPEN (The British Association for Parenteral and Enteral Nutrition); 2007. Jordan S, Philpin S, Warring J, Cheung WJ, Williams J. J Adv Nurs. 2006;56:270–81. Klose J, Heldwein W, Rafferzeder M, Sernetz F, Gross M, Loeschke K. Dig Dis Sci. 2003;48:2057–63. Lochs H, Dejong C, Hammarqvist F, Hebuterne X, Leon-Sanz M, Schϋtz T, Van Gemert W, Van Gossum A, Valentini L, DGEM (German Society for Nutritional Medicine), Lϋbke H, Bischoff S, Engelmann N, Thul P; ESPEN (European Society for Parenteral and Enteral Nutrition). Clin Nutr. 2006;25:260–74. Loeser C, von Herz U, Küchler T, Rzehak P, Müller MJ. Nutrition. 2003;19:605–11. Magnuson BL, Clifford TM, Hoskins LA, Bernard AC. Nutr Clin Pract. 2005; 20:618–24. Majani G, Callegari S. Test SAT-P. Soddisfazione Soggettiva e Qualità della Vita. Trento: Centro Studi Erickson; 1998. Malone M. J Parenter Enteral Nutr. 2002;26:164–8. Marin Caro MM, Laviano A, Pichard C. Clin Nutr. 2007;26:289–301. Mitchell SL, Teno JM, Roy J, Kabumoto G, Mor V. JAMA. 2003;290:73–80. Morello M, Marcon ML, Pizzolato D, Baruffi C, Giometto M, Favaro V, Rebuffi S, Cenerelli P, Calabrò M, Spinella N, Tessarin M, Paccagnella A. Nutr Clin Pract. 2009;24:635–41. National Institute for Health and Clinical Excellence (NICE). Nutrition support in adults: oral supplements, enteral tube feeding and parenteral feeding. London: Second draft for consultation; 2005. Paccagnella A, Mauri A, Berto R, Falchero S, Baruffi C, Marcon ML, Faronato PP, Dal Ben G, Foscolo G. Minerva Med. 2007;98:5–17. Paccagnella A, Baruffi C, Pizzolato D, Favaro V, Marcon ML, Morello M, Semenzin M, Rebuffi S, Fossa E, Baronato P, Spinella N, Tessarin M, Foscolo G. Clin Nutr. 2008;27:378–85. Revicki DA, Osoba D, Fairclough D, Barofsky I, Berzon R, Leidy NK, Rothman M. Qual Life Res. 2000;9:887–900. Ryff CD, Keyes CL. J Pers Soc Psychol. 1995;69:719–27. Roberge C, Tran M, Massoud C, Poirée B, Duval N, Damecour E, Frout D, Malvy D, Joly F, Lebailly P, Henry-Amar M. Br J Cancer. 2000;82:263–9. Rogers SN, Thomson R, O’Toole P, Lowe D. Oral Oncol. 2007;43: 499–507. Sampson EL, Candy B, Jones L. Enteral tube feeding for older people with advanced dementia review. Cochrane Database Syst Rev. 2. 2009; Art.No.:CD007209. DOI:10.1002/14651858.CD007209.pub2.
193 Biopsychosocial, Behavioural Aspects and Quality of Life with Home Enteral Nutrition
3135
Schneider SM, Raina C, Pugliese P, Pouget I, Rampal P, Hebuterne X. J Parenter Enteral Nutr. 2001;25:203–9. Schrag SP, Sharma R, Jaik NP, Seamon MJ, Lukaszczyk JJ, Martin ND, Hoey BA, Stawicki SP. J Gastrointestin Liver Dis. 2007;16:407–18. Silver HJ, Wellman NS. J Am Diet Assoc. 2002;102:831–6. Solomon MZ, O’Donnell L, Jennings B, Guilfoy V, Wolf SM, Nolan K, Jackson R, Koch-Weser D, Donnelley S. Am J Public Health. 1993;83:14–23. Terrell JE, Ronis DL, Fowler KE, Bradford CR, Chepeha DB, Prince ME, Teknos TN, Wolf GT, Duffy SA. Arch Otolaryngol Head Neck Surg. 2004;130:401–8. Verhoef MJ, Van Rosendaal GM. J Clin Gastroenterol. 2001. 32:49–53. Ware JE, Snow KK, Kosinski M, Gandek B. SF-36 health survey: manual and interpretation guide. Boston: The Health Institute; 1993. Weaver J.P., Odell P, Nelson C Arch. Fam. Med. 1993;2:953–56. World Health Organization (WHO). Ottawa charter for health promotion: an international conference on health promotion, the move towards a new public health. Geneva: World Health Organization; 1986.
wwwwwwwwwwwwwww
Chapter 194
Quality of Life, Diet, and Behavior in Cancer Brenda Larson and Aminah Jatoi
Abbreviations CACS TNF-alpha IL-1 IL-6 IFN-gamma
Cancer-related anorexia cachexia syndrome Tumor necrosis factor alpha Interleukin 1 Interleukin 6 Interferon gamma
194.1 Introduction Anorexia and weight loss are common in patients with cancer and carry a tremendous impact on the social and psychosocial aspects of cancer care. Studies in patients with advanced disease have suggested that as many as 80% of patients experience anorexia, or loss of appetite, and as many as 82% experience weight loss (Poole and Froggatt 2002). This symptom and sign have major psychosocial ramifications for patients.
194.2 Impact of Anorexia and Weight Loss Anorexia and weight loss have ramifications relevant to patients’ prognosis and quality of life. Dewys and others conducted the first major study to support the prognostic significance of weight loss in cancer patients. Focusing upon 3041 patients treated on 12 Eastern Cooperative Oncology Group protocols, these investigators observed that weight loss of >5% of baseline weight prior to the start of cancer treatment predicted a shortened survival (Table 194.1). Patients with such weight loss also manifested lower response rates to chemotherapy. Weight loss tended to correlate with performance status (Dewys et al. 1980). Along these same lines, a study in which a questionnaire was administered to 1115 patients with advanced cancer, showed that a patient’s own assessment of his or her nutritional status, including their own opinion of their appetite and food intake, carried prognostic
A. Jatoi (*) Department of Oncology, Mayo Clinic, 200 First Street SW, Rochester, Minnesota, USA e-mail:
[email protected] V.R. Preedy et al. (eds.), Handbook of Behavior, Food and Nutrition, DOI 10.1007/978-0-387-92271-3_194, © Springer Science+Business Media, LLC 2011
3137
3138
B. Larson and A. Jatoi
Table 194.1 Weight loss is associated with major differences in survival in cancer patients (Adapted from Dewys et al. 1980) Weight loss and early death Median survival (weeks) Cancer No weight loss Weight loss P-value Lung, small cell 34 27 35 with Major risk factors
Healthy eating advice
Fig. 211.1 Standard procedure for treating obesity in Type 2 diabetes. The standard procedure takes the Type 2 diabetes patients into account as well and constitutes a simple straightforward approach to successful long-term treatment of the obese Type 2 diabetes patients. To integrate the factors of diet, exercise, drug therapy, and environment, this procedure is a key to successful long-term treatment for obesity
central nervous system and influences hunger and meal-seeking behavior (Campfield et al. 2003). Blood glucose dynamics and diet are linked in a complex relationship, and strategies to regulate blood glucose levels must be designed depending on the diet. Controversy still exists regarding whether lowering of dietary fat intake enhances weight loss by itself, rather than through energy restriction. Majority of the evidence shows that energy restriction is the overriding factor linked to weight loss in type 2 diabetes after the consumption of usual lowfat, high-carbohydrate diets (NIH 1998). While weight loss relates to energy deficit alone, the composition of the weight-reducing diet can, however, affect glycemia and lipid levels. Replacement of saturated fat with either monounsaturated fat or carbohydrates was shown to equally affect weight loss in obese diabetic subjects and positively affect lipid levels, supporting other evidence that energy content is the major determinant of weight loss (Gumbiner 1999); the replacement of saturated fats was achieved by advising the subjects to replace certain foods. The study questions the dogma that still governs dietary instruction in diabetes, and emphasizes the need to make dietary changes as simple as possible for diabetic patients already burdened with the procedures associated with managing diabetes (NIH 1998). As yet, no scientific evidence exists to show that greater or more sustained weight loss occurs in diabetes by such manipulations as the intake of foods with low glycemic index or raising the protein proportion in the diet, although there is some concern with regard to the possible adverse effects associated with the consumption of high-protein foods by diabetic subjects. Similarly, very-low-calorie diets result in more rapid weight loss initially, but the weight loss is not sustained in the long term (Heilbronn et al. 1999). Such low-calorie diets could predispose to greater binge-eating activity and are recommended in medical emergencies, and should be accompanied with careful monitoring of glucose levels and early reduction in medication (Garner et al. 1998). To conclude, multiple components of a healthy diet, e.g., high fiber and low saturated fats, reduce diabetes risk and contribute to sustained weight loss; they should therefore be included in long-term
3378
Y.-X. Li et al.
lifestyle interventions. Dietary composition plays a key role in diabetes prevention, primarily by sustaining long-term weight loss and secondarily by initiating weight loss and reducing the risk of diabetes.
211.2.4 Nutritional Education and Coping Skills Training in Diabetes Prediabetic individuals or those with diabetes should receive individualized nutritional education, preferably provided by a registered dietitian familiar with diabetes nutritional education. Behavioral factors (including missing meals, alcohol intake, lack of exercise, and poor recognition of the need to self-treat) also contribute to hypoglycemia risk. These risk factors can be modified with patient education (Boyle et al. 2007). Autonomy in daily treatment will enable patients to introduce behavioral adaptations in their daily life, thereby allowing them to control the disease throughout their life-span. Nutritional education would allow for a permanent, voluntary, and conscious change in habits and eating behaviors. Eating behavior appears to be a reflection of the individual’s personality, with its strengths and weaknesses; rational beliefs; and family and personal history. Major barriers to weight loss that are often overlooked are psychological and behavioral factors. Many people eat in response to negative emotions. To save time, questionnaires can be sent out to the patient as a means of self-evaluation before their weight management appointment; the successful completion of the questionnaires can serve as an indicator of the patient’s level of motivation and compliance. Psychological evaluation should include determining possible underlying depression. If the patient is found to be suffering from depression, this factor should be considered as part of the patient’s weight management (Plodkowski et al. 2003). The six key behaviors (strategies) that successful weight-loss maintainers use include (i) a lowcalorie, low-fat diet; (ii) eating breakfast; (iii) consistent patterns of eating; (iv) high levels of physical activity; (v) regular self-monitoring of weight; and (vi) preventing weight regains from turning into big relapses (Wing et al. 2001). Diabetes requires a substantial degree of patient involvement for effective self-management. Although diabetes education has been the standard of care, it is clear that provision of knowledge alone does not change behavior. Coping skills training (Fig. 211.2) is a cognitive–behavioral intervention that focuses on improving competence and mastery by retraining
Illnes or treatment attribution of symptoms Illness representation Phyical symptoms and severity
Coping
Evaluation Emotional Representation Distress Anxiety and depression
Coping Illnes beliefs and consequence
Fig. 211.2 A dynamic model of coping skills training in health and Illness. This model includes the treatment perceptions and the common-sense model of self-regulation behaviors. It reflects a dynamic interaction between the three stages (representation, coping, and appraisal) and parallel processing: cognitive and emotional processes may operate independently, although they generally interact
211 Treatment of Diet-Related Disorders in Adult Diabetes
3379
inappropriate or nonconstructive coping styles and patterns of behavior into more constructive behavior. Cognitive–behavioral modification is comprised of three steps: (i) recognition of thoughts and feelings; (ii) problem solving; and (iii) guided self-dialogue. The first step is encouraging the individual to reflect on how they think and then respond to situations. The individual’s thoughts are then assessed to determine whether the thoughts are based on fact or assumption. Once the thoughts are assessed, the next step is to solve the patient’s social problem(s). The third step is teaching the individual to use his or her thoughts to guide the decision made in the previous step. A pen and paper should be used when teaching this skill. Group members can list their negative thoughts and then the member and the group can formulate alternate positive thoughts to counter the negative thoughts (Grey et al. 2004). However, further studies on the relationship between behavioral interventions using problem solving and glycemic and psychosocial outcomes in adults should be carried out. Coping skills training may still be an important methodology for health-care providers to use in assisting patients to control blood glucose levels and lose weight, in combination with nutritional education and exercise to decrease BMI, hypertension, or hyperlipidemia (Grey et al. 2004).
211.3 Gastrointestinal Complications in Diabetes 211.3.1 Overview GI motility disturbances, including esophageal motor dysfunction, gastroparesis, intestinal enteropathy (which can cause diarrhea and constipation), are often found in a substantial number of patients with diabetes mellitus (Fig. 211.3). Cross-sectional epidemiological studies suggest that the prevalence of GI symptoms is increased in both type 1 and 2 diabetic subjects (Bytzer et al. 2001a). Some studies have noted that the incidence of GI motility disturbances is higher in female diabetic patients than in males (Bytzer et al. 2001b). Gastropathy may result in delayed emptying of solids and ingestible particles, rapid emptying of liquids, bezoar formation, malnutrition, and weight loss. The causes of GI motility disturbances remain unclear. Autoimmune damage; metabolic insults that alter critical cellular pathways and essential trophic factor signaling, resulting in smooth muscle atrophy and neural apoptosis; and transdifferentiation (Horvath et al. 2006; Rayner et al. 2006) have been suggested as causes in the literature. Other potential etiologies of intestinal dysfunction in patients with diabetes include ischemia and hypoxia due to microvascular disease of the GI tract; mitochondrial dysfunction (Leinninger et al. 2006); the formation of irreversible, advanced glycation end products; and peroxynitrite-mediated endothelial and enteric neuron damage (Hoeldtke et al. 2002). In vitro studies also suggest that hyperglycemia can induce apoptosis of enteric neurons and impair phosphatidylinositol 3-kinase (PI3K) pathway activity. However, apoptosis can be prevented by glial cell-derived neurotrophic factors (Rayner et al. 2006; Leinninger et al. 2006). Several issues should be considered while treating a patient with GI dysmotility, such as the patient’s nutritional status, pain management, prokinetic therapy, symptom suppression, and the consideration of endoscopic or surgical management. Attention to nutritional status is of paramount importance in the management of these patients. Specific deficiencies should be identified and appropriate replacements instituted (Sellin et al. 2008).
3380
Y.-X. Li et al.
Dysphagia
Gastroesophageal Reflux disease Gastroparesis Upper gastrointestinal tract Lower digestive tract Constipation
Diabetic diarrhea Celiac disease
Fig. 211.3 Possible complications involved in gastrointestinal motility disorder. Diabetes can potentially affect any part of the gastrointestinal tract. The common gastrointestinal complications of diabetes include dysphagia, gastroesophageal reflux disease, gastroparesis, diarrhea, and constipation
211.3.2 Esophageal Complications and Treatment Up to 50% of diabetic patients have esophageal abnormalities, including GI motility disturbances and/or acid reflux (Kinekawa et al. 2001). In a previous study, esophageal symptoms were found to be slightly more prevalent in patients with diabetes than in control subjects; however, they were proportionately less common than measured changes in motility (Talley 2003). Esophageal complications of diabetes can be successfully treated. Gastroesophageal reflux disease (GERD) can be managed effectively with medicines that are conventionally used to reduce acid reflux. Antireflux surgery, however, is recommended only for patients with disease that is refractory to medical treatment. Patients with oral and esophageal candidiasis are treated effectively by improving glycemic control and administering oral antifungal agents, such as fluconazole. Symptomatic dysphagia, however, is more difficult to manage, particularly in the presence of advanced diabetic motor neuropathy. In these cases, early diagnosis of dysphagia is important, and glycemic control should be resorted to for treating this reversible condition (Sellin et al. 2008).
211.3.3 Diabetic Gastroparesis and Treatments 211.3.3.1 Symptoms and Diagnosis Diabetic gastroparesis usually occurs during the onset of chronic diabetes; it is a frequent and sometimes troubling complication of the disease. Epigastric fullness, bloating, nausea, and vomiting may be indications of gastroparesis. These symptoms, although nonspecific, are caused by delayed gastric emptying (Lysy et al. 2006). Recurrent symptoms have a deleterious impact on nutrition and
211 Treatment of Diet-Related Disorders in Adult Diabetes
3381
limit the ability of oral hypoglycemic agents to control blood sugar. Recurrent episodes of hyperglycemia have been linked to delay gastric emptying (Lysy et al. 2006). The assumed pathogenesis of this link is that chronic hyperglycemia leads to neuropathic changes and dysfunctional innervations of the stomach and, therefore, to delayed gastric emptying and alteration of antroduodenal motility (Horowitz et al. 2005), which induces a sense of fullness. Hyperglycemia can limit the efficacy of some prokinetic agents (Rayner et al. 2005). These findings were found to be statistically significant, but it is not clear whether they are clinically meaningful. Nevertheless, glycemic control remains the bedrock of therapy for diabetic gastroparesis. Hormonal changes might also be an important factor in diabetic gastroparesis. A diagnosis of diabetic gastroparesis is usually made on the basis of the patient’s medical history. Physical examination does not usually aid in diagnosis. Delayed emptying of a labeled solid material in the absence of any anatomic abnormalities is considered diagnostic for gastroparesis. Concomitant GI can be ruled out through endoscopy. The most useful diagnostic technique for gastroparesis is nuclear scintigraphy. Other techniques that have been used for diagnosing gastric emptying in previous studies include breath tests, the use of radiopaque markers, electrogastrograms, capsule endoscopy, and measurement of antroduodenal motility (Sellin et al. 2008).
211.3.3.2 Dietary Modification Products of fat digestion are known to slow gastric emptying, while nondigestible solids may predispose to gastric bezoar formation. Therefore, small-volume, frequent meals, that are low in insoluble fiber and fat are generally recommended, despite a lack of evidence to support this approach (Rayner et al. 2005; Abell et al. 2006) Thorough chewing, remaining upright for 1–2 h postprandially, and supplementation with multivitamins have also been advocated. Increasing the proportion of energy provided as liquids rather than solids may be beneficial because delayed liquid emptying is rare. An elemental diet is limited by unpalatability (Kuo et al. 2007) but may be a short-term option, despite the lack of evidence to support its superiority over polymeric feeding. Total parenteral nutrition is expensive and impractical, and is associated with potentially serious complications, including sepsis. Indications for nutritional supplementation include weight loss of ³10% during a period of 3–6 months, inability to maintain recommended body weight, and severe symptoms requiring hospitalization or nonpharmacological intervention (e.g., nasogastric tube to relieve nausea and vomiting) (35,36 Rayner et al. 2005; Abell et al. 2006).
211.3.3.3 Therapeutic Options Therapy depends on the severity of symptoms, the ability of patients to maintain adequate nutrition, and their responsiveness to therapy. The primary aim of therapy is to optimize glycemic control. Hyperglycemia induces the development of autonomic neuropathy; blood sugar control can be used as a possible approach to reverse abnormal motility. Beyond the critical factor of glycemic control, the treatment of diabetic gastroparesis is similar, in general, to that of regular gastroparesis. Therapies may involve diet modification, pharmacotherapy, and more invasive approaches to treat “gastric failure.” A low-fat, lowresidue diet accompanied by frequent small meals can minimize postprandial fullness. Avoidance of nutrients that delay gastric emptying (fat, fiber) may also improve gastroparesis (Abell et al. 2006). Several therapeutic agents (Tables 211.1 and 211.2) are available to treat patients with diabetic gastropathy. These include prokinetic and antiemetic agents, as well as analgesics for pain management.
3382
Y.-X. Li et al.
Table 211.1 Treatment options for gastroparesis disorder Treatment Mechanism Adverse effects Evidence comments Symptoms improved in Dystonic reactions, Serotonin(5-HT3)receptor Metoclopramide 25–62% of patients tardive dyskinesia antagonist, central (Reglan), 10 mg extrapyramidal dopanine(102)receptor four times daily symptoms antagonist hyperprolactinemia Physicians should discuss the Normalize gastric slow-wave risk of tardive dyskinesia dysrhythmias by inhibiting with their patients and gastric smooth muscle document this discussion relaxation produced by in the medical records dopamine Most studies are open-label Nausca, vomiting Erythromycin 250 mg Motilin receptor agonist design abdominal pain: three times daily Prokinetic effects via action antibiotic resistants on gastroduodenal motilin receptors Not a ture prokinetic agent Salivation : blurred Nonspecific chalinergic Bethanechal vision : abdominal muscarinic receptor (Urechaline) 25 mg cramps and bladder agonist four times daily spasm Most studies are open-lable Inhibits acetylchaline release Batinlitmen laxin design from synaptic vosicles in type A pylorus (Botox) No well-designed studies for Surgery Gastric decompression, partial diabetic gastroparesis gastrectomy with studies are nonrandomoized, Roux-en-Y unblendded, or care series gastriojejunostomy Possible infection, No well-designed studies; Gastric electric Electric stimulation gastric erosion more data are needed stimulation with high-energy, long-duration pulses This table lists the key facts of treatment options for gastroparesis. Treatments are listed in order from most to least likely to be used (Sellin et al. 2008). Therapeutic considerations depend on the severity of symptoms, the ability of the patient to maintain adequate nutrition and their responsiveness to therapy. Generally, more severe symptoms still require pharmacologic intervention
Table 211.2 Common used prokinetic drugs Drug Erythromycin
Mechanism of action Motilin agonist
Cisapride
5-HT4-receptor agonist: 5-HT3-receptor antagonist D2-receptor antagonist: 5-HT3-receptor antagonist; 5-HT4-receptor agonist D2-receptor antagonist
Metoclopramide
Domperidone
Administration route IV, oral Oral
IV, SC, IM, oral
Oral
Dose(mg) 50–250 (3-4 times per day) 10-20 (2-4 times per day)
Adverse reactions Nausea, vomiting, abdominal pain, arrhythmia Arrhythmia, abdominal pain, diarrhoea, headache
10
Dystomia, tardive dyskinesia, sedation, hyperprolactinaemia
10–20 Hyperprolactinaemia, dry mouth, headache (2-4 times per day) This table lists the key facts of the most commonly used prokinetic agents include metoclopramide, domperidone, erythromycin and cisapride. The mainstay of pharmacological therapy of the gastroparesis is the use of prokinetic agents. The aim of therapy is to improve symptoms by accelerating gastric emptying, despite a poor correlation between the two (Kuo et al. 2007) 5-HT serotonin, D dopamine, IM intramuscular, IV intravenous, SC subcutaneous
211 Treatment of Diet-Related Disorders in Adult Diabetes
3383
The prokinetic effect is theoretically elicited through a coordinated set of pressure waves moving through the antrum, pylorus, and duodenum. However, there is little evidence of their efficacy in GI motor disorders; further, they are associated with a high prevalence of side effects outside of the GI tract. Both metoclopramide and domperidone are antagonists of dopamine receptors that may inhibit gastric emptying, although the dopamine receptor-mediated pathway in the brain stem may regulate nausea. Metoclopramide and domperidone exert their therapeutic effect by blocking these actions. Metoclopramide can also act as an agonist of 5-hydroxytryptamine receptor subtype 4 (5HT-4) to stimulate the cholinergic neural pathway in the stomach (Kuo et al. 2007). Metoclopramide is the most widely used drug for the treatment of diabetic gastroparesis because it brings about short-term improvements in both symptoms and gastric emptying rates. Patients are often treated with metoclopramide for long-term management of diabetic gastroparesis, but its real efficacy has not been proven (Camilleri et al. 2007). Erythromycin acts as a molecular mimic of motilin to stimulate motor activity, primarily in the upper GI tract. Erythromycin binds to the motilin receptor and initiates similar biologic effects. Erythromycin administered in low doses (50–100 mg) will effectively correct gastroparesis even in patients with refractory symptoms. Erythromycin administered intravenously and as an oral liquid suspension are more potent prokinetic agents than the tablet form (Abell et al. 2003). Cisapride and tegasorod are 5HT-4 agonists that induce the release of acetylcholine from myenteric cholinergic neurons along the GI tract. They are available in both tablet and suspension form (Lata et al. 2003). There are also some innovative approaches to treat diabetic gastroparesis. An intrapyloric injection of botulinum toxin may decrease pylorospasms associated with gastroparesis and can improve symptoms (Jones 2002). Moreover, no serious adverse effects have been noted. Gastric electrical stimulation as a therapeutic modality for diabetic gastroparesis is an attractive concept that involves surgical intervention; however, the device sometimes has to be removed because of complications (Bromer et al. 2005). In individuals who are undernourished, enteral feeding can be an option. Nevertheless, surgery for treating gastroparesis should be resorted to as a last option; however, postoperative complications limit the efficacy of surgical intervention in the treatment of diabetic gastroparesis (Abell et al. 2003). In sum, complications can most often be resolved with simple approaches such as dietary modification. As the severity of symptoms increases, a more aggressive approach must be considered if patients fail to respond to more conventional approaches. However, possible therapeutic options are limited and are not necessarily evidence based.
211.3.4 Intestinal Complications and Treatments 211.3.4.1 Diarrhea Diarrhea is a more clinically relevant and troubling problem in patients with diabetes. There are at least two possible causes of diabetic diarrhea. The first is impaired adrenergic regulation of fluid and electrolyte transport secondary to decreased a2-adrenergic input. The second is slow intestinal transit and subsequent bacterial overgrowth. Nonetheless, multiple other etiologic factors must be considered in the clinical evaluation of diabetic diarrhea. Possible etiologies include rapid intestinal transit, bacterial overgrowth, medications (e.g., metformin, acarbose, or miglito), the use of artificial sweeteners, celiac disease, pancreatic insufficiency, or nondiabetic diarrhea. It is commonly assumed that the rate of small bowel transit is reduced, leading to bacterial overgrowth in patients with diabetes. However, rapid gastric emptying can also accompany decreased intestinal transit times. Individuals with diarrhea and associated autonomic neuropathy, primarily
3384
Y.-X. Li et al.
diabetic, can have an extremely rapid orocecal transit time ( 75), and the most common PDs in this sample were avoidant (39.39%), histrionic (39.39%), and self-defeating PDs (39.39%), followed by passive–aggressive (36.36%) and schizoid (30.30%) PDs. In the study by Godt (2002), 31% of patients with BN had at least one PD. The most common was avoidant PD (18.5%), followed by dependent (13, 6%), histrionic (7.4%), and borderline (6.2%) PDs. Comorbidity of BN also varies according to sex. Thus, in men with BN there is a higher rate of homosexuality and perfectionism and susceptibility features appear more often (Herzog et al. 1984; Schneider and Agras 1987). In women with BN, in contrast, weight concerns and an obsession with thinness dominated (Joiner et al. 2000). Finally, as in people with AN, patients with BN who were treated successfully were diagnosed with a lower comorbidity. Thus, according to the study quoted above by Matsunaga et al. (2000), 21% of patients who recovered from bulimia met the criteria for a diagnosis of at least one PD.
216.3.4 Eating Disorders Not Otherwise Specified and Personality Disorders Morbid obesity has been studied from this perspective. According to Black et al. (1992), 72% of individuals are also affected by PDs (50% more than one). As regards the actual type of deterioration, in general there is some relationship between the behavioral characteristics of borderline personality (impulsivity, self-defeating behaviors, etc.) and morbid obesity, at least in women (Sansone et al. 1996). However, in other studies on obesity the data is more variable. Thus, disorders submitted by the sample from the study of Black et al. (1992) are diverse (histrionic, borderline, passive–aggressive), without a clear predominance of one over the other. By contrast, the study by Grana et al. (1989) concerning morbid obesity showed that the most prevalent PDs were others, specifically avoidant, antisocial, and obsessive personality. In the study by Guisado et al. (2001), the most prevalent PDs were paranoid, anancastic, schizoid, and anxious. In conclusion, the prevalence rate of PDs in obese individuals is very high, but very variable too, depending on the methodology and the type of sample used in the studies. However, we cannot establish a clear association between a specific type of disorder and morbid obesity. Conversely, when there is an association between morbid obesity and compulsive overeating, the results are clearer. According to Yanovski et al. (1993), 16% of the obese sample presents PDs, but this rate rises to 35% in the obese who binge. In the latter case, the PDs most frequently involved are borderline (14%) and avoidant (9%). Finally, it should be noted that in the study by Godt (2002), 36.2% of patients with a non-specified eating disorder had at least one PD. The most common was avoidant PD (27.7%), followed by borderline PD (6.4%).
216.3.5 Personality Disorders and Eating Disorders When studies focus on PDs and, as a result, on the comorbidity with Axis I disorders, EDs correlate with group B personality disorders (Modestin et al. 1997). More specifically, borderline PD has a higher rate of comorbidity with BN (it can range from 2% to 47%) than with the rest of the Axis I disorder (Wonderlich 1995). This broad range is related to
216 Use of the International Personality Disorder Examination (IPDE) and the Millon Clinical Multiaxial
3463
the methodological problems of studies (e.g., inexact diagnoses and lack of control groups) and, above all, to some overlap of diagnostic criteria that are common in the borderline disorder (e.g., criterion 4: binging, which is the result of impulsivity) and BN. This overlap in criteria poses a risk of overdiagnosis of borderline PD in subjects with bulimia. In fact, there is usually a decline in the rate of borderline PDs after short-term treatment focused on bulimia symptoms. The comorbidity of borderline PD may be mediated by the sex of the patients. Specifically, impulsivity (typical of this disorder) is expressed in males as alcohol or drug abuse (men: 63.3–81.9%; women: 38.1–59.1%) and in women in the form of binge eating or unspecified problems (men: 10.8–13.6%; women: 29.5–30.4%) (Zanarini et al. 1998; Zlotnick et al. 2002). With regard to the comorbidity between BN and borderline disorder, a point of interest is the possible existence of sexual abuse in childhood. In several previous studies it has been indicated that a history of sexual abuse is three times more common in patients with BN than in normal people (Garfinkel et al. 1995) and patients with restrictive AN (Coovert et al. 1989). More recently, however, it has been indicated that bulimics with a previous history of sexual abuse have a more severe clinical picture and a higher rate of comorbidity with the borderline disorder (Claridge et al. 1998). It should be noted that, in the study conducted by Marino and Zanarini (2001), 33% of patients with borderline PD met criteria for an ED not otherwise specified. Of these patients, 20% had restrictive patterns without weight loss, 37% purges without binging, 37% a pattern of binging without purging, and the remaining 33% had weight loss without menstruation.
216.3.6 Research The data of comorbidity between EDs and PDs outlined in various studies do not correspond. The discrepancies obtained correspond to various reasons: the type of samples (outpatients or hospitalized patients), the sex and age of the patients, the degree of evolution of the disorder (in an acute period, in a situation of chronic phase or in remission), the diagnostic tools used to assess personality disorders (self-report questionnaires or interviews) and the differences between the diagnostic criteria. An important methodological limitation is the absence of clinical control groups (patients with mental disorders other than anorexia or bulimia) and normal subjects (subjects from the normal population). This means that one should exercise a degree of caution as regards the conclusions reached. We therefore consider it important now to bring to attention the study from the doctoral thesis of the first author of this chapter (Marañon et al. 2007a, b). This study forms part of a large investigation whose purpose is to measure the comorbidity between PDs and EDs, assessed by the MCMI-II and the IPDE. The main contribution of this study to provide better knowledge of comorbidity between PDs and ED is related to the specific method used. What this means is that, apart from the ED group, there were clinical and normative control groups. The aim of this procedure was to discover whether the frequency and profile of PDs among EDs were different from the normal population and from non-ED patients seeking treatment for another Axis I mental disorder. According to the results in previous studies of PDs in patients with EDs, this study found that the prevalence rates of comorbidity were high both in IPDE (54.8%) and MCMI-II (77.4%) (see Table 216.5). As expected in previous studies, the highest rates of PD in patients with ED were obtained with the MCMI-II. The correspondence between the IPDE and the MCMI-II in diagnosing personality disorders in patients with eating disorders is very low. The MCMI-II tends to overdiagnose specific PDs and, as a result, it is not a good assessment measure for obtaining PD diagnosis. This finding is consistent with that of Kennedy et al. (1995) using the SCID-II and the MCMI-II to
3464
I. Marañon and I. Gonzalez
Table 216.5 Key features of International Personality Disorder Examination 1. IPDE is a semi-structured interview 2. IPDE has two versions that the interviewer can choose: one for the ICD-10 personality disorder criteria and the other for the DSM-IV personality disorder criteria 3. It consists of a self-screening questionnaire whose answers are either true or false 4. The screening questionnaire has no diagnostic value but can be used to select which part of the interview will be carried out 5. The whole interview can be used or it can also be used to identify specific disorders that have given positive in the self-administered questionnaire 6. The patient is asked to answer the questions taking into account how he or she usually is or acts. He or she is also asked to tell us if he or she has experienced a permanent change in his or her way of acting 7. The questions are organized under the following headings: work, self, interpersonal relationships, feelings, reality testing, and impulse control 8. The various sections of the interview begin with open questions that allow the patient to answer with as much information as desired and the interviewer to situate the context for understanding the subsequent responses to specific questions 9. The clinician assesses the responses with a score of 0 if the evaluated behavior is absent, with a score of 1 if the behavior is increased and with a score of 2 if the behavior meets diagnostic criteria or reaches pathological levels 10. IPDE can be administered to subjects aged over 18 11. To implement reliable results the interviewer needs to be trained 12. Qualified interviewers administer IPDE in about 45–60 min. Those who are not trained require more than 90 min In this table the major features of IPDE are listed, as well as the steps and requirements for administration and correction
assess PDs in patients with EDs. On the other hand, as Wetzler and Dubro (1990) observed in their study, the MCMI-II detects the subjects who may have a possible PD. Therefore, we can conclude that the MCMI-II does not replace a diagnostic interview. The MCMI-II can be used as a screening tool, but not as a personality disorder diagnostic instrument. The high negative predictive value indicates that the MCMI-II performs well in indicating when the disorder was not present during IPDE. In contrast, if the MCMI-II obtained a PD diagnosis, it would probably be a false positive. It would be possible to say that PD measurement with the MCMI-II is a statistic artifact. It can be understood clearly if we consider that 58% of subjects with at least one PD according to the MCMI-II, have more than four PDs altogether. Another piece of evidence for the above affirmation is that 32% of the AN patients have a schizoid PD assessed with the MCMI-II and not one of the same patients has the same diagnosis measured with the IPDE. In any case these results might also reflect problems in the PD diagnostic criteria, or they could be due to genuine high PD base rates (and comorbidity rates) in ED patients. For the interpretation of these differing results between the IPDE and the MCMI-II, the validity of the PD construct has been taken into account. We cannot forget that the ambiguity of the PD definition could be one of the reasons for the lack of correspondence between instruments. Taking the above into account, and therefore considering the data from personality disorders obtained by IPDE, the most relevant conclusion of this study was that more than half of subjects with ED (55%) met DSM-IV-TR diagnostic criteria for at least one PD, compared to 22% of the non-ED patients and 9% of the normative control group (See Table 216.6). The main contribution of this study, consequently, is to have proven that this high rate of comorbidity with PDs is specific in EDs and much higher than in other Axis I mental disorders. This finding is consistent with those of previous reports using structured interviews to assess PDs in EDs (Gartner et al. 1989; Matsunaga et al. 1998). It is also consistent with those using only one of the control groups we used (Díaz-Marsá et al. 2000; Grilo et al. 1996; Herpertz-Dahlmann et al. 2001). This fact is a challenge for clinical practices because the presence of a PD in a patient with AN or BN complicates treatment, and the prognosis of the ED becomes poorer.
216 Use of the International Personality Disorder Examination (IPDE) and the Millon Clinical Multiaxial
3465
Table 216.6 Key features of Millon Clinical Multiaxial Inventory 1. MCMI is a self-administered questionnaire 2. The subject must answer 175 questions in a true–false answer format 3. Before beginning to answer the questionnaire the clinician should make sure that the subject has understood the instructions 4. Once the questionnaire is answered, correction can be done manually (with template correction) via Internet or using a local software 5. It should be taken into account whether the subject is a person of the general population, if he or she is an adult undergoing psychiatric treatment in a hospital, or if he/she is an adult undergoing psychiatric outpatient treatment 6. While correcting, the sex of the subject must be specified 7. MCMI can be administered to subjects aged over 18 with a reading level equivalent to 8 years of schooling 8. The subjects typically need about 25–30 min to answer all questions 9. Those whose scales give a score greater than 84 BR are considered pathological In this table the main characteristics of MCMI are listed, as well as the steps and requirements for administration and correction
216.4 Application to Other Areas of Health and Disease Beyond the different data found in the published studies, EDs are disorders which are rarely found in a psychopathologically pure state: they are often found with Axis II clinical disorders. The presence of a PD with an ED complicates the clinical picture. Specifically, it makes the early detection of the problem more difficult, as well as making the treatment difficult. It also hinders therapeutic prognosis (cited in Echeburúa and Marañon 2001); in fact, patients with AN and BN who are affected by a PD have a greater incidence of binging, vomiting, symptoms of anxiety and depression, and excessive alcohol consumption, as well as greater difficulties in social integration and an increased frequency of suicide attempts (Gartner et al. 1989; Wonderlich et al. 1990; Braun et al. 1994; Steiger and Stotland 1996; Kozyk et al. 1998; Matsunaga et al. 1998). This fact should be taken into account when planning treatment. To do so, the design of intervention programs that consider personality aspects would be useful, just as the development of specific therapeutic programs for EDs comorbid with PDs is a challenge for clinical research (Palmer et al. 2003). On the other hand, it seems that the MCMI is not a good diagnostic tool to diagnose PDs in EDs. Therefore, to obtain a reliable diagnosis to determine whether a patient has a PD or not, we recommend evaluation using a clinical interview. In addition, using an interview will allow us to obtain valuable information in order to design specific treatment for a patient, as it will allow us to discover the patient’s strong points and his or her weaknesses. However, MCMI may function best as a screening tool and more conservative clinical norms/cut-offs could help correct overdiagnosis problems.
216.5 Conclusion In this chapter, according to the diagnostic philosophy contained within DSM-IV-TR, PDs have been considered in a categorical way, that is, as discrete entities of abnormal behavior patterns. However, although DSM-IV-TR approaches clinical diagnoses from a categorical perspective, there is increasing empirical research approaching the clinical assessment of PDs from a dimensional perspective (Segal and Coolidge 1998), since one of the most perplexing difficulties in this clinical field is the definition and measurement of personality dysfunction (Sansone et al. 2005). This point needs to be researched further. In general, the comorbidity of EDs and PDs is very high: it can range from 20% to 80%. When considering PDs by type (A, B, and C), we found that the most prevalent PDs in AN are those from
Table 216.7 Agreement between IPDE and MCMI-II personality disorders DIAGNOSTIC IPDE positive (%) MCMI positive (%) Agreement (%) KAPPA SEN (%) SPE (%) PPV (%) NPV (%) Cluster A 1.2 32.1 66.66 –0.023 0 67.46 0 98.24 Paranoid 1.2 9.5 89.28 –0.022 0 90.36 0 98.69 Schizoid 0 21.4 78.57 – – 78.57 0 100 Schizotypal 0 4.8 95.23 – – 95.23 0 100 Cluster B 20.2 33.3 67.85 0.198 52.94 71.64 32.14 85.71 Antisocial 0 13.1 86.90 – – 86.90 0 100 Borderline 19 8.3 79.76 0.164 18.75 94.11 42.85 83.11 Histrionic 2.4 25 77.38 0.136* 100 76.82 9.52 100 Narcissistic 1.2 11.9 86.90 –0.022 0 87.79 0 98.64 Cluster C 31 28.6 66.66 0.203 42.30 77.58 45.83 75 Avoidant 16.7 14.3 85.71 0.455** 50 92.85 58.33 90.27 Dependent 2.4 11.9 88.09 0.132 50 89.02 10 98.64 Obsessive– 22.6 11.9 70.23 –0.021 10.52 87.69 20 77.02 compulsive Any disorder 54.8 77.4 79.76 0.322 91.30 39.47 64.61 78.94 The frequency of the 12 personality disorders and three clusters, the agreement percentage between the IPDE and the MCMI-II, values for Kappa, sensitivity (SEN), specificity (SPE), positive predictive value (PPV), and negative predictive value (NPV) are presented in this table. Correspondence between the two measures was only marginal. In most of the personality scales the kappa value was around 0, so correspondence between the two measures was only at chance level. Sensitivity refers to the likelihood that the MCMI-II will be positive when there is a diagnostic present according to the IPDE. Specificity refers to the likelihood that the MCMI-II will be negative when the IPDE is negative for a particular diagnosis. Positive predictive value is the probability that the IPDE will be positive when the MCMI-II is positive. Negative predictive value is the probability that the IPDE will be negative when this disorder is absent according to the MCMI-II * p