Home Care Enteral Feeding
Nestlé Nutrition Workshop Series Clinical & Performance Program, Vol. 10
Home Care Enteral Feeding Editors H. Lochs, Berlin, Germany D.R. Thomas, Saint Louis, Mo., USA
Nestec Ltd., 55 Avenue Nestlé, CH–1800 Vevey (Switzerland) S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland) www.karger.com © 2005 Nestec Ltd., Vevey (Switzerland) and S. Karger AG, Basel (Switzerland). All rights reserved. This book is protected by copyright. No part of it may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, or recording, or otherwise, without the written permission of the publisher. Printed in Switzerland on acid-free paper by Reinhardt Druck, Basel ISBN 3–8055–7850–4 ISSN 1422–7584 Library of Congress Cataloging-in-Publication Data Nestlé Nutrition Workshop of the Clinical and Performance Nutrition Program (10th : 2004 : Potsdam, Germany) Home care enteral feeding / editors, H. Lochs, D.R. Thomas. p. ; cm. – (Nestle nutrition workshop series. Clinical & performance program, ISSN 1422-7584 ; v. 10) “10th Nestlé Nutrition Workshop of the Clinical and Performance Nutrition Program, which took place in June 2004 in Potsdam, Germany”–Foreword. Includes bibliographical references and index. ISBN 3-8055-7850-4 (hard cover : alk. paper) 1. Older people–Nutrition–Congresses. 2. Enteral feeding–Congresses. 3. Older people–Home care–Congresses. 4. Malnutrition–Congresses. [DNLM: 1. Enteral Nutrition–Aged. 2. Home Care Services–Aged. 3. Malnutrition–epidemiology–Aged. 4. Malnutrition–therapy–Aged. WB 410 N4685h 2005] I. Lochs, H. II. Thomas, D. R. (David R.) III. Title. IV. Nestlé Nutrition workshop series. Clinical & performance programme ; v. 10. RC953.8.D54N47 2004 618.97⬘639–dc22 2004026200
Basel · Freiburg · Paris · London · New York · Bangalore · Bangkok · Singapore · Tokyo · Sydney
The material contained in this volume was submitted as previously unpublished material, except in the instances in which credit has been given to the source from which some of the illustrative material was derived. Great care has been taken to maintain the accuracy of the information contained in the volume. However, neither Nestec Ltd. nor S. Karger AG can be held responsible for errors or for any consequences arising from the use of the information contained herein.
Contents
VII Preface IX Foreword XI Contributors 1 Malnutrition in the Developing World: The Lack of Food Scenario Labadarios, D. (Tygerberg) 15 Are Older People Starving to Death in a World of Plenty? Thomas, D.R. (St. Louis, Mo.) 31 Physiological vs. Pathological Changes of Nutritional Status over Life Time Müller, M.J.; Bosy-Westphal, A.; Geisler, C.; Onur, S. (Kiel) 45 Home Enteral Nutrition Demographics and Utilization in the United States DeLegge, M.H. (Charleston, S.C.) 59 Home Enteral Nutrition Epidemiology and Legislation in Europe Van Gossum, A. (Brussels) 73 When Does Malnutrition Become a Risk? Genton, L.; van Gemert, W.G.; Dejong, C.H.; Cox-Reijven, P.L.; Soeters, P.B. (Maastricht)
V
Contents 89 What Are the Goals of Nutritional Support? The Example of Home Enteral Nutrition Hébuterne, X.; Schneider, S.M. (Nice) 103 Oral Protein and Energy Supplementation in Older People: A Systematic Review of Randomized Trials Milne, A.C.; Avenell, A. (Aberdeen); Potter, J. (Glasgow) 127 Efficacy of Enteral and Parenteral Nutrition in Cancer Patients Bozzetti, F. (Prato); Bozzetti, V. (Milan) 143 Ethics and Economics in Nutritional Support Buchman, A.L. (Chicago, Ill.) 167 Pathophysiology of Weight Loss in Older Persons Morley, J.E. (St. Louis, Mo.) 179 Interaction between Nutrition, Intestinal Flora and the Gastrointestinal Immune System Lochs, H. (Berlin) 189 Psychoimmunology of Nutrition Lesourd, B. (Clermont-Ferrand) 203 How Can We Impact the Immune System with Pre- and Probiotics? Schiffrin, E.J.; Donnet, A.; Blum, S. (Lausanne) 219 How Can We Modulate Cytokine Production and Action? Cynober, L. (Paris) 233 How Can We Improve Functional Outcomes? Elia, M. (Southampton) 249 Concluding Remarks 251 Subject Index
VI
Preface
The nutritional status of older adults living at home is poor. On average, persons over the age of 70 years consume one third less calories compared to younger persons. In dietary intake studies, 10% of older men and 20% of older women have intakes of protein below the United States Recommended Daily Allowance (RDA), and one third consume fewer calories than the RDA. Fifty percent of older adults have intakes of minerals and vitamins of less than the RDA and 10–30% have subnormal levels of minerals and vitamins. Sixteen to 18 percent of community-dwelling elderly persons consume less than 1,000 kcal daily. The reasons for this decline in nutrient intake are myriad, including loss of appetite (anorexia), inadequate food resources (starvation), physiological reductions in anabolic hormones or energy expenditure (sarcopenia), and disease-related loss of weight (cachexia). In ambulatory clinic patients, a pathological cause of undernutrition can be identified in 93% of older persons and 90% of younger persons. Most of these patients (89%) will have potentially treatable causes for their undernutrition. The syndrome of cachexia, the cytokine-induced wasting of protein and energy stores, is due to the effects of disease. Cytokines are related to a number of disease conditions, including cancer, end-stage renal disease, chronic pulmonary disease, congestive heart failure, rheumatoid arthritis, and AIDS. Systemic inflammation mediated through cell injury or activation of the immune system triggers an acute inflammatory response. In this volume, a distinguished panel of international experts in nutrition address the state-of-the-art in understanding nutrition in communitydwelling older adults. The epidemiology of undernutrition in the developing and developed world are addressed by Professors Labadarios and Thomas. The utilization of home enteral nutrition in the United States and in Europe are reviewed by Professors DeLegge and Van Gossum. The physiological changes in nutritional status are surveyed by Professors Müller, Soeters, and Morley. Professors Hébuterne, Milne, and Bozzetti address the practical evaluation of nutritional support in community-dwelling older persons, and Professor Buchanan examines the ethical considerations of enteral feeding. VII
Preface The impact of psychological factors, probiotics and the immune system, intestinal flora, the relationship of cytokines in nutrition, and functional outcomes are described by Professors Lesourd, Schiffrin, Lochs, Cynober, and Elia. We wish to thank the distinguished faculty for sharing their insights and their expertise in this volume. We commend this volume to you as an in depth review of the state-of-the-art in community nutrition. Both what is known and directions for future research are explored. David R. Thomas and Herbert Lochs
VIII
Foreword
As clinical nutrition plays an increasing role in the management of patients at home, often for a much longer period than for a hospital stay, the topic ‘Home Care Enteral Feeding’ was chosen for this 10th Nestlé Nutrition Workshop of the Clinical and Performance Nutrition Program, which took place in June 2004 in Potsdam, Germany. Physicians and dieticians have the important task of detecting malnutrition of patients in time, and subsequently to improve their nutritional status. In this workshop a lot was learnt about the nutritional status of people in different parts of the world, the reasons for malnutrition, the various possibilities on how to support nutrition, the regulation of appetite, and the potential modulation of the intestinal flora and immune functions; the aim being to improve the functional outcome of patients. I would like to thank the chairmen, Prof. Herbert Lochs, Berlin, Germany, and Prof. David Thomas, St. Louis, Mo., USA, who are recognized experts in the field of long-term enteral nutrition, for consolidating the program and inviting the opinion leaders in various divisions of enteral nutrition as speakers. Physicians from 14 countries contributed to the discussions that are published in this book. Dr. Mike Possner and his team from Nestlé Nutrition Germany provided all logistical support, enabling the participants to enjoy the German hospitality. Dr. Philippe Steenhout from Nestlé’s Nutrition Strategic Business Division in Lausanne, Switzerland, was responsible for the scientific coordination. His cooperation with the chairpersons was essential for the success of this workshop. Prof. Wolf Endres, MD Vice-President Nestec Ltd., Lausanne, Switzerland
IX
10th Nestlé Nutrition Workshop Clinical & Performance Program Berlin, June 13–17, 2004
Contributors
Chairpersons and Speakers Prof. Federico Bozzetti Residenza Le Querce IT–20090 Segrate Italy Tel. ⫹39 2 2390 2354 Fax ⫹39 2 2641 0267 E-Mail
[email protected] Prof. Mark H. Delegge Medical University of South Carolina 96 Jonathan Lucas Street Charleston, SC 29425 USA Tel. ⫹1 843 9716202 Fax ⫹1 888 6099745 E-Mail
[email protected] Prof. Alan Buchman Northwestern University Medical School 676 N. St. Clair Street Suite 1400 Chicago, IL 60611 USA Tel. ⫹1 312 6954514 Fax ⫹1 312 6953999 E-Mail
[email protected] Prof. Marinos Elia Institute of Human Nutrition Mailpoint 113, Tremona Road Southampton SO16 6YD UK Tel. ⫹44 2380 794277 Fax ⫹44 2380 794277 E-Mail
[email protected] Prof. Xavier Hébuterne Prof. Luc Cynober Laboratoire de Biologie de la Nutrition Faculté de Pharmacie Université Paris V 4, avenue de l’Observatoire FR–75270 Paris Cedex 06 France Tel. ⫹33 1 4234 8260 Fax ⫹33 1 4234 8612 E-Mail solange.ngon@ htd.ap-hop-paris.fr
Hôpital de l’Archet 2 Hepato-Gastroentérologie et Nutrition Clinique 151, route Saint-Antoine de Ginestière FR–06200 Nice France Tel. ⫹33 4 9203 6168 Fax ⫹33 4 9203 6575 E-Mail
[email protected] XI
Contributors Prof. Demetre Labadarios Department of Human Nutrition University of Stellenbosch and Tygerberg Academic Hospital PO Box 19063, Tygerberg 7505 South Africa Tel. ⫹27 21 938 9259 Fax ⫹27 21 933 2991 E-Mail
[email protected] Prof. Manfred Müller Institut für Humanernährung und Lebensmittelkunde Düsternbrooker Weg 17 DE–24105 Kiel Germany Tel. ⫹49 431 880 5670 Fax ⫹49 431 880 5679 E-Mail mmueller@ nutrfoodsc.uni-kiel.de
Prof. Bruno Lesourd Département de Gérontologie Faculté de Médecine de Clermont-Ferrand 28, place Henri-Dunant FR–63001 Clermont-Ferrand Cedex 1 France Tel. ⫹33 473 177 945 Fax ⫹33 473 750 836 E-Mail blesourd@ chu-clermontferrand.fr
Prof. Herbert Lochs Medizinische Klinik Schwerpunkt Gastroenterologie/ Hepatologie/Endokrinologie Schumannstrasse 20–21 DE–10117 Berlin Germany Tel. ⫹49 30 450 514 021 Fax ⫹49 30 450 514 923 E-Mail
[email protected] Mrs. Anne Milne Health Services Research Unit University of Aberdeen Polwarth Building, Foresterhill Aberdeen AB25 2ZD Scotland, UK Tel. ⫹44 12 24 559 799 Fax ⫹44 12 24 554 580 E-Mail
[email protected] Prof. John Morley Division of Geriatric Medicine Saint Louis University Health Sciences Center 1402 S. Grand Blvd, Room M238 St. Louis, MO 63104 USA Tel. ⫹1 314 977 8462 Fax ⫹1 314 771 8575 E-Mail morley@slu-edu
XII
Dr. Eduardo Schiffrin Nestlé Research Center, Nestec Ltd. PO Box 44, Vers-chez-les-Blanc CH–1000 Lausanne 26 Switzerland Tel. ⫹41 21 785 8671 Fax ⫹41 21 785 8565 E-Mail eduardo.schiffrin@ rdls.nestle.com
Prof. Peter Boudewijn Soeters Department of Surgery Academic Hospital Maastricht PO Box 5800 NL–6202 AZ Maastricht The Netherlands Tel. ⫹31 43 387 7489 Fax ⫹31 43 387 5473 E-Mail
[email protected] Prof. David Thomas Internal Medicine/Geriatric Medicine St. Louis University 1402 Grand Boulevard, Room M238 St. Louis, MO 63104 USA Tel. ⫹1 314 577 8462 Fax ⫹1 314 771 8575 E-Mail
[email protected] Prof. André Van Gossum Department of Gastroenterology Hôpital Erasme 808, route de Lennik BE–1070 Brussels Belgium Tel. ⫹32 2 555 3712 Fax ⫹32 2 555 4697 E-Mail
[email protected] Contributors Moderators Prof. Hans Konrad Biesalski Institute for Biological Chemistry and Nutritional Science University Hohenheim Garbenstrasse 30, DE–70593 Stuttgart Germany Tel. ⫹49 711 4 59 41 12 Fax ⫹49 711 4 59 38 22 E-Mail
[email protected] Prof. Peter Fürst Institute for Nutritional Science Rheinische Friedrich Wilhelms University, Endenicher Allee 11–13 DE–53115 Bonn Germany Tel. ⫹49 2 28 73 38 14 Fax ⫹49 2 28 73 64 83 E-Mail
[email protected] Prof. Ulrich Oltersdorf Institute of Nutritional Economics and Sociology Federal Research Centre for Nutrition and Food Haid-und-Neu-Strasse 9
DE–76131 Karlsruhe Germany Tel. ⫹49 721 6 62 55 50 Fax ⫹49 721 6 62 55 52 E-Mail ulrich.oltersdorf@ bfe.uni-karlsruhe.de
Prof. Elisabeth Steinhagen-Thiessen Evangelic Geriatric Centre Virchow University Hospital Reinickendorfer Strasse 61 DE–13347 Berlin Germany Tel. ⫹49 30 45 94 19 00 Fax ⫹49 30 45 94 19 38 E-Mail elisabeth.steinhagen-thiessen@ charite.de
Dr. Dorothee Volkert Geriatric Nutrition Endenicher Allee 11–13 DE–53115 Bonn Germany Tel. ⫹49 2 28 73 20 18 Fax ⫹49 2 28 73 32 17 E-Mail
[email protected] Invited attendees Dr. Walter Pauwels / Belgium Prof. Isabel Correia / Brazil Dr. David Armstrong / Canada Dr. Sami Antoun / France Prof. Eric Bertin / France Dr. Anne-Grit Bialojan / Germany Mr. Wolfgang Kühn / Germany Prof. Christian Löser / Germany Dr. Kristina Norman / Germany Dr. Johann Ockenga / Germany Dr. Matthias Pirlich / Germany Dr. Ingolf Schiefke / Germany Dr. Ralf-Joachim Schulz / Germany Prof. Dieter Schwab / Germany Dr. Luzia Valentini / Germany
Dr. Wolfram Weinrebe / Germany Dr. Maria Ilia / Greece Dr. Mairi Koulentaki / Greece Dr. Ioannis Nikolaou / Greece Dr. Efstathios Papavassiliou / Greece Dr. Ioannis Triantafyllidis / Greece Dr. Anna Demagistris / Italy Dr. Maria Estrella Petrina Jauregui / Spain Dr. Juana Maria Rabat Restrepo / Spain Dr. Sekela Mwakyusa / Tanzania Prof. Jeremy Powell-Tuck / UK Dr. Timothy Bowling / UK
XIII
Contributors Nestlé participants Dr. Frank-Arnaud Battandier / France Prof. Ferdinand Haschke / Germany Dr. Mike Possner / Germany Dr. Nicole Handschel / Germany Mr. Fernando Carvalho / Portugal Prof. Wolf Endres / Switzerland Dr. Claudia Roessle / Switzerland Dr. Philippe Steenhout / Switzerland Ms. Sue Jones / UK
XIV
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 1–13, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
Malnutrition in the Developing World: The Lack of Food Scenario D. Labadarios Department of Human Nutrition, University of Stellenbosch and Tygerberg Academic Hospital, Tygerberg, South Africa
Introduction The current literature concurs that the prevalence of malnutrition, and stunting among children in particular, is a reasonably accurate reflection on the prevailing socioeconomic status in a given environment. In this regard, malnutrition is known to adversely impact on mental development, scholastic achievement, productivity, child and woman morbidity and mortality as well as the risk for infection. As such, it is hardly surprising that nutritional status is one of the key Millennium Development Goals [1]. Poverty and food insecurity are almost universally accepted to be very significant contributory factors. Although slow, but welcome improvements are being achieved in many developing countries, child malnutrition, poverty and household food insecurity appear to be actually increasing (fig. 1) in India and Sub-Saharan Africa [2]. In South Africa for instance and at the national level, three of four households have been reported to be either ‘hungry’ or at ‘risk of hunger’ (table 1), the extent of hunger being related to adverse growth patterns (fig. 2), lower energy intake (fig. 3), lower income as well as a lower intake of a number of nutrients among children 1–9 years of age [3]. In this regard, the World Bank and the International Monetary Fund have recently predicted [4] that although the income poverty goal is likely to be achieved at the global level, Africa will fall well short. The predicted shortfalls appear especially serious for the health and environmental goals, namely child and maternal mortality, and access to safe drinking water and basic sanitation. The likely increasing number of malnourished people in these regions assumes even greater importance against the background of the HIV/AIDS pandemic in such developing countries, since HIV/AIDS exacerbates food insecurity by decreasing the available agricultural labor force and food production [2]. 1
Malnutrition in the Developing World: The Lack of Food Scenario 1990–1992 to 1995–1997 (millions)
1995–1997 to 1999–2001 (millions) China* India Southeast Asia South America West Africa North Africa North America Central America Caribbean Southern Africa Other East Asia Other South Asia Near East East Africa Central Africa
⫺50 ⫺40 ⫺30⫺20 ⫺10
0
Reduction (progress) Increase (setback)
*includes Taiwan Province of China ⫺20 ⫺10
10 20
0
10 20
Source: FAO
Fig. 1. Changes in the number of undernourished people in developing sub-regions of the world. From the FAO [2] with permission from the publishers.
Table 1. Household hunger risk classification in South Africa, 1999 Hunger risk Province classification EC FS
G/ KZN TENG
Number (n) 398 209 409 Food 4.3 45.5 36.7 secure, % At risk of 12.6 16.8 21.5 hunger, % Experience 83.2 37.8 41.8 hunger, %
M/GA NC
NP
NW
WC
RSA
525 150 144 332 226 342 2,735 26.7 21.3 13.2 19.3 13.3 39.8 25.0 25.9
26.0
23.6
26.2
25.2
29.0
22.9
47.4
52.7
63.2
54.5
61.5
31.3
52.2
EC ⫽ Eastern Cape; FS ⫽ Free State; G/TENG ⫽ Gauteng; KZN ⫽ KwaZulu-Natal; M/GA ⫽ Mpumalanga; NC ⫽ Northern Cape; NP ⫽ Northern Province (now Limpopo); NW ⫽ North West Province; WC ⫽ Western Cape; RSA ⫽ Republic of South Africa. From Labadarios [3] with permission from the publishers. *Significant difference between provinces, p ⬍ 0.01, 2 test.
Child Growth in Relation to Food Insecurity Against this background, few will question the importance and necessity of providing or creating access to food as a means of alleviating the effects of poverty and food insecurity with a view to improving nutritional status. Nevertheless, and despite the available evidence indicating that malnutrition is being steadily reduced in much of the world, in some parts of the African continent nutritional status improvement indicators remain unfavorable or further deteriorate, despite continuing, and expensive, nutritional interventions. 2
Mean Z score
Malnutrition in the Developing World: The Lack of Food Scenario
0.2 0 ⫺0.2 ⫺0.4 ⫺0.6 ⫺0.8 ⫺1 ⫺1.2
Experience hunger At risk of hunger Food secure Height-for age
Weight-forage
Weight-forheight
Food secure At risk of hunger Experience hunger
Mean energy intake (kJ)
Fig. 2. Hunger risk classification as related to anthropometric status nationally in children aged 1–9 years: South Africa 1999 (p ⬍ 0.001). From Labadarios [3] with permission from the publishers.
Food secure At risk of hunger Experience hunger 6,000 5,000 4,000 3,000 2,000 1,000 0
Experience hunger At risk of hunger Food secure Urban
Rural
National
Fig. 3. Hunger risk classification as related to the intake of energy nationally and by area of residence in children aged 1–9 years: South Africa 1999 (p ⬍ 0.001 urban vs. rural). From Labadarios [3] with permission from the publishers.
In this regard, the role of nutrition, on its own, as the factor predisposing to poor growth or for that matter adverse pregnancy outcomes [5, 6], has been frequently questioned [7, 8]. The better understanding of the causes of stunting and its consequences on short- and long-term health, therefore, is not only of scientific interest but it also has important implications in monetary and policy considerations. The factors thought to have been contributory to poor growth improvement outcomes include the timing of such interventions, the role of specific macro- and micronutrients, energy intake, dietary protein quality, intervention leakages as well as diarrheal disease [7]. Importantly though, available evidence indicates that improvement in living conditions 3
Malnutrition in the Developing World: The Lack of Food Scenario either through migration or adoption is associated with marked improvements in growth patterns [10, 11]. Although the latter may be due to combined nutritional and environmental improvements, the importance of environmental improvements should not be underestimated [12]. In fact, it has been proposed [13] that the failure of children to grow adequately in developing countries may be due to the presence of the acute phase response, without necessarily the manifestation of any overt clinical signs, induced by adverse environmental conditions.
Child Growth in Relation to Chronic Inflammation Global estimates for the year 2000 indicate that 135 and 162 million children under the age of 5 years were respectively underweight and stunted [14] of which 31 and 45 million respectively are African children. Also globally and on the basis of the analysis of 39 nationally representative surveys from developing countries, the pattern of growth faltering appears to be remarkably similar [15]. Thus weight for age starts to falter at approximately 3 months of age whereas length/height for age starts to falter immediately after birth. In Africa and more specifically in the Gambia, the available evidence indicates that the marginal adequacy of the infants’ diet could not, on its own, explain the adverse growth patterns of such children [16]. Indeed, the outcomes of a number of nutritional intervention studies aimed at improving growth have been neither consistent nor significant, with only a few such studies showing a measure of growth improvements [16, 17]. Furthermore and although diarrheal disease, as opposed to gastrointestinal disease, may be associated with acute weight loss, the impact of the former on long-term growth faltering has been questioned [18, 19]. The latter, namely gastrointestinal disease diagnosed as an enteropathy on the basis of impaired gastrointestinal mucosa permeability using the lactulose:mannitol test, has been associated with poor infant growth and is reported to explain more than 43% of the growth faltering of Gambian infants [20, 21]. The proposed mechanism of the reported association includes villar atrophy with attendant maldigestion and malabsorption as well as translocation of macromolecules with the attendant induction of local and systemic immune and inflammatory response [20]. Of further interest is that the enteropathy has recently been histologically described and includes crypt hyperplasia, villous stunting, a high number of intraepithelial lymphocytes and an increase in cytokine immunoreactive mononuclear cells with a proinflammatory response [22]. The presence of the inflammatory response in relation to contrasting environments has also been reported in stunted Nepali children [23], and impaired intestinal permeability has also been reported in stunted rural Bangladeshi and mildly stunted Nepali infants and children [24]. Moreover a longitudinal study [25] in rural Gambian infants has documented evidence of chronic 4
Malnutrition in the Developing World: The Lack of Food Scenario immunostimulation (raised white cell, lymphocyte and platelet counts together with frequently elevated C-reactive protein), impaired intestinal permeability associated with impaired growth, and raised endotoxin and immunoglobulin (Ig)G-endotoxin core antibody which was related both to growth and measures of mucosal enteropathy. It would thus appear that the consequences of the presence of the inflammatory response may not only be of equal importance as that of dietary inadequacies, but, additionally may, in the long-term, impact adversely on health and well being.
Proposed Consequences of Impaired Child Growth The human and financial costs of malnutrition, apart from those welldescribed and related to short-term morbidity and mortality aspects, are, in the longer-term, likely to be considerably higher than currently estimated. Emerging evidence indicates that malnutrition early in life appears to be associated, among emerging populations in the nutrition transition, with an increased risk for the so-called degenerative diseases of the developed world. It has been postulated that, in countries undergoing the nutrition transition, stunting appears to be associated with overweight later in life [26]. Stunted children in South Africa, China and Russia appear to be at a two- to sevenfold increased relative risk of being overweight. In Brazil, the greatest increase in the prevalence of obesity has been reported among poor women from the most affluent region of the country, the prevalence of obesity increasing with income among men and poverty among women [27]. Furthermore, in a longitudinal study over a period of 2 years among school girls in Sao Paulo documented a higher increase in weight-for-height in stunted as compared with non-stunted girls when an increased amount of energy was derived from fat [28]. Also of interest is that increasing body fatness together with dietary changes to a more Western type of diet and decreased physical activity in Brazil appears to be associated closely with an increasing prevalence of overweight, type-2 diabetes and cardiovascular disease [29]. Stunting has also been associated with the metabolic syndrome [30], and observations from Brazil indicate that stunted individuals have higher fasting plasma glucose, triglyceride, total cholesterol and low-density lipoprotein cholesterol when compared with individuals of normal stature [31].
Chronic Inflammation and Degenerative Diseases The available evidence indicates that inflammation may play an important and significant role in the etiology of the so-called chronic degenerative diseases in the Western world. In the case of atherosclerosis, elevated 5
Malnutrition in the Developing World: The Lack of Food Scenario concentrations of such inflammatory markers as C-reactive protein have been reported to be at least as strong, if not stronger, predictors of cardiovascular events [32]. Endothelial cell activation and the inflammatory response are thought to be initiated by oxidized lipoproteins, hypertension, diabetes mellitus as well as obesity [33]. Systemic inflammation has also been reported to be associated with insulin resistance and incident cardiovascular disease and diabetes [34]. In this regard, the apparent predisposition of the undernourished child to become overweight/obese later in life, especially in countries undergoing the nutrition transition with the attendant urbanization and changes in dietary patterns, has been proposed as a possible mechanism to explain the rapidly and consistently increasing prevalence of diabetes and cardiovascular disease [29, 35]. Certainly, preliminary evidence indicates that circulating levels of the proinflammatory cytokines interleukin-6 and tumor necrosis factor-␣ were significantly higher in urban middle class and urban slum dwellers and were related to body mass index when compared with rural village dwellers. Moreover, in Indian Asians living in England the geometric mean C-reactive protein concentrations have been reported to be 17% higher than those in European whites affording an estimated 14% increase in population coronary heart disease risk among Indian Asians when compared with European whites [36].
Common Pathways of Under- and Over-Nutrition The consistency of the available evidence is supportive of the role of chronic inflammation in the diseases of affluence [32–34]. In relation to food and poor food choices and their role in increasing the risk of these diseases, the concept is emerging that some foods may have greater or lesser proinflammatory properties. For instance, a diet with a high glycemic load has been documented to predict plasma C-reactive protein concentrations in apparently healthy middle-aged women [37], independent of the conventional risk factors for ischemic heart disease, especially in overweight women prone to insulin resistance. This of course contrasts the reported anti-inflammatory properties of n-3 fatty acids [38] which, in some people at least, are known to suppress tumor necrosis factor-␣ by peripheral blood mononuclear cells in healthy men [39]. In the developing world, chronic inflammation appears to be associated with undernutrition, within the context of food insecurity and unfavorable environmental influences. Especially in countries undergoing the nutrition transition with the attendant urbanization and very significant adverse dietary changes, the apparent tendency of the previously malnourished to the metabolic syndrome and obesity may in itself and by itself be mediated by chronic inflammation. Adipose tissue-derived cytokines [40] for instance may contribute to a proinflammatory environment which may be further accentuated by adverse environmental and low grade infectious 6
Malnutrition in the Developing World: The Lack of Food Scenario
Undernutrition
Overnutrition
Chronic inflammation, environmental and dietary influences
Growth faltering
Chronic disease
Fig. 4. Possible common pathways to under- and over-nutrition.
influences (fig. 4). These latter associations underscore the importance, and urgency, for an improved understanding of the multifactorial causes of malnutrition with a view to establishing cost-effective interventions for its prevention, be it among the over- or undernourished.
References 1 United Nations Standing Committee on Nutrition: 5th Report on the World Nutrition Situation: Nutrition for Improved Development Outcomes. Geneva, Standing Committee on Nutrition, 2004. 2 Food and Agriculture Organization: The State of Food Insecurity in the World 2003. Monitoring Progress Towards the World Food Summit and Millennium Development Goals. Rome, FAO, 2003. 3 Labadarios D, supported by Steyn N, Maunder E, MacIntyre U, et al: The National Food Consumption Survey (NFCS): Children Aged 1–9 years, South Africa, 1999. Stellenbosch, University of Stellenbosch Printers, 2000. 4 Development Committee (Joint Ministerial Committee of the Boards of Governors of the Bank and the Fund on the Transfer of Real Resources to Developing Countries): Global Monitoring Report 2004. Policies and Actions for Achieving the MDGs and Related Outcomes. New York, International Monetary Fund/World Bank, 2004. 5 Kramer MS: Balanced Protein/Energy Supplementation in Pregnancy (Cochrane Review). Cochrane Library, issue 4. Oxford, Update Software, 2001. 6 Kramer MS: High Protein Supplementation in Pregnancy (Cochrane Review). Cochrane Library, issue 3. Oxford, Update Software, 2000. 7 Allen LH: Nutritional influences on linear growth: A general review. Eur J Clin Nutr 1994; 48(suppl 1):S75–S89. 8 Beckett C, Durnin JVGA, Aitchison TC, Pollitt E: Effects of an energy and micronutrient supplement in undernourished children in Indonesia. Eur J Clin Nutr 2000;54(suppl 2):S52–S59. 9 Martorell R, Kettel Khan L, Schroeder DG: Reversibility of stunting: Epidemiological findings in children from developing countries. Eur J Clin Nutr 1994;48(suppl 1):S45–S57. 10 Yip R, Scanlon K, Trowbridge F: Improving growth status of Asian refugee children in the United Sates. JAMA 1992;267:937–940. 11 Proos LA, Hofvander Y, Wennqvist K, Tuvemo T: A longitudinal study on anthropometric and clinical development of Indian children adopted in Sweden II: Growth, morbidity and
7
Malnutrition in the Developing World: The Lack of Food Scenario
12
13 14 15 16 17 18 19
20 21 22
23 24
25
26 27 28
29 30 31 32
33 34
8
development during two years after arrival in Sweden; in Proos LA (ed): Growth and Development of Indian Children Adopted in Sweden. Uppsala, Lindbergs Grafiska, 1992, pp 1–18. Martorell R, Habicht JP, Klein RE: Anthropometric indicators of changes in nutritional status in malnourished populations; in Underwood BA (ed): Methodologies for Human Population Studies in Nutrition Related to Health. NIA Publication No 82-2452. Washington, US Government Printing Office, 1982, pp 96–110. Solomons NW, Mazariegos M, Brown KH, Klassing K: The underprivileged, developing country child: Environmental contamination and growth failure revisited. Nutr Rev 1993;51:327–332. De Onis M, Blossner M: The World Health Organization Global database on child growth and malnutrition: Methodology and applications. Int J Epidemiol 2003;32:518–526. Shrimpton R, Victora CG, de Onis M, et al: Worldwide timing of growth faltering: Implications for national interventions. Pediatrics 2001;107:e75. Prentice A: Nutrient requirements for growth pregnancy and lactation: The Keneba experience. S Afr J Clin Nutr 1993;6:33–38. Schroeder DG, Martorell R, Rivera JA, et al: Age differences in the impact of nutritional supplementation on growth. J Nutr 1995;125:1051S–1059S. Briend A, Hasan KZ, Aziz KMA, Hoque BA: Are diarrhoea control programmes likely to reduce childhood malnutrition? Observations from rural Bangladesh. Lancet 1989;ii:319–322. Lunn PG, Northrop-Clewes CA, Downes RM: Long-term growth faltering in Gambian infants is related to intestinal damage but not diarrhoeal prevalence. Trans R Soc Trop Med Hyg 1993;87:371. Lunn PG, Northrop-Clewes CA, Downes RM: Intestinal permeability, mucosal injury and growth faltering I Gambian children. Lancet 1991;338:907–910. Lunn PG: The impact of infection and nutrition on gut function and growth in childhood. Proc Nutr Soc 2000;59:147–154. Campbell DI, Murch SH, Elia M, et al: Chronic T-cell mediated enteropathy in rural West African children: Relationship with nutritional status and small bowel function. Pediatr Res 2003;54:306–311. Panter-Brick C, Lunn PG, Baker R, Todd A: Elevated acute phase protein in stunted children reporting low morbidity: Different rural and urban profiles. Br J Nutr 2001;85:125–131. Goto R, Panter-Brick C, Northrop-Clewes CA, et al: Poor intestinal permeability in mildly stunted Nepali children: Associations with weaning practices and Giardia lamblia infection. Br J Nutr 2002;88:141–149. Campbell DI, Elia M, Lunn PG: Growth faltering in rural Gambian children is associated with impaired small intestinal function leading to endotoxemia and systemic inflammation. J Nutr 2003;133:1332–1338. Popkin BM, Richards MK, Montiero CA: Stunting is associated with overweight in children of four nations undergoing the nutrition transition. J Nutr 1996;126:3009–3016. Monteiro CA, Conde WL, Popkin BM: Secular trends in obesity by social class: Northeast and South east of Brazil, 1975–1989–1997. Arq Braz Endocrinol Metab 1999;43:1–14. Sawaya AI, Grillo LP, Verreschi I, et al: Mild stunting is associated with higher susceptibility to the effects of high fat diets: Studies in a shantytown population in Sao Paulo, Brazil. J Nutr 1998;128:415–420. Sawaya AL, Martins P, Hoffman D, Roberts SB: The link between childhood undernutrition and risk of chronic diseases in adulthood: A case study of Brazil. Nutr Rev 2003;61:168–175. Ford ES, Giles WH, Dietz WH: Prevalence of the metabolic syndrome among US adults. JAMA 2002;287:356–359. Valasquez-Melendez G, Martins IS, Cervato AM, et al: Relationship between stature, overweight and central obesity in the adult population in Sao Paulo, Brazil. Int J Obes 1999;23:639–644. Ridker PM, Rifai N, Rose L, et al: Comparison of C-reactive protein and low density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N Engl J Med 2002;347: 1557–1565. Libby P, Ridker PM: Inflammation and atheroscherosis: Role of C-reactive protein in risk assessment. Am J Med 2004;116(suppl 6A):9S–16S. Ridker PM, Buring JE, Cook NR, Rifai N: C-reactive protein, the metabolic syndrome, and risk of incident cardiovascular events: An 8 year follow-up of 14719 initially healthy American women. Circulation 2003;107:391–397.
Malnutrition in the Developing World: The Lack of Food Scenario 35 Yajnik CS: The insulin resistance epidemic in India: Fetal origins, later lifestyle, or both? Nutr Rev 2001;59:1–9. 36 Chambers JC, Eda S, Bassett P, et al: C-reactive protein, insulin resistance, central obesity and coronary heart disease risk in Indian Asians from the United Kingdom compared with European whites. Circulation 2001;104:145–150. 37 Liu S, Manson JE, Buring JE, et al: Relation between a diet with a high glycaemic load on plasma concentrations of high-sensitivity C-reactive protein in middle-aged women. Am J Clin Nutr 2002;75:492–498. 38 Calder PC, Zurier RB: Polyunsaturated fatty acids and rheumatoid arthritis. Curr Opin Clin Nutr Metab Care 2001;4:115–121. 39 Grimble RF, Howell WM, O’Reilly G, et al: the ability of fish oil to suppress tumor necrosis factor ␣ production by peripheral blood mononuclear cells in healthy men is associated with polymorphisms in genes that influence tumor necrosis factor ␣ production. Am J Clin Nutr 2002;76:454–459. 40 Mohamed-Ali V, Pinklney JH, Coppack SW: Adipose tissue as an endocrine and paracrine organ. Int J Obes Relat Metab Disord 1998;22:1145–1158.
Discussion Dr. Haschke: Can you elaborate on anti-chymotrypsin? What does it really tell you, and is there anything in the elderly which is related to the activity? Dr. Labadarios: Anti-chymotrypsin is one of what we call a positive acute phase protein. It arises in the presence of infection or inflammation, just like C-reactive protein (CRP). The advantage of the chymotrypsin is that it stays elevated much longer than CRP, so there is the advantage of actually picking up an infectious or an inflammatory state for longer after the process of inflammation or infection has actually stopped. As for data in the elderly, I do not know of any that have been done in normal or apparently healthy elderly. The only relation that I can give you in that regard is the data of Ridker et al [1] from the United States which includes a component of the elderly population at risk of myocardial infarction, but that is in a group sense rather than in a sort of study in which a specific relationship has been studied. I don’t know whether anybody else here can help. I don’t know of any such data. Nevertheless, one must not forget that aging is an inflammatory process. Dr. Elia: You eluded to this syndrome, that seems to be quite common in developing countries, of stunting in overweight children. Presumably there is sufficient energy available, but the children are not growing. What do you think the mechanism might be? Why aren’t these children growing? Why are they stunted while at the same time they have excess energy? Presumably the dietary intake of energy is adequate but this does not allow normal growth to proceed. What is blocking it? Dr. Labadarios: I think your data give us a little bit of a glimpse of what may be happening. The energy question of course is a little bit more complicated and that was my purpose in showing those slides on weight loss in relation to CRP and glycemic load. It may be the case that it is not so much how much you eat, but what you eat. There was an excellent publication recently [2] regarding the obesity of poverty because obesity is equally common if not more common among poor people. So the question that arises here is that of the energy density of food and, in relation to that, we know that the more energy dense the food the greater the risk of weight gain, and in relation to inflammation, the greater the risk of having elevated inflammatory markers, at least CRP, that we know about. I would think the energy density of food is something that is emerging as a real entity in the developing world. The reason people eat energy-dense food is not by choice; so it happens that energy dense food is the cheapest food to eat, if you compare maize for instance with fruits and vegetables: people
9
Malnutrition in the Developing World: The Lack of Food Scenario cannot afford fruits and vegetables. So it is a matter of choice, it is a matter of circumstances, it is a matter of education, and quite honestly very often it is a matter of what is available. In the survey that I showed, we went into households that had no food in their cupboards; none. On the basis of a food inventory questionnaire we administered, in addition to other data. So if you have nothing to eat the cheapest food available is the one that you go for and with that goes energy density. I think that it is not the whole concept that is emerging; they are stunted and if they cannot grow in height, I suppose they can only ‘grow’ weight. And if they happen to eat dense food then the process is accelerated. Now I am quite sure it is not as simple as that, but at least that is what one can say at the moment. Dr. Powell-Tuck: I enjoyed your talk immensely and I obviously look at this sort of problem from the outside, working in the UK. But over the years we have been talked into distinguishing between kwashiorkor and marasmus: kwashiorkor on one hand being hypoalbuminemic, and marasmic children less so and with greater stunting. Yet in your studies and those that you quoted by Panter-Brick et al. [3], we are seeing that this hypoalbuminemia, low blood albumin is associated with cytokine drive, etc. I wonder if the classification is changing now or whether the emphasis on a distinction between kwashiorkor and marasmus is still as strong as it was traditionally 20 years ago? Would you like to comment on that? Again it is about stunting versus hypoalbuminemia and perhaps inflammation. Dr. Labadarios: I think the easiest point of reference would be the perfectly normal adult who encounters some kind of accident and is admitted to an intensive care unit. Now the albumin of that well-nourished person would obviously be well within the normal range, and yet if you measure the same albumin the following day a difference of 7–8 g can be found, and in my experience that depends on the severity of the injury. Granted there are a lot of fluid shifts in relation to trauma, and part of the decrease in plasma albumin is actually accounted for by that, but not all of it. At the moment, as best as we understand it, the mechanism is that the acute phase response switches off albumin synthesis in the liver in favor of the acute phase proteins synthesis which really explains what we observe. Now in analogy to what I described in the study in Nepal, I would not be surprised if the mechanism is not the same, except that the grade of inflammation is much lower. I am not saying that this is what it is, but I am saying that it is a real possibility because there are data from another field of nutrition [4] and the authors went to great pains to explain that there was no food shortage at the time of their study; so food was not plentiful but sufficient to meet nutrient requirements. As to the definition, I actually prefer to keep the definitions for marasmus and kwashiorkor until such time that we understand things a little better, because if I was to chose between the two I would rather chose to be marasmic than have kwashiorkor in terms of morbidity and mortality, particularly mortality. So having said that it is really a matter of time before these definitions actually change. I think Dr. Elia can help me here. In the study that I saw from Campbell et al. [5] late last year, they had a few marasmic children in their study, and interestingly their inflammatory response was suppressed. So it is not an easy sort of thing to come to grips with until we understand things a little bit better. Dr. Elia: I would just like to comment on that because of the complexity of the etiology of these overlapping syndromes. If you have a child in England who may be edematous and short as in the nephrotic syndrome, for example, what is the difference between that and kwashiorkor? There are difficulties in the definitions, and for this reason international agencies have tended to move a little away from classifying children as kwashiorkor and marasmic or marasmic kwashiorkor, and put marasmus on more descriptive terms such as edematous and non-edematous malnutrition for children. This is the position that is taken by the World Health Organization, for example, in their recent protocols about how to manage malnourished children.
10
Malnutrition in the Developing World: The Lack of Food Scenario Dr. Buchman: You mentioned the increasing intestinal permeability in these children and young adults in Africa. I have always had a hard time trying to resolve what this means and whether the increasing intestinal permeability seen in Crohn’s disease [6, 7], for example, which is also another intestinal inflammatory disorder, is really more of a cause and effect. But we do know that lactulose and mannitol are very small and that endotoxin is thousands of times larger than either of those molecules, so that increased permeability to these micromolecules does not suggest that there is an increased permeability to endotoxin and even bacteria for that matter. We know from studies done 25 years ago in Nigerian children that chronic parasitic infections are associated with hypoplasia as well as neutrophilic inflammation in the small intestine [8–10], and rather than the intestinal inflammation, as you suggested, being an effect of dietary antigen intake, I wonder if a better and perhaps alternative explanation would be chronic parasitic inflammation? Dr. Labadarios: The interesting part of your question regarding inflammatory bowel disease is that one of the major problems is failure to grow or failure to thrive. So at least we have one common sort of outcome. What Campbell et al.’s study does not show is a cause and effect relationship, but on the other hand there are the regressions that they reported. There is an overlap of permeability at the IgG antibody and growth and, if I remember correctly, that accounts for some 77% of the lack of growth or the slow growth. They seem to be part of one another rather than independent entities contributing to growth failure. In fact the article confirms that malnutrition affects the integrity of the gut, but now the concept is emerging that in fact it is the inflammatory process in the first instance that impairs permeability. In my opinion it is obviously an original and very interesting finding. It will have to be verified but we already have verification in terms of permeability from other countries like Bangladesh and Nepal. So the emerging concept is actually the reverse of what was previously thought to be the case, namely that the inflammatory process is started and actually induces the alterations in the structure function of the gut. Dr. Morley: I enjoyed your talk. I have two comments. The first is if we want to try and make an integrated approach to looking at malnutrition, the easier definitions in kwashiorkor and marasmus are fundamentally to talk about lack of food intake plus minus cytokine excess because I think it is the cytokine excess that creates the real differences. But listening to your talk reminds me a little bit of the work of Westerdope [11] in Holland who has been trying to prove why women who have very few babies live longer than those who have many babies, provided this happened before the 20th century. So you have got to understand that the problems of the so-called disposable theory only work up until modern times. He went to West Africa and looked at people who have a proinflammatory cytokine excess and used TNF-␣, and in those who have an anti-inflammatory, he used IL-10 as an anti-inflammatory. It turns out that if you are lucky enough to have a gene that allows you to produce a lot of cytokines you will in fact overcome a meningococcal infection and other infections. So if you start to look at it that way, it suggests that if you live in an angry environment, a place where there are lots of infections and very poor medicine, you would likely produce a lot of cytokines which would actually cause you to have many abortions as you got older because high cytokines produce spontaneous abortions, and you would live longer under those circumstances. If you make the mistake of then moving from this angry environment to a happy environment where you have good physicians with antibiotics and very few infections in the community because of good vaccinations, good housing and so on, you would fundamentally almost be the opposite, you now want an antiinflammatory environment. I think this is what we are seeing in the developed world; we now see this with the aging process which, I agree with you, is predominantly a cytokine-based process or the acceleration of it. Many people are now talking about
11
Malnutrition in the Developing World: The Lack of Food Scenario a cytokine-related aging process, but what is being seen is that as you get older, if you have this proinflammatory approach, you now do worse in a good environment. So I think we have got to start looking at two totally separate environments and understand that the so-called kwashiorkor child did worse than the marasmic child; the kwashiorkor child was almost always someone who had been exposed to an infection. Within that group of exposed subjects, we almost have to look for who does better and who does worse, and it may well be that those who actually look worse do slightly better. My memory from when I practiced medicine in South Africa a long time ago is that some of the children who I thought had no hope and had very low albumin actually did better in overcoming their infections than some of the others, using albumin as a marker for cytokine. So I think we can start to put it together. I know I haven’t done a good job of explaining Westerdope’s data, but every time I listen to him I get confused so I am not doing any worse most probably than he is, but I do think this proanti-inflammatory cytokine theory is very important for understanding how we interact with our environment. Dr. Labadarios: I think we are going to hear a lot about inflammation, we are going to hear a lot more about environment, and we are going to hear a lot more about genetics or nutrigenomics. But within the context of what we now know, your comments are extremely pertinent and I thank you for that. There are data available from India actually showing that if you are large and you stay in that percentile of largeness, or if you are small and you stay in that percentile of smallness, then you are OK, which implies a same environment. The minute you walk out of those percentiles then diabetes is 3 or 4 times higher. Of course regarding your comment of the adverse nutritional influences on top of an inflammatory background, I think that is a very pertinent and it made me include those two slides, one from the States and the other one from South America, I think it was Guatemala. We are going to hear a lot more about the role of chronic inflammation in relation to chronic disease. I never understood what hypertension has to do with smoking or with cholesterol. I never understood why we should ask people to exercise, and I don’t understand why red wine is good for you, which I incidentally like. Data are coming showing that alcohol reduces CRP levels. But don’t take that as a rule. There was a publication in the Archives of Internal Medicine last year showing that exercise reduces CRP levels [12]. So we are definitely going to hear a lot more about inflammation. Dr. Lochs: You first showed this nice correlation between permeability and inflammation, and now you mention a very interesting fact, alcohol and exercise increase permeability, and they still reduce the general inflammatory response, they reduce the CRP levels. How do you get that together? Dr. Labadarios: I don’t know. I did say they are emerging facts. The type of exercise in the study I referred to is not the “madness” that one sees in marathon running and all that, it is ordinary walking and so forth. I am not defending the study; I am just trying to put some background to your question. It is not severe exercise that will be associated with the type of permeability changes that you refer to. If there is any permeability change then of course we will wait to have a verification of the study and see what the exact mechanism is. At the moment all we know is that if one exercises regularly the chances are that CRP levels may be lower than those in individuals who do not exercise. We don’t understand it, but at least in relation to the point I made for the first time we appear to have a unifying sort of entity, including aging, that actually tries to put things together and I hope we will not be in too much of a hurry to put things together quickly and be wrong as we have been in the past in relation to food. Food on its own is extremely important, nobody doubts that; starvation kills, we know that. Dr. Oltersdorf: Thank you very much for this good discussion and I enjoyed it because I think nutrition is, as you said, at the end very easy. Food is important, no
12
Malnutrition in the Developing World: The Lack of Food Scenario food is death. But in between, things are quite complicated and of course I know red wine is good but too much red wine is not; permeability is important for nutrient, but too much permeability is worse; too much discussion is worse and therefore we stop here.
References 1 Ridker PM, Rifai N, Rose L, et al: Comparision of C-reactive protein and low-density lipoprotein cholesterol levels in the predition of first cardiovascular events. N Engl J Med 2002;347: 1557–1565. 2 Drewnowski A, Specter SE: Poverty and obesity: The role of energy density and energy costs. Am J Clin Nutr 2004;79:6–16. 3 Panter-Brick C, Lunn PG, Baker R, Todd A: Elevated acute-phase protein in stunted Nepali children reporting low morbidity: Different rural and urban profiles. Br J Nutr 2001;85: 125–131. 4 Campbell DI, Lunn PG, Elia M: Age-relaed association of small intestinal mucosal enteropathy with nutritional status in rural Gambian children. Br J Nutr 2002;88:499–505. 5 Campbell DI, Murch SH, Elia M, et al: Chronic T cell-mediated enteropathy in rural west African children: Relationship with nutritional status and small bowel function. Pediatr Res 2003;54:306–311. 6 Secondulfo M, de Magistris L, Fiandra R, et al: Intestinal permeability in Crohn’s disease patients and their first degree relatives. Dig Liver Dis 2001;33:680–685. 7 D’Inca R, Di Leo V, Corrao G, et al: Intestinal permeability test as a predictor of clinical course in Crohn’s disease. Am J Gastroenterol 1999;94:2956–2960. 8 Burman D: The jejunal mucosa in kwashiorkor. Arch Dis Child 1965;40:526–531. 9 Stanfield JP, Hutt MS, Tunnicliffe R: Intestinal biopsy in kwashiorkor. Lancet 1965;ii:519–523. 10 Campos JV, Neto UF, Patricio FR, et al: Jejunal mucosa in marasmic children. Clinical, pathological, and fine structural evaluation of the effect of protein-energy malnutrition and environmental contamination. Am J Clin Nutr 1979;32:1575–1591. 11 Westendorp RG: Are we becoming less disposable? EMBO Rep 2004;5:2–6. 12 Rothenbacher D, Hoffmeister A, Brenner H, Koenig W: Physical activity coronary heart disease and inflammatory response. Arch Intern Med 2003;163:1200–1205.
13
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 15–29, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
Are Older People Starving to Death in a World of Plenty? David R. Thomas Division of Geriatric Medicine, Saint Louis Health Sciences Center, Saint Louis, Mo., USA
‘Thousands of patients are annually starved in the midst of plenty.’ Florence Nightingale, 1859 ‘Doctors and nurses frequently fail to recognize undernourishment because they are not trained to look for it.’ J.E. Lennard-Jones, 1992
The indictment by Nightingale has persisted for over a century, aided by the inadequate attention paid by physicians to nutritional status. Dietary restrictions, improper dietary prescriptions, and keeping patients non per ora for considerable lengths of time have contributed to nutritional problems in the healthcare system [1, 2]. The nutritional status of older adults living at home is poor. On average, persons over the age of 70 years consume one third less calories compared to younger persons. Energy intakes of older men (40–74 years old) range from 2,100 to 2,300 cal/day compared to younger men (24–34 years old) who consume 2,700 cal/day [3]. Ten percent of older men and 20% of older women have intakes of protein below the US recommended daily allowance (RDA), and one third consume fewer calories than the RDA. Fifty percent of older adults have intakes of minerals and vitamins less than the RDA and 10–30% have subnormal levels of minerals and vitamins [4]. Sixteen to eighteen percent of community-dwelling elderly persons consume less than 1,000 kcal daily [5]. The drive to find food, designated by the term ‘hunger’, is essential in all species. Hunger is controlled by chemical mediators, signaling when to stop eating (‘satiation’), and when to resume searching for food (‘satiety’) which 15
Are Older People Starving to Death in a World of Plenty? Table 1. Conditions associated with undernutrition 1 Social factors, poverty, lack of socialization, help with meals 2 Mechanical barriers a Poor oral health status or hygiene, eyesight, motor coordination, taste alterations b Slow eating pace c Ethnic preferences or culturally acceptable food d Therapeutic or mechanically altered diet 3 Medical conditions leading to increased energy requirements a Cancer b Infections (acute and chronic) c Chronic obstructive pulmonary disease d Wounds, burns, and fractures 4 Medical conditions leading to interference with eating a Congestive heart failure b Malabsorption syndromes c Diabetic gastroparesis d Cholelithiasis 5 Psychological conditions a Depression b Dementia c Late-life paranoia d Anorexia nervosa
defines the interval between meals. ‘Appetite’, the enjoyment of food for itself, rather than for physiological need, is conditioned by a number of social, cultural, and psychological factors, as well as by disease states. Accumulating evidence points towards ‘anorexia’, the decline in appetite, as a major contributor to weight loss and undernutrition of older persons [6].
Sociological and Physiological Causes of Anorexia Caloric intake in older adults is greatly influenced by sociological, psychological, and physiological factors (table 1) [7]. Sociological factors include food preferences, chiefly determined by cultural circumstances. Older persons modulate food intake by time of day, number of people present, pre-meal stomach contents and their subjective state of hunger in ways similar to that of younger persons. Larger meals (10%) are eaten on weekends than weekdays, and larger meals are eaten later in the day. Provision of pleasant, welllighted, unhurried mealtimes in a social environment increases caloric intake [8]. Socialization greatly increases caloric intake. Women eat more (13%) when men are present, both genders eat more (23%) with family present. Meals eaten in group tend to be up to 44% larger than meals eaten alone [9]. 16
Are Older People Starving to Death in a World of Plenty? If the persons delivering meals-on-wheels deliveries stays while the older person eats, nutritional risk is reduced [9]. Physiological changes in the hedonic qualities of food occur universally with aging. The physiological alterations in taste and smell makes food appear less ‘tasty’ as we age. The major change in taste with aging is an increase in the taste threshold. The alteration in the taste threshold means that food presentation and food choice play a more important role than does the actual taste of the food. Flavor amplification may enhance food palatability and acceptance, stimulate salivary flow and reduce complaints concerning the oral cavity. Flavor enhancers have been shown to produce a tendency for ingestion of greater quantities of food and improved food preference [10].
The Influence of Medical Illness on Undernutrition Acute illness is characterized by a spontaneous decrease in food intake [11], a paradoxical response in the face of a need for increased nutrients during healing. A reduction in food intake accompanying acute illness occurs both before and during hospitalization. In the month before hospitalization, 65% of the males and 69% of the females had an insufficient energy intake, and undernutrition was present in 53% of males and 61% of females by the time of admission to the hospital [12]. Inadequate intake of nutrients often continues during hospitalization. In 286 general medical subjects, 27% became malnourished during hospital admission. These subjects were more likely to consume less than 40% of prescribed food, and were more likely to have lower Mini-Mental Status Examination scores, functional impairment, lower total lymphocyte counts, and lower serum albumin levels [13]. Over 90% of older persons admitted to a skilled care facility after hospitalization either have malnutrition or are at high risk of undernutrition [14]. Similarly, persons returning home from the hospital after an acute illness are at high risk of undernutrition. The differential diagnosis and management of undernutrition has been published as an algorithm [15]. Depression is one of the most common reversible causes of weight loss in elderly persons, accounting for up to 30% of undernutrition in medical outpatients [16], and 36% of residents in nursing homes [17]. Anorexia is frequently caused by medications [14]. Overzealous restriction in diet by physicians is a common reason for reduced caloric intake [18]. Special or restrictive diets (low cholesterol, low salt, no concentrated sweets) often reduce food intake without significantly helping the clinical status of the patient. For example, a regular diet does not affect glucose control in institutionalized diabetic older adults [19]. Recently evidence has emerged suggesting that inadequate caloric intake is associated with an excess production of cytokines. The syndrome of 17
Are Older People Starving to Death in a World of Plenty? Table 2. Distinguishing starvation from cachexia Starvation
Cachexia
Appetite Body mass index Serum albumin Cholesterol Total lymphocyte count
Suppressed in late phase Not predictive of mortality Low in late phase May remain normal Low, responds to refeeding
Cytokines Inflammatory disease Response to refeeding
Little data Usually not present Reversible
Suppressed in early phase Predictive of mortality Low in early phase Low Low, unresponsive to refeeding Elevated Present Resistant
cachexia, the cytokine-induced wasting of protein and energy stores, is related to a number of disease conditions including cancer [20], end-stage renal disease [21], chronic pulmonary disease [22], congestive heart failure [23], rheumatoid arthritis [24], and AIDS [25]. Cytokines associated with disease states directly result in feeding suppression and lower intake of nutrients. Interleukin-1 and tumor necrosis factor act on the glucose-sensitive neurons in the ventromedial hypothalamic nucleus (a ‘satiety’ site) and the lateral hypothalamic area (a ‘hunger’ site) [26]. The data suggest that cytokine levels are commonly associated with disease conditions characterized by cachexia, and may play a role in appetite suppression, mortality, and weight loss. Cytokine-induced anorexia is the most common cause of poor caloric intake observed in the acute care setting [27], and affects community-dwelling older persons as well.
Distinguishing Starvation from Cachexia Are patients starving in the midst of plenty? Or rather, is the high prevalence of undernutrition reported in hospital and community settings due to disease? Simplistically, the reduction in food intake seen in older adults is often equated with starvation. Worldwide, starvation is most often caused by lack of food. However, a remarkably similar incidence of undernutrition among hospitalized adults has been noted worldwide, suggesting that disease rather than lack of food may be the common denominator in developed countries [28] (table 2). The distinction is important since the strategies to correct cachexia may be different from the simple correction of starvation. Clinicians frequently confuse the effect of cytokine-induced acute-phase reactants on biochemical variables with undernutrition. However, these biochemical markers are not specific for nutritional status. For example, 18
Are Older People Starving to Death in a World of Plenty? hypoalbuminemia occurs in disease states such as hepatic disease, renal disease, and congestive heart failure [29], stress [30], and occurs after 8 h of bed rest [31]. The total lymphocyte count correlates poorly with both the body cell mass and the nutritional state measured by the Nae to Ke ratio, producing a false-positive rate of 34% and a false-negative rate of 50% for diagnosing undernutrition [32]. In contrast to cytokine-mediated disease states, few changes occur in biochemical markers in simple starvation. Serum albumin remains normal in both short-term and long-term fasting [33]. In fact, after 9 weeks on a diet of about half the normal dietary intake of protein, serum albumin remained normal despite changes in lean body mass and immune status [34]. Serum albumin levels in anorexia nervosa, a condition of chronic energy deficiency, remain normal and serum cholesterol levels increase in one third of anorexia patients. However, chronic inadequate intake of protein (kwashiorkor) does lead to a decline in serum albumin levels. Use of biochemical markers as nutritional parameters can potentially lead to over-diagnosis of undernutrition [35]. A screening instrument that has less reliance on biochemical markers, such as the Mini-Nutritional Assessment, has been shown to be effective in older adults [36].
Nutritional Interventions The first response to clinical signs of undernutrition, whether due to starvation or cachexia, is to increase nutrient intake. Starvation due to inadequate food sources responds to hypercaloric feeding in both children and adults. Provided absorption is intact, repletion of serum albumin, cholesterol, improved immune function, and weight gain should be achievable in starvation states [37]. Increased voluntary consumption of adequate calories has been effective in producing weight gain. In a meta-analysis of 15 randomized, controlled clinical trials of dietary advice with or without nutritional supplements, the nutritionally supplemented group had a gain in weight (weighted mean difference ⫺1.14 kg [95% CI ⫺1.94, ⫺0.33]) [38]. Oral supplementation improved outcome in older patients with a body mass index of less than the 75th percentile, who were free of cancer or dysphagia. Compared to control subjects, patients who received a sip feed had weight gain and improved energy intake. Mortality and functional status were improved in the most undernourished patients, although total group mortality did not change (odds ratio 0.62, 95% CI 0.35, 1.13). In contrast to starvation, cachexia is remarkably resistant to hypercaloric feeding (table 3). The provision of additional calories and protein alone has not been shown to be efficacious in patients with cancer cachexia [39], in renal, or in cardiac cachexia [14]. 19
Are Older People Starving to Death in a World of Plenty? Table 3. Frequency of moderate and/or severe undernutrition in acute hospitals Admission site
Moderate or severe SGA
Country
Year
Acute hospital (CRI admission) Acute hospital (n ⫽ 2) Acute hospital Acute geriatric hospital Acute hospital (n ⫽ 13) Acute hospital Acute hospital Acute hospital (dialysis) Acute geriatric hospital Geriatric long-term care Oncology [48]
28% (severe) 36% (severe) 45% (severe) 41% 50% 53% 61% 65% 69% 70% 76%
Australia Australia Netherlands USA Latin America Sweden Switzerland United Kingdom Sweden USA Australia
2001 1997 1997 1999 2001 1996 2002 1997 2002 2000 2002
Undernutrition was assessed using the Subjective Global Assessment (SGA). Several studies reported combined moderate and severe undernutrition, while others reported only severe undernutrition. Adapted from Thomas [28].
Pharmacological Interventions Depression is the most common treatable cause of undernutrition, and its treatment may lead to reversal of weight loss. Tricyclic antidepressants and monoamine oxidase inhibitors are more likely to produce weight gain than the selective serotonin reuptake inhibitors or the newer antidepressants. Mirtazapine appears to be particularly useful in stimulating appetite [40]. Orexigenic drugs have shown promise in improving appetite and producing weight gain in older adults. Cannabinoids (dronabinol, marinol, and nabilone) have improved appetite in cancer patients [41, 42] and in AIDS cachexia [43]. Megestrol acetate has been shown to improve appetite and stop weight loss in older adults with cancer, AIDS, and weight loss [44]. Interestingly, an anticytokine effect may be responsible for weight gain in older adults treated with megestrol acetate [45]. Melatonin, at a dose of 20 mg/day, stabilized weight and decreased tumor necrosis factor levels in cancer-related weight loss compared to controls [46]. Treatment with an etanercept, an anticytokine drug, has been demonstrated to results in a significant in left ventricular structure and function and a trend toward improvement in patient functional status in patients with advanced heart failure [47]. These results suggest that anticytokine treatment may be effective in the treatment cachexia-mediated syndromes. Although much work remains to be done, anticytokine drugs appear to be a promising avenue for the treatment of involuntary weight loss. 20
Are Older People Starving to Death in a World of Plenty? Conclusion It is possible to starve in a world of plenty. However, in the presence of adequate food, the failure to gain weight most often is due to cytokineassociated suppression of appetite. Assessment of changes in appetite are essential to evaluating older persons with weight loss. When anorexic changes are identified, a search for reversible causes should be instituted. Intervention should first be aimed at the provision of adequate calories and protein, often in the form of high-density nutritional supplements. The chief difference between starvation and cachexia is that refeeding reverses starvation, but is less effective for cachexia. If weight loss continues, use of an orexigenic drug should be considered. Although much work remains to be done, anticytokine drugs appear to be a promising avenue for the treatment of involuntary weight loss. References 1 Sullivan DH, Moriarty MS, Chernoff R, et al: Patterns of care: An analysis of the quality of nutritional care routinely provided to elderly hospitalized veterans. JPEN J Parenter Enteral Nutr 1989;13:249–254. 2 Kamath SK, Lawler M, Smith AE, et al: Hospital malnutrition: A 33-hospital screening study. J Am Diet Assoc 1986;86:203–206. 3 McGandy RB, Barrows CH Jr, Spanias A, et al: Nutrient intake and energy expenditure in men of different ages. J Gerontol 1966;21:581–587. 4 Ritchie CR, Thomas DR: Aging; in Heimburger DC, Weinsier RL (eds): Handbook of Clinical Nutrition, ed 3. St Louis, Mosby, 1997. 5 Abraham S, Carroll MD, Dresser CM, et al: Dietary Intake of Persons 1–74 Years of Age in the United States. Advance Data from Vital and Health Statistics of the National Center for Health Statistics No. G. Rockville, Public Health Service, 1977. 6 Morley JE: Decreased food intake with aging. J Gerontol Ser A Biol Sci Med Sci 2001;56:81–88. 7 Morley JE, Thomas DR: Anorexia and aging: Pathophysiology. Nutrition 1999;15:499–503. 8 Kayser-Jones J: Mealtime in nursing homes: The importance of individualized care. J Gerontol Nurs 1996;22:26–31. 9 de Castro JM, Stroebele N: Food intake in the real world: Implications for nutrition and aging. Clin Geriatr Med 2002;18:685. 10 Schiffman SS, Warwick ZS: Effect of flavor enhancement of foods for the elderly on nutritional status: Food intake, biochemical indices, and anthropometric measures. Physiol Behav 1993;53:395–402. 11 Plata-Salaman CR: Anorexia during acute and chronic disease. Nutrition 1996;12:69–76. 12 Mowe M, Bohmer T, Kindt E: Reduced nutritional status in an elderly population (⬎70 y) is probable before disease and possibly contributes to the development of disease. Am J Clin Nutr 1994;59:317–24. 13 Incalzi RA, Gemma A, Capparella O, et al: Energy intake and in-hospital starvation. A clinically relevant relationship. Arch Intern Med 1996;156:425–429. 14 Thomas DR, Zdrowski CD, Wilson MM, et al: Malnutrition in subacute care. Am J Clin Nutr 2002;75:308–313. 15 Thomas DR, Ashmen W, Morley JE, Evans JE: Nutritional management in long-term care: Development of a clinical guideline. J Gerontol A Biol Sci Med Sci 2000;55:725–734. 16 Wilson MM, Vaswani S, Liu D, et al: Prevalence and causes of undernutrition in medical outpatients. Am J Med 1998;104:56–63. 17 Morley JE, Kraenzle D: Causes of weight loss in a community nursing home. J Am Geriatr Soc 1994;42:583–585.
21
Are Older People Starving to Death in a World of Plenty? 18 Buckler DA, Kelber ST, Goodwin JS: The use of dietary restrictions in malnourished nursing home patients. J Am Geriatr Soc 1994;42:1100–1102. 19 Tariq S, Karcic E, Thomas DR, et al: The use of a no-concentrated-sweets diet in the management of type 2 diabetes in nursing home patients. J Am Diet Assoc 2001;101:1463–1466. 20 Shike M, Russell DM, Detsky AS, et al: Changes in body composition in patients with smallcell cancer. The effect of total parenteral nutrition as an adjunct to chemotherapy. Ann Intern Med 1984;101:303–309. 21 Mitch WE: Mechanisms causing loss of lean body mass in kidney disease. Am J Clin Nutr 1998;67:359–366. 22 Hedlund J, Hansson LO, Ortqvist A: Short- and long-term prognosis for middle-aged and elderly patients hospitalized with community-acquired pneumonia: Impact of nutritional and inflammatory factors. Scand J Infect Dis 1995;27:32–37. 23 Toth MJ, Gottlieb SS, Goran MI, et al: Daily energy expenditure in free-living heart failure patients. Am J Physiol 1997;272:469–475. 24 Roubenoff R, Roubenoff RA, Cannon JG, et al: Rheumatoid cachexia: Cytokine-driven hypermetalboism accompanying reduced body cell mass in chronic inflammation. J Clin Invest 1994;93:2379–2386. 25 Kotler DP, Wang J, Pierson RN: Body composition studies in patients with the acquired immunodeficiency syndrome. Am J Clin Nutr 1985;42:1255–1265. 26 Espat NJ, Moldawer LL, Copeland EM 3rd: Cytokine-mediated alterations in host metabolism prevent nutritional repletion in cachectic cancer patients. J Surg Oncol 1995;58:77–82. 27 Rote NS: Inflammation; in McCance KL, Huether SE (eds): Pathophysiology: The Biological Basis for Disease in Adults and Children. St Louis, Mosby, 1998, pp 205–236. 28 Thomas DR: Starving in the hospital. Nutrition 2003;19:907–908. 29 Friedman PJ, Campbell AJ, Caradoc-Davies TH: Hypoalbuminemia in the elderly is due to disease not malnutrition. J Clin Exp Gerontol 1985;7:191–293. 30 Ingenbleek Y, Carpentier YA: A prognostic inflammatory and nutritional index scoring critically ill patients. Int J Vitam Nutr Res 1985;55:91–101. 31 Hyltoft PP, Felding P, Horder M, Tryding N: Effects of posture on concentration of serum proteins in healthy adults. Scand J Clin Lab Invest 1980;40:623–628. 32 Forse RA, Rompre C, Crosilla P, et al: Reliability of the total lymphocyte count as a parameter of nutrition. Can J Surg 1985;28:216–219. 33 Hoffer LJ, Bistrian BR, Young VR, et al: Metabolic effects of carbohydrate in low-calorie diets. Metabolism 1984;33:820–825. 34 Castaneda C, Charnley JM, Evans WJ, Crim MC: Elderly women accommodate to a lowprotein diet with losses of body cell mass, muscle function, and immune response. Am J Clin Nutr 1995;62:30–39. 35 Rosenthal AJ, Sanders KM, McMurtry CT, et al: Is malnutrition overdiagnosed in older hospitalized patients? Association between the soluble interleukin-2 receptor and serum markers of malnutrition. J Geront A Biol Med Sci 1998;53:M81–M86. 36 Guigoz Y, Vellas B: The mini-nutritional assessment for grading the nutritional state of elderly patients: Presentation of the MNA, history and validation; in Vellas B, Garry PJ, Guigoz Y (eds): Mini Nutritional Assessment (MNA): Research and Practice. Nestlé Nutrition Workshop Series Clinical and Performance Programme. Vevey, Nestec/Basel, Karger, 1998, vol 1, pp 3–12. 37 Mehler PS, Gray MC, Schulte M: Medical complications of anorexia nervosa. J Womens Health 1997;6:533–541. 38 Anonymous: The Cochrane Database of Systematic Reviews, issue 2, 2001. 39 Fearon KC, Barber MD, Moses AG: The cancer cachexia syndrome. Surg Oncol Clin North Am 2001;10:109–126. 40 Fava M: Weight gain and antidepressants. J Clin Psychiatry 2000;61(suppl 11):37–41. 41 Plassee TF, Gorter RW, Krasnow SH, et al: Recent clinical experience with dronabinol. Pharmacol Biochem Behav 1991;40:695–700. 42 Nelson K, Walsh D, Deeter P, Sheehan F: A phase II study of delta-nine-tetrahydrocannabinol for appetite stimulation in cancer-associated anorexia. J Palliat Care 1994;10:14–18. 43 Beal JE, Olson R, Laubenstein L, et al: Dronabinol as a treatment for anorexia associated with weight loss in patients with AIDS. J Pain Symptom Manage 1995;10:89–97. 44 Thomas DR, Ashmen W, Morley JE, Evans WJ: Nutritional management in long-term care: Development of a clinical guideline. J Gerontol A Biol Sci Med Sci 2000;55:M725–M734.
22
Are Older People Starving to Death in a World of Plenty? 45 Yeh SS, Wu SY, Levine DM, et al: The correlation of cytokine levels with body weight after megestrol acetate treatment in geriatric patients. J Gerontol A Biol Sci Med Sci 2001; 56:M48–M54. 46 Lissoni P, Paolorossi F, Tancini G, et al: Is there a role for melatonin in the treatment of neoplastic cachexia? Eur J Cancer 1996;32A:1340–1343. 47 Bozkurt B, Torre-Amione G, Warren MS, et al: Results of targeted anti-tumor necrosis factor therapy with etanercept (ENBREL) in patients with advanced heart failure. Circulation 2001;103:1044–1047. 48 Bauer J, Capra S, Ferguson M: Use of the scored Patient-Generated Subjective Global Assessment (PG-SGA) as a nutrition assessment tool in patients with cancer. Eur J Clin Nutr 2002;56:779–785.
Discussion Dr. Haschke: Concerning anti-inflammatory effects, could you elaborate on n-3 fatty acids? Have further studies been done in a more general population? The second question: is there also something in the literature regarding hormonal treatment with testosterone and growth hormone? Dr. Thomas: Testosterone has been looked at in sarcopenic older men and in hypogonadal older men [1, 2]. Testosterone will produce weight gain and increase lean body mass. There are some concerns that limit the use of testosterone in this population, particularly related to the question of prostatic hypertrophy, prostatic cancer, and the effect on the hematological system. Testosterone does not increase appetite. It produces weight gain without necessarily increasing nutrient intake. For that reason it is acting potentially as an anabolic steroid rather than an anti-cytokine drug. The lack of effect on appetite at least suggests that it is probably not suppressing the cytokine-induced anorexia. Testosterone clearly improves functional outcome. Growth hormone will do the same thing, acting as an anabolic steroid to produce weight gain. Growth hormone has been shown to produce small gains in functional outcomes in elderly patients, particularly on knee strength [3, 4]. The problem with the use of growth hormone has been some unpleasant side effects and the obvious problem of the cost of the drug. The chief factor in the United States that has limited the use of growth hormone has been the fact that when it was used in a critical care situation there was an increase in mortality. This leads to some speculation. One universal phenomenon that occurs is that when you get acutely ill, you decrease your food intake. This may be either a side effect of cytokine production or it may be a teleological protective effect, implying that you should go through a period of starvation when you get critically ill. If you are able to suppress the cytokines in that population and you are able to increase nutrient intake, you actually may do some harm. This is speculative but at least there is concern that there may be a protective effect of inflammation in critically ill patients. Dr. Ockenga: One problem is that in acute inflammation we also mobilize a lot of antigen substrates like glucose, etc. In intensive care patients, external energy or external food may not really have a beneficial effect. So I think we also have to focus on the metabolic changes due to chronic inflammation and in this situation we have to think about whether increasing food may really be of benefit to our patients. Dr. Thomas: I think that is a very good point. When we talk about inflammation and cytokine excess we need to distinguish acute from chronic states. I think they are very different fields. I am not sure that I know what we ought to do in the acute setting. If we target decreasing cytokines in an acute setting we may actually do harm. However, in chronic settings, such as in children with chronic inflammatory disease who may not be able to absorb nutrients and in older patients with chronic inflammatory
23
Are Older People Starving to Death in a World of Plenty? disease, we often don’t improve nutritional status even though we supply increased nutrition. So I think that we must distinguish chronic disease from acute illness. Dr. Ockenga: I would also like to mention the recent article in Science [5] describing the changes in muscle related to age, increasing the fat content of the muscle and increasing peripheral insulin resistance, so there is also a change in the metabolic as well as the hormonal status of these patients. This may also be a link to our failure to improve the nutritional status of the patients. Dr. Thomas: I agree. Dr. Schiffrin: If we go back to the aspect of cytokines and sarcopenia, I would like to know what are the most important in terms for inducing sarcopenia? Does that mean a reaction of the cytokines with receptors in muscle cells, and if so what is the effect? Is the intracellular effect due to a lack of synthesis or increased catabolism of proteins? What pathway of proteolytic events are induced by this cytokine reaction? Dr. Thomas: I don’t think we know the mechanisms as specifically as we would like. At the cellular level I cannot tell you what the pathways are; someone else may have a comment. Most of the literature has been observational associations. In other words, we find that the patients who have elevated C-reactive proteins also have an increase in heart disease. In patients who have renal cachexia, for example, IL-2 and IL-6 are usually associated with weight loss and IL-1 is not. In other studies of renal cachexia, other cytokines have and have not been associated with weight loss. If you look there is a lot of variability in the literature and I don’t think that a specific cytokine can be consistently associated with a specific cachectic state. Dr. Morley: Major work has been done in this area by Roubenoff [6], suggested that it is low grade proteolysis for sarcopenia as opposed to high grade. In addition there is an alteration in the IgF within the muscle, and the work by Musaro et al. [7] in Rome has shown that if you give old animals stem cells with IgF you can basically reverse the sarcopenia that you see as opposed to cachexia. So it looks as if the IgF and the cytokines are playing a role in sarcopenia. The third component of sarcopenia is testosterone and Herbst and Bhasin [8] have recently shown very nicely that at the stem cell level testosterone actually suppresses the production of adipocytes and increases the production of satellite cells. The satellite cells then fuse with the muscle and allow repair of the muscle. This really leads to one of the conundrums with testosterone and perhaps also with growth hormone. Both of these cause a much bigger muscle mass gain than they do a strength gain, and it appears that repair takes place but you don’t necessarily improve the true quality of the fibers when you repair with hormones. Dr. Bozzetti: I would like you to expand a bit more on the effect of megestrol in geriatric wasting and especially what the effect on body composition is? We have experience in the oncological field that usually megestrol works to increase body weight but the composition is mainly made up of water or fat tissue. So what is the effect in elderly subjects? Dr. Thomas: Very interesting question. It was generally assumed that as someone diets or starves, they lose fat mass and then muscle mass. This is the reason why we all diet, to get rid of fat. We maintain our muscle and then we look nice. As you recover from starvation, what should happen teleologically is that you should repair muscle mass first, get strong, and then you should go on to gain fat mass and replace your shortage of fat. Paradoxically, what really happens in recovery from starvation is that fat mass increases first before lean body mass, whether you are doing this nutritionally or with megestrol acetate or some other drugs. This has lead to the idea that we should use anabolic steroids, such as testosterone or growth hormone, to increase lean body mass first because that will give us an improvement in function, and then we can go on to replete the patient nutritionally and regain fat mass. This is a great
24
Are Older People Starving to Death in a World of Plenty? scenario but it doesn’t work very well in practice. Regarding megestrol acetate specifically, the gain is predominantly fat mass over lean mass and is not solely due to fluid gain [9, 10]. I suggest that megestrol acetate may suppress a chronic cytokine excess and improve appetite and energy intake. The fat mass increase that occurs may or may not be associated with improvements in outcome, and that is what is not clear. Dr. Cynober: I would like to comment on 3 points. First, the study you mentioned about testosterone supplementation in elderly subjects was a short-term administration, I think up to 1 month, with very impressive results [11]. My concern is that sarcopenia is a long-term disease and nobody can imagine providing intravenous testosterone for years. Second, just a complement to the debate about the mechanism of action of cytokines on muscle protein turnover; it is mainly a proteasome ubiquitin-dependent system but probably not directly mediated by cytokines but by the interaction between cortisol and cytokines at the muscle level. And third, I agree with all the hypotheses discussed about a hormone-related mechanism for sarcopenia including IGF1, human growth hormone, insulin sensitivity, but I would like to call to your attention that to make proteins you need amino acids. It has been repetitively observed by Volpi et al. [12] in the United States and Boirie et al. [13] in ClermontFerrand that in elderly subjects the sequestration of amino acids after a meal is doubled compared to young adults. As a consequence the amount of amino acids in the periphery available for protein synthesis is truly decreased in elderly. In addition, when amino acids are provided by the parenteral route, in this condition the fractional synthetic rate of protein in the muscle is normal in these elderly subjects, and this has also been shown in elderly rats. Dr. Morley: Can I just comment on the testosterone studies. The original studies were short-term and one of ours was 3 months, but subsequently we published studies of 1 year in length [1, 14, 15] and there are now 2 studies in the literature of 3 years in length. All show a reversal of sarcopenia and in the middle-aged population. Wang et al. [16] now have follow-up data for up to 5 years I think, so realistically testosterone does remain a possibility as a treatment. There is no good long-term study looking at the side effects which we are left with, and that is a major need to decide whether this is a reasonable approach. But the studies are as long as we see with anything else of the stage. Dr. Thomas: Let me also add that not all men get hypogonadal as they age. In fact bioavailable testosterone stays normal in a fairly substantial number of men as they age. There is little data suggesting that testosterone or other anabolic steroids have a lot of effect in those patients. So we are talking about patients who have true hypogonadism, whose sarcopenia is not due to amino acid problems or starvation problems or other problems, but it is strictly due to the lack of testosterone. In those patients we certainly can improve functional status, muscle mass and produce weight gain. This is limited to patients who have low bioavailable testosterone. Dr. Volkert: I enjoyed your talk very much and would like to emphasize the importance of physical activity besides the supply of amino acids and enough nutrients. Especially in nursing home patients it is very important that they are physically active in order to increase the fat-free mass and body cell mass. Dr. Thomas: Thank you very much for making that point because this is very important. One of the clearly demonstrated things is that physical activity will lower cytokines and that may be the source of the benefit. It is critical to remember that exercise is a good way to lower cytokines without drugs. Dr. DeLegge: I wanted to tap your brain for a second with regard to supplementation that you touched upon in the elderly. I am not aware of any data that have ever shown that enteral supplementation in patients who can swallow changes outcome. In a nursing home population the response of most physicians or caregivers for someone
25
Are Older People Starving to Death in a World of Plenty? who is elderly and able to eat is to provide them with some sort of supplementation either in a can or some other consistency. My question is, is that the right approach to these patients? Dr. Thomas: This is one of the problems that catches your attention in the literature on nutritional studies. In a meta-analysis of quality studies, the best evidence is that there is an effect of nutritional supplements in lowering mortality. There is also an effect to produce gain in body weight. Even in persons 90 years old, adding nutritional supplements to physical activity has an effect to improve functional status and weight gain. So I think we have got data. Part of the issue is that the magnitude of the effect is somewhat small. My hypothesis is that we are not properly analyzing the population. We are observing a general population in which we have some people who have no response to nutrition interventions and some people who have a good response, and when we pull them together we observe a small positive effect. What I would like to see us do is to see if we can divide the population who have a cachexia syndrome from the population that does not have cytokine excess and see if there is not a better improvement in the non-cachectic patients. Perhaps the poor improvement occurs only in the cachectic syndrome. We are in the process now of trying to set up a study to do that and hopefully we are going to get some data that will answer this question because I think that is where the real question lies. Clearly you get a positive but small magnitude effect in a general population and I think that the small magnitude effect may be because of the inclusion of cachectic non-responders. Dr. Labadarios: My question relates to the weight gain concept. Should we continue to talk about weight gain or should we in the future be a little bit more fancy as opposed to talking about other things like body composition and particularly muscle mass since we know weight per se, particularly fat weight, may not actually be such a good idea. Dr. Thomas: I agree, and I think it is time that we move on. We have had a lot of studies that measure weight and body composition. I think where we need to start trying to really look at outcome. What we want to be able to do is to either improve quality of life or functional status. Those are the two issues. If we can improve quality of life in our elderly population, does it matter whether they gain weight? If enjoyment of life improves, does it matter whether their body composition is fat or lean? So I think you are right and that is the next step. I think we can, but to do this we need to demonstrate it. Dr. Elia: I agree with you that the most important next step is to try and get evidence of functional benefits as a result of interventions. In relation to the previous question that was raised about enteral tube feeding in those who don’t have swallowing difficulties, we have undertaken a meta-analysis that shows functional improvements and a reduction in complications. The question I wanted to ask was about the issue of lean and fat mass loss during starvation. You indicated that starvation conserves lean mass and depletes fat mass. However, over the years, we and others have proposed alternative models in which lean and fat tissues talk to each other, so that the proportion of lean and fat that is lost during starvation depends on the initial body composition: A lean person loses a greater proportion of lean to fat tissue during starvation than an obese individual. The differences have implications for prolonged survival. Dr. Thomas: Thank you for clarifying that because what we were led to believe was that you would lose fat and conserve lean body mass. In fact, in starvation you lose both lean and fat mass, and there are differences in the portions of how much you lose. But obviously it is not fat followed by lean. You lose both at the same time and you lose functional ability at the same time. This has enormous implications for the
26
Are Older People Starving to Death in a World of Plenty? obese person. I have talked mostly about undernutrition but in obese people who lose weight that is a significant problem. Dr. Elia: I think it also has implications for certain chronic diseases including HIV where these issues are being explored. Dr. Schwab: I would like to come back to the pharmaceutical interventions influencing outcome. I am thinking about an agent that was demonstrated to reduce TNF-␣ on the one hand and has another capacity in terms of positively influencing sleeping disorders, which are actually quite common in elderly people. What I am talking about is thalidomide which was shown to reduce TNF-␣ and was used positively in Crohn’s disease for instance. Could you comment on this agent in a rather low dose because in inflammatory bowel disease the toxicity is rather large in a high dose of 150–300 mg. But at a lower dose, let’s say 50 or 100 mg, the toxicity is less. Could you comment on this agent please? Dr. Thomas: A very good point and it was one of the drugs I did not mention. There are very few drugs that we know of that have a strong anti-cytokine effect. Thalidomide is certainly one of them. My understanding is that we can get some anticytokine suppression and lose some of the side effects if we deal with isomers, but clearly it is an effective anti-cytokine drug and there are good data suggesting that you can improve weight in patients who have cachexia syndrome. The other drug is pentoxifylline, which is another unusual drug that reduces anti-cytokines. In general what we are seeing is a clinical observation. We have these drugs some of which improve appetite and weight. What they have in common is an anti-cytokine effect. None of these drugs were designed to do that. What I would like to see is that the experts here come up with ideas to develop a better drug that we can give along with nutritional supplements and thus solve the problem of weight loss. Dr. Lochs: About the cytokines: are we talking about one single mechanism which is probably mediated by different cytokines or could it be that we are talking about different mechanisms? As Dr. Morley said, one cytokine might influence proteolysis, another cytokine might influence appetite regulation, and a third cytokine might influence absorption in the intestine. If so, could you tell us which cytokine is doing what? Dr. Thomas: At least from my standpoint, I think we are in an observational mode now. As you know great advances in clinical medicine are made by people who just simply observe things that they see going on clinically. In the current literature what we are observing is a relationship. We are observing a profound relationship between C-reactive protein, which is very nonspecific, and heart disease. We are seeing a profound relationship between certain cytokines and renal cachexia. However, the relationship is confusing. One study says IL-2, one says IL-6, another says it is TNF-␣. I don’t think we know. The concept is that there is a chronic inflammatory disease that may be occurring in the gut of malnourished children, and a cachexia syndrome occurring in our older patients. We now have anti-cytokine drugs that can suppress cytokines, for example in rheumatoid arthritis. It would be wonderful to see what happens to body weight in that population. Obviously function improves, but does body weight or appetite improve with these drugs? I don’t think we know. We don’t know all the cytokines that are involved. Can we manipulate cytokines in such a way that we could have a positive outcome? Dr. Labadarios: I enjoyed your comment and your thoughts on the question of cytokines, the pharmacological manipulation of cytokines. Is it a question of wanting to be able to alter cytokine profiles or is it a question of whether we are knowledgeable enough to interfere with the cytokines? In the acute situation we know manipulation with growth hormone is associated with significant adverse clinical outcomes [17]. Now what do we know in terms of whether it is desirable to actually manipulate the cytokine balance, and obviously what are the implications of that question?
27
Are Older People Starving to Death in a World of Plenty? Dr. Thomas: An excellent observation, I can tell you that we do not know. The state of the art is advancing, but this is obviously very empirical, very observational, and the actual question that needs to be answered is the one you posed. Does manipulation of this chronic inflammatory state improve outcome? We know we can manipulate body composition, we know we can manipulate anabolic hormones, but that may not be good and I think that is where we need to go. Dr. Ockenga: May I just add a recent article by Schwenk et al. [17] from England as well as one from Paton et al. [18] showing the treatment of chronic tuberculosis. The patients were followed for half a year after sufficient antimicrobial therapy, and it was shown that they gain weight, but they gain fat mass and not muscle mass, so function did not really improve. We can assume that in these patients the inflammatory response is decreased. Still this alone would not explain why they did not gain muscle mass. Dr. Thomas: It may be related to exercise in that situation. Perhaps you need physical training plus the manipulation of the weight gain. Dr. Schiffrin: If we can speculate a little bit, one tries to imagine what is feeding this cytokine production. For example, we have discussed previously that the gut or bacterial translocation or endotoxemia could be a factor playing a role here. Do we have any data showing that the gut is probably a source of keeping this low noise inflammatory condition well known in the elderly? Dr. Thomas: The short answer is I don’t think we have a lot of data. We do have some observational data. I asked Dr. Labadarios whether he has tried to treat these young patients with anti-inflammatory drugs to see if we could not remove the problem by treating the cytokines. Do you want to comment on that? Dr. Labadarios: Very important question, but I don’t think in the end we are going to have a simple answer that it is one organ and nothing else, it is the gut and nothing else. I mentioned earlier that we don’t know what cholesterol and hypertension or other such risk factors for heart disease have in common. One of the risk factors also for heart disease is periodontal disease. People with periodontal disease have significantly higher C-reactive protein levels than those without periodontal disease [20]. So I think any low grade inflammation is theoretically important in the initiation or progression of disease. Periodontal disease in relation to heart disease makes dentistry and dental hygiene, in terms of environmental influences, crucially important and would be applicable in the elderly. Dr. Thomas: One of the other examples of anti-cytokine drugs is statins, the cholesterol-lowering agents. It is now clear that if you give one of these drugs to a patient with acute coronary syndrome on admission to the hospital, you can decrease the risk of subsequent myocardial damage by 29%. This has nothing to do with the cholesterol effect. It is postulated now that it is an anti-cytokine effect. So there is another anti-cytokine drug that we ought to think about exploring.
References 1 Sih R, Morley JE, Kaiser FE, et al: Testosterone replacement in older hypogonadal men: A 12-month randomized controlled trial. J Clin Endocrinol Metab 1997;82:1661–1667. 2 Morley JE, Perry HM 3rd: Androgen deficiency in aging men: Role of testosterone replacement therapy. J Lab Clin Med 2000;135:370–378. 3 Hennessey JV, Chromiak JA, DellaVentura S, et al: Growth hormone administration and exercise effects on muscle fiber type and diameter in moderately frail older people. J Am Geriatr Soc 2001;49:852–858.
28
Are Older People Starving to Death in a World of Plenty? 4 Lange KH, Andersen JL, Beyer N, et al: GH administration changes myosin heavy chain isoforms in skeletal muscle but does not augment muscle strength or hypertrophy, either alone or combined with resistance exercise training in healthy elderly men. J Clin Endocrinol Metab 2002;87:513–523. 5 Petersen KF, Befroy D, Dufour S, et al: Mitochondrial dysfunction in the elderly: Possible role in insulin resistance. Science 2003;300:1140–1142. 6 Roubenoff R. Sarcopenia: Effects on body composition and function. J Gerontol A Biol Sci Med Sci 2003;58:1012–1017. 7 Musaro A, Giacinti C, Borsellino G, et al: Stem cell-mediated muscle regeneration is enhanced by local isoform of insulin-like growth factor 1. Proc Natl Acad Sci USA 2004;101:1206–1210. 8 Herbst KL, Bhasin S: Testosterone action on skeletal muscle. Curr Opin Clin Nutr Metab Care 2004;7:271–277. 9 Oster MH, Enders SR, Samuels SJ, et al: Megestrol acetate in patients with AIDS and cachexia. Ann Intern Med 1994;121:400–408. 10 Loprinzi CL, Schaid DJ, Dose AM, Burnham et al: Body-composition changes in patients who gain weight while receiving megestrol acetate. J Clin Oncol 1993;11:152–154. 11 Urban RJ, Bodenburg YH, Gilkison C, et al: Testosterone administration to elderly men increases skeletal muscle strength and protein synthesis. Am J Physiol 1995;269:E820–E826. 12 Volpi E, Sheffield-Moore M, Rasmussen BB, Wolfe RR: Basal muscle amino acid kinetics and protein synthesis in healthy young and older men. JAMA 2001;286:1206–1212. 13 Boirie Y, Gachon P, Beaufrere B: Splanchnic and whole-body leucine kinetics in young and elderly men. Am J Clin Nutr 1997;65:489–495. 14 Wittert GA, Chapman IM, Haren MT, et al: Oral testosterone supplementation increases muscle and decreases fat mass in healthy elderly males with low-normal gonadal status. J Gerontol A Biol Sci Med Sci 2003;58:618–625. 15 Hajjar RR, Kaiser FE, Morley JE: Outcomes of long-term testosterone replacement in older hypogonadal males: A retrospective analysis. J Clin Endocrinol Metab 1997;82:3793–3796. 16 Wang C, Swedloff RS, Iranmanesh A, et al: Transdermal testosterone gel improves sexual function, mood, muscle strength, and body composition parameters in hypogonadal men. Testosterone Gel Study Group. J Clin Endocrinol Metab 2000;85:2839–2853. 17 Takala J, Ruokonen E, Webster NR, et al: Increased mortality with growth hormone treatment in critically ill adults. N Engl J Med 1999;341:785–792. 18 Schwenk A, Hodgson L, Wright A, et al: Nutrient partitioning during treatment of tuberculosis: Gain in body fat mass but not in protein mass. Am J Clin Nutr 2004;79:1006–1012. 19 Paton NI, Chua YK, Earnest A, Chee CB: Randomized controlled trial of nutritional supplementation in patients with newly diagnosed tuberculosis and wasting. Am J Clin Nutr 2004;80:460–465. 20 Slade GD, Ghezzi EM, Heiss G, et al: Relationship between periodontal disease and C-reactive protein among adults in the atherosclerosis risk in communities study. Arch Intern Med 2003;163:1172–1179.
29
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 31–43, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
Physiological vs. Pathological Changes of Nutritional Status over Life Time Manfred J. Müller, Anja Bosy-Westphal, Corinna Geisler and Simone Onur Institut für Humanernährung und Lebensmittelkunde, Christian-Albrechts-Universität zu Kiel, Kiel, Germany
Introduction Age-related changes in nutritional status include weight loss as well as weight gain. In the EURONUT-Seneca study on elderly subjects (age 75–80 years) who were investigated in different parts of Europe, the prevalence of a 5-year weight gain of ⬎5 kg was highly variable and reached a highest prevalence of 9 and 17% in females and males, respectively [1]. By contrast the prevalence of age-related weight loss (⬎5 kg in 5 years) reached 27 or 37% in females and males, respectively. Concomitantly, underweight (i.e. a body mass index (BMI of ⬍20 kg/m2) was seen in up to 24% of elderly subjects, whereas obesity was observed in up to 38% [1]. These data suggest that in the elderly there are two different clinically relevant manifestations of nutritional status. First, there is frequent weight loss and sarcopenia, which is explained by anorexia related to chronic diseases, depression or age-related changes in physiologic functions (e.g. impaired regulation of appetite). Weight loss and sarcopenia are associated with functional impairment, reduced muscle strength and immune function, and thus increased morbidity and mortality. Second, weight gain and an increase in fat mass (FM) are due to a sedentary life style. This manifestation is associated with a high prevalence of chronic metabolic diseases (e.g. non-insulin-dependent diabetes mellitus and metabolic syndrome). Age-related changes in nutritional status include alterations in body composition. Thus, FM as well as fat-free mass (FFM) and thus body cell mass all 31
Physiological vs. Pathological Changes
Drugs Physical activity
BCM
FM FFM Nutrition
Heritability
Body composition
MM/OM Hydration Disease Age
Fig. 1. Determinants of body composition. FM ⫽ Fat mass; FFM ⫽ fat-free mass; BCM ⫽ body cell mass; OM ⫽ organ masses.
change with age. There are a number of determinants of body composition including dietary intake, physical activity and inactivity, heritability, age, chronic diseases and drugs (fig. 1). The contribution of the different determinants differs between individuals and also between different age groups resulting in a heterogeneity of physiological and pathophysiological changes in nutritional status over a life time.
Methodological Issues Different methods can be used to characterize nutritional status. Faced with recent developments in body composition research, there is no gold standard in techniques used to assess different body compartments. In addition there is no unique parameter characterizing all aspects of body composition. Considering age-related changes in nutritional status, the different methods also differ with respect to accuracy and validity. Comparing BMI, bioelectrical impedance-derived FM and FM measured by densitometry (which is considered as one of the most suitable methods to assess fat mass) [2] in a group of 129 healthy women in Kiel (age range 18–84 years) the agerelated increase in FM can only be seen by the use of densitometry (fig. 2). These data suggest that considering age-related changes in body composition, one has not only to consider the changes in the ‘true’ phenomena but also the methodological limitations of the methods used to assess body composition. Using bioelectrical impedance analysis (BIA) in a greater group of subjects differing with respect to BMI and age, it becomes evident that BIA-derived FM increases with BMI. However, at a given BMI, FM decreases with age. The age-related differences in FM may reach 5%. BIA measures two variables, 32
Physiological vs. Pathological Changes BMI vs. age
40
FMBIA (%)
BMI (kg/m2)
50
30 20 10 0 0
20
40
60
Age (years)
%Fat massBIA vs. age
70
80
100
60
60
50
50
40 30
40 30
20
20
10
10
0
0 0
20
40 60 80 Age (years)
%Fat massADP vs. age
70
FMADP (%)
60
100
y ⫽0.2166x ⫹ 28.181 2 R ⫽ 0.15 0
20
40
60
80
100
Age (years)
Fig. 2. Age-related changes in body mass index (BMI), %fat massBIA and %fat massADP in 129 women in Kiel, Germany. BIA ⫽ Bioelectrical impedance analysis; ADP ⫽ air-displacement plethysmography, densitometry.
i.e. resistance (the detected current is weaker than the source current) and reactance (i.e. the detected current lags behind the source current). Standard BIA algorithms only use resistance but ignore reactance. Considering the BMI- and age-related changes in resistance (R) and reactance (Xc), R to height and Xc to height both decrease with BMI. However, at a given BMI there is no effect of age on R/height. By contrast at a given BMI, Xc to height decreases with age. It becomes evident that the R/Xc vector downslopes with age. Thus standard BIA algorithms cannot be used to address age-related changes in body composition. Age-related changes in Xc reflect changes in body cell mass and cannot be used as measures of FM or other obesity-related components of body composition. Taken together this method is also not suitable to measure interrelations between energy intake, output and stores. Besides the high quality of BIA measurements (i.e. its high precision), the accuracy (i.e. the level of agreement between the measured value and the true value) is low and differs between age groups.
Metabolic Impact of Age-Related Changes in Nutritional Status Age-related changes in BMI, FFM and FM are highly variable. Considering cross-sectional data, FFM and FM in kilograms both increase up to age 40–60 years and then decrease again. When compared to FM this decrease is more pronounced in FFM. Considering the age-related changes in FFM and FM as percent body weight, there is a continuous decrease in FFM with age. By contrast FM increases with age. Both changes are more pronounced in subjects ⬎60 years. FFM is the major determinant of resting energy expenditure (REE). Throughout the lifespan REE increases (up to age 40–60 years) and decreases 33
Physiological vs. Pathological Changes 16
18–39 years y⫽0.0968x ⫹ 1.5317 R2⫽0.60
14
REE (MJ/day)
12 40–59 years y ⫽ 0.0837x⫹ 2.0255 R2⫽ 0.67
10 8
60–69 years y ⫽0.0719x ⫹ 2.6956 R2⫽ 0.54 70–79 years y ⫽0.0653x ⫹ 2.9193 ⬎80 years R2⫽ 0.51 y⫽ 0.0426x ⫹3.6555 2 R ⫽ 0.25
6 4 2 0
20
40
60
80
100
120
140
FFM (kg)
Fig. 3. Age dependency of the resting energy expenditure (REE) vs. fat-free mass (FFM) relationship. From Müller et al. [8].
(⬎60 years). Adjusting REE for FFM and/or FFM ⫹ FM, most of the agerelated changes in REE disappear suggesting that age-related changes in metabolic rate are most likely explained by changes in body composition. There is a close association between FFM and REE (R2 between 0.50 and 0.80) [3–7]. Considering the association between REE and FFM in different age groups there is a progressive sloping of the REE–FFM relationship in adults of different ages [8]. These data also suggest that the ‘specific’ metabolic rate (i.e. REE/kg FFM) decreases with increasing age (fig. 3). This finding can be explained by two different factors. First, there may be an age-related decrease in specific metabolic rate. Second, the composition of FFM may change with age thus affecting metabolic rate. Anatomically as well as metabolically FFM is heterogeneous. About 50% of FFM can be explained by metabolically active tissues and organs. By contrast bone, extracellular mass and plasma volume have a very low or no metabolic activity (i.e. ⬍1 kcal/kg FFM ⫻ day but add to 50% of FFM. In a normal-weight man skeletal muscle mass adds to about 88% of metabolically active FFM. By contrast the brain, heart, liver, kidneys and other visceral organs reach only 12% of metabolically active FFM (FFMm). Since skeletal muscle has a specific metabolic activity of about 15 kcal/kg FFMm ⫻ day⫺1 but visceral organs reach a mean value of 468 kcal/kg FFMm ⫻ day⫺1, the composition of FFM (i.e. the ratio between organs with a low and organs with a high metabolic rate) adds to the metabolic variance between subjects [4]. 34
Physiological vs. Pathological Changes Table 1. In vivo body composition assessment in young and elderly subjects Females
FMDXA, kg Muscle massDXA, kg Bone massDXA, kg Organ massMRI, kg Brain, kg Heart, kg Liver, kg Kidneys, kg Residuals, kg
Males
young
elderly
young
elderly
19.2 ⫾ 5.7 20.7 ⫾ 2.7 8.6 ⫾ 1.1 3.8 ⫾ 0.3 1.5 ⫾ 0.1 0.28 ⫾ 0.0 1.5 ⫾ 0.2 0.30 ⫾ 0.0 10.3 ⫾ 1.8
27.6 ⫾ 5.4** 16.5 ⫾ 2.1* 7.2 ⫾ 0.9* 2.9 ⫾ 0.2* 1.1 ⫾ 0.1** 0.36 ⫾ 0.1** 1.1 ⫾ 0.1** 0.23 ⫾ 0.0* 11.4 ⫾ 3.4
12.9 ⫾ 4.3 33.6 ⫾ 3.3 11.4 ⫾ 2.1 4.3 ⫾ 0.5 1.6 ⫾ 0.2 0.37 ⫾ 0.0 1.7 ⫾ 0.3 0.39 ⫾ 0.1 18.0 ⫾ 3.5
19.6 ⫾ 5.1** 24.8 ⫾ 4.6* 9.8 ⫾ 1.7* 3.4 ⫾ 0.4* 1.2 ⫾ 0.1** 0.47 ⫾ 0.1** 1.2 ⫾ 0.3** 0.31 ⫾ 0.1* 18.6 ⫾ 4.4
According to Bosy-Westphal et al. [6]. *p ⬍ 0.05; **p ⬍ 0.01.
Using modern body composition assessment technologies, one can directly address this question in young and elderly subjects [4]. Skeletal muscle and bone mass can be measured by DEXA. In addition magnetic resonance imaging (MRI) can be used to take transversal images at different distances at a slice thickness of 6–8 mm with a technical error of 0.7–1.1%. Determination of cross-sectional organ areas can be done by hand or using a computer program. Calculation of organ volumes is performed from the sum of crosssectional areas ⫻ slice thickness. Organ mass is then calculated from organ volumes ⫻ organ densities as taken from the literature (e.g. 1.036 g/cm3 for brain and 1.06 g/cm3 for heart muscle). Table 1 shows detailed body composition data in young and elderly subjects [6]. Since REE is a function of body composition, it can be modeled from detailed body composition analysis: REEc (kJ/day) ⫽ 1,008 ⫻ brain ⫹ 840 ⫻ liver ⫹ 1,848 ⫻ heart ⫹ 1,848 ⫻ kidneys ⫹ 55 ⫻ muscle ⫹ 19 ⫻ FM ⫹ 1 ⫻ bone ⫹ 50 ⫻ residual mass. REEm (measured REE) was lower than REEc. When compared with young adults the difference was significantly greater in the elderly (0.60 vs. 0.11 MJ/day, p ⬍ 0.01) [6]. FFM showed a close association with the sum of organs, liver, kidney and spleen masses. Comparing the regression lines observed in young and elderly subjects, there were no differences in the slopes but organ masses at a given FFM were lower in the elderly (fig. 4). By contrast heart mass per FFM was higher in the elderly. In elderly subjects 60% of the variance in REEm–REEc was explained by heart mass. The difference between REEm and REEc between age groups disappeared after exclusion of elderly subjects with cardiac hypertrophy. Exclusion of 5 subjects with cardiac hypertrophy (i.e. a heart weight of ⬎500 g) resulted in a reduced REE prediction error (i.e. in 35
Physiological vs. Pathological Changes
Elderly subjects 0.8
y ⫽ 0.030x ⫹2.07 R2⫽0.63
5
y⫽0.007x⫹0.08 R2⫽0.54
0.7 0.6 Heart massMRI (kg)
OMMRI (brain⫹ liver⫹ kidneys ⫹spleen) (kg)
Young subjects 6
4 3 2
0.5 0.4 0.3 0.2
y ⫽0.024x ⫹ 1.56 R2⫽0.68
1
y ⫽0.004x⫹0.101 R2⫽0.69
0.1
0
0.0 0
50 FFMDXA (kg)
100
0
50 FFMDXA (kg)
100
Fig. 4. Fat-free mass (FFM) vs. organ or heart mass in young and elderly subjects. From Bosy-Westphal et al. [6].
the elderly the difference between REEm and REEc reached ⫺0.10 MJ/day which is close to the data observed in young subjects) [6]. The study has certain limitations. It was assumed that specific organ metabolic rates are constant with increasing age and increasing organ mass. In addition organ composition and density was assumed to be constant too. To summarize, the age-related decline in REE is mainly explained by a reduction in FFM as well as by alterations in FFM composition. Overestimation of REE in subjects with cardiac hypertrophy suggests a decrease in specific heart metabolic rate with increasing heart mass.
Heritability of Nutritional Status and Metabolic Function There are genetic effects on nutritional status and metabolism. Heritability describes the additive genetic effect. It can be estimated from data on twins or parent–child relationships [9]. Heritability is a population parameter reflecting the extent of the contribution of genes to the individual phenotype. Heritability does not address genotype–environmental interactions (i.e. differences in the sensitivity of individuals to environmental or lifestyle changes). Heritability adds to physiological and pathophysiological changes in nutritional status and metabolism during the lifespan. Preliminary analysis of data from an ongoing family path study in Kiel which included data sets from 53 families (grandparents, parents, children) suggests heritability estimates of 0.42, 0.44 and 0.47 36
Physiological vs. Pathological Changes Unexplained ~36% FFM
Age ~21%
Heritability ~43%
~50% REE
Age
Heritability
~4%
~21%
Fig. 5. Explained and unexplained variance in fat-free mass (FFM) and resting energy expenditure (REE). From Geisler et al., unpublished data.
~25% Unexplained
Nutritional Status Age-related changes in nutritional status Lifestyle/SES-related changes in nutritional status
~50–60%
Disease-related changes in nutritional status Heritability ~ 40–50% Metabolic function: Resting Energy Expenditure Age-related changes in metabolic function? Lifestyle/SES-related changes in metabolic function
~29%
Disease-related changes in metabolic function Heritability ~ 21% ⴙ Nutritional Status ~ 50% ⴝ ~ 71%
20
30
40
50
60
70
Age (years)
Fig. 6. Physiological vs. pathological changes in nutritional status over lifetime.
for BMI, FFM and waist circumference, respectively. Thus heritability of nutritional status is between 40 and 50%. This is close to the estimates of the Quebec family path study [10]. By contrast 50–60% of the age-related changes in nutritional status are due to the effects of age itself, lifestyle and disease-related factors. These factors also include possible gene–environment interactions. Regarding metabolism the heritability of REE was 0.53. After adjustment of REE for FFM, heritability decreased to 0.42. These results suggest that 42% of the individual differences in REE, after accounting for the influence of body composition are genotype-dependent. Since both FFM (i.e. the major determinant of REE) and REE adjusted for FFM have heritabilities of about 43 and 21%, respectively, there is a considerable genetic effect on metabolic rate. These 37
Physiological vs. Pathological Changes data are again close to previous estimates [10]. By contrast there is only a minor effect of age on REE (about 4% of its variance is explained by age). However, age explained 21% of the variance in FFM. Taken together these data suggest that the age-related changes in REE are more likely due to age-related changes in nutritional status. Figure 5 shows the different estimates of our calculation. These data also suggest that heritability estimates for nutritional status and metabolism differ. When compared with REE the heritability of nutritional status is higher (i.e. 40–50 vs. about 20%) suggesting that lifestyle- and diseaserelated factors add more to the variance in metabolism than to the variance in nutritional status. It is evident that heritability adds to physiological and pathophysiological changes of nutritional status and metabolism (fig. 6). Summary To summarize, body composition methods have to be selected carefully to assess age-related changes in nutritional status. Age-related changes in metabolism are likely explained by age-related changes in nutritional status and diseases. There is a certain heritability in nutritional status and metabolism (about 40–50 and 20%, respectively) which tracks throughout life. Acknowledgement Supported by DFG Mü 8–1; Precon, Bickenbach; BMBF ‘Nahrungsfette und Genvariabilität’.
References 1 SENECA: Nutrition and the elderly in Europe. Follow-up study and longitudinal analysis. Eur J Clin Nutr 1996;50(suppl 2):S1–S127. 2 Pierson N Jr, Wang J, Thornton JC: Body composition comes of age: A modest proposal for the next generation. The new reference man. Ann NY Acad Sci 2000;904:1–11. 3 Wang Z, Heshka S, Gallagher D, et al: Resting energy expenditure-fat-free mass relationship: New insights provided by body composition modeling. Am J Physiol 2000;279,E539–E545. 4 Müller MJ, Bosy-Westphal A, Kutzner D, Heller M: Metabolically active components of fatfree mass and resting energy expenditure in humans: Recent lessons from imaging technologies. Obes Rev 2002;3:113–122. 5 Illner K, Brinkmann G, Heller M, et al: Metabolically active components of fat-free mass and resting energy expenditure in humans. Am J Physiol Endocrinol Metab 2000;278:E308–E315. 6 Bosy-Westphal A, Eichhorn C, Kutzner D, et al: The age-related decline in resting energy expenditure in humans is due to the loss of fat-free mass and to alterations in its metabolically active components. J Nutr 2003;133:2356–2362. 7 Bosy-Westphal A, Reinecke U, Schlörke T, et al: Effect of organ and tissue masses on resting energy expenditure in underweight, normal weight and obese adults. Int J Obes 2004;28:72–79. 8 Müller MJ, Bosy-Westphal A, Klaus S, et al: WHO equations have shortcomings to predict resting energy expenditure in individuals from a modern affluent population – Generation of a new reference standard from a retrospective analysis of an actual German database of resting energy expenditure. Am J Clin Nutr, 2004.
38
Physiological vs. Pathological Changes 9 Bouchard C, Tremblay A, Nadeau A, et al: Genetic effect in resting and exercise metabolic rates. Metabolism 1989;38:364–370. 10 Bouchard C, Deriaz O, Perusse L, Tremblay A: Genetics of energy expenditure in humans; in Bouchard C (ed): The Genetics of Obesity. Boca Raton, CRC Press, 1994, pp 135–146.
Discussion Dr. Elia: Thank you very much for asking me to comment on the organ-specific constancy that you were using in your studies. These were largely derived from studies of arteriovenous oxygen differences in younger groups of people. If one is to use those and extrapolate to elderly subjects there is potential for error. For example, I actually don’t exactly know what a healthy old person is because disabilities increase with age, affecting large segments of the ‘normal’ population. You mentioned blood pressure and ischemic heart disease were present in one of your studies. It could be that the values obtained from younger healthy people may be different from those in the elderly population who had such problems. In addition there are some fluid changes in older people, which could confound some of the interpretations. But this age-specific effect also has relevance to other data that you showed, such as bioelectrical impedance measurements. These are largely empirically based relationships and it is important to know whether they apply to other age groups. Other studies using reference body composition techniques show an increase in percentage fat with age, even after adjusting for body mass index (BMI). The question I want to ask is about the heritability of BMI because if one looks at the literature the estimates of heritability have ranged from somewhere between 10 and 90%, with quite a few around the middle range like the one you reported. Could you comment as to the reasons for this large variability and the type of models used to establish these heritability factors? Dr. Müller: Heritability estimates vary between the different types of studies. There are two types of study that have been used in the literature to calculate the heritability of parameters of nutritional status and also parameters of metabolic function [1]. The family path study is one possible model for calculations like this on the heritability of phenotype between generations. If DNA probes are available one can also estimate the heritability of the genotype of interest. When compared with heritability obtained in the family studies the highest heritability is reached using twins, monozygotic and dizygotic twins, and the highest numbers are reached when dizygotic twins are used to calculate the heritability between subjects. This is not a big surprise when you look at twins and you are totally convinced that there must be a high heritability in the parameters. With regard to family studies, I am only aware of 3 family studies which are cited in the area of body weight and body composition. These studies are also used as a basis for all the molecular studies which we are faced with now. I am not very sure that these 2 or 3 family studies can really be used as a reference standard since heritability varies for BMI for instance, so one has to be careful. In the case of the Quebec study by Bouchard [2], which is most frequently cited, he only calculated the parent–child relationship which is different to what I have shown where we have grandparents, parents and children. So we have one generation more, and it is no surprise that our data may differ to some extent from the data of the Quebec study [2]. On the one hand there is a clear need to bring these studies together and on the other hand to be careful with all these analyses. We have just ended heritability assessments with DNA probes on all these subjects which is part of the human genome on the metabolic syndrome. The data we found here are within the magnitudes which have been reported. Perhaps they can be used to differentiate
39
Physiological vs. Pathological Changes between nutritional status-dependent changes and nutritional status-independent changes in the variables of interest here; perhaps we can get an idea from these data. Dr. Oltersdorf: I know that there should be many more longitudinal nutritional studies. I know that there are still the data on the Framingham offspring [3]. So it is longitudinal. Has this generation study not been done? I am surprised. Dr. Müller: I have never seen heritability data or the longitudinal data of the Framingham study. I am not aware of this. Dr. Oltersdorf: Nobody knows about it because they select many things, and so it should be easy to analyze the generation. Dr. Steinhagen-Thiessen: I think the Baltimore longitudinal study [4] has those data. Dr. Müller: But I think they don’t have body composition data, they only have BMI data. Dr. Steinhagen-Thiessen: I am not sure because there was a time where they were very much focused on water and fat consumption and so on. Dr. Morley: The Baltimore longitudinal study from Saint Louis does have data but it is based on skin-fold thickness, the data were published perhaps 10–15 years ago [5]. The problem always with the Baltimore longitudinal data is that very little of what they published is longitudinal so they often published cross-sectional and a little bit of longitudinal buried in the cross-sectional. It was in the Journal of Gerontology I think about 10–15 years ago, and as I said, it was based skin folds not on any of the more modern techniques. Dr. Lochs: When I look at your data on age, it seems that this is a rather uniform behavior and that there are not 2 groups which behave differently from one another. Do you think it is correct then to conclude that these are physiological changes because if there were a sick group and a healthy group then you would expect that these curves would grow apart with increasing age. Is that correct or not? Dr. Müller: Let me go back to Dr. Elia’s question. These are self-reports of health, and if you assume something like a constant factor, e.g. heritability, for example the probability of a silent disease in all subjects increases with age. So then these are not really health-related changes. What we are seeing is a whole picture. I have no idea whether the data can be divided into subgroups. So what I see here is a more spontaneous situation in a heterogeneous group of subjects who are healthy with respect to their self-reported health, but there may be a contribution of certain diseases in some people. But you are right, there is a big variance in the data. Dr. Lochs: But still it looks quite uniform. It does not look as though it is divided up into groups. But of course you took only healthy people or subjectively healthy people, so it would be nice to see how nursing home subjects change or what their nutritional status is. Dr. Müller: The decrease in the second part of life is accelerated in this group; this is what we would expect in this group. Dr. Oltersdorf: Perhaps you can explain a little bit about how you got the marvelous database 1 as Precon data? What kind of people are they? Dr. Müller: Precon is a company involved in body weight reduction programs for overweight and obese people. This part of our study was in cooperation with this company which has a huge database in Germany of more than 100,000 impedance measurements. They gave us the data because we have the so-called German Reference Center of the Body Composition Research here. They didn’t know what to do with all these data and asked us if we could find anything. We tried to systematically analyze the data, and found that there are true limits to this analysis because these are field measurements performed on numerous subjects. Dr. Oltersdorf: Do you only know the age of these people? Dr. Müller: We have age, weight, height, sex and the impedance data.
40
Physiological vs. Pathological Changes Dr. Labadarios: A very valid and important point you made regarding methodology and its limitations or advantages. Could I please have your thoughts on your approach to the criteria you use in accepting or ignoring data when you read publications on bioelectrical impedance? Dr. Müller: There are numerous problems with impedance measurements, with the method itself, and the major problem is with the algorithm you use to calculate variables like fat-free mass or fat mass. Our personal decision in my group was to use raw data instead of the calculated data. When you start with impedance measurements, you need a reference standard and a reference population that fits your population of interest. There is some discussion about a suitable reference standard for impedance measurements. I think at present there is no real gold standard, but we are all sure that the four-compartment model, which uses different methods to combine these variables, is the most robust model we have for body composition research. For me, if you don’t have the four-compartment model, the best way is to speak about reference method-specific impedance numbers. So if you use impedance and your reference method is DEXA you have fat mass measured by impedance based on an algorithm generated by DEXA. You also have to say what your reference population is. In Germany when I work with children, the problem is that for instance in Jena, in the eastern part of Germany, I cannot use an algorithm that was generated in Kiel. So if we try to apply these algorithms or these formulas generated from measurements of densitometry, and we both (Kiel and Jena) use densitometry as a reference, we come up with different algorithms. So there is an absolute need to generate population-appropriate algorithms and to work with specific numbers. The results are not absolute numbers, i.e. this is not absolute fat mass, you don’t reach absolute fat mass. I think we have no gold standards, we have specific DEXA values, we have specific densitometry values, we have isotope dilution-specific values. They are close to each other, no question about that, but they are not really true numbers. Dr. Bozzetti: As far as I know from all the clinical studies made by Roza and Shizgal [1] 20 years ago, the resting energy expenditure to body cell mass ratio measured by total exchangeable potassium usually remains constant during different intervals of age. Is this still correct or not? Dr. Müller: It cannot be correct, when you look at the regression line between fat-free mass or body cell mass and resting energy expenditure, it doesn’t cross the zero point. It is slightly higher, at about 1, 2, 3 J/day. So this association causes some doubts in simply dividing resting energy expenditure by body cell mass or fat-free mass. The problem can be addressed by regression procedures and Roza and Shizgal [1] did not do this at that time. So even in the case of healthy subjects and also in patients, the simple ratio of resting energy expenditure to body cell mass or resting energy expenditure to fat-free mass ratio does not allow a comparison of different groups with respect to energy expenditure. If you look at your own patients with cancer cachexia, you cannot compare this group with another group of normal weight subjects with respect to the ratio of resting energy expenditure to fat-free mass. Dr. Elia: One of the potential variables here is the same issue that has been raised with fat-free mass. Body cell mass is a mixture of low energy expenditure muscle tissue and high energy expenditure organs; so if there is an alteration in the proportion of these such that the total body cell mass remains the same you will still get some differences in energy expenditure. Body cell mass is a summary measure of all the tissues with widely different metabolic rates and this can be disturbed in disease and possibly with aging as well. Dr. Ockenga: Just another topic, this project is part of the worldwide genome project and the aim of this project is to get information about the genotype–phenotype
41
Physiological vs. Pathological Changes association especially in diseases. Do you think that a further analysis of your data would give us data which would identify patients at risk for obesity as well as patients at risk for malnutrition under special clinical circumstances? Dr. Müller: This is what molecular biologists have been dreaming about for many years. Faced with the heterogeneity phenomena in this area. I am not very sure that they will come to a positive end with that kind of research. We are mainly interested in fatty acid binding proteins and fatty acid transport proteins. We look for genetic variances between subjects and between generations, and based on the data of these 53 families and these datasets and we are going to include 200 families. This is the upper level that we can reach in this study. So I am not very optimistic about this kind of study, we have to look further for the truth. I think that the influence of lifestyle and socioeconomic status is much greater. So if we are interested in the prevention of metabolic syndrome, overweight and obesity, we should look for socioeconomic factors and think about both education and improvement in the economic situation of the people, which is much more effective than selective prevention in subgroups characterized by different genetic makeup. This is not the way we can really solve the problem of overweight endemia in our society. Dr. Thomas: You have commented on population variability and the software analysis regression equations. Can you comment on the reliability of repeated measures of bioelectrical impedance in the same individual? What is your standard of accuracy between two separate measurements with different time variance? Dr. Müller: I consider the inter-individual variance as very low. It is about 2–3% within a subject if you measure him twice within let’s say 1 h. At our institute, in this setting, the intra-individual variance for resting energy expenditure is lower, about 2–5%. But if you ask the subject to come back a week later, his variance is higher. If you have him come back many times over a period of a year, you get something like that. It differs if he comes with a bicycle or on foot. There are a number of things which have an effect on the inter-individual variance of these data. So for the field situation there are problems with respect to the interpretation of difference between groups or within subjects over a long time period. We try to standardize the measurement condition by letting the subject lay down for at least 15 min to measure him and we try to take into account a number of factors which influence the measurement. But I am not sure whether we can really address all these factors which add to the variance. Dr. Morley: First of all in answer to Dr. Lochs who wanted to know the difference between a nursing home population and a general population. This was studied many years ago by a methodology that I think is somewhat suspect, but the differences were enormous. The generally healthy elderly are very different from the population in a nursing home. I think that is what we can say as far as safety of the numbers. The question I really have, which has driven me crazy, is how should you measure body composition if you actually want to do it in a cross-sectional or longitudinal study? We know, for instance with DEXA, that there are real problems with measuring visceral fat as some studies show that if you put fat on the stomach you underestimate it dramatically in those circumstances. Even if you go to something like stable isotope studies there is sometimes a 10–15% variance in the methodology which makes it very difficult if you really want to look at smaller numbers which is necessary. For instance the genome projects have outgrown our ability to do something that we think is very simple, which is to measure body composition, which may be a lot harder than measuring parts of the genome. Dr. Müller: If you are not a specialist in this field, I would advise you to start with one method, to try to standardize the method as well as you can in your hands and work with specific numbers which you can compare within your population. It is quite
42
Physiological vs. Pathological Changes correct to use anthropometric data, you can use impedance data, you can use DEXA data. The accuracy of DEXA is above body impedance analysis. But even if you have DEXA or densitometry data you have a lot of limitations with their interpretation which are due to the algorithms you use which for instance cannot be used in all age groups in the same way. So if you go on and on, you have to make a number of assumptions. If you use the Heymsfield [6] data for instance: in one of his studies on organ mass and the modeling of resting energy expenditure, he had at least 26 assumptions in his algorithms, which have not been proven at all. There is quite clearly some plausibility, but there are a lot of assumptions and this is something for the specialist, not the greater audience. So use one method which can be applied in your setting, try to standardize it and work with it over the years. Dr. Morley: At present in the United States it is very difficult to get anything funded if you are not also to get CT or MRI data. How do you feel that interferes with each of the different methods? It is very clear that a 70-year-old has much more fat in his muscle than a 35-year-old and that is one of the problems we are now dealing with in addition to the rehydration status. Dr. Müller: I cannot really imagine a context where you need CT or MRI measurements of body composition in a clinical setting or a field study. If you are interested in age-related changes, not specifically in changes in visceral fat mass or bone mass for instance, you can work with a standard method. I don’t see a need for these imaging technologies in this context. If you have specific patients or specific questions (e.g. regarding visceral fat mass) there is some need for these imaging technologies, but I feel that even in this case DEXA is enough today. You don’t need CT for visceral fat mass and you don’t need MRI for the calculation or measurement of muscle and/or organ mass because these are really specific questions.
References 1 Roza AM, Shizgal HM: The Harris Benedict equation reevaluated: Resting energy requirements and the body cell mass. Am J Clin Nutr 1984;40:168–182. 2 Bouchard C: Heredity and the path to overweight and obesity. Med Sci Sports Exerc 1991;23:285–291. 3 Hood MY, Moore LL, Sundarajan-Ramamurti A, et al: Parental eating attitudes and the development of obesity in children. The Framingham Children’s Study. Int J Obes Relat Metab Disord 2000;24:1319–1325. 4 Hirsch R, Lethbridge-Cejku M, Hanson R, et al: Familial aggregation of osteoarthritis: Data from the Baltimore Longitudinal Study on Aging. Arthritis Rheum 1998;41:1227–1232. 5 Hallfrisch J, Muller D, Drinkwater D, et al: Continuing diet trends in men: The Baltimore Longitudinal Study of Aging (1961–1987). J Gerontol 1990;45:M186–M191. 6 Heymsfield SB, Gallagher D, Kotler DP, et al: Body-size dependence of resting energy expenditure can be attributed to nonenergetic homogeneity of fat-free mass. Am J Physiol Endocrinol Metab 2002;282:E132–E138.
43
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 45–58, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
Home Enteral Nutrition Demographics and Utilization in the United States
Mark H. DeLegge Medical University of South Carolina, Charleston, S.C., USA
Enteral nutrition is the act of receiving nutrients through the gut, either orally or through an enteral access device. It has long been used in the hospital and nursing home setting. The number of patients receiving home enteral nutrition (HEN) in the United States has progressively increased, yet the absolute numbers remain difficult to determine. The use of enteral nutrition has a long history. In 1790, Hunter was first reported to use a mixture of jellies, milk, eggs, sugar and wine through a whale bone covered with eel skin attached to a bladder pump [1]. Einhorn [2] in 1910 provided medical practitioners with guidelines for gastric and intestinal enteral feedings. In the 1950s, attention was placed on the development of enteral formulas. Pareira et al. [3] and Meade Johnson (Evansville, Ind., USA) developed an early enteral formula consisting of milk, milk solids, calcium caseinate, dextrose, maltodextrose, vitamins and minerals. At the same time at Henry Ford Hospital, Barron et al. [4] reported on the use of tube feedings made in the kitchen by blenderizing and straining table food. In the late 1950s and early 1960s a large study on the use of an elemental-based diet was conducted by the Vivonex Corporation (Mountain View, Calif., USA) and the National Institutes of Health. These diets were shown to provide adequate nutrition and maintain a patient’s well-being [5]. The last 3 decades have shown an explosion in the enteral formula industry with the development of many enteral products, some disease-specific. This has provided some specificity in a clinician’s ability to use enteral nutrition as a disease-specific tool, but has complicated the home care industry by the sheer volume of enteral formula choices available to patients and their providers. HEN has become a growing segment of the total home care arena in the United States. In 1992 it was estimated that there were approximately 73,000 HEN patients with an average yearly expenditure of USD 136 million [6]. Today’s expenditure estimates are very difficult to obtain, not only from the 45
Home Enteral Nutrition Table 1. Common home enteral nutrition diagnoses
Neoplasm Swallowing disorders Motility disorders Crohn’s disease Congenital bowel defects AIDS
Medicare population, but also for the Medicaid and private insurer populations. The general consensus is that the HEN population continues to grow at a rapid rate both in numbers and in dollars spent. There are many common patient disease groups that make up the majority of the HEN population including patients with neurological dysfunction, upper gastrointestinal cancers, anorexia and failure to thrive. These patients, in general, have difficulty with transfer of food from the oral cavity to the stomach. A review of the available data from 1987 to 1991 gives an excellent sample of the types of patients who were sent home on enteral nutrition [7] (table 1). In the late 1980s and early 1990s, a review of the outcomes of these HEN patients noted that at 1 year between 45 and 60% had died. The overall actual number of patients who are able to go back to oral nutrition ranged between 19 and 30%. In the neurologic-based disease patient group at 1 year, approximately 25% of the patients continued on HEN. The percentage of patients with cancer continuing on HEN therapy at the end of 1 year was much less. Further review of the Medicare HEN patient population data noted that neuromuscular-based disease patients had approximately 1.2 complications/year, with the majority of these complications related to their primary disease. Those patients with cancer had approximately 3.1 complications/ year, again the majority of these were related to their overall primary cancer process. This helped establish the safety of HEN. More recent data have been collected by Coram Healthcare, Denver, Colo., USA. 17,014 patients were cared for between 1998 and 2002. The mean age was 46.6 years. Thirty-six percent of the patients were more than 65 years of age and 28% of patients were less than 16 years of age. Fifty-four percent of patients were male and 44% were female. The top 5 diagnostic ICD-9 codes were: gastrointestinal dysfunction; protein calorie malnutrition; metabolic or developmental symptoms; intestinal malabsorption, and diseases of the esophagus. The average length of therapy was 357 days. Nineteen percent of patients were receiving concurrent antibiotic therapy and 3% concurrent pain management. The suspected feeding tube infection rate was 5.7% and feeding tube occlusion rate was 6.5%. Rehospitalizations occurred in 9.5% of patients within 30 days of hospital discharge, although the vast 46
Home Enteral Nutrition majority of these admissions were for the patient’s primary disease process, not the HEN. The quality of life of the HEN patient can be difficult to obtain, often because this patient population can be very difficult to assess. A study of 83 HEN patients used a rough quality of life tool [8]. The patients were assessed at initiation of their HEN and at 1 year, documenting if the patient was better, the same, or worse than they were before starting HEN. Approximately 30% of the original population was available for re-assessment at 1 year, the remaining HEN patients having died. At 1 year, 40% of these patients felt they were better, 40% had no change, and 10% felt worse. Another study examined 38 adult HEN patients who had been on therapy for 25 months. Most of them had neurologic disease, head and neck cancers or a decreased level of consciousness [9]. Quality of life was measured using the SF-36. Also, the European Quality of Life scoring system and an EQID visual analog scale were used to determine quality of life in these same patients. If the patients could not answer questions because of a decreased level of consciousness, the family was used to measure a subjective quality of life. Some of the patients were completely independent for all their HEN needs, while others had someone at home caring for and helping them. A number of patients had more than 1 person who was participating in their care at home. These HEN patients spent approximately 1.9% of their time in the hospital, or approximately two of every 100 days of therapy. Of those hospital days, 50% were based on a HEN complication, and 50% were based on the patient’s overall primary disease process. Quality of life scores measured in these HEN patients were low as compared to a normal control group, most likely because of the significant medical disabilities in the patients receiving HEN. When the quality of life tools were applied to the families of HEN patients who could not respond to quality of life questions, the majority of the families thought the patients were better on therapy than at initiation of therapy. Interestingly, patient quality of life was significantly improved if the patient’s age was less than 45 years and if they had one or more caregivers available at home to assist with the HEN therapy. Additional outcome studies of HEN patients are small in number. The main questions are: does HEN result in a beneficial outcome for patients, and what are the therapy-associated complications? Globally, 18–44% of adults on HEN are ultimately able to gain weight [10]. Only 14% of older adults achieve partial or complete rehabilitation from their primary medical problem compared to 55% of younger patients [10]. The goal of transitioning adults to full or partial oral nutrition occurs in 10–30% of patients [10]. A study by Wilcock et al. [11] focused on HEN complications in 19 patients over 1 year, a very small number. Enteral stomal infections frequently resulted in a visit to the physician or the healthcare worker. The same was true for the onset of diarrhea. Flatulence, although a common complaint, rarely resulted in a physician visit. They reported 14 incidents where patients couldn’t obtain their tube feeding 47
Home Enteral Nutrition from their home provider: 2 cases due to equipment failure, and 10 cases due to obstruction of the feeding tube. Only one of those problems resulted in a visit to a healthcare provider. Schattner et al. [12] reported on HEN use in patients with dysphagia. This was a retrospective review over the course of 8 years. The patient’s nutrition therapy was followed by a nutrition support team, who also monitored their outcomes. There were 82 patients with a mean age of 61 years. Most of them had head and neck cancer. Most of the patients were at home on gastric feedings, the minority of them were home on small-bowel feedings. Diarrhea occurred in approximately 20% of the whole group. Tube site stomal infection or stomal irritation and leakage occurred in 6–7% of the patient group. A separate study was published on 416 adult patients with multiple primary diseases and an age of 65 years or older [13]. Their mean home enteral feeding duration was 242 days. The 1-month, 1-year, and 5-year mortality was monitored. Death was associated with the patient’s primary disease such as dementia, neurologic disease, head and neck cancer, AIDS, or an age of ⬎70 years. At 1 year, 55% of the patients had died. At 5 years of HEN, 75% of the patients had died. The overall long-term survival of patients on HEN demonstrated a rapidly diminishing group with the passage of time. Only 13% of the patients were transitioned back to oral nutrition. This underscores the fact that the likelihood of getting someone off tube feeding once they’re sent home, is quite low. Another study focused on 14 patients with cystic fibrosis who put their own nasogastric (NG) tube in each evening [14]. They received approximately 1,200 cal/day and were followed for 14.5 months. This high-risk group of patients was able to gain weight. Their lung function also improved as their nutritional status improved. Sixty-four percent of this group developed problems with hyperglycemia. The other major complication was nasal pharyngitis, a complication that would be expected from nightly placement of an NG tube. In a preoperative head and neck cancer study, preoperative enteral nutrition support was given to a group of patients with weight loss [15]. Forty-six patients received an NG tube and 43 patients received a percutaneous endoscopic gastrostomy (PEG). All received HEN training at home. Their pretreatment weight loss was approximately 12%. On follow-up, there were no significant HEN complications. The NG tube-fed patients only had a 15% compliance rate with NG feedings. Those who were fed with a PEG had a 68% compliance with their home tube feeding regime. Because of nutrition compliance issues, the PEG-fed patients had a 30% reduction in their postoperative hospital stay as compared to the NG-fed patients. In a separate study, 39 patients with head and neck cancer had a PEG placed in the hospital for nutritional support [16]. Ten patients died before they left the hospital. The overall median survival was 176 days. An examination of hospital readmissions after discharge revealed that approximately one 48
Home Enteral Nutrition Table 2. Outcome studies in pediatric HEN patients Study
Patient populations
Outcome
Kang et al. [19], 1998
Low normal weight Malnourished Stunted Cystic fibrosis
No weight change Weight gain Weight gain Weight gain Improved lung function
Steinkampf et al. [16], 1994
third of the patients had no readmission, one third had 1 hospital readmission and another third came back for 2 or more hospital readmissions, generally related to their head and neck cancer, occasionally related to their nutritional difficulties. Outcome studies for HEN in the pediatric population exist, although they are few in number (table 2) [14, 17]. In a quality of life evaluation of HEN in children, a questionnaire was administered to 70 families of pediatric patients with HEN using an NG tube [18]. These patients received enteral nutrition for 11,000 days. Both the parent and the child were asked how their feeding tube was working and if the tube feeding was well tolerated. One hundred percent of the parents said that all was going well. Fifty percent of the children stated that all was going well with the nasoenteric tube and tube feeding while the other 50% were unsatisfied with the nasoenteric tube and tube feeding. Interestingly, 10 of 70 of the children had a nocturnal cough only when they were tube fed, raising a concern for tube feeding regurgitation and aspiration. The older children in this study were self-conscious about going to school with a nasoenteric tube, thus opting for nighttime NG tube placement and overnight feedings. Sleep disturbance occurred in parents and children on home overnight NG tube feeds. It was recorded that approximately 60% of the parents were up during the night, often not because the child had a problem, but because the parents wanted to avoid complications. The impact of overnight tube feedings may have a significant effect on the quality of life from the perspective of the child and the parent. This always needs to be taken into consideration. The management of HEN requires decisions that are made early on by clinicians regarding enteral access, enteral formula, route of formula delivery, monitoring, complication management, and appropriateness for therapy. Careful attention must be given to each of these decisions in order to prevent complications and to ensure favorable patient outcomes. There are a number of enteral access devices available for home enteral feedings [19] (table 3). A low profile gastric device, or skin level device, is more esthetic than a PEG and is used commonly in the pediatric population (fig. 1) However, it requires the use of a specialized access tube to feed the 49
Home Enteral Nutrition Table 3. Enteral access options
Gastric Naso/orogastric PEG, PRG or surgical gastrostomy Low profile device (skin level) Small bowel Naso/oro jejunal PEG/J or PRG/J DPEJ or surgical jejunostomy
Time of use
Problems
Advantages
⬍30 days
Poor patient tolerance Wound infection
Economical
⬎30 days
Need connecting tubing to access device
Cosmetic
⬍30 days
Poor patient tolerance J-tube migration and obstruction Dislodgement and obstruction
Economical
⬎30 day
⬎30 days ⬍6 months ⬎6 months
Reliable
Gastric and jejunal access Reliable
PEG ⫽ Percutaneous endoscopic gastrostomy; PRG ⫽ percutaneous radiologic gastrostomy; PEG/J ⫽ percutaneous endoscopic gastro/jejunostomy; PRGJ ⫽ percutaneous radiologic gastro/jejunostomy; DPEJ ⫽ direct percutaneous endoscopic jejunostomy.
Fig. 1. Low profile gastrostomy device.
patient. Many of the replacement gastrostomy and jejunostomy tubes are balloon-type devices, similar to a Foley catheter, easily replaced at the bedside in the home (fig. 2). However, they have their own related problems including balloon deflation and tube dislodgement. A thorough knowledge of enteral access devices is imperative to be able to understand the problems encountered by HEN patients. 50
Home Enteral Nutrition
Fig. 2. Balloon gastrostomy replacement tube.
The choice of enteral formula is very patient-specific [20]. The clinician may choose either blenderized, standard, fiber-supplemented, elemental or a specialty formulation. Decisions regarding the appropriate use of a highnitrogen or a high-calorie formula are often determined by a patient’s protein, calorie and volume needs. Electrolyte, mineral and water content also varies between formulas. A clear understanding of the various enteral formula available and their specific contents will allow the clinician to choose the most cost-effective and safest formula. Delivery methods for HEN are another critical decision for the clinician. This decision would include whether to use bolus feedings, gravity feedings or pump feedings. The range of home enteral pumps varies tremendously (table 4). Each modality has its advantages and disadvantages for the patient, including time commitment, portability, expense and prevention of problems such as aspiration and diarrhea. In general, mobile patients are usually bolus fed and bed bound, or immobile patients are often fed with the use of a pump. These decisions are often based on lifestyle decisions by both the patient and their caregiver. Monitoring guidelines for HEN patients remain poorly described. Longterm monitoring is required in patients alive at 1 year, approximately 30–35% of patients who originally were started on HEN therapy. These patients generally have significant comorbid diseases. We should be concerned about the patient’s weight fluctuations over time. Daily fluid intake and output, and calorie and protein intake are also important. Laboratory analyses that should 51
Home Enteral Nutrition Table 4. Home enteral pumps Flow rates cm3/h Kendall (Mansfield, Mass.) Kangaroo 22
Battery life, h
Special feature
5–295
3
Cigarette lighter adapter
Kangaroo 224/324 Kangaroo Entriflush Kangaroo Pet 2100 Ross (Columbus, Ohio) Flexiflo III Flexiflo Companion Flexiflo Quantam
5–300 1–300 1–400
24 8 14
1–300 5–300 1–300
8 8 8
Flexiflo Patrol Zevex (Salt Lake City, Utah) Enteralite EZ Enteral Pump IVAC (San Diego, Calif.) Keofeed II Abbott (Chicago, Ill.) Breeze Lifecare 175 Elan Pharma (Dublin, Ireland) KM60,70,80,85
1–300
3
1–600 1–295
24 6
1–300
8
Quick recharge
1–999
5
Quick recharge
5–2,000
6.5
Automatic flush Ambulatory Ambulatory Automatic water infusion Pediatrics Ambulatory
be considered include albumin, serum electrolytes, glucose, magnesium, phosphorous, calcium, blood urea nitrogen and creatinine. Vitamin or mineral deficiency is rare unless the patient has other significant comorbid disease processes contributing to these deficiencies. Tolerance to enteral tube feedings should be documented. This includes problems with diarrhea, nausea and vomiting, abdominal bloating and cramping. Although more than 70% of the complications discussed here can be resolved in the home, 25% require visits to or by the physician and an additional 5% hospital admission [21, 22]. Other problems related to HEN use in the United States are secondary to its classification as durable medical equipment by Medicare. Enteral nutrition components fall into the same category as wheelchairs or walkers. Often the enteral formulas, enteral delivery bags and enteral tubing are sent to the patient’s home; this being the only contact between the patient and the home care provider. Nursing is only involved if the patient requires other skilled nursing services. This may lead to some difficulties in attempting to manage and monitor complications of HEN patients. From the clinician’s viewpoint, monitoring HEN is not a reimbursable activity in the United States. In addition, reviewing and signing the required insurance documentation for HEN is also a non-reimbursed activity. The care of HEN patients can be very labor intensive. Many physicians view 52
Home Enteral Nutrition HEN as not routinely requiring monitoring, compared to home parenteral nutrition patients. Although this may occasionally be true, most long-term HEN patients are debilitated and require routine clinical monitoring. Also, many clinicians may be uncomfortable with treating the complications of HEN because they are not familiar with enteral access devices or enteral formulations. In conclusion, the HEN population is a growing. Substantial nutritional benefit has been documented in appropriate patient populations. The HEN outcome data are very often consistent with the patient’s primary disease process. Nutrition is not a cure for their primary disease process. Reported complication rates vary tremendously. A standardized approach to reporting complications of HEN patients needs to be developed. Long-term HEN patients do exist, although they are in the minority of the patients started on enteral nutrition in the hospital. HEN patient monitoring is often not perceived to be as important as with home parenteral nutrition patients. The HEN patient population is a high-risk group that often requires careful monitoring. There are multiple medical decisions to be made by the clinician prescribing HEN. These decisions often have a tremendous impact on the patient’s outcome. Only with adequate oversight and rigorous patient policies and procedures can satisfactory HEN outcomes be obtained. References 1 Harkness L: The history of enteral nutrition therapy: From raw eggs and nasal tubes to purified amino acids and early postoperative jejunal delivery. J Am Diet Assoc 2002;102:399–404. 2 Einhorn M: Duodenal alimentation. Med Rec 1910;78:92–94. 3 Pareira MD, Conrad EJ, Hicks W, et al: Therapeutic nutrition with tube feeding. JAMA 1954; 156:810–816. 4 Barron J, Predergast JJ, Jocz MW: Food pump: New approach to tube feeding. JAMA 1956; 161:621–622. 5 Butler FS: Precision liquid diet helps physically, mentally ill. Hosp Topics 1967;45:60–61. 6 Howard L, Ament M, Fleming CR, et al: Current use and clinical outcomes of home parenteral and enteral therapies in the United States. Gastroenterology 1995;109:355–365. 7 Howard L, Heaphey L, Fleming CR, et al: Four years of North American registry home parenteral nutrition outcome data and their implications for patient management. JPEN J Parenter Enteral Nutr 1991;15:384–393. 8 Weaver JP, Odell P, Nelson C: Evaluation of the benefits of gastric tube feeding in an elderly population. Arch Fam Med 1993;2:953–956. 9 Schneider SM, Pouget I, Staccini P, et al: Quality of life in long-term home enteral nutrition patients. Clin Nutr 2000;91:25–28. 10 Silver HJ, Wellman NS: Family caregiver training is needed to improve outcomes for older adults using home care technologies. J Am Diet Assoc 2002;102:831–836. 11 Wilcock H, Armstrong J, Cottee S, et al: Parenteral Nutrition support for the Cambridge Health District. Health Trends 1991;23:93–100. 12 Schattner M, Barrera R, Nygard S, et al: Outcome of home enteral nutrition in patients with malignant dysphagia. NCP 2001;16:292–295. 13 Light VL, Slezak FA, Porter JA, et al: Predictive factors for early mortality after percutaneous endoscopic gastrostomy. Gastrointest Endosc 1995;42:330–334. 14 Steinkamp G, von der Hardt H: Improvement of nutritional status and lung function after long-term nocturnal gastrostomy feedings in cystic fibrosis. J Pediatr 1994;124:244–248.
53
Home Enteral Nutrition 15 Lee JH, Machtay M, Unger LD, et al: Prophylactic gastrostomy tubes in patients undergoing intensive irradiation for cancer of the head and neck. Arch Otolaryngol Head Neck Surg 1998;124:871–875. 16 Urban RG, Terris DJ: Percutaneous endoscopic gastrostomy by head and neck surgeons. Arch Otolaryngol Head Neck Surg 1997;116:489–492. 17 Kang A, Zamera SA, Scott B, et al: Catch-up growth in children treated with home enteral nutrition. Pediatrics 1998;102:951–955. 18 Holden CE, Puntis JWL, Charlton CPL, Booth IW: Nasogastric feeding at home: Acceptability and safety. Arch Dis Child 1991;66:148–151. 19 Kirby DF, DeLegge MH, Fleming CR: American Gastroenterological Association technical review on tube feeding for enteral nutrition. Gastroenterology 1995;108:1282–1301. 20 DeLegge MH, Kirby DF: Enteral feeding, formula, delivery and complications. Pract Gastroenterol 1992;16:32–44. 21 Wolfsen HC, Kozareck RA, Ball TJ, et al: Long-term survival in patients undergoing percutaneous endoscopic gastrostomy and jejunostomy. Am J Gastroenterol 1990;85: 1120–1122. 22 Larson DE, Burton DD, Schroeder KW, et al: Percutaneous endoscopic gastrostomy: Indications, success, complications and mortality in 314 consecutive patients. Gastroenterology 1987;93:48–52.
Discussion Dr. Buchman: One of the things that I will touch on briefly in my talk tomorrow is something that I just recently found out and perhaps you could comment on. In nursing homes, especially where a lot of patients receive enteral nutrition, generally via percutaneous endoscopic gastrostomy (PEG), reimbursement is substantially greater for those patients receiving enteral feeding via a PEG than those receiving the same enteral feeding via a nasogastric (NG) tube. In both cases reimbursement is substantially greater than for those patients who don’t receive any enteral nutrition for whom staff must be hired to actually feed them. So do you think the fact that nursing homes make a lot of money out of enteral nutrition drives the significant increase in the number of enterally fed patients in the nursing homes and the numbers that you stated on the doubling of patients in the US on enteral nutrition? Does that actually largely reflect the increased volume from the nursing home? Dr. DeLegge: The whole market does it. I think the whole market has a lot to do with consumers because many people are asking for things. Families ask for it because they have heard of it from other people, and also the fact that physicians by and large are a little more knowledgeable about what they have available. But I agree with you, the nursing home situation in the US is a mess, and it is kind of a square box that is meant to keep you in and not able to get out. If I send someone to a nursing home with an NG tube, compared to a PEG, the nursing home gets less reimbursement for enteral nutrition. If I send someone to a nursing home with oral feeding, meaning that they are going to be fed my mouth, this means that someone has to warm up their food, someone has to sit at the bedside, place the food in their mouth, watch them swallow and thus give them some sort of bedside feeding. In fact because of the time constraints, often those patients by and large ultimately won’t be fed adequately. So there is a real drive from the nursing homes in the US to put in PEG tubes as often as possible in anybody with any dysphagia simply because they don’t have the resources to do what they would need to do otherwise. The Federal Government, and the Commissions of Accreditation for nursing homes inspect them and will give major negative points for nursing homes with patients who are losing weight. So although the patient may be losing weight appropriately, meaning that they have end-stage
54
Home Enteral Nutrition dementia and are in their last 2 months of life, the nursing home would be criticized because the patient was allowed to lose weight. Dr. Lochs: Did your data include liquid oral supplements? If one looks at the literature it seems that oral supplements are even more effective than tube feeding which might be due to the argument that in the earliest stage there is still some reversal of malnutrition possible or something like that. Dr. DeLegge: Unfortunately the database that I have does not include those patients who went home on oral supplements. My own preliminary data from Charleston justify what I was just trying to say earlier today. If I have a patient with a reversible process, meaning perhaps that they have had pancreatic surgery for cancer and are going to go home but are not consuming enough calories, the utilization of an oral supplement seems to be very beneficial. When I have elderly patients who have started to stop eating, I have not had good success with providing oral supplements to that population because it doesn’t change what they have actually taken in. They use the supplements as a meal substitute. Dr. Bowling: Forgive me if I misunderstood this. But did you say that 16% of these patients require parenteral feeding as well? So 1 in 6 of your patients are on two methods of nutritional support? Dr. DeLegge: No, that is not correct. What happens is 1 in 6 are going from enteral to parenteral and back to enteral, based on the physician’s decision for whatever reason that is. This is a relatively high amount. Dr. Elia: I was interested to hear about the readmission rates of 10–15% in the US, which are in excess of what we see in the UK and probably other European countries. This raises the question as to whether the patients are appropriately trained before they go home, and whether there are appropriate plans in place to ensure continuity of care. Do you think there is a lack of an appropriate policy, and if so what are the reasons behind this? Dr. DeLegge: It is a secondary defect. The amount of education that is done prior to discharge is very minimal, the amount of actual clinical intervention at home by a clinician, once the patients are discharged, is again pretty minimal. You are absolutely correct, this is an area where we need to have standardized protocols for intervention. We do have some standardized protocols at our own center. But I can assure you that this is a rarity for what happens in the US, so we are sorely lacking in patient follow-up. Dr. Labadarios: Do you give enteral nutrition as a means of sustenance just to survive, or do you give disease specific enteral nutrition, for example in dementia? If you do it from the point of view of altering the course of disease, whatever that may be, do you have any data on that aspect of your presentation? Dr. DeLegge: Within which group, the dementia patients? Dr. Labadarios: Any group in terms of complications, and in terms of outcomes. Dr. DeLegge: The few studies that I showed you in which the complications were very low in number included head and neck cancer, pharyngeal cancer, and cystic fibrosis; definitively good data and outcomes. Dr. Bozzetti: I would like to come back to the patients with the incurable disease on enteral nutrition. Why don’t you accept that the final survival rate is an indicator of outcome? If you consider that healthy people undergoing total macronutrient starvation would survive only 2 or 3 months, it is obvious that if there is a group of patients who survive more than 3 months, this can be possible only thanks to the parenteral or enteral support. Dr. DeLegge: I agree with you. I was questioned a month ago by a major health insurer in the United States as to why I wanted to place a feeding tube in a patient with terminal cancer. This patient was still functional and able to walk, so I wanted to
55
Home Enteral Nutrition provide him with home enteral nutrition (HEN). I was told that HEN wasn’t going to change his outcome. So the patient went home anyway on HEN but had to pay for it out of his own pocket. But you are absolutely correct that there are good data showing that the terminally ill who are still functional can in fact benefit from nutritional support. Dr. Hébuterne: You mentioned quality of life as an important issue in HEN patients, but we have two major problems in assessing quality of life. We have no specific tool and many patients are incompetent, patients with neurological disease are incompetent, are not able to understand. I would be happy to have your opinion on this point, how to assess quality of life in these patients. Dr. DeLegge: I readily admit that our ability to assess quality of life in those patients is poor. There is a very good recent scoring system looking at pain in patients with dementia [1]. Additionally it measures some quality of life issues, and it has been standardized. However, I would agree with you, most of what we do is talking and asking the family if the patient is doing better? Dr. Hébuterne: I think it is really important to have the patient’s opinion. For example in my center if we ask a doctor about the incidence of diarrhea and constipation in HEN patients it is between 15 and 20%, but if you ask the patients 60% of them are not happy because they have constipation and 40% of them have diarrhea, nobody is happy. So the patient’s opinion is probably different from that of the caregiver. Dr. DeLegge: I agree. Dr. Morley: I don’t think it is fair to blame the nursing homes for the high amount of tube feedings in the United States. In the United States we have a medical legal system that rules by its own laws that virtually force you to have tubes put into people if the family wants it. So the number one thing is that the family has to be asked and really the reason is the family more than anything else. The second reason, if you want to say it has anything to do with money, is gastroenterologists who have no problem whatsoever in putting tubes into people who I think are dead. But they put them in and sent the patient back to me in the nursing home. The third problem is so to speak the therapist who has learned that any dysphagia requires altered food. We all aspirate a little bit but in a nursing home everybody aspirates and therefore not everybody should be fed. And if you go down that road you go very rapidly to having no choice but to put in a tube. I don’t think that I have ever seen a nursing home administrator come to me and say we will make more money if you put in a tube than if we feed people; I have never had that pressure, but I have certainly had other pressure. So I think we need to be very careful not to blame the nursing homes. The second part of it, which I think is also important, is you said you had a 15% readmission rate. I presume that is for everything, not just tube-related. That is where the difference to England is, we put tubes into people who would never ever get a tube anywhere else in the world. I mean these people wouldn’t be discharged either, we discharge people after 4–5 days and we send them home tremendously sick and that 15% admission rate for this population is low. For our subacute care we run a 30% readmission rate because of the type of people we are discharging. So I think again we should not blame the tube for a 15% readmission rate. Dr. Buchman: I would like to address Dr. Morley’s comment. Being a gastroenterologist I can tell you there are people in my medical center that would put tubes in people who are nearly brain dead because reimbursement is significant for PEG placement. However, it doesn’t take away from the fact that the nursing homes do make a lot of money from patients who are fed via a PEG and they drive this industry. Nurses also often refuse to accept a patient from the hospital without a PEG I suspect often because of reimbursement issues. If patients are in an acute care hospital in the US,
56
Home Enteral Nutrition they can’t stay there, they have got to go somewhere. In fact there is a survey of 42 states in which all 42 state that the reimbursement was significantly greater for having the PEG tube [2]. In fact in the US it almost seems by law that we must spend 90% of our health care dollars in the last 30 days of life. It contrasts significantly to Europe because many of the patients who get tubes in the US never would have even been admitted to hospital in Europe. But we are forced to take care of these patients, not so much because of the family but because of where they ultimately go. Our physicians can convince the family of anything depending on how they bias or slant the facts. Dr. Correia: I would like to go back to the advanced cancer patients, the terminally ill patients as Dr. Bozzetti mentioned. I think we cannot measure outcomes in terms of survival or quality of life because as we have seen it is very complicated to define quality of life, etc. But we do have a very good argument and it comes from hunger strikers; as we have seen in several natural trials if they don’t eat they die after 6 weeks [3]. So if we don’t feed these patients I cannot show insurance companies or any other person that they would do better, but they certainly would not die of passive euthanasia. From what I know in most of our countries euthanasia is not allowed or in terms of law it is not accepted. So the justification to these insurance companies is just that if you don’t allow enteral nutrition you are pro-euthanasia, which again is not legal. Dr. DeLegge: I will try that next time. Dr. Schwab: I would like to come to this extraordinary mortality rate after PEG placement which to me is actually a disaster. You have to imagine a procedure is indicated in patients and you think you can supply nutrition for at least 1 month, but a fifth of them is dead after this time. Dr. DeLegge: They are not dying from the tube, they are dying from their primary disease. Dr. Schwab: The problem is that these are probably very valid data because I recall 4 articles [4–7] in which exactly the same 20% is stated, not only in your country but in ours also. So the question is what are we doing wrong? Is our selection probably wrong? I recall just one idea that inpatients are doing much worse than outpatients. There was a study from New York [4] showing this quite nicely. But I am not aware of any other risk factors for early mortality because we have to exclude these patients because it is, as everybody would probably agree, just nonsense to put a tube in these patients. Dr. DeLegge: Currently we have a prospective study going on in our hospital using something called the Charlson co-morbidity index, which basically is going to score patients for severity of disease prior to them getting a feeding tube. We will look for an association between that index and patient outcome to see if perhaps that can help better chose patients who should or should not get a tube. In the US generally what happens is that the family doctor and the family have already decided that they want a feeding tube. The specialists, either the surgeon or the gastroenterologist, is consulted to place the feeding tube. If the specialist decides a tube is not warranted, it can be very difficult to convince the patient and/or their families who have already made up their minds. So part of the issue in the US is education, not so much at the specialist level, but at the primary care level. Dr. Powell-Tuck: The proportion of patients fed for cancers has dropped over the last 20 years or so. I think it was something like 55% and went down to 20%. Is this just a trick of proportions or is it a reduction in numbers? Dr. DeLegge: That is a great question. The data I did not show you were with regard to patients receiving home parenteral nutrition. There has been some shifting of patients getting home enteral to the home parenteral side, and there has also been
57
Home Enteral Nutrition some drop off of people who were previously getting tube feeding for obviously terminal disease. The other issue is the fact that the neurologic group as a whole has just exploded in numbers, which makes the cancer population by and large look smaller too.
References 1 Villanueva MR, Smith TL: The PADE: A new instrument to assess pain in the demented elderly. AAHPM Bull 2003;fall:4–6. 2 Mitchell SL, Buchanan JL, Littlehale S, Hamel MB: Tube-feeding versus hand-feeding nursing home residents with advanced dementia: A cost comparison. J Am Med Dir Assoc 2003;4:27–33. 3 Allison SP: Malnutrition, disease, and outcome. Nutrition 2000;16:590–593. 4 Abuksis G, Mor M, Segal N, et al: Percutaneous endoscopic gastrostomy: High mortality rates in hospitalized patients. Am J Gastroenterol 2000;95:128–132. 5 Loser C, Wolters S, Folsch UR: Enteral long-term nutrition via percutaneous endoscopic gastrostomy (PEG) in 210 patients: A four-year prospective study. Dig Dis Sci 1998;43: 2549–2557. 6 Sanders DS, Carter MJ, D’Silva J, et al: Percutaneous endoscopic gastrostomy: A prospective analysis of hospital support required and complications following discharge to the community. Eur J Clin Nutr 2001;55:610–614. 7 Wolfsen HC, Kozarek RA, Ball TJ, et al: Long-term survival in patients undergoing percutaneous endoscopic gastrostomy and jejunostomy. Am J Gastroenterol 1990;85:1120–1122.
58
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 59–71, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
Home Enteral Nutrition Epidemiology and Legislation in Europe
A. Van Gossum Department of Gastroenterology and Hepatopancreatology, Hôpital Erasme, Université Libre de Bruxelles, Bruxelles, Belgique
Introduction Enteral nutrition is the preferred way of feeding patients who cannot maintain sufficient oral intake but have a functioning gastrointestinal tract. Enteral feeding has been used for several decades for hospitalized patients, but for the last 20 years home enteral nutrition (HEN) has been expanding in home care in many industrialized countries throughout the world. Despite the fact that the number of patients on HEN is now much higher than patients on home parenteral nutrition (HPN), scientific interest and medical concern were initially much higher for HPN [1–3]. The number of scientific publications on the use of HEN is quite weak. In many European countries, legislation on HEN has only quite recently come into effect or is lacking. There are no real guidelines on the correct use of HEN while guidelines for enteral nutrition in hospitalized patients have recently been published [4]. In this chapter, we have tried to collect data on epidemiology, legislation and the current use of HEN throughout Europe.
Definition In 1999 a survey was performed by the ESPEN-HAN working group on legislation of home artificial nutrition in different European countries [5]. It appeared that there was little agreement about what constitutes HEN. Indeed, while in Italy, France and the UK only tube feedings covering ⬎75% of requirements are considered as enteral nutrition, in 6 other countries (Belgium, Czech Republic, Denmark, Israel, Poland and Spain) both tube and 59
Home Enteral Nutrition oral feeding covering ⬎75% requirements apply for this consideration, and in two countries (Austria and Croatia) any kind of enteral diet or supplement is considered as enteral nutrition. In the European epidemiological survey that was also done in 1999 by the ESPEN-HAN working group, HEN was defined as the provision of enteral diets as the main source of daily intake at home [6]. In an article entitled ‘Standards for home nutrition support’, the ASPEN defined enteral nutrition as ‘nutrition provided via the gastrointestinal tract’ either ‘orally’, meaning enteral nutrition taken by mouth, or ‘by tube’, meaning enteral nutrition provided through a tube or catheter or stoma, delivering nutrients to the oral cavity [7]. The BANS report on home artificial nutrition provided data on parenteral nutrition and enteral tube feeding [8]. In a recent report done by the NADYA-SENPE working group (Spain), Planas et al. [9] described 2,986 patients who were enrolled in HEN during the year 2000. Data were collected from 22 hospitals. Among these patients who were considered to be on HEN, oral nutrition was the preferential route in 50.8%, followed by nasoenteral tube (30%), and in 17.4% ostomy tubes were placed. A prospective observational study was also performed in the Valladolid area in Spain between January 1999 and December 2001 showing that HEN was administered orally to 79% of the patients [10]. The lack of agreement on the definition of HEN between the different European countries obviously hampers interpretation of the epidemiological data.
Incidence – Prevalence The real incidence and/or prevalence of HEN in Europe are not really known. The main reasons are the large number of HEN centers in many countries and the lack of a registry including all newly enrolled patients. Obviously, the incidence may differ regarding the definition of HEN per se as well as the consideration of adult patients only or children (as is the case in the BAPEN report). So, estimation of the incidence is generally based on clinical practice in some areas. In the European epidemiological survey, the incidence was considered to be reliable in 8 limited areas in which HEN was covered by a specific center [6]. The median yearly incidence of adult patients receiving HEN was 163 patients/million population/ year but the range was between 62/million/year in Turin and 457/million/ year in Ivrea (Italy). In the Valladolid area in Spain, the incidence was 150/million/year in 1999, 213/million/year in 2000 and 95/million/year in 2001 [10]. The BAPEN report showed that the point prevalence of registered adult patients receiving home enteral tube feeding (HETF) between the end of 60
Home Enteral Nutrition 1996 and the end of 2002 increased from 4,236 to 15,148 patients [8]. In the same report, they showed that, at the end of 2002, there were 4,219 children (⬍16 years) registered as receiving HETF from 170 reporting pediatric centers (⬃11% growth in point prevalence compared with 2001). Although the number of reporting pediatric centers submitting to BANS has remained similar over the past 7 years (1996–2001), the number of new registrations increased by 50% during this time. In the Spanish annual report, data were based on 22 reporting hospitals which included 2,986 patients during the year 2000 [9]. So in the UK the number of registered patients (according to the period prevalence) was 50 per center, while in Spain this number reached 130/center. The difference is probably due to selection of the Spanish centers that were more interested in participating in the survey. Although there are no official data available for Germany, it seems that 140,000 percutaneous endoscopic gastrostomies (PEGs) are implanted every year, and 100,000 patients are on HEN including 40% with deep neurological alterations. Patient’s Characteristics In the European survey, the age distribution for adult patients was: 16–40 years (7.5%); 41–65 years (39%); 66–80 years (34%), and ⬎80 years (21%) [6]. The male/female ratio was 1.62. In the UK, the age distribution for adult was quite similar: 16–40 years (8.9%); 41–60 (20.6%); 61–70 (19.5%), and ⬎71 years (51%). In this survey, compared to the patients already receiving HEN, the new registration consisted of a greater proportion of elderly patients (⬎70 years) and a smaller proportion of younger people (⬍50 years). In the UK in 1999, the age distribution for children was: 0–1 years (37%); 1–2 years (13.4%); 3–5 years (16.8%); 6–12 years (19.4%), and 13–16 years (13%) [8]. The point prevalence remained stable between 1996 and 1999. The age distribution may be related to the underlying disease of the patients requiring HEN. The Dutch national registry reported their experience over the period 1998–2000 for patients with head and neck cancer [11, 12]. The mean age was 60 ⫾ 12 years for 1,094 patients. In the Spanish national register, the mean age of 2,986 patients was 65 ⫾ 19 years.
Underlying Diseases and Indications In the European survey, underlying diseases for HEN in the 1,397 patients were: neurological diseases (n ⫽ 620); head and neck cancer (n ⫽ 423); 61
Home Enteral Nutrition benign digestive disease (n ⫽ 148); geriatric disease (n ⫽ 101); AIDS (n ⫽ 10), and miscellaneous (n ⫽ 95) [6]. Dysphagia secondary to a swallowing disorder was the principal reason for HEN in 81% of the patients, oral failure (without dysphagia or a digestive disorder) in 14.4% and partial intestinal failure in 4.8%. In the Spanish registry (2000), among 2,986 patients, 41.2% were diagnosed with neurological diseases and 33% with cancer [9]. In the BANS report (2002) [8], the underlying diseases were: cardiac disease (0.9%); central nervous system problems (60.5%); gastrointestinal diseases including esophageal cancer (24.2%); renal diseases (0.4%); respiratory diseases (1.7%), and other diseases (12.3%). Swallowing disorder was the main indication in 72.6% [8]. In the pediatric population, the distribution of underlying diseases was as follows: cardiac (9.2%); central nervous system (35.6%); renal diseases (2.1%); gastrointestinal diseases (16.6%); respiratory (6.2%) including 3.7% with cystic fibrosis, and other diseases (29.4%). The reasons for nutritional support were: failure to thrive (39.8%); swallowing disorder (21.1%); to improve nutritional status (24.3%), and malabsorption in 3.2%. We obtained data from the Children’s Memorial Health Institute in Warsaw about 101 children with PEG suffering from primary neurological diseases (52%), metabolic diseases with secondary encephalopathy (20%), gastrointestinal diseases (22%), neoplastic diseases (3%), anorexic patients, and microsomic patients (3%).
Technical Aspects (Route/Formula) In the 1,397 patients described in the European survey, HEN was administered through PEG in 813 patients (58.2%), a nasogastric tube in 410 patients (29.3%), surgical jejunostomy in 76 patients (5.4%) and by other access in 3.4% [6]. A pump was used in 43% whereas 57% received HEN by gravity without a pump. The proportion of patients fed via PEG and with a pump was different between countries. The mode of administration of HEN was cyclic nocturnal for 36.7%, cyclic diurnal for 24.8%, continuous for 4.4%, and by bolus in 34.1%. There were important variations between the centers concerning the mode of administration of nutrients. In some centers, most patients were on cyclic enteral nutrition whereas in others most were fed by bolus. Commercial rather than ‘home-brewed’ preparations were used in almost all patients; the feeds were standard or high energy (⬎1 kcal/ml) in 65.3%, enriched with fibers in 24.5%, elemental or semi-elemental in 5%, or modified in some other way in 5.2%. In the Spanish registry, polymeric formula composition was used in 83% [9]. 62
Home Enteral Nutrition In Europe, the HEN technique was performed by the patients (17.1%), relative/caregivers (38.1%), nurse (35%), or others (9.8%), but important variations were observed throughout the centers. In the 2002 BANS report [8], 21% of the patients were independent and able to manage their nutritional support, 19.6% required some help, and 58.7% required total help. Patients were treated in a nursing home (38%), in their own home (54%), or other location (8%). On the contrary, for children, 97% were managed in their own home. In a local series (Brussels) of 210 patients in whom a PEG was placed for enteral feeding, 30 (15%) were taught to manage their feeding independently (personal data). In the Dutch registry, Van Reeuwyk-Werkhorst et al. [11] showed that feeding characteristics at baseline differ significantly between the hospitals. General hospitals (57%) used PEG at baseline more often than academic hospitals (32%) and oncology hospitals (41%). In all groups, HEN was administered mostly by bolus (36–64%) except for females in academic hospitals in whom HEN was mostly administered intermittently (49%). The most frequently used diameter of the feeding tube was 8 french (46%) in oncology hospitals and 14 french (33%) in academic hospitals. The medical devices most frequently prescribed were a syringe in oncologic hospitals (58%), a feeding pump (57%) in general hospitals and in academic hospitals a clamp for males (41%) and a feeding pump for females (39%) were used. In the population aged ⬍16 years, the use of nasogastric tube in patients was 92% and the use of PEG was 7%. HEN was delivered either by bolus in 73% or was intermittent in 17%. In infants, a syringe was mainly used (39%) and in children a pump with a rucksack (59%).
Legislation According to the survey performed in 1999, five countries (Austria, Croatia, Germany, Poland and the UK) have no legislation regarding HEN [5]. Italy and France were the first countries to pass legislation on reimbursement policies in 1988 (table 1). This legislation is applied nation-wide in 6 countries and is regional in Italy. Except for Croatia and Belgium, it applies for both children and adults. In Belgium, children ⬍2 years receive special reimbursement consideration. In Switzerland, there is also well-defined legislation. In most of the countries (Austria, Belgium, Denmark, Germany, Israel, Italy, Poland, Spain and the UK) any hospital can take care of HEN patients. In France and the Czech Republic, HEN is restricted to a certain number of centers for geographical reasons, and in Croatia it is based on the hospital size or specialization. Nevertheless, almost any physician can prescribe HEN 63
Home Enteral Nutrition Table 1. Date of approval of home enteral nutrition legislation
Year
Country
1988 1990 1996 1997 1998 1999
Italy, France Czech Republic Belgium Israel Spain Denmark
From Moreno et al. [5].
except in the Czech Republic and Croatia, although preferably members of a nutritional support team (NST). In general, once the patient is sent home the general practitioner or the NST must take care of the medical follow-up. HEN is restricted to a limited number of diseases or conditions in 5 countries (Belgium, Croatia, France, Czech Republic and Spain) as well as in Switzerland. Any marketed enteral diet can be prescribed, except in Denmark, Croatia and Spain. Funding Regarding funding arrangements, the answers vary widely [5] (table 2). There are no clear indications about disposables and infusion pumps in most of the countries. Provision of an enteral diet and equipment does not follow a unique pattern, as shown in table 3. We asked for the existence of written guidelines both for health care workers and for patients, and we did not receive a uniform answer. These data were confirmed in the survey performed by the ESPEN-HAN group in 1999 [6]. Daily costs of HEN were not available in centers from Denmark and the UK. In the other center the daily overall costs of HEN varied from EUR 7 to 25. The daily cost of HEN was EUR 12.4 in Belgium, EUR 10.0 in France, EUR 23.3 in Germany, EUR 24.2 in Italy, EUR 12.0 in Poland, and EUR 16.7 in Spain. These costs include the formula, the infusion pump, micronutrients, and all the necessary equipment: bags, tubing and dressings. They do not include the cost of the caregiver, cost of rehospitalizations, and medical monitoring. In Poland, the situation is ambiguous. HEN is not considered to be an approved medical service although many patients are receiving HEN. Patients must pay for the feeds that they can buy open in a pharmacy, but they have difficulty with tube feeding. In conclusion, the use of HEN is expanding in many European countries. It is surprising that the definition of HEN differs from one to another country (oral, tube or both). The real incidence and/or prevalence of HEN in Europe 64
Home Enteral Nutrition Table 2. Home enteral nutrition funding arrangements Formula
Disposable
National health system Totally
Croatia, Czech Same Republic, France, Germany, Italy, Spain and UK Partially Austria, Belgium Same Denmark Private insurance Germany Germany Patient Totally Israel and Israel and Poland Poland Partially
Austria, Belgium and Denmark
Medical care Nursing or dietician care
All except Belgium and Israel
All except Belgium, Israel and Poland
Germany
Germany
Belgium and Israel
Belgium, Israel and Poland –
Austria, Belgium – and Denmark
Belgium has specific regulation for children and varies according the type of formula. Germany has 100% national health system (NHS) or private insurance. Co-payments: Belgium NHS 30%, patient 70%; Denmark NHS 60%, patient 40%; Austria NHS 60–70%, patient 30–40% (for tube feedings 100% NHS). From Moreno et al. [5].
Table 3. Provision of enteral diet, disposables and pumps Enteral diets
Disposables
Pumps
Hospital
Croatia, Czech Republic, France and Spain
Croatia, Czech Republic, France, Poland and Spain
Private pharmacists
Austria, Belgium, Czech Republic, Germany, Israel Poland, Spain and UK Austria, Belgium, Germany, Israel, UK and Italy
Austria, Belgium, Czech Republic, Denmark, Germany, Israel and Poland Belgium, Denmark, Germany and UK
Belgium, Croatia, Czech Republic, France, Spain and UK –
Home care firms Primary care
Italy and UK
Austria, Belgium, France, Germany and UK Italy
Enteral diets: Croatia and France 100% hospital; Italy 100% primary care; data not available from Denmark. Disposables: Croatia, France and Spain 100% hospital; Italy 100% primary care. Pumps were not used. Poland and Israel used others. From Moreno et al. [5].
65
Home Enteral Nutrition are unknown; the incidence of HEN is estimated to be approximately 160/million inhabitants/year. While the main indication is a swallowing disorder, the principal underlying diseases are neurological disorders and head and neck cancer. In an adult population, the percentage of patients older than 70 years is higher than 50%. Legislation and funding for HEN differs in several countries and are still lacking in some.
References 1 Van Gossum A, Bakker H, Bozzetti F, et al: Home parenteral nutrition in adults: A European multicentre survey in 1997. Clin Nutr 1999;18:135–140. 2 Howard L, Ament M, Fleming CR, et al: Current use and clinical outcome of home parenteral and enteral nutrition therapies in the United States. Gastroenterology 1995;109:355–365. 3 Elia M: An international perspective on artificial nutritional support in the community. Lancet 1995;345:1345–1349. 4 Stroud M, Duncan H, Nightingale J, British Society of Gastroenterology: Guidelines for enteral feeding in adult hospital patients. Gut 2003;52(suppl 7):vii1–vii12. 5 Moreno JM, Shaffer J, Staun M, et al, Home Artificial Nutrition Working Group – ESPEN: Survey on legislation and funding of home artificial nutrition in different European countries. Clin Nutr 2001;20:117–123. 6 Hebuterne X, Bozzetti F, Moreno Villares JM, et al, ESPEN – Home Artificial Nutrition Working Group: Home enteral nutrition in adults: A European multicentre survey. Clin Nutr 2003;22:261–266. 7 ASPEN Board of Directors: Standards of home nutrition support. Nutr Clin Pract 1999;14:151–162. 8 Glencorse C, Meadows N, Holden C: A report by the British Artificial Nutrition Survey (BANS) a Committee of BAPEN. Trends in Home Artificial Nutrition Support in the UK during 1996–2002. 9 Planas M, Castella M, Garcia Luna PP, et al: Enteral nutrition at home: National register for the year 2000 (in Spanish). Nutr Hosp 2003;18:34–38. 10 de Luis DA, Aller R, de Luis J, et al: Clinical and biochemical characteristics of patients with home enteral nutrition in an area of Spain. Eur J Clin Nutr 2003;57:612–615. 11 Van Reeuwyk-Werkhorst J, Beaumont M, Nieboer M: Home enteral nutrition in patients with head and neck cancer in The Netherlands. Clin Nutr 2002;21(suppl 1):37. 12 Beaumont M, Van Reeuwyk-Werkhorst J, De Hullu A: Home enteral nutrition in infants and children in The Netherlands. Clin Nutr 2002;21(suppl 1):218.
Discussion Dr. Hébuterne: I have a question for both speakers about the incidence. The incidence in Europe is somewhere between 100 and 400 but probably closer to 100 than 400, and in the USA probably more than 400. Could you explain why there is this difference between Europe and the USA, and who is right? Dr. Van Gossum: I know that the difference is the same for home parenteral feeding. We expect that there are 10 times more patients on home parenteral feeding in the USA than in Europe, and it is the same for enteral feeding. The distribution of the underlying disease is similar, so it is not a question of disease. The age of the patient is also similar, so it is probably a question of routine practice and, as you mentioned, probably that in the USA the patients are discharged home much faster than in Europe. For me this is the only reason.
66
Home Enteral Nutrition Dr. Steinhagen-Thiessen: May I also ask you a question on the same topic? There were data about increasing enteral feeding and, if I remember correctly, in the adult it was an increase of 26% and in children of 14%. I can understand in the adults, but are there more children or, is this a growing population of the children? Dr. Van Gossum: Those are data from the UK and perhaps we can ask our colleagues to give some more details about these data. Dr. Elia: We have been amalgamating data in the British Artificial Nutrition Survey since 1996. In the UK the single commonest diagnosis for home enteral tube feeding in children is cerebral palsy, and there has been a tendency to have a lower threshold to start this. In the same way there has been a lower threshold for starting tube feeding in elderly, even when taking into account specific age categories. I would just like to raise one point about the amount of tube feeding taking place. When I look at different values from various parts of the world, I sometimes find it difficult to know exactly what they mean. It seems that they are sometimes referring to number of new cases over a period of time (period prevalence), at other times to the prevalence over a period of time (period prevalence) and yet at other times to the prevalence at a point in time (point prevalence). I feel that the distinction between these is sometimes confused, making it difficult to make direct comparisons. Dr. Van Gossum: It is true that it is confusing and it is also true in other fields of medicine about epidemiology, the terms of incidence, prevalence, point prevalence, are probably not correctly understood. We are talking about the yearly incidence of new patients, and it is true that the number is probably different when we are talking about point prevalence. According to the data that I got from the UK for adults, there is a point prevalence of about 9,000, but the number, the approximation of the number of patients on home enteral nutrition was about 15,000 patients. Dr. Elia: It has been growing since the 1999 or 2000 report that you refer to. Interestingly enough the growth has been increasing but beginning to slow down in percentage terms. We do not have the full picture although data from over 250 centers in the UK suggest that the point prevalence is now in excess of 25,000. Dr. Thomas: Just a couple of comments regarding what we are seeing at least in the geriatric population because I think we have a slightly skewed view in most of the patients who we start feeding enterally very closed to the end of life. In all of the prevalence data, one of the things that we need to do is to throw out that population. For example in the United States over half of the patients are referred to hospice care in the last 2 weeks, so it is a very sort of failing end of life decision to put a tube in. It makes the family feel more comfortable, is often demanded, and so it is a term event. It is not really done for any sort of staying reason or rational therapeutic purpose. So I think if you remove those, which is a substantial percentage, then you are closer to the true number that are started. For that reason if you look at the data in long-term care, 40% of the patients who start enteral feeding in the geriatric population die within 6–12 months, and again it is because of this sort of false bias that occurs in this population. One of the reasons why we have more enteral tubes put in in the United States is simply because in nursing homes when people are restrained nasogastric tubes get pulled out very quickly, whereas the enteral tube tends to stay in longer. So it probably has more to do with that than it has to do with the 2 or 3 dollars difference in reimbursement. Dr. Bowling: I would appreciate your views on providing the service. We have heard from the United States that patients are pretty much left on their own, plus or minus an enthusiastic physician, if they are lucky. In the UK it is certainly very patchy. If someone is very lucky there will be a full multidisciplinary nutrition team in the community, but they are the absolute exception rather than the rule. Sometimes it is just a dietician, sometimes it is just a nurse, sometimes it is a general practitioner and
67
Home Enteral Nutrition very often it is nobody at all. So I was just wondering what your perception was around Europe as to providing home care? Dr. Van Gossum: According to the survey that we performed it was also very variable from one center to another. There are some really well-organized centers, such as in Nice with Dr. Hébuterne, and they have the whole strategy to follow the patients. But we have to recognize that in many centers it is totally different. Sometimes there is a nutrition team who will join up with the home care but in many other centers there is no organization, and it is true that it depends on the general practitioner. Sometimes there is a dietician at home but it seems very infrequent to have a reevaluation of the nutritional needs of the patients once he is at home. The main problem is technical, but I am aware of at least one study showing that there is only a small percentage of patients on home enteral nutrition who will be reassessed for the need for the feeding [1]. So they are discharged with 1,500 kcal and then that is all, they will maintain that for 1 or 2 years, and nobody will take care about that, which is very common unfortunately. Dr. Bowling: I think I have raised a very important issue as to removing a percutaneous endoscopic gastrostomy tube once the patients are out of the hospital setting. I am sure there are many patients who probably do recover their ability to swallow and can eat and drink but don’t get reassessed. There are a number of patients I have been contacted about, perhaps 3 or 4 years down the line, who are no longer using their tube and are eating quite normally, but no one has thought to take it out. Dr. Van Gossum: That is the reality. I think that it is very important to provide the patients or their relatives with some information when they are discharged home. But it is not so common, in some hospitals it is not usual to have such a recommendation. Dr. DeLegge: There is an important fact I forgot to mention before in patients with dementia. In the United States upon entering a hospital, the patient signs a living will which outlines whether the patient wants aggressive interventional care such as respirators or tube feedings. Additionally there is a code status assigned to each patient detailing whether aggressive life support is desired or warranted for a particular patient. Recent trends have shown that if the physician or nurse took the time to make sure that the family sat down and addressed the patient’s living will and code status upon hospital admission, then there would be a significant drop in feeding tube placement as compared to statistics from 5 years previously when these items were not addressed. Family pressure to do something has a significant influence on whether physicians will or will not place feeding tubes. Dr. Buchman: One of the comments though in terms of family pressure has again to do with how the physician explains the situation. In fact there is a study that was done with a group of patients, if they had a cardiac arrest would they want cardiopulmonary resuscitation, and I think something like 70% of the patients said yes. Then they re-asked the question after telling them that the outcome at 1 year is not very good and the majority of those 70% said actually we changed our mind, we don’t want to have cardiopulmonary resuscitation. A similar survey was also done in nursing home patients with tube feeding and of those patients who were not demented and could understand the question, something like 55 or 60% said yes, they would consent to tube feeding, but then when they were told that actually with tube feeding a lot of patients had to be restrained, only 25% now wanted tube feeding. So a part of the problem here in terms of the family issue is that the physician has not appropriately informed the family of the situation. The physician often biases the family, using what his beliefs are and what he wants to do. If I want to put a percutaneous endoscopic gastrostomy tube into a patient, almost 100% of the time I can convince the family by slanting the facts of how it is to their benefit and the patient’s benefit to put a tube in. If the family wants a tube and I don’t want to do it, 90% of the time I can convince
68
Home Enteral Nutrition them why I don’t want to do it because I will emphasize the risks. Now I am rare in doing that because that means that I just make a few hundred dollars less with that patient and I get paid USD 30 for convincing him not to do it. Most gastroenterologists won’t do that; they would rather put the tube in and make a couple of hundred dollars. But the fact is that I keep hearing about the family, the family is important but it is rare in my experience. I work in a hospital where we have patients of a very high income level and very high education level, and it is very rare that the patients ask for a tube. It is the physician who suggests it, often the primary care physician, and then when the gastroenterologist gets there who is he to say that the primary care physician is wrong and the patient should not have a tube. So it is really the physician who shapes the family’s wishes. Dr. Pierlich: I have a question regarding the legal situation. In Germany economic pressure is increasing as in other European countries and so there is a new law in preparation which limits the funding for enteral feeding in certain situations. For instance the joint commission of physicians and insurance companies suggested to stop malnutrition and accept as an indication enteral feeding for funding. And now we are arguing against it and are showing data but they try to focus on certain diseases. So my question is how is it in other European countries? Dr. Van Gossum: I tried to show you that in some countries, and I know the case much better in Belgium. There is a list of underlying diseases which has been accepted. Obviously this list of underlying diseases will include the majority of the patients. In some other countries there is no restriction, you may use enteral feeding for all the patients, but for example in Belgium for dementia it is not normally recognized, you may always try to use it. So it is true that there is some economic pressure and as doctors we have to be careful about such a list. I agree with you because for some patients there are some limitations that we could not accept but that is always the border between medicine, economics and ethics, and I think there will be a topic devoted to ethics in nutrition. Dr. Steinhagen-Thiessen: Besides this problem that was just mentioned, we have another problem in our country: we have two different budgets. We have the hospital budget and we have the budget of all these things which are done by the family doctors on the ambulant track. These two budgets are very much in competition, and there we have huge problems. I also think that all these data you have shown us from the different European countries are very difficult to compare because we have such different systems in medicine, how we are organized, for example. In our country the work the family doctor is doing is far separated from the work we are doing in the hospitals. Dr. Van Gossum: I agree, it is true that in the ESPEN-HAN group, which includes doctors, dieticians and nurses from at least 10 European countries, when we compare legislations it is totally different and also changing. From this time there is some new legislation in different countries. In Italy for example, Dr. Bozzetti perhaps you could give some details, I know that it is from one region to another, it is not at the national level in Italy. Is that true? Dr. Bozzetti: Yes, you are right. There are regulations which are different from region to region, and I would say also from town to town in the same region. For instance in Prato, the town where I am working now, the company responsible for the hospital is also responsible for public health in the region, so the budget is unique. In this situation it is easier to have funding for home enteral nutrition because it is the same budget as the hospital. So there is no difference in spending this money for the hospital or for people outside the hospital. But it changes from region to region. Dr. Labadarios: In relation to the comment made regarding variability between countries: how do you plan to actually use all these data? Do you see the European
69
Home Enteral Nutrition Union or the European Council approach as a way of standardization? Are there any moves towards that, or would you rather not go that route? Dr. Van Gossum: We didn’t plan that but it seems to be useful at least to provide such data and such legislation in Europe to some countries, previous eastern European countries where there is no legislation and where home enteral feeding is totally unknown and unrecognized. I talked with Dr. Pertkiewich from Poland who was involved in home enteral and parenteral feeding, and he explained to me recently that he is still fighting to have such reimbursement. So we hope that such a survey could be helpful for these countries, especially now because they have very recently joined the European Community. When I was asked to prepare this talk I sent an e-mail to different colleagues in former eastern European countries and in most of them there is no reimbursement and no practice of home enteral feeding. So throughout Europe we have to improve a lot in the use of such therapy. Dr. Labadarios: I don’t want to preempt Dr. Buchman’s forthcoming talk but there is the question of human right of access to food. So one wonders how can one reconcile that with what is happening in different countries? Dr. Van Gossum: I have no answer but maybe some reaction? Dr. DeLegge: It is a double-edged sword. I will give you a classic example in the US with home parenteral nutrition. Back in the 1980s there was a great deal of time spent collecting data on home parenteral nutrition utilization in the US, and the outcome was that the average number of days patients were on parenteral nutrition, whether they died or came off therapy, was 60 days. Subsequently Medicare, the federal insurance system for seniors and disabled patients, came out with their new criteria for patients to receive parenteral nutrition in the US. One of the criteria was that the physician had to attest to the fact that all patients would require parenteral nutrition for 90 days or greater. They decided to elevate the bar in the US for approval of parenteral nutrition therapy at home. Dr. Elia: Is there a risk under those circumstances that people may be kept on parenteral nutrition for longer to fulfill the 30-day requirement? Dr. DeLegge: Yes, absolutely. If the patient comes off home parenteral nutrition before 90 days, the home parenteral nutrition therapy payment is subject to disqualification. Dr. Morley: It seems to me that we have an absolute posit of data saying who benefits or doesn’t benefit from home enteral feeding. Therefore what is happening is if there is money available to pay for it people get it, and if there is no money they don’t get it. We clearly have made decisions for unconscious people who get fed enterally forever. They are enterally fed forever which certainly doesn’t suggest any quality of life issue there. It would seem that you are in the position to go to the EU now with your data and to at least collect basic quality of life data from around the EU, looking at people going on it and saying does it at all improve the outcome and is it different in various countries. If there was no improvement in the outcome even without a control, which would be ideal but let’s accept that you could not do the control yet, that would make us feel that perhaps we are over-feeding people with enteral feeding. On the other hand if quality of life truly goes up then maybe we should be pushing to give more people enteral feeding, but it is seems from where you are coming from that you are in the ideal position to go to Brussels now and say let’s really find out how much we should be spending. Dr. Van Gossum: Yes it is true, but it is a difficult topic because it is really the border between caring for the patients and ethics. And when we are talking about orally feeding patients, it is true that for many patients when we start either parenteral or enteral feeding, we don’t know exactly the life expectancy of these patients, and most of the time we over estimate the life duration. But it is very difficult to make
70
Home Enteral Nutrition the decision not to feed the patient because of the question of duration of life, because there are some cultural, some religious, some traditions of feeding patients, but there will be a topic about ethics and nutrition. Dr. Ockenga: We recognize that there are several differences in funding as well as in clinical practice in all of our countries, in the US or in Europe. So one step may be just to improve the quality of clinical care for all patients and we could do this by introducing standards, and that is what we would like to do on the European level. We finished the Germany standards on enteral nutrition and now we will introduce these German standards or clinical practice guidelines into the European area, and we will hopefully finish that this year. So that will be a first step to improve the quality, and then we can check if really by doing medicine on equal standards everywhere we can improve the outcome of our patients. Dr. Steinhagen-Thiessen: Do you have any knowledge about the comorbidity of all these patients, and especially do you have any knowledge about those patients who get tube feeding and also had pressure? Dr. Van Gossum: This is quite common but we don’t have data about that. When the patients are discharged home it is quite difficult to have a good follow-up, so I have no answer.
Reference 1 Loeser C, von Herz U, Kuchler T, et al: Quality of life and nutritional state in patients on home enteral tube feeding. Nutrition 2003;19:605–611.
71
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 73–88, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
When Does Malnutrition Become a Risk? L. Gentona, W.G. van Gemerta, C.H. Dejonga, P.L. Cox-Reijvenb and P.B. Soetersa aDepartment
of Surgery, University Hospital, and bNUTRIM Institute, Maastricht University, Maastricht, The Netherlands
Introduction Malnutrition has been defined as a deficiency of energy, protein or other types of nutrients, which produces alterations in body function, is associated with worse outcome from illness and is reversible by nutritional support [1]. Global malnutrition generally results from simple starvation or stress starvation and has to be distinguished from deficiency of one micronutrient or vitamin [2]. Starvation results from a pure deficit of all macro- and micronutrients and occurs for instance in hunger strikers, persons with anorexia, or patients with intestinal diseases leading to malabsorption. Stress starvation, or cachexia, represents the accelerated loss of muscle mass in response to metabolic stress and generally affects patients with inflammatory or neoplastic diseases. Some authors report a third type of malnutrition, sarcopenia [3], which reflects loss of skeletal muscle in the elderly as well as in persons who repeatedly try to lose weight by dieting and in people with growth hormone deficiency, immobilization or arthritis. However, it is questionable whether cachexia and sarcopenia are totally different entities because they may share one or more common pathophysiologic causes including metabolic stress. Whatever the specific etiology is, global malnutrition results in decreased body cell mass (BCM) and fat-free mass (FFM) and leads to diminished body function. FFM can routinely be assessed by several methods, including measurements of skinfold thickness, arm circumference, bioelectrical impedance analysis (BIA) or dual energy X-ray absorptiometry (DXA). In many studies, parallel decreases in BCM and quality of life have been demonstrated [4–7]
73
When Does Malnutrition Become a Risk?
Starvation
Cachexia
• ↓ Energy and protein intakes • ↓ Insulin • ↑ Catecholamines • ↑ Glucagon • ↓ Leptin?
• ↑ Pro-inflammatory cytokines (IL-1, IL-6, TNF-␣) • ↓ CNS input (loss of motor neurons, motor unit activation) • ↑ Glucocorticoids • ↓ Growth hormone • ↓ Estrogen, androgen • ↓ Inactivity • Anorexia • ↓ Protein intakes? • ↓ Leptin?
Loss of body cell mass
Weakness
• • • •
↓ Physical functioning Physical disability ↓ Quality of life ↑ Mortality
Fig. 1. The physiological mechanisms of loss of FFM are complex and associated with inflammatory, hormonal and neuronal interactions. All lead to loss of FFM which eventually increases morbidity and mortality.
but in only very few studies has a critical size of BCM been determined to sustain health [8–10]. In this chapter, we will review the factors responsible for malnutrition, the prevalence of malnutrition in hospitalized patients, and the methods available to assess malnutrition. We will also try to link the loss of BCM with physical disability, impairment of physical functioning, immune function, quality of life and mortality.
Physiological Mechanisms of Malnutrition The mechanisms and causes of malnutrition are shown in figure 1. The calorie deficit occurring during chronic starvation forces the human body to rely on its own energy stores for survival. The priority of the organism is to maintain euglycemia in order to ensure proper function of vital organs such as the brain, heart, kidney and muscles. At the beginning of starvation, the human body reduces insulin secretion and stimulates glucagon and 74
When Does Malnutrition Become a Risk? catecholamine secretion to maintain euglycemia. 75% of glucose is produced by liver glycogenolysis and 25% by gluconeogenesis. In the fasted state, hepatic glycogen stores are depleted within 24 h and gluconeogenesis remains the only glucose source thereafter. The substrates used for gluconeogenesis are amino acids (especially alanine and glutamine) resulting mainly from protein breakdown in skeletal muscle and glycerol produced by lipolysis. After a few days (or weeks) of starvation, gluconeogenesis decreases in response to an unknown mechanism. The brain adapts by using ketone bodies derived from non-esterified fatty acids as metabolic fuel. The metabolic alterations occurring with fasting have been linked, in rodents, to decreased levels of leptin, an adipocyte-derived cytokine [11]. In contrast to pure starvation, muscle wasting occurring in cachexia (stress starvation) can appear despite adequate energy and protein intake. It is the complex result of neuronal, inflammatory and hormonal interactions occurring during metabolic stress. Its pathogenesis has been related to increased levels of proinflammatory cytokines (IL-1, IL-6, TNF-␣, IFN-␥), reduced secretions of insulin-like growth factor-1 (IGF-1) and gonadotropic hormones, and increased secretions of glucocorticoids [12]. Cachexia is often associated with physical inactivity, leading to muscle atrophy and anorexia, an abnormal loss of appetite for food, which complicates the interpretation of physiological findings in humans. The major characteristic of cachexia is a stimulation of skeletal muscle degradation and an increased synthesis of the hepatic proteins implicated in the acute phase response. This re-prioritization of nitrogen utilization accelerates weight loss in cachectic fasting subjects. This results from a mismatch between the amino acid composition of muscle and that of acute phase proteins. Indeed, it has been calculated that 2.6 g of muscle protein has to be catabolized to synthesize 1 g of fibrinogen [13]. Another reason for weight loss during metabolic stress is that a major part of the branched chain amino acids derived from protein degradation are irreversibly degraded to yield glutamine and alanine, so that these branched chain amino acids cannot be utilized for protein synthesis. Changes in fat metabolism also occur and include hyperlipidemia, increased lipolysis, increased de novo triglyceride synthesis, increased free fatty acid turnover and decreased lipoprotein lipase activity. Peripheral insulin resistance, hyperinsulinemia and glucose intolerance form an integral part of the inflammatory response. Recent studies suggest that the loss of skeletal muscle mass occurring with aging also arises from an inflammatory catabolic signal, which could either increase catabolism directly or through the inhibition of anabolic stimuli like diet and physical exercise. This signal may be regulated by IL-6, whose production has been shown to increase with aging [14]. However, sarcopenia may also result from physical inactivity or comorbid factors like arthrosis, heart and lung failure, which also induce an inflammatory response. 75
When Does Malnutrition Become a Risk? Table 1. Changes in body composition and laboratory values in starvation and cachexia Variable
Starvation
Cachexia
Weight Fat mass Fat-free mass Total water Body cell mass Intracellular potassium Extracellular water
↓↓ ↓↓ → or ↓ ↑ ↓ ↓ ↑
↓ or → or ↑ ↓ or → or ↑ ↓↓ ↓↓ ↓↓ ↓ → or ↑
Clinical Consequences of Malnutrition The human body responds to malnutrition with changes in body composition eventually affecting muscle function. These changes are usually reflected in abnormal laboratory values. At the tissue level, body composition can be altered with regard to fat mass (FM) or FFM, which is body mass devoid of all extractable fat and regroups bone mineral mass, non-bone mineral lean body mass, and total body water. At the cellular level, BCM (the active, energy-consuming, protein-rich and potassium-rich intracellular tissue) may be distinguished from extracellular fluids and extracellular solids. BCM is viewed by many as the most relevant body compartment to assess malnutrition. In table 1, the changes occurring in the different types of malnutrition are summarized. Starvation translates into decreased resting energy expenditure, body weight, and a predominant reduction of FM over FFM [3]. The loss of FFM, which presents as muscle atrophy, is inversely correlated with initial adiposity. A closer examination of the composition of the FFM loss shows a reduced bone mass and density and a decreased BCM [3]. The reduction in BCM reflects atrophy and hypoplasia of muscle, intestinal, liver, splenic and immune cells. Total body water as a percentage of body weight is increased with an expansion of extracellular water. Blood values may remain within the normal range, at least during the first 6 weeks of starvation [15]. Refeeding is possible and the resulting gain in body weight is half FM and half FFM. Cachexia differs from starvation by an increased resting energy expenditure corrected for FFM or BCM but reduced total energy expenditure, resulting probably from limited physical activity [16]. Weight loss is common but not universal. There is a decrease in cellular mass, like in starvation, but which occurs much faster and is generally accompanied by hepatomegaly. It is unclear whether hepatomegaly results solely from steatosis or also reflects an increase in cell size or number. In rats, liver size and protein content have been shown to increase after endotoxin injection [17]. In addition, the liver produces increased amounts of plasma proteins, including C-reactive protein, 76
When Does Malnutrition Become a Risk? fibrinogen, complement factors and others. Although plasma albumin levels are decreased, its fractional synthesis rate increases. The amount of extracellular water remains stable or is expanded [16]. The increased protein degradation in muscle cannot be overcome by the anabolic stimulus of feeding, precluding the reversal of the BCM loss. A decrease in BCM and an expansion of extracellular fluid has also been reported in sarcopenia. Histological data report a myocyte hypoplasia and atrophy [18]. Weight and FM loss are not especially associated with sarcopenia. Indeed, in healthy Swiss subjects, FFM decreases but FM and weight increase on average until the age of 74 years [19]. Plasma levels of albumin have been reported to decrease with aging independently of other factors affecting FFM and albumin levels [20]. However, in that study, inflammatory parameters have not been assessed. To summarize, all types of malnutrition show decreased BCM, but the quantitative and qualitative alterations differ. It is of interest that patients may be malnourished despite an increase in body weight (sarcopenia, edema). FFM does not reflect BCM and is therefore, especially in the depleted state and in severe illness, an invalid measure of BCM.
Assessment of Malnutrition Malnutrition is not yet well recognized by many health care professionals. This may be due to a lack of training or interest. In addition, the situation is complicated by the fact that the diagnosis of malnutrition does not rely on one single parameter but requires the integration of nutritional markers and the clinical history and physical examination of the patient. Presently, the most often used clinical tools to detect malnutrition are anthropometric parameters, body composition, laboratory values (plasma albumin, prealbumin, transferrin, C-reactive protein, IGF-1, total lymphocyte count) and nutritional indices (e.g. Subjective Global Assessment, Mini Nutritional Assessment, Nutritional Risk Index). Functional assessment includes testing of muscle function, either by dynamometry for handgrip strength or electrical stimulation, or simply by observation of the physical capacities of the patient [21]. In this context, it must be realized that some of these assumed markers of nutritional status, such as serum albumin, are not nutritional markers, but are predominantly determined by the inflammatory status of the accompanying disease, leading to increased capillary permeability and transcapillary leakage of protein, fluid and electrolytes and to a change in the kinetics of albumin. The identification of malnutrition requires thorough assessment of body compartments. Methods presently available in clinical routine measure two, three or four compartments, as indicated in figure 2. Skinfold thickness (e.g. triceps skinfold thickness) allows estimation of total body fat using previously published empirical equations. Mid-arm circumference is used to 77
When Does Malnutrition Become a Risk?
Fat mass (skinfold thicknesses, BIA, DXA) Fat free mass (arm circumference, BIA)
Non mineral fat free mass (DXA)
Body cell mass (BIA)
Extracellular fluids (BIA?)
2 compartments
Bone mineral content (DXA)
Extracellular solids (BIA?)
3 compartments
4 compartments
Fig. 2. The body has been separated into compartments which can be measured routinely by the methods shown in italics.
calculate lean tissue and muscle mass. Both methods are inexpensive and relatively easy to perform but are limited by inter-observer variability, variations in skin compressibility, abnormalities of fat and water distribution in disease (e.g. lipodystrophy, ascites, edema), and inaccessibility of sites (e.g. intensive care patients in a supine position). Their precision is 5% maximum when performed by trained researchers. BIA is easy, fast, noninvasive and inexpensive. Self-adhesive electrodes are put on the right hand, wrist, ankle and foot. A generator applies an alternating current (0.8 mA, 50 kHz). Resistance and reactance are measured and converted to total body water, FFM and skeletal muscle mass by validated equations. The results are influenced by posture, hydration and ion status, body temperature and body geometry. Its reproducibility is 0.8–4.2% for measures of FFM. Multi-frequency bioimpedance analysis has been used to differentiate between total body water and its extra- and intracellular subcompartments. The theory is that impedance at low frequencies is correlated with extracellular water and impedance at high frequencies with total body water. However, its use to assess BCM in patients is still not accepted because present equations yield variable results when validated against dilution methods [22]. Disease changes the total protein content and composition as well as the membrane capacitance which affect the resistance. As a consequence, equations to determine FFM are invalid [22]. DXA is a scanning technique measuring FM, FFM and bone mass. Although not yet considered the ‘gold’ standard, it is one of the reference methods for measuring body composition. It uses X-rays of two different energy levels. The attenuation of these X-rays 78
When Does Malnutrition Become a Risk? Table 2. Prevalence of malnutrition in hospitalized adults (1990–2004)
Medicine
Surgery
Geriatrics
References
Year
Total patients
% malnourished
Isabel [36] Kyle et al. [19 ] Waitzberg [37] Edington [38] McWhirter [39] Larsson [40] Willard [41] Kyle et al. [19] McWirther [39] Postma [42] Hall [43] Isabel [36] Guigoz [44] Thomas [45] Sullivan [46] Contans [47] Füllop [48] Larsson [40]
2003 2001 2001 2000 1994 1993 1990 2001 1994 1993 1990 2003 2002 2002 1994 1992 1991 1990
NA
52 36 33 20 45 29 32 31 33 23 29 53 53 29 38 37 34 29
618 4,000 850 300 382 200 253 200 422 367 NA ⬎10,000 837 110 324 552 500
NA ⫽ Not available.
as they pass through the body allow differentiation of bone, FM and FFM. For measuring FFM, the accuracy of DXA is 2–3% and its reproducibility 1.5–3%, for bone mass, its accuracy is 30 g and the reproducibility for bone density is 0.8% [23]. Other methods are available for measuring body composition, such as total body potassium, isotope dilution of deuterate-, titrate- or 18-oxygen-enriched H2O, CT scans, magnetic resonance and ultrasonography, but they are not practical for use in clinical routine. The distinction between starvation and cachexia can be made according to the changes in body composition (table 1), plasma albumin and the clinical history of the patient. In summary, BCM is the crucial determinant of depletion. However, at present, there are no reliable methods to measure BCM in clinical routine. Therefore, in the literature, several surrogates of BCM are used.
Prevalence of Malnutrition The prevalence of malnutrition among hospitalized patients varies between 20 and 50%, depending on the methods used to assess malnutrition, the age and the primary disease of the subjects (table 2). The occurrence of malnutrition in the future will probably even rise because of aging of the population, increased frequency of chronic diseases, and the ability to perform more and more invasive and prolonged medico-surgical interventions. 79
When Does Malnutrition Become a Risk? In practice, malnutrition is often only detected upon hospital admission when complications necessitating hospitalization have occurred. However, the WHO defines malnutrition as a risk to health, or in other terms, as a ‘factor that raises the probability of an adverse health outcome’. This definition implies that malnutrition should be diagnosed without the presence of malnutrition-related complications. While cachexia can be expected in many subjects who will undergo major surgery, the detection of malnutrition in outpatients with chronic diseases remains problematic. Therefore, the prevalence of malnutrition mentioned above may in fact only indicate the prevalence of malnutrition-related complications.
Health Risks Related with Malnutrition Most of the studies regarding malnutrition-related risks have been performed in hospital settings and thus do relate to cachexia. Malnutrition leads to muscle atrophy and weakness. It mainly decreases peripheral skeletal muscle mass, which induces loss of mobility, an increased risk of falling, and probably also plays a role in equilibrium disorders. In addition, it also affects respiratory muscle mass. This translates clinically into impaired exercise tolerance [24], forms an impediment to weaning patients from ventilatory support and an impossibility to adequately mobilize patients postoperatively. This may also constitute an increased risk of community-acquired and nosocomial pneumonia [25]. In the cardiovascular system, malnutrition may lead to abnormalities of mitral valve motion, reduction in left ventricular mass and filling, and increased peripheral resistance [4]. Besides the effects on muscle, malnutrition delays wound healing and leads to immune dysfunction, which both favor infection and promote further catabolism [5]. It has been hypothesized that diminished muscle mass and immune dysfunction are directly linked because muscle is an important supplier of substrate serving to fuel the immune response. Examples of these substrates are glutamine, alanine and possibly arginine. As a consequence of these complications, malnutrition reduces quality of life. It thereby contributes to an increase in the length of hospital stay and rehabilitation, which raises health-related costs, as well as mortality.
Cutoff Points for Health-Related Risks It is difficult to determine at what point malnutrition becomes a risk for health because malnutrition is diagnosed by parameters which are in turn influenced directly by other factors such as primary diseases, drugs and physical activity. Muscle atrophy decreases quality of life and increases mortality. Death results from a loss of 30–50% of BCM. A loss of BCM has been associated with 80
When Does Malnutrition Become a Risk? fatigue, global distress, depressive symptoms and reduced life satisfaction in, e.g., HIV patients [6]. A parallel decrease in muscle mass, physical performance and quality of life has been described in patients with renal transplantation [7], post-poliomyelitis syndrome [26], and elderly subjects [27]. In contrast, treatments which increase FFM and weight, as for example anabolic agents, cytokine inhibitors, and resistance training, have been shown to increase quality of life in diseased subjects. The question arises whether a subject with decreased BCM due to cachexia (depletion due to inflammation) responds similarly to a metabolic challenge than a subject with simple starvation. This does not appear to be the case. Indeed, inflammation impairs the formation of healthy granulation tissue, healthy re-epithelialization, the growth of hair, nails and skin, the healing of anastomosis and the production of fibrin. Furthermore, preoperative albumin, which is always low in cachexia but not in simple starvation, has been shown to correlate inversely with complications, length of stay, postoperative stay, intensive care unit stay, mortality, and resumption of oral intake [28]. A low subjective global assessment, indicating depletion, has also been related to a higher risk of infectious complications, respiratory failure and cardiac failure, as well as wound dehiscence [29]. Only very few studies have related changes in anthropometrics or body composition to mortality, physical disability or impairment of physical functioning defined as limitations in mobility performance. Engeland et al. [30] showed a U-shaped or J-shaped relationship between body mass index (BMI) and mortality. Mortality was lowest with a BMI of 22–25 kg/m2 and appeared to rise sharply with a BMI below 18 kg/m2 in men and women. The optimal BMI increased between the ages of 20–29 and 70–74 years from 21.6 to 24.0 kg/m2 in men and from 22.2 to 25.7 kg/m2 in women [30]. Rosenbaum et al. [31] determined the 95% confidence intervals for normal changes in body weight and found that a normal rate of weight loss would be ⫾2% in 1 month, ⫾3.5% in 3 months, ⫾5% in 6 months and ⫾10% in 1 year. Many consider weight loss beyond these limits as malnutrition. However, small to moderate weight losses are not always related to changes in function and fatigue and quality of life. Minor weight changes are not a good indicator of malnutrition but large changes should be considered seriously. In 1936, Studley [32] observed a mortality rate of 33% in patients who underwent elective surgery for peptic ulcer disease and lost 20% of their body weight postoperatively. In contrast, a loss of 10% of body weight was not associated with a higher mortality [32]. Nevertheless, modifications in body composition and function assessed by quality of life may better reflect the consequences of malnutrition. This is also apparent if one realizes that sick patients may show an increase in weight, which is merely caused by an accumulation of body water (edema). Janssen et al. [8] recently published skeletal muscle cutoff points for physical disability in the elderly. Physical disability was defined as an impairment 81
When Does Malnutrition Become a Risk? or health problem requiring help for eating, bathing, dressing or moving around at home or for household tasks, necessary business or shopping. They measured skeletal muscle mass with BIA and normalized it for height (skeletal muscle mass index ⫽ SMI). They demonstrated moderate and high risk of physical disability with an SMI ⬍6.75 kg/m2 in women and ⬍10.75 kg/m2 in men. Other studies have reported physical disability [9] or decreased physical functioning [10] but they used an arbitrary cutoff point to define sarcopenia (SMI ⬍2 standard deviations below the mean of young adults).
Detection of Malnutrition in Outpatients Malnutrition is difficult to diagnose in outpatients, especially in patients with short bowel syndrome. They may report loss of body weight and function which can reflect acute metabolic stress or simply a metabolic adaptation of the organism to a shorter small bowel. Indeed, after small bowel resection, body weight and cell mass decrease in response to reduced intestinal energy absorption and stabilize when energy uptake balances basal and exerciseinduced energy expenditure. It would be reasonable to assume that the body weight which is then reached determines the quality of life and the subsequent indication for nutritional support. However, in patients with a small bowel length between 48 and 60 cm who were followed for a mean of 418 days, the lower weights after surgery (47.6 ⫾ 7.8 kg) than at admission (53.5 ⫾ 2.1 kg) did not seem to affect quality of life [33]. Wilmore et al. [34] treated 45 patients with a jejunum-ileum length of ⬍50 cm and a portion of colon in continuity, with growth hormone, glutamine and a diet high in complex carbohydrates and low in fat for 4 weeks. They showed that independency of nutritional support was predicted by lower body weight, greater small bowel, and greater bowel length–body weight ratio [34]. The presence of terminal ileum or colon in continuity also increases the chance of independence from parenteral nutrition [35]. According to these studies, minor weight loss does not seem to determine quality of life or necessity of nutritional support. However, they did not report BMI or body composition or evaluated weight loss in function of BMI. We suggest that a given weight loss does not have the same consequences on quality of life in persons with low BMI and BCM than in those with high BMI and high BCM. Since no studies have yet addressed this issue in short bowel patients, we presently rely on our clinical experience to determine when a patient needs nutritional support. A low original BMI associated with weight loss, as defined by Rosenbaum et al. [31], and the appearance of physical disabilities (like impossibility to stand, walk or perform domestic tasks) are elements indicating the need for nutritional support. The cutoff points that are used to install artificial nutrition are variable because they are based on the subjective wellbeing of the patient and on his desire to benefit from nutritional support or not. 82
When Does Malnutrition Become a Risk? Conclusion Malnutrition leads to changes in body composition, especially in FFM, and body function. A decrease in FFM, which can be assessed by bioelectrical impedance, DXA, isotope dilution or measurement of intracellular potassium, affects muscle strength, decreases quality of life and physical functioning, and increases physical disability. Although it is generally accepted that severe malnutrition presents a health risk and is accompanied by a diminished quality of life, at present very few studies have tried to determine the level of malnutrition at which this risk is increased and quality of life decreases. Further studies are necessary to confirm such cutoff points in terms of body function, quality of life, complications and mortality rates. At present, decision regarding institution of artificial nutritional support in weight-losing patients are predominantly made intuitively on the basis of body weight and quality of life. References 1 Allison SP: Malnutrition, disease, and outcome. Nutrition 2000;16:590–593. 2 Barendregt K, Soeters PB: Stress starvation; in Sobotka L (ed): Basics in Clinical Nutrition. Prague, Galen, 2000. 3 Kotler DP: Cachexia. Ann Intern Med 2000;133:622–634. 4 de Simone G, Scalfi L, Galderisi M, et al: Cardiac abnormalities in young women with anorexia nervosa. Br Heart J 1994;71:287–292. 5 Felblinger DM: Malnutrition, infection, and sepsis in acute and chronic illness. Crit Care Nurs Clin North Am 2003;15:71–78. 6 Wagner GJ, Ferrando SJ, Rabkin JG: Psychological and physical health correlates of body cell mass depletion among HIV⫹ men. J Psychosom Res 2000;49:55–57. 7 Painter PL, Topp KS, Krasnoff JB, et al: Health-related fitness and quality of life following steroid withdrawal in renal transplant recipients. Kidney Int 2003;63:2309–2316. 8 Janssen I, Baumgartner RN, Ross R, et al: Skeletal muscle cutpoints associated with elevated physical disability risk in older men and women. Am J Epidemiol 2004;159:413–421. 9 Baumgartner RN, Koehler KM, Gallagher D, et al: Epidemiology of sarcopenia among the elderly in New Mexico. Am J Epidemiol 1998;147:755–763. 10 Melton LJ 3rd, Khosla S, Crowson CS, et al: Epidemiology of sarcopenia. J Am Geriatr Soc 2000;48:625–630. 11 Neary NM, Goldstone AP, Bloom SR: Appetite regulation: from the gut to the hypothalamus. Clin Endocrinol (Oxf) 2004;60:153–160. 12 Wray CJ, Mammen JM, Hasselgren PO: Catabolic response to stress and potential benefits of nutrition support. Nutrition 2002;18:971–977. 13 Preston T, Slater C, McMillan DC, et al: Fibrinogen synthesis is elevated in fasting cancer patients with an acute phase response. J Nutr 1998;128:1355–1360. 14 Ferrucci L, Harris TB, Guralnik JM, et al: Serum IL-6 level and the development of disability in older persons. J Am Geriatr Soc 1999;47:639–646. 15 Faintuch J, Soriano FG, Ladeira JP, et al: Refeeding procedures after 43 days of total fasting. Nutrition 2001;17:100–104. 16 Moses AW, Slater C, Preston T, et al: Reduced total energy expenditure and physical activity in cachectic patients with pancreatic cancer can be modulated by an energy and protein dense oral supplement enriched with n-3 fatty acids. Br J Cancer 2004;90:996–1002. 17 Rooyackers OE, Saris WH, Soeters PB, Wagenmakers AJ: Prolonged changes in protein and amino acid metabolism after zymosan treatment in rats. Clin Sci (Lond) 1994;87:619–626. 18 Roubenoff R: Catabolism of aging: Is it an inflammatory process? Curr Opin Clin Nutr Metab Care 2003;6:295–299.
83
When Does Malnutrition Become a Risk? 19 Kyle UG, Genton L, Slosman DO, Pichard C: Fat-free and fat mass percentiles in 5225 healthy subjects aged 15 to 98 years. Nutrition 2001;17:534–541. 20 Baumgartner RN, Koehler KM, Romero L, Garry PJ: Serum albumin is associated with skeletal muscle in elderly men and women. Am J Clin Nutr 1996;64:552–558. 21 Jeejeebhoy KN: How should we monitor nutritional support: Structure or function? New Horiz 1994;2:131–138. 22 Cox-Reijven PL, van Kreel B, Soeters PB: Bioelectrical impedance measurements in patients with gastrointestinal disease: Validation of the spectrum approach and a comparison of different methods for screening for nutritional depletion. Am J Clin Nutr 2003;78:1111–1119. 23 Pichard C, Kyle UG: Body composition measurements during wasting diseases. Curr Opin Clin Nutr Metab Care 1998;1:357–361. 24 Birmingham CL, Tan AO: Respiratory muscle weakness and anorexia nervosa. Int J Eat Disord 2003;33:230–233. 25 Laaban JP: Nutrition and chronic respiratory failure. Ann Med Interne (Paris) 2000;151:542–548. 26 Nollet F, Beelen A, Twisk JW, et al: Perceived health and physical functioning in postpoliomyelitis syndrome: A 6-year prospective follow-up study. Arch Phys Med Rehabil 2003; 84:1048–1056. 27 Iannuzzi-Sucich M, Prestwood KM, Kenny AM: Prevalence of sarcopenia and predictors of skeletal muscle mass in healthy, older men and women. J Gerontol A Biol Sci Med Sci 2002;57: M772–M777. 28 Kudsk KA, Tolley EA, DeWitt RC, et al: Preoperative albumin and surgical site identify surgical risk for major postoperative complications. JPEN J Parenter Enteral Nutr 2003;27:1–9. 29 Correia MI, Waitzberg DL: The impact of malnutrition on morbidity, mortality, length of hospital stay and costs evaluated through a multivariate model analysis. Clin Nutr 2003;22: 235–239. 30 Engeland A, Bjorge T, Selmer RM, Tverdal A: Height and body mass index in relation to total mortality. Epidemiology 2003;14:293–299. 31 Rosenbaum K, Wang J, Pierson RN Jr, Kotler DP: Time-dependent variation in weight and body composition in healthy adults. JPEN J Parenter Enteral Nutr 2000;24:52–55. 32 Studley H: Percentage of weight loss with physical impairment: A basic indicator of surgical risk in patients with chronic peptic ulcer. JAMA 1936;106:458–460. 33 Sales TR, Torres HO, Couto CM, Carvalho EB: Intestinal adaptation in short bowel syndrome without tube feeding or home parenteral nutrition: Report of four consecutive cases. Nutrition 1998;14:508–512. 34 Wilmore DW, Lacey JM, Soultanakis RP, et al: Factors predicting a successful outcome after pharmacologic bowel compensation. Ann Surg 1997;226:288–293. 35 Messing B, Crenn P, Beau P, et al: Long-term survival and parenteral nutrition dependence in adult patients with the short bowel syndrome. Gastroenterology 1999;117:1043–1050.
Discussion Dr. Steinhagen-Thiessen: If, at this stage now, you had to give us a good definition of malnutrition to make a huge register, what would you say? What parameters are absolutely necessary? Dr. Soeters: I think anthropometrics, weight loss. I would include hemoglobin, and I would certainly include albumin. Dr. Steinhagen-Thiessen: Not prealbumin? Dr. Soeters: No, it is a short-term risk. Prealbumin can change over night. Albumin gives a much better reflection of what has been going on over time. Dr. Steinhagen-Thiessen: These are also clinical parameters so they are not difficult to get. It is easy. Dr. Soeters: I am just mentioning those because they are easy to do. Dr. Buchman: You don’t think that the albumin status can change over night, especially postoperatively because of intravascular fluid shifts? Extracellular albumin
84
When Does Malnutrition Become a Risk? can decrease by 30% in patients who come in with a variceal bleed. At baseline their albumin is 2.8 g/l, and with gastrointestinal bleeding it is 1.9 g/l. The liver doesn’t just stop producing albumin, and there are tremendous intra- and extravascular fluid shifts. I think in the acute setting the serum albumin concentration is close to useless as a measure of nutritional status. It is a better indicator of inpatient mortality than anything else [1, 2]. Would you agree? Dr. Soeters: I am not saying that I use albumin as a nutritional indicator; I am saying that to calculate risk you need first of all to know what the body cell mass is. But if that is not possible you have to use weight loss and body weight or body mass index. But then in addition you need to know about albumin, and then if the body cell mass is low or anthropometrics are low and albumin is decreased, then the patient is in a bad shape. You should then try in some way to decrease the inflammatory state. I agree with you about let’s say 48-hour shifts, but in the intensive care patients I mentioned these changes took place over weeks, and then I think it is a very reliable measure. In the acute situation, just after an operation, after a really septic episode, then albumin by itself only says that there is an inflammation. Dr. Morley: I would like to suggest that for undernutrition the only useful parameters are anthropometrics, and that would be weight loss, perhaps skin-fold thickness and/or mid-arm circumference. All the other parameters that we have used classically, whether cholesterol, prealbumin, or albumin, have a cytokine component involved in their production. Dr. Soeters: You are saying the same thing as I am. Dr. Morley: So I would say if we were going to have a definition, the definition for undernutrition should exclude those factors, unless you measure something like C-reactive protein or serum albumin protein, to say that they are not elevated, in which case they suddenly all become very good parameters. I think where we get ourselves into trouble, particularly in acute hospitals, is that we tend to use things like the serum albumin as something that says now we should feed this person aggressively because it is low and clearly often that is not the major thing that that person needs at that time. Dr. Soeters: I tried to say that. You really want to know what the body composition is, and that is very difficult. You still have to guess in some way or use the easy parameters that are there. I was asked to talk about risk, and then the factors like albumin have to be added to give you an idea about increased risk. It doesn’t tell you much about body composition, it will only tell you if albumin drops, then it is very likely that body cell mass is also decreasing and you still have to deal with a catabolic patient. So I think we are saying the same thing. Dr. Ockenga: If you have already identified a patient as malnourished, as depleted, and he is coming to you for elective surgery, maybe a pancreas head resection, what would your concept be for this patient? Do you do any perioperative support or something else? Dr. Soeters: If I am dealing with a pancreatic patient or an esophageal patient, these are the patients who rapidly lose weight, not the colonic cancer patients in general, we would certainly use the preoperative time to replete them because they cannot be admitted immediately. They are generally in negative energy balance anyway, so we will try to replete them using a tube, a nasal tube if it is esophageal cancer for instance. We replenish the pancreatic enzymes in patients with pancreatic cancer, which already generally takes away part of their anorexia. We try to insert tube, intubate the tumor so that their bile will flow again, or we collect it and ask them to drink it or add it to the nasal tube so that bile and nutrition goes in. So those things we certainly do in this kind of patient, not in patients with non-obstructive pancreatic cancer without weight loss obviously.
85
When Does Malnutrition Become a Risk? Dr. Buchman: Do you ever use the delayed skin hypersensitivity response or even a total lymphocyte count as a measure of nutrition status? Although of course they have a number of things that can affect these besides nutrition. Dr. Soeters: No, we don’t use it but it is a pity that it has not been developed further. We have tried to do some of it by vaccination giving several types of antigens to assess the immune system via several pathways, but delayed cutaneous hypersensitivity had the same character and that is why I said, it is a functional measure. I don’t know why we left it and why we did not try to improve it. There is so much difficulty in assessing the immune status and immune competence. It is a relatively easy measure and it is the only measure that truly has at least some functional character. Lymphocytes are interesting because they are also always low in depleted patients, they are always low in patients with inflammation, so it is also a relatively reliable sign. It scored high in our study and in others as well. Dr. Bozzetti: As you know many surgical trials on perioperative nutritional support had some eligibility criteria as body weight loss, 10 or 15%. The level of serum albumin and total lymphocyte count are the most common parameters which have been used. In some of these trials the results of nutritional support were positive, in others negative. Quite recently Kudsk et al. [3] published their experience in Annals of Surgery and they were able to show that the same risk factor, i.e. weight loss and serum albumin, had a different impact in different patient populations. For instance in pancreatic cancer they predicted a higher rate of complications than in colonic cancer patients. So do you think that for any type of surgical operation we should have a set of markers which defines the risk and the potential benefit of nutritional support, because this would be the message of Kudsk et al. [3]? Dr. Soeters: We are becoming very clinical now. In pancreatic cancer, the obstructed pancreas and the lack of bile in the digestive tract cause an inflammatory state all by themselves. But if you replete the patients and re-infuse bile or allow bile to be added to the normal route, you are certainly doing something good. In colonic cancer, the low albumin in the inflammatory state generally has to do with the fact that you are dealing with a growing gangrenous carcinoma or with necrotic areas in the carcinoma, which is a different situation. So I think that you may have to separate these different cancers from each other. Dr. Labadarios: My point really continues on what Dr. Bozzetti just said. It is really reassuring to hear a surgeon not considering albumin as an index of nutritional status. At least in my experience, one can define the inflammatory response on clinical grounds reasonably well from serum albumin and at the same time serum cholesterol. The minute those two turn upwards you know the patient is improving. But in the acute situation as you were showing in your slide, and that relates to Dr. Bozzetti’s question, although you can have a measure of the intensity of the inflammatory response, can you really have a measure of undernutrition in that type of patient on the basis of weight loss? I don’t know how you determine weight loss in that type of patient with the likely huge third space that you showed us. I don’t know how you do anthropometrics in an edematous patient. Dr. Soeters: But that is the message of my talk that in acute disease there is such an enormous increase in the extracellular water compartment that total body water is not a valid measure for lean body mass, especially because the ratio between intracellular water and extracellular water is different, it is not a normal ratio. So you need to look at something different and then you come up with more sophisticated techniques. Dr. Labadarios: What I am trying to say is that are we need to be careful when trying to apply the same parameters in different clinical settings?
86
When Does Malnutrition Become a Risk? Dr. Soeters: I showed a patient not because I applied those parameters in her, but to show that the truly nutritional state parameters cannot be applied to her. Dr. Cynober: I have several comments. First, I don’t think that it is a problem of clinical setting; it is a problem of the goal. In many cases simple diagnosis of malnutrition does not require anything, even body composition. Now if you are looking at patients at risk it is excellent because it takes into consideration, as mentioned by Dr. Soeters, all the components of risk. For me what is most important in terms of follow-up of patients is to follow the parameters such as prealbumin. When it is repeated with time, taking into consideration the inflammation component, and has been measured in parallel, C-reactive protein is absolutely excellent even in burned patients. Now if we are discussing the follow-up of patients, especially critically ill patients, since the key is protein depletion, there is nitrogen balance which is a very good parameter provided that you are really measuring total nitrogen in most cases; of course there are some limitations. Now that we have entered the third millennium, there are specific criteria to look at in patients at risk. Since we know now that perhaps 20% of the population in Western countries has a gene polymorphism, especially for proinflammatory cytokines and also for nitric oxide synthesis, it becomes feasible, not easy but feasible, to determine the gene polymorphism for instance for tumor necrosis factor or IL-6 in patients admitted to the intensive care unit. I am certain that really determining patients at risk, healthy and unhealthy patients, will be the key for the future. Dr. Soeters: My talk was meant to try to be practical and to really address the issues that we have in daily practice. The polymorphism of tumor necrosis factor and IL-6 is very well taken and that is the future. But regarding nitrogen balance it has become somewhat fashionable to say that nitrogen balance doesn’t tell you anything. Regarding protein turnover measurements, I think it depends very much on whether it is done with phenylalanine or leucine. Different measures occur and they are not practical either. I agree with you that nitrogen balance is perhaps laborious and is not so easy for the nurses in intensive care. It may be easier especially in patients who do not have wounds, who do not have great losses outside their urine, stools, and in whom you can adequately measure how much they eat or how much parenteral nutrition is given. Then you can calculate a relatively reliable nitrogen balance, but it takes work. I agree with you that it would be a very good measure to see if the patient is improving or not, but as long as it is so laborious I think it will not be very popular in the intensive care units and certainly under budget restraints. That is why we use albumin, again not as a nutritional parameter but rather as a sign of becoming better, and if the patients get better I am convinced that they are receiving adequate nutrition either enterally, parenterally, or both. Then I am sure that these patients are becoming anabolic. But again, it is well taken. I think nitrogen balance would still be a good measure. Dr. Cynober: What do you mean laborious for the nurse? Do you mean urinary collection or the determination of nitrogen balance? Dr. Soeters: Very rigid discipline must be installed in the intensive care unit. Urine, excretions and wound fluid should be sampled; everything flowing in should be calculated. You should preferably refrain from infusing albumin or plasma or erythrocytes because you don’t know how much that is going to influence nitrogen balance. It is laborious in the sense that it is work and it needs discipline, but it can be done. Dr. Fürst: All stress situations are associated with a decrease in membrane potential. In the late 1970s and early 1980s we measured muscle biopsies, the actual membrane potential, and it was decreased from 87 to 70. Dr. Soeters: Columbia and Sweden.
87
When Does Malnutrition Become a Risk? Dr. Fürst: Columbia and Sweden, yes. Now even in a moderate stress situation like immobilization, we could measure significantly decreased intracellular water. The water distribution is highly affected by changes in membrane potential together with a redistribution of electrolytes like potassium and magnesium. Today we might have the possibility of measuring intracellular water with the multi-frequency bioimpedance. What do you think about measuring intracellular water as a marker of stress? Dr. Soeters: I would like to look at the sick cell. In fact the organs, the parts of the body that are ill are cells. They get leaky, they open up channels, they pump harder, they try to maintain their membrane potential and they leak, and that leads to a different distribution of ions outside and inside the cell. So if we were better able to measure for instance membrane potential in an easy way, it is not easy to do, certainly not in patients, then you might have a disease indicator which is much more reliable than anything else. Dr. Fürst: A membrane potential can be measured very easily. For instance in Texas, there is a quite easy routinely manageable method. You were looking for a very routine method and bioimpedance is such a method. We measured intracellular metabolites, we measured the energy status of the muscle, there are significant changes but they are difficult, they are not routinely manageable, but today bioimpedance is indeed available. Dr. Soeters: But you have to validate it.
References 1 McCluskey A, Thomas AN, Bowles BJ, Kishen R: The prognostic value of serial measurements of serum albumin concentration in patients admitted to an intensive care unit. Anaesthesia 1996;51:724–727. 2 Friedamn PJ: A prospective comparison of methods to identify lethal wasting malnutrition. Nutr Res 1986;6:193–146. 3 Kudsk KA, Tolley EA, DeWitt RC, et al: Preoperative albumin and surgical site identify surgical risk for major postoperative complications. JPEN J Parenter Enteral Nutr 2003;27:1–9.
88
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 89–102, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
What Are the Goals of Nutritional Support? The Example of Home Enteral Nutrition
Xavier Hébuterne and Stéphane M. Schneider Gastroenterology and Clinical Nutrition, Archet 2 Hospital, Nice, France
Artificial nutrition is a supportive medical therapy aiming at achieving predefined objectives, which should be adjusted for changing clinical situations [1]. The goals of nutritional support must be clearly predefined within a global therapeutic plan. First of all it is necessary to determine whether nutritional support is a medical treatment or basic human care. If it is considered a medical treatment, the goal of this treatment is to improve some parameters related to disease progression and/or malnutrition (considered as a disease). For the latter the goal of nutritional support is only to provide nutrients to patients unable to reach their nutritional requirements. However, it appears that nutritional support per se may be harmful and it seems inappropriate to provide artificial nutrition when the burdens of this treatment outweigh its benefits. For this reason we believe that nutritional support must be considered as a treatment. Therefore, clear and loyal evaluation of the risk/benefit ratio must be done. A recent survey from the ESPEN Home Artificial Nutrition Group [2] has demonstrated that in Europe home enteral nutrition (HEN) is mainly prescribed for dysphagic patients with neurological disorders or cancer, using a standard feed via percutaneous endoscopic gastrostomy (PEG). However, there were important differences between the countries regarding the underlying diseases and age of patients [2]. In the two national registers available in Europe, similar differences were observed. In the UK, during the period 1996–1999, only 146 patients with head and neck cancer where on HEN, compared to 5,037 with cerebrovascular disease [3]. In Italy, during the period 1992–1999, of the 7,111 patients examined 1,900 suffered from head and neck cancer and 1,647 from cerebrovascular disease [4]. Moreover, the use of HEN is approximately two to three times lower in Europe than in the US [2]. There is no evidence that the differences observed between European centers and between Europe and the US are 89
What Are the Goals of Nutritional Support? related to differences in expertise in artificial nutrition, training centers, or to local variations in the incidences of certain diseases. Inter-country differences appear to be due to recognized differences in medical practice. We believe that the clear determination of the goals of nutritional support may be helpful to evaluate the rationale of starting nutritional support at a patient level. Several goals may be achieved; some of them are necessary, other may be optional. The main questions should be the following: (1) Will nutritional support be well tolerated by my patient? (2) Will nutritional support improve or sustain the nutritional status of my patient? (3) Will nutritional support improve or sustain the quality of life of my patient? (4) Will nutritional support improve the outcome and extend the survival of my patient? As there are many clinical situations we will take the example of HEN to answer these questions.
Will HEN Be Well Tolerated by My Patient? Tolerance of artificial nutrition is essential because adverse effects may affect the predefined goals of nutritional support. Before discharging a patient from the hospital ward on long-term HEN, it is essential to verify the tolerance of nutritional therapy in the hospital during at least 7 days. In some cases, enteral nutrition (EN) and/or gastrostomy are responsible for complications that alter the prognosis and reduce the expected benefits. Procedurerelated mortality due to PEG is in the region of 0.5–1%, with major and minor complications occurring in about 9 and 6% of cases [5]. Major complications include wound-related problems (major infections, septicemia, dehiscence), aspiration, peritonitis, hemorrhage, tube dislodgement and respiratory failure. Minor complications include stomal soreness and minor infections, stomal leakage, tube blockage and gastrointestinal upset. In a recent study [6] we prospectively assessed complications and 1-month mortality after PEG in patients selected by a nutrition team. 106 PEG tubes were placed in 70 men and 36 women aged 63 (22–93) years. The indications were dysphagia in 92 patients (53 neurological disorders, 37 head and neck tumors, 2 traumas), and anorexia in 14 patients. There were 33 minor complications in 30 patients, which all resolved under symptomatic treatment: abdominal pain (n ⫽ 11); local infection (n ⫽ 6); hyperthermia (n ⫽ 5); peristomal pain (n ⫽ 3); leakage (n ⫽ 3); intestinal obstruction (n ⫽ 3); buried bumper syndrome (n ⫽ 1), and tube dislodgement (n ⫽ 1). However, 13 major complications occurred in the first 30 days in 13 patients (12.3%). Ten patients had pneumonia (aspiration pneumonia was confirmed in 3 cases), which was fatal in 6 cases and resolved in 4 cases. Two patients had peritonitis 90
What Are the Goals of Nutritional Support? Table 1. Factors associated with high risks of complications and/or early mortality after PEG
Age ⬎80 years Dementia BMI ⬍16.5 Recent bronchopulmonary infection Albuminemia ⬍30 g/l
(fatal in 1 case and resolved with surgery in 1 case), and 1 patient had severe gastric bleeding which resolved under medical treatment. Sixteen patients (15.1%) died during the 1st month after the procedure. In 9 cases death was probably due to the initial disease whereas in 7 cases it was possibly induced by intervention (in 6 cases due to nosocomial pneumonia and in 1 case due to peritonitis). In another study PEG was proposed to 46 nursing home residents with a mean age of 74 years (52% had dementia). Authors observed 34 complications related to tube feeding in 16 patients (35% of patients) with tube obstruction (30% of patients), tube migration (17%), buried bumper syndrome (6.5%), wound infection (4.3%), and aspiration pneumonia occurred in 20% of the patients [7]. In a 11-month period 70 tube-fed patients aged 65–95 years were studied prospectively. In patients with a nasogastric tube, agitation and self-extubation occurred in 67% and aspiration pneumonia in 43%. In PEG patients the most common early problem was aspiration (44%) [8]. Tolerance of EN and/or PEG is greatly dependent on the condition of the patient. Several factors have been shown to be associated with a high level of complications and early mortality after PEG (table 1) and should be considered. In cases of severe malnutrition it may be useful to feed the patient via a nasogastric tube during 2–3 weeks prior to inserting a PEG tube.
Will HEN Improve or Sustain the Nutritional Status of My Patient? HEN is generally considered for patients with oral failure. We recommend oral failure as a new criteria for choosing and deciding the enteral route of nutrition. To date, symmetrical to intestinal failure, we define oral failure as the inappropriate and involuntary reduction in the oral intake below the minimal amount necessary for the maintenance of energy-protein equilibrium. There is no doubt that in a patient with severe or total dysphagia EN will be able to provide nutrients and sustain nutritional status indefinitely. In our experience [9] 5.5% of HEN patients are indefinitely dependent and maintain their nutritional status with HEN (table 2). Nutritional support is clearly indicated when adequate food intake is not possible for long periods. In patients with permanent neurological impairment, or oropharyngeal dysfunction, and in premature infants, long-term nutritional support is needed to prevent death from starvation. 91
What Are the Goals of Nutritional Support? Table 2. Long-term outcome of home enteral nutrition patients
Patients n % Duration of of HEN, days1 Age, years1
Indefinitely HENdependent
Resumed full oral nutrition
Died during the 1st month of HEN
Died after the 1st month of HEN
Off HEN for other reasons
23 5.5 1,854 ⫾ 827
136 32.6 111 ⫾ 176
84 20.2 13 ⫾ 8
146 35.0 219 ⫾ 257
28 6.7 360 ⫾ 453
50.3 ⫾ 28.1
54.0 ⫾ 28.1
76.0 ⫾ 17.0
70.0 ⫾ 19.0
58.1 ⫾ 24.2
⫾ SD. Adapted from Schneider et al. [9]. 1Mean
The nutritional effect of EN is well demonstrated. Cyclic EN has been shown to improve nutritional parameters [10] as well as body composition [11]. This effect has also been observed in patients with moderate intestinal failure such as gastrectomy or pancreatectomy [12]. In children with Crohn’s disease, it has been shown that home nocturnal nasogastric feedings can achieve a dramatic improvement in weight gain and linear growth [13]. However, the nutritional effect of EN is dependent on the patient’s condition. In a study of 97 middle-aged and elderly undernourished patients we have demonstrated that cyclic EN (with similar protein and energy amounts) was more effective in younger patients than in the elderly and that the recovery from malnutrition was more difficult in the elderly than in young people [10]. Moreover, in the elderly, after 4 weeks of intensive enteral feeding, the probability of not failing (to be alive and in remission) at 1 year was only 25% (95% confidence interval 1–52%) when during the last week of re-nutrition, oral energy intakes were lower than the resting energy expenditure, and 80% (95% confidence interval 65–95%) when oral energy intakes were higher than the resting energy expenditure (p ⫽ 0.0001; fig. 1). The practical consequences of this finding are that if a patient is able to resume normal oral intake during EN, it probably will not be necessary to continue nutritional support. However, if oral intake is still under the requirements after a period of 2–4 weeks of tube feeding, the prolongation of EN is mandatory. In this case a PEG should be discussed with the patient and/or his/her family [14]. Restoration of normal oral intake is an important goal in tube feeding. In our experience [9] 33% of HEN patients were able to resume full oral nutrition after a mean of 6 months HEN (table 2). If we analyze the long-term outcome of elderly patients on HEN for primary or secondary anorexia (without dysphagia or digestive disease), 56% of them will resume full oral nutrition 92
% of patients alive and in remission
What Are the Goals of Nutritional Support? 100 Oral intake⬎ REE 80 60
p⬍ 0.0001
40 Oral intake⬍ REE 20 0 0
10
20
30
40
50
Weeks
Fig. 1. Kaplan-Meier analysis of recurrence or death after cyclic enteral nutrition in the elderly patients with an oral intake higher than the resting energy expenditure (oral intake ⬎REE) or lower than the resting energy expenditure (oral intake ⬍REE) during the last week of refeeding. Adapted from Hébuterne et al. [2].
and in this case, the probabilities of being alive 1 and 5 years after initiation of HEN were 65 and 43%, respectively.
Will HEN Improve or Sustain the Quality of Life of My Patient? Quality of life has become a major concern in the treatment of patients with chronic diseases [15]. Little information is available concerning HEN patients. Although it can be assumed that HEN improves quality of life by maintaining patients at home, where premorbid landmarks and the family are present, the resultant alteration in their relations to food and dependency on a feeding tube and an infusion pump can be expected to deteriorate their quality of life. In a recent study we have evaluated the quality of life of 38 patients aged 56 ⫾ 5 years who had been on HEN for more than 25 ⫾ 5 months [16]. Analysis of generic questionnaires (SF-36 and EuroQol questionnaires) revealed poorer quality of life parameters in comparison to a general population. Nevertheless, the patients’ subjective assessment of the changes in their quality of life since beginning HEN was generally good, with most patients reporting improved or stable mental and physical well-being. In another study we evaluated the quality of life of 22 patients with head and neck cancer or neurological diseases who were discharged from hospital with a PEG. They were evaluated at the time of discharge and 3 months later [17]. Both mental and physical health slightly improved between the two evaluations (table 3). In 39 consecutive head and neck cancer patients quality of 93
What Are the Goals of Nutritional Support? Table 3. Effects of home enteral nutrition on the quality of life of 22 patients
MCS (n ⫽ 50 ⫾ 10) PCS (n ⫽ 50 ⫾ 10) EQ-5D (n ⫽ 0.85 ⫾ 0.25) VAS (n ⫽ 82.5 ⫾ 17.0)
Evaluation at discharge
Second evaluation
Evolution, %
p
35.1 ⫾ 10.9 33.4 ⫾ 10.0 0.49 ⫾ 0.39 50.5 ⫾ 25.0
39.9 ⫾ 13.4 34.6 ⫾ 7.3 0.59 ⫾ 0.35 54.7 ⫾ 22.9
⫹14.7 ⫾ 27.6 ⫹8.1 ⫾ 24.2 ⫹23.6 ⫾ 59.5 ⫹17.3 ⫾ 40.1
0.0001 0.006 0.07 0.001
MCS ⫽ Mental component scale; PCS ⫽ Physical component scale; EQ-5D ⫽ EuroQol 5-item scale; VAS ⫽ EuroQol visual analog scale; n ⫽ normal value. Adapted from Schneider et al. [17].
life was evaluated after discharge and 3 weeks later using a self-administered questionnaire of the EORTC. Overall, the global health status and quality of life scale score slightly improved but some patients experienced psychosocial distress [18]. In an interesting study, Loeser et al. [19] showed that quality of life was reduced in HEN patients and that part of this effect was explained by malnutrition. In 28 competent and 28 non-competent patients the Karnofsky index improved in 4 months: physical functioning improved and fatigue decreased. Interestingly 50% of the non-competent patients became competent. To determine whether HEN merely prolongs a life of poor quality or can really improve quality of life, a longitudinal and multicenter study of a larger number of patients on HEN is needed. For a given patient the answer to this question may be yes if HEN is well tolerated and if it improves nutritional status. If HEN is not well tolerated it will probably alter the quality of life of the patient. If HEN does not improve nutritional status but only sustains it, the burdens of HEN may alter quality of life and should be evaluated. We recommend the routine and longitudinal evaluation of quality of life with a tool as simple as a visual analogical scale in all patients on HEN. Will HEN Improve Outcome and Extend Survival of My Patient? This is probably the main question when considering a patient for HEN. The life expectancy of HEN patients is generally poor [20], suggesting that in many cases HEN has no effect on the underlying disease (fig. 2). In our overall experience, HEN patients had a poor outcome, and the probabilities of being alive 1 month, 1 year and 5 years after initiation of HEN were only 80.0, 41.7 and 25%, respectively [9]. An explanation for this poor outcome is the age of these patients. For our overall population, the probabilities of being alive at 1 year were 88% for children, 47% for patients between 16 and 70 years, and 30% for elderly patients over 70 years of age (fig. 3). Moreover, age over 70 was an independent factor influencing mortality. This poor outcome was also influenced by the severity of the underlying disease (cancer, 94
What Are the Goals of Nutritional Support?
Oropharyngeal malignancy Stroke with dysphagia Dementia Miscellaneous (e.g. head injury, motor neurone disease)
Probability of survival
100
%
75
50
Group 3 vs other groups p ⬍0.0001
25
0 0
6
12
18
24
30
36 42 Months
48
54
60
66
72
Fig. 2. Survival probability of patients on home enteral nutrition as a function of underlying diseases. Adapted from Sanders et al. [20].
stroke, amyotrophic lateral sclerosis, etc.), and most deaths were ascribed to the primary disease. In the United States, 1 year after initiation of HEN, 48% of patients with neurological swallowing disorders and 59% of cancer patients had died [21]. Less than one third of HEN patients resumed oral nutrition and more than 50% died during HEN [9]. Despite the selection of candidates for HEN by a professional nutritional support team, approximately 20% of patients die during the first month of HEN. In these patients HEN probably provides no benefit (terminal patients) or is even harmful (demented patients). The selection process is rigorous; however, it may in fact suffer from our overoptimism. Indeed, a recent study evaluating the physicians’ prognostic accuracy in terminally ill patients revealed overoptimistic predictions of survival in 63% of cases, with only 20% of accurate predictions [22]. Clearly, outcome and life expectancy on HEN depend on the patient’s condition and many data are now available. In a recent study, PEG was performed in 23 patients with dementia; 18 other patients met the medical criteria for PEG tube placement, but surrogates refused placement. The median survival for the 23 patients who underwent PEG was 59 days; the median survival for the 18 patients who did not undergo PEG was 60 days [23]. In another study the 6-month mortality of 55 demented elderly with PEG was 44%, and 26% among 33 controls comparable in age, gender, and comorbidities [24]. In this situation we believe, as others [25], that a comprehensive, motivated, conscientious program of hand feeding is the proper treatment. Tube feeding might be considered as an empirical treatment if the family is clearly advised that the best evidence suggests it will not help. In elderly patients with dysphagia, some recent studies suggest that tube feeding might be helpful. Rudberg et al. [26] compared the mortality outcomes of nursing home patients with or without tube feeding placed at the 95
What Are the Goals of Nutritional Support?
1 ⬍16 yr
0.8
0.6
0.4
16–17 yr
0.2 ⬎70 yr 0 0
20
40
60
80
100
Time (months)
Fig. 3. Survival probability of patients on home enteral nutrition as a function of age. Adapted from Schneider et al. [9].
time that they became unable to feed themselves and who had swallowing disorders. Estimated survival at 1 year was 39% for those without tube feeding and 50% for those with feeding tube. The precocity of nutritional intervention and the route of EN may also affect outcome and mortality. In patients with dysphagic stroke, mortality at 6 weeks was significantly lower in the PEG group (12%) compared with patients fed via a nasogastric tube (57%) [27]. In another study, a group of amyotrophic lateral sclerosis (ALS) patients were offered PEG when weight loss exceeded 5% of normal body weight or if symptomatic dysphagia occurred. Approximately half of the patients rejected the offer of a PEG and thus constituted a self-selected untreated control group. The others received EN via a PEG. Both groups were similar in terms of ALS severity, body mass index, percent of weight loss, and forced vital capacity. Patients receiving HEN maintained body weight, whereas the untreated control subjects experienced a progressive weight loss. Approximately 40% of PEG-fed patients were alive at 2 years, whereas survival in controls was only 5% [28]. Another frequent situation is patients with head and neck cancer. Aggressive surgical resection followed by soft tissue and bone reconstruction associated with radiotherapy and chemotherapy is the gold standard for advanced cancer. However, the incidence of complications is high and reaches 20–50%. Studies have reported malnutrition in 35–50% of head and neck patients at the time of 96
What Are the Goals of Nutritional Support? diagnosis and malnutrition is recognized as a factor of poor prognosis in cancer treatment-related morbidity and mortality. The surgical treatment includes tracheostomy, glossectomy, mandibulectomy, surgery of the palate, and total or partial laryngectomy. In addition, oropharyngeal function can be hindered further by radiation and chemotherapy, which cause mucositis, xerostomia, dysgeusia, nausea, vomiting and anorexia [29]. It has been shown that the incidence of severe weight loss during radiotherapy is about 33% with a 10% rate of rehospitalization. Postoperative tube feeding significantly reduces the incidence of weight loss and hospitalization during radiotherapy [30]. The effectiveness of early nutritional intervention in oropharyngeal cancer patients undergoing radiotherapy has been confirmed in another study [31]. In head and neck cancer patients at high risk of malnutrition we propose PEG before surgery/radiotherapy/chemotherapy in order to reduce weight loss and complications and therefore to optimize cancer treatment. In conclusion, the goals of HEN are dependent of the patient’s condition and the medical situation. In all cases tolerance of EN is an issue. In patients at high risk of complications such as aspiration, EN should not be proposed. If EN is well tolerated it will be able to maintain or to improve nutritional status. In patients with severe oral failure due to dysphagia sustaining nutritional status extends life duration. In other cases evaluation of the effects of HEN on quality of life is certainly necessary. In some cases good evidence exists in the literature, suggesting that EN may have a positive (dysphagia after stroke, ALS, head and neck cancer) or a negative (dementia) effect on outcome. On every occasion, questions on the goals of nutritional support should be clearly asked and discussed with the patient and/or his/her family/ caregivers in order to evaluate the risks/benefits ratio of the treatment. References 1 Planas M, Camilo ME: Artificial nutrition: Dilemmas in decision-making. Clin Nutr 2002; 21:355–361. 2 Hébuterne X, Bozzetti F, Moreno Villares JM, et al: Nutrition Working Group. Home enteral nutrition in adults: A European multicentre survey. Clin Nutr 2003;22:261–266. 3 Elia M, Stratton RJ, Holden C, et al: Home enteral tube feeding following cerebrovascular accident. Clin Nutr 2001;20:27–30. 4 Gaggiotti G, Orlandoni P, Ambrosi S, Catani M: Italian home enteral nutrition register: Data collections and aims. Clin Nutr 2001;20(suppl 2):69–72. 5 Skelly RH: Are we using percutaneous endoscopic gastrostomy appropriately in the elderly. Curr Opin Clin Nutr Metab Care 2002;5:35–42. 6 Schneider SM, Arab K, Filippi J, et al: One-month mortality after percutaneous endoscopic gastrostomy: Effects of screening on survival factors (abstract). Gastroenterology 2002;122 (suppl):T13. 7 Kaw MK, Sekas G: Long-term follow-up of consequences of percutaneous endoscopic gastrostomy tubes in nursing home patients. Dig Dis Sci 1994;39:738–743. 8 Ciocon JO, Silverstone FA, Graver LM, Foley CJ: Tube feedings in elderly patients. Indications, benefits, and complications. Arch Intern Med 1988;148:429–433. 9 Schneider SM, Raina C, Pugliese P, et al: Outcome of patients treated with home enteral nutrition. JPEN J Parenter Enteral Nutr 2001;25:203–209.
97
What Are the Goals of Nutritional Support? 10 Hébuterne X, Broussard JF, Rampal P: Acute renutrition by cyclic enteral nutrition in elderly and younger patients. JAMA 1995;273:638–643. 11 Hébuterne X, Péroux JL, Schneider SM, Rampal P: Effects of refeeding by cyclic enteral nutrition on body composition: Comparative study of elderly and younger patients. Clin Nutr 1997;16:283–289. 12 Hébuterne X, Vaillon F, Péroux JL, Rampal P: Correction of malnutrition following gastrectomy with cyclic enteral nutrition. Dig Dis Sci 1999:44;1875–1882. 13 Aiges H, Markowitz J, Rosa J, Daum F: Home nocturnal supplemental nasogastric feedings in growth-retarded adolescents with Crohn’s disease. Gastroenterology 1989;97:905–910. 14 Kirby DF, DeLegge MH, Fleming CR: American Gastroenterological Association technical review on tube feeding for enteral nutrition. Gastroenterology 1995;108:1282–1301. 15 Testa MA, Simonson DC: Assessment of quality of life outcomes. N Engl J Med 1996;334: 835–840. 16 Schneider SM, Pouget I, Staccini P, et al: Quality of life in long-term home enteral nutrition patients. Clin Nutr 2000;19:23–28. 17 Schneider SM, Pouget I, Pivot X, et al: Improvement of quality of life in home enteral nutrition patients: A prospective study (abstract). Clin Nutr 2000;19(suppl 1):56–57. 18 Roberge C, Tran M, Poirée B, et al: Quality of life and home enteral tube feeding: A French prospective study in patients with head and neck oesophageal cancer. Br J Cancer 2000;82:263–269. 19 Loeser C, von Herz U, Küchler T, et al: Quality of life and nutritional state in patients on home enteral tube feeding. Nutrition 2003;19:605–611. 20 Sanders DS, Carter MJ, D’Silva J, et al: Survival analysis in percutaneous endoscopic gastrostomy feeding: A worse outcome in patients with dementia. Am J Gastroenterol 2000;95: 1472–1475. 21 Howard L, Ament M, Fleming C, et al: Current use and clinical outcome of home parenteral and enteral nutrition therapies in the United States. Gastroenterology 1995;109:355–365. 22 Christakis NA, Lamont EB: Extent and determinants of error in doctors’ prognoses in terminally ill patients: Prospective cohort study. BMJ 2000;320:469–472. 23 Murphy LM, Lipman TO: Percutaneous endoscopic gastrostomy does not prolong survival in patients with dementia. Arch Intern Med 2003;163:1351–1353. 24 Nair S, Hertan H, Pitchumoni CS: Hypoalbuminemia is a poor predictor of survival after percutaneous endoscopic gastrostomy in elderly patients with dementia. Am J Gastroenterol 2000;95:133–136. 25 Finucare TE, Christmas C, Travis K: Tube feeding in patients with advanced dementia: A review of the evidence. JAMA 1999;282:1365–1370. 26 Rudberd MA, Egleston BL, Grant MD, Brody JA: Effectiveness of feeding tubes in nursing home residents with swallowing disorders. JPEN J Parenter Enteral Nutr 2000;24:97–102. 27 Norton B, Homer-Ward M, Donnelly MT, et al: A randomised prospective comparison of percutaneous endoscopic gastrostomy and nasogastric tube feeding after dyspahgic stroke. BMJ 1996;312:13–16. 28 Mazzini L, Corra T, Zaccala M, et al: Percutaneous endoscopic gastrostomy and enteral nutrition in amyotrophic lateral sclerosis. J Neurol 1995;242:695–698. 29 Coti Bertrand P, Piquet MA, Bordier I, et al: Preoperative nutritional support at home in head and neck cancer patients: From nutritional benefits to the prevention of the alcohol withdrawal syndrome. Curr Opin Clin Nutr Metab Care 2002;5:435–440. 30 Beaver MES, Matheny KE, Roberts DB, Myers JN: Predictors of weight loss during radiation therapy. Otolaryngol Head Neck Surg 2001;125:645–648. 31 Piquet MA, Ozsahim M, Larpin, et al: Early nutritional intervention in oropharyngeal cancer patients undergoing radiotherapy. Support Care Cancer 2002;10:502–504.
Discussion Dr. Cynober: You started by mentioning that there are two approaches: enteral nutrition at home as basic therapy or medical therapy. You immediately discard basic
98
What Are the Goals of Nutritional Support? therapy and then you presented a lot of interesting data on tolerance, quality of life, and I come to the conclusion that actually what you presented is evidence from basic therapy. My question is, as a goal for example, what about healing in head and neck cancer patients or supporting the immune function in order to avoid secondary infections and so on? Dr. Hébuterne: If you try to avoid some complications it is a medical treatment, and if you start antibiotics because of infection it is a medical treatment, so it is the same. I know it is not easy to differentiate basic treatment and medical treatment and the problem may be in patients in a vegetative state, it is probably the most difficult question. I am not sure that the response is medical, it may be ethical and in this case the family’s opinion is very important. In our experience we have some patients who have been on home enteral nutrition for more than 20 years due to a motorcycle accident, and we have to change the tube every 2 years. In this case the pressure of the family is terrible, they are very anxious about changing the tube in this vegetative patient. Dr. Biesalski: May I go into more detail on the question from Dr. Cynober? I think what you meant was if you take enteral or parenteral nutrition as a medical treatment what about antioxidants, what about immune enhancing substitutes, and what about, for example, special amino acids or whatever, if I take it as a treatment? Dr. Hébuterne: Yes, we have some recommended dietary allowances (RDAs). We need to give the patient the RDA, and if we put 1,500 cal into a patient we have RDAs for vitamins, for everything. After that if you have to give more of something like -carotene or vitamin E or selenium, I think it is a medical treatment because you give more than expected from the RDAs. But the question is are the RDAs right, what is the amount of vitamin C we need every day just to avoid complications due to vitamin C deficiency or to prevent coronary heart disease – that is a problem. Dr. Fürst: My comment is directly associated with the previous question. We have a great problem. You defined the basic treatment to ascertain nutrition requirement. Now we don’t know the nutritional requirement in a disease condition. For instance for amino acids we have no data. What is the requirement? Our moderator may confirm my statement that we have no idea how much vitamins and how much antioxidants we should give to a septic patient. We were sitting in the WHO FO Committee to define the nutritional requirement for amino acids and we only came to the healthy population, so we were not able to give a recommendation concerning sick patients. You nicely showed that by using nutritional support we are able to improve outcome, but I would like to emphasize strongly that probably by knowing the proper requirement we could further improve treatment. This is a priority for future research to define the requirement, because if you do not have the basic treatment, your medical treatment has no baseline on which to build up the appropriate treatment. Dr. Hébuterne: I don’t totally agree with you. I know that it is not a scientific demonstration, that it is an empirical demonstration, but we now have some experience with patients on long-term enteral nutrition. My oldest patient started enteral nutrition in 1985 without receiving anything by mouth, she cannot eat anything, and she cannot swallow. She is like you and me, she can walk, she is a grandmother now, it was after an accident, and she received 1,500 to 2,000 cal/day of standard polymeric diet with no supplementation, without any problems for 20 years. We have several patients like that and I think we can use these patients to discover the requirement. You don’t know the requirement, I don’t know the requirement, but she knows her requirement because when her grandson is expected to come home she takes 2,000 calories to improve her body weight and to have more time to take care of her grandson when he is there. So these kinds of patients know exactly how to manage
99
What Are the Goals of Nutritional Support? enteral nutrition, I think this is interesting. But I agree with you in this case it is not sick patients, it is normal patients. We also have some experience with sick patients after surgery for example, and it is not very difficult to maintain nutritional status and body weight in chronic patients, but of course in acute patients it is a little bit different. Dr. Van Gossum: You mentioned that for patients with a high risk of aspiration home enteral nutrition should perhaps not be initiated, but it is true that it is the main cause of mortality in these patients. What are your criteria to define a high risk of aspiration, and do you think that we could prevent such an event and such complication by using a specific enteral access such as percutaneous endoscopic gastrostomy (PEG) or percutaneous endoscopic jejunostomy? Dr. Hébuterne: The patient at risk of aspiration is a patient with aspiration before enteral nutrition, and this is the case in demented patients. I think most physicians agree that enteral nutrition is not good for these patients. We have conscious patients who cannot swallow; they are of course at risk of aspiration and in this case it may be useful just to put in a PEG. There are also patients who are between non-competent and competent patients. For example the patients of Loeser et al. [1], during the follow-up they are incompetent at the beginning and they become competent. These kinds of patients are very interesting because it is possible to improve their outcome. In this case it may be interesting to use jejunostomy or jejunal PEG or direct percutaneous endoscopic jejunostomy, and according to our experience it is not always very easy to perform. With direct percutaneous endoscopic jejunostomy we have some leakage problems in 15–20% of the patients so we don’t chose it very often. What we try to do is to have a good protocol to avoid aspiration. First of all is to use a pump: it was well demonstrated that a bolus of nutrients in the stomach diminishes the pressure of the lower esophageal sphincter. When enteral nutrition is put in with a pump at a regular flow it is better, and in this case we try to avoid enteral nutrition during sleep, it is better during the day, and also small procedures like that. Dr. Bozzetti: I perfectly agree with your conclusions because they are supported by the data. Nevertheless I also agree with Dr. Cynober and I think we should be reminded of the statement of the ASPEN Board of Directors in their clinical guidelines [2]. When they classify their recommendations for total parenteral nutrition they say that there is some difference between a drug and nutrition because healthy people do not need drugs but healthy people need nutrition to survive. So this is the main difference between basic support and a drug. I think that we are not in the position to demonstrate scientifically through randomized clinical trials the efficacy of nutritional support in aphagic patients because we cannot have a control group that doesn’t receive nutritional support. This is the main reason why I think that nutritional support should be classified in aphagic malnourished patients as a basic support rather than as a clinical treatment. Dr. Correia: I totally agree with most of your final conclusions. I just would like to make two comments based on dementia patients and on patients with a high risk of aspiration. The reason for my comments on that is because most of our decisions are based on scientific reasoning and most of the studies [3, 4] showing that dementia patients do not benefit from nutrition were not randomized and therefore most of the conclusions were biased. That means that most of the patients who received enteral nutrition had it at the very end of their lives, so definitely nutrition will not change the course of their illness. This seems to be the same attitude as oncologists in my hospital when I suggest gastrostomy for head and neck cancer patients, they always say no because they will have a lot of complications due to the gastrostomy. I tell them that the patients are not having complications due to the gastrostomy, they are having complications because they are receiving the gastrostomy when they are already
100
What Are the Goals of Nutritional Support? severely malnourished. So the complications are not due to the gastrostomy per se but to their nutritional status. So again, I am very much afraid when we as specialists working with nutrition say watch out for dementia patients because most of these results are biased: they were not randomized studies, and the patients were chosen early in the initial state of disease so that those who received nutrition did better. The second remark has to do with aspiration pneumonia. Of course if the patients are at high risk of aspiration pneumonia, they will still be at high risk of aspiration pneumonia no matter whether they receive enteral diets or not, and no matter if it is in the stomach or in the jejunum because they will aspirate saliva anyway. So I think that, although they are at high risk, they should still receive nutrition because otherwise they are going to die of starvation rather than their disease. I totally agree with your results, but we must be cautious when we make statements such as watch out for dementia patients or high risk patients because those outside will just use that against us or against our patients. Dr. Hébuterne: I totally agree with you it is very important to start as early as possible, and then we never regret placing a PEG when the disease just starts, at the beginning, it is really the best way to do. But the problem is that sometimes we first see the patient at the end of the disease, a 90-year-old patient with a body mass index of 12, and we are asked to do something. What can we do, if we just touch the patient he will die, so it is a problem. The studies on amyotrophic lateral sclerosis (ALS) patients demonstrate that if we start at the beginning it is probably better at this time. Of course if you do very well, nutritional support may improve the outcome, it may improve survival. I am sure that we improve survival in some patients if we start at the beginning but I am also sure that we can kill the patient if we start later. The selection of the patients is the most difficult point. When we started our program of enteral nutrition the 1-month mortality was 25% as in the literature. It was not too bad but we thought it was bad, so we tried to improve that. We have now a policy for selecting the patients. It is impossible for us to go down to a 15% 1-month mortality in these patients. Why, because when you ask a physician to assess the survival of patients, 50% of the time the assessment is not good, and in this case there is always an overoptimistic assessment. This was demonstrated several years ago. So we expect the patient will live 2–5 months, 5 months if we are optimistic. In this case there is doubt about the probability of survival of the patients and the doctors prefer to try the procedure rather than not try it, but this is difficult and in our experience less than 15% of the 1-month mortality is not possible. Dr. Labadarios: In your first slide I found it very interesting that you said, or I think I understood you saying, that malnutrition is a disease. Can you elaborate on that? Dr. Hébuterne: I am happy that you pointed this out because I think we have to explain to other physicians and to politicians that malnutrition is a disease. For example we are going to change the legislation on home enteral nutrition in France for the third time in 10 years. In France we like to change the law regularly, and at this time we are trying to avoid a list of diseases, and we are trying to say that the indication of enteral nutrition is first malnutrition. So if the problem is malnutrition we have to treat it, and it is logical to say that nutritional support is a treatment if malnutrition is a disease. Secondly patients are at risk of malnutrition when they are not able to cover their energy needs with food. We are trying to avoid a list of diseases because when you make a list of diseases there is always one disease which is not on the list, and that is a problem. Dr. Labadarios: Actually I agree with the speaker that malnutrition is a disease, and what is important is that it is the number 1 disease worldwide, which is a neglected entity in any case.
101
What Are the Goals of Nutritional Support? References 1 Loeser C, von Herz U, Kuchler T, et al: Quality of life and nutritional state in patients on home enteral tube feeding. Nutrition 2003;19:605–611. 2 ASPEN Board of Directors and the Clinical Guidelines Task Force: Guidelines for the use of parenteral and enteral nutrition in adult and pediatric patients. JPEN J Parenter Enteral Nutr 2002;26(suppl):1SA–138SA. 3 Kaw M, Sekas G: Long-term follow-up of consequences of percutaneous endoscopic gastrostomy (PEG) tubes in nursing home patients. Dig Dis Sci 1994;39:738–743. 4 Hull MA, Rawlings J, Murray FE, et al: Audit of outcome of long-term enteral nutrition by percutaneous endoscopic gastrostomy. Lancet 1993;341:869–872.
102
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 103–125, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
Oral Protein and Energy Supplementation in Older People: A Systematic Review of Randomized Trials Anne C. Milnea, Alison Avenella and Jan Potterb aHealth
Services Research Unit, University of Aberdeen, Aberdeen, and of Medicine for the Elderly, Victoria Infirmary, South Glasgow University Hospitals Trust, Glasgow, UK
bDepartment
Introduction A report of the UK Royal College of Physicians entitled ‘Nutrition and Patients: A Doctor’s Responsibility’, published in July 2002 [1], highlighted again the importance of nutritional care for vulnerable groups such as undernourished elderly people, particularly those who are hospitalized or institutionalized. Older people have longer periods of illness and longer hospital stays [2], and data show that up to 55% of elderly hospitalized patients are undernourished on admission [3, 4]. Malnutrition is associated with poorer recovery in a broad range of patients and conditions [5, 6], and furthermore, nutritional status has been shown to decline during hospitalization [3, 7]. This has led to the hypothesis that providing nutritional support to those at risk of malnutrition will have beneficial effects on outcome. The best way to establish this is to demonstrate it in a randomized controlled trial (RCT). A systematic review of RCTs by Potter et al. [8] in 1998 examined the effects of oral and enteral protein and energy supplementation in trials involving adults of all ages with any medical or surgical condition. Supplementation seemed to improve the nutritional indices of adult patients, but there were insufficient data in trials which met strict methodological criteria to be certain if mortality was reduced. A Cochrane Review of dietary advice for illness-related malnutrition in adults of all ages was also carried out by Baldwin et al. [9]. There was insufficient evidence to establish whether dietary advice alone improves the outcomes, however the limited information available suggested that weight gain was greater in people who also received supplements, with or without advice. 103
Oral Protein and Energy Supplementation in Older People Oral nutritional supplements are widely prescribed for older people both in hospital and in the community and it is important to establish whether they are clinically effective, by reviewing trials of oral protein and energy supplementation and including more recent better quality trials. Methods Studies were identified, and the analyses were carried out according Cochrane methodology [10]. The search (until the end of 2002), included the following databases: the Cochrane Controlled Trials Register, Medline, Embase, Biosis and CAB abstracts. Selection of Studies We included randomized or quasi-randomized trials with a minimum intervention of 1 week. Groups of study participants had to have a minimum average age of 65 years. All patient groups were included, with the exception of people in critical care or recovering from cancer treatment. We included commercial sip feeds, other milk-based supplements and fortification of normal food sources. We excluded studies of specially designed immunomodulatory supplements or supplements of specific amino acids. The full description of the search strategy is available elsewhere [11]. If necessary, trialists were contacted for further information on methodology and data. The following outcomes were examined as pre-specified in our protocol: all-cause mortality; number of people with morbidity or complications; length of hospital stay; functional status; participants’ perceived quality of life; percentage weight change; percentage change in mid-arm muscle circumference (MAMC); acceptance of the supplement, and side effects. Extraction of outcome data from the included trials and quality assessment of trials was carried out independently by 2 reviewers. All differences were resolved by discussion. Data were combined for meta-analysis for the dichotomous variables mortality and number of people with complications. For each study relative risks (RRs) were calculated, the results were combined using fixed effects models with 95% confidence limits. Because length of hospital stay data were likely to be positively skewed and therefore not suitable for meta-analysis, the effect of supplementation on length of stay was tested using the Wilcoxon signed ranks test. Weighted mean difference and 95% confidence intervals were calculated for percentage weight change and percentage MAMC change using a fixed effects model. For meta-analyses the same methods were used to standardize for missing data as previously used by Potter et al. [8]. Where there was evidence of heterogeneity in the meta-analyses a random effects model was applied. Heterogeneity between comparable trials was explored using a standard 2 test, the level of statistical significance used was p ⬍ 0.1 (two-sided). Pre-specified subgroup analyses of the mortality data were carried out: (a) baseline nutritional status as reported by the investigators (nourished, undernourished); (b) mean age (⬍75 years, ⱖ75 years); (c) amount of kilojoules provided in supplement (⬍1,674 kJ (400 kcal), ⱖ1,674 kJ); (d) duration of intervention (⬍35 days, ⱖ35 days), and (e) environment (hospitalized, not hospitalized). An exploratory subgroup analysis was also carried out on the basis of diagnostic group, providing 3 or more trials were available. The diagnostic groups were patients with hip fracture, chest conditions, stroke, congestive heart failure, geriatric conditions (trials which included frail patients with a variety of pathologies), and older surgical patients.
104
Oral Protein and Energy Supplementation in Older People A sensitivity analysis was carried out including only trials which reported clearly concealed randomization. Description of Studies More than 200 potentially relevant trials were identified from the searches. Two reviewers independently assessed these trials. As a result of mutual agreement, 35 have been included in the review with a total of 3,242 individually randomized participants (table 1). Most were patients with various pathologies in long-stay/care of the elderly/hospital wards or nursing homes (15 trials), or older people at home (4 trials) or in a metabolic unit (1 trial). The remaining trials included particular groups with hip fracture (6 trials), stroke (1 trial), congestive heart failure (1 trial), chest conditions (5 trials), or older surgical patients (2 trials). Most trials used commercial supplements. About 50% of trials offered ⱖ1,674 kJ/day (400 kcal). The period of the intervention ranged from 7 days to 18 months. The intervention time was ⱖ35 days for about 50% of trials.
Results Methodological Quality of Included Studies Concealment of allocation was reported in 9 studies. An intention-to-treat analysis was reported in 17 studies. Only 7 studies clearly reported taking action to blind outcome assessors such that bias was unlikely (table 1). Full details of the quality assessment are available elsewhere [11]. Outcomes Mortality: Results for the meta-analysis were obtained from 24 trials (2,387 participants). Nutritional supplementation by the end of follow-up was associated with reduced mortality (RR 0.69, 95% CI 0.54–0.88; fig. 1). Subgroup Analyses (table 2) The results were consistently statistically significant when limited to trials where participants were undernourished, ⬎75 years, were offered ⱖ400 kcal (1,674 kJ)/day in the supplement, when supplementation was continued for ⱖ35 days, and when participants were hospitalized. The results were also statistically significant when limited to trials including frail patients with a variety of geriatric conditions. Sensitivity Analyses: The results for mortality for all trials were consistent when analysis was restricted to trials with clearly concealed randomization (n ⫽ 1,536; RR 0.63, 95% CI 0.48–0.83). Number of Complications: Data were combined for meta-analysis for the total number of complications from 10 studies (n ⫽ 876). The risk of complications in supplemented groups was not significantly different from the control groups (RR 0.92, 95% CI 0.78–1.09; fig. 2). Length of Stay: Data on length of hospital stay were combined from 8 studies (1,135 patients; table 3), including 1 study with 3 subgroups [12]. The mean length of stay was shorter for the supplemented groups (18.9 versus 25.4 days) but this was not statistically significant, (Wilcoxon signed ranks test p ⫽ 0.113). 105
106
101/103 65
9/13
5/5
23/23
37
Barr et al. [23], 2000
Broqvist et al. [24], 1994
Brown and Seabrook [25, 1992
Carver and Dobson [26], 1995
Deletter [27], 1991
67
Not given
Not given
72
81
33/30
Banerjee [22], 1978
Mean age years
N/Na
Reference, year
Thin elderly, hip fracture
Severe congestive heart failure
Healthy
Consent only
Inclusion criteria
Home
Outpatients, COPD
Care of the Senile elderly wards dementia, malnourished
Inpatients
Possibly outpatients
At home
Long-stay wards
Environment
Table 1. Description of included trials
9 weeks
12 weeks, 2,519 kJ/day
Until discharge
8 weeks, 3.138 kJ and 30 g Pr/day
12 weeks, 6.8 dl low fat milk/day
1
0
0
1
1
2
1
2
2
1
2
0
1
2
2
0
0
Mortality, anthropometry, dietary intake Weight change, Barthel index, quality of life, dietary intake Mortality, number of complications, weight change, anthropometry, dietary intake Mortality, length of stay, functional status, anthropometry, weight, change, dietary intake Mortality, weight change, anthropometry, compliance Functional status, weight change, anthropometry, dietary intake
AAc ITTd 0Be Outcomesf
14 weeks, 1,109 kJ 1 and 18.6 g Pr/day
Interventionb
Oral Protein and Energy Supplementation in Older People
27/32
24/26
21/21
7/9
25/25
10/10
Delmi [28], 1990
Fiatarone et al. [20], 1994
Gariballa et al. [29], 1998
Gegerle et al. [30], 1986
Gray et al. [14], 1995
Hankey et al. [31], 1993
81
78
77
79
87
82
Inpatients
At home
Inpatients
Inpatients
Long-term care residents
Orthopaedic unit
Continuing care, elderly malnourished, over 75 years
Nutritionally ‘at risk’
Hip fracture
Thin stroke patients, able to swallow
Over 70 years, able to walk 6 m
Hip fracture, over 60 years
8 weeks
Until discharge, 1,063 kJ and 20 g Pr/day 12 weeks, 1,960–2,090 kJ and 17–26 g Pr/day
4 weeks, 2,510 kJ and 20 g Pr/day
10 weeks, 1,506 kJ/day
Until discharge, 1,060 kJ and 20.4 g Pr/day
1
1
1
2
1
1
0
2
2
2
2
1
0
1
0
1
0
0
Mortality, quality of life, functional status, handgrip, anthropometry, weight change, dietary intake Anthropometry, weight change, dietary intake
Mortality, number of complications, side effects, length of stay, dietary intake Mortality, functional status, side effects, weight change, anthropometry, dietary intake Mortality, number of complications, length of stay, Barthel index, weight change, anthropometry, dietary intake Dietary intake
Oral Protein and Energy Supplementation in Older People
107
108
13/12
Knowles et al. [37], 1988
35/36
28/24
At home
unclear Over 75
Jensen and Hessov [34], 1997 Krondl et al. [35], 1999
Kwok et al. [36], 2001
At home
16/16
Nursing home
Metabolic unit
80
⬃69
70
Not given Hospitalized geriatric
Inpatients
Hubsch et al. [33], 1992
86
17/15
Environment
Hankins [32], 1996
Mean age years
N/Na
Reference, year
Table 1 (continued)
8 weeks
7 weeks
BMI ⬍27
Malnourished with severe COPD
16 weeks, 983 kJ and 11.8 g Pr/day
Until discharge, 996 kJ and 20 g Pr/day 110 days
1
1
0
2
1
2
1
0
1
2
2
2
2
0
1
0
0
Anthropometry, weight change, dietary intake Mortality, quality of life, weight change, dietary intake Mortality, grip strength, anthropometry, weight change, dietary intake, side effects Functional status, anthropometry, dietary intake
Mortality, number of complications, side effects, length of stay, Barthel index, weight change, anthropometry, dietary intake Mortality, weight change
AAc ITTd 0Be Outcomesf
30 days, 1,710 kJ 2 and 22.5 g Pr/day
Interventionb
Healthy elderly
After surgery
Undernourished
Thin, hip fracture, over 65 years
Inclusion criteria
Oral Protein and Energy Supplementation in Older People
McEvoy and 26/25 James [41], 1982 McWhirter and 35/26 Pennington [42], 1996
Unclear Not given
Madigan [40], 1994
Inpatients
⬃71 Malnourished medical patients
Acute geriatric Poorly ward nourished
Hip fracture, over 60 years
Gastrointestinal surgery
Nutritionally ‘at risk’
Newly admitted to long-term care
Not given
Inpatients
Out- and inpatients
⬃65
24/24/ 27/25
MacFie et al. [39], 2000
Nursing home
19/22
Lauque et al. [38], 2000
Inpatients
84
197/238 ⬃80
Larsson et al. [7], 1990
1
2
1
2
Minimum of 7 days to meet nutritional needs
1
4 weeks, 2,694 kJ 1 and 36.4 g Pr/day
10 days, 1,300 kJ and 16 g Pr/day
Outpatient preoperatively until day before surgery, 2,092–2,510 kJ and 10–20 g Pr/day
60 days
Up to 8 weeks, 1,674 kJ and 16 g Pr/day
2
2
0
1
1
0
0
0
0
0
0
0
Mortality, number of complications, functional status Mortality, handgrip, weight change, dietary intake Mortality, number of complications, length of stay, side effects, handgrip quality of life, anthropometry, weight change, dietary intake Mortality, number of complications, length of stay, functional status, anthropometry, dietary intake, compliance Mortality, anthropometry, weight change Number of side effects, weight change, anthropometry, dietary intake
Oral Protein and Energy Supplementation in Older People
109
110
186/195 83
Potter et al. [12], 2001
Nutritionally ‘at risk’
Nutritionally ‘at risk’
Very healthy
Inclusion criteria
Acute geriatric Subgroup ward nutritionally ‘at risk’
At home
⬃80
43/46
Payette et al. [13], 2002
Inpatients
⬃74
17/17
Ovesen [44], 1992
Metabolic unit
⬃67
6/6
Meredith et al. [43], 1992
Environment
Mean age years
N/Na
Reference, year
Table 1 (continued)
1
Until discharge, 2,259 kJ and 22.5 g Pr/day
2
16 weeks, 1 maximum tolerable intake to gain 0.5 kg body weight/week
10 days
2
2
1
1
1
1
1
0
Mortality functional status, anthropometry, weight change, dietary intake Mortality, side effects, dietary intake, compliance Mortality, weight change, dietary intake, functional status, anthropometry, handgrip, quality of life Mortality, number of complications, length of stay, Barthel index, anthropometry, weight change, dietary intake
AAc ITTd 0Be Outcomesf
12 weeks, 33 kJ 1 and 0.33 g Pr/day, per kg ideal body weight
Interventionb
Oral Protein and Energy Supplementation in Older People
At home
85
⬃73
35/37
40/41
Long-term care
275/274 66
Vlaming et al. [46], 2001 Volkert et al. [17], 1996
Woo et al. [16], 1994
Inpatients
Unclear 82
COPD
Inpatients
1
1
2
1
Until 2 discharge 6 months, 2,092 kJ 1 and 30 g Pr/day
10 days, 1,340 kJ and 18.5 g Pr/day
8 weeks, 1,757 kJ and 14 g Pr/day
14 days
After inpatient stay 1 month, 2,092 kJ with chest infection and 17 g Pr/day
Nutritionally ‘at risk’ Nutritionally ‘at risk’
Hip fracture, over 65 years
Rehabilitation COPD center
Stableforth [19], 1986
65
72/63
Inpatients
Schols et al. [45], 1995
69
17/16
Saundy et al. [18], 1997
2
1
2
1
1
1
2
0
2
0
2
1
Mortality, number of complications, length of stay, functional status, quality of life, anthropometry, weight change, dietary intake Mortality, functional status, anthropometry, weight change, dietary intake, side effects Mortality, number of complications, dietary intake Mortality, length of stay Mortality, Barthel index, anthropometry, weight change, dietary intake, compliance Mortality, Barthel index, anthropometry, weight change, dietary intake
Oral Protein and Energy Supplementation in Older People
111
112
32/30
Yamaguchi et al. [47], 1998
Environment
At home
Mean age years
⬃78 Home-bound starting to receive meals-on-wheels
Inclusion criteria 0
0
Weight change, dietary intake
AAc ITTd 0Be Outcomesf
18 months, 2,510 kJ 2 and 30 g Pr/day
Interventionb
COPD ⫽ Chronic obstructive pulmonary disease; BMI ⫽ body mass index. aIntervention/control. bDuration of intervention and composition of supplement if reported. cAllocation concealed: 2 ⫽ method did not allow disclosure of assignment, 1 ⫽ small but possible chance of disclosure, or states random but no description, 0 ⫽ quasi-randomized. dIntention-to-treat analysis: 2 ⫽ carried out for all possible cases, 1 ⫽ states number and reasons for withdrawal but ITT not possible, 0 ⫽ not mentioned or not possible. eBlinding of outcome assessors: 2 ⫽ action taken, or outcomes such that bias unlikely, 1 ⫽ small or moderate chance of unblinding of assessors, 0 ⫽ not mentioned. fData unavailable or incomplete for some outcomes, results from dietary intake not reported here.
N/Na
Reference, year
Table 1 (continued)
Oral Protein and Energy Supplementation in Older People
Oral Protein and Energy Supplementation in Older People Comparison: Outcome:
Oral protein and energy versus routine care 01 Mortality
Study or sub-category Broqvist 1994 Brown 1992 Carver 1995 Deletter 1991 Delmi 1990 Fiatarone 1994 Gariballa 1998 Gray-Donald 1995 Hankins 1996 Hubsch 1992 Krondl 1999 Kwok 2001 Lauque 2000 MacFie 2000 Madigan 1994 McEvoy 1982 Meredith 1992 Ovesen 1992 Payette 2002 SG Larsson malnour SG Larsson nourished SG Potter malnourish SG Potter nourished Stableforth 1986 Vlaming 2001 Volkert 1996
Treatment n/N
Control n/N
1/9 0/5 0/20 0/18 6/27 1/49 2/20 3/25 2/17 0/16 0/35 1/28 0/19 4/75 4/18 0/26 0/6 0/17 0/43 17/59 12/138 13/124 8/62 0/24 12/275 4/35
1/13 0/5 0/20 0/17 10/32 1/51 7/20 1/25 4/14 0/16 0/36 0/24 0/22 1/25 0/12 0/25 0/5 0/17 0/46 21/56 34/182 27/127 6/68 0/34 14/274 8/37
RR (fixed) 95% CI
Weight % 0.62
10.70 5.94
1.44[0.10,20.21] Not estimable Not estimable Not estimable 0.71[0.30,1.70] 1.04[0.70,16.18] 0.29[0.07,1.21] 3.00[0.33,26.92] 0.41[0.09,1.93] Not estimable Not estimable 2.59[0.11,60.69] Not estimable 1.33[0.16,11.38] 6.16[0.36,104.90] Not estimable Not estimable Not estimable Not estimable 0.77[0.45,1.30] 0.47[0.25,0.87] 0.49[0.27,0.91] 1.46[0.54,3.98] Not estimable 0.85[0.40,1.81] 0.53[0.17,1.60]
100.0
0.69[0.54,0.88]
6.98 0.75 5.34 0.76 3.35 0.41 1.14 0.45
16.44 22.38 20.36 4.37
Total (95%CI) 1190 1203 Total events: 90 (Treatment), 135 (Control) 2 Test for heterogeneity: Chi ⫽ 12.86, df⫽14 (P⫽ 0.54) Test for overall effect: Z⫽ 2.94 (P⫽0.003)
RR (fixed) 95% CI
0.1 0.2 0.5 1 2 5 10 Favours treatment Favours control
Fig. 1. Oral protein and energy versus routine care: mortality. Table 2. Subgroup analysis of mortality Subgroup analysis for mortality Nutritional status Undernourished Not undernourished Age Mean age ⱖ75 years Mean age ⬍75 years kcal offered/day ⱖ400 kcal/day ⬍400 kcal/day Period of supplementation ⱖ35 days ⬍35 days Location Hospital inpatients Community Disease state Geriatric conditions Hip fracture
Relative risk
95% CI
Number of studies (patients) included
0.65 0.76
0.48, 0.89 0.50, 1.14
15 (1,486) 10 (901)
0.63 0.93
0.48, 0.82 0.47, 1.84
15 (1,484) 7 (822)
0.65 1.12
0.50, 0.85 0.53, 2.34
14 (1,906) 8 (436)
0.66 0.67
0.46, 0.95 0.42, 1.09
12 (1,012) 9 (954)
0.66 1.54
0.50, 0.85 0.31, 7.68
14 (1,791) 7 (3,740)
0.68 0.85
0.52, 0.90 0.42, 1.70
13 (1,920) 5 (188)
113
Oral Protein and Energy Supplementation in Older People Review: Comparison: Outcome:
Protein and energy supplementation in elderly people at risk from malnutrition (Version 15) 01 Oral protein and energy versus routine care 11 Number of complications
Study or sub-category
Treatment n/N
Control n/N
Broqvist 1994 Delmi 1990 Gariballa 1998 Hankins 1996 Larsson 1990 MacFie 2000 Madigan 1994 Potter 2001 Saudny 1997 Stableforth 1986
2/9 4/25 9/20 5/17 67/116 19/75 6/18 37/130 0/14 0/24
0/13 10/27 11/20 6/12 83/137 3/25 4/12 44/138 1/10 0/34
RR (fixed) (95% CI)
Weight % 0.26 6.09 6.97 4.46 48.20 2.85 3.04 27.03 1.10
Total (95%CI) 448 428 Total events: 149 (Treatment), 162 (Control) 2 Test for heterogeneity: Chi ⫽7.91, df⫽ 8 (P⫽ 0.44), I2 ⫽ 0% Test for overall effect: Z⫽ 0.94 (P⫽0.35)
100.00
RR (fixed) 95% CI 7.00 [0.38, 130.56] 0.43 [0.16, 1.20] 0.82 [0.44, 1.53] 0.59 [0.23, 1.49] 0.95 [0.78, 1.17] 2.11 [0.68, 6.54] 1.00 [0.36, 2.81] 0.89 [0.62, 1.29] 0.24 [0.01, 5.45] Not estimable 0.92 [0.78, 1.09]
0.1 0.2 0.5 1 2 5 10 Favours treatment Favours control
Fig. 2. Oral protein and energy versus routine care: number of complications.
Functional Status: Functional status or quality of life measures were reported in 21 studies, the outcomes measured were diverse, and few suggested any functional benefit. Fewer bed disability days were reported in the study by Payette et al. [13]. Gray et al. [14] reported that the number of falls was lower among those receiving supplements. Larsson et al. [7] reported a significant improvement in the activity rating in the supplemented group at 8 weeks compared to the control group. The Barthel activities of daily living (ADL) index [15] was measured in 6 studies. Only 1 study [16] reported a lower level of functional ability in the control group. Potter et al. [12] reported a significant improvement with supplements only in a subgroup of very malnourished patients. Volkert et al. [17] found an improvement in the ADL index only a subgroup with good acceptance of the supplement. Four studies in patients with chronic obstructive pulmonary disease measured lung function. Only 1 study [18] found a sustained statistically significant improvement in forced vital capacity with supplements. No statistically significant effects between groups were reported from quality of life questionnaires. Weight Change: In data from 24 trials (1,867 participants) the pooled weighted mean difference for percentage weight change showed a benefit from supplementation of 2.4% (95% CI 1.9–3.0; fig. 3). MAMC: In data from 13 trials (1,187 participants) the pooled weighted mean difference for percentage MAMC change showed a benefit from supplementation of 1.4% (95% CI 0.19–2.6; fig. 4). 114
*Missing data.
Vlaming et al. [46], 2001
Potter et al. [12], 2001
Potter et al. [12], 2001
Potter et al. [12], 2001
MacFie et al. [39], 2000 Madigan [40], 1994
Gariballa et al. [29], 1998 Hankins [32], 1996
Fractured femur, acute hospital stay (mean and SD) Severely undernourished, emergency admission to care of the elderly unit (median and range) Moderately undernourished, emergency admission to care of the elderly unit (median and range) Adequately nourished, emergency admission to care of the elderly unit (median and range) Nutritionally ‘at risk’ hospital patients (mean and SD)
Fractured femur, days from admission to discharge from orthopedic surgeon’s care (mean and range) Fractured femur, days in orthopedic unit and recovery hospital (median) Stroke patients (median and interquartile range) Fractured femur, mean acute hospital and rehabilitation hospital stay Gastrointestinal surgery, postoperative stay
Brown and Seabrook [25], 1992
Delmi et al. [28], 1990
Description
Reference, year
274
54
82
29
18
14.2
13.5
18.5
17
16
11
24
17 75
24
24
27
20
21
5
24.9
3–62
3–141
4–100
8
*
6–50
3–122
13–157
17–37
274
62
74
26
12
25
14
20
28
5
11.4
21
16.5
17.5
15
13
26
42
40
48
LOS days
n
range
n
LOS days
Control
Supplement
Table 3. Effect of oral protein and energy supplementation on length of hospital stay (LOS)
16.4
2–69
3–62
2–76
11
*
3–60
3–77
10–259
11–85
range
Oral Protein and Energy Supplementation in Older People
115
Oral Protein and Energy Supplementation in Older People Comparison: Oral protein and energy versus routine care Outcome: % Weight change Treatment Control Study n Mean(sd) n Mean(sd) Barr 2000 Broqvist 1994 Brown 1992 Carver 1995 Deletter 1991 Fiatarone 1994 Gariballa 1998 Gray-Donald 1995 Hankey 1993 Hubsch 1992 Krondl 1999 Kwok 2001 Lauque 2000 MacFie 2000 McEvoy 1982 McWhirter 1996 Meredith 1992 Payette 2002 Potter 2001 SG Larsson malnour SG Larsson nourished SG Volkert comply SG Volkert non compl Schols 1995 Woo 1994 Yamaguchi 1998
101 1.93(10.00) 7 1.17(10.00) 5 ⫺2.60(2.30) 20 7.50(10.00) 18 1.96(10.00) 24 1.50(3.40) 18 0.35(10.00) 22 4.38(4.80) 7 2.83(10.00) 16 ⫺0.33(10.00) 35 0.00(10.00) 25 3.37(10.00) 13 2.60(10.00) 75 ⫺6.20(10.00) 26 4.33(4.00) 35 2.90(10.00) 6 2.98(10.00) 42 3.00(10.00) 142 1.00(5.60) 59 0.05(0.19) 138 ⫺1.89(6.84) 7 8.20(10.00) 6 3.30(10.00) 33 1.56(3.40) 40 4.70(10.00) 11 4.80(10.00)
103 12 5 20 17 26 13 24 7 16 36 20 22 25 25 26 5 41 151 56 182 9 10 38 41 6
WMD (95%Cl Fixed)
1.02(10.00) ⫺0.26(10.00) ⫺9.10(7.90) 1.32(10.00) 0.00(10.00) ⫺0.80(3.10) ⫺1.23(10.00) 1.23(3.28) ⫺0.53(10.00) 0.33(10.00) 0.00(1000) ⫺0.70(10.00) ⫺2.48(10.00) ⫺4.30(10.00) ⫺0.33(2.48) ⫺2.50(10.00) ⫺2.03(10.00) 0.00(10.00) ⫺1.00(6.00) ⫺1.96(4.00) ⫺6.49(28.80) 6.45(10.00) 6.45(10.00) ⫺0.54(3.20) 2.70(1000) ⫺5.30(10.00)
Total (95%CI) 931 936 Test for heterogeneity chi-square⫽23.52 df⫽ 25 p⫽ 0.55 Test for overall effect z⫽8.60 p⬍0.00001 ⫺10 ⫺5 Favours control
0
Weight %
WMD (95% Cl Fixed)
4.0 0.3 0.6 0.8 0.7 9.2 0.6 5.2 0.3 0.6 1.4 0.9 0.6 1.5 9.1 1.2 0.2 1.6 17.1 27.4 1.6 0.3 0.3 12.6 1.6 0.3
0.91[⫺1.83,3.65] 1.43[⫺7.89,10.75] 6.50[⫺0.71,13.71] 6.18[⫺0.02,12.38] 1.96[⫺4.67,8.59] 2.30[0.49,4.11] 1.58[⫺5.55,8.71] 3.15[0.75,5.55] 3.36[⫺7.12,13.84] ⫺0.66[⫺7.59,6.27] 0.00[⫺4.65,4.65] 4.07[⫺1.81,9.95] 5.08[⫺1.78,11.94] ⫺1.90[⫺6.43,2.63] 4.66[2.84,6.48] 5.40[0.33,10.47] 5.01[⫺6.86,16.88] 3.00[⫺1.30,7.30] 2.00[0.67,3.33] 2.01[0.96,3.06] 4.60[0.26,8.94] 1.75[⫺8.13,11.63] ⫺3.15[⫺13.27,6.97] 2.10[0.56,3.64] 2.00[⫺2.36,6.36] 10.10[10.15,20.05]
100.0
2.41[1.86,2.96]
5 10 Favours treatment
Fig. 3. Oral protein and energy versus routine care: % weight change.
Acceptance of the Supplement and Side Effects Supplemented intake was significantly greater than non-supplemented intake in most studies. Fiatarone et al. [20], however, found that the increase in intake from the supplements had been offset by a reduction in normal food intake. Thirteen studies reported that the supplements were well accepted by most patients. Problems with acceptance were reported by Gray et al. [14] where 36% of potentially eligible participants refused to participate mainly because they did not wish to take a supplement. Volkert et al. [17] found that 45% of participants had poor acceptance. Problems with nausea, gastrointestinal discomfort and diarrhea attributed to the supplements were also reported in a 4 studies. Discussion This review suggests that there may be a beneficial effect of supplementation on mortality. However doubts remain because of the poor quality of most included trials. 116
Oral Protein and Energy Supplementation in Older People Comparison: Outcome:
Oral protein and energy versus routine care % Arm muscle circumference change
Study
Treatment Control n Mean(sd) n Mean(sd)
Broqvist 1994 7 Brown 1992 5 Carver 1995 20 Gariballa 1998 18 Hankey 1993 7 Knowles 1988 13 Kwok 2001 25 McEvoy 1982 26 McWhirter 1996 35 Payette 2002 42 Potter 2001 142 SG Larsson malnor 59 SG Larsson nourished 138 Woo 1994 40
⫺0.39(10.00) ⫺2.01(2.41) 2.69(4.84) 0.00(10.00) ⫺1.00(10.00) 7.39(10.00) 0.00(10.00) 1.00(3.75) 1.66(10.00) 0.00(10.00) ⫺0.40(6.60) ⫺5.07(10.70) ⫺1.10(4.10) ⫺0.45(10.00)
12 5 20 13 7 12 21 25 26 41 149 56 182 41
WMD (95% Cl Random)
1.6 ⫺0.39[⫺9.71,8.93] 10.8 2.92[0.16,5.68] 4.9 2.69[⫺6.27,7.99] 2.6 0.86[⫺6.27,7.99] 1.3 ⫺1.60[⫺12.08,8.88] 2.1 7.82[⫺0.03,15.67] 3.7 1.47[⫺4.33,7.27] 18.1 1.00[⫺0.49,2.49] 4.6 ⫺4.45[⫺0.62,9.52] 5.9 1.00[⫺3.30,5.30] 18.3 1.20[⫺0.26,2.66] ⫺9.7 ⫺3.79[⫺6.81,⫺0.77] 10.7 3.69[0.92,6.46] 5.8 0.65[⫺3.71,5.01]
0.00(10.00) ⫺4.93(2.02) 0.00(10.00) ⫺0.86(10.00) 0.60(10.00) ⫺0.43(10.00) ⫺1.47(10.00) 0.00(1.00) ⫺2.79(10.00) ⫺1.00(10.00) ⫺1.60(6.10) ⫺1.28(4.93) ⫺4.79(18.50) ⫺1.10(10.00)
Total(95%Cl) 577 610 Test for heterogeneity chi-square⫽ 20.25 df⫽13 p⫽ 0.089 Test for overall effect z⫽ 2.27 p⫽ 0.02
⫺10
Weight WMD % (95% Cl Random)
100.0 1.39[0.19, 2.60]
⫺5
Favours control
0
5
10
Favours treatment
Fig. 4. Oral protein and energy versus routine care: % arm muscle circumference change.
The data were unable to suggest reduced hospital stay or complications, which might be expected to accompany reduced mortality. The duration of the intervention was probably too short in most trials to realistically improve these outcomes. However, results of the subgroup analysis of mortality data do follow a pattern of benefit which may be expected clinically. Supplementation produced a small but consistent weight gain, which could be fat, muscle or water. A gain of fat mass or water will not improve muscle strength. The data from MAMC also suggests a small gain, possibly in muscle mass. However, Fiatarone et al. [20] proposed that exercise is also required to produce a significant improvement in muscle strength and function. Since few studies used placebo supplements, bias may result from supplemented patients receiving a higher standard of care, or from outcome assessors being aware of treatment status. Only 7 trials reported that outcome assessors were blinded. Nearly 50% of trials did not report an intention-to-treat analysis and participants in some studies were excluded from the analysis because they were unable to take the supplements. It is well known that frail elderly patients have low intakes and can find it difficult to consume oral supplements. Indeed, it has been suggested that even with extra feeding support from specially trained staff, improvement in the nutritional status of older patients may not be achieved [21]. Problems with 117
Oral Protein and Energy Supplementation in Older People acceptance of the supplement were reported in some trials, but the methods of delivery of the supplement were poorly described. The mode and timing of distribution, and the volumes offered may be key factors in maximizing acceptance. A unique pragmatic approach used by Potter et al. [12] was to administer a small volume (120 ml) during the drug round as a part of routine medical care. This was well accepted and resulted in a significantly higher energy intake and weight gain compared to the control group. However, there are major organizational and practical challenges to achieving improvements in nutritional intake, and the best methods have still to be established. The main concern of most trials was the effect of supplementation on nutritional status. These results suggest that supplements can have a useful role to play in improving the nutritional status of elderly people. However, older people are very heterogeneous, as reflected in the trials, and some patients groups may benefit more than others. There were insufficient trials of most diagnostic subgroups (such as patients with stroke) to undertake separate meta-analyses. Although the evidence was limited, the meta-analysis suggested that hospitalized patients were more likely to benefit from supplementation than older people in the community, even though most oral supplements are prescribed in primary care. It is essential that we aim to find evidence-based methods to improve the nutritional status of those ‘at risk’ in the community, before they become malnourished or are admitted to hospital. In conclusion, adequately powered, well-designed trials with clearly defined patient groups, both in hospital and in the community, are still required to determine whether supplements can be beneficial in terms of mortality, complications, length of hospital stay, functional status, quality of life and cost-effectiveness.
Acknowledgements We would like to thank the authors who took time to provide further information about the trials. Also, Magnus McGee, Craig Ramsay and Marion Campbell for statistical advice and Dr. Helen Handoll who was also involved in the data extraction of trials. Extramural sources of support were the Medical Research Council UK; Chief Scientist Office of the Scottish Executive Health Department, UK, and the Student Awards Agency for Scotland, UK. The Health Services Research Unit is core funded by the Chief Scientists Office of the Scottish Executive Health Department. These funding sources had no involvement in the study design, collection, analysis and interpretation of data, in writing the report, or in the decision to submit the paper for publication.
Conflict of Interest Statement One reviewer (J.P.) is also an author of one of the trials included [12] but was not involved in the data abstraction for this trial. There are no other conflicts of interest.
118
Oral Protein and Energy Supplementation in Older People References 1 Royal College of Physicians: Nutrition and Patients: A Doctor’s Responsibility. Report of the working party of the Royal College of Physicians. London, Royal College of Physicians, 2003. 2 Agency for Healthcare Research and Quality: HCUPnet: A tool for identifying, tracking and analyzing hospital statistics. Retrieved July 2003 from the World Wide Web:http/ /hcup.ahrq.gov/HCUPNet.asp 3 McWhirter JP, Pennington CR: The incidence and recognition of malnutrition in hospital. BMJ 1994;308:945–948. 4 Weekes E: The incidence of malnutrition in medical patients admitted to hospital in south London. Proc Nutr Soc 1999;58:126A. 5 Potter J, Klipstein K, Reilly JJ, Roberts M: The nutritional status and clinical course of acute admissions to a geriatric unit. Age Ageing 1995;24:131–136. 6 Robinson G, Goldstein M, Levine GM: Impact of nutritional status on DRG length of stay. JPEN J Parenter Enteral Nutr 1987;11:49–51. 7 Larsson J, Unosson M, Ek AC, et al: Effect of dietary supplementation on nutritional status and clinical outcome in 501 geriatric patients – A randomised study. Clin Nutr 1990;9:179–184. 8 Potter J, Langhorne P, Roberts M: Routine protein energy supplementation in adults: Systematic review. BMJ 1998;317:495–501. 9 Baldwin C, Parsons T, Logan S: Dietary advice and oral nutritional supplements for illnessrelated malnutrition in adults (Cochrane Review). Cochrane Library, issue 3. Chichester, Wiley, 2004. 10 Clarke M, Oxman AD. Cochrane Reviewers Handbook 4.1.2. Cochrane Library. Chichester, Wiley, 2001. 11 Milne AC, Potter J, Avenell A: Protein and energy supplementation in elderly people at risk from malnutrition. (Cochrane Review). Cochrane Library, issue 3. Chichester, Wiley, 2004. 12 Potter JM, Roberts MA, McColl JH, Reily JJ: Protein energy supplements in unwell elderly patients – A randomized controlled trial. JPEN J Parenter Enteral Nutr 2001;25:323–329. 13 Payette H, Boutier V, Coulombe C, Gray DK: Benefits of nutritional supplementation in freeliving, frail, undernourished elderly people: A prospective randomized community trial. J Am Diet Assoc 2002;102:1088–1095. 14 Gray DK, Payette H, Boutier V: Randomized clinical trial of nutritional supplementation shows little effect on functional status among free-living frail elderly. J Nutr 1995;25: 2965–2971. 15 Mahoney FI, Barthel D: Functional evaluation: The Barthel Index. Maryland State Med J 1965;14:56–61. 16 Woo J, Ho SC, Mak YT, et al: Nutritional status of elderly patients during recovery from chest infection. Age Ageing 1994;23:40–48. 17 Volkert D, Hubsch S, Oster P, Schlierf G: Nutritional support and functional status in undernourished geriatric patients during hospitalization and 6-month follow-up. Aging Clin Exp Res 1996;8:386–395. 18 Saudny UH, Martin JG, Gray DK: Impact of nutritional support on functional status during acute exacerbation of chronic obstructive pulmonary disease. Am J Resp Crit Care Med 1997;156:794–799. 19 Stableforth PG: Supplement feeds and nitrogen and calorie balance following femoral neck fracture. Br J Surg 1986;73:651–655. 20 Fiatarone MA, O’Neill EF, Ryan ND, et al: Exercise training and nutritional supplementation for physical frailty in very elderly people. N Engl J Med 1994;330:1769–1775. 21 Hickson M, Bulpitt C, Nunes M, et al: Does additional feeding support provided by health care assistants improve nutritional status and outcome in acutely ill older in-patients? A randomized controlled trial 2004;23:69–77. 22 Banerjee AK: Nutritional status of long-stay geriatric inpatients: Effects of a food supplement (Complan). Age Ageing 1978;7:237–243. 23 Barr SI, McCarron DA, Heaney RP, et al: Effects of increased consumption of fluid milk on energy and nutrient intake, body weight and cardiovascular risk factors in healthy older adults. J Am Diet Assoc 1999;7:810–817. 24 Broqvist M, Arnqvist H, Dahlstrom U, et al: Nutritional assessment and muscle energy metabolism in severe chronic congestive heart failure – Effects of long term dietary supplementation. Eur Heart J 1994;15:1641–1650.
119
Oral Protein and Energy Supplementation in Older People 25 Brown KM, Seabrook NA: Nutritional influences on recovery and length of hospital stay in elderly women following femoral fracture. Proc Nutr Soc 1992;51:132A. 26 Carver AD, Dobson AM: Effect of dietary supplementation of elderly demented hospital residents. J Hum Nutr Diet 1995;8:389–394. 27 Deletter MC: A Nutritional Intervention for Persons with Chronic Airflow Limitation; PhD thesis, Lexington, University of Kentucky, 1991. 28 Delmi M, Rapin CH, Bengoa JM, et al: Dietary supplementation in elderly patients with fractured neck of the femur. Lancet 1990;335:1013–1016. 29 Gariballa SE, Parker SG, Taub N, Castleden CM: A randomized controlled a single-blind trial of nutritional supplementation after acute stroke. JPEN J Parenter Enteral Nutr 1998; 22:315–319. 30 Gegerle P, Bengoa JM, Delmi M, et al: Dietary survey on the effect of an oral nutritional supplement after femoral neck fracture. Schweiz Rundsch Med Prax 1986;75:933–935. 31 Hankey CR, Summerbell J, Wynne HA: The effect of dietary supplementation in continuing care elderly people: Nutritional anthropometric and biochemical parameters. J Hum Nutr Diet 1993;6:317–322. 32 Hankins C: Dietary Supplementation with Sustagen in Elderly Patients with Fractured Neck of Femur; PhD thesis, Sydney, Sydney University, 1996. 33 Hubsch S, Volkert D, Oster P, Schlierf G: Effect of dietary supplementation on weight and protein status in undernourished geriatric patients. Clin Nutr 1992;11:97. 34 Jensen MB, Hessov I: Dietary supplementation at home improves the regain of lean body mass after surgery. Nutrition 1997;13:422–430. 35 Krondl M, Coleman PH, Bradley CL, et al: Subjectively healthy elderly consuming a liquid nutrition supplement maintained body mass index and improved some nutritional parameters and perceived well-being. J Am Diet Assoc 1999;99:1542–1548. 36 Kwok T, Woo J, Kwan M: Does low lactose milk powder improve the nutritional intake and nutritional status of frail older Chinese people living in nursing homes? J Nutr Health Aging 2001;5:17–21. 37 Knowles JB, Fairbarn MS, Wiggs BJ, et al: Dietary supplementation and respiratory muscle performance in patients with COPD. Chest 1988;93:977–983. 38 Lauque S, Arnaud BF, Mansourian R, et al: Protein-energy oral supplementation in malnourished nursing-home residents. A controlled trial. Age Ageing 2000;29:51–56. 39 MacFie J, Woodcock NP, Palmer MD, et al: Oral dietary supplements in pre- and postoperative surgical patients: A prospective and randomized clinical trial. Nutrition 2000;16:723–728. 40 Madigan C: Benefits of Dietary Supplementation in Elderly Patients with Fractured Neck of Femur; MSc dissertation, Sydney University, Sydney, 1994. 41 McEvoy AW, James OF: The effect of a dietary supplement (Build-up) on nutritional status in hospitalized elderly patients. Hum Nutr Appl Nutr 1982;36A:374–376. 42 McWhirter JP, Pennington CR: A comparison between oral and enteral nutritional supplements in malnourished patients. Nurition 1996;12:502–506. 43 Meredith CN, Frontera WR, O’Reilly KP, Evans WJ: Body composition in elderly men: Effect of dietary modification during strength training. J Am Geriatr Soc 1992;40:155–162. 44 Ovesen L: The effect of a supplement which is nutrient dense compared to standard concentration on the total nutritional intake of anorectic patients. Clin Nutr 1992;11:154–167. 45 Schols AM, Soeters PB, Mostert R, et al: Physiologic effects of nutritional support and anabolic steroids in patients with chronic obstructive pulmonary disease. A placebo-controlled randomized trial. Am J Resp Crit Care Med 1995;152:1268–1274. 46 Vlaming S, Biehling A, Hennessey EM, et al: Should the food intake of patients admitted to acute hospital services be routinely supplemented? A randomized placebo controlled trial. Clin Nutr 2001;20:517–526. 47 Yamaguchi LY, Coulston AM, Lu NC, et al: Improvement in nutrient intake by elderly mealson-wheels participants receiving a liquid nutrition supplement. Nutr Today 1998;33:37–44.
Discussion Dr. Lochs: My question is, are we looking at the right parameters? We nutritionoriented people are always very cautious about making statements. If I were at an
120
Oral Protein and Energy Supplementation in Older People oncology conference, they would say look at the wonderful results we have, we have a 2-week later disease progression. There was of course no effect on mortality and there was a longer stay in the hospital due to chemotherapy, but this is wonderful and the whole world will start to use this treatment. Now we look at the hardest end point, we say well mortality, even if there is a small extension of life it is not that important and there are complications. How was functionality in old patients, this is very important; how was the quality of life, and what end point should we rather look at with nutrition? Dr. Milne: I absolutely agree with what you say. The problem with doing systematic reviews is that you can only use what the trials have provided in terms of outcomes. A lot of these trials have been designed to look at nutritional outcomes and have reported mortality. Some of the more recent ones are focusing more on outcomes that are important to patients and treatment providers. There are more recent trials that have looked at functional outcomes such as handgrip and quality of life outcomes [1, 2]. Just recently I have been able to look at a meta-analysis of handgrip strength studying the changes in handgrip, but there was no significant effect on handgrip. I will also present the findings in terms of quality of life and functional outcomes in a descriptive section of the updated Cochrane Review rather than using a meta-analysis method. Dr. Lochs: Couldn’t that be one thing this conference could do to make recommendations? What should we study in the future? Dr. Hébuterne said nutritional status is more or less what we are looking for. You are also looking in the same direction and then we end up with a very thin database. Perhaps we should then say our recommendations for future studies would be to look at this and this and this parameter. I don’t know but perhaps it is that something we could end up with this conference. Dr. Milne: That would be very useful. I think we also have to take into account the different diagnostic groups and the different outcomes which are relevant to those, for example, for patients with chronic obstructive pulmonary disease (COPD) you may be interested in lung function, and in patients with hip fracture you may be interested in how soon they are up and walking and their ability to walk and that sort of thing. So it will be difficult to find something that suits all these different trials, but I think it will be useful to get some kind of consistency so we can actually look more carefully at these outcomes. Dr. Roessle: It is already difficult to design double-blind randomized clinical trials for parenteral and enteral nutrition. If you are designing trials for oral nutrition the task is even more complicated because by definition you have no placebo treatment and you cannot run a true double-blind randomized clinical trial. I wonder whether in your meta-analysis you have been really fair in picking up the trials because the only way to do this is to compare with no treatment and this is not a blinded way to analyze. If you look at all the other trials which are not randomized or not comparative, they might also give valuable results because the ideal trial is not possible. I would like to have your advice on how to design future trials with oral supplements to be as fair as possible. Dr. Milne: I think in some way it is easier to design trials of oral supplements than it is to design trials of enteral or parenteral feeding. What you need to do is compare your provision of supplements with usual care. I think the problem is that there are deficiencies in the way we use supplements at the moment and there are ways in which we can provide better care in terms of oral supplementation. So if we design an intervention which we think is practical and usable on a ward situation, hopefully that will go beyond what is usual practice at the moment, and I don’t think there is any ethical problem then with developing a trial. I can give you the example of one of my coworkers. Potter et al. [3] recently ran a trial of oral supplements which was designed to compare usual practice, which may include interventions by the dietician and
121
Oral Protein and Energy Supplementation in Older People anything that would normally happen on the ward, with providing small quantities of more energy-dense oral supplements which were prescribed on the drug cardex and then administered to the patients. This was a very pragmatic approach, but a lot of the problems with supplements have to do with wastage, meaning that these supplements may sit on the bedside lockers and don’t get consumed, so there is wastage. So it helps overcome such problems. I agree that we also need to take other kinds of nonrandomized trials into account and not just focus completely on randomized trials. Dr. Elia: One of the advantages of undertaking meta-analyses such as these is that you may be able to identify groups of patients who may benefit from particular interventions and those who don’t. In relation to functional outcomes it seems reasonable to assume that those people who aren’t malnourished may not respond to the supplement in the same way as those who are malnourished and therefore have already lost some function. Some studies and meta-analyses have split the groups of patients according to some form of index of nutritional status, one of the simplest of which being body mass index. Have you considered this issue in your meta-analysis? Dr. Milne: I haven’t looked at functional outcomes in terms of nutritional status, but I think it would be useful to do that. Dr. Elia: All outcomes including mortality would be interesting to look at. Dr. Milne: I have looked at mortality and divided the results looking at the subgroup analyses of nutritional status, those who were defined as at nutritional risk compared with others who were not but who hadn’t been reported as being at nutritional risk. Certainly the evidence for mortality suggested that those who had been defined as being at nutritional risk were benefiting more. Dr. Powell-Tuck: Many congratulations on this meticulous work, it is very helpful indeed. Firstly a follow-up on what Dr. Elia was saying. I think there are some patients in whom we are trying to maintain weight, maintain quality of life, and in these of course you are not going to expect to see a change. The positive outcome is no change, in other words it is not worsening. That is the observation. The question is, you had 8 trials for respiratory diseases and I wondered why you haven’t gone ahead and analyzed those, and whether you could make any comments about that particular group of illnesses? Dr. Milne: There were 8 trials of COPD patients. I think all these diagnostic groups may have different responses to supplementation and that is certainly true in patients with COPD. I didn’t look at mortality as a subgroup because there was not sufficient information available for the outcomes of interest, in that mortality data were not there. In the review I reported the functional outcomes which were things like change and lung performance, but again from the way that the results were provided in the articles, I was not able to combine them. So it will be descriptive rather than using a meta-analysis. Dr. DeLegge: Just two questions for you. One is, would these people in the control groups have free access to food, meaning if they wanted an extra candy bar or Snickers or Coca Cola they could have it if they wanted it? The second question is, one of the more difficult problems we have with oral supplements is compliance and in fact our recent hospital-based survey showed that compliance with oral supplements was about 18% of the patients. So I don’t think you presented the intention-to-treat data, but if you look at intention-to-treat sub-analyses, did you have similar results? Dr. Milne: The results were presented as they were described in the trials, so some used an intention-to-treat analysis and some did not. So I didn’t split them by an intention-to-treat analysis. But I agree that a lot of the problems with our oral supplements is the fact that patients aren’t very compliant so in some ways we need to find better ways of providing oral nutrition support. There is a lack of trials which look at other methods, for example fortifying normal food sources, or even trials which provide extra support or extra help for patients in terms of helping them eat their meals
122
Oral Protein and Energy Supplementation in Older People or take their supplements. So there are different ways that we should be looking at in terms of helping patients in hospital and, also more importantly, the evidence is much poorer in terms of community-living older people, and trying to tackle nutritional risk before patients are admitted to the hospital in the first place. Dr. DeLegge: Just make sure you keep a positive spin. Dr. Thomas: A really nice job and I know you are limited into a large extent by the trials you have to work with. One of things that seems clear from looking at the data is that there are populations of people who do not respond at all to nutritional supplements and there are populations that appear to respond. Since you did have a positive effect, all be it a small effect, it would be nice to see if the trials would allow you to eliminate certain cachexia states that have not been shown to respond at all, and then see if you have enough strength left in the trials of people who are likely to respond to increase the magnitude of the effect. That may or may not be possible but I would love to see you try to do that. Dr. Milne: I think the inflammatory response was discussed yesterday; people with different diseases will have different responses. From what I understood, practically every disease has some degree of inflammatory response but perhaps some diseases like COPD will be more affected. At the moment I don’t feel I have enough knowledge to try and start splitting them by those that may have that response. It is something that will perhaps be done in the future. Dr. Labadarios: I would like to support Dr. Loch’s proposal regarding the criteria for such studies. It is rather important that we collectively do something about it otherwise in 10 years time we will probably be sitting in a different place but we will have the same results. The quality of the trials is too poor to make a difference or to draw a conclusion, and I think that it reflects badly on the leadership of nutrition quite honestly. Somehow everybody around this table had input in getting these protocols approved. So I think it is a constructive and worthwhile undertaking which should be supported. The question is, and I hope I understood your slide correctly, the in-hospital population appeared to do significantly better than the community-based population. Could you expand a little bit on this, were you able to decide why that was the case? Dr. Milne: There were 10 trials in the hospital and only 5 in the community, so actually there is less evidence from the community as opposed to evidence of no effect. In terms of compliance it seems to be more of a problem with patients in the community taking their supplements than those being supervised in hospital. Some of the recent trials with community patients would suggest that compliance is a problem. Dr. Bozzetti: I have a couple of questions. First, what is the reason for dividing patients into groups receiving more or less than 400 kcal? Would it not be more informative or useful to express the kilocalories per kilogram of body weight? Secondly, is there a minimum time necessary to observe a benefit with supplements? Dr. Milne: These were rather arbitrary cutoffs established before we actually looked at the trials. We didn’t know how many calories were going to be provided. But just looking at those who were given fewer calories compared to more calories, I think there is a limit to how much you can read from the results because a lot of it depends on compliance, and even if you have fewer calories but actually consume them all, it is likely to be more important than being offered 1,000 kcal/day and only taking 500. I think it is probably more important to look at the length of time patients were supplemented and also the length of time patients were followed up, which is another deficiency in the trials included because really we are interested in outcome at 6 months, a year down the line rather than within a month of supplementation. So it was an attempt to try and look quantitatively at those who were given either more or less supplements for longer or shorter periods, but there is a limit to how much you can read into the results, I agree.
123
Oral Protein and Energy Supplementation in Older People Dr. Correia: I am going to go further into the Dr. Labadario’s and Dr. Loch’s comments. As those who are leaders in thinking in the area of nutrition, we should try to establish some basic concepts. I go further by saying that we are nowadays discussing molecular nutrition, and we forget basic things such as should we calculate our energy requirements based on actual or ideal body weight. We even forget or we even don’t know exactly the definition of what malnutrition is and how to assess malnutrition. So I think for the future, in order to have better studies, we definitely should define these basic issues so that when we analyze independently, whether it is a meta-analysis or not, we can come to the same conclusion based on what we are referring to as malnutrition. We have recently reviewed data on oral supplements because my group is completing a paper on the use of oral supplements. One of the difficulties we found is that the articles do say that they gave an average of 100 kcal as a supplement, but most of them forgot to tell us that if by feeding these patients with oral supplements what the amount of the oral food was. So my question is, when we give supplements aren’t we changing food for another artificial nutrition? Aren’t these patients decreasing what they are eating? From the articles I have read, I can tell you that 90% of them did not have a simple sentence about the amounts of calories that these patients took while eating regular food. So I think it is very difficult to analyze all of these data based on results if we don’t measure what the patients really eat. Dr. Milne: A lot of the trials that I included made some attempt to measure the intake of the patients. I admit that the methods they used were not ideal and there is probably a potential for quite a lot of error, but the results would suggest that patients do increase their intake with oral supplements although, as you can see from the effect on weight change, it is only 2.5%, which for a 50-kg person may mean that they only gain less than 1.5 kg. If they were consuming all the supplements in addition to a normal diet you would expect the weight change between the two groups to be more. So there is probably some compensation going on in terms of normal diet but I think we can see that they actually do increase their intake, but probably not as much as we would like, and most studies reported good compliance with supplements. I think a difference may be between what happens in the trials and what happens in a normal ward situation. Dr. Elia: We have looked at the effect of supplements in suppressing food intake. The results from the literature are varied, but in general it seems that most of the supplements in the studies that we were able to analyze added to food intake. This is interesting in itself, but the extent to which it was additive varied according to nutritional status. Those who were undernourished, defined as a body mass index of less than 20, were more likely to have a higher additive intake than those who were wellnourished. The effect tended to decrease with time. Dr. Milne: I accept your point. Dr. Biesalski: There might be one other problem with those studies. If you include fortified food, etc., and keep in mind that the balanced or misbalanced micronutrient status might have a big impact on quality of life, B vitamins and depression or lowdose multivitamins for example might have an influence on the immune system. So the studies have to be separated between those who add some B vitamins and those who allow that they might be added. This is another point that might have a big impact on the outcome of these studies. Dr. Powell-Tuck: I want to follow on Dr. Elia’s comment, and I actually disagree with Dr. Correia because I think the way you design a trial depends on the question you are asking. If firstly, as Dr. Elia said, we have excellent evidence not only from his group’s work but also from your own group’s work, I think that supplementation is effective in increasing intake. So that may not be the question that you are seeking to answer again because it has been answered very fully already. Also if one was on full dietary assessment of every patient, one is liable to confine one’s trials to rather small
124
Oral Protein and Energy Supplementation in Older People numbers because it becomes impractical to do this on a large scale. So if you are actually asking the question whether a particular intervention with a particular supplement is beneficial on length of stay, it may be quite unnecessary to make the measurements of additional food intake. You are interested in the outcome of that intervention and that way you may have the chance to do the big scale studies, which I believe we need in this field on a practical and financially basis. Dr. Milne: I absolutely agree, and this is the type of trial that Potter et al. [3] did with acute elderly patients. It was a pragmatic practical trial and they measured the clinical outcomes. So basically we get the answer that we need rather than looking at whether the patients have increased their intake or whether their nutritional status has changed. Dr. Lochs: I agree with you, but doesn’t it help to interpret the results if you know the supplements reduced the oral intake, and then you have an easier time to interpret if you find something or not? Dr. Milne: One time when it will be really useful to do, is if you run a trial and find no clinical effects at all. At least if you have actually looked at changes in intake then you can perhaps say the reason why we didn’t find an effect was that patients didn’t actually increase their intake, and I think that is an important point. Dr. Powell-Tuck: May I just come back. In the ideal world of course one would love to have all the data, but what I worry about is the practicalities here. You have to decide what you are going to do; you decide your budget; what you can achieve, and I think it is very important sometimes to simplify things down so that you could do large numbers to really answer the question does an intervention have an effect. I think we sometimes focus too much on nutritional outcome and not enough on economically important outcome. That is the point I am trying to make. Dr. DeLegge: That is why I think in these particular trials the intention-to-treat seat is important because you flash out all those ideas; meaning if you have a new drug but the compliance at home is poor when the drug has to be taken 6 times a day, the drug, although it may be great, may not be very effective practically. So is there any possibility of reanalyzing the intention-to-treat studies that you have? Dr. Milne: I think it is something that would be a possibility. The data are limited so you are starting to getting down to only a few studies. But I absolutely agree, I think a lot of the older studies were done on efficacy rather than effectiveness, and there are not many trials which you could say were pragmatic trials where they haven’t actually brought in people to support the trial and make sure the patients take their supplements. But that still leaves us with the problem of not really knowing what happens in practice and not having interventions which we know if they are put in place on a ward will have a beneficial effect because they take into the account the problems that happen on the wards. Furthermore, you can’t look at which interventions are likely to be more cost-effective unless you look at pragmatic trials.
References 1 Edington J, Barnes R, Bryan F, et al: A prospective randomised controlled trial of nutritional supplementation in malnourished elderly in the community: Clinical and health economic outcomes. Clin Nutr 2004;23:195–204. 2 Payette H, Boutier V, Coulombe C, Gray-Donald K: Benefits of nutritional supplementation in free-living, frail, undernourished elderly people: A prospective randomized community trial. J Am Diet Assoc 2002;102:1088–1095. 3 Potter JM, Roberts MA, Reilly JJ, McCoil JH: An evaluation of protein energy supplementation in medically ill admissions to a geriatric unit. Proc Nutr Soc 1998;57:88A.
125
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 127–142, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
Efficacy of Enteral and Parenteral Nutrition in Cancer Patients Federico Bozzettia and Valentina Bozzettib aOspedale
di Prato, Prato, and bUniversità Vita e Salute, San Raffaele, Milan, Italy
Before attempting to analyze the potential efficacy of the two different routes of administering nutrients to cancer patients, enteral nutrition (EN) versus total parenteral nutrition (TPN), it should be appreciated that patients who are usually fed via a vein are not the same as those who receive EN. If fact, nowadays, the option for TPN only emerges if a patient is not suitable for EN because his/her gut is not working. Consequently the different effects of EN and TPN cannot be attributed only to the administration route of nutrients, but also to the different basic conditions of these 2 groups of patients. There are, however, some exceptions: at the beginning of the era of artificial nutrition, TPN was more developed than EN probably because the impetus of this new approach relied on an urgent need to solve the problems of patients with intestinal failure. In fact, initially TPN was developed in surgical departments to meet the nutritional requirements of patients with short-bowel syndrome or abdominal catastrophes. This translated into the use of TPN also in patients with a working gut, and finally rendered a comparison between TPN and EN not only possible, but even scientifically rational and ethically acceptable. At this point we were roughly in the 1980s. Another field where TPN and EN are frequently compared is the perioperative setting. Nutritional support in the perioperative period is not nutrition in the traditional meaning of the word: it is the attempt to control the metabolic reaction to surgical trauma and to potentiate the defenses of the patient through the administration of nutrients. The indication for nutritional support in this patient population was questioned for a long time and the simple provision of a saline solution for a few days was usually accepted in many institutions. Consequently in the literature there are studies comparing not only different nutritional regimens and 127
EN vs. TPN in Cancer Patients routes of administration in the perioperative period, but also nutritional support with saline solutions. Finally, today the vast majority of patients receiving artificial nutrition is fed enterally or intravenously depending on whether the gut is working or not. Moreover, the current idea nowadays is that maintaining a working gut is beneficial for patients and they are often administered minimal enteral feeding in order to accomplish this aim even though macronutrient requirements are mainly covered by the intravenous route. EN and TPN have now completely reversed their original positions: in the 1970s, TPN was given to some patients even though they were able to be fed enterally, whereas now strong emphasis is put on giving some nutrients enterally even if the majority of the nutritional support can only be administered venous route.
Prospective Controlled Studies on Metabolic Effects of TPN and EN Only a few randomized [1–3] or prospective and controlled [4] studies have compared TPN and EN (table 1). It appears that TPN was more effective than EN in promoting weight gain [3] even though this gain may simply be due to accrual of fat or water. Nevertheless TPN was able to maintain a better nitrogen balance and plasma amino acid level [1, 3] and a positive balance of K, Na, Cl, P and Mg [4]. In their study in 1983 Burt et al. [3] did not find any difference between EN and TPN as far as whole body flux, protein synthesis and catabolism were concerned. However, during the control (basal) period, whole-body protein catabolism was uniformly and significantly higher than synthesis, and during the period of nutritional repletion through EN or TPN the rate of whole-body synthesis tended to be greater than that of catabolism. In addition, during the period of nutritional support, the percentage of nitrogen entering the metabolic pool that was derived from catabolism of protein was significantly decreased in both EN and TPN groups. The study by Dresler et al. [5] showed that before nutritional repletion, the flux of nitrogen entering the metabolic pool originated solely from the breakdown of body protein and about 75% of this flux was utilized in protein synthesis. However, when an exogenous supply of amino acids was introduced in cancer patients, utilization efficiency decreased to 58% during EN and to 39–43% in cancer patients during TPN [3, 6]. There appeared to be a definite advantage for the enteral route in terms of utilization for synthesis of body proteins. Overall, both EN and TPN tend to stabilize the nutritional status and whole-body protein economics and, depending on the type of metabolic parameter adopted, TPN or EN appears to be slightly more effective. 128
EN vs. TPN in Cancer Patients Table 1. Prospective studies on the metabolic effects of EN vs TPN [1–4] Nutritional parameter Balance of N P K Na Cl Ca Mg Change in H2O Albumin CHI TSF MMA N balance Weight Albumin Glucose turnover rate Gluconeogenesis from alanine Plasma amino acid level Weight N balance Albumin Transferrin Ceruloplasmim TBK 3 Met-His
EN
TPN
positive positive negative negative negative positive negative
positive more positive positive positive positive negative positive
positive no change positive positive positive positive after 7 days ↑ 1% after 30 days ↑ 7.4% ↑⫻4 suppressed maintained no change no change ↓ no change no change no change no change
positive positive positive positive positive positive after 1 day ↑ 6% after 30 days ↑ 6.3% ↑⫻4 suppressed increased ↑ ↑ no change no change no change no change no change
p
0.05 0.05 0.05 0.05 0.05
0.05 0.03
Nixon et al. [4]: EN vs. TPN (kg/day): 38 kcal and 0.9–1.2 vs. 1.2–1.8 g AA. Lim et al. [1]: EN vs. TPN (kg/day): 63 vs. 62 kcal and 2.1 vs. 1.5 g AA. Burt et al. [3]: EN vs. TPN (kg/day): 49 vs. 49 kcal and 1.5–2 vs. 2.5 g AA. CHI ⫽ Creatinine/height index; TSF ⫽ triceps skinfold; MMA ⫽ muscle mass area; TBK ⫽ total body potassium; 3 Met-His ⫽ 3-methylhistidine.
Randomized Controlled Trials on the Efficacy of Perioperative TPN and EN Whereas there are a large number of articles comparing EN to TPN, the number decreases when randomized studies are considered. These studies only include cancer patients (both as a total or as the majority of the series) are homogeneous with regard to the nutritional status and are particularly well matched as far as the calorie and nitrogen intake of the nutritional regimens of the 2 groups are concerned. 129
EN vs. TPN in Cancer Patients Table 2. RCT comparing EN (tube feeding) with TPN surgical cancer patients Reference, year
Malnourished Hamaoui et al. [9], 1990 Iovinelli et al. [10], 1993 Sand et al. [11], 1997 Shirabe et al. [12], 1997 Total Mean Non-malnourished Lim et al. [1], 1981 Reynolds et al. [13], 1997 Sako et al. [14], 1981 Von Meyenfeldt et al. [15], 1992 Total Mean
Number of patients
19 48 29 26 122
24 67 69 101 261
Complications, %
Mortality, %
EN
EN
TPN
TPN
18 25 46 8
12.5 54 50 61.5
9 0 0 0
0 0 6 0
24.2
44
9
6
100 72 27 80
75 74 29 108
17 6 0 8
8 3 8 4
68
71
10
6
There are in fact some excellent metabolic studies [7, 8] demonstrating the benefits on whole-body protein kinetics and the hormonal pattern with postoperative EN as compared with simple standard intravenous nutrition. However, they do not solve the dilemma of whether the benefit is due to the quality of the nutritional regimen or to the route of administration of nutrients. Table 2 summarizes the clinical outcome in 8 randomized controlled trials (RCTs) [1, 9–15] comparing EN (tube feeding) and TPN in malnourished and non-malnourished cancer patients. Interpretation of these studies, which belong to the past century is not easy: some authors counted the number of patients with one or more complications, others computed the number of single complications. Moreover these studies only included a small number of patients, hence the statistic power appears to be very small. To my knowledge there are only 3 RCTs that compared two isoenergetic and isoprotein regimens, one by vein and the other by the enteral route, in a large number of patients with gastrointestinal cancer. A study published by Braga et al. [16] included 257 patients, less than half of whom had weight loss of ⬎10%, and serum albumin was approximately 3.7 g/dl. The nutritional intake was 24.4 and 23 kcal/kg/day in the TPN and EN groups, respectively. Overall, complications occurred in 56.4 and 49.2% of TPN and EN patients, respectively (not statistically significant). The authors considered the subgroup of malnourished patients in a post hoc analysis, and found a significantly shorter duration of hospital stay in EN than TPN patients. 130
EN vs. TPN in Cancer Patients In the second study, Pacelli et al. [17] randomly assigned 241 patients (⬎90% with gastrointestinal cancer) with moderate malnutrition (15% body weight loss and albumin 3.6 g/dl) to a daily nutritional regimen of 24.8 (TPN) or 25.3 kcal/kg/day (EN), respectively. Major complications occurred in 39.3 and 37.8% of TPN and EN patients (not statistically significant), respectively, with no difference in mortality (2.5 and 5.9%, respectively). The duration of postoperative stay was 15 and 16 days in the EN and TPN groups, respectively. The third study was launched by the Italian Society for Parenteral and Enteral Nutrition (SINPE) [18] and was conducted in 317 patients with moderate to severe malnutrition (13–14% body weight loss, serum albumin 3.4–3.5 g/dl and 1,541–1,573 peripheral lymphocytes/mm3). Those patients were randomly assigned to early EN or TPN (mean 26 kcal/kg/day and 1.4 amino acids/kg/day) for 1 week, starting on the 1st postoperative day. There was a benefit for the EN versus TPN patients with respect to postoperative complications (34.0 and 49.4%, respectively; risk differential 15%; p ⱕ 0.02) and to postoperative stay (13.4 and 15.0 days, respectively; p ⫽ 0.009). Adverse effects of nutritional support were more frequent in EN patients (35%) than in TPN patients (14%; p ⬍ 0.001), however. How can these conflicting results be reconciled? A possible explanation is that the advantage of EN is confined to malnourished patients. The majority of patients in the study of Braga et al. [16] were not malnourished, those in the study of Pacelli et al. [17] were moderately malnourished, and those in the SINPE study [18] were more malnourished because their serum albumin was somewhat lower and peripheral lymphocytes were often below the normal range; only within this latter group of patients was there a clinical benefit from EN. Quite recently a large study [19] on 54,215 patients reported that in regression models, the albumin level was the strongest predicting factor of surgical mortality and morbidity, and similarly Kudsk et al. [20] confirmed on 526 candidates for abdominal surgery that preoperative albumin correlated inversely with complications, length of stay, postoperative stay, intensive care unit stay, mortality and resumption of oral intake. They write that ‘this lack of appreciation for nutritional risk … can explain discrepancies in outcome noted in several randomized, prospective nutrition studies’. More precisely, by categorizing the patients according to the level of serum albumin at intervals of 0.5 g/dl, they were able to demonstrate that each class of patients had a statistically different frequency of major complications. Second, malnourished patients need to receive an adequate amount of protein if they are to benefit from EN. It is worth noting that the SINPE study [18] patients received 30% more nitrogen daily in comparison with those from the study by Pacelli et al. [17]. This might account for the quicker and safer recovery from surgery in the former group. Finally, criticism concerning the quality of complications that could account for the discrepancy in benefits from EN [21] is not substantiated by 131
EN vs. TPN in Cancer Patients 1.0
Probability of complications
0.9 0.8 0.7 0.6 0.5 0.4 0.3 SIF TPN EN IEEN
0.2 0.1 0.0 0
5
10
15
20
25
30
35
Weight loss (%)
Fig. 1. Probability of postoperative complications by the percentage of weight loss of two patients’ populations (upper curve at high surgical risk, lower curve at low surgical risk) depending on the type of nutritional support: SIF ⫽ Standard isotonic fluid; TPN ⫽ total parenteral nutrition; EN ⫽ enteral nutrition; IEEN ⫽ immune enhancing enteral nutrition.
the fact that patients randomized to EN in the SINPE trial also had a shorter postoperative stay. The reasons for the beneficial effects of EN versus TPN in this setting are largely unknown. It is interesting, however, that Reynolds et al. [13], who conducted a RCT of early EN feeding versus TPN (an almost isocaloric and isonitrogenous regimen) in 67 patients after major gastrointestinal surgery, observed no significant difference between the groups with respect to intestinal permeability, acute-phase response and endotoxin exposure. In an attempt to further clarify the complex relationship between nutritional status (weight loss), postoperative complications and type of nutritional support, data on 1,410 patients entered in RCTs on perioperative nutrition, coordinated by the Istituto Tumori of Milan and San Raffaele Hospital of Milan over the last 8 years, were reviewed. Patient data were obtained from electronic databases specifically created to achieve charts of all those subjects included in RCTs designed to test the effect of different nutritional approaches on clinical outcome [16, 18, 22–25]. Figure 1 shows that, in groups of patients at varying risk of surgical complications, low or high (depending on level of serum albumin, age and type of surgery), the probability of complications changes according to the degree of weight loss, and EN and immune-enhancing EN are always better than TPN and standard isotonic fluids. 132
EN vs. TPN in Cancer Patients Home EN and TPN Survival of cancer patients on home artificial nutrition is largely dependent on the severity of the basic disease. Unfortunately many series mix together patients with simple sequelae of cancer, i.e. radiation enteropathy or chronic surgical complications or mutilations, patients with active cancer who are receiving oncologic therapy and finally patients with incurable cancer. For this last group there are divergent opinions regarding the indication both between different countries and different institutions within the same country. Therefore, disparity in the composition of the series and indications for home artificial nutrition can account for different survival rates in reports on patients receiving home TPN or home EN, and when comparing the two procedures. In a series of small retrospective studies of patients receiving home TPN [26–32], the median survival ranged from 53 to 120 days. More reliable are studies reporting large institutional experiences or data from national registers, or pooling data from several nations. Table 3 shows that most of the 1-year survival rates range between 20 and 30% [33–37]. This figure is in keeping with data of the British Artificial Nutrition Survey [38] which in 1999 reported a 26% survival at 1 year. If one considers more favorably selected series of patients (non-terminal cancer patients), a survival rate of 38% at 5 years is observed [39]. The survival rate from the Register of the Italian Society for Parenteral and Enteral Nutrition is shown in figure 2. The survival rate in some series of patients receiving home EN is reported in table 3 [33–35]. It is noteworthy that in some of them survival at 1-year surpasses 30%. Moreover data from the British Artificial Nutrition Survey show that in patients with upper gastrointestinal cancer the 1-year survival is 70–80% [41], and in the series of the Nice group [42, 43] the 1- and the 4- to 5-year survival rates were 38.8 and 23.8–21.7% respectively. Data on quality of life (QoL) are more sparse. There are some retrospective analyses [27, 29, 39] which would indicate a very limited benefit in patients on home TPN: the Karnofsky performance score increased in 7% of patients after 1 month [28], and in 68% of patients surviving longer than 3 months [29], the ability to sustain daily activities and oral intake improved in 27% of patients [44]. According to the data of the North American Register, 31% of patients appeared to be completely rehabilitated at 1 year. An ad hoc study has recently been published by Bozzetti et al. [31]. Sixty-nine advanced cancer patients enrolled in a program of home parenteral nutrition in 6 different Italian centers were prospectively studied with regard to nutritional status (body weight, serum albumin, serum transferrin and total lymphocyte count), length of survival and QoL through the Rotterdam Symptom Checklist questionnaire. These variables were collected at the start of home parenteral nutrition and then at monthly intervals. All these patients were severely malnourished, almost aphagic, and beyond any possibility of cure. They reported a median survival of 4 (range 1–14) months with about 133
EN vs. TPN in Cancer Patients Table 3. Survival of cancer patients on home TPN and EN Reference, year
Period of study
Number of patients
Survival
Home TPN Howard [33], 1993
1985–1990
1,672
Howard et al. [34], 1995 Messing et al. [35], 1998 Van Gossum et al. [36], 1999 Howard [37], 2000
1985–1992
2,122
28% at 1 year; median 6 months; mean 4 months 37% at 1 year
1993–1995
524
19.5% at 6 months
1997
200
26% at 6–12 months
1984–1988
1,073
SINPE Register 2004a
1980–2003
1,103
Home EN Howard [33], 1993
1985–1990
1,296
1989–1992
1,644
1992–1999
3,690
Howard et al. [34], 1995 Gaggiotti et al. [40], 2001
aCourtesy
25% at 1 year; median 6 months 20% at 1 year; median 6 months 32% at 1 year; mean/ median 6 months 41% at 1 year Mean Head-neck Esophagus-stomach Colon-rectum Biliary Lung
18.3 13.5 21.8 16.8 10.7
of Dr. A. Palmo.
one third of the patients surviving more than 7 months, and nutritional indices maintained stable until death. QoL parameters remained stable until 2–3 months before death. The authors concluded that home TPN may benefit a limited percentage of patients who may survive longer than the time allowed by a condition of starvation and depletion. Provided that these patients survive longer than 3 months, there was some evidence that QoL remains stable for some months and acceptable for the patients. With reference to home EN, it appears from data of the National Register Information of the USA [32] that 79% of patients were able to achieve full rehabilitation. Similarly some retrospective studies report that 75% of cancer patients on home EN had a subjective improvement of QoL [46]. In a prospective study including a mixed population of cancer and non-cancer patients, Loser et al. [42] reported an improvement in Karnofsky and Spitzer indices and in the EORTC QLQ C 30 self-rating questionnaires. 134
EN vs. TPN in Cancer Patients 1.0
Survival
0.8 0.6 0.4 0.2 0.0 0
1
2
3
4
Years
Fig. 2. Survival of incurable cancer patients on home TPN.
Schneider et al. [43] prospectively studied the QoL-related parameters in 38 patients (11 with cancer) and reported that patients on home EN have a poorer QoL than the age- and sex-matched general population, probably because of their underlying disease. A more complete study is that by Robergé [49] who prospectively investigated 39 cancer patients using 3 questionnaires, EORTC QLQ-C30, EORTC HEN 35, OES 24, that were to be filled in after 1 week of hospital discharge and 3 weeks later. They found a slight improvement in QoL and in symptoms such as constipation, coughing, social functioning and body image/sexuality. However, one third of patients felt uncomfortable with body image.
Conclusions The scenario of enteral and parenteral nutrition in cancer patients is not uniform. Not only do the final effects depend on the type of evaluation, metabolic versus clinical, but for both the outcome is affected by the duration of the nutritional support and by the biological-clinical aggressiveness of the malignancy. The less aggressive the tumor, the less the impact on the nutritional status is and the longer the time allotted to the artificial nutrition is to improve the general status of the patients. In the short-term, both EN and TPN seem to be equivalent. The metabolic benefits sometimes described for TPN versus EN or vice versa are not strong enough to endorse the choice of one route over the other. Should both routes be available, the option may depend on the availability of proper access, the need of particular formulations and other clinical and logistic factors which may render nutritional support more practical and easier in that single case. Table 4 summarizes the advantages and disadvantages of both approaches with a particular emphasis on some of these points that can finally lead to the choice of one or the other approach. 135
EN vs. TPN in Cancer Patients Table 4. Advantages and disadvantages of TPN and EN Advantages TPN As much as is needed can be given A working gut is not required Regimen may be adjusted without withdrawal Better modulation of substrates
Disadvantages More expensive More demanding Potentially more dangerous More prone to stimulate tumor growth
Compliance may be better (patients may harbor a CVC for other purposes) EN Simple Low cost Metabolically better Less stimulation of tumor growth
It requires a functioning gut NG tube is often required Critical volume to meet the nutritional requirements Adverse effects force to withdraw nutrition Compliance may be poor
In the perioperative setting there is compelling evidence that the enteral route is the first choice. It is much more evident now that perioperative EN works more as a complex drug able to favorably modulate the cytokine pattern and the immune response than as a support capable of restoring a depleted nutritional status. When patients need a long-term nutritional support at home, the choice is definitely conditioned by the degree of intestinal failure. If this is not total, the enteral supply of small quantities of nutrients is usually recommended not only to integrate TPN but also to prevent metabolic complications. There are, however, some clinical circumstances where both procedures could be applied. In such cases consideration of patient preference should be included in decisions on the method of feeding. In an ad hoc study Scolapio et al. [50] explored the patient’s preference of therapy. They distributed a written questionnaire to 101 hospitalized oncology patients and 98 outpatients without gastrointestinal illness. Patients were asked their preference of nasogastric versus TPN feeding. In both groups, the majority preferred TPN over EN. Preference was related to patient perception of the comfort of these interventions. In logistic regression analyses, the strongest influence on preference were age and perceived comfort of TPN feeding. Consequently, awareness of patient preference should be taken into account when making decisions regarding the route of nutrient delivery. 136
EN vs. TPN in Cancer Patients References 1 Lim STK, Choa RG, Lan KH, Ong GB: Total parenteral nutrition versus gastrostomy in the preoperative preparation of patients with carcinoma of the oesophagus. Br J Surg 1981;68:69–72. 2 Pearlstone DB, Lee J, Alexander RH, et al: Effect of enteral and parenteral nutrition on amino acid levels in cancer patients. JPEN J Parenter Enteral Nutr 1995;19:204–208. 3 Burt ME, Stein TP, Brennan MF: A controlled, randomized trial evaluating the effects of enteral and parenteral nutrition on protein metabolism in cancer-bearing man. J Surg Res 1983;34:303–314. 4 Nixon DW, Lawson DH, Kutner M, et al: Hyperalimentation of the cancer patient with proteincalorie undernutrition. Cancer Res 1981;41:2038–2045. 5 Dresler CM, Jeevanandam M, Brennan MF: Metabolic efficacy of enteral feeding in malnourished cancer and noncancer patients. Metabolism 1987;36:82–88. 6 Jeevanandam M, Horowitz GD, Lowry SF, et al: Cancer cachexia: Effect of total parenteral nutrition on whole body protein kinetics in man. JPEN J Parenter Enteral Nutr 1985;9:108. 7 Harrison LE, Hochwald SN, Heslin MJ, et al: Early postoperative enteral nutrition improves peripheral protein kinetics in upper gastrointestinal cancer patients undergoing complete resection: A randomized trial. JPEN J Parenter Enteral Nutr 1997;21:202–207. 8 Hochwald SN, Harrison LE, Heslin MJ, et al: Early postoperative enteral feeding improves whole body protein kinetics in upper gastrointestinal cancer patients. Am J Surg 1997; 174:325–330. 9 Hamaoui E, Lefkowitz R, Olender L, et al: Enteral nutrition in the early postoperative period: A new semi-elemental formula versus total parenteral nutrition. JPEN J Parenter Enteral Nutr 1990;14:501–507. 10 Iovinelli G, Marsi I, Varassi G: Nutrition support after total laryngectomy. JPEN J Parenter Enteral Nutr 1993;17:445–448. 11 Sand J, Luorastinen M, Matikainen M: Enteral or parenteral feeding after total gastrectomy: Prospective randomised pilot study. Eur J Surg 1997;163:761–766. 12 Shirabe K, Matsumata T, Shimada M, et al: A comparison of parenteral hyperalimentation and early enteral feeding regarding systemic immunity after major hepatic resection-the results of a randomized prospective study. Hepatogastroenterology 1997;44:205–209. 13 Reynolds JV, Kanwar S, Welsch FKS, et al: Does the route of feeding modify gut barrier function and clinical outcome in patients after major upper gastrointestinal surgery? JPEN J Parenter Enteral Nutr 1997;21:196–201. 14 Sako K, Lore JM, Kaufman S, et al: Parenteral hyperalimentation in surgical patients with head and neck cancer: A randomized study. J Surg Oncol 1981;16:391–402. 15 Von Meyenfeldt MF, Meijerink WJHJ, Rouflart MMJ, et al: Perioperative nutritional support: A randomized clinical trial. Clin Nutr 1992;11:180–186. 16 Braga M, Gianotti L, Gentilini O, et al: Early postoperative enteral nutrition improves gut oxygenation and reduces costs compared with total parenteral nutrition. Crit Care Med 2001;29:242–248. 17 Pacelli F, Bossola M, Papa V, et al: Enteral vs parenteral nutrition after major abdominal surgery: An even match. Arch Surg. 2001;136:933–936. 18 Bozzetti F, Braga M, Gianotti L, et al: Postoperative enteral versus parenteral nutrition in malnourished patients with gastrointestinal cancer: A randomized multicenter trial. Lancet 2001;358:1487–1492. 19 Gibbs J, Cull W, Henderson W, et al: Preoperative serum albumin level as a predictor of operative mortality and morbidity: Results from the National VA Surgical Risk Study. Arch Surg 1999;134:36–42. 20 Kudsk KA, Tolley EA, DeWitt RC, et al: Preoperative albumin and surgical site identify surgical risk major postoperative complications. JPEN J Parenter Enteral Nutr 2003;27:1–9. 21 Pacelli F, Bossola M, Papa V, et al: Postoperative enteral versus parenteral nutrition in malnourished patients with gastrointestinal cancer: A randomized multicenter trial. Lancet 2002;359:1697–1698. 22 Gianotti L, Braga M, Vignali A, et al: Effect of route of delivery and formulation of postoperative nutritional support in patients undergoing major operations for malignancy. Arch Surg 1997;132:1222–1230.
137
EN vs. TPN in Cancer Patients 23 Braga M, Gianotti L, Radaelli G, et al: Perioperative immunonutrition in patients undergoing cancer surgery: Results of randomized double-blind phase III trial. Arch Surg 1999;134: 428–433. 24 Bozzetti F, Gavazzi C, Miceli R, et al: Perioperative total parenteral nutrition in malnourished, gastrointestinal cancer patients: A randomized, clinical trial. JPEN J Parenter Enteral Nutr 2000;24:7–14. 25 Braga M, Gianotti L, Nespoli L, et al: Nutritional approach in malnourished surgical patients: a prospective randomized trial. Arch Surg 2002;137:174–180. 26 August DA, Thorn D, Fisher RI, et al: Home parenteral nutrition for patients with inoperable malignant bowel obstruction. JPEN J Parenter Enteral Nutr 1991;15:323–327. 27 King LA, Carson LF, Konstantinides N, et al: Outcome assessment of home parenteral nutrition in patients with gynaecologic malignancies: What have we learned in a decade of experience? Gynecol Oncol 1993;51:377–382. 28 Pironi L, Ruggeri E, Paganelli F, et al: Impact of home parenteral nutrition on performance status in advanced cancer. Clin Nutr 1999;18(suppl 1):52. 29 Cozzaglio L, Balzola F, Cosentino F, et al: Outcome of cancer patients receiving home parenteral nutrition. Italia Society of Parenteral and Enteral Nutrition (SINPE). JPEN J Parenter Enteral Nutr 1997;21:339–342. 30 Hurley RS, Campbell SM, Mirtallo JM, et al: Outcomes of cancer and noncancer patients on HPN. Nutr Clin Pract 1990;5:59–62. 31 Bozzetti F, Cozzaglio L, Biganzoli E, et al: Quality of life and length of survival in advanced cancer patients on home parenteral nutrition. Clin Nutr 2002;21:281–288. 32 Pasanisi F, Orban A, Scalfi L, et al: Predictors of survival in terminal-cancer patients with irreversible bowel obstruction receiving home parenteral nutrition. Nutrition 2002;17:581–584. 33 Howard L: Home parenteral and enteral nutrition in cancer patients. Cancer 1993;72 (suppl 11):3531–3541. 34 Howard L, Ament M, Fleming CR, et al: Current use and clinical outcome of home parenteral and enteral nutrition therapies in the United States. Gastroenterology 1995;109:335–365. 35 Messing B, Barnoud D, Beau P, et al: Données épidémiologiques 1993–1995 de la nutrition parentérale à domicile en centres agréés chez l’adulte en France. Gastroenterol Clin Biol 1998;22:413–418. 36 Van Gossum A, Bakker H, Bozzetti F, et al: Home parenteral nutrition in adults: A European multicentre survey in 1997. Clin Nutr 1999;18:135–140. 37 Howard L: A global perspective of home parenteral and enteral nutrition. Nutrition 2000;16:625–628. 38 Elia M, Russell CA, Stratton RJ, et al: Trends in home artificial nutrition support in the UK during 1996–1999. A report by the British Artificial Nutrition Survey (BANS). The British Association for Parenteral and Enteral Nutrition. 39 Scolapio JS, Fleming CR, Kelly DG, et al: Survival of home parenteral nutrition-treated patients: 20 years of experience at the Mayo Clinic. Mayo Clin Proc 1999;74:217–222. 40 Gaggiotti G, Orlandoni P, Ambrosi S, et al: Italian Home Parenteral Nutrition (IHEN) Register: Data collection and aims. Clin Nutr 2001;20:69–72. 41 British Artificial Nutrition Survey (BANS) Annual Report, PO Box 992, Maidenhead Berks, SL6 4SH UK (BAPEN), 1999. 42 Werhlen-Martini S, Hébuterne X, Pugliese P, et al: Bilan des 47 premiers mois d’activité d’un centre de nutrition entérale à domicile et devenir des patients pris en charge. Nutr Clin Metabol 1997;11:7–17. 43 Schneider SM, Raina C, Pugliese P, et al: Outcome of patients treated with home parenteral nutrition. JPEN J Parenter Enteral Nutr 2001;25:203–209. 44 Torelli GF, Campos AC, Meguid MM: Use of TPN in terminally ill cancer patients. Nutrition 1999;15:665–667. 45 Howard L, Malone M: Clinical outcome of geriatric patients in the United States receiving home parenteral and enteral nutrition. Am J Clin Nutr 1997;66:1364–1370. 46 Sami H, Saint-Aubert B, Szwalowski AW, et al: Home enteral nutrition system: One patient, one daily ration, an ‘all-in-one’ sterile and modular formula in a single container. JPEN J Parenter Enteral Nutr 1990;14:173–176. 47 Loeser C, von Herz U, Kuchlet T, et al: Quality of life and nutritional state in patients on home enteral tube feeding. Nutrition 2003;19:605–611.
138
EN vs. TPN in Cancer Patients 48 Schneider SM, Pouget I, Staccini P, et al: Quality of life in long-term home enteral nutrition patients. Clin Nutr 2000;19:23–28. 49 Robergé C, Tran M, Massoud C, et al: Quality of life and home enteral tube feeding: A French prospective study in patients with head and neck or oesophageal cancer. Br J Cancer 200; 82:263–269. 50 Scolapio JS, Picco MF, Tarrosa VB: Enteral versus parenteral nutrition: The patient’s preference. JPEN J Parenter Enteral Nutr 2002;26:248–250.
Discussion Dr. Powell-Tuck: I enjoyed your very elegant presentation as much as ever. The difference internationally is largely around the use of home nutritional support for cancer patients. I obviously come from a country that uses this very little, although in parts of London occasional patients with perhaps a very high stoma or a patient with sclerosing peritonitis are supported with home parenteral nutrition (PN). I have a little bit more difficulty in really understanding how home PN and home enteral nutrition (EN) are being used by those that advocate them more widely. I wondered if you could fill in color in some of the indications for home care in your own practice in Italy which might differ from this more extreme care that we use in the UK? What do you think the difference is for home nutritional support between the countries? Dr. Bozzetti: I think that there is large variation in the indications not only between different European countries but also between different regions in Italy or even institutions in the same town; this reflects cultural attitude. I am a oncological surgeon, and I know that when a patient has a cancer many general surgeons consider that there is nothing more to do, and this is just the opposite if we take care of a cancer patient in a cancer center. So I admit we have an aggressive policy. As oncologists we are also more and more conditioned by the fact that our cancer therapy is very often palliative. We see that we are not able to cure a cancer of the pancreas for instance, and we always need to follow our patients with a palliative approach as much as possible. But coming to your question, I think that the patients who really could benefit from this approach are patients who are more likely to die due to starvation than tumor progression. So I think that our major effort should be to try to identify the patient population who will be dying from starvation and theoretically could benefit from nutritional support. These are the patients who have an advanced tumor without involvement of the vital organs, patients without liver metastases, patients without lung metastases; on the contrary young and relatively healthy patients with abdominal carcinomatosis from a slow growing tumor. A classical example is a young female patient with ovarian cancer and abdominal carcinomatosis from ovarian cancer, and these patients do not need an excess of palliative care. Patients who receive strong analgesic therapy, which could also complicate their quality of life because they are lying in bed all the time and are treated with sedatives and this type of drugs, and patients with ascites or respiratory problems should not be considered because their condition could be worsened by the administration of fluid. So I think that perhaps if we try to select a favorable group of patients there is a benefit. Of course I agree with you that now we don’t have a clear parameter to specifically select this group of patients, but I think it should be a focus for future research. Dr. Lochs: I was quite surprised by your very nice summary. Thank you for summarizing all the literature so well, because our standard assumption is that EN is better metabolically and immunologically and usually this is not true in cancer patients. Why is that so? Is that probably the main reason that most cancer patients do not have a functioning gut or is there another reason that the immune situation of
139
EN vs. TPN in Cancer Patients the cancer patients is so deteriorated that even EN does not save them from complications? Because this was really surprising to me. Metabolically you showed that PN is better and quality of life, so there seems to be almost no reason to start EN in these patients. Dr. Bozzetti: No, I think that EN is better. I think that in the cancer patient population perhaps the most frequent indication is an obstruction or a sub-obstruction of the bowel so in this kind of patients you cannot use this route. The other problem is that if the patients are anorexic for a variety of reasons and they have a central venous catheter, sometimes it is practically easier to give them nutritional support by a vein because the alternative would be to place a nasogastric tube and this is psychologically bad. So in many patients I support this way because I agree with the statement made by the study of Scolapio [1] that fluid via a vein is a well-accepted procedure. But I agree with you, from the physiological metabolic point of view, especially for long-term nutrition, EN is definitively better and probably also with regard to tumor growth. In the literature there are 7 or 8 studies comparing tumor growth by the incorporation of specific indicators and the results show that PN better supports tumor growth than EN. So it is the type of population that forces you to modulate the nutritional approach, but I agree with you that EN is better than PN. Dr. DeLegge: I have a statement and a question. The statement is that your data on home PN are very similar to those published August et al. [2] back in the early 1990s regarding malignant bowel obstruction and home total PN (TPN) where they also showed that if the patients had a good Karnofsky scale prior to starting home TPN they did quite well at home. That is a suggestion we should make use of and in fact perhaps we need to use it for home EN evaluation, a functional scale. In the case of someone having a malignant biliary obstruction there wouldn’t be very much hesitation in the US to put in a metal biliary stand at the cost of thousands of dollars, but yet there is a reluctance to utilize home nutrition support in a very similar patient population. I don’t see the difference. My question to you is invariably on the PN to EN studies. Patients on EN receive less calories because of the fact that someone is withholding their tube feeding for a gastric residual or perhaps going off to the X-ray suite. So in most of those studies although they were supposed to get similar amounts of calories, was that really so? Meaning, in the studies you showed us, the TPN patients and the PN patients invariably received more calories and more nitrogen per day than the EN patients, but yet they still had very similar results. Dr. Bozzetti: With the exception of the study by Lim et al. [3] in 1981 in the British Journal of Surgery where the TPN included a hypercaloric regimen, both the studies by Nixon et al. [4] and Brennan et al. [5] perfectly matched the quantity of nitrogen and calories and there were no significant differences between the groups. But in the practice of home TPN and home EN, I agree with you that patients receiving EN usually receive less calories and this may be a crucial point. Especially if you have a gastric approach the patients may experience a full satiety feeling and you are obliged to stop the administration of nutrients. This doesn’t occur with TPN. So you are obliged to use a hypercaloric formula. Dr. Labadarios: I just want to pick up from Dr. DeLegge’s point. You know there is a lot of lip service paid to nutrition, but when it actually comes to budgetary issues, both at the policy and the practice level, there is very little that is actually made available to support nutrition. Now at the policy level, we have tremendous problems to convince the administrators that nutrition is important is life-saving, and is a costsaving practice. But I wonder to what extent is it appropriate for us in the nutrition field to call any form of nutritional support expensive. In what sense is it expensive? What is the reference point for comparison? One wonders, or is it just a point of debate? It is a life-saving therapy that needs to be administered like any other therapy, and yet
140
EN vs. TPN in Cancer Patients when it comes to nutrition support, it is expensive. I have a problem with that, or a little bit of difficulty in understanding the approach to it. Dr. Thomas: I have a couple of questions with regard to the route of administration and duration. This was mostly in oncology patients and the mean duration is fairly short; there is another group of patients who are more long-term in home care. So the question would be first, in the preference study you showed from the Mayo Clinic was for nasogastric intubation versus vein, and no mention of percutaneous intubation. I wonder if it is different when you give that option rather than nasogastric? Second, can you comment on the maximum duration or main duration that you are able to maintain these patients on in terms of parenteral nutrition? Did you use percutaneous tubes, the jejunostomy tubes? What is the duration you can expect to be able to maintain these tubes? Dr. Bozzetti: With regard to the Scolapio [1] study, you are right, Scolapio investigated the preference of the patients and the choice was between intravenous catheters and the nasogastric tube. As an abdominal surgeon I think there are very few patients with advanced disease, anorectic and malnourished, who could benefit from a gastrostomy or a jejunostomy because the main indication for nutritional support is undernutrition due to intestinal obstruction or peritoneal carcinomatosis. I don’t exclude this, but there is little room for this approach. In my experience it was possible to perform surgically a jejunostomy in some patients because when during the operation we realized that there was no possibility of cure. There is a small percentage of the patients who are managed by jejunostomy which, in this condition, is not so different from the patients receiving TPN. The long survival of patients receiving EN is due to the fact that these patients had a less aggressive tumor of the upper digestive tract with no problem in the abdominal cavity. In my experience gastrostomy or jejunostomy can be performed for this purpose in a few patients who are candidates for nutritional support, so I don’t exclude it but it is a rare occurrence. Dr. Buchman: Tumors are rapidly growing in dividing cells and therefore they act as nutrient sinks. Do you think therefore that there are actually any carefully controlled studies that clearly demonstrate that the tumors of patients who were fed by PN grew at a more rapid rate than those in patients fed enterally? But in any case do you think that the fact that invariably in every study that has ever been published (there is one unpublished study that I am aware of), patients who received PN always receive more calories than EN, as Dr. DeLegge has pointed out? Do you think this could be the reason for the fact that, if it is true, tumor growth is enhanced with PN? In Europe glutamine dipeptide has been used in great proportions, and given the fact that some tumors are exclusively sensitive to glutamine, and in animal models tumor growth is enhanced by glutamine [6], do you think that glutamine-supplemented TPN would also be part of the reason that tumor growth could be enhanced? Dr. Bozzetti: Quite interesting questions. With regard to the first one: I can say that I disregard many of the studies in experimental tumors. Experimental tumors are different from the human tumors. In experimental tumor the weight of the tumor can account for 20 or 30% of the weight of the carcass and so when we feed a tumorbearing animal we directly feed the tumor. The condition in humans is totally different because even in patients with a big tumor it usually never accounts more than 0.1% of the body weight of the patients. With reference to your question we have tried to explore this problem by giving a glucose-based regimen or a lipid-based regimen to cancer patients with liver metastases and to study the fluorodeoxyglucose uptake of the tumor with positron emission tomography [7]. The two regimens were quite identical but all the non-protein calories were given in the lipid group as fat, and no glucose was given at all. We saw that the utilization of glucose was the same in the patients receiving the mixed regimen and the lipid regimen. Even if the tumor is
141
EN vs. TPN in Cancer Patients utilizing glucose and not lipid, we cannot support the statement that a glucose-based nutrition is dangerous for the patient, because when the size of the tumor and its metabolic capacity are calculated we realize that the quantity of the glucose necessary to support tumor growth is very small, and anyway the small quantity of glucose made by the gluconeogenesis is enough to support tumor growth. What we have also observed is that tumor growth within the liver had a such a high metabolic capitation of the glucose that it seems to have worked at the maximal capacity possible. So it was not affected at all by the administration of glucose or the administration of a lipid regimen. The second question: theoretically you are right, in the laboratory it is possible to stimulate tumor growth and the growth of some cells by adding glutamine. I am not skeptical; I accept that cancer patients could receive glutamine for two reasons. One because as a surgeon I accept the risk; I accept that when I perform an operation there can be some infection, there can be some complications or some morbidity and so on, so I can also accept that if the patient needs glutamine for bone marrow function or for immune system function, this could be deleterious for the tumor growth. Secondly, I am aware of a wonderful study published in Cancer Research by Holm et al. [8] who measured the flux across the tumor in vivo and the dynamic balance of the substrate. They were able to demonstrate that glutamine is not introduced and utilized by the cancer cell, at least colon cancer cell and gastric cancer cell; the colon cancer cell because the study was done in patients with colon cancer. So I think that this is a theoretical possibility that glutamine stimulates tumor growth, but it has not been substantiated in human studies. Probably glutamine can be utilized and can be useful for other purposes, so I don’t see any major contraindication to use it.
References 1 Scolapio JS: A review of the trends in the use of enteral and parenteral nutrition support. J Clin Gastroenterol 2004;38:403–407. 2 August DA, Thorn D, Fisher RL, Welchek CM: Home parenteral nutrition for patients with inoperable malignant bowel obstruction. JPEN J Parenter Enteral Nutr 1991;15:323–327. 3 Lim ST, Choa RG, Lam KH, et al: Total parenteral nutrition versus gastrostomy in the preoperative preparation of patients with carcinoma of the oesophagus. Br J Surg 1981;68:69–72. 4 Nixon DW, Lawson DH, Kutner M, et al: Hyperalimentation of the cancer patient with proteincalorie undernutrition. Cancer Res 1981;41:2038–2045. 5 Brennan MF, Pisters PW, Posner M, et al: A prospective randomized trial of total parenteral nutrition after major pancreatic resection for malignancy. Ann Surg 1994;220:436–444. 6 Chance WT, Cao LQ, Fischer JE: Response of tumor and host to hyperalimentation and antiglutamine treatments. JPEN J Parenter Enteral Nutr 1990;14:122–128. 7 Bozzetti F, Gavazzi C, Mariani L, Crippa F: Glucose-based total parenteral nutrition does not stimulate glucose uptake by humans tumours. Clin Nutr 2004;23:417–421. 8 Holm E, Hagmuller E, Staedt U, et al: Substrate balances across colonic carcinomas in humans. Cancer Res 1995;55:1373–1378.
142
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 143–166, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
Ethics and Economics in Nutritional Support Alan L. Buchman Division of Gastroenterology, Feinberg School of Medicine, Northwestern University, Chicago, Ill., USA
Introduction Enteral and parenteral nutritional support was developed to provide sustenance to patients who ordinarily were unable to meet their nutritional needs either because of an inability to eat or an inability to digest and absorb food. Undernutrition has been associated with increased infection risk, poor wound healing, increased postoperative complications, prolonged hospital stay, respiratory, cardiac, and hepatic dysfunction, as well as increased mortality. A clinical decision must be made about the provision of both nutrition as well as hydration fluids for patients otherwise able to eat in the usual fashion. Patients who had intestinal failure from either short bowel syndrome (congenital or acquired) or severe dysmotility disorders were doomed in the absence of parenteral nutrition. Soon after the advent of total parenteral nutrition (TPN) in the late 1960s and early 1970s, it was thought that TPN was the cure-all for many ills and it became integral to the management of the chemotherapy patient, bone marrow transplant patient, preoperative patient, burn patient, and trauma patient, among other patient subsets. It was once hypothesized that, because patients who received TPN had a significantly decreased serum cholesterol concentration, TPN might be a reasonable therapy for atherosclerotic heart disease [1]. As this therapy came under increasingly more widespread use, evidencebased medicine with regard to TPN began to evolve. It was found TPN did not increase the effectiveness of chemotherapy, improve bone marrow engraftment, result in decreased postoperative infections in only mildly or non-malnourished patients, and did not improve outcome in trauma (other than head trauma) or burn patients, and in fact was associated with significant treatment-related complications in some patients [2]. The survival rate 143
Ethics and Economics in Nutritional Support Table 1. Considerations for nutritional support • • • • • • • •
Patient’s medical prognosis and accuracy of the diagnosis Patient’s quality of life Patient’s life expectancy Potential benefits of therapy Potential risks of therapy Other available options Patient’s wishes for the use of life-sustaining therapy Family’s wishes if patient’s wishes unknown or unobtainable
Table 2. Criteria for determining the appropriateness of a cancer patient for home TPN [20] • Requirement for intravenous fluid in order to maintain fluid and nutritional equilibrium • Patients capable of self-care and spend a minimum of 50% of their waking hours out of bed • The patient is able to physically as well as mentally and emotionally stop, start, and control their TPN • Survival of ⱖ3 months is anticipated • The patient is aware of the diagnosis and prognosis, and can therefore make an informed decision as to use TPN • There is no alternative to TPN
for patients with malignancy who receive home TPN was 32% in an Italian study, with satisfactory social activity achieved in only 25% of these patients [3]. Survival is only 15% for patients with metastatic disease who receive home TPN [4], and the length of survival is most often 2–3 months or less [5–7]. Home TPN does not improve quality of life for the majority of cancer patients [7], although quality of life may be improved for those patients who are more functional at baseline and in whom more prolonged survival is anticipated [3, 6]. Specific criteria have been proposed to identify cancer patients who might be most likely to benefit from home TPN (table 2) [8]. Similarly, enteral nutrition (EN), administered by enteric tube in demented patients has not been demonstrated to improve survival, prevent infection or prevent pressure ulcers [9].
Should Nutritional Support Be Initiated? The decision to initiate nutritional support should be based on the potential risks and benefits in conjunction with the patient’s and, to a lesser degree, the 144
Ethics and Economics in Nutritional Support Table 3. Arguments for initiation or continuation of nutritional support • Nourishment is within the scope of minimal care and is required regardless of how hopeless recovery may be • Dignity required feeding • Nutritional support may be provided in a noninvasive manor • Nourishment, provided by nutritional support, is not medical therapy • Withdrawal of nutritional support will precipitate painful and agonizing death from starvation and/or dehydration • Nourishment via nutritional support allows life to continue until the underlying illness runs its natural course
family’s desires. These desires may be tempered by religious and cultural beliefs. Regardless, it must be realized that the goals of nutritional support may evolve and change during the course of a given patient’s illness. It is unclear whether patients are actually treated with nutritional support or simply maintained. Is nutritional support a treatment in and of itself? If not, can it be considered to have treatment attributes if its use contributes to the treatment of underlying pathology, or is considered a part of the success of other therapies (table 3)? The initiation of nutritional support should be accompanied by informed patient consent, although such consent need not be in writing. This consent requires the patient to have an awareness as well as an understanding of their prognosis, the requirements of the proposed therapy, the potential for success with nutritional support and what that ‘success’ entails, and the possible risks inherent with the therapy. This principle of beneficence weighs the benefits of an intervention versus the burdens produced by such an intervention. The principle of beneficence requires the patient to be informed of the potential medical benefits and risks prior to initiation of this nutritional support; ‘every human being of adult years and sound mind has a right to determine what shall be done with his own body’ [10]. If the patient is not properly informed, he or she is incapable of making an informed consent [11]. The potential for benefit must be discussed in terms of improved survival, improved quality of life, comfort, or correction of metabolic abnormalities that have a material impact on the quality, appropriateness, and success of the medical care rendered. According to the Helsinki accords a physician is free to propose therapy, but not necessarily to use such therapy. Informed decisions about what may be justified in a research setting are not necessarily the same as what is required in a clinical setting. A physician who cannot explain or will not explain the unbiased risks and benefits should remove themselves from the consent process. Notwithstanding the explanation of the consent, the physician is free to render an opinion or advise to the patient. That said, it is important to differentiate opinions, 145
Ethics and Economics in Nutritional Support which may be construed by the sequence of discussion or the emphasis the physician places on a particular modality, from concealment. Concealing information to promote certain treatment options is unethical. There is no evidence that concealing what admittedly may be a poor option provides benefit to the patient. In a study of 421 randomly selected, competent patients living in 49 different nursing homes were surveyed about feeding tubes. 24% of the eligible patients refused to participate, but only one third of the 379 patients who completed the survey stated they would agree to feeding tube placement if they were unable to eat sufficiently, and 25% of those who agreed changed their minds when told that feeding tube placement and use might involve restraints [12]. The associated use of restraints is often a required, but has a negative influence on the quality of life in the cognitively impaired patient [13]. It was not made clear whether tube feeding would have been via an nasogastric tube or percutaneous endoscopic gastrostomy (PEG). Males, African-Americans, absence of a living will, and no previous discussions of life-sustaining therapy with family or staff were all associated with a greater likelihood to prefer tube feeding [12]. Similarly, in another study, elderly patients who initially indicated that they wanted cardiopulmonary resuscitation in the event of cardiac arrest rescinded their wish when told of the poor outcome of attempted resuscitation [14]. Physicians and family must be careful to avoid intrusion on which, by all rights, is the territory of the patient in question. This is the concept of selfdetermination. In cases where the value of nutritional support is unclear or undefined, a limited trial may be considered, with efficacy to be evaluated and goals to be potentially reevaluated. Nutritional support is not mandatory when it is burdensome or of no proven medical value [15]. TPN is associated with numerous potentially life-threatening complications that include bleeding, infection, thrombosis, metabolic abnormalities, hepatic failure, renal disease, and osteoporosis [16]. EN is also associated with numerous potentially lifethreatening complications, including infection, perforation of abdominal viscera, aspiration, metabolic abnormalities, and diarrhea and dehydration [17]. One does not omit feeding in a conscious attempt to cause death, although death may ultimately result. Therefore, withholding nutritional support is not a decision to kill by lethal omission. PEG or home TPN are seldom indicated in the patient with advanced cancer associated with a significantly diminished performance status, or in those patients whose unresectable disease is unresponsive or not judged appropriate for radiation therapy or chemotherapy. In these cases, the patient will not be able to adequately use the nutrition even if provided, and there is little or no chance quality of life will improve and a not insignificant chance that quality of life may be further decimated. In a study of 135 nursing home residents who had a feeding tube placed because of severe cognitive impairment, there was no increase in survival [18]. The only determinants of survival were the 146
Ethics and Economics in Nutritional Support Clinical category Anorexia-cachexia syndrome?
Clinical guideline
Ethical rationale for guidelines
Yes
Do not offer PEG
Patient unable to make use of nutrients
Yes
Offer and recommend against PEG
Patient unable to experience any quality of life
Yes
Offer and recommend PEG
Patient unequivocally benefits from PEG
Yes
Discuss no PEG vs trial of PEG
Patient equivocally benefits from PEG and potential exists for loss of quality of life
No Permanent vegetative state? No
Dysphagia without complications? No
Dysphagia with complications?
Fig. 1. Decision-making algorithm for PEG tube placement. From Rabeneck et al. [19]. Reprinted with permission.
patient’s underlying medical condition, advanced age, and do not resuscitate status. Other studies have shown no benefit of EN support on the healing of pressure sores in similar patient groups [9]. Rabeneck et al. [19] suggested that the initial determination of whether a patient should have a PEG placed is to determine whether their cachexia-anorexia is due to metabolic alterations that will be unresponsive to nutritional support. They proposed a decision-making algorithm for PEG tube placement (fig. 1). These authors indicated that the physician is under no obligation to place the PEG if no significant physiological benefit is anticipated. PEG tubes and central venous catheters are often placed inappropriately because of unrealistic expectations on the part of either the health care provider or the patient, or both. Issues of perceived improvement in quality of life and improvement in outcome of malnutrition without regard to the underlying disease process, no matter how unrealistic or impractical, may cause considerable confusion among patients and their families and support. Appropriate indications for PEG include dysphagia secondary to reversible disease, incurable disease with survival potential, loss of the ability to eat, primary neurological disorders with the likelihood of prolonged survival and improved quality of life, severe upper gastrointestinal motility disorders (often for decompression) or obstruction, and growth failure (children). It should be noted that PEG does not prevent aspiration or aspiration pneumonia [20, 21]. 147
Ethics and Economics in Nutritional Support The Role of Morality, Religion, and Patient Self-Determination: Who Should Not Receive Nutritional Support? Morality Some regard feeding as a moral and emotional commitment rather than a medical therapy. The provision of food and fluids has often been associated with nurturing and caring. It has been widely stated that one cannot allow a patient to ‘starve to death’, or to ‘allow the patient to die of thirst’. Food intake is also often viewed within a social context, and is associated with pleasurable phenomenon including taste, smell, and socialization according to Lipman [22]. However, consumption of continuous and non-orally introduced food and fluid can hardly be seen as socially normative. Does eating require the monitoring of electrolytes, blood urea nitrogen and creatinine? Does eating require the flushing of the mouth? Does eating require a permanently attached fork to the mouth in order for nourishment be delivered? Lipman [22] has referred to this as non-volitional, forced, invasive delivery of artificial, limited, and fixed substrates associated with finite morbidity and mortality. Some might also add that if one does not want extraordinary care, then why even come to the hospital? What is the definition of the life that is being sustained? Life has different meanings to different people. It may be a single cell, or it may be defined in the context of a functioning groups of cells, or at a higher level including other living creatures or a society where each life plays a unique role (fig. 2). When does life cease? Is it when one part of society, some tissue from an organism, or some cells from tissue become separated, lost and perhaps die? Kantian ethics considers the act itself is intrinsically right or wrong regardless of the outcome. This concept is based on the idea that what happens in life is often governed by chance and circumstance (i.e. a lucky shot in billiards) and therefore, the motivation for the act becomes the differentiating factor. Therefore, nutrition should be provided for moral reasons and omission or withdrawal would be considered unethical. Utilitarian ethics suggests the means to the end are not important, it is the end result that matters most. Therefore, no act is intrinsically good or bad, but the goodness or the badness, and hence, the morality of an act is determined from the outcome. If happiness or a positive outcome is achieved, the act would be considered ethical and moral. Religion Historically, Judeo-Christian litany has proclaimed the sanctity of human life. The physician is the healer and woe is thy name who seeks to end life prematurely by omitting or withdrawing life-sustaining nutrition and hydration. In Judaism, it is of the utmost importance to preserve and sustain life. However, in cases of terminal illness, where therapy would be expected to increase the patient’s suffering or be considered futile, the patient may be 148
Ethics and Economics in Nutritional Support Death
Life
Society Individual humanity Organism Organ Tissue Cell DNA
Levels of Life
Fig. 2. The definition of life and death. Adapted from Belgumd et al: J Religious Health 1980;28:125–137.
entitled to reject such therapy in advance just as with other therapeutic interventions [23]. In general, even Orthodox Judaism argues against interventions that prolong suffering and impede dying [23, 24]. In Judaism, the physician may be permitted to use heroic methods to save and prolong life, but may not be required to do so. Some, however, consider tube feeding basic care that may not be rejected; other Jewish theologians point out that God provided the materials and know-how to develop feeding catheters which therefore should be used to prolong life whenever necessary. However, it must be realized that the development of today’s technology of artificial feeding was unforeseen by our ancestors. In Catholicism, extraordinary care is optional and the patient may refuse it, although ordinary care may not be refused. Sometimes the definition of extraordinary must be viewed operationally, or within the context of a particular situation. For example heart valve replacement may be considered ordinary care in an otherwise healthy high school student, but would be considered extraordinary in a debilitated 90-year-old individual. Proportionate care has been defined as that capable of providing a reasonable chance for a substantial and sufficient benefit as to justify the risks (including pain, expense, and inconvenience) and complications inherent in such therapy. Otherwise, care is known as ‘disproportionate’ when it does not offer reasonable hope of benefit, and patients may refuse it. Extraordinary care is disproportionate care that may not be justifiable in a risk-benefit analysis. 149
Ethics and Economics in Nutritional Support In general, religious, ethnic, and cultural beliefs and preferences should be respected when the initiation or withdrawal of nutritional support is considered, especially when such therapy may be life-sustaining. For example Americans place a high value on individual autonomy, the individual’s ability to make decisions for themselves. However, in order to make such decisions, the patient must be educated about their disorder, treatment options, and potential outcomes. In other cultures, such as Japan and other Asian countries, as well as Mexico, it is the family that bears the brunt of the decision-making process, and the patient may not even be aware of a terminal diagnosis and prognosis. Self-Determination and Medical Decision Making When patients are near death and are given the opportunity to refuse nutritional support, the majority do so [25]. Studies have shown that, at least in some circumstances, physicians may underestimate patients’ quality of life when compared to patients’ own ratings [26]. Capacity refers to an individual’s ability to make an informed decision as assessed by health care providers rather than a judge. The physician should form an opinion about a patient’s capacity to provide an informed consent. If an individual does not have the capacity to make a medical decision about their care, a decision-making algorithm is displayed in figure 3. If the patient is mentally incapacitated, an appropriate proxy must be identified. If a patient has the capacity of make an informed decision, the principle of autonomy requires respect for that decision. The combination of patient refusal for one life-sustaining therapy such as surgery, but agreement for other life-sustaining therapy such as nutritional support should not be considered incomprehensible, but should be respected by the health care provider. The caveat is that the patient should be properly educated in order to make such a decision. In cases where the patient is able to make an informed decision with regard to treatment, a distinction must be made between refusal of nutritional support because of unpleasant side effects, and refusal of such therapy because quality of life will not be improved in the long-term. It also may be difficult to determine when depression is sufficiently severe as to impair decision-making capacity. Some physicians consider TPN and EN, as well as hydration to provide a basic human need to avoid starvation and dehydration. Views among patients and physicians and other medical personnel may vary across cultures and nationalities. For example, patient self-determination is of utmost importance in the United States where given the choice, many terminal patients may choose to forgo life-sustaining therapies, including artificial nutrition. This decision process is supported by the Federal Patient Self-Determination Act. In other cultures, such as Japan, it is not customary to inform the patient of a terminal diagnosis and therefore, the patient may wish to 150
Ethics and Economics in Nutritional Support
Medical decision making with an incompetent patient
Optimize patient function
Assess capacity
Competent
Incompetent
Determine patient’s wishes
Identify appropriate proxy
Available
Unavailable
Full discussion with proxy
Use best medical judgment while locating or arranging for court appointment
Substituted judgment
Best interests
Fig. 3. Medical decision-making with an incompetent patient. From Goldstein MK: Ethics; in Ham RJ, Sloane PD (eds): Primary Care Geriatics: A Case-Based Approach, ed 2. St Louis, Mosby-Yearbook, 1992, p 222. Reprinted with permission.
continue therapy. In the situation of a terminally ill patient (a 65-year-old businessman with metastatic gastric cancer with a 1-month expected survival) who is unaware of his diagnosis, 67.5% of surveyed multi-specialty Japanese physicians would provide TPN for malnutrition although only 36% of those surveyed would want such therapy for themselves in a similar situation (and only 5% of Japanese-American physicians; fig. 4) [27]. On the contrary, only 33% of Japanese-American general practice and internal medicine physicians would recommend TPN to treat malnutrition in a terminal condition. 36% of the Japanese physicians indicated they would ignore the patient’s request to withdraw TPN if the physician thought it necessary, although only 6.5% of Japanese-American physicians would do so. For patients in a persistent vegetative state during which the patient has eyesopen unconsciousness, periods of wakefulness and physiologic sleep-wake cycles, but shows no evidence of self- or environmental awareness, opinions of American neurologists and Medical Directors are even more clear-cut: 89% of 490 respondents in a national survey believed the withdrawal of 151
Physicians who would provide TPN, %
Ethics and Economics in Nutritional Support 70
Japanese physicians
60
Japanese-American physicians
50 40 30 20 10 0 For patient
For self
Would ignore patients refusal
Fig. 4. 65-year-old businessman with metastatic cancer and 1 month expected survival. Adapted from Asai et al. [28].
artificial nutrition and hydration from such patients was ethical [28]. A survey of 580 internists in the USA revealed similar opinions [29]. It has been estimated that the cost of providing futile care for these patients even some 10 years ago was USD 1–7 billion/year [30]. However, treating physicians may often find it difficult to realize and understand the value of an individual to their family and friends, even in a persistent vegetative state.
What Is the Goal of Nutritional Support in the Terminal Patient? The history of nutritional support in the terminal patient is similar to cardiopulmonary resuscitation; it has been tried in virtually every patient who died, despite having been invented for specific, reversible conditions, and was deemed unsuccessful in most [31]. Patients with terminal illness are often comfortable with very limited food and water intake [32]. TPN and EN may adversely affect the body’s normal adaptation to starvation [32–34]. Anticipation of a poor outcome does not engender the treating physician to act to prevent such an outcome. One must consider the fact that nutritional support itself, especially in the patient in whom the medical prognosis is not expected to reasonably improve, may be a form of torture. In the absence of evidence-based medicine to confirm a substantial benefit of nutritional support in a terminally ill patient (quality of life included), rather than providing the patient the option to refuse therapy should be replaced with a question to the physician provider: why was nutritional support proposed in the first place? The President’s Commission for the Study of Ethical Problems in Medicine and Biomedical Research expressed the principle that physicians are not required to provide futile therapies [35]. Futility has been medically 152
Ethics and Economics in Nutritional Support defined as a treatment that fails to improve the patient’s prognosis, comfort, well-being, or general state of health for which there is no possibility of a meaningful recovery [36]. Actually withholding nutritional support may enhance patient well-being and comfort in some cases, most notably those who are terminally ill. A primary goal of nutritional support should be an improvement in the patient’s quality of life. However, quality of life is a value judgment and as such lacks a common definition so that a value is difficult to assign to improvements or decrements. There may also be inconsistencies in the interpretation of quality of life; there exist both objective and subjective components. It is difficult to judge quality of life in another without interference from one’s own bias. Therefore, decisions by surrogate decision makers with regard to initiating or withdrawing nutritional support in order to effect improved quality of life may be invalid.
Nutritional Support in the Developmentally Disabled and the Non-Terminal, but ‘Vegetative’ Patient Although the medical literature contains references to a number of cases where patients arose from seemingly comatose, vegetative states, the general medical consensus is that a persistent unconscious condition is irreversible and there is no reasonable likelihood of recovery of cognitive brain function. Apply the concept of ‘beyond reasonable doubt’. However, notwithstanding the evidence, the decision to withdraw or withhold nutritional support should be made on an individual case-specific basis. The decision to initiate or to withdraw nutritional support in the patient lacking the capacity to make such a decision themselves, rests on the relative weight assigned to benefits to the family such as prolonged life versus the burdens on the patients such as complications. For example if the family wants mother to live as long as possible, complications from nutritional support may be viewed as acceptable consequences. Regardless, the nutritional plan should have a commitment to discontinue treatment if anticipated measurable physiological outcomes are not achieved if appropriate and sufficient nutritional support has been provided for a sufficient length of time. The informed consent for the delivery of such nutritional support should include the prospect of medical uncertainty. The developmentally disabled may never have had the capacity to make a decision about their medical care. Therefore, it is difficult for surrogate decision makers with power of attorney to make decisions based on what the patient may have wanted. However, such patients do have an interest in having a minimum of suffering in relation to treatment benefits and to be as pain-free as possible. 153
Ethics and Economics in Nutritional Support Can Nutritional Support Be Withdrawn? Is there a difference between failure to initiate therapy and an action to discontinue therapy? Is shutting off tube feeding or TPN similar to shutting off the heat or air conditioning or oxygen delivered via nasal cannula? Some studies have shown that physicians are often reluctant to withdraw therapy even when the risks outweigh the potential benefits [37]. Professional organizations such as the American Medical Association have issued guidelines that ‘Artificial administration of nutrition and fluids is a life-prolonging treatment. As such, it is subject to the same principles for decisions as other treatments’ [35]. Most would agree that there is no moral obligation to continue medical care at all, including nutritional support, when inevitable death is imminent and continued treatment would only prolong life under such circumstances. However, there is a moral obligation to care for, including medical care, and to provide nutritional support if necessary, for those patients who may otherwise be incapacitated, are helpless, or even chronically ill. However, this moral obligation does not necessarily exist towards patients who are in a persistent vegetative state or in whom no possibility of meaningful recovery exists or in whom a terminal illness has been diagnosed with a prognosis that includes either a minimal survival time or minimal quality survival time. Medical treatment, including nutritional support if necessary, is appropriate in order to provide some relief from suffering or an improvement in the underlying disease state that may result in improved quality of life (table 3). Withdrawal of nutritional support should be accompanied by ‘comfort care’, which includes appropriate pain management. It might be said that comfort care avoids abandonment and a human bond so that other care may be morally withdrawn. However, there cannot always be sufficient evidence that the patient is suffering pain from nutrition support or its delivery.
Legal Considerations Laws supply secular morality that may often cross cultural boundaries. Laws vary between countries however, and judgments from one country cannot necessarily be applied as precedent for another. Recent court cases in both the United States and the United Kingdom have generally upheld the withdrawal of nutritional support in severely demented patients with essentially no chance for recovery and for whom no benefit from nutritional support was demonstrated [38]. Withholding and withdrawal of therapy are legally equivalent. Competence and capacity for judgment refer to a given patient’s ability to exercise good judgment and make an informed decision on their own behalf. Competence is actually a legal definition that defines an individual’s perceived ability to make appropriate decisions. Competence or incompetence are not 154
Ethics and Economics in Nutritional Support all-inclusive. For example, an individual may be incompetent to handle their finances, but competent to make medical decisions. Laws generally govern the mechanism for transfer of decision-making power. An advanced directive, such as a living will, takes precedence over family relationships. If a proxy becomes involved they are asked to provide substituted judgment – to decide as best they can how the patient themselves would have decided if he or she had been able. In the United States, advance directive laws are in affect in each state. These statutes enable each patient to voice their decision about life-sustaining therapies prior to becoming incapacitated and unable to render such opinions, although patients are often not instructed to comment on more specific items such as nutritional support. The United States Supreme Court has ruled the provision of nutrition and hydration are medical treatment and as such, may be legally withheld or withdrawn if there is an appropriate medical and ethical situation [39]. In addition, the withdrawal of life-sustaining therapy has been made legally distinct from homicide since 1976 in the United States [40]. However, in such countries as Germany, the withdrawal of nutrition in patients suffering from terminal illness may be considered active euthanasia unless death is imminent [41], and invokes the potential of civil or even criminal liability [42]. So-called ‘living wills’ permit an individual to provide advance directives should they become incapacitated. Nutritional support should be specifically addressed in advanced directives and living wills. Such directives may include the disallowing of certain aspects of treatment and may also include provision for a proxy decision maker, the latter including the use of durable powers of attorney. These directives allow the individual to identify an individual who has intimate knowledge of their desires who can make health care decisions in the event the individual is unable to make such informed decisions. Incumbent on such decision making is the obligation that the decision maker should express the desires of the patient as best they can rather than their own desires. Most, if not all states require that hospitals, nursing homes, and other similar institutions develop and use a policy covering advance directives.
Nutritional Support in Children Parental consent is required for medical treatment [43]. Parents are responsible to make healthcare decisions about their children. Courts have consistently upheld parents’ rights to serve as surrogate decision makers for their children indicating that they are better placed to represent the child’s best interests. Courts have held that as the appropriate decision makers for the children, parents may refuse life-sustaining therapy as long as such refusal does not constitute neglect [44–46]. However, parents are not given 155
Ethics and Economics in Nutritional Support carte blanche as society has established protections designed to prevent child neglect. ‘Baby Doe’ amendments in 1984 to the 1973 Federal Child Abuse and Neglect Prevention Act mandate that infants with severe lifelimiting handicaps still receive appropriate nutrition and hydration under all circumstances, although the word ‘appropriate’ does allow some latitude in interpretation [47, 48]. Physicians can invoke the power of the state to initiate or to continue treatment, but only when such treatment is clearly beneficial to the patient and such treatment is considered the appropriate standard of care. In that situation, refusal of treatment would be grounds for medical neglect. However, both ethical and legal consensus has emerged that concludes nutritional support and hydration may be omitted if justification can be made [49–51].
Economic Considerations in Nutritional Support One may question whether it is ethical to consider costs of nutritional support to either the individual or society in terms of money and resources. However, economics can be defined by a variety of cost analyses (table 4). In 1990, the yearly cost for hospital-based TPN in the USA was USD 6 billion [52]. Trujillo et al. [53] reported that 23% of 209 inpatients begun on TPN at the Brigham and Women’s Hospital in Boston over a 5-month period had alternative means of nutritional support and 15% were not indicated at all under the American Society for Parenteral and Enteral Nutrition (ASPEN) guidelines [54]. Avoidable charges (although not costs) for preventable TPN days were USD 510,746 when extrapolated to a full year. 35% of patients had short-term TPN use (⬍5 days) for which risks may be increased, but the potential benefit is nil. National guidelines governing home EN are present in 3 European countries and guidelines for home PN in 4 European countries [55]. Guidelines are present in the United States, but unlike Europe, most patients are not cared for at centers of expertise. Interestingly, Medicare, Medicaid, and private insurance companies in the United States support the financing of hospitalbased as well as home nutritional support, but have shown no interest in assuring that patients receive the highest quality and cost-effective care. Rankings of hotels, airlines, football teams, schools, and hospitals are commonplace in American, but there is no ranking of homecare agencies or of professionals involved in the delivery of homecare. In fact, nutrition as a whole is a completely unregulated field. Medicare has specific guidelines for the indications of both home TPN and home EN, otherwise financial coverage is not provided. For TPN, these include: (1) permanent (ⱖ3 months) disease whereby small intestinal absorption or transport is impaired; (2) enterectomy within the last 3 months leaving ⱕ1.5 m of residual small intestine; (3) with an oral/enteral intake of 2.5–3.0 liters/day, enteral losses are ⱖ50% 156
Ethics and Economics in Nutritional Support Table 4. Cost-analysis definitions • Cost-identification: Direct cost for therapy or service regardless of outcome • Cost-effectiveness: Cost of one intervention compared to the costs of another • Cost-utility: Degree of outcome improvement provided for by the chosen intervention • Cost-benefit: Ratio of the benefits achieved from the chosen intervention versus the costs (both direct costs as well as from complications related to the intervention) otherwise known as ‘burden’ Adapted from Goff KL: Cost and cost-benefit of enteral nutrition. Gastrointestinal Endosc Clin North Am 1998;8:733–744, table 1.
and urine output is ⱕ1 liters/day; (4) energy needs are 25–35 kcal/kg/day; (5) mechanical small bowel obstruction and surgery is not an option, or (6) 10% weight loss over last 3 months, serum albumin concentration is ⱕ3.4 g/dl, and fecal fat losses are 50% of intake, although exceptions can be granted for less severe malabsorption. For EN, this includes: (1) permanent (ⱖ3 months) disease or inability to swallow or get nutrients into the small intestine, and (2) a disease that impairs digestion and absorption of an oral diet where the patient is unable to maintain weight and strength from an oral diet. Tube feeding has not been validated for efficacy in demented nursing home patients. Despite that, it has become commonplace and in addition, undue emphasis should be given to staff convenience. Randomized, controlled clinical trials of the efficacy of tube feeding in the demented nursing home patient should be funded and undertaken. Few studies have used actual costs rather than charges or published wholesale charges, the latter of which are subject to negotiated contracts. In addition, estimates for the cost of parenteral nutrition and EN generally have excluded the costs of hospitalization and treatment for complications related to these therapies. For example, patients are typically hospitalized 3–5 days/year because of complications related to home TPN; hospitalization may be required to treat complications of home EN, but is much less common [56]. The cost to an individual’s lifestyle should also be considered when determining economic costs (table 5). This includes physical and emotional problems stemming from nutritional therapy, sleep deprivation, loss of social prestige, changes in employment and leisure activity including exercise and travel, and time involved in required self-care. Nursing home costs of care are higher for patients not tube fed versus those that receive tube feeding [54]. Much of the increased cost is caused by the increased nursing staff and nursing assistant time to feed patients; food costs were similar between groups. Total costs including hospitalization and treatment of tube-feeding complications resulted in substantially increased overall costs in the tube-fed group. However, the brunt of these costs were 157
Ethics and Economics in Nutritional Support Table 5. Example of items in cost-benefit categories for home enteral nutrition
Direct
Indirect
Intangible
Cost
Benefits
Nutritional product Feeding set or syringe Pump and pole rental (if continuous drip) Tube site dressing supplies Nurse visits for instruction and follow-up Transportation to physician office for follow-up
Resources not spent on hospitalization costs Room Monitoring services
Personnel time for stocking and delivering the product and supplies Procedures for tube replacement (radiology, endoscopy) Time off work for family and friends to care for patient or transportation to physician follow-up visits Travel time for follow-up visits Loss of income Pain from disease or feeding tube Altered body image Worry over prognosis or length of time out of work Worry about burden of care place on family Worry about financial burden if enteral nutrition not covered by payer
Other costs Reduced risk of nosocomial complications Cost to payer is less (assuming family or patient able to provide own care and patient has full coverage)
Expense to patient may be less or more depending on coverage plan benefits Patient or family may be able to return to work
Patient may be able to resume hobbies
Patient able to be in familiar environment, benefits patient and family Patient feel more independent Patient has more control over daily activities
From Goff KL: Gastrointest Endosc Clin North Am 1998;8:733–744. Reprinted with permission.
not born by the nursing home. In addition, in at least 32 states, government reimbursement for the tube-fed patient is substantially greater than for the non-tube-fed patient [57]. Therefore, it appears that a financial incentive exists for nursing homes to require a PEG prior to acceptance from an acute care facility. In addition, state regulatory agencies often use weight measurement as a surrogate sole measure of nutritional status and therefore nursing 158
Ethics and Economics in Nutritional Support homes are pressured to maintain ‘normal’ weight and to prevent weight loss via tube feeding.
References 1 Dudrick SJ, Latifi R, Adams PR: Arrest and reversal of atherosclerosis with parenteral nutrition. Surg Clin North Am 1991;71:665–675. 2 Koretz RL, Lipman TO, Klein S, the American Gastroenterological Association: AGA technical review on parenteral nutrition. Gastroenterology 2001;121:970–1001. 3 Cozzaglio L, Balzola F, Cosention F, et al: Outcome of cancer patients receiving home parenteral nutrition: Italian Society of Parenteral and Enteral Nutrition (SINPE). JPEN J Parenter Enteral Nutr 1997;21:339–342. 4 Malcolm R, Robson JRK, Vanderveen TW, O’Neil PM: Psychosocial aspects of total parenteral nutrition. Psychosomatic 1980;21:115–125. 5 Sharp JW, Roncagli T: HPN in advanced malignancies. JPEN J Parenter Enteral Nutr 1992;16:190–191. 6 King LA, Carson LF, Konstantinides N, et al: Outcome assessment of home parenteral nutrition in patients with gynecologic malignancies: What have we learned in a decade of experience? Gynecol Oncol 1993;51:377–382. 7 Bozzetti F, Cozzaglio L, Biganzoli E, et al: Quality of life and length of survival in advanced cancer patients on home parenteral nutrition. Clin Nutr 2002;21:281–288. 8 Weiss SM, Worthington PH, Prioleau M, Rosato FE: Home total parenteral nutrition in cancer patients. Cancer 1982;50:1210–1213. 9 Finucane TE, Christmas C, Travis K: Tube feeding in patients with advanced dementia: A review of the evidence. JAMA 1999;282:1365–1370. 10 Schloendorff vs. Society of NewYork Hospital, 105 N.E. 92, 93, New York, 1914. 11 Buchman AL: Must every cancer patient die with a central venous catheter? Clin Nutr 2002;21:269–271. 12 O’Brien LA, Siegert EA, Grisso JA, et al: Tube feeding preferences among nursing home residents. J Gen Intern Med 1997;12:364–371. 13 Quill TE: Utilization of nasogastric feeding tubes in a group of chronically ill, elderly patients in a community hospital. Arch Intern Med 1989;149:1937–1944. 14 Murphy DJ, Burrows D, Santilli S, et al: The influence of the probability of survival on patients’ preferences regarding cardiopulmonary resuscitation. N Engl J Med 1994;330:545–549. 15 Lynn J, Childress JF: Must patients always be given food and water? Hastings Cent Rep 1983;13:17–21. 16 Buchman AL: Complications of home parenteral nutrition. Dig Dis Sci 2001;46:1–18. 17 Buchman AL: Practical Nutritional Support Techniques, ed 2. Thorofare, Slack, 2003, pp 63–68. 18 Mitchell SL, Kiely DK, Lipsitz LA: The risk factors and impact of survival of feeding tube placement in nursing home residents with severe cognitive impairment. Arch Intern Med 1997;157:327–332. 19 Rabeneck L, McCullough LB, Wray NP: Ethically justified, clinically comprehensive guidelines for percutaneous endoscopic gastrostomy tube placement. Lancet 1997;349:496–498. 20 Finucane TE, Bynum JPW: Use of tube feeding to prevent aspiration pneumonia. Lancet 1996;348:1421–1424. 21 Patel PH, Thomas E: Risk factors for pneumonia after percutaneous endoscopic gastrostomy. J Clin Gastroenterol 1990;12:389–392. 22 Lipman TO: The chicken soup paradigm and nutrition support: Rethinking terminology. JPEN J Parenter Enteral Nutr 2003;27:93–94. 23 Schostak Z: Jewish ethical guidelines for resuscitation and artificial nutrition and hydration of the dying elderly. J Med Ethics 1994;20:93–100. 24 Rosin AJ, Sonnenblick M: Autonomy and paternalism in geriatric medicine: The Jewish ethical approach to issues of feeding terminally ill patients, and to cardiopulmonary resuscitation. J Med Ethics 1998;24:44–48.
159
Ethics and Economics in Nutritional Support 25 Lo B, McLeod G, Saika G: Patient attitudes to discussing life-sustaining treatment. Arch intern Med 1986;146:1613–1615. 26 Pearlman RA, Uhlmann RF: Quality of life in chronic diseases: Perceptions of elderly patients. J Gerontol 1988;43:M25–M30. 27 Asai A, Fukuhara S, Lo B: Attitudes of Japanese and Japanese-American physicians towards life-sustaining treatment. Lancet 1995;346:356–359. 28 Payne K, Taylor RM, Stocking C, Sachs GA: Physician’s attitudes about the care of patients in the persistent vegetative state: A national survey. Ann Intern Med 1996;125:104–110. 29 Hodges MO, Tolle SW, Stocking C, Cassel CK: Tube feeding. Internists’ attitudes regarding ethical obligations. Arch Intern Med 1994;154:1013–1020. 30 Withdrawal of life-support from patients in a persistent vegetative state. Institute of Medical Ethics Working Party on the Ethics of Prolonging Life and Assisting Death. Lancet 1991;337:96–98. 31 Blackhall LJ: Must we always use CPR? N Engl J Med 1987;317:1281–1285. 32 McCann RM, Hall WJ, Groth-Juncker A: Comfort care for terminally ill patients. The appropriate use of nutrition and hydration. JAMA 1994;272:1263–1266. 33 Owen OE, Caprio S, Reichard GA, et al: Ketosis of starvation. Clin Endocrinol Metabol 1983;12:357–359. 34 Heber D, Byerley LO, Chi J, et al: Pathophysiology of malnutrition in the adult cancer patient. Cancer 1986;58:1867–1873. 35 American College of Physicians: Ethics manual, ed 4. Ann Intern Med 1998;128:576–594. 36 Schneiderman LJ, Jecker NS, Jonsen AR: Medical futility: Its meaning and ethical implications. Ann Intern Med 1990;112:949–954. 37 Solomon MZ, O’Donnell L, Jennings B, et al: Decisions near the end of life: Professional views on life-sustaining treatments. Am J Public Health 1993;83:14–23. 38 Goodhall L: Tube feeding dilemmas: Can artificial nutrition and hydration be legally or ethically withheld or withdrawn. J Adv Nurs 1997;25:217–222. 39 Cruzan vs. Director Missouri Department of Health. 497 US 261, 110 S. Ct 2841, 111 L. Ed 2nd 224, 1990. 40 In re Quinlin, 70 N.J. 10, 355 A.2d 647, cert. Denied sub nom. Garger vs. New Jersey, 429 U.S. 922, 1976. 41 Bundesärztekammer, Richtlinien der Bundsärztekammer für die Sterbebegleitung, DÄBL 90 B1791–2, 1993. 42 Schmidt P, Dettmeyer R, Madea B: Withdrawal of artificial nutrition in the persistent vegetative state: A continuous controversy. Forensic Sci Int 2000;113:505–509. 43 American Academy of Pediatrics Committee on Bioethics: Informed consent, parental permission, and assent in pediatric practice. Pediatrics 1995;95:314–317. 44 In re. Guardianship of Crum, 61 Ohio Misc. 2d 596; 580 N.E.2d 876 P.Ct. Franklin County, Ohio, 1991. 45 In re. Joelle Rosebush, 195 Mich. App. 675, 491 NW.2d 633:Michigan Appeals Court; 1991. 46 In re. Guardianship of Andrew James Barry, 445 So. 2d 365:Florida Appeals Court, 1984. 47 Leicher CR, DiMario FJ Jr: Termination of nutrition and hydration in a child with vegetative state. Arch Pediatr Adolesc Med 1994;148:87–92. 48 Kopelman LM, Irons TG, Kopelman AE: Neonatologists judge the ‘Baby Doe’ regulations. N Engl J Med 1988;318:677–683. 49 American Academy of Pediatrics Committee on Bioethics: Guidelines for foregoing lifesustaining medical treatment. Pediatrics 1994;93:532–536. 50 Nelson LJ, Rushton CH, Cranford RE, et al: Forgoing medically provided nutrition and hydration in pediatric patients. J Law Med Ethics 1995;23:33–46. 51 Miraie ED: Withholding nutrition from seriously ill newborn infants: A parents’ perspective. J Pediatr 1988;113:262–265. 52 Goel V: Economics of total parenteral nutrition. Nutrition 1990;6:332–335. 53 Trujillo EB, Young LS, Chertow GM, et al: Metabolic and monetary costs of avoidable parenteral nutrition use. JPEN J Parenter Enteral Nutr 1999;23:109–113. 54 ASPEN Board of Directors: Guidelines for the use of parenteral and enteral nutrition in adult and pediatric patients. JPEN J Parenter Enteral Nutr 1993;17:1SA–52SA 55 Moreno JM, Shaffer J, Staun M, et al: Survey on legislation and funding of home artificial nutrition in different European countries. Clin Nutr 2001;20:117–123.
160
Ethics and Economics in Nutritional Support 56 Reddy P, Malone M: Cost and outcome analysis of home parenteral and enteral nutrition. JPEN J Parenter Enteral Nutr 1998;22:302–310. 57 Mitchell SL, Buchanan JL, Littlehale S, Hamel MB: Tube-feeding versus hand-feeding nursing home residents with advanced dementia: A cost comparison. J Am Med Dir Assoc 2003; 4:27–33.
Discussion Dr. Bozzetti: I have a short comment and a question. The comment is that in my presentation I omitted a point that you have stressed which is when and how to withdraw nutritional support in this situation. With reference to the cancer patients, I think that the correct policy is to discuss the potential indications for nutritional support and for withdrawal of total parenteral nutrition (TPN) at the same time. I do this when I propose TPN at home, and the reasons for doing this is that if we are not able to reach our endpoint after 2 weeks or 2 months of nutrition, we have already agreed earlier that if there is no evidence of efficacy or benefit of TPN, we are authorized to withdraw the nutritional support. So this is a very important point when faced with competent patients or the family to discuss this point before starting home TPN. I was impressed by the article published in the Lancet by Chochinov et al. [1] a few years ago in which they analyzed the living wills of 168 patients. They were able to show how the living will changes during the day, during the week and during the month. So I question how we can rely on the living wills made by persons who are not ill when they make this decision; perhaps when they are sick they will change their ideas as sick people also change their ideas in the different periods of their disease. Dr. Buchman: Individuals are always free to change their living wills, but I have not had the experience of seeing this happen very frequently. In fact in our hospital it is required when a patient enters the hospital that there is some advanced directive noted on the front of the chart as to whether these patients, once they had cardiopulmonary resuscitation (CPR) for example, are to be transferred to an intensive care unit, this sort of thing. Of course one of the issues that we also have to be careful of is, for example, an individual who comes to the emergency room with a broken leg and says ‘just kill me, let me die, I can’t take this pain anymore’; well he really doesn’t want to be dead. But I think the living will still has to be respected because regardless of whether it changes a 100 times that is the patient’s decision and the most current decision is the most current decision that we have to respect. When they complete a living will, the medical issues need to be discussed with these patients by a health care provider who is duly informed so that the patients know exactly what it is that they are agreeing to or what they are not agreeing to, and the living will can also be very specific. For example it might allow a patient to be on a ventilator but not allow CPR, it might allow dialysis but not CPR and a ventilator. There are various combinations, and again this is the patient’s will and we as physicians have to respect that. It is a legal document and so it is not really up to us to determine whether it is rational or justified since it is a legal document. For us to determine if something is rational or justified such as a living will, I think is also somewhat inappropriate because that is a value judgment that we see only with our own eyes, and therefore it is by definition biased. Dr. Thomas: I just want to clarify a couple of things about the living will which I think will be helpful for those of you who may not be familiar with it. The legislation that covers the living will is convoluted and typically crazy. It requires that when a patient is admitted to a hospital he/she will be asked whether they have a living will or not. It does not require that they have a living will, and so there is a form that is filled
161
Ethics and Economics in Nutritional Support out by the admissions clerk buried in the chart that most physicians never see, and most patients in fact don’t have a living will. In the geriatric population we find that if you do surveys and questions you find very few people over the age of 65 or in nursing homes that want a living will or will fill one out even if help is provided. The law also requires that the hospital provide assistance to anyone who wishes to complete a living will on admission to the hospital. In fact that rarely happens. The second issue of the living will is that it only applies when two physicians determine that the patient is hopelessly, terminally and irreversibly ill. So, even though we typically view it as describing what people would like to have, in fact it does not take effect until the physician determines that there is no hope of recovery. Therefore, it only springs into effect at that late, late, late junction. The advanced directive laws in the United States are different; they allow someone to appoint a surrogate decision maker in case they can’t make a decision themselves. The living will is often used as a sort of guidance, but in fact most Americans don’t have a living will and most Americans don’t want to make a living will, and it is certainly not a requirement at admission. Dr. Buchman: I would certainly agree with that although I would suspect that the number of individuals in the United States with living wills have significantly increased over the last few years, because there has been a public effort to educate individuals about that. Certainly when they come to the hospital for the first time perhaps they may actually receive their initial education about it at that point. Dr. Bowling: I have a statement and a question. In English law we don’t provide for surrogate decision makers and it is actually down to the senior clinician involved; unlike Scotland where surrogates are involved. The question actually goes back to this conversation on advanced directives. I accept that they can be quite specific, but it is certainly my experience that they are often very nonspecific and are drawn up by an individual with or without some legal advice but usually not with medical advice. So, for example, you might be presented with an advanced directive that says ‘if I am rendered incapable I do not wish to receive nutritional support’. Now the problem that that can put you in is if a patient comes in with a stroke and you know that there may be a reasonable chance of recovery, perhaps several weeks down the line, but you are faced with an advanced directive saying I do not wish to be fed. What do you do? Do you try to persuade the family that actually this advanced directive is perhaps not appropriate, and then you might be sued if the patient does recover and has found out that he received nutritional support. I think it is very difficult. Dr. Buchman: That is a very difficult question and again the world of ethics is not to provide answers, it is actually to provide questions, and that makes my talk much more different from the others that will be presented. I think clearly all these issues are on the table, but one of the things that as health care providers we need to be careful of is not to allow our own value judgments to interfere with our recommendations for the patients; our job is to provide them with the data and they then make the decision. As I mentioned yesterday, I think there is hardly a patient that I can’t convince to have a gastrostomy tube at some point, but it may take a few visits. It all depends on tact and how it is explained. Dr. Bowling: But it can be pretty tough on families to put the decision making issue into their hands. It is a very difficult emotional and traumatic time for them, and I think we have got to be careful that we are not just passing the buck because we don’t want to get into difficult situations and difficult decisions. Dr. Buchman: I agree, but one of the things you also have to do is to support whatever decision the family makes because that allows the family to have some feeling that they have made the right decision, when in fact there may be no right or wrong decision. But you have got to support their decision. There is a difference between supporting their decision and making the decision for them. Often times you
162
Ethics and Economics in Nutritional Support will be presented with a family who wants you to make the decision because either they don’t know or more likely they don’t want to actually make the decision themselves. Again your job is to provide them with the amount of reasonable information that they need in order to make that decision. Now you can provide some guidance but that guidance should be nothing more than guidance, not making the decision for them. Dr. DeLegge: I have two questions for you, ethically or I guess legally with regard to home parenteral nutrition. For someone who is about to go home and has not been on home parenteral nutrition before, do you think that the informed consent process should be more formal, meaning rather than just an explanation to the patient? Similar to other therapies that you might do, for example a liver resection, the consent process would be a lot more formal, there would be a lot more said about it, it would be written and very detailed. That was question one. The question two is in regard to one of the situations that I always have the most difficulty with; it is a rare instance but it happens that a family wants to withdraw nutrition support, that I may or may not agree with. In those situations do you see any role for a hospital-based ethics committee to help and evaluate the situation from outside? Dr. Buchman: I will answer the second question first. There are certainly times when, like you, I have also perhaps disagreed to some degree or even completely with the families’ desires, and that is a situation where you need to remove yourself from the decision-making process. I often start by getting a second opinion from a colleague, and certainly a hospital-based ethics council is also appropriate, but remember that an ethics council is not necessarily going to provide you with an answer. They are going to provide you with the appropriate questions that you and the family and the other providers need to think about. The family and team meetings can also be useful for everyone to have a chance to voice their opinions and disagreements, and that would include the family and in many cases the patients themselves, the nurse, perhaps the dietician, and whoever else is involved in the patient’s care so that a consensus opinion in terms of the direction of treatment can be derived. I personally hate those sorts of things because they normally take at least an hour away from your usual work, but sometimes there is no alternative. In terms of your first question with the consent for nutritional support; many of you may know that in the United States the regulations in the last couple of years in health care have become overwhelming and much of this has come about because there were some indignities involved in terms of health care and health care research: there were patients at John Hopkins and other places who were in experimental studies to which they had never even consented to be in. Similar things went on here 50 years ago, perhaps even to an almost similar degree. These are very isolated events, but unfortunately what has come out of those is a whole set of requirements and privacy statements and so forth that has made actually doing clinical research extraordinarily difficult. In fact even taking care of patients is difficult. We can’t even talk to a consultant now and tell them about the patients they are about to see unless the patient has actually signed a document and that document is faxed to the physician. A patient can e-mail me with a question and legally I can’t even e-mail an answer back unless they fax a signed consent form allowing them to communicate by e-mail, even though they initiated the contact. So I shudder at the idea of adding another written document to the overwhelming number of documents now, but I think it probably would be better then. Perhaps it is not inappropriate actually to have a specific document for an informed consent for nutritional support itself just as you would for a surgical procedure or endoscopy for example, but I think it is important for these patients because they are very few. The patients who require home TPN for example are almost 40,000 in the US but probably there are around 5,000 to 10,000 requiring this on a permanent basis. These patients should
163
Ethics and Economics in Nutritional Support be taken care of by perhaps 10 centers across the US, as much as they are in France and other countries. Europe seems to be backing a way from that, but I think once the health care providers have a particular expertise in nutritional support, for example, they are uniquely qualified in order to provide the patient with information such as the cost and benefits that they may achieve. In my opinion, I think that is even more effective than having a signed consent because patients can sign anything put in front of them, but its value is only how well they understand the consent and how well it is explained to them; it also reflects whether a level of coercion is involved. I don’t do what we call open colonoscopy. Screening colonoscopy has become a big thing in the United States, it is a money-maker for the gastroenterologists and a big thing for patients. Often an internist will simply check colonoscopy off on the lab sheet so to speak, as though it was a complete blood count. The gastroenterologist meets the patients for the first time when an intravenous catheter has already been placed. The patients then sign an informed consent form after they have already drunk the prep or they have had sat on the john for 2 days straight. That is coercion because those patients know they have already done all this. They are not going to refuse because they have already done everything. Often they don’t even read the consent. I don’t practice that because that is how lawsuits can happen. My practice is that I must actually see every patient beforehand, and not 5 min before the procedure but at least a week beforehand. The procedure is explained to the patient regarding the indications, the risks and alternatives, and I document these as well as the fact that I answered all their questions. An exception is emergency procedures obviously. They get a copy of the consent form and they can take it home and read it over 25 times if they want, call up and ask questions about it. Therefore, if we are going to provide informed consent to the patients minutes before we start TPN or before we fill out the TPN order form or before we put the nasogastric tube in, that is not an appropriate informed consent anyway; it is really a kind of coercion when you wave a nasogastric tube in front of them. This is actually a very complicated issue, I am not sure what the right answer is as I have some of my own opinions, but I don’t know if they are correct. Dr. Morley: First I have a comment and then a question. The comment again is the durable power of attorney of living wills. As a geriatrician I actually deal with these far more often than most people, I have most probably read 1,000 plus of them and I have never been able to interpret at the end what I am supposed to do because they are written by lawyers who say but if, and there is always an if and a whatever. They are totally unintelligible documents when you read them carefully; they are virtually never provided by the patient after discussion with a physician or any health care provider; they are done by people who have no concept. So you will see many people say, I really want somebody to push on my chest if my heart stops but I don’t want ventilation. Certainly with nutrition it is: ‘do you ever want a feeding tube’, ‘yes though not if I have a stroke’. So I really think there are two documents created by lawyers and ethicists which like everything done in my opinion by lawyers and ethicists is most probably to the detriment of the patient. That is a sweeping statement that I don’t totally believe, but nevertheless it is done by those two groups of people with their inability to come to grips with what you really should do. What I do with most of my patients is ask them early on basically what they want. We go through and have this discussion which takes up to an hour, which is why not every patient always get it, and I try to do this long before the patient gets into hospital. At that stage I at least have a baseline of where they want to go, and I send them home to discuss it with their family and ask them to come back and tell me what their family feels about it. If you don’t do that I think we are not talking informed consent, we are making it up from their own. So I think we have got huge problems, and quite honestly I would trust an Italian physician who makes the decision without asking the patient more
164
Ethics and Economics in Nutritional Support than I trust the American legal system, but that is another way of looking at. But more important are the patients who decide not to have nutritional support. So they then become ill, go into hospital and the first thing that happens is they get intravenous fluid, and quite honestly I don’t believe that dying from starvation, having watched many patients do this, is a nice thing. They die with multiple complications and I don’t think they do very well. I think dying from dehydration is a wonderful way to die. I cannot understand why basically in every major hospital in the US the first thing that is done when somebody comes in with withdrawal of nutritional support is to give them an intravenous, to give them fluids, and I would like your comments on the fluid part of it. Dr. Buchman: To comment on the first part of your statement, I completely agree with you. I think in theory the living wills are wonderful documents but the problem is that again a lot of lawyers try to usurp the authority from physicians. Although there are some that I have dealt with that are plainly written and very easy to follow, I think it is something that needs to be worked on. I think the idea, the concept is good, but it needs to be amended. In terms of the issue of starvation, I didn’t really mention it in my talk but there have been studies done that have looked at discomfort during starvation [2]. We think that it is a terrible thing to die of starvation, it is painful and we should be providing comfort care. But there is actually evidence that would suggest that in starvation endorphin levels are increased in the blood and the body actually sort of takes care of itself, and that ketone production occurs in starvation because it is also associated with a decrease in pain sensation. Surveys have been done in nursing home patients who have been starved [2–4]. Only a very small minority has complained of hunger-like symptoms and they don’t have thirst either, they don’t complain of thirst symptoms. This is contrary to what makes common sense, but unfortunately in medicine there are some things that make common sense that actually aren’t true and there are some things that make no sense at all that are actually true. This unfortunately happens to be the case, at least on the evidence-based data that exist in the literature. On the other hand mouth care is very important. Putting Vaseline® ointment to keep the mouth moist, allowing them some drops of water, ice cubes for example, that actually is very important. Again you have much more clinical experience than I, I am unbiased because I am just simply reading the evidence-based literature and I have not been biased by the clinical experience because by the time a patient just requires any Vaseline and a couple of ice cubes in the mouth in the nursing home I am long gone from their care. Perhaps if your experience is different, then you should convene another study. There are currently several present in the literature. Dr. Morley: Can I just follow up on that because the studies in the literature are absolutely horrendously badly done. They die of their hunger; no one says that in dying from starvation hunger is the problem. The problem in my patients who die in the nursing home or die in home care if I withdraw nutritional support is that it takes them sometimes 4–6 months because they are usually eating a little bit, they are drinking a little bit, they may have 8–10 hospital admissions. They will certainly finish up with a pressure also somewhere along the line that we could argue whether that was due to the nutritional support or to the lack of it. They will end up with a variety of skin tests, a variety of other things that will create problems. That is never taken into account by someone who goes in and asks if the people are hungry. Well I don’t care whether people are hungry, if they are happy not being hungry, it doesn’t matter to me or even being hungry, but certainly when I watch what happens to my patients when they starve, they don’t die quickly, and that is why I really believe in dehydration. Dehydration, nobody goes longer than a week if you stop giving them fluids and most go in less than that and they go very quickly. In many of the cases where people
165
Ethics and Economics in Nutritional Support have gone in and done these studies, they don’t distinguish between people who are truly dehydrated, who have no response at all. Many of these hunger studies are done in mental patients, and quite honestly I have no idea how to ask mental patients whether they are hungry or not. But again the studies you are talking about were done purely by ethicists, most of whom I know very well and they have never worked in a nursing home. It is amazing how people who never worked in a place can come up with great answers; when they work there they seem to change their mind. Dr. Buchman: I think you have some excellent points here. One of the things though in terms of the efficacy that you alluded to is there is a new cubism. There is cubism in art but there is also cubism in nutritional support. That would be in contrast to the American court system for example where one is innocent until proven guilty. In Cuba in the court system one is guilty until proven innocent. That is how I view nutritional support. It doesn’t work until it is proven to work, and it has not been proven to decrease pressure sores or otherwise improve survival in nursing home patients. In fact, as pointed out by the study that was presented yesterday, survival and morbidity are decreased in nursing home patients in relation to their age, their underlying diagnosis and their code status, whether they are ‘do not resuscitate’. Those are the only three things that actually have importance in terms of survival and perhaps in the well-being of these patients. Perhaps in the future we will find that there are some other things and perhaps nutrition does play a role, but unfortunately at the present time nutrition just doesn’t have it and we just don’t get it. That is what is in the literature. Dr. Bozzetti: Just a short comment because Dr. Morley mentioned the Italian physician. I think that informed consent is very important, but if the patient does not want to be informed I will respect his desire. Dr. Buchman: I would agree, that is part of patient autonomy. If the patient says ‘tell my family that I have got metastatic liver cancer, don’t tell me’, or ‘if it is really bad, don’t tell me’, and they truly believe this, do you have to respect all their decisions? If they have the capacity to understand what you are saying, one has to respect their judgment as well.
References 1 Chochinov HM, Tataryn D, Clinch JJ, Dudgeon D: Will to live in the terminally ill. Lancet 1999;354:816–819. 2 McCann RM, Hall WJ, Groth-Juncker A: Comfort care for terminally ill patients. The appropriate use of nutrition and hydration. JAMA 1994;272:1263–1266. 3 Owen OE, Caprio S, Reichard GA Jr, et al: Ketosis of starvation: A revisit and new perspectives. Clin Endocrinol Metab 1983;12:359–379. 4 Heber D, Byerley LO, Chi J, et al: Pathophysiology of malnutrition in the adult cancer patient. Cancer 1986;58(suppl):1867–1873.
166
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 167–178, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
Pathophysiology of Weight Loss in Older Persons John E. Morley GRECC, VA Medical Center, and Saint Louis University School of Medicine, St. Louis, Mo., USA
‘… for in old men, even without any disease, owing to their being near the close of life, the appetite is nearly gone.’ Aretaees the Cappadocian
It is now well recognized that a physiological anorexia of aging occurs that is associated with gradual weight loss in older persons [1, 2]. A number of epidemiological studies have shown that weight loss in older persons leads to death (fig. 1) [3–6]. In addition, weight loss has been shown to be associated with hip fracture [7]. When an older person develops a disease, the disease interacts with the propensity for anorexia, leading to severe weight loss and cachexia. The Hebrew physician, Maimorides differentiated between sarcopenia and cachexia: ‘…for wasting which resembles old age (sarcopenia) and wasting which is secondary to fever (cachexia) and wasting which is called doalgashi (starvation)’. The differences between the three forms of weight loss are delineated in table 1. In nursing home residents who were losing weight, it was shown that those who could reverse their weight loss had a much lower mortality rate than those who continued to lose weight [8]. The Cochrane collaboration found that caloric supplementation decreased mortality [9].
The Physiological Anorexia of Aging The anorexia of aging is due in part to social factors, especially isolation. Having someone eating with an older persons [10] or enhancing the environment of the dining room [11] have both been shown to enhance food intake. 167
Pathophysiology of Weight Loss in Older Persons Pamuk et al: Am J Epidemiol 1992;136:686. BMI
Reynolds et al: JAGS 1999;48:1409. BMI
1.8
⬍26
2.7 2.1 1.9 2.0
26–29 ⬎29
2.0 Odds ratio
3.06 3.84
23–28 2.8
1.0
⬍23
2.53
⬎28 1.0
3.0
2.0 3.0 Odds ratio
4.0
1.69 1.62
Lozonczy et al: Am J Epldemiol 1995;141:312.
2.2
Degroot et al: J Nutr Hlth Age 2002;6:4–8.
1.3 1.0
2.0 Odds ratio
3.0
Fig. 1. Weight loss leads to death.
With aging there is an increase in taste sensory threshold and a decrease in smell. Both smoking and drugs can further decrease the ability to enjoy food. Zinc deficiency, which is common in diabetics and persons taking diuretics, can lead to hypogeusia. The use of taste enhancers has been shown to reverse weight loss [12]. The major reason for the early satiation that occurs in older persons is altered processing of food in the stomach [13]. Large caloric meals slow gastric emptying. With aging there is a decrease in fundal compliance resulting in more rapid antral filling and earlier satiation [14, 15]. Decreased release of nitric oxide in the fundus in response to food appears to be the reason for the decline in fundal compliance [16, 17]. In addition, older persons have an increase in cholecystokinin levels. This increase is particularly marked in response to fat [18]. Cholecystokinin is also a more potent satiating agent in older animals [19] and humans [20]. We have demonstrated that caloric supplements given 60 min before a meal are more effective at increasing caloric intake than when they are given with the meal [21]. Males develop greater anorexia with aging. This appears to be due to the declining testosterone levels leading to an increase in leptin [22]. Leptin is a peptide hormone that causes anorexia and an increase in metabolic rate. The anorectic effects of leptin are attenuated in the presence of hypertriglyceridemia. Within the central nervous system a variety of neurotransmitters, such as norepinephrine, neuropeptide Y, serotonin and nitric oxide are responsible for transducing the effects of these peripheral signals. 168
Pathophysiology of Weight Loss in Older Persons Table 1. Comparison of anorexia (starvation), sarcopenia and cachexia
Body mass Fat-free mass Body fat Food intake Resting metabolic rate Physical activity Proteolysis Insulin resistance Cortisol Triglycerides Cytokines Leptin
Anorexia
Sarcopenia
Cachexia
⫺⫺ ⫺ ⫺⫺⫺ ⫺⫺⫺ ⫺ ⫺ ⫺ 0 ⫾ 0 ⫾ ⫺
⫺ ⫺⫺ 0 0 ⫺ ⫺ ⫹ 0 ⫾ 0 ⫹ 0
⫺⫺⫺ ⫺⫺⫺ ⫺⫺ ⫺⫺ ⫹⫹ ⫺⫺ ⫹⫹ ⫹ ⫹⫹ ⫹⫹ ⫹⫹⫹ ⫹
⫺ ⫽ Decrease; ⫹ ⫽ increase.
Sarcopenia Sarcopenia (lack of flesh) refers to the loss of muscle mass that occurs with aging [23]. Sarcopenia is associated with decreased strength, frailty and disability. Persons who remain obese while losing muscle mass – ‘the obese sarcopenia’ – do particularly poorly as they age. Numerous factors are involved in the pathogenesis of sarcopenia. A decline in testosterone in males appears to be particularly important in the pathogenesis of sarcopenia [24]. Other factors involved include decreased food intake, decreased physical activity, a decline in muscle insulin-like growth factor-1, lack of appropriate neural input and poor vascular supply. Interleukin-6, the ‘geriatric cytokine,’ has also been implicated in the pathogenesis of sarcopenia.
Cachexia Cachexia comes from the Greek meaning ‘poor condition’. Cachexia is a condition of severe wasting which occurs in the presence of chronic inflammation. Cachexia is associated with a redistribution of protein, with loss of protein from muscle and an increase in hepatic protein synthesis, leading to an increase in circulating acute phase proteins, such as C-reactive protein and serum amyloid protein. Persons with cachexia lose fat-free mass mainly from muscle as well as an equal amount of fat. They maintain extracellular water. Cachectic patients usually have low serum albumin and cholesterol levels. Cachectic patients have an increase in resting metabolic rate but a decrease in physical activity. 169
Pathophysiology of Weight Loss in Older Persons Table 2. Reversible causes of weight loss – the MEALS-ON-WHEELS mnemonic Medications Emotional (depression) Alcoholism, elder abuse, anorexia tardive Late life paranoia Swallowing disorders Oral problems Nosocomial infections, no money (poverty) Wandering and other dementia related behaviors Hyperthyroidism, hypoadrenalism, hyperglycemia, hypercalcemia Enteral problems (gluten enteropathy, pancreatic insufficiency) Eating problems Low salt, low cholesterol diet Stones (gallstones), shopping problems
Cachexia is triggered by the release of cytokines. Tumor necrosis factor-␣, interleukin-1 and interleukin-6 are the major cytokines involved in the pathogenesis of cachexia. Patients with cachexia often have an associated anorexia. Within cells protein degradation occurs when proteins bind to ubiquitin and are then transported to a proteasome, where proteolysis occurs. Cortisol, tumor necrosis factor-␣, and interleukin-1 stimulate this pathway. In addition, cachexia is associated with decreased hepatic lipoprotein lipase activity and increased triglycerides (table 2). An associated insulin resistance decreases glucose going to muscle while making it available for the liver and other organs.
Pathological Causes of Weight Loss The majority of causes of weight loss in older persons are potentially reversible. Nearly a third of older persons with weight loss have depression [25, 26]. Depression classically causes anorexia. Treatment of depression reverses weight loss. In severe cases this requires electroconvulsive therapy. Numerous medications produce anorexia including digoxin, theophylline, cimetidine, and fluoxetine. Elder abuse occurs in 5% of older persons and may result in weight loss. Late life alcoholism is also a cause of weight loss. Females who had asthenia (anorexia nervosa) when young may redevelop an anorexia tardive in old age. Cholesterol phobia has also been associated with weight loss. Late life paranoia can result in the rejection of food. Dysphagia can result in aversion to eating. Oral disease can lead to a decrease in intake of about 100 cal/day. Nosocomial infections such as tuberculosis, recurrent Clostridium deficile, and Helicobacter pylori, are common 170
Pathophysiology of Weight Loss in Older Persons chronic infections that lead to weight loss. Poverty can lead to insufficient funds to purchase food. Some elderly have problems with shopping or food preparation related to a decline in instrumental activities of daily living. Persons with a decline in basic activities of daily living may require assistance with eating. It can take up to 45 min a meal to feed an older person. Persons with dementia may forget to eat. They can also develop apraxia of swallowing. Demented persons who wander excessively may need a marked increase in calories. A number of metabolic conditions are associated with weight loss. Hypercalcemia leads to anorexia. Apathetic hyperthyroidism is a classical cause of weight loss. Persons with Addison’s disease can have abdominal pain, diarrhea and weight loss. Older persons with pheochromocytoma continue to be hypertensive in the face of severe weight loss. Enteral problems leading to malabsorption include gluten enteropathy and pancreatic insufficiency. Persons with Parkinson’s disease and other tremors may find it difficult to get food to their mouth. Therapeutic diets have been shown to be associated with weight loss. There is no evidence that these diets have positive effects on disease processes [27]. It is no longer recommended to use therapeutic diets in nursing homes. Gallstones can present with early satiation. Persons with severe cardiac disease cannot only be anorectic, but with hepatic congestion decrease albumin production and lose protein from the gut. Diabetes, when uncontrolled, is associated with weight loss. Persons with chronic obstructive pulmonary disease have an increased resting metabolic rate but decrease their physical activity. These persons become hypoxic while eating because of the oxygen necessary to maintain the thermic energy of eating. This leads to food avoidance and severe anorexia. Cancer accounts for 10% of the causes of weight loss in older persons.
Conclusion Weight loss is a major barometer of impending frailty in older persons. Numerous causes of weight loss are reversible. We have recently found that an appetite questionnaire can predict weight loss. All older persons should have their appetite regularly assessed and if they have a decline in appetite, reversible causes should be considered.
References 1 Morley JE: Anorexia and weight loss in older persons. J Gerontol Med Sci 2003;58A:131–137. 2 Morley JE, Silver AJ: Anorexia in the elderly. Neurobiol Aging 1988;9:9–16. 3 Wedick NM, Barrett-Connor E, Knoke JD, Wingard DL: The relationship between weight loss and all-cause mortality in older men and women with and without diabetes mellitus: The Rancho Bernardo study. J Am Geriatr Soc 2002;50:1810–1815.
171
Pathophysiology of Weight Loss in Older Persons 4 Reynolds MW, Fredman L, Langenberg P, Magaziner J: Weight, weight change, mortality in a random sample of older community-dwelling women. J Am Geriatr Soc 1999;47:1409–1414. 5 Ryan C, Bryant E, Eleazer P, et al: Unintentional weight loss in long-term care: Predictor of mortality in the elderly. South Med J 1995;88:721–724. 6 Crotty M, Miller M, Giles L, et al: Australian Longitudinal Study of Ageing: Prospective evaluation of anthropometric indices in terms of four year mortality in community-living older adults. J Nutr Health Aging 2002;6:20–23. 7 Ensrud KE, Ewing SK, Stone KL, et al: Intentional and unintentional weight loss increase bone loss and hip fracture risk in older women. J Am Geriatr Soc 2003;51:1740–1747. 8 Sullivan DH, Morley JE, Johnson LE, et al: The GAIN (Geriatric Anorexia Nutrition) Registry: The impact of appetite and weight on mortality in a long-term care population. J Nutr Health Aging 2002;6:275–281. 9 Milne AC, Potter J, Avenell A: Protein and energy supplementation in elderly people at risk from malnutrition. Cochrane Library. Cochrane Review, issue 1. Chichester, Wiley, 2004. 10 Suda Y, Marske CE, Flaherty JH, et al: Examining the effect of intervention to nutritional problems of the elderly living in an inner city area: A pilot project. J Nutr Health Aging 2001;5:118–123. 11 Mathey MF, Vanneste VG, de Graaf C, et al: Health effect of improved meal ambiance in a Dutch nursing home: A 1-year intervention study. Prev Med 2001;32:416–423. 12 Mathey MF, Siebelink E, de Graaf C, van Staveren WA: Flavor enhancement of food improves dietary intake and nutritional status of elderly nursing home residents. J Gerontol Med Sci 2001;56:M200–M205. 13 Clarkston WK, Pantano MM, Morley JE, et al: Evidence for the anorexia of aging: Gastrointestinal transit and hunger in healthy elderly vs. young adults. Am J Physiol 1997;272:R243–R248. 14 Rayner CK, MacIntosh CG, Chapman IM, et al: Effects of age on proximal gastric motor and sensory function. Scand J Gastroenterol 2000;35:1041–1047. 15 Morley JE: Anorexia of aging: physiologic and pathologic. Am J Clin Nutr 1997;66:760–773. 16 Chapman IM, MacIntosh CG, Morley JE, Horowitz M: The anorexia of ageing. Biogerontology 2002;3:67–71. 17 Morley JE: Pathophysiology of anorexia. Clin Geriatr Med 2002;18:661–673. 18 McIntosh CG, Horowitz M, Verhagen MA, et al: Effect of small intestinal nutrient infusion on appetite, gastrointestinal hormone release, and gastric myoelectrical activity in young and older men. Am J Gastroenterol 2001;96:997–1007. 19 Silver AJ, Flood JF, Morley JE: Effect of gastrointestinal peptides on ingestion in old and young mice. Peptides 1988;9:221–225. 20 McIntosh CG, Morley JE, Wishart J, et al: Effect of exogenous cholecystokinin (CCK)-8 on food intake and plasma CCK, leptin, and insulin concentrations in older and young adults: Evidence for increased CCK activity as a cause of the anorexia of aging. J Clin Endocrin Metab 2001;86:5830–5837. 21 Wilson MM, Purushothaman R, Morley JE: Effect of liquid dietary supplements on energy intake in the elderly. Am J Clin Nutr 2002;75:944–947. 22 Morley JE, Perry HM, Baumgartner RP, Garry PJ: Leptin, adipose tissue and aging – Is there a role for testosterone? J Gerontol Biol Sci 1999;54:B108–B109. 23 Morley JE, Baumgartner RN, Roubenoff R, et al: Sarcopenia. J Lab Clin Med 2001;137:231–243. 24 Morley JE: Anorexia, sarcopenia, and aging. Nutrition 2001;17:660–663. 25 Wilson MM, Vaswani S, Liu D, et al: Prevalence and causes of undernutrition in medical outpatients. Am J Med 1998;104:56–63. 26 Morley JE, Kraenzle D: Causes of weight loss in a community nursing home. J Am Geriatr Soc 1994;42:583–585. 27 Tariq SH, Karcic E, Thomas DR, et al: The use of a no-concentrated-sweets diet in the management of type 2 diabetes in nursing homes. J Am Diet Assoc 2001;101:1463–1466.
Discussion Dr. Elia: I was particularly interested in your comments on the anorexia of aging which might be expected to produce weight loss. However, we are increasingly seeing
172
Pathophysiology of Weight Loss in Older Persons elderly people who are overweight/obese. So, is the anorexia of aging defined in relation to the appetite behavior of younger individuals or to the behavior of those concerned? In addition, do your comments relate to older individuals who are ‘healthy’ or those with disability and disease? Dr. Morley: The first part of my lecture was about healthy people and we have actually looked at extraordinarily healthy people, and there is no question that if you take a 20- to 30-year-old and a healthy 70-year-old, the 70-year-old eats much less. Now the 70-year-old also exercises much less. So what you are looking at is that over the lifetime food intake should decrease along with the decrease in physical activity which has been shown in many studies. The epidemic of obesity we are seeing at the moment is occurring predominantly in middle-age people, both in the United Kingdom and the United States, and these people are already eating less than you would expect. So now you have got to stop and say what is wrong with the scenario when the CDC in the United States says we have got to eat less. What is wrong with it if that is not the problem? The problem is we are not exercising enough. Our whole lifestyle has changed from the beginning of the 20th century to now. At the beginning of the 20th century most people didn’t sit in a hall for 3 days thinking that this was hard work; basically they would have been out there working hard because the jobs were hard. The jobs have changed; what people do has changed. In the United States no one walks anywhere; a car is taken to every place. We have stopped exercising and I think what we are failing to do is recognize that exercise is the key to weight reduction and it has very little to do with food intake other than that we are over-eating. If you go back to the original studies by Willet in which he showed that high cholesterol was bad for the heart, he couldn’t do this until he took one factor out of his epidemiological analysis. This is why I hate epidemiology because epidemiologists have no trouble in throwing things that they don’t understand out. In his older biography Willet pointed out that they had to take out this one thing: the more you ate the less heart disease you had, and that was very clear in all the early epidemiological studies. Now we know obesity produces heart disease. We know that fundamentally this does not make sense, but it does if you think of it, because people who eat a lot usually exercise a lot as well. So exercise becomes important and food in itself may be protective. What you find here again is that weight loss is bad for you. Now if weight loss is bad for you, ethically we should not be telling people to lose weight by not eating. I think we should be telling them to exercise, and I don’t think that I have seen any data to support that long-term dieting is a good way to do anything. It has been shown that with dieting you lose muscle mass as well as fat mass, and if you are like me when I diet, eventually put it back on. You actually put on more fat than you put on muscle, you put on about 15% muscle when you go back. In our studies in New Mexico we have shown that the obese sarcopenic subjects had by far the worst outcome. All this suggests to me that we are talking about an exercise problem with obesity or a physical activity problem, and I think that is what we should beginning with rather than looking for a food intake problem; lack of food intake is bad for us, exercise is good for us. I don’t know if that answers your question but that is what I have come to terms with because I have to tell all my patients not to diet anymore. Dr. Elia: In terms of the weight maintenance that is very important. The issue of appetite regulation is very complex. Dr. Morley: I wouldn’t disagree with that. I have spent 20 years to prove it isn’t and it really is. Dr. Armstrong: Do you have any data on the effect of exercise on appetite in the elderly or in any populations? Dr. Morley: There are actually minuscule data on the effect of exercise on appetite in all the people. It is interesting that it suggests that there is an early decline when
173
Pathophysiology of Weight Loss in Older Persons you start exercising followed by an increase. Particularly if you have exercised a lot as I have at different times in my life, you recognize basically that you are often very anorexic when you finish exercising at a high level. This is most probably because of an increase in corticotropin-releasing factor (CRF) which occurs in major exercise. The increase in CRF is very anorexic for animals and actually in humans. So I think you get short-term anorexia, long-term with bouts in between you tend to increase your appetite and eat more. That is a long-term effect as perhaps the changes in leptin and issues like that. In addition in Diabetes [1] we very recently showed that hypertriglyceridemia actually blocks the leptin effect. In fact when people try to give leptin to obese people what happens is that basically it doesn’t decrease their appetite unless they were truly leptin-deficient which very few people are. Nobody could understand that. we believe that it is the hypertriglyceridemia in fat people and therefore when you exercise you get rid of some of the fat over time. You lower your hypertriglyceridemia so you start to get an effect of leptin but in the intermediate period you actually have an increase in appetite and then it balances is out. It is extraordinarily complex. The whole problem is that every single thing about appetite and its relationships with the rest of the body are complex, and if you only look at one thing you get the answer wrong. Dr. Biesalski: Regarding some of the interventions you described in nursing homes: my view of nursing homes is that lots of people are sitting around in chairs not doing very much. Perhaps the sort of chronic exercise you describe is affecting appetite and may also play a role in improving people’s nutritional status. Dr. Morley: We spent a lot of time in nursing homes trying to get all the people in the nursing homes to be involved in activity, to pull people out of bed. People who stay in bed do poorly; immobility is poor; people need to actually be moved around. A meta-analysis of the long-term effects of exercise basically came to the conclusion that it did nothing. Now the problem here is that long-term effects require people with intention-to-treat to stay on the exercise program, and most human beings are not very good and they should believe in exercise and stay on the exercise program. So in fact the drop off when you stop exercise is very rapid. Dr. Biesalski: Would you generally recommend that the elderly take multivitamins or multi-micronutrient supplements? Dr. Morley: That is a very tough question because if you are eating adequately you will get adequate micronutrients. What I try to tell my patients is that they should eat adequately and that I believe, though I have absolutely no evidence, that micronutrients when given in food are better than micronutrients when taken in a tablet. If somebody is anorexic and not eating they certainly need a vitamin supplement, but there is no evidence one way or the other whether or not you should give free radical inhibitors. We use ␣-lipoic acid rather than vitamin E because some of our basic science studies show that in Alzheimer’s animals it actually reverses Alzheimer’s disease. The animal data are very good for using free radical inhibitors; the human data are not nearly as good and I think that is where we run into problems. There are some data on ␣-lipoic acid for Alzheimer’s disease now, but at least the Cochrane database [2], which I believe in fundamentally, says that there are insufficient data to say it is good or bad. I think we have to live with that until we have better data. So if people are starving, they need micronutrients. I think we are now giving much more intravenous multivitamins as people come to hospital because almost all of those people over the age of 70 are vitamindeficient, but again there are no data, just a belief. This field is total belief unfortunately. Dr. Biesalski: There is an argument frequently used that a diet or an energy intake below 1,500 kcal is not sufficient with respect to micronutrients. Can you ensure that all the elderly in nursing homes have an energy intake that is more than 1,500 up to 2,000 kcal?
174
Pathophysiology of Weight Loss in Older Persons Dr. Morley: Again it depends on the diet. The arguments I have seen are that around 1,000 cal are essential and it depends on what is in that diet. Obviously in most of our American nursing home diets 1,000 cal makes the daily allowance, just. So I can say that if we are treating people properly they get about 1,000 cal. If they are not being properly treated, then I need to be aggressively treating them or they need to be on a multivitamin. But as soon as people lose weight they should be on a multivitamin and one exception I didn’t mention is calcium and vitamin D. Clearly studies in nursing homes have now shown that calcium and vitamin D decrease hip fracture. There is no excuse for any older person in a nursing home not to be on calcium and vitamin D, and women most probably should be doing this from 50 years old. 85% of the women in nursing homes in the United States are osteoporotic; so with those figures everybody should be getting calcium and vitamin D. But there are no easy answers and the more I look at the literature, the more confused I get. Every time I read an article I change my mind, which says we don’t have enough literature. Dr. Labadarios: How careful should we be in assuming or saying that people who eat adequately don’t need micronutrients? I do not subscribe to the believe that everyone who is in an institution needs micronutrients. I am making the comment because if one reads the literature and takes the people or the population in the world that probably eats the most, that is the USA, there is still a significant proportion of that population in which micronutrient intake right across the ages would actually be inadequate if it was not from the ingestion of supplements, that is point number 1. Point number 2, if you look at the data from Europe, particularly from Germany, they are really not much different. Point 3, in view of the emerging evidence on the role of micronutrients, and I am not talking about antioxidant micronutrients only particularly the B group of vitamins, and their role or their possible role in chronic disease, can we afford to assume that people who eat adequately don’t need micronutrient supplements? Dr. Morley: It is a very good point and we don’t know the answer. First of all in the Unites States flours are fortified with folate. If I was in Europe I would have said in answer to you that there is not as much fortification. Certainly folate B12 needs to be looked at much more carefully. We have looked at our patients with B12 deficiency or elevated homocysteine, which is different, and in the Unites States it appears to be mostly due to renal failure rather than to folate B12 because high doses won’t bring it down. B12 cures a lot of things. So I think we have got to think about this, I think we don’t know exactly when we are giving people the right amount. The problem I have is giving a tablet to someone who is already on 9 tablets. We did a study 15 years ago in which we looked at people on over 10 drugs, and we reduced the drugs by 5 and reduced hospitalization and death by a third, and the number 1 drug we stopped was multivitamins. The reason was that every older person takes multivitamins. What was happening is they take the multivitamin but they wouldn’t take the diuretics, so by stopping it we improved compliance. So every time you add a tablet, even if you think it is totally innocuous, you have got to realize that in an older person this can really interfere with compliance. I don’t know the answer. Dr. Labadarios: My other point relates to weight loss. In the data that you showed us there were two types of weight loss: the involuntary one which had a much higher risk of mortality, and the voluntary one which was associated with a relative risk of 1.78. Dr. Morley: Most of the data were basically considered as a mix of voluntary and involuntary. The only purely voluntary one was with hip fracture and basically there were twice as many hip fractures in people who had voluntary weight loss as in the involuntary. So it was exactly the opposite way around in the one piece of data that you saw. Dr. Labadarios: So the weight loss that you were talking about was a diseaseassociated weight loss?
175
Pathophysiology of Weight Loss in Older Persons Dr. Morley: Not necessarily and I think it is very important. Now again there are always little bits of disease. These studies are incredibly difficult to do but there is no evidence that truly voluntary weight loss has any advantage at all in older people. If you look at the data Dr. Thomas showed on the first day, older people lose weight. What happens is they can never catch up again, they have a real problem as opposed to younger people who adjust go back to a normal appropriate level for their body, older people don’t seem to be able to do that. Dr. Labadarios: My question still remains. Is there a different mechanism for weight loss in the elderly as opposed to the young adult? Do we know anything about that? The evidence that we are beginning to have now is that weight loss is associated with certain beneficial effects, not necessarily only for heart disease but even at the cytokine level. Now what is the relationship of that to the elderly? If they are obese and lose weight are they still at an increased risk of mortality? Dr. Morley: The answer would appear to be yes, but it is based predominantly on the single hip fracture study [3], which is the best study available looking at voluntary weight loss. You would think with voluntary weight loss in an obese woman that basically they would be less likely to have hip fractures and would do better. Now the problem with that study is that one of the ways to avoid hip fracture is to keep fat on the hips. We now give hip fats to people who are malnourished to try to protect them from fracturing their hips. So I could argue that that is not the answer. The problem with all the weight loss studies that I have seen is that they are all short-term, meaning 1 or 2 years, and it is a little bit like that wonderful drug that lowers cholesterol, increases high-density lipoprotein cholesterol, causes you to have nice wide arteries, but unfortunately increases heart disease. These intermediate end points are frightening things that we tend to use. We use end points that we think are important and then turn out not necessarily to be important, and we must be very careful about that. So I am not aware of long-term starvation studies in a big enough population or weight reduction studies without exercise that have shown anything. What I can tell you is that in our New Mexico population, the people who have done that, almost all of them have gone back to being obese sarcopenics who have by far the highest death rate. So it really worries me because the mechanism of the fat frail or obese sarcopenic is dieting and failing to keep to the diet. Now if the patient can promise me that they can diet, do exercise and keep to it, then I think that we have got a great approach and I would not disagree with that one, but unfortunately human beings are very frail. Dr. DeLegge: I need you to help me to understand something. I am going to pick a specific disease state in the nursing home, which is dementia. There are many people who feel that weight loss is a normal part of the normal process of patients with dementia for whatever reason, whether they forgot how to eat or they can’t swallow or it may be medications, whatever it is, realizing there are many different stages of dementia. Can you give me some advice regarding weight loss in that population on outcome? Dr. Morley: It is not as easy as you would like. First of all there are many different demented patients and a large number of the patients in nursing homes who are demented turn out to be depressed and have a number of treatable diseases because people in nursing homes often have many diseases. So the first thing when you see a demented patient is not to assume that the weight loss is fundamentally because they have dementia. I would say 20–30% of the demented patients I see for the first time with weight loss turn out to have another cause. So that is your starting point, look for all the other causes. We have had a number of people with Helicobacter pylori infection in this group who basically, once H. pylori is treated, don’t get bad indigestion anymore, and they eat. We have demented patients with severe pain that nobody is recognizing and treating, that is why they are not eating. So the first thing is to treat
176
Pathophysiology of Weight Loss in Older Persons the treatable causes. You are now left with a natural history of dementia which is a hyperphagia often early on in the disease followed by weight loss which eventually becomes lethal, and it is most probably one of the reasons why many of these people die. Now against that you put extraordinarily high quality care. Perhaps the best studies are from a northern Italian nursing home, where everybody was fed, no tubes whatsoever, and they showed that they could actually reverse weight loss and appeared to increase longevity, although the data were not as good. Now increasing longevity in demented patients may or may not be a positive outcome, but if the family has decided they want the person alive we will assume it is a positive outcome. So a lot of the weight loss in demented patients who forget to eat is because nobody feeds them. So much of the answer is that with good feeding weight is not as dramatic. But you do reach a stage eventually in dementia where people lose weight; they lose it appropriately and they are going to die. I don’t think it is appropriate at that stage to put a tube in if that is the question; I think it is the worst time. Where I use tube feeding in my demented patients, where I am given the choice as opposed to the family, is in the person who gets sick, has a cytokine excess episode, is in hospital with a pneumonia or something like that, and is doing poorly. In those people if you put a tube in for 4–6 months and feed them they will often do well. They are also the people who do very well with megestrol acetate. So it is looking at that group and saying what is appropriate for each person. But dementia is a huge group of people and I am very hesitant to say there is one approach for everybody. Unfortunately the patients also have adult children, and the pain of my life is adult children. If you look after old people, the problem is that the adult children have totally different ideas about what their parents might have wanted. I don’t know if that helps but it is where it is. Dr. Schwab: I learned quite a few things about appetite regulation in terms of quantity, but I am not quite sure about the quality. I have two questions concerning this. The first is, is there a difference? I guess there must be because my grandparents behave differently according to the selection and preference of food than for instance my children, my wife or myself. The second one is how is this actually regulated? Why is this lady singing because of the ice cream and not because of the roast beef? Dr. Morley: These are very good questions. For the first one I would refer you to the Seneca data. If you go through Europe looking at different countries there are huge differences in caloric intake that appear to be voluntary, there are differences in the kinds of food, and the amount of macro- and micronutrients, the amount of calcium in the diet is absolutely dramatically different, and it is not always what you would expect. That is part of what I see around and I don’t even understand how some of these differences came about. So each individual chooses based on environmental pressure and on a learned pressure to eat something that they think is appropriate for them. There are obviously all manner of decisions on what you are going to eat; what is in the store makes a big difference; how easy it is to prepare, but beyond that there are also studies in animals, and virtually none in humans, that show that food choice is actually controlled by the brain. So for instance a caribou living on the side of Lake Superior will eat two thirds of its food from right on the lake to get enough salt, and then it is going to eat the other third away from the lake because now it has enough salt. So that is a salt-siting behavior, the salt is driving the mix of the two different kinds of food that the caribou eats, so we know there are driving forces. In humans we have virtually no data to tell us what drives that. We do know that there are big differences in what people eat, which is why we aim for a reasonable intake to cover all the micronutrients, hoping that we are covering everything. I don’t think we have an idea of what the ideal diet is for any given individual. Does that answer your question? I know it is unsatisfying.
177
Pathophysiology of Weight Loss in Older Persons References 1 Banks WA, Coon AB, Robinson SM, et al: Triglycerides induce leptin resistance at the bloodbrain barrier. Diabetes 2004;53:1253–1260. 2 Sauer J, Tabet N, Howard R: Alpha lipoic acid for dementia. Cochrane Database Syst Rev 2004;1:CD004244. 3 Ensrud KE, Ewing SK, Stone KL, et al: Intentional and unintentional weight loss increase bone loss and hip fracture risk in older women. J Am Geriatr Soc 2003;51:1740–1747.
178
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 179–188, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
Interaction between Nutrition, Intestinal Flora and the Gastrointestinal Immune System H. Lochs Medizinische Klinik mit Schwerpunkt Gastroenterologie, Hepatologie und Endokrinologie, Charité, Berlin, Germany
Introduction The intestinal mucosa is the biggest surface of the body, which is constantly in close contact with a high number of different bacteria and food antigens. Furthermore it has to absorb nutrients and in this process to differentiate between those molecules which have to pass the mucosal barrier and be taken up as nutrition and those molecules and organisms which have to be kept out to maintain the sterile condition in the organism. This is a complex function which is regulated by different layers of the intestinal barrier as well as specific transport systems.
Structure of the Intestinal Barrier The intestinal barrier consists of the mucous layer, the epithelial cells and the intestinal immune system. The mucous layer represents a physical barrier against bacteria as well as an immunological barrier due to its high content of secretory IgA. Specific mucins secreted into the mucous layer inhibit the adherence of bacteria to the epithelial cells [1]. The mucosal epithelium is also a physical barrier against invasion of bacteria. The epithelial cells are connected by specific structures the tight junctions regulating the intercellular influx of molecules. Located between the epithelial cells are the so-called M cells, specialized cells sampling luminal antigens like intestinal bacteria and presenting them to the intestinal macrophages in the submucosal department. Furthermore, dendritic cells enter in the tight junctions between epithelial cells, also sampling antigens and bacteria from the lumen. In rodents 179
Nutrition, Intestinal Flora and the Gastrointestinal Immune System translocation of bacteria through the epithelial cells of the intestine has frequently been described, however apart from intestinal infections this appears to be a rare phenomenon in man. The intestinal immune system is the third part of the barrier, which can react to invasion of bacteria with inflammation or tolerance. In the healthy organism the intestinal immune system is tolerant against commensal luminal bacteria and food antigens. Furthermore the barrier function of the mucous and epithelial cells is sufficient, so that the intestinal immune system is not activated, as has been shown by measurement of cytokine production by mucosa-associated immune cells [2]. However there are different conditions in which the barrier function is disturbed, and a temporary or chronic activation of the intestinal immune system is observed [3]. During gastrointestinal infections, in malnutrition, critical illness, and in inflammatory bowel disease, the intestinal barrier appears to be leaky allowing bacteria to adhere to the mucosa or even invade the mucosa. Consequently the intestinal immune system reacts with inflammation. In self-limiting infections this inflammation is quickly downregulated by the release of anti-inflammatory cytokines [4]. However, in inflammatory bowel disease the downregulation of the inflammation is inadequate and a continuous inflammatory activation of the immune system results. This could be due to a genetically mediated disturbance of the intestinal barrier function allowing adherence or even translocation of bacteria without infection with pathogenetic strains [5]. A similar situation might also develop during critical illness and even during malnutrition-associated damages to the intestinal barrier. In vitro studies demonstrated different effects of different bacterial strains. The functional assessment of the intestinal barrier is difficult due to its complexity. However so-called permeability tests using different sugars or polyethylene glycol (PEG) have been developed, which reflect the integrity of barrier function [6]. These tests seem to reflect clinically relevant changes in barrier function, however whether they allow conclusions on bacterial translocation is still an open question.
Mucosal Flora in Health and Disease Although the concentration of bacteria in the human colon is about 1012/ml, it appears that the colonic mucosa in healthy persons is sterile due to the excellent barrier function of the mucous layer. In endoscopic biopsies from the human colon of control persons none or only very few bacteria have been found [5]. In contrast in gastrointestinal infections and in inflammatory bowel disease, a high number of bacteria is adherent to the mucosa. Thereby inflammatory bowel disease might be a model for a disturbance of the intestinal barrier and might offer the opportunity to study immune reactions to 180
Nutrition, Intestinal Flora and the Gastrointestinal Immune System chronic barrier leakiness. It appears, however, that the concentration and the composition of the mucosal bacteria in inflammatory bowel disease is different than in all other conditions [5]. Since the early studies on bacterial translocation and the development of the gut sepsis hypothesis, the question was raised whether malnutrition might lead to similar defects in the intestinal barrier and thereby also allow bacteria to adhere or invade the mucosa and lead to an inflammatory reaction or to an infection. Although bacterial translocation appears to be an infrequent event in man, it has been shown by permeability tests that intestinal permeability increases with the degree of malnutrition [7]. This increased permeability is also accompanied by HLA-DR expression, indicating that immune activation is taking place [7]. It is noteworthy that such findings have only been reported in patients and not in healthy malnourished people. In contrast, several articles have shown a reduction in intestinal permeability in underweight healthy persons such as anorectic women [8]. It might be concluded that malnutrition and a second insult such as disease stress is necessary to disturb the intestinal barrier and allow bacteria to adhere. This hypothesis is supported by data from animal experiments showing a deterioration in the mucous layer, followed by adhesion of bacteria to the intestinal epithelium of rats exposed to psychic stress [9].
Effect of Substrates and Bacteria on Barrier Function and Intestinal Flora These changes seem to be influenced independently by the nutritional situation and the stress factor. Alscher et al. [10] demonstrated that injection of endotoxin was followed by an increase in intestinal permeability, compared to sham injection, in fed animals. In fasted animals, however, intestinal permeability was also increased after sham injection. By the combination of both endotoxin injection and fasting, the permeability increased significantly more than in fed endotoxin-injected animals. This experiment showed that both starvation and stress (endotoxin injection) independently disturb the intestinal barrier. The mechanism of this barrier disruption was studied by Spitz et al. [11]. They demonstrated a significant reduction in the secretory IgA concentration in the mucous layer by injection of dexamethasone. This effect was further increased when animals were starved. They also showed that bacterial adherence correlated very well with the IgA concentration in the mucous layer. It appears therefore that both malnutrition and disease stress impair the intestinal barrier; however, stress and malnutrition have a potentiating effect. From these data it was concluded that nutrition in severely ill, and also in other malnourished patients, is essential to maintain the intestinal barrier and avoid gut-derived sepsis or complications. This hypothesis is supported 181
Nutrition, Intestinal Flora and the Gastrointestinal Immune System by a number of studies showing an improved barrier function and reduction of complications in patients with enteral nutrition as compared to starved patients. However, the question has to be raised, which route and composition of nutrition might be optimal to maintain the intestinal barrier and at the same time induce a positive reaction of the gastrointestinal immune system rather than continuous inflammation. Luminal nutrition clearly appears to be better in maintaining the integrity of the intestinal mucosa than parenteral nutrition. Luminal nutrition leads to proliferation of the intestinal epithelium as has been shown in a number of studies. However, it is not only the mucosal integrity which is influenced by nutrition but also the composition of luminal flora. Schneider et al. [12] demonstrated in healthy controls and patients that both total enteral nutrition and total parenteral nutrition were followed by changes in the composition of the fecal flora. While the concentration of anaerobes was reduced by both total enteral nutrition and total parenteral nutrition, aerobes were only reduced by parenteral nutrition [12]. Such changes in the barrier and the composition of the intestinal flora appear also to be followed by immune reactions. Ikeda et al. [13] showed that the expression of MADCAM-1, an adhesion molecule in Peyer’s patches, was significantly reduced by parenteral nutrition as well as by elemental diets and intragastric application of parenteral nutrition. In addition to the substrate effects which were considered to be mediated by gut hormones, probiotics seem to have a profound effect on the intestinal barrier as well as the intestinal immune system. In the coculture model in which immune cells are cultured under a layer of epithelial cells, it has been shown that the addition of bacteria to the epithelial cell compartment does induce secretion of different cytokines by the immune cells, which do not have direct contact with the bacteria [14]. This raised the question whether changes in the luminal flora of the intestine might also have effects on the intestinal immune system. Actually Roller et al. [15] demonstrated that IgA secretion into the mucous layer as well as systemic IL10 secretion were increased by feeding pro- and prebiotics as well as symbiotics to animals. In contrast, the oxidative burst in intestinal immune cells was reduced. Some recent studies demonstrated that several probiotics have profound effects on the intestinal immune system when fed to IL-10 knockout animals [16]. These animals have an increased tumor necrosis factor (TNF␣) secretion in basal conditions as well as after lipopolysaccharide (LPS) stimulation. By prefeeding them with probiotics, the TNF␣ production was reduced in both the basal state as well as after LPS stimulation. This effect apparently cannot only be induced by whole bacteria but also by bacterial DNA. Rachmilewitz et al. [17] showed that feeding of a 22-basepair-long bacterial DNA to animals with experimental colitis could significantly reduce disease activity, weight loss and inflammatory changes. Similar effects have been demonstrated by Jijon et al. [18] in HT29 cell cultures. They were able to demonstrate that 182
Nutrition, Intestinal Flora and the Gastrointestinal Immune System bacterial DNA inhibited proteasome activity, interferon-␥ secretion and IL-8 secretion in those cells after infection with Salmonella typhimurium DNA [18].
Clinical Consequences Of course, the question has to be raised as to whether the above-mentioned data are relevant in clinical situations for patients. There are few data on the effect of enteral nutrition on the intestinal flora or bacterial translocation in man. However Sedman et al. [19] showed no difference in the frequency of bacterial translocation in patients whether they were fed enterally or not. The clinical effects of enteral nutrition on outcome are limited and furthermore cannot clearly be attributed to an effect on the intestinal barrier or flora. A number of publications, however, have recently shown that addition of bacteria or bacterial DNA to nutrition has a profound impact on the clinical course, infections and diarrhea in patients. Treatment with Escherichia coli Nissle maintains remission in ulcerative colitis as well as 5ASA medication [20]. The addition of Saccharomyces boulardii to enteral nutrition of critically ill patients significantly reduced the incidence of diarrhea [21]. Feeding of lactobacillus reduced the number of infections after liver transplantation as compared to treatment with antibiotics as well as enteral nutrition without probiotics [22, 23]. This appears to be an especially interesting study since it randomly compared patients who were fed standard enteral nutrition with patients who received the lactobacillus, indicating that the effect of the probiotics outweighs the effect of substrates. Also in acute pancreatitis the addition of lactobacillus to feeding significantly reduced the numbers of septic complications and positive aspiration cultures [24]. Therefore, the question that must be raised is whether enteral nutrition should be supplemented by probiotics and/or bacterial DNA in patients on long-term nutrition who run a higher risk for infection. However, although the positive effect of some probiotics has been demonstrated in several disease states, it is not clear which probiotic or which bacterial DNA would be most suitable as a general supplement to enteral nutrition. In vitro experiments as well as clinical studies have shown that not all probiotics act similarly in all situations. Care has therefore to be taken not to consider all probiotics equally. While positive effects have been demonstrated in different conditions with E. coli Nissle and VSL3 and some lactobacilli in some situations as described above, one study showed a negative effect with Lactobacillus GG in postoperative Crohn’s disease. The patients receiving Lactobacillus GG had more relapses than the control patients [25]. Before recommending probiotics, careful studies should therefore be performed. However, due to the impressive effects in some studies, it appears to be quite important to further investigate the interaction between bacteria, the 183
Nutrition, Intestinal Flora and the Gastrointestinal Immune System intestinal barrier and intestinal immune system to improve the composition of enteral nutrition. In summary a strong effect of both substrates as well as bacteria and bacterial DNA on the intestinal barrier and immune system has been demonstrated, and this appears to be clinically relevant. Although we do not yet know all the conditions in which such an effect might be crucial for the outcome of the patient, studies are essential to investigate which composition of enteral nutrition is advisable in which situation. References 1 Mack DR, Michail S, Wei S, et al: Probiotics inhibit enteropathogenic E. coli adherence in vitro by inducing intestinal mucin gene expression. Am J Physiol 1999;276:G941–G950. 2 Nikolaus S, Bauditz J, Gionchetti P, et al: Increased secretion of pro-inflammatory cytokines by circulating polymorphonuclear neutrophils and regulation by interleukin 10 during intestinal inflammation. Gut 1998;42:470–476. 3 Hollander D: Intestinal permeability, leaky gut, and intestinal disorders. Curr Gastroenterol Rep 1999;1:410–416. 4 Shanahan F: Crohn’s disease. Lancet 2002;359:62–69. 5 Swidsinski A, Ladhoff A, Pernthaler A, et al: Mucosal flora in inflammatory bowel disease. Gastroenterology 2002;122:44–54. 6 Wyatt J, Vogelsang H, Hübl W, et al: Intestinal permeability and the prediction of relapse in Crohn’s disease. Lancet 1993;341:1437–1439. 7 Welsh FK, Farmery SM, Maclennan K, et al: Gut barrier function in malnourished patients. Gut 1998;42:396–401. 8 Monteleone P, Carratu R, Carteni M, et al: Intestinal permeability is decreased in anorexia nervosa. Mol Psychiatry 2004;9:76–80. 9 Soderholm JD, Yang PC, Ceponis P, et al: Chronic stress induces mast cell-dependent bacterial adherence and initiates mucosal inflammation in rat intestine. Gastroenterology 2002;123: 1099–1108. 10 Alscher KT, Phang PT, McDonald TE, Walley KR: Enteral feeding decreases gut apoptosis, permeability, and lung inflammation during murine endotoxemia. Am J Physiol Gastrointest Liver Physiol 2001;281:G569–G576. 11 Spitz JC, Ghandi S, Taveras M, et al: Characteristics of the intestinal epithelial barrier during dietary manipulation and glucocorticoid stress. Crit Care Med 1996;24:635–641. 12 Schneider SM, Le Gali P, Girard-Pipau F, et al: Total artificial nutrition is associated with major changes in the fecal flora. Eur J Nutr 2000;39:248–255. 13 Ikeda S, Kudsk KA, Fukatsu K, et al: Enteral feeding preserves mucosal immunity despite in vivo MAdCAM-1 blockade of lymphocyte homing. Ann Surg 2003;237:677–685. 14 Haller D, Bode C, Hammes WP, et al: Non-pathogenic bacteria elicit a differential cytokine response by intestinal epithelial cell/leucocyte co-cultures. Gut 2000;47:79–87. 15 Roller M, Rechkemmer G, Watzl B: Prebiotic inulin enriched with oligofructose in combination with the probiotics Lactobacillus rhamnosus and Bifidobacterium lactis modulates intestinal immune functions in rats. J Nutr 2004;134:153–156. 16 Madsen K, Cornish A, Soper P, et al: Probiotic bacteria enhance murine and human intestinal epithelial barrier function. Gastroenterology 2001;121:580–591. 17 Rachmilewitz D, Karmeli F, Takabayashi K, et al: Immunostimulatory DNA ameliorates experimental and spontaneous murine colitis. Gastroenterology 2002;122:1428–1441. 18 Jijon H, Backer J, Diaz H, et al: DNA from probiotic bacteria modulates murine and human epithelial and immune function. Gastroenterology 2004;126:1358–1373. 19 Sedman PC, Macfie J, Sagar P, et al: The prevalence of gut translocation in humans. Gastroenterology 1994;107:643–649. 20 Rembacken BJ, Snelling AM, Hawkey PM, et al: Non-pathogenic Escherichia coli versus mesalazine for the treatment of ulcerative colitis: A randomized trial. Lancet 1999;354:635–639.
184
Nutrition, Intestinal Flora and the Gastrointestinal Immune System 21 Bleichner G, Blehaut H, Mentec H, Moyse D: Saccharomyces boulardii prevents diarrhea in critically ill tube-fed patients. A multicenter, randomized, double-blind placebo-controlled trial. Intensive Care Med 1997;23:517–523. 22 Rayes N, Seehofer D, Hansen S, et al: Early enteral supply of lactobacillus and fiber versus selective bowel decontamination: A controlled trial in liver transplant recipients. Transplantation 2002;74:123–127. 23 Rayes N, Hansen S, Seehofer D, et al: Early enteral supply of fiber and lactobacilli versus conventional nutrition: A controlled trial in patients with major abdominal surgery. Nutrition 2002;18:609–615. 24 Olah A, Belagyi T, Issekutz A, et al: Randomized clinical trial of specific lactobacillus and fibre supplement to early enteral nutrition in patients with acute pancreatitis. Br J Surg 2002;89: 1103–1107. 25 Prantera C, Scribano ML, Falasco G, et al: Ineffectiveness of probiotics in preventing recurrence after curative resection for Crohn’s disease: A randomized controlled trial with Lactobacillus GG. Gut 2002;51:405–409.
Discussion Dr. Cynober: I was interested by the provocative data you showed us about the decrease in the lactulose mannitol test in patients suffering from anorexia nervosa, and a lot of data you show are based on this test. This test is based on the variations in lactulose and mannitol, which lead to variations in the ratio. I would like to know what your feeling is about this test and the interpretation and the degree of confidence we can achieve with this test? Dr. Lochs: The major questions are, if permeability tests are reproducible and if they reflect clinically relevant changes of the intestinal barrier. The first question has been investigated and showed that the lactulose mannitol test is nicely reproducible in the same person. Regarding the second question, some studies e.g. the study by Welch et al. [1] showed a good correlation of intestinal immune reactions to the permeability test. Welch showed that HLA-DR expression and other inflammatory signs correlated with increased lactulose permeability index. Similar correlations have been shown with other permeability tests like the polyethylene glycol test. It has not been demonstrated that permeability tests correlate with bacterial translocation in humans, however this seems to be a rare event. Furthermore it is not clear if the translocation of whole bacteria is necessary to induce the complications attributed to increased intestinal permeability or if adherence of bacteria to the intestinal mucosa with initiation of a proinflammatory immune reaction and translocation of bacterial toxins like lipopolysaccharide are enough to induce infectious complications. In summary right now the permeability tests are the best parameter we have to investigate disturbances of the intestinal mucosa. Dr. Schiffrin: I would like to ask you whether you see some differences in the studies that you have shown and those of Jijon et al. [2] with bacterial DNA from VSL3 on anti-inflammatory activities, and the study that has been reported by Rachmilewitz et al. [3] involving CpG motifs-Toll-like receptor-9 reactions and also anti-inflammation? Do you need to be very specific in the bacterial DNA that is interacting with the pattern recognition receptors to start this anti-inflammatory activity? Dr. Lochs: I did not want to go into that because I thought, you would deal with it. The studies up to now used very different bacterial DNA or even parts of it like the study of Rachmilewith et al. [3] who used a 22 base-pair CpG. There is evidence that certain motifs in the bacterial DNA seem to be responsible for the reaction with the Toll-like receptor-9. These motifs seem to be present in the DNA of different bacteria. This probably also explains the data on the reduction of atopia in children: if you need the mothers during pregnancy with probiotics their children get these motifs as well,
185
Nutrition, Intestinal Flora and the Gastrointestinal Immune System and this creates a different immune reaction. However, there might be much more DNA motifs which induce different immune reactions we have not yet investigated. Dr. Van Gossum: You have shown that malnutrition may alter intestinal permeability, but is it really due to malnutrition itself or to the fact of not being correctly fed? That is the first question. The second question is about the role of intestinal content because there are many reports showing that, for example in Crohn’s disease, with a stoma. If one takes biopsies from a segment distal from the stoma there is a decreased immune reaction. So what is the role of intestinal fluid, and not only the feeding itself? The third question, you didn’t talk about any specific nutrients, such as the role of glutamine, for improving the intestinal permeability. Are there some data about that, could you speculate on that? Dr. Lochs: I absolutely agree with Dr. Labadarios that malnutrition by itself is not enough to damage the intestinal barrier; there needs to be something else and apparently this is some inflammatory stimulus, and bacteria are essential for this inflammatory stimulus. This has been shown in patients, as you mentioned, if a stoma is created then these inflammatory changes are not found distally from the stoma. In animal experiments, if the animals are bred sterile they do not develop inflammation, but as soon as some bacteria are added to the environment then inflammation is created. Of course there are a number of substrates influencing the intestinal barries but this will be covered in the next session. There are experiments on glutamine, on glycine, on other substrates influencing the barrier, and I think this is one of the major things we can do with nutrition, but I do not want to take this into my talk because I am sure that this will be coming. Dr. Labadarios: I am interested in the study on the stressed rats you referred to and the apparent adhesion of microorganisms on the mucosa. You know there are stressed humans as well. The recent literature [4], actually indicates that stress is a significant component of the irritable bowel syndrome (IBS). Do you have any thoughts on this? Dr. Lochs: It was in exactly this context that Soderholm et al. [5] conducted their study. They were looking for a model of IBS and found that these rats, if some stress was produced, developed neurological changes in the intestine similar to IBS, and then they found that bacteria adhered. Their hypothesis is that this happens also in men. Now we looked at the mucosa of IBS patients, and in fact a good number of IBS patients have bacteria on the mucosa similar to inflammatory bowel disease patients, so obviously in a good number of patients there is a breakdown of this barrier. If you look at other parameters what do you find histologically? Most of these patients have microscopic inflammation, so this fits very well. Hospitalization is usually a stress to the patient, and this might be some reason why the patients, even without malnutrition, do develop problems with their intestinal barrier when they get into hospital. Dr. Thomas: Can you comment on the current thinking about the intestinal barrier and increased infection complications even in acutely ill patients? That seems to have gone back and forth in the literature over the last several years about the concept of early feeding, even hypocaloric feeding, decreasing septic complications. Dr. Lochs: I think it is well established that feeding is really decreasing. The question which is raised now is how does the food have to be composed and do we need to add either probiotic bacteria or at least CpGs of probiotic bacteria? There are very nice studies comparing standard enteral nutrition versus probiotics on infections after liver transplantation, and during acute pancreatitis the probiotics do better than standard enteral feeding. So it seems that it is not just the food in the lumen, it has to be composed in a certain way to create certain immune responses. Dr. Elia: Do you have any information about the effect of starvation and nutrition on the barrier function of other epithelial surfaces such as the respiratory epithelium or genitourinary epithelium? Are there common overacting concepts?
186
Nutrition, Intestinal Flora and the Gastrointestinal Immune System Dr. Lochs: I don’t know but that sounds like a very interesting question, if that is regulated in a similar way, I have no information about that. Dr. Lesourd: Regarding the data you presented about the numbers of bacteria that link to the mucosa with a different regimen, it seems logical to me that when you re-feed just with food you probably have less adherence of the bacteria. But why is there so much difference between the two tube feedings, with percutaneous endoscopic gastrostomy and nasal tube feeding? Dr. Lochs: This study was not an intestinal bacterial adherence. It was just in the oropharynx that pathogenic bacteria were found, not commensal but pathogenic bacteria. Apparently if the patient eats you find much less pathogenic bacteria in the oropharynx as compared to tube feeding. The interesting thing to me was that the tube which goes through the naso-oropharynx into the stomach causes the highest number of pathogenic bacteria being there, not adhering. The surface was not looked at, a swab was taken to look at the composition. Dr. Endres: I enjoyed your talk very much; especially when you spoke about whether probiotic bacteria have to be alive. We know that from strain to strain it is different. It has been demonstrated with various bacteria that they can have some effects upon the immune system even after heat inactivation, whatever the effective agent may have been, e.g. cell membranes, DNA, etc. Thus not to find living probiotic bacteria in the feces of patients having received probiotics does not necessarily mean that there has not been a functional effect. What counts is the clinical effect which has to be proven. Would you agree that probiotic bacteria having to be alive is questionable nowadays? Dr. Lochs: I agree, I think it is questionable. However, there seem to be different effects that can be created with live bacteria and with DNA. For example live bacteria change the composition of the mucins in the mucin layer. If some lactobacilli are added, for example mucin-5, which has the ability to avoid adherence is increased in comparison to other mucins. So it seems that the DNA or the CpGs are important for some immune reactions but live bacteria seem to interact with the mucosa and create other effects. We know very little about that but we do know that there are some effects dependent on live bacteria. Dr. Arnaud-Battandier: I would like to know about food antigen when a patient in the intensive care unit or the ward has a leaky gut, what about gluten? Do you think gluten can provoke an immune reaction and then have a definitive consequence? Dr. Lochs: There are two nice reviews about this question. The problem with gluten is that HLA DR2 is needed because gliadin has a form that fits in the HLA DR2 to be ideally presented. Therefore people who do not have HLA DR2 do not respond to gluten as HLA DR2-positive people do. But if you reduce the question and ask how the development of celiac disease is explained, then the hypothesis is that by some infection you reduce the barrier at some stage and then gluten enters and if the person is HLA DR2 positive the immune reaction starts. Does that answer your question? Dr. Arnaud-Battandier: Yes partly. So this doesn’t happen in the majority of the patients, only in the DR2-positive patients? Dr. Lochs: Only in the genetically predisposed patients, that is what we know right now. Dr. Van Gossum: You mentioned malnutrition combined with any kind of stress for alteration of the intestinal barrier, but what about the role of drugs such as NSAIDs that elderly patients are taking a lot? We know that they could also alter the intestinal barrier. Dr. Lochs: You are absolutely right. Some drugs constitute a classical stress to the intestinal barrier. This has been shown for NSAIDs but also for other medications. There are very nice experiments showing that, for example, antibiotics damage the barrier really profoundly. So there is a double effect with antibiotics: on the one hand you damage the barrier and on the other hand you keep the bacteria out.
187
Nutrition, Intestinal Flora and the Gastrointestinal Immune System Dr. Bowling: The gut is full of millions and millions of bacteria, most of which are commensal. How can you tell when you just do sampling if you are looking at commensal or pathogenic bacteria? We all have Escherichia coli in our gut, but what makes E. coli become pathogenic in healthy individuals, and is that relevant here? Dr. Lochs: For research you can look at it using different probes. We use about 80 different probes for different species. The interesting thing is how the immune system knows this and it seems to be very nicely regulated. We are all tolerant to our own flora, and interestingly if the flora from a healthy person is brought into contact with the immune cells of another healthy person, the immune cells will be activated by the flora of the other healthy person but not their own flora. So there is a learning of tolerance in early childhood and whenever a foreign antigen arrives, it then activates the immune system. Of course there are bacteria which invade the cells and are immediately recognized as pathogenic. If bacteria are presented via the M cells they usually create tolerance, but if bacteria invade via the epithelial cells or via the tight junctions they create an immune response with secretion of IL1 or IL6. Dr. Morley: One of the things I have always struggled with is the intestinal barrier. Is there another group of people besides the malnourished who have a very leaky intestinal barrier e.g. diabetics who are usually very obese and often over eat? I am wondering how you put the two very different groups or disparate groups together when you come up with theories of how this works? Dr. Lochs: You are right, this is an interesting question, but diabetics are prone to infections and we have never really known why. Perhaps this disturbed intestinal barrier is contributing. Dr. Armstrong: My question probably reflects ignorance. I have always had difficulty understanding the concept of a leaky gut as a cause of problems. It is actually a pathogenic mechanism or an indicator of a problem going on. The reason is that it strikes me that the gut always has bacteria as we heard; it is always repairing itself, and it is always potentially leaky. There are numerable things that can alter gut permeability: drugs, viral infections, celiac disease, a whole lot of things that can alter permeability which do not necessarily seem to lead inexorably to the sort of manifestations and diseases that we have heard of, and it may therefore be changes in gut permeability or other processes. We may be barking at the wrong tree by trying to improve or reduce permeability in some of the conditions that we heard about. I would be interested in your comments. Dr. Lochs: I agree there is not enough evidence to show a causal relationship. However, it goes very nicely in parallel; so the leakiness of the gut goes in parallel with the increased complications and the tightening of the gut goes in parallel with a reduction in complications and infections. So it is the best parameter we have right now to look at, but I absolutely agree, it is not clear if that is the first step or if it is only one step in a series.
References 1 Welsh FK, Farmery SM, MacLennan K, et al: Gut barrier function in malnourished patients. Gut 1998;42:396–401. 2 Jijon H, Backer J, Diaz H, et al: DNA from probiotic bacteria modulates murine and human epithelial and immune function. Gastroenterology 2004;126:1358–1373. 3 Rachmilewitz D, Katakura K, Karmeli F, et al: Toll-like receptor 9 signaling mediates the antiinflammatory effects of probiotics in murine experimental colitis. Gastroenterology 2004;126: 520–528. 4 Maxwell PR, Mendoll MA, Kumar D: Irritable bowel syndrome. Lancet 1997;350:1691–1695. 5 Soderholm JD, Yang PC, Ceponis P, et al: Chronic stress induces mast cell-dependent bacterial adherence and initiates mucosal inflammation in rat intestine. Gastroenterology 2002;123: 1099–1108.
188
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 189–202, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
Psychoimmunology of Nutrition Bruno Lesourd Faculté Médecine Clermont-Ferrand, Paris, France
The relationships between behavior and immune responses have been studied for a long time. It is well recognized that persons under psychological stresses, such as depression, marital problems, bereavement or alcoholism, are more susceptible to infections. Nevertheless only a few studies were conducted before the 1970s. Since then, progress in knowledge of the immune system, improvement in questionnaires assessing the characterization and quantification of different moods, and very recently the link between a psychological approach and brain neuron cell functions (neuroimmunology) have pushed to extend research in this new field, called psychoneuroimmunology. This review will briefly describe the immune system and its roles. Thereafter the most important findings linking psychological stresses and immune responses will be discussed. Recent research at the neuron cell level is reported in order to present the current hypotheses being investigated.
Immune Responses The immune system is an important body part (3 kg of lymphocytes, 5% of body weight) and is widely spread throughout the body. This large distribution of immune cells enables the development of an immune response at any body site where aggression occurs [1–3]. When activated, the immune system releases numerous mediators (cytokines, immunoglobulins, prostaglandins, etc.) that inform the entire body of the aggression and enabling a large response that mobilizes all the body reserves, including the immune cells and also nutritional body reserves. Such a global response is directly, or indirectly through hormonal response, activated by cytokines inducing inflammatory processes. Two types of immune responses may reflect immune activation. The more primitive is called nonspecific immunity. This includes responses from monocytes-macrophages, polymorphonuclear cells and a lymphocyte subset: natural killer (NK) cells. This immune response induces destruction of 189
Psychoimmunology of Nutrition
Stress
⫺
Food intakes
Hypothalamus ⫹ Adrenals ⫹
IL-1, TNF Macrophages ⫹
Glucocorticoids ⫹
Amino acids
Muscle catabolism
Fig. 1. Regulation of macrophage functions during stress. Activated macrophages exert a retroactive feedback on hypothalamic functions and therefore on food intakes.
contaminating agents, whether intra- or extracellular, through phagocytosis and killing, in a safe manner, mainly by free radical activity within intracellular vacuoles. The specific immune system may help by circulating mediators, such as immunoglobulins or complement factors, for phagocytosis. This immune system is mainly non-antigen-specific and works without any antigen recognition. Part of this primitive system, namely monocytes-macrophages, also plays an important role in more sophisticated immune responses: (1) through partial intracellular digestion of the aggressive agents enabling the presentation of antigen molecules to lymphocytes, this antigen presentation is able to induce specific antigen responses, and (2) by inducing directly (through proinflammatory cytokines such as IL-1, IL-6 or TNF) [4] or indirectly (through hormonal responses; fig. 1) a global metabolic response called hypercatabolism that mobilizes nutritional body reserves as well as brain functions to face the aggression. Very little is still known about the importance of brain activation during such responses. The other immune system is associated with specific antigen recognition at the molecular level. Antigen specificity is related to specific lymphocyte differentiation: on its membrane a lymphocyte bears receptors specific for only one antigen. This system is dual. (1) B lymphocytes act through secretions of soluble receptors (immunoglobulins) that help to phagocytose the aggressive agents or permit direct killing through the complement cascade. Humoral immunity is mainly active against cells, in particular bacteria. (2) T lymphocytes, more recent in ontogenesis, act in a more sophisticated way. This immune response is called cell-mediated immunity (CMI). The CD4⫹ cells recognize the antigen presented by the monocytes-macrophages 190
Psychoimmunology of Nutrition
Macrophages IL-1, IL-6, IL-12
PGE2, IL-10 ⫹
⫹
TH2
TH1 ⫺
⫺
IL-4, IL-5
IL-2, IFN␥ ⫹ TL Cytotoxique
⫹
⫺
⫺
BL
Fig. 2. Activation of T helper (TH1 and TH2) by macrophages. The activation of T cells depends on macrophage functions. Acute activation of macrophages induces TH1 through proinflammatory monokine boost while chronic activation induces a stronger boost of TH2 through anti-inflammatory monokines.
and are then transformed into active cells by the macrophage cytokines. In fact the activity of the CD4⫹ cells depends on the type of macrophage activation (fig. 2). The most frequent activation leads to TH1-activated lymphocytes that, in cascade, activate cytotoxic T lymphocytes (CD8⫹). The cytotoxic T lymphocytes directly kill the infected (or transformed) cells that present an abnormal antigen at the membrane level. This system is the most potent destroying system within the body. It prevents not only proliferation of hidden (intracellular) pathogens but it also enables the elimination of abnormal cells (tumor cells) by inducing apoptosis of these cells. TH2 lymphocytes are also activated by macrophages but through different mechanisms. They activate B lymphocytes and induce the production of more sophisticated immunoglobulins, such as IgG or IgA. Activation of cytotoxic T lymphocytes and/or B lymphocytes depends on the cytokines released by TH1 and TH2 lymphocytes. The TH1/TH2 equilibrium is evaluated by quantification of the cytokines released by each lymphocyte type: IL-2, IFN␥ or IL-12 for TH1, and IL-4, IL-5 or IL-10 for TH2. This TH1/TH2 equilibrium seems to be of great importance for CMI efficiency (fig. 2). In fact TH1 lymphocytes are predominant in the first part of life during which responses to unknown antigens are driven. TH1 responses decline with age, so that TH2 responses become predominant in old age. TH1 responses decline more rapidly 191
Psychoimmunology of Nutrition if antigen boosts are frequent and at a high level (multi-infected patients) during life. CMI (functions of the T-lymphocyte) is mostly active against intracellular pathogens and transformed cells while humoral immunity mostly acts against extracellular pathogens. T lymphocytes are rapidly dividing cells. They may induce autoproliferation through IL-2 secretion than binds the IL-2 receptor to the membrane. This rapid proliferation of T lymphocytes is of major importance in the face of aggression. Indeed, there is competition between the replication speed of the aggressive agent and the proliferation of the T lymphocytes. Every disturbance, such as undernutrition, stress, behavioral modification, that slows the proliferation rate, renders the individual more susceptible to infections. This is why lymphocyte proliferation is often used as a pertinent evaluation of the immune system.
Psychological Stresses and Immune Responses In the late 1970s and early 1980s, it was shown in laboratory rodents that different stresses induce decreases in CMI, namely lymphocyte proliferation and/or natural immunity, i.e. NK cell activity. Intense auditory stimulation in mice [5], electric shocks in rats [6] and foot shocks in rats [7] induce similar decreases in CMI. It was shown that these effects might be modulated by opioid peptides [7], drawing evidence from the modulation of immune responses by neuromediators. Nevertheless stress was already known to induce multimetabolic changes within the body, i.e. changes in nutritional and hormonal status that also affect the immune responses. Such metabolic changes may be more important than behavioral modifications in modulating immune responses. The importance of conducting human studies in which psychological changes may be quantified in more sophisticated ways has arisen. Glaser et al. [8] conducted several studies in healthy students who clearly showed that psychological stress induced by examination sessions may influence immune responses. In the same students, they showed that the numbers of T lymphocytes (CD3⫹, CD4⫹ and CD8⫹) are decreased during examination periods when compared to immune responses after vacation. In addition lymphocyte proliferations were also lower [8]. Such decreases in CMI have been linked to higher depression, anxiety and hostility scores. Decreased immune responses were observed not only for CMI but also for nonspecific immunity (NK cells) [9] and for humoral immunity [10]. In fact, Epstein-Barr virus antibodies were higher during examination sessions, showing that this immune deficit may lead to reactivation of a permanent virus [9]. The relationship between decreases in IFN␥ production and in NK activity points out the importance of a decreased CMI in diminished nonspecific immune responses [9]. Similar findings were observed in medical students feeling loneliness [11], in psychiatric patients feeling alone, in separated/divorced females, in spouses 192
Psychoimmunology of Nutrition with poorer marital quality, in males with marital discord, and in caregivers of dementia patients who are at high risk of depression [12]. The same group explored similar changes over an entire academic year. They were able to show that such decreased immune responses during examination periods were repetitive over the year, and that in between the immune responses recover [13]. They also showed that cytokine production, i.e. IFN and LIF (leukocyte inhibitory factor) followed the same pattern as lymphocyte proliferation. In addition, students with lower immune responses during examination periods had a higher incidence of self-reported symptoms of infectious illness. This points out the strong relationships between examination-induced psychological stress and immune responses. Decreased immune responses related to such psychological stress have been linked to hormonal metabolic changes since urinary cortisol is increased during examination periods [14]. More recently it was shown that psychological stress has clinical consequences by inducing a slowing of wound healing [15], a process dependent on macrophage functions [16]. They also alter dynamic immune responses: immune responses to influenza vaccine are lower in stressed elderly [17], a human group in which immune responses are particularly sensitive to environmental factors [18]. This effect may be more intense in elderly subjects in whom stress induces longer hormonal responses as shown by longer periods of elevated cortisol levels [19]. It appears that stress-induced elevated serum cortisol is probably of major importance for the stress-induced decrease in immune responses. This is probably not related to the nutritional effect of stress since decreased immune responses in caregivers of dementia patients are related to a higher increase in salivary cortisol [20], a parameter not dependent on corticosteroid-binding protein and therefore on protein nutritional status. The importance of stress-induced hormonal changes on immune responses have also been observed in depressive and alcoholic patients, and both diseases exert cumulative effects on NK cell activity [21]. It is well known that alcoholism is often caused by psychological stress and is also often associated with undernutrition which is a consequence of such behavior. Many publications refer to decreased immune responses in relation to undernutrition [for references see, 22]. It was shown that, in the elderly, major undernutrition (protein-energy malnutrition) as well as moderate undernutrition, i.e. micronutrient deficits, are both associated with decreased immune responses [23]. In addition, refeeding is also associated with improved immune responses, showing the strong interrelationships between nutrition and immunity [24].
Mechanisms Stresses, including psychological stresses (see above), induce chronic changes in nutritional behavior and hormonal status, i.e. the cortisol level. 193
Psychoimmunology of Nutrition The effect of psychological stresses may be related to the chronic effect of a high hormonal level that may induce changes in cell receptor density and/or sensitivity. It was shown that in the elderly, lymphocytes have lower levels of membrane transduction markers, such as CD38⫹, or of cytotoxic markers (CD56⫹) on CD8⫹ cells [25]. This is more pronounced in elderly caregivers who are under chronic stress [25]. Furthermore, lymphocytes from persons under chronic depression (stress) become resistant to the immunosuppressive effect of steroids [26]. This was reported in depressive patients in whom chronically high cortisol levels are noticed [27]. In addition these patients show a failure to suppress endogenous cortisol levels following dexamethasone administration [28], indicating that the chronic stimulation of glucocorticoid receptors induces changes in receptor affinity. Bauer et al. [20] conducted a study in not undernourished caregivers of dementia patients. They expressed greater anxiety and depression (Savage Personality Screening Scale [29]) and greater stress scores (Global Measure of Perceived Stress Scale [30]) than non-caregivers. They also had lower lymphocyte proliferation and lower IL-2 production that was related to higher salivary cortisol levels. In in vitro mononuclear cell cultures, additive suppression of lymphocyte proliferation required higher doses of dexamethasone in caregivers than in non-caregivers, showing the corticoid resistance to glucocorticoid in persons who are under chronic stress. From such findings it appears that chronic stress is related not only to cortisone-induced decreased immune responses but also to some degree of stress resistance that pushes lymphocytes to be less reactive to further glucocorticoid challenge. Therefore, part of the mechanism of stress-induced decreases in immune responses is related to chronic changes in glucocorticoid levels. Are those changes due to brain activation? Acute gastric injection of alcohol into rats induces decreased immune responses and direct activation of the hypothalamic paraventricular nucleus (PVN): increases in corticotropinreleasing factor mRNA through the activation of early genes such as c-fos (fig. 3) [31, 32]. In contrast, chronic acute injection decreased the response of the hypothalamic-pituitary-adrenal (HPA) axis to further alcohol ingestion but not to stress of another nature, such as immune challenge or electroshocks [33]. This is associated with a lower ACTH response to vasopressin [34] and a lower level of serotonin (a neuromediator) in the PVN [34], indicating a relationship between neuromediator release and HPA axis modulation. A second alcohol challenge, few days after the first, does not lead to activation of early genes but in contrast to lower activation [35]. Such decreased responses may be overpassed by N-nitro-L-arginine methyl ester that blocks nitric oxide formation [35] but is not related to higher serotonin secretion in the hypothalamic PVN [36]. This group pointed out the importance of chronic alcohol ingestion-induced nitric oxide production in the PVN [37] and the correlation with HPA axis modulation. Nevertheless, they have been unable to show a correlation with serotonin release, at least in some 194
Psychoimmunology of Nutrition
STRESS
Hypothamalus
⫺
Pituitary Cytokines ACTH
Prolactin ⫹
Adrenals
Immune system Glucocorticoids
⫺
Fig. 3. Influence of stress on the hypothalamic-pituitary-adrenal axis. Macrophages exert a feedback control at the hypothalamic level. This feedback is under the control of glucocorticoid release by adrenal glands.
stress situations [38]. The link between neurotransmission and HPA axis modulation remains unclear even though the site of such a link is known. In summary, it was shown that alcohol activates the HPA axis response by inducing the release of corticotropin-releasing factor (and as a result glucocorticoid at the blood level), thus mimicking a stress effect. Even though the mechanism remains unclear, the site at which such an alcohol stress acts, the hypothalamic PVN, may explain many of the behavioral effects of alcohol since the PVN is a central passage for mood adaptation related to stress and the link between HPA axis modulation and the changes in nutritional intakes. Many changes observed in alcoholic patients, such as behavior, decreased food intakes, mood disorders, act in relation to this mechanism. Chronic insufficient intakes lead to micronutrient deficiency and later to protein-energy malnutrition, both being associated with decreases in immune responses [22–24]. Chronic insufficient intakes are related to behavioral changes [39] such as depression, showing the importance of psychological stress, metabolic changes and immune responses. Numerous articles refer to the importance of neurotransmission changes in eating behavior, showing the major influence of neurotransmission on immune responses [40]. This interrelationship is more important in aged subjects who are more susceptible to mood changes [40]. Recently, several experiments in rats have investigated these mechanisms, trying to understand which metabolic changes in which brain sites may be associated with changes in immune responses. In healthy albinos rats, it was shown that an increase in age from 3 to 9 months is associated with increased GABAergic hypothalamic activity and no change in immune responses, i.e. lymphocyte proliferation [41], while later on 195
Psychoimmunology of Nutrition (9–18 months) GABAergic hypothalamic activity and immune responses both decline while circulatory corticosterone increases [42]. The authors point out the strong link of both changes between the ages of 3 and 18 months [41]. This activity may be altered by changes in protein diet on a long-term (30 days) basis [41, 42]. Short-term (7 days) changes in protein diet have little effect. In contrast long-term (30 days) changes in protein diet induce alterations in immune responses, in circulatory corticosterone and in GABAergic activities. A low protein diet (5% casein) delays the age decline in immune response (lymphocyte proliferation) and in parallel decreases GABAergic activity. In contrast, a high protein diet (40% casein) induces decreases in immune responses and increases GABAergic activities [43], while circulatory corticosterone increases [42]. Those changes in GABAergic activities, mainly related to GABA receptor density [43], are only observed in the hypothalamus and not in the cerebellum or pons medulla showing that hypothalamic activity is of major importance for changes in immunity. This brain area is of major importance, in particular since it is an important site for food intake control. Such findings are even more impressive when one knows that changes in diets induce changes in hypothalamic activation as measured by early gene (Fos) activation. Horn and Friedman [44] investigated brain activation in relation to diet type, i.e. high fat/low carbohydrate (HF/LC) versus low fat/high carbohydrate (LF/HC) diet, in middle-aged Sprague-Dawley rats. Rats fed the LF/HC diet increase food intakes after administration of a fructose analog, 2,5-anhydro-D-mannitol. This treatment induces activation of the PVN of the hypothalamus, and does not have any effect in rats fed the HF/LC diet. In contrast in rats fed the HF/LC diet, the same effect is seen when they are treated with methyl palmorixate, an inhibitor of fatty acid oxidation, but the activation concerns another part of the brain: the solitary tract [44]. Those effects can be blocked by vagotomy, showing the importance of nerve transmission to the brain [45].
Conclusion The importance of hypothalamic functions in food intake and in immune response activities is emphasized by these findings. The activity of the hypothalamus is probably of importance in this regulation, most likely through still unclear modulation of HPA axis responses. The strong simultaneous activities observed in hypothalamic PVN, in HPA axis and in immune responses point to the importance of PVN activation in the regulation of stress responses. The possible neuromediators as well as the receptors involved in this regulation are presently being investigated. This brings new approaches to the understanding of the regulation of psychological stress and its interrelationship with immune responses. In the near future, this approach will probably help to modify metabolic and behavior stress consequences. 196
Psychoimmunology of Nutrition References 1 Roitt IM, Delves PJ (eds): Roitt’s Essential Immunology, ed 10. Oxford, Blackwell, 2001. 2 Scrimshaw NS, SanGiovanni JP: Synergism of nutrition, infection and immunity: An overview. Am J Clin Nutr 1997;66:464S–477S. 3 Cakman I, Rohwer J, Schtüz RM, et al: Dysregulation between TH1 and TH2 T cell subpopulations in the elderly. Mech Aging Dev 1996;87:197–209. 4 Lesourd B: Conséquences nutritionnelles des cytokines: facteur de gravité des hypercatabolismes chez le sujet âgé. Age Nutr 1992;3:100–109. 5 Monjan AA, Collector MI: Stress-induced modulation of the immune response. Science 1977;196:307–308. 6 Keller SE, Weiss JM, Schleifer SJ, et al: Suppression of immunity by stress: Effect of a graded series of stressor on lymphocyte stimulation in the rat. Science 1981;213: 1397–1400. 7 Shavit Y, Lewis JW, Terman GW, et al: Opioid peptides mediate the suppressive effect of stress on natural killer cell cytotoxicity. Science 1984;223:188–190. 8 Glaser R, Kiecolt-Glaser JK, Stout JC, et al: Stress-related impairments in cellular immunity. Psychiatr Res 1985;16:223–239. 9 Glaser R, Rice J, Speicher CE, et al: Stress depresses interferon production concomitant with a decrease in natural killer activity. Behav Neurosci 1986;100:675–678. 10 Glaser R, Kiecolt-Glaser JK, Speicher CE, Holliday JE: Stress, loneliness and changes in herpes virus latency. J Behav Med 1985;8:249–260. 11 Kiecolt-Glaser JK, Garner W, Speicher CE, et al: Psychosocial modifiers of immunocompetence in medical students. Psychosom Med 1984;46:7–14. 12 Kennedy S, Kiecolt-Glaser JK, Glaser R: Immunological consequences of acute and chronic stressors: Mediating role of interpersonal relationships. Br J Med Psychol 1988;61: 77–85. 13 Glaser R, Rice J, Sheridan J, et al: Stress-related immune suppression: Health implications. Brain Behav Immun 1987;1:7–20. 14 Kiecolt-Glaser JK, Ricker D, Messick G, et al: Urinary cortisol, cellular immunocompetency and loneliness in psychiatric inpatients. Psychosom Med 1984;46:15–23. 15 Kiecolt-Glaser JK, Marucha P, Malarkey W, et al: Slowing of wound healing by psychological stress. Lancet 1995;346:1194–1196. 16 Lesourd B: Immune response during disease and recovery in the elderly. Proc Nutr Soc 1999;58:85–98. 17 Kiecolt-Glaser JK, Glaser R, Gravenstein S, et al: Chronic stress alters immune response to influenza virus vaccine in older adults. Proc Natl Acad Sci USA 1996;93:3043–3047. 18 Mazari L, Lesourd B: Nutritional influence on immune response in healthy aged persons. Mech Ageing Dev 1998;100:17–32. 19 Sapolski RM, Krey LC McEwen BF: The neuroendocrinology of stress and aging: The glucocorticoid cascade hypothesis. Endocr Rev 1986;7:284–301. 20 Bauer ME, Vedhara K, Perks P, et al: Chronic stress in caregivers of dementia patients is associated with reduced lymphocyte sensitivity to glucocorticoids. J Neuroimmunol 200; 103:84–92. 21 Irwin M, Caldwell C, Smith TL, et al: Major depressive disorder, alcoholism, and reduced natural killer cytotoxicity. Role of severity of depressive symptoms and alcohol consumption. Arch Gen Psychiatry 1990;47:713–719. 22 Lesourd B, Raynaud-Simon A, Mazari M: Nutrition and ageing of the immune system; in Calder PC, Field CJ, Gill HS (eds): Nutrition and Immune Function. Abingdon, CABI, 2002, pp 357–374. 23 Lesourd BM, Mazari L, Ferry M: The role of nutrition in immunity in the aged. Nutr Rev 1998;56:S113–S125. 24 Lesourd BM, Mazari L: Immune responses during recovery from protein energy malnutrition. Clin Nutr 1997;16:37–46. 25 Castle S, Wiljins S, Heck E, et al: Depression in caregivers of demented patients is associated with altered immunity: Impaired proliferative capacity, increased CD8⫹, and a decline in lymphocyte with surface signal transduction molecules (CD38⫹) and a cytotoxicity marker (CD56⫹CD8⫹). Clin Exp Immunol 1995;101:487–493.
197
Psychoimmunology of Nutrition 26 Kok F, Heijnen C, Bruijn J, et al: Immunoglobulin production in vitro in major depression: A pilot study on the modulating action of endogenous cortisol. Biol Psychiatry 1995;38:217–226. 27 Wodarz N, Rupprecht R, Kornhuber J, et al: Normal lymphocyte responsiveness to lectins but impaired sensitivity to in vitro glucocorticoids in major depression. J Affect Disord 1991;22:241–248. 28 Pedersen CA, Folds JD, Evans DL: Dexamethasone effects on number of cells in lymphocyte subpopulations: Changes associated with major depression and DST nonsuppression. Prog Neuropsychopharmacol Biol Psychiatry 1989;13:895–906. 29 Savage R: Savage Personality Screening Scale (SSPS-20). Perth, Murdoch University. 30 Cohen S, Kamarak T, Mermelstein R: A global measure of perceived stress. J Health Soc Behav 1983;24:385–396. 31 Rivier C, Lee S: Acute alcohol administration stimulates the activity of hypothalamic neurons that express corticotropin-releasing factor and vasopressin. Brain Res 1996;726:1–10. 32 Lee S, Rivier C: An initial, three-day long treatment with alcohol induces a long-lasting phenomenon of selective tolerance in the activity of the rat hypothalamic-pituitary-adrenal axis. J Neurosci 1997;17:8856–8866. 33 Davis SL: Environmental modulation of the immune system via the endocrine system. Domest Anim Endocrinol 1998;15:283–289. 34 Lee S, Rivier C: Altered ACTH and corticosterone responses to interleukin-1 in male exposed to an alcohol diet: Possible role of vasopressin and testosterone. Alcohol Clin Exp Res 1995;16:539–541. 35 Woods JM, Druse MJ: Effects of chronic ethanol consumption and aging on dopamine, serotonin and metabolites. J Neurochem 1996;66:2168–2178. 36 Rivier C, Shen G: In the rat, endogenous nitric oxide modulates the response on the hypothalamic-pituitary-adrenal axis to interleukin-1, vasopressin and oxytocin. J Neurosci 1994;14:1985–1993. 37 Rivier C: Adult male rats exposed to an alcohol diet exhibit a blunted ACTH responses to immune or physical stress: Possible role of nitric oxide. Alcohol Clin Exp Res 1995;19:1474–1479. 38 Ogilvie K, Lee S, Weiss B, Rivier C: Mechanisms mediating the influence of alcohol on the hypothalamic-pituitary-adrenal axis responses to immune and nonimmune signals. Alcohol Clin Exp Res 1998;22:243S–247S. 39 Lesourd B, Raynaud-Simon A, Mathey M: Comment favoriser la prise alimentaire des sujets âgés. Nutr Clin Métabol 2001;15:177–188. 40 Harrold JA: Hypothalamic control of energy balance. Curr Drug Targets 2004;5:207–219. 41 Bandyopadhyay BC, Poddar MK: Is hypothalamic GABA involved in immune function in relation to dietary protein during aging. Neurochem Res 1997;22:1135–1144. 42 Poddar MK, Bandyopadhyay BC, Chakrabarti L: Dietary protein alters age-induced change in hypothalamic GABA and immune response. Neuroscience 2000;97:405–409. 43 Food control and hypothalamus. 44 Horn CC, Friedman MI: Metabolic inhibition increases feeding and brain Fos-like immunoreactivity as a function of diet. Am J Physiol 1998;275:5448–5459. 45 Ritter STT, Friedman MI: Induction of Fos-like immunoreactivity (Fos-li) and stimulation of feeding by 2,5-anhydro-D-mannitol (2,5-AM) require the vagus nerve. Brain Res 1994; 646:53–64.
Discussion Dr. Cynober: I really enjoyed your talk. Congratulations because it was probably a really difficult task. From the data you presented I think that the key point, the central element, is certainly cortisol. Are you aware of experiments trying to block the cortisol action, for instance using RU486? Dr. Lesourd: That was not done. The only thing that has been seen is that there is a blockade when dexamethasone is added after chronic stress, not after a
198
Psychoimmunology of Nutrition shock stress. But to block the production of cortisol for a long period, that was not done. Dr. Elia: You focused on cortisol, but of course stress is associated with all kinds of other hormones including catecholamines and so on. Do they play a role? Dr. Lesourd: I don’t know. I mentioned that there are actions on the pituitary hormones secreted during stress, but that was not looked at, and the picture is not clear for catecholamines. Dr. Morley: For the sake of our chairman I do have to point out that ‘psychoimmunology’ was the first used in the late 1950s and early 1960s when people with rheumatoid arthritis were studied. It was shown that with stress the immune function response is altered. After a few experiments the term was changed to psychoneuroimmunology. So that is I suppose the epistemology of this terminology. I have a couple of points about other hormones, both for catecholamines. Actually very interesting work has been done in rats in which the catecholamine response in the spleen was looked at and the alterations during stress were related to immune function changes within the spleen. So there is literature, it is almost all animal literature at that level. The other, relatively well-studied hormone is endorphin which has almost the opposite effects acutely to cortisol. For instance if you do exercise stress the endorphin increases and NK cells increase in proliferation together with the endorphin. If you do long-term morphine addiction studies, you find tolerance and downregulation of the cells. There has been a big argument about whether the changes are all cortisol or tolerance; both of those have to be looked at. But the stress of increased exercise can be blocked with naloxone. So there are a lot of interesting interactions; I think it is very much like feeding. There are so many different things that can cause problems. There are clearly also multiple effects of cytokines on the central nervous system, not only all the effects on feeding but also on memory. Recently it was shown that these effects can be produced both by cytokines directly crossing the blood barrier and also by stimulating the ascending fibers of the autonomic nervous system. It becomes very complex and I think when you have to add nutrition to this, and certainly nutrition alters every one of these factors, it makes it very difficult for any single human being or even all of us together to really understand how these interact. That is why I think you did an incredibly good job of giving us a good concept to pay attention to the role of the psyche along with nutrition and immunology. Dr. Lesourd: I have also looked at the catecholamine publications. It is far more complicated to really understand what is going on and it is impossible to get a really clear picture from that; it is very confusing. This is why I didn’t talk on that. Dr. Labadarios: Have you noticed how President Bush’s hair has gone white over the past 2 months? Regarding the slide that you had on the mechanism where there is actually a blockage in melanocyte-stimulating hormone, do you think psychoimmunonutrition will help President Bush? Dr. Morley: But every United States President goes gray when they are president, there is no question, no matter who they are. But I have also noticed that many South Africans are very gray, so I would guess this is not unique. Dr. Labadarios: If you are referring to our president you are quite right. Actually this is serious, people under stress get gray hair faster for reasons we don’t know, and that is a consistant observation. Now on the more serious side, at least serious until we know better, would you comment on the literature that indicates that stress is a proinflammatory state? Dr. Lesourd: When we are talking about chronic cortisol secretion, for me it is chronic stress. It is chronic deregulation of hormonal production most likely at the level of regulation of the hypothalamus, and the paraventricular nucleus is probably
199
Psychoimmunology of Nutrition the central place to look for this. That is exactly the chronic stress we are talking about. Stress was measured by cortisol production for a very long time and, even though we no longer use it, it is still a good reflection of the acute phase reaction. Dr. Labadarios: A recent article [1] actually associates stress and depression with increased C-reactive protein levels, which really adds to the comment that you are making. Dr. Thomas: Just to follow up on that. As we look at the causes of anorexia and weight loss, particularly in long-term care chronic settings, they are predominantly associated with depressive reactions such as a chronic stress state, certainly mediated by the catecholamines and serotonin, dopamine, etc., which gives us a real opportunity to improve nutritional status and improve intake by the treatment of the depression. Dr. Lesourd: I don’t have any experience with dopamine. But I do know that when chronically undernourished patients are treated, at the beginning there is always an increase in cortisol and C-reactive protein, a very low level, 20–25 mg/l, but an increase. When the patients are treated and are recovering with the food intake, then it decreases. Dr. Ockenga: I have a further comment on the link between catecholamines and catabolism as well as psycho-immunology. My former colleagues in Hanover did some elegant studies with students and they showed that increased catecholamines also increase the circulating lymphocytes by decreasing the adherence of molecules [2]. They also did studies on chronic stress and found comparable results [3]. We have just finished some studies in which we found that an increase in sympathetic activity is related to the leptin system, and 2 years ago we published a study in Gastroenterology [4] on the leptin system associated with resting energy expenditure and catabolism. There are probably several links as already mentioned here between sympathetic activity as well as all our regulation systems of food intake and energy homeostasis. Dr. Lesourd: But you know that leptin could also be activated by the sympathetic system; so it is a link. Dr. Ockenga: Sure, and at least this has been done in animals. If the sympathetic nervous system in the brain is cut, then the effect of leptin in liver metabolism is regulated, as well as resting energy expenditure and the further release of several hormones. Dr. Schiffrin: To bring the intestine back into the picture since it is such an important immune organ. Increased bacterial adhesion during stress has been reported. Is stress promoting mast cell degradation? The barrier then becomes leaky? What happens with the stress on the mucosal immune system? Dr. Lesourd: I have not seen a study that tried to link stress and the function of intestine, even for the immune response. Dr. Morley: There are a couple of studies, depending on how you want to look at it, on the pure mucosal barrier. Lower down in the gut there are no studies. But looking at the mucosal barrier within the stomach, stress releases and reduces the mucosal barrier and is one of the reasons why it produces stress; the other is alteration in blood flow. Both are most probably more important than increased gastric acid. In addition to that there are fairly good studies looking at the effect of stress on gastrointestinal motility. I think most of them show a decline in gastrointestinal motility. We see both but I think the majority suggested a slowing, but there may be others. Dr. Lesourd: And a decline in gastric emptying. Dr. Morley: Yes, certainly a decline in gastric emptying. Dr. Labadarios: In relation to the last comment: the available motility studies are actually variable and to my understanding that is why we are not trying to classify irritable bowel syndrome into its components of diarrhea or constipation. At the functional
200
Psychoimmunology of Nutrition level that is probably the most clearly known association between so-called stress and irritable bowel syndrome. Dr. Armstrong: I congratulate you on the wonderful job you did in trying to put together all the data around the effects of stress. For me the question is whether or not we should be trying to redefine or categorize stress, because what you described is a whole slew of different stresses to the body which are difficult to compare directly from the starvation studies of Soderholm et al. [5] through to food deprivation, through to alcoholism. These all would be expected to produce different sorts of responses in organisms that have adapted over millions of years to life around them. Stress responses presumably have evolved to cope with a number of different external and internal stresses, and to assume that each of them is going to be comparable and is going to work by comparable mechanisms I think does the mechanisms a disservice and makes life very difficult for us to try to differentiate what is going on under different circumstances. The question therefore is, is there a way of separating out different types of stress or categorizing so that we can get a degree of uniformity and dissect the mechanisms, or is that being over optimistic? Dr. Lesourd: What I tried to do is to present stress without any nutritional response because that is really what we would like to know. If there is something that the nutrition response is not involved in, and I know that there are a lot of other stresses where there is a very important change in nutritional behavior, so it is probably different, except for the mechanisms I tried to show you. Dr. Lochs: Just a comment on stress and the gastrointestinal barrier. There are a number of studies showing that different kinds of stress create ulcers, and this also differentiates what kind of stress. Immobilization stress was mainly used in rats. Rats were put in casts so that they couldn’t move their legs which creates an enormous stress and they develop ulcers very quickly. How ulcers develop was also studied: by a reduction in mucosal blood flow. So they have mucosal ischemia and then they get ulcers. To my knowledge, as Dr. Morley said, no studies on the intestine have been done but it is most likely that similar effects happen in the intestine: blood flow is reduced and is one of the major determinants of barrier. You showed very nicely that in young rats a low protein diet causes an increase in cortisol and a decrease in lymphocyte proliferation that could even be dependent on the increase in cortisol, but in older rats this was the opposite, the lymphocytes increased and cortisol. What causes this change? You would expect that if you treat somebody with steroids that the lymphocyte proliferation would decrease? How is that regulated since this regulation is no longer present in older rats? Dr. Lesourd: I have also shown that the lymphocyte at the peripheral level became totally insensitive to dexamethasone, so probably the same occurs at the brain level. This means that cortisol is no longer active and if it does increase it doesn’t do anything. Dr. Lochs: But why isn’t it active? Are there cortisol receptors on the nucleus, or what is it? Dr. Lesourd: There are very few data on that. It has been shown that the reappearance of cortisol receptors is very low and the number of receptors that reappear is far lower than what you have before the stress. So it is probably something that is going on inside the cells that makes the receptors unable to be re-synthesized or it should be modulation of NF-B or other mediators within the cell but we don’t know which mediators. Dr. Morley: Just to follow up for Dr. Armstrong’s question about defining stress and what kinds of stress. If you take depression as a stress factor in young people, the problem is that about 90% will basically put on weight if they are depressed, and 10% will lose weight. In an older population it will turn out to be that 60% will lose weight
201
Psychoimmunology of Nutrition and 30% won’t. So age interacts with the stress and in addition you also have the genomic effect. Within the immune system there are a number of studies in which the coping indexes have been looked at. The coping index is a psychological term which says how well you handle stress, and it turns out that people who can cope with a similar stress will have much fewer changes in the immune system than people who can’t cope. So we have an enormous challenge if we are trying to look at this because we have got to go from the genome to the brain and the environment all at the same time. Many years ago Pugh, an English man, said that if the human brain was so simple that we could understand it, we would be so simple that we couldn’t. That may be the best way to look at this whole area; it is very difficult to understand. Dr. Armstrong: A part of the problem then comes down to the definition of depression because depression is not necessarily the stress itself. Presumably some stress or something has lead to depression and whatever it was that lead to the depression may have been different initially and therefore the response may be different. In addition, as you said, there are coping skills, and something with which you can cope is much less likely then to be a stress. So that again is the problem with chronic stress because chronic exposure to something that might cause stress will not cause stress if you learn how to deal with it. So a constant stimulus will produce a changing effect depending on the individual organism’s ability to adapt to that. My question was not really with the expectation that we were going to be able to simplify this, but if we are stacked in a soup of different stresses and different responses, it is going to be very difficult to tease out the particular things we are trying to deal with. Dr. Morley: Psychologists who work in this area have tried to use defined stresses, such as mental arithmetic and telling the person their answer was wrong even if it was right. They look at the coping skills of the person up front, exercise to define physical stress. I think that is the way to start to tease out an area like this, to use very defined stresses that we can all understand and that are relatively simple. The problem is long-term, it is much harder to manipulate people and stress them for long-term than to define a stressor. We have examples of wars and things like that. In fact during an earthquake at UCLA in Los Angeles, a scientist actually measured the immune system responses to the earthquake and showed major changes that came back over an about 6-month period. So there are in fact ways of looking at those, but it is very difficult and I think we are both saying the same thing, it is an extraordinary complex area. I think that is what you said at the beginning and you did very well.
References 1 Ford DE, Erlinger TP: Depression and C-reactive protein in US adults. Arch Intern Med 2004;164:1010–1014. 2 Schedlowski M, Falk A, Rohne A, et al: Catecholamines induce alterations of distribution and activity of human natural killer (NK) cells. J Clin Immunol 1993;13:344–351. 3 Benschop RJ, Jacobs R, Sommer B, et al: Modulation of the immunologic response to acute stress in humans by beta-blockade or benzodiazepines. FASEB J 1996;10:517–524. 4 Ockenga J, Bischoff SC, Tillmann HL, et al: Elevated bound leptin correlates with energy expenditure in cirrhotics. Gastroenterology 2000;119:1656–1662. 5 Soderholm JD, Yang PC, Ceponis P, et al: Chronic stress induces mast cell-dependent bacterial adherence and initiates mucosal inflammation in rat intestine. Gastroenterology 2002; 123:1099–1108.
202
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 203–217, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
How Can We Impact the Immune System with Pre- and Probiotics? E.J. Schiffrin, A. Donnet and S. Blum Nestlé Research Center, Lausanne, Switzerland
Background In recent years there has been a growing interest in understanding the influence of intestinal microbiota on the physiology of the body. Moreover, with the available genomic studies, it is now possible to analyze how components of the intestinal microbiota modulate features of human postnatal development and physiology [1]. An area of major interest has been the relationship between the gut bacteria and the immune system, both at the intestinal and systemic level [2]. Changes in the microbiologic content of the intestine can be induced by the administration of selected bacterial inoculums as part of a normal diet or as dietary supplements. The health-promoting microorganisms are called probiotics. The administration of specific fibers in the diet called prebiotics can also modify the intestinal ecology by promoting the growth of some particular components of the intestinal microbiota, such as bifidobacteria. On the one hand, there is an immune activation which is associated with improved mucosal defenses against pathogens and responses to oral vaccines. On the other, a modified immune reactivity which preserves homeostasis in mucosal tissues confronted with a constantly changing environment. Not only does the latter avoid an excessive reaction and inflammatory damage in the local environment, it also influences the homeostasis of the systemic immune system and prevents the development of allergic or autoimmune diseases. It is difficult to provide a simple mechanistic explanation for the underlying cellular and molecular events that support these apparently opposing effects. However, a brief overview of the evolving models that have been postulated to explain basic immune function, may help us understand how intestinal bacteria effect the mucosal and systemic immune systems [3, 4]. The most important of these are the following. 203
Intestinal Microbiota and Immune Function (1) Burnet proposed the self-non-self theory in which antigen-specific immune cells (single cells) are activated following recognition of non-self. The model explains activation at the single cell level but fails to explain the mechanisms underlying the maintenance of homeostasis. (2) The Janeway model subsequently proposed that initial events in immune activation depend on stimulation of accessory cells or antigenpresenting cells (APCs) [5]. APCs are not antigen-specific cells. They can recognize highly conserved pathogen, or commensal, microbial-associated molecular patterns through a limited number of germ-line encoded molecules, called pattern recognition receptors (PRRs). Recognition and subsequent activation are part of the innate response. Thus, the innate immune system is able to discriminate between ‘infectious non-self’ and ‘noninfectious self’. APCs in this model are preferentially dendritic cells (DCs) which upon activation can present antigen and activate naïve antigen-specific T cells. This model marked the progression from the self-non-self theory to the infectious non-self model. (3) The most recently proposed is the ‘danger model’ by Matzinger [3]. This model proposes that resting tissue APCs, particularly DCs, have a ‘sentinel’ function and detect not only infectious non-self but also danger signals generated by, for example, damaged cells of the host [3]. Highly conserved, ancient PRR molecules called toll-like receptors (TLRs) are expressed on DCs and allow recognition of conserved bacterial motifs present on pathogens and commensals, and endogenous danger signals [3, 4, 6, 7]. This process results in maturation and activation of the DCs [5, 8] that depend on the nuclear translocation of the transcription factor NF-B and the subsequent activation of gene products involved in defenses, innate response and inflammation. In this last controversial model, the innate activation of the ‘sentinel’ DCs may be transient and resolve after clearance of infectious non-self or components of damaged self. This short-lived innate response is probably sufficient to protect against the majority of invading microorganisms and is not accompanied by an adaptive immune response since immature DCs, resident in peripheral tissues, are unable to induce effector responses of naïve T cells. Another possibility is that incompletely activated DCs produce interleukin (IL)-10 and thereby induce or prime regulatory T cells [9] instead of effector T cells. In either case, immune tolerance results. However, if pathogens or endogenous cell injury leads to persistent inflammatory reactions, the DCs become fully mature and activated [4, 10]. In turn, they activate naïve T cells to initiate the antigen-specific immune response through effector and memory T cells. Furthermore, they can integrate different stimuli and elicit Th1 or Th2 responses depending on the environmental signals [11]. In conclusion, DCs represent a heterogenous lineage of cells that attain different levels of maturation and activation depending on the nature of the stim204
Intestinal Microbiota and Immune Function uli in their microenvironment. This diversity in DC functional status elicits different, and often contradictory, types of T-cell responses which vary from deletion in the thymus to generation of effector and memory T cells, either Th1 or Th2, and T regulatory cells that participate in peripheral tolerance [12]. Application of the general models of immune function described above to the immune system of the intestinal mucosa has received little attention. No other place in the body has probably such an important interaction with nonself, or even with ‘infectious-non-self’, and yet, for the most part, it remains tolerant to commensal bacteria. An original feature of the intestinal mucosa is that the epithelial cells are in permanent contact with the luminal contents and, as such, bacterial products. Thus, intestinal epithelial cells (IECs), in addition to the ‘sentinel’ DCs, may also sense bacterial signals in the lumen, participate in innate immune responses and, through their secretory products, indirectly influence the stimulation of a subsequent adaptive immune reaction. Are the mechanisms underlying mucosal immune activation and immune homeostasis opposing processes or rather a range of responses that need to be finely tuned in order to cope with danger and yet prevent an overreaction to harmless or even beneficial, non-self symbionts? It is likely that in this exquisite physiological process, both innate and adaptive responses are operating in a coordinated manner which is in symbiosis with the intestinal microbiota.
Microbiota–Host Interactions Anatomical, Cellular and Molecular Bases of Microbiota–Host Interactions The interactions between the intestinal bacteria and the host occur in different mucosal environments. In very general terms, bacteria–eukaryotic cell interactions take place at the absorptive mucosa, in the areas devoid of lymphoid-organized structures and in the gut-associated lymphoid tissue (GALT), such as the Peyer’s patches in the small bowel. At both sites, bacterial cells or their components interact preferentially with IECs or DCs. In the GALT, commensals and pathogens target specialized epithelial M cells, which are present only in the follicular-associated epithelium and which facilitate their uptake across the epithelial layer. Once in the dome of the GALT, the bacteria interact with immature DCs. Whether pathogenic or not, bacteria, virus and other bugs have a number of molecular signatures, from proteins to nucleic acids, that are ligands for TLRs [13]. Epithelial cells and DCs are positioned to detect bacterial signals from both the complex variety of normal commensal microbiota and the superimposed pathogenic bacteria. Bacterial molecular signatures recognized by epithelial cells [14, 15] and DCs, although conserved and 205
Intestinal Microbiota and Immune Function sIgA secretion Commensals or probiotics
Mo
B
T iDC
iDC
Th3 cell, TGF- production Naïve T cell
Direct stimulation of B cell by DC and sIgA production
T regulatory cell, IL-10 production
Fig. 1. Interaction of the mucosal surface with commensals or probiotics. Bacterial cells come in contact with immune cells preferentially at the Peyer’s patches through the epithelium of M cells. T cells differentiate into regulatory T cells, which produce IL-10 and/or T helper (TH) 3 cells, which produce TGF-. B cells produce secretory IgA that will result in immune exclusion of bacteria and reinforcement of the mucosal barrier to prevent bacterial translocation. Overall the immunological consequences are local IgA production, systemic tolerance and local immune homeostasis.
of a limited repertoire, can induce different types of cellular and host responses [16]. The IECs, either primary cultures or cell lines, constitutively express TLRs, such as TLR2 and 4, which can bind bacterial products like peptidoglycans and lipopolysaccharide (LPS), respectively [16], and TLR3 and TLR5 which recognize double-stranded viral DNA and bacterial flagellin. The nature of the TLRs engaged influences the type of epithelial response [14, 16]. Cellular signalling initiated by non-pathogenic bacterial products binding to TLRs leads to nuclear factor B (NF-B) and mitogen-activated protein kinase activation [10], and expression of inflammatory genes such as chemokines. In normal situations, this physiological challenge and its response are of short duration and, as such, are well tolerated. They lead to a state of hyporesponsiveness. The length of the inflammatory response is apparently linked to the limited duration of the NF-B nuclear translocation [17]. Other molecular mechanisms of epithelial tolerance to commensal challenge have been suggested. They depend on low surface expression of TLRs and the MD-2 adaptor molecule, as well as upregulation of the regulatory intermediate Tollip, which inhibits TLR expression [18]. Furthermore, antiinflammatory downregulatory signals such as IL-10 or transforming growth 206
Intestinal Microbiota and Immune Function factor- (TGF-), originating from other mucosal cell types present in the lamina propria, may also be involved [19, 20]. However, hyper-reactivity may also be prevented by the production of soluble TLRs, specifically sTLR-2 [21], which are released upon stimulation with bacterial products such as LPS. Intestinal DCs can come in contact with commensal bacterial cells or their products (a) in the Peyer’s patches, (b) in the lamina propria due to ‘physiological’ bacterial translocation, or (c) through an active sampling by transepithelial DCs [22, 23]. Thereafter, the DCs can either participate in an innate reaction to prevent infection, initiate a protective secretory immune response, or participate in the induction and maintenance of immunological tolerance (peripheral tolerance) towards the organism [22]. DCs can also participate in the initiation and perpetuation of a proinflammatory immune reaction (fig. 1). In summary, both IECs and DCs are involved in the interaction with the intestinal microbiota, and play a role in the defense and homeostatic responses. Although both cell types participate in innate responses, only DCs play a direct role in the initiation of antigen-specific responses. A limited, inflammatory innate response to commensals is crucial to prevent mucosal damage. In addition, it has been suggested that commensals may play an active role in the modulation of sentinel reactivity to pathogens and proinflammatory molecules (see below). Differential Innate Response to Intestinal Bugs: From Pathogens to Different Types of Commensals Pathogenic microbes subvert nonspecific host-defenses to reach specific cellular receptors on host tissues. They thereby attain a niche which provides ecological advantages and improved survival [24]. The strategies and cellular receptors exploited by pathogens to interact with intestinal mucosal cells are varied. They bind to the same PRRs that are involved in the interactions with commensals. However, due to either the different cellular signalling or cytopathic effects they induce, they stimulate a different innate response. For the most part, the pathogenic response involves a persistent activation of NF-B [25], chemokine production and leukocyte recruitment with inflammatory tissue damage. Working together, the innate and adaptive immune responses ultimately clear the pathogen. Although under physiological conditions, interactions between commensals and the host can be varied, these organisms do not alter the integrity of the intestinal mucosa. Commensals are primarily embedded in the mucous layer, are coated by secretory antibodies, and very seldom do they seem to have direct contact with enterocytes. When this rare event happens, host epithelial cells detect or sense bacterial signals and react to them. It is possible that bacterial sensing by the host is in part independent of direct contact with the epithelial membrane since soluble PRRs, such as sCD14 and sTLRs, may play a sentinel role in the lumen of the gut [26]. Consequently, the normal flora seldom has direct contact with enterocytes. 207
Intestinal Microbiota and Immune Function The immune responses of the mucosal surface induced by interaction with commensal microbiota can be considered as antipathogenic and immunomodulatory. Bacteroides thetaiotaomicron has been used as a model of symbiotic bacteria in several experimental studies. Colonization with this organism reinforces the mucosal barrier innate defenses through the secretion of a bactericidal product, Ang4, by Paneth cells [27]. Furthermore, both Gramnegative and Gram-positive bacteria, LPS, lipoteichoic acid, lipid A and muramyl dipeptide, elicit antimicrobial peptides, such as ␣-defensins, that keep bacterial populations in check [28]. In addition, it has recently been reported that Paneth cell antimicrobial cryptdins play a role as paracrine regulators of the intestinal innate response to bacteria [29]. Thus sensing bacteria by cells of the mucosal surface at some compartments initiate a defensive cellular response and also a coordinated innate host reaction. The result of bacterial changes at the intestinal level can result not only in local responses but they can also have an influence on the systemic innate reactivity of the host. It has been observed that probiotic administration to healthy adult volunteers increased the phagocytic capacity of blood granulocytes and monocytes [30, 31]. In addition to the aforementioned antipathogenic effects commensals play an immunomodulatory activity. In vitro studies have shown that single epithelial cells responded differently to Gram-negative non-pathogenic enterobacteria and Gram-positive lactic acid bacteria (LAB). Only the former were able to clearly induce inflammatory gene activation probably through NF-B nuclear translocation [32]. More recent studies, performed with IEC-6 and primary rat epithelial cells confirmed a proinflammatory response of epithelial cells challenged by Gramnegative bacteria and a lack of response to Gram-positive nonpathogens [33]. Furthermore, it was shown that nonpathogenic Gram-negative bacteria had the capacity to signal IECs through TLR-4 and the NF-B system, specifically through phosphorylation and nuclear translocation of the RelA subunit. Studies using Caco-2/PBMC cocultures showed that epithelial cells responded differently to nonpathogenic bacteria and produced a transient, proinflammatory response to Gram-negatives [34]. When pathogenic Escherichia coli were used for comparison, the inflammatory response lasted longer than that induced by nonpathogenic E. coli. If an inflammatory response was induced to LAB using the same model, the reaction was transient. Of note, some specific strains of LAB did not induce an inflammatory response but increased expression of the gene for TGF-. TGF- is central to the maintenance of mucosal immune homeostasis. Interestingly, it inhibits innate immune activation via TLR-4 [35]. Furthermore, it is able to inhibit LPS or bacterially induced recruitment of NF-B to the IL-6 gene promoter through modulation of histone acetylation [20]. 208
Intestinal Microbiota and Immune Function Attenuated Salmonella prevents nuclear translocation of NF-B by inhibiting IB-␣ degradation and, moreover, prevents proinflammatory gene activation induced by a wild-type Salmonella [36]. In these experiments colonization of the monolayers with the attenuated strains also inhibited IL-8 induction by TNF-␣. B. thetaiotaomicron, a dominant component of human anaerobic microflora, was able to attenuate the proinflammatory reaction induced in vitro and in vivo by Salmonella enteritidis as well as other proinflammatory agonists by regulating the nuclear export of RelA complexed to peroxisome proliferator-activated receptor-␥ [17]. In in vitro IECs/immune cell coculture assays, secretory products, such as IL-10 produced by immune cells in the basolateral compartment, have also been shown to play a role in dampening the transient inflammatory reaction of epithelial cells. Thus commensals and selected probiotic strains can induce a defensive reaction of the IECs and also downregulatory responses to proinflammatory stimuli. Adaptive Immune Response to Commensals Intestinal components of the microbiota reside in the lumen, embedded in the mucus covering the mucosal surface and coated by secretory immunoglobulin A (sIgA) [37]. Although a symbiotic relation between the host and its microbiota was thought to be based on immunological tolerance, antibody and cellular responses to commensal bacterial antigens have been reported by different authors. Stimulation of IgA secretion by non-pathogens in the intestinal lumen may provide a shield against bacterial dissemination or translocation. It may also prevent the antigenic dissemination that could stimulate overwhelming immune reactions and autoimmunity. A large proportion of intestinal IgA are specifically directed to commensal cell wall antigens and it is independent of T-cell help [38, 39]. Moreover specific commensal antibodies are found in intestinal secretions but seldom in serum. Occasionally it has been reported that healthy human subjects can have serum antibody responses to the predominant intestinal lactobacilli and bifidobacteria of their own microbiota, thus although local responses seem to be prevalent, some leakage of nonpathogenic bacterial antigens cannot be excluded [40]. The induction of the local secretory immune response is initiated by DC sampling of commensal bacterial cells and triggering a local protective immune response in the lamina propria and in the mesenteric lymph nodes without systemic dissemination and immune activation. Thus symbiosis between the host and its microflora is supported by the competence of the host mucosal immune system [39]. It is not known whether this secretory immune response limits bacterial growth or also inhibits penetration of the mucosal surface. In any case the response does not result in prevention of intestinal colonization. 209
Intestinal Microbiota and Immune Function The interaction of the intestinal microbiota and the host adaptive immune system depends on cell populations, mainly DCs, that can recognize dangerous and harmless environmental signals. In the absence of inflammation or another type of tissue damage, DCs in contact with commensal bacteria will attain a partial degree of maturation and stimulate regulatory T-cell differentiation, B-cell IgA production, immunological tolerance at the systemic level and homeostasis of the immune function [41]. In contrast if pathogens are encountered or an inflammatory reaction is ongoing, then DCs will prime effector cells, most probably of the Th1 subtype, and thereby contribute to further inflammation and breakdown of oral tolerance. It has been observed that the administration of probiotics can have an immunoadjuvant effect during oral vaccination against Salmonella in healthy human subjects [42]. The cellular events of the adjuvant activity have not been characterized but were not associated with any inflammatory reaction. Recently it was reported that the administration of Enterococcus faecium to young dogs induced an increase in polyclonal fecal IgA and specific antibodies against canine distemper virus vaccination [43].
From Cellular Events and Experimental Studies to Health Benefits The physiology of the innate and adaptive intestinal immune response cannot be evaluated independent of the multiple and complex interactions that the host entertains with its microbiota. Epidemiological evidence suggests that an improved standard of life is associated with an increased incidence of diseases with underlying immunopathological mechanisms. Allergy, autoimmunity, inflammatory bowel disease are all mediated by pathological immune responses. Very clean or ‘germ-free’ environments seem to contribute to a huge increase in diseases such as allergy and asthma. In addition to this cleaner environment, it is possible that sterile foods, prescribed antibiotics or their inadvertent presence in our diet, all contribute to the reduced Trophism of the intestinal microbiota. As the latter can be considered a real ‘organ’ of the body, an effect of this magnitude modifies several important functions, most particularly the immune response. Several factors are prompting the medical community to examine ways in which to modify or to feed this ‘organ’ and thereby achieve health benefits. There are emerging scientific and clinical evidence to support the use of probiotics to control diarrhea, allergy, chronic inflammatory bowel disease and gastrointestinal infections. The mechanisms are only partially understood, but immune function can be positively modified by the administration of carefully selected strains [44].
210
Intestinal Microbiota and Immune Function It has also been shown that prebiotics can modulate the host intestinal immune response. Their administration leads to an increment in the intestinal IgA response [45]. Furthermore, given to elderly humans they change the intestinal ecology and have an influence on the systemic inflammatory status that is frequently observed with increasing age [46]. Thus modulation of the innate immune response at the mucosal surface is not a total abrogation of reactivity, strong responses can still be initiated if the surface is exposed to proinflammatory cytokines or microbial products that may be ‘interpreted’ by the sentinel cells to be virulence factors or danger signals. Immune regulation does not occur without activation. Some level of intermediate activation or priming is required if the host is to respond quickly to pathogens. Fortunately, this activation is tightly controlled and occurs in the absence of major inflammatory damage. It is achieved through the intervention of commensal or probiotic bacteria that induce the production of modulatory mediators such as TGF-, IL-10, and IL1-ra.
References 1 Xu J, Gordon JI: Honor thy symbionts. Proc Natl Acad Sci USA 2003;100:10452–10459. 2 Köhler H, McCormick BA, Walker WA: Bacterial-enterocyte crosstalk: Cellular mechanisms in health and disease. J Pediatr Gastroenterol Nutr 2003;36:175–185. 3 Matzinger P: The danger model: A renewed sense of self. Science 2002;296:301–305. 4 Matzinger P: Tolerance, danger, and the extended family. Annu Rev Immunol 1994;12: 991–1045. 5 Medzhitov R, Janeway CA Jr: Decoding the patterns of self and nonself by the innate immune system. Science 2002;296:298–300. 6 Kamradt T, Mitchison NA: Tolerance and autoimmunity. N Engl J Med 2001;344:645–664. 7 Kopp E, Medzhitov R: Recognition of microbial infection by Toll-like receptors. Curr Opin Immunol 2003;15:396–401. 8 Pasare C, Medzhitov R: Tool-like receptors and acquired immunity. Semin Immunol 2004;16:23–26. 9 Jonuleit H, Schmitt E, Schuler G, et al: Induction of interleukin 10-producing nonproliferating CD4⫹ T cells with regulatory properties by repetitive stimulation with allogeneic immature human dendritic cells. J Exp Med 2000;192:1213–1222. 10 Chang L, Karin M: Mammalian MAP kinase signaling cascades. Nature 2001;410:37–40. 11 Gagliardi MC, Sallusto F, Marinaro M, et al: Cholera toxin induces maturation of human dendritic cells and licenses them for Th2 priming. Eur J Immunol 2000;30:2394–2403. 12 Lanzavecchia A, Sallusto F: The instructive role of dendritic cells on the T cell response: Lineages, plasticity and kinetics. Curr Opin Immunol 2001;13:291–298. 13 Reis e Sousa C: Toll-like receptors and dendritic cells: For whom the bug tolls. Semin Immunol 2004;16:27–34. 14 Cario E, Podolsky DK: Differential alteration in intestinal cell expression of toll-like receptor 3 (TLR3) and TLR4 in inflammatory bowel disease. Infect Immun 2000;68:7010–7017. 15 Cario E, Gerken G, Podolsky DA: ‘For whom the bell tolls’ – Innate defense mechanisms and survival strategies of the intestinal epithelium against lumenal pathogens. Z Gastroenterol 2002;40:983–990. 16 Cario E, Brown D, McKee M, et al: Commensal-associated molecular patterns induce selective toll-like receptor trafficking from apical membrane to cytoplasmic compartments in polarized intestinal epithelium Am J Pathol 2002;160:165–173.
211
Intestinal Microbiota and Immune Function 17 Kelly D, Campbell JI, King TP, et al: Commensal anaerobic gut bacteria attenuate inflammation by regulating nuclear-cytoplasmic shuttling of PPAR-␥ and RelA. Nat Immunol 2004;5:104–112. 18 Otte J-M, Cario E, Podolsky DK: Mechanisms of cross hyporesponsiveness to Toll-like receptor bacterial ligands in intestinal epithelial cells. Gastroenterology 2004;126:1054–1070. 19 Haller D, Serrant P, Perruisseau G, et al: IL-10 producing CD14 low monocytes inhibit lymphocyte-dependent activation of intestinal epithelial cells by commensal bacteria. Microbiol Immunol 2002;46:195–205. 20 Haller D, Holt L, Kim SC, et al: Transforming growth factor -1 inhibits non-pathogenic Gramnegative bacteria-induced NFB recruitment to the interleukin-6 gene promoter in intestinal epithelial cells through modulation of histone acetylation. J Biol Chem 2003;278: 23851–23860. 21 LeBoulder E, Rey-Nores JE, Rushmere NK, et al: Soluble forms of toll-like receptor (TLR)2 capable of modulating TLR2 signaling are present in human plasma and breast milk. J Immunol 2003;171:6680–6689. 22 Uhlig HH, Powrie F: Dendritic cells and the intestinal microflora: A role for localized immune responses. J Clin Invest 2003;112:648–651. 23 Rescigno M, Urbano M, Valzasina B, et al: Dendritic cell express tight junction proteins and penetrate gut epithelial monolayers to sample bacteria. Nat Immunol 2001;2:361–367. 24 Gruenheid S, Finlay BB: Microbial pathogenesis and cytoskeletal function. Nature 2003; 422:775–781. 25 Kim JG, Lee SJ, Kagnoff MF: Nod 1 is an essential signal transducer in intestinal epithelial cells infected with bacteria that avoid recognition by toll-like receptors. Infect Immun 2004;72:1487–1495. 26 Labéta MO, Vidal K, Rey Nores JE, et al: Innate recognition of bacteria in human milk is mediated by a milk-derived highly expressed pattern recognition receptor, soluble sCD14. J Exp Med 2000;191:1807–1812. 27 Hooper LV, Stappenbeck TS, Hong CV, Gordon JI: Angiogenins: A new class of microbicidal proteins involved in innate immunity. Nat Immunol 2003;4:269–273. 28 Ayabe T, Satchell DP, Wilson CL, et al: Secretion of microbicidal alpha-defensins by intestinal Paneth cells in response to bacteria. Nat Immunol 2000;1:113–118. 29 Lin PW, Simon PO, Gewirtz AT, et al: Paneth cell cryptdins act in citro as apical paracrine regulators of the innate inflammatory response. J Biol Chem 2004;279:19902–19907. 30 Schiffrin EJ, Rochat F, Link-Amster H, et al: Immuno modulation of human blood cells following the ingestion of lactic acid bacteria. J Dairy Sc 1995;78:491–497. 31 Schiffrin EJ, Brassart D, Servin A, et al: Immune modulation of blood leukocytes in man by lactic acid bacteria: Criteria for the strain selection. Am J Clin Nutr 1997;66:515S–520S. 32 Delneste Y, Donnet-Hughes A, Schiffrin EJ: Functional foods: Mechanisms of action on immunocompetent cells. Nutr Rev 1998;56:S93–S98. 33 Haller D, Russo MP, Sartor RB, Jobin C: I and phosphatidylinositol 3-Kinase/Akt participate in non-pathogenic gram-negative enteric bacteria-induced RelA phosphorylation and NF-B activation in both primary and intestinal epithelial cell lines. J Biol Chem 2002;41:38168–38178. 34 Haller D, Bode C, Hammes WP, et al: Non-pathogenic bacteria elicit a differential cytokine response by intestinal epithelial cell/leukocyte co-cultures. Gut 2000;47:79–87. 35 McCartney FN, Jin W, Wahl SM: Aberrant Toll receptor expression and endotoxin hypersensitivity in mice lacking a functional TGF-beta 1 signaling pathway. J Immunol 2004;172: 3814–3821. 36 Neish AS, Gewirtz AT, Zeng H, et al: Prokaryotic regulation of epithelial responses by inhibition of IB-␣ ubiquitination. Science 2000;289:1560–1563. 37 van der Waaij LA, Limburg PC, Mesander G, van der Waaij D: In vivo IgA coating of anaerobic bacteria in faeces. Gut 1996;38:348–354. 38 Macpherson AJ, Gatto D, Sainsbury E, et al: A primitive T cell-independent mechanism of intestinal mucosal IgA responses to commensal bacteria. Science 2000;288:2222–2226. 39 Macpherson AJ, Uhr T: Induction of protective IgA by intestinal dendritic cells carrying commensal bacteria. Science 2004;303:1662–1665. 40 Kimura K, McCartney AL, McConnell MA, Tannock GW: Analysis of fecal populations of bifidobacteria and lactobacilli and investigation of the immunological responses to their human host to the predominant strains. Appl Environ Microbiol 1997;63:3394–3398.
212
Intestinal Microbiota and Immune Function 41 Mowat AM: Anatomical basis of tolerance and immunity to intestinal antigens. Nat Rev Immunol 2003;3:331–341. 42 Link-Amster H, Rochat F, Saudan KX, et al: Modulation of a specific humoral immune response and changes in intestinal flora mediated through fermented milk intake. FEMS Immunol Med Microbiol 1994;10:55–63. 43 Benyacoub J, Czarnecki-Maulden GL, Cavadini C, et al: Supplementation of food with Entercoccus faecium (SF68) stimulates immune functions in young dogs. J Nutr 2003;113: 1158–1162. 44 Reid G, Jass J, Sebulky MT, McCormick JK: Potential uses of probiotics in clinical practice. Clin Microbiol Rev 2003;16:658–672. 45 Hosono A, Ozawa A, Kato R, et al: Dietary fructooligosaccharides induce immunoregulation of intestinal IgA secretion by murine Peyer’s patch cells. Biosci Biotechnol Biochem 2003; 67:758–764. 46 Guigoz Y, Rochat F, Perruisseau-Carrier G, et al: Effects of oligosaccharide on the faecal flora and non-specific immune system in elderly people. Nutr Res 2002;22:13–25.
Discussion Dr. Powell-Tuck: I think what one is hearing here is very little short of the very fundamental basis of gastroenterology. This is what gut disease and gut health is all about, and the further understanding of this will undoubtedly be the way forward in the future. Just going on though from that, if we are going to treat gastrointestinal disease, we have got to alter these responses, we have got to improve mucosal health. How can we beneficially influence the functions of these dendritic cells and all the cytokines, apart from probiotics which are going to have a modest effect? I am aware of the huge effects of antibiotics in hospital care which massively outweigh the effect of prebiotics and probiotics, but how else do you think we can influence this fundamental cellular function? Dr. Schiffrin: I don’t know very well where we are going. This is a critical moment for the whole subject because having jumped into probiotics and prebiotics some years ago, we honestly had the illusion that they were good for everything. But then the more we knew about them the more we realized that sometimes they can be pushing the immune system towards a Th1 or a Th2 response or promoting or not oral tolerance. I think this raises some concerns and also the necessity to make better research. I don’t know if the industry will have the energy to support that or who is going to support that, but I think today the most rational intelligent selection of probiotics for specific problems will come from a genomic evaluation that a lot of people have done. We can see that bacteria that are taxonomically close can interact with cells or with animals or with humans in totally different ways. So this is the problem today, we don’t know how to predict the biological activity, it is not like a single molecule such as glutamine, we are dealing with something complex and really at the moment the selection of probiotics still remains very empiric. Then of course we have to be sure that they are safe first, and then we do our best guess. But I don’t think yet that we are doing really an intelligent selection of probiotics for specific problems. Dr. Endres: You showed a lot of interactions of probiotics and the immune system. Concerning the prebiotics the data were more rough clinical results. Could you please outline a little bit more the mechanistic ideas about how prebiotics work? Is it via probiotics or is it a direct effect or possibly both? Dr. Schiffrin: We have assumed, probably from a simplistic point of view, that immune modulation with nutrients such as probiotics is more achievable in the small bowel. Bacterial colonization is lower and there are major immune cell populations in
213
Intestinal Microbiota and Immune Function the mucosa. In contrast, to change the ecological condition of the colon containing massive bacterial populations and, in addition, poorer immune cell compartments seems a more difficult task. Any nutritional bacterial inoculum will probably have a minor consequence in changing the endogenous bacterial communities of the colon. Are probiotics able to grow and establish in an already overcrowded habitat? I think for this particular purpose that prebiotics may represent an interesting possibility. The example of the probiotic blend VSL3 as an efficient way to control pouchitis may in part be due to the lack of colon and thus ingested bacteria get to the distal reconstructed intestine faster. I don’t know if other bacterial blends could also have some beneficial effect in this particular condition. Another example of the successful use of probiotics is in the case of prevention or even treatment of the Th2-driven immune reactions in atopy, but our guess is that the action takes place in the small bowel. In the case of nutrition for a change in the colonic ecology, prebiotics are interesting candidates. It is true that I have shown more mechanistic explanations for the potential benefits induced by probiotics and more clinical evidence without a detailed mechanistic approach of how prebiotics work. Our first guess is of course that they can change bacterial communities and their metabolic activities in an altered ecological system due to the clinical conditions and their treatments, in particular antibiotics. Dr. Morley: That was a brilliant lecture and I really enjoyed it although I am not claiming I understood all of it. If we go back historically to about the early 1900s, Medziekov decided that because Bulgarians ate a lot of yogurts and seemed to live forever, though they didn’t, that basically we could actually get longevity by altering gut flora. The problem I have as I look at the literature is that we are almost at the same stage now as we were 105 years ago, and I am wondering if we are tackling the wrong problems. Whether the evidence is true or not, I think many of us will believe that giving yogurt will certainly make a difference, or taking antibiotics particularly in older people as far as diarrhea is concerned. But the situations where I would assume that we are liable to see a major effect would be in irritable bowel syndrome and gas production, things which are very dependent upon changes in gastrointestinal motility in which an alteration in cytokines could easily alter the system. Many years ago we showed that exorphins in the food fed to a person will basically alter gastrointestinal motility and therefore one would assume that you can do the same sort of thing with pre- and probiotics depending on how you alter the cytokine milieu that has been produced in the gut. I wonder how much data there are to support this sort of concept. Certainly there were data many years ago, which nobody ever followed up as far as I know, showing that with bacterial overgrowth you actually get severe anorexia and weight loss which is curable by antibiotics. I wonder if that area is not where we should be going a little bit and are there data to support going down that sort of pathway? Do we have good animal models to perhaps look at gastrointestinal motility in a different biotic structure at this moment in time? Dr. Schiffrin: I think this is a very good point and irritable bowel syndrome is a major problem in gastroenterology. There is a group in France that is working along these lines, trying to relate bacterial colonization or some probiotics and intestinal motility [Bueno L., personal commun.]. Nowadays we have irritable bowel syndrome with and without inflammation; we have a heterogenous picture with different underlying mechanisms. Cytokine production is playing a role. I think that in very few months we will see this information in the literature showing the interest in this field. Dr. Bowling: You have shown us an awful lot of animal data. Of course if this is going to translate into useful products along these lines, we have got to have in vivo human data. The problem with a lot of the studies, and certainly some that you
214
Intestinal Microbiota and Immune Function presented, is that we are looking at short-chain fatty acid concentrations in the stools. However, the most important part of the colon is the ascending colon, and what is coming out of the bottom end is not necessarily any kind of reflection of what is going on physiologically much further into the system. I am just wondering how much we actually know of what is going on in that part of the human colon in vivo? Dr. Schiffrin: I agree with you. We know that we will be losing information analyzing the fecal short-chain fatty acids instead of going to the place where their generation is taking place. There are some studies that have tried to sample using magnetic capsules that can be given to the volunteer, followed by radioscopy and sampling at the right place, and analysis of short-chain fatty acids was then made. Macfarlane et al. [1] did studies like that. So it is true, those are very complicated studies, you have to have the volunteers for a couple of days and of course give the diet many days before, but then at the moment of sampling you have to be very careful. So these things are going on, I don’t expect to see many studies trying to clarify these points. I guess that we will have to be satisfied with just a few but of course these points were raised. Dr. Armstrong: This is highly speculative. I was wondering if you could speculate on how important prebiotics and probiotics are going to be in early life in the immediate postnatal period. I remember a person saying that our gut flora is pretty well determined within a few days of birth, or within a few weeks, and we also heard this morning that there are differences between twins with respect, for example, to calcium and vitamin D status decades into life. So my question is, is the behavior of our gut and the way that it deals with its microflora and the consequences determined by the nutrient status, the initial colonization? Do you think that there is a role for investigating or dealing with people’s problems very early on with prebiotics and probiotics? Dr. Schiffrin: I think that your question is fascinating because we don’t yet understand the physiology of neonatal colonization. I guess that what we were discussing earlier today regarding the challenge to have nutritional effects despite gene diversity is also applicable to the immune system. However, in this latter case the possibility of immune education seems to exist. The immune system has a memory, so if you really do the real good points at the beginning of life or in the critical moments early in life, I don’t know when this time is, perhaps 7 days after birth or at weaning, then I think that you can improve a lot of things in the immune system despite the fact that you will always be dealing with genetic diversity. Now the question is do we intervene already in children, in neonates? I think a lot of people are starting to do that with reasonable success. I still think though that we need to do some homework to understand the physiological events of neonatal colonization with breast feeding. We still don’t know what the relevant points are there, and I think until we understand that, the temptation to intervene without having the frame of what is physiological is not the right way to do things. Of course there are such artificial situations in the neonatal intensive care unit in which we know antibiotics will be used. The neonates will not have breast milk, they will be colonized by necrotizing enterocolitis promoting bacteria, then in that case of course the benefit could be so good that you are tempted to do something. But for the general population I think that we only partially understand the problem for the moment. Dr. Endres: It is known that breast-fed infants have a predominance of bifidobacteria in their feces [2–5] and it is also known that infants fed with a whey-predominant formula have some bifidobacteria [3], whereas those fed with a casein-predominant formula have particularly no bifidobacteria [6]. So the industry is trying to imitate this status seen in breast-fed infants, and so far we are trying to reach this goal by using Bifidobacterium lactis in infant formulae. A positive effect is that with this you can
215
Intestinal Microbiota and Immune Function reduce the prevalence of diarrhea, the prevalence of constipation, but when it comes to the very critically ill preterm babies there are two studies, one with Lactobacillus GG [7] and the other with the Lactobacillus acidophilus [8] showing that there is a colonization, which is not really a colonization but counting the bugs in the stools, and the authors have shown that the bacteria given are found in the feces but the negative bacteria are not diminished. There are two unsuccessful clinical studies in preterm infants trying to reduce the rate of necrotizing enterocolitis, one using Lactobacillus GG [9], the other using B. lactis [10]. I think this is really the group of patients who could have benefited from a formula or expressed breast milk fortified with probiotics. Dr. Schiffrin: Yes, in addition, there is a study in Colombia with about 1,200 preterm babies given Lactobacillus GG with less incidence of necrotizing enterocolitis [11]. So absolutely, this population seems to be very candidate. Dr. Ockenga: Are you aware of any major clinically relevant side effects when using probiotics? Dr. Schiffrin: I know a few cases. For example there are reports of fungemia with Saccharomyces boulardii in critically ill patients [12]. S. boulardii was given to prevent diarrhea, it has been used with good efficacy, and apparently they don’t know whether this was really a translocation of the yeast or if it was a spray when facets were opened. There are some doubts about the use of live bacteria in the intensive care unit due to the spray possibility. Sometimes we don’t want even to think of probiotics or how to use a probiotic in the intensive care unit because if you can culture from a catheter or if during the manipulation there is a mistake then the use of probiotics will be questioned. Then there is a case report of a liver abscess with GG in a diabetic woman consuming enormous quantities of GG-containing products [13]. There are now very interesting studies going on in Finland. A study has recently been published in which they looked for sepsis due to gram-positive lactobacilli and in 5 years they had 26 cases due to Lactobacillus rhamnosus and in 11 of those cases they have a fingerprint with a profile which is similar to GG [14]. So there are things coming out. Of course sometimes there are the immunodepressed, the immunosuppressed patients, and safety considerations are always a priority. Dr. Labadarios: As far as normality is concerned the side effects are actually very few. Regarding the immunocompromised patient that is another matter, and I think Dr. Schiffrin just shown that there is a lot of caution to be exercised. Dr. Schiffrin: I think that the benefit that we could have in immunocompromised patients is potentially enormous using probiotics since we know that the wasting syndrome of immunocompromised patients depends on the colonization of Escherichia coli or translocation of E. coli. On the other hand 2 patients with short bowel had sepsis due to GG. So I think we could really have a benefit but of course who will dare to go first, the nutrition company or the pharmaceutical company, we are in a nutrition world so it is really difficult. We are always a little bit blocked, since nutritional interventions are usually not established to tackle very severe health problems. Dr. Roessle: At Nestlé, we have seriously considered proposing products for enteral nutrition with probiotics. If you listen to the experts there are two typical reactions: most gastroenterologists are rather favorable, but the physicians in intensive care have problems handling the paradox of using alive bacteria in an environment which is meant to be germ-free or sterile, and adding probiotics to tube feeding which, by definition, is almost a sterile product is a kind of contradiction. Of course there are some occasional side effects reported in the literature [15, 16], but if you look at that you also have to discuss the risk-benefit potential and as soon as you have therapeutic means to fight against the side effects by using antibiotics then perhaps the benefits will overtake. So our opinion is still in favor of trying to use probiotics in selected clinical conditions.
216
Intestinal Microbiota and Immune Function References 1 Macfarlane GT, Gibson GR, Cummings JH: Comparison of fermentation reactions in different regions of the human colon. J Appl Bacteriol 1992;72:57–64. 2 Benno Y, Sawada K, Mitsuoka T: The intestinal microflora of infants: Composition of fecal flora in breast-fed and bottle-fed infants. Microbiol Immunol 1984;28:975–986. 3 Bisicchia R, Sozzi T, Bertelli M, et al: Bifidobacteria colonization in diversely-fed newborn infants (in Italian). Pediatr Med Chir 1984;6:395–400. 4 Nagendra R, Vishwanatha S, Rao SV, Ravish SR: Effect of feeding infant formula containing lactulose on intestinal flora in the infant. Indian J Pediatr 1992;59:763–766. 5 Roberts AK, Chierici R, Sawatzki G, et al: Supplementation of an adapted formula with bovine lactoferrin: 1. Effect on the infant faecal flora. Acta Paediatr 1992;81:119–124. 6 Simhon A, Douglas JR, Drasar BS, Soothill JF: Effect of feeding on infants’ faecal flora. Arch Dis Child 1982;57:54–58. 7 Millar MR, Bacon C, Smith SL, et al: Enteral feeding of premature infants with Lactobacillus GG. Arch Dis Child 1993;69:483–487. 8 Reuman PD, Duckworth DH, Smith KL, et al: Lack of effect of Lactobacillus on gastrointestinal bacterial colonization in premature infants. Pediatr Infect Dis 1986;5:663–668. 9 Dani C, Biadaioli R, Bertini G, et al: Probiotics feeding in prevention of urinary tract infection, bacterial sepsis and necrotizing enterocolitis in preterm infants. A prospective double-blind study. Biol Neonate 2002;82:103–108. 10 Mihatsch WA, Vossbeck S, Franz AR, et al: Effect of enteral administration of a probiotic strain of bifidobacterium on the incidence of nosocomial infections in preterm infants. Pediatr Res 2004;55:448A. 11 Hoyos AB: Reduced incidence of necrotizing enterocolitis associated with enteral administration of Lactobacillus acidophilus and Bifidobacterium infantis to neonates in an intensive care unit. Int J Infect Dis 1999;3:197–202. 12 Lherm T, Monet C, Nougiere B, et al: Seven cases of fungemia with Saccharomyces boulardii in critically ill patients. Intensive Care Med 2002;28:797–801. 13 Sipsas NV, Zonios DI, Kordossis T: Safety of Lactobacillus strains used as probiotic agents. Clin Infect Dis 2002;34:1283–1284. 14 Salminen MK, Tynkkynen S, Rautelin H, et al: Lactobacillus bacteremia during a rapid increase in probiotic use of Lactobacillus rhamnosus GG in Finland. Clin Infect Dis 2002; 35:1155–1160. 15 Salminen S, von Wright A, Morelli L, et al: Demonstration of safety of probiotics – A review. Int J Food Microbiol 1998;44:93–106. 16 Borriello SP, Hammes WP, Holzapfel W, et al: Safety of probiotics that contain lactobacilli or bifidobacteria. Clin Infect Dis 2003;36:775–780.
217
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 219–232, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
How Can We Modulate Cytokine Production and Action? Luc Cynober Biochemistry Laboratory, Hôtel-Dieu Hospital AP-HP and Laboratory of Biological Nutrition, EA 2498, Pharmacy School, Paris University, Paris, France
Introduction: Role of Cytokines in Wasting Diseases The loss of body weight and development of cachexia are common signs associated with several diseases. Net muscle protein catabolism is the result of a neuronal and endocrinological response, the main hormone involved in this process being cortisol [1]. Besides this, a number of pathological situations (e.g. cancer, infection, trauma, surgery) lead to activation of the immunological system which, in particular, involves the release of mediators. Among these, cytokines play a preeminent role. Cytokines are now classified according to the cell subset synthesizing them (i.e. Th1 and Th2) and their main action (i.e. pro- versus anti-inflammatory). The main Th1 cytokines are tumor necrosis factor-␣ (TNF-␣), interleukin (IL)-1␣ and  and interferon-␥ (IFN-␥). The main Th2 cytokines are IL-4 and IL-10. A notable exception to this classification is IL-6, which is synthesized by Th2 cells but is more a proinflammatory cytokine (PIC) [2]. Th1 cytokines inhibit Th2 cytokine production and vice versa. PICs contribute to protein wasting via several mechanisms, including a direct effect on: (i) protein turnover (i.e. net protein catabolism) increasing protein catabolism through NF-B activation [3] and activation of the ubiquitinproteasome system [4, 5], especially activation of E3 ligases atrogin-1 and MURF-1 [6–8], decreasing protein synthesis through inhibition of eIF2B [9] or other factors involved in the translation process such as 4E-BP1 [10], and (ii) amino acid metabolism and oxidation through gluconeogenesis [11]. Also, cytokines potentiate cortisol and glucagon action at the target tissue level (i.e. muscle and liver, respectively) and blunt IGF-1 production and action in muscle [6]. However, the problem is complicated by the fact that each individual PIC has a more or less significant effect on specific aspects of protein 219
How Can We Modulate Cytokine Production and Action? metabolism (e.g. protein catabolism, protein synthesis, amino acid oxidation or tissue efflux) [12]. There are very few studies concerning the production of cytokines and its consequences in patients under long-term artificial nutrition at home, but there is no reason to think that this may be different from what happens at the hospital once the disease evolves. In a recent study, Hise et al. [13] studied well-nourished stable patients on home parenteral nutrition (10 short bowel syndrome, 2 dysmotility syndromes). TNF-␣, IL-6 and C-reactive protein plasma levels were not different from controls. However, soluble TNF-␣ receptors p55 and p75 and sICAM-1 were significantly higher. This suggests that even in stable well-nourished patients the method of feeding may alter the immunological status and thus nutritional status. In this short review, various means for modulating PIC are discussed, focusing specifically on the regulation of protein metabolism. Occasionally, the effects of manipulations on survival are described. In order to remain as clear as possible, manipulations are described according to the target (i.e. cytokine production, cytokine action). Figures 1 and 2 summarize the regulation of cytokine synthesis and actions.
Limiting PIC Production It is possible to counteract PIC production using physiopathological approaches. Cortisol exerts a retro-inhibitory effect on PIC synthesis [14]. However, the use of glucocorticoids to inhibit PIC synthesis is unsuitable as an anti-cachexia therapy since glucocorticoids themselves are potent inducers of protein catabolism [14]. Several drugs are able to decrease PIC production. Among them, pentoxifylline (PTX), a phosphodiesterase inhibitor, has been the most extensively studied. PTX inhibited TNF-␣ production, restored IL-2 synthesis and improved survival in a burn-injury mouse model [15]. In a model of rats infected by living Escherichia coli, PTX suppressed the increase in plasma concentrations of TNF-␣, and partially prevented the inhibition of protein synthesis [16, 17] and stimulation of protein degradation [16]. Finally, PTX decreased anorexia, but the effects on protein turnover described above were not dependent on food intake [17]. The action of PTX does not seem to be related to antioxidant properties since this drug failed to modify the glutathione decrease and malonyldialdehyde increase in a model of liver ischemia/reperfusion injury [18]. Curcumin (diferuloylmethane) is a dietary pigment that gives curry its yellow color. Its action relies on its ability to block the activity of the NF-B inhibitor, IB [19] or upstream at MEKK-1 level [20] (see fig. 2 for details about NF-B processing and action). Curcumin administration stimulates 220
How Can We Modulate Cytokine Production and Action? Stimuli
GLN CYS GLY
Vitamin E Glutathione
ROS
⫹
ALLN Curcumin Vitamin C
NFB
Producing cell
Isohelenin Cloricromen Gene transcription Proinflammatory cytokine (PIC) Blood stream
⫹
PIC
TNF␣ab
Cortisol
IL-1ra IL-6ra IL-1rab
RU38486 Target cell
Receptor Ubiquitin proteasome system
ARA
Cox
PGE2
eIF2B 4E-BP1
Indomethacin ibuprofen
Fig. 1. A simplified presentation of cytokine processing and action, and how drugs or nutrients can interfere with cytokine production or action. Stimuli such as LPS lead to reactive oxygen species (ROS) and NF-B activation (see figure 2 for details), leading to gene transcription and cytokine synthesis. Cytokines bind to ubiquitous receptors eliciting signals responsible for alterations in protein metabolism. Cortisol may potentiate certain actions of cytokines. Drugs and nutrients can interfere at every step of the process. GLN ⫽ Glutamine; CYS ⫽ cysteine; GLY ⫽ glycine; ALLN ⫽ N-acetylleucinyl-leucinyl-norleucinal; ab ⫽ antibody; ra ⫽ receptor agonist; rab ⫽ receptor antibody; Cox ⫽ cyclooxygenase; ARA ⫽ arachidonic acid.
muscle regeneration after injury [19]. Curcumin inhibits IL-1-mediated ICAM-1 and IL-8 gene expression in intestinal cells [20]. N-Acetyl-leucinyl-leucinyl-norleucinal (ALLN) is a potent inhibitor of proteolysis catalyzed by proteasomes. By this mechanism, ALLN inhibits the degradation of IB and the proteolytic cleavage of p105 to p50 and therefore the formation of the active p50/p65 NF-B heterodimer [21]. Hence, ALLN counteracts IL-6 and TNF-␣ production by macrophages in vitro or in vivo in response of mice to LPS [21]. Cloricromene is a semi-synthetic non-anticoagulant coumarine derivative with anti-platelet and anti-leukocyte properties. Cloricromene inhibits LPSinduced TNF-␣ release by rat macrophages in a dose-dependent manner and inhibits LPS-induced expression of TNF-␣ mRNA [22]. In addition, in this situation cloricromene inhibits NF-B activation as a result of an inhibition of LPS-induced cellular oxidative activity [22]. 221
How Can We Modulate Cytokine Production and Action? Stimuli ⫹
MEKK-1 P
NIK Curcumin Pentoxifylline (?)
IB kinase IB
IB
P
Vitamin C ALLN
P65 P50 Isohelenin
P65 P50 Gene transcription Cytoplasm
Nucleus Gene expression
Fig. 2. Simplified description of activation of NF-B and the target of various inhibitors. NF-B is a heterodimer composed of two subunits, namely p65 and p50. IB complexes and sequesters NF-B in the cytoplasm, preventing NF-B action. Following stimulation (e.g. LPS, cytokines) IB is phosphorated, which initiates its ubiquination and degradation. Subsequently released NF-B translocates to the nucleus and activates the transcription of multiple genes, including TNF-␣, IL-6, IL-8, iNOS, COX-2, etc. NIK ⫽ NF-B-inducing kinase.
Various inflammatory mediators, including IL-6 and TNF-␣, cause a cascade of events which lead to phosphorylation of IB␣ and its degradation. NF-B is then free to translocate to the nucleus, thereby initiating transcription for various PICs, iNOS and adhesion molecules [23]. In a model of endotoxemic shock, isohelenin, a sesquiterpene lactone, inhibited nucleus translocation of ⭈ NF-B without modification of IB␣ degradation. Interestingly, plasma NO was lower and survival was higher in isohelenin-treated animals [23].
Limiting PIC Interaction with Target Cells The administration of antibodies against PIC will neutralize them. Hence, the administration of an anti-TNF-␣ IgG to tumor-bearing rats decreased protein degradation rates in skeletal muscle, heart and liver compared with controls [24], and this action may be related to the fact that TNF-␣ antibody abolishes the increase in muscle ubiquitin gene expression observed in cancer-bearing animals [5]. However, blocking TNF-␣ with antibodies may prove detrimental: for example, in rats with caerulein-induced pancreatitis, blockade of TNF-␣ activity was found to increase edema formation in both the pulmonary and pancreatic microvascular beds [25]. 222
How Can We Modulate Cytokine Production and Action? Antibodies directed towards PIC receptors can also be used. Hence, in a model of lethal endotoxemia in mice, murine monoclonal IL-1 receptor antibody (IL-1rab) dramatically improved survival [26]; of note, both plasma IL-6 and IL-6 gene expression in the liver were decreased in IL-1rab-treated animals [26]. In a very elegant and interesting study, Tsujnaka et al. [27] showed that mice overexpressing IL-6 exhibit muscle atrophy and that treatment with IL-6 receptor antibody totally counteracted the effects of IL-6 on muscle weight and cathepsin activity. IL-1 receptor antagonist (IL-1ra) is produced by lymphocytes and phagocytes; its amino acid sequence is very similar to that of IL-1 and it blocks both type-I and type-II IL-1 receptors without exhibiting any agonist activity [28]. In an experimental model of endotoxemia in rats, it was shown that recombinant IL-1ra administration blunts the increase in LPS-mediated protein breakdown [29]. In a model of chronic abdominal sepsis, IL-1ra administration counteracted the sepsis-induced decrease in the rate of protein synthesis in muscle [30]. Another way to block PIC action is to block the cortisol receptor, since the interaction between PIC and cortisol at the target cell receptor level could potentiate the effects of both molecule types. However, RU 38486 failed to block IL-1-induced muscle proteolysis [31]. That said, the effect may be different from one PIC to another. For example, whereas both TNF-␣ and IL-1 administration inhibit amino acid uptake by rat muscle , only TNF-␣ action was counteracted by RU 38486 [32].
Blockage of PIC Message within Target Cells PICs activate PGE2 synthesis in skeletal muscle, which in turn may activate lysosome-mediated proteolysis. However, indomethacin, an inhibitor of cyclooxygenase, failed to counteract the augmentation of amino acid leg efflux in septic patients [33]. In contrast, perioperative ibuprofen administration in patients undergoing cholecystectomy blunted hyperglycemia and led to smaller changes in IL-6, ACTH and cortisol than in controls, whereas C-reactive protein was similar in the 2 groups [34].
Limiting PIC Production and Action with Anti-Inflammatory Cytokines The main anti-inflammatory cytokines are IL-4 and IL-10. As mentioned above, these cytokines are produced as a result of immune activation by a subpopulation of helper T cells (Th2) and inhibit the production of PICs by Th1 cells. Hence, IL-10 administration reduced lung dysfunction and mortality in a murine model of multiple organ dysfunction induced by zymosan 223
How Can We Modulate Cytokine Production and Action? treatment [35]. More specifically, LPS (from E. coli, 10 mg/kg) administration to rats decreased the force-frequency curves and half-relaxation time of the diaphragm muscle. When IL-10 was injected intraperitoneally 5 min after LPS, LPS-mediated effects were counteracted, and this was associated with ⭈ a significant decrease in NO production [36]. However, because of the broad immunosuppressive and anti-inflammatory properties of IL-10, its potential for resulting in secondary infections is a matter of concern: administration of IL-10 may suppress T-cell function and further increase the risk of opportunist infections in patients with preexisting immune suppression [37]. Moreover, blocking IL-10 production by using a chemical agent (e.g. AS101) can either improve or worsen survival of mice in a model of cecal ligation and puncture, depending on the time of administration [38].
Effects of Specific Nutrients Glutamine (GLN) There are several studies supporting evidence that GLN (in GLN-enriched parenteral nutrition) acts as an anti-inflammatory agent. For example, De Beaux et al. [39] showed that total parenteral nutrition enriched with GLN blunts the IL-8 response to stress in patients with pancreatitis. Given that GLN is a key substrate for activated immune cells, it is not surprising that in vitro GLN availability modulates cytokine production. However, diverse results have been obtained: e.g. an increase in IL-1 production [40] or a decrease in TNF-␣ production [41] by macrophages in response to increasing GLN levels in the medium. This variability is dependent on several factors. (1) The agent (e.g. concanavalin A, LPS) used to stimulate cells [42]. (2) The amount of arginine in the culture medium [41]. In fact, argininederived nitric oxide production can interfere with GLN action, since nitric oxide is also a potent modulator of NF-B. (3) The cell type: the production of anti-inflammatory cytokines by T-lymphocytes is highly stimulated by GLN, and in parallel the production of PICs by monocytes is significantly less stimulated [42]. Hence, GLN action is the result of modulation of anti-inflammatory/inflammatory balance. Interestingly, the GLN effect is not limited to immune cells: biopsies of human duodenum in culture release less IL-6 and IL-8 when incubated with GLN compared to GLN-free conditions [43]. In addition, when biopsies are incubated in the presence of IL-1, GLN dose-dependently decreased the IL-1-mediated increase in IL-6 and IL-8 production [44]. In Caco-2 cells, GLN blunted LPS-mediated IL-8 production and this effect was not related to NF-B [45]. These effects of GLN could be direct, depending on the regulation of cell volume for example, or indirect, related to the production of key molecules, 224
How Can We Modulate Cytokine Production and Action? e.g. inhibiting nitric oxide production [46] or activating glutathione synthesis. As a matter of fact, GLN forms the precursor pool of glutamate, which is involved in glutathione synthesis, the major antioxidant system. In postoperative patients, GLN supplementation of total parenteral nutrition counteracts the injury-related decrease in total and reduced glutathione [47]. Glutathione is actually a repressor of NF-B activation [28], the main transduction factor mediating PIC action. n-3 Fatty Acids Production of cytokines and their cellular effects is mediated and modulated by a number of compounds formed from the hydrolysis of membrane phospholipids by phospholipases A2 and C, and by activation of protein kinase C. Hence, alterations in phospholipid fatty acid composition will change the nature of substrates for the action of phospholipases [28], leading to the production of different eicosanoids with variable pro- or anti-inflammatory properties. n-3 polyunsaturated fatty acids (PUFAs), derived from linolenic acid, inhibit PIC production whereas n-6 PUFAs increase it [6]. This could explain why fish oil supplementation of the diet of patients with pancreatic cancer cachexia has been reported to stop weight loss [28]. In patients with sepsis, parenteral nutrition with fish oil resulted in lower PIC production by activated mononuclear leukocytes compared to cells from patients perfused with a conventional (n-6) lipid emulsion [48]. Others In vitro, but at dosages compatible with physiology, vitamin C inhibits the activation of NF-B by IL-1and TNF-␣ in endothelial cells. This action is related to an inhibition of phosphorylation and degradation of IB␣ [49]. Other antioxidants such as vitamin E, zinc and selenium are able to decrease PIC production by quenching reactive oxygen species production, and thus limiting the activation of transcription factors (e.g. NF-B, NF-IL-6) [28]. Conversely, the production of TNF-␣ by mice and IL-1 by rats in response to endotoxin is suppressed by desferrioxamine, an iron chelator [28].
Conclusion There are a number of experimental studies but none of the agents potentially able to modulate PIC production and/or action has proven its efficacy in regulating muscle hypercatabolism in stressed patients. In some cases, we are even faced with a yin-yang situation: whereas anti-glucocorticoid receptors blunt protein catabolism, glucocorticoids themselves repress PIC production. The association of cytokines, different dosages and timely administration may also influence the final result of the manipulation [50]. The fact that there are a number of players in the game acting as a network with 225
How Can We Modulate Cytokine Production and Action? numerous positive or negative regulatory loops further complicates the clarification of the usefulness of manipulations. Therefore, manipulating the inflammatory/immune response to stress presents a risk of serious unexpected and uncontrolled side effects, especially in the long-term. Therefore, further mechanistic studies are warranted before we can expect safe and efficient clinical application of the modulation of PIC production and action using drugs or antibodies. In contrast, the use of modulating nutrients (e.g. GLN and n-3 PUFAs) appears to be safe and efficient.
Acknowledgement We would like to thank Mrs. Solange Ngon for her expert secretarial assistance.
References 1 Hall-Angerås M, Angerås U, Zamir O, et al: Effect of the glucocorticoid receptor antagonist RU 38486 on muscle protein breakdown in sepsis. Surgery 1991;109:468–473. 2 Cavaillon JM: Pro- versus anti-inflammatory cytokines: Myth or reality. Cell Mol Biol (Noisyle-grand) 2001;47:695–702. 3 Li YP, Reid MP: NF-B mediates the protein loss induced by TNF-␣ in differentiated skeletal muscle myotubes. Am J Physiol 2000;279:R1165–R1170. 4 Price SR, Mitch WE: Mechanisms stimulating protein degradation to cause muscle atrophy. Curr Opin Clin Nutr Metab Care 1998;1:79–83. 5 Llovera M, Carbo N, Garcia-Martinez C, et al: Anti-TNF treatment reverts increased muscle ubiquitin gene expression in tumour-bearing rats. Biochem Biophys Res Commun 1996;221: 653–655. 6 Späte U, Schulze PC: Pro-inflammatory cytokines and skeletal muscle. Curr Opin Clin Nutr Metab Care 2004;8:265–269. 7 Tsujinaka T, Fujita J, Ebisuir C, et al: Interleukin-6 receptor antibody inhibits muscle atrophy and modulates proteolytic systems in interleukin-6 transgenic mice. J Clin Invest 1996;97: 244–249. 8 Van der Poll T, Sauerwein HP: Tumor necrosis factor-␣: Its role in the metabolic response in sepsis. Clin Sci 1993;84:247–256. 9 Cooney RN, Maish GO III, Gilpin T, et al: Mechanism of IL-1 induced inhibition of protein synthesis in skeletal muscle. Shock 1999;11:235–241. 10 Lang CH, Frost RM, Nairn AC, et al: TNF-␣ impairs heart and skeletal muscle protein synthesis by altering translation initiation. Am J Physiol 2002;282:E336–E347. 11 Grimble RF: Interaction between nutrients, pro-inflammatory cytokines and inflammation. Clin Sci 1996;91:121–130. 12 Bistrian BR, Schwartz J, Istfan NW: Cytokines, muscle proteolysis, and the catabolic response to infection and inflammation. Proc Soc Exp Biol Med 1992;200:220–223. 13 Hise ME, Compher C, Harlan L, et al: Inflammatory mediators in home parenteral nutrition patients (abstract). JPEN J Parenter Enteral Nutr 2004;28(suppl 1):S5. 14 Dunn AJ: Action de l’interleukine-1 sur les sécrétions hormonales. Nutr Clin Métabol 1990;4:75–82. 15 Holzheimer RG, Molloy RG, O’Riordain DS, et al: Long-term immunotherapeutic intervention with pentoxifylline in a mouse model of thermal injury and infection. J Trauma 1995;38: 757–762. 16 Vary T, Dardevet D, Grizard J, et al: Pentoxifylline improves insulin action limiting skeletal muscle catabolism after infection. J Endocrinol 1999;163:15–24.
226
How Can We Modulate Cytokine Production and Action? 17 Breuillé D, Farges MC, Rosé F, et al: Pentoxifylline decreases body weight loss and muscle protein wasting characteristics of sepsis. Am J Physiol 1993;265:E660–E666. 18 Demir S, Inal-Erden S: Pentoxifylline and N-acetylcysteine in hepatic ischemia/reperfusion injury. Clin Chim Acta 1998;275:127–135. 19 Thaloor D, Miller KJ, Gephart J, et al: Systemic administration of the NF-B inhibitor curcumin stimulates muscle regeneration after traumatic injury. Am J Physiol 1999;277: C320–C329. 20 Jobin C, Bradham CA, Russo MP: Curcumin blocks cytokine-mediated NF-B activation and proinflammatory gene expression by inhibiting inhibitory factor I-B kinase activity. J Immunol 1999;163:3474–3483. 21 Schow SR, Joly A: N-acetyl-leucinyl-leucinyl-norleucinal inhibits lipopolysaccharide-induced NFB activation and prevents TNF and IL-6 synthesis in vivo. Cell Immunol 1997;175: 199–2002. 22 Corsini E, Lucchi L, Binaglia M, et al: Cloricromene, a semi-synthetic coumarin derivative, inhibits tumor necrosis factor-␣ production at a pre-transcriptional level. Eur J Pharmacol 2001;418:231–237. 23 Sheehan M, Wong HR, Hake PW, Zingarelli B: Protective effects of isohelenin, an ihibitor of nuclear factor B, in endotoxemic shock in rats. J Endotoxin Res 2002;8:99–107. 24 Costelli P, Carbo N, Tessitore L, et al: Tumor necrosis factor-␣ mediates changes in tissue protein turnover in a rat cancer cachexia model. J Clin Invest 1993;92:2783–2789. 25 Guice KS, Oldham KT, Remick DG, et al: Anti-tumor necrosis factor antibody augments edema formation in caerulein-induced acute pancreatitis. J Surg Res 1991;51:495–499. 26 McNamara MJ, Norton JA, Nauta RJ, Alexander HR: Interleukin-1 receptor antibody (IL-1rab) protection and treatment against lethal endotoxemia in mice. J Surg Res 1993;54: 316–321. 27 Tsujinaka T, Fujita J, Ebisui C, et al: Interleukin 6 receptor antibody inhibits muscle atrophy and modulates proteolytic systems in interleukin 6 transgenic mice. J Clin Invest 1996;97: 244–249. 28 Grimble RF: Interaction between nutrients, pro-inflammatory cytokines and inflammation. Clin Sci 1996;91:121–130. 29 Zamir O, Hasselgren P-O, O’Brien W, et al: Muscle protein breakdown during endotoxemia in rats and after treatment with interleukin-1 receptor antagonist (IL-1ra). Ann Surg 1992; 216:381–387. 30 Cooney R, Owens E, Jurasinski C, et al: Interleukin-1 receptor antagonist prevents sepsisinduced inhibition of protein synthesis. Am J Physiol 1994;267:E636–E641. 31 Zamir O, Hasselgren P-O, Von Allmen D, Fischer JE: The effect of interleudin-1 and the glucocorticoid receptor blocker RU 38486 on total and myofibrillar protein breakdown in skeletal muscle. J Surg Res 1991;50:579–583. 32 Zamir O, Hasselgren P-O, James H, et al: Effect of tumor necrosis factor or interleukin-1 on muscle amino acid uptake and the role of glucocorticoids. Surg Gynecol Obstet 1993;177:27–32. 33 Wahren J, Von Euler C, Eriksson LS: Effect of endomethacin on leg exchange of amino acids in septic traumatism patients. Clin Nutr 1988;7:203–206. 34 Chambrier C, Chassard D, Bienvenu J, et al: Cytokine and hormonal changes after cholecystectomy. Effect of ibuprofen pretreatment. Ann Surg 1996;224:178–182. 35 Ferrer TJ, Webb JW, Wallace BH, et al: Interleudin-10 reduces morbidity and mortality in murine multiple organ dysfunction syndrome. J Surg Res 1998;77:157–164. 36 Taneda A, Shindoh C, Ohuchi Y, Shirato K: Protective effects of interleudin-10 on diaphragm muscle in a septic animal model. Tohoku J Exp Med 1998;185:45–54. 37 Huber TS, Gaines GC, Welborn MB III, et al: Anticytokine therapies for acute inflammation and the systemic inflammatory response syndrome: IL-10 and ischemia/reperfusion injury as a new paradigm. Shock 2000;13:425–434. 38 Kalechman Y, Gafter U, Gal R, et al: Anti-IL-10 therapy strategy using the immunomodulator AS101 in protecting mice from sepsis-induced death: Dependence on timing of immunomodulating intervention. J Immunol 2002;169:384–392. 39 De Beaux AC, O’Riordain MG, Ross JA, et al: Glutamine-supplemented total parenteral nutrition reduces blood mononuclear cell interleukin-8 release in severe acute pancreatitis. Nutrition 1998;14:261–265. 40 Wallace C, Keast D: Glutamine and macrophage function. Metabolism 1992;41:1016–1020.
227
How Can We Modulate Cytokine Production and Action? 41 Blanc MC, Moinard C, Béziel A, et al: Arginine and glutamine availability and macrophage functions in the obese insulin-resistant Zucker rat. J Cell Physiol 2004, in press. 42 Yaqoob J, Calder PC: Cytokine production by human peripheral blood mononuclear cells: Differential sensitivity to glutamine availability. Cytokine 1998;10:790–794. 43 Coëffier M, Miralles-Barrachina O, Le Pessot F, et al: Influence of glutamine on cytokine production by human gut in vitro. Cytokine 2001;13:148–154. 44 Coëffier M, Marion R, Ducrotté P, Déchelotte P: Modulating effect of glutamine on IL-1 induced cytokine production by human gut. Clin Nutr 2003;22:407–413. 45 Huang Y, Li N, Liboni K, Neu J: Glutamine decreases lipopolysaccharide-induced IL-8 production in Caco-2 cells through a non-NF-B p50 mechanism. Cytokine 2003;22:77–83. 46 Roth E: L-Arginine-nitric oxide metabolism. Glutamine: A new player in this metabolic game? Clin Nutr 1998;17:1–2. 47 Fläring UB, Rooyackers OE, Wernerman J, Hammarqvist F: Glutamine attenuates posttraumatic glutathione depletion in human muscle. Clin Sci 2003;104:275–282. 48 Mayer K, Gokorsch S, Fegbeutel C, et al: Parenteral nutrition with fish oil modulates cytokine response in patients with sepsis. Am J Respir Crit Care Med 2003;167:1321–1328. 49 Bowie AR, O’Neill LAS: Vitamin C inhibits NF-B activation by TNF via the activation of p38 mitogen-activated protein kinase. J Immunol 2000;165:7180–7188. 50 O’Suilleabhain C, O’Sullivan ST, Kelly JL, et al: Interleukin-12 treatment restores normal resistance to bacterial challenge after burn injury. Surgery 1996;120:290–296.
Discussion Dr. Bowling: You mentioned at the beginning that your talk was about modulating cytokines and then you said why should we do it. These are clearly mechanisms essential for homeostasis. Is it therefore going to be that by looking at all the various pathways and the various kind of stages that you can intervene, that if you block one pathway or another you are simply going to upregulate or downregulate the system elsewhere, and in terms of therapeutic potential, do you actually think there is going to be an awful lot to be achieved? Dr. Cynober: That is the point. If, for example, you are blocking tumor necrosis factor (TNF), you can expect to decrease some specific pathways in certain conditions, but then you must be aware than TNF in the short-term has the possibility of stimulating the production of other cytokines, for example IL-8. Up to now it is alright, it is good to limit or to block TNF production, but in the middle term there are some regulatory effects which in the end make TNF block the proinflammatory response and block the response of some other cytokines by some other cells. Therefore by doing that some other pathways are de-repressed and the reverse effect to that expected occurs. I am not really confident that in the near future, even using modern tools such as transgenic mice, we will have a definite answer about what is the target because, as I mentioned, it is a total network and when you are making something somewhere you have unexpected effects in another tissue or another organ. Dr. Elia: The situation may even be more complex than you have indicated for a variety of reasons, one of which is the genetic cytokine polymorphism. There is a growing list of associations between high producers and low producers or both pro- and anti-inflammatory cytokines, which means that there is a differential response to injury according to the individual’s genotype. Do you think there is information to show that there is a differential response to treatment according to the genotype of individuals? Dr. Cynober: Of course it has been demonstrated as you mentioned for example for TNF that perhaps 10–20% of the Caucasian population overexpress TNF production
228
How Can We Modulate Cytokine Production and Action? and this overexpression of TNF is clearly associated with poor prognosis in intensive care units. More recently this consideration has been extended to cardiovascular disease that in the end is an inflammatory chronic disease. As I mentioned before, I think that this is a possibility of the near future to screen the patients upon entrance to the intensive care unit to determine whether or not they are suitable for specific immunomodulatory treatment. This is the first point, but note that in some cases the polymorphism is protective. This has been demonstrated for IL-6 which protects the patients against malaria complications. Now the second part of the question: Grimble [1] in Southampton made some studies showing that the ability of n-3 fatty acids to avoid inflammatory properties may be dependent upon the gene polymorphism. Perhaps you know better than I do. Dr. Elia: I know that work very well. But of course if one is going to extend this, for example, to the intensive care unit or specific conditions, such as malaria, it is necessary to demonstrate that the results of the interventions differ either in a clinically beneficial or a negative way, according to the genetic polymorphism. I think that perhaps more work still needs to be done. Dr. Cynober: But if it was the sense of your question, I am not aware of similar results in the field of amino acid therapy, for example arginine or others. Dr. Lochs: Isn’t it the case that whatever we eat or feed a patient, we are interfering with cytokine production? We have a lot of information about specific substrates, if you eat long-chain triglycerides or unsaturated medium chain or if you feed more glycine or more glutamine. Are we still in the position that we can close our eyes to this information and feed a patient a so-called standard food or do we have to decide and say well, in this situation, let’s say sepsis, we do not give this and this because we know this has this and this consequence? In another situation, let’s say inflammatory bowel disease, we have good information that this and this composition is improving, and in alcoholic liver disease we have good information that this and this composition is improving. So is it still possible that in a hospital we just say well, there is some standard enteral food and let’s give it to everybody, and there is some standard composition of parenteral food, or do we have to make a more detailed recommendation based on these data you presented? Dr. Cynober: It is a problem of a disease-specific regimen. It was not my purpose to say that. If I take the best example which is glutamine, because it is safe and was demonstrated to be efficient in a number of situations, we can now say alright, glutamine must be added as a pharmacological drug, not only as a nutrient, to patients in intensive care units in certain postoperative situations, and why not in patients with sepsis. On the contrary we can say that to date nobody has shown interest in providing glutamine in Crohn’s disease, which is another form of inflammation. I think that we can start to make a number of statements and make recommendations for the use of such drugs or pharmaconutrients or specific nutrients. But we need evidence, we need data, not speculations. Just an example, and I will be very provocative, can you explain why in the same diet huge amounts of n-3 fatty acids, which may be immunosuppressive, and arginine, which is immunostimulative, are provided? It is provocative but we can argue like that, it is simply the period in the evolution of the patient that will indicate that such products may be useful or harmful, but we need data. Dr. Biesalski: I think we have to keep in mind what we are dealing with: nutritional pharmacology. Let me give you an example which fits cytokine production relatively well. Retinoic acid critically regulates the expression of the redox-sensitive transcription factors like AP1 and NF-B, but it would not make any sense to give vitamin A as a precursor because it is absolutely metabolically regulated. I think we have to keep in mind what we are doing, whether we give the active metabolite, for example with respect to vitamin D you have the same, they interact with the cytokine
229
How Can We Modulate Cytokine Production and Action? production, or we give a nutrition where we have the pre-drug or whatever. Another example is the interaction that Grimble [1] showed, this nice balance of antioxidants and NF-B activation. If you give -carotene as provitamin A it fulfills two roles as an antioxidant and probably as vitamin A precursor in another way because it circumvents the metabolic control of vitamin A via the liver. So I think we should also address the question what is nutrition on the one hand and what is probably circumventing metabolic pathways and can be useful to interact with processes like this. I have one question: do you have any evidence or know of a clinical study with cytokine production and retinoic acid? It would make sense from basic knowledge and from in vitro animal experience. Dr. Cynober: I confess my ignorance. It is out of my major field of interest. I don’t know. Dr. Thomas: There are a number of drugs that are becoming available that are sort of anti-cytokine in a general sense, for example in inflammatory rheumatologic diseases such as cardiac cachexia. Dr. Cynober: Which drug? Dr. Thomas: The one that is available for cardiac cachexia, I think it is Inalopret, it is only available in Europe, not in the USA. Can you comment on the empirical data showing that the use of some of these anti-cytokine drugs have an effect to produce weight gain and improve nutritional parameters in these patients when they use them? Dr. Cynober: It depends on the type of drug. For example you mentioned on the first day of the meeting the possible use of statin to modulate cytokine production, and this has nothing to do directly with cholesterol. In fact, just as an example, it has an indirect effect because statin increases the activity of nitric oxide synthesis in the endothelium, which in turn allows generation of a higher quantity of nitric oxide and in this given condition it is very helpful because it will decrease the amount of oxidized low-density lipoprotein. Therefore that is a clear example of pharmacological nutrition, and you can provide both arginine and statin. We made the experiment in hypercholesterolemic rabbits, and by associating the statin and the arginine there is a dramatic effect. I am absolutely not certain that the therapy will be useful to modulate muscle protein metabolism and so on. In my opinion the larger the target, the more you will be efficient and the more side effects you will have; the more the target is limited, the more you will have specific effects and few side effects. Dr. Ockenga: Ten years ago we did a study in HIV patients with wasting and tried to use ketotifen as anti-TNF treatment [2], and indeed in the pilot study we showed a decrease in TNF released and stimulated with lipopolysaccharide, and we saw an effect on weight gain in this small clinical study. There are some other studies with thalidomide as well as pentoxifylline in AIDS patients [2] and a small clinical effect was seen. But I see some danger in using anti-TNF. Probably most of us know the study on septic patients in the intensive care unit using a TNF antibody [3]. There are 2 large acute studies showing an increased mortality using TNF antibody in this specific clinical situation [3]. So probably we have only a very limited therapeutic window we can use and we have to choose between the side effects and the effects we have on the nutrition side. Dr. Cynober: This reminds me of another study by Takala et al. [4] about the administration of human growth hormone to patients in the intensive care unit. This large multicenter trial had to be stopped because the mortality was higher in the human growth hormone-treated group. Now nutrition was rather hypocaloric and hypoproteic, and a possible explanation is that blocking protein catabolism and efflux of amino acids from the muscle completely starves the splanchnic tissues, i.e. the liver and intestine. This is another aspect. Again should we block cytokine production?
230
How Can We Modulate Cytokine Production and Action? Dr. Labadarios: Thank you for your welcome comments, cautions and the philosophy of what we should or should not do. Of particular interest to me are all these new compounds that you mentioned because there is this a shift from the orthodox to a complimentary type of medicine containing plant extracts and similar compounds. In the experiments that you conducted and apart from the measurements that you showed, did you actually include, anything in terms of clinical outcomes? Let me tell you why I am asking this question. Some 10 years ago there was an article from Japan [5] that made an impression on me in the sense that they did something very simple. They made a surgical incision in rats and then fed them with an n-3-rich diet. There was no difference in the healing time as best as I can remember, but when they actually measured the strength of the wound it was significantly diminished both in terms of elasticity and strength. So apart from the things that you mentioned, are there any other such data in relation to what you presented? Dr. Cynober: Do you mean about exotic medical products or in general? There are a lot of data in the literature and most of them are of course experimental studies for various reasons. Specifically to comment on the study on the n-3 fatty acids in wound healing in rats. I am not surprised at all because these lipids may be immunosuppressive. Now to have wound healing you need the migration of immune cells into the wound and also the release of proinflammatory cytokines, but not only, in order to synthesize new collagen and so on. I am not saying we must not use n-3 fatty acids, I am saying that probably the use of these various therapeutic agents is time related, and if you make some manipulations at not the most adequate time you can achieve a true disaster. Now there are plenty of studies with burn rats, with lipopolysaccharidetreated rats, looking at different pharmacological effects, I just made a selection. Dr. Roessle: I wonder whether you would agree to adding the amino acid cysteine to the list of logical ingredients? As glutamine, it is a precursor of glutathione and, in the same animal model as you have shown for the data on pentoxifylline, it has been shown to downregulate TNF-␣ and IL-6 production without shutting them off. It also shows improved long-term outcome on sepsis, and it would be a more logical nutritional ingredient which we might consider to use in the future. Dr. Cynober: I totally agree, and if you look at the abstract book, cysteine is mentioned together with glycine. Of course there are a lot of studies [6], especially using the stable precursor form acetylcysteine, which demonstrate that you can decrease the activation of NF-B by mainly stimulating glutathione synthesis, but perhaps not only. Dr. Lesourd: Don’t you think that the side effects you observed in the assays are probably related to some concept that was really wrong? We tried to block some reactions and the problem is probably not to block it but to decrease it in order for the body to go back to its own equilibrium. Probably we have been going too far in the treatment we have given in the different assays. What do you think? Dr. Cynober: Yes, it makes sense. I will give another example of how with nutritional manipulation you can achieve something or exactly the reverse. If, for example, you use rats and take the liver and store it at 4⬚C and then reperfuse the stored liver, you have a certain degree of ischemia reperfusion injury. Now if you starve the rats for 2 days before taking the liver you have a huge ischemia reperfusion injury simply because you have decreased the antioxidant defenses. Now, if you starve the rats for 8 days you have no reperfusion injury simply because you have depleted the Kupfer cells and there is no longer a reaction (Charrueau, personal commun.). Dr. Lesourd: So we are probably doing the same when we are refeeding the patients. We would like to continue the same regimen any time and that is probably not the right way to do it. Dr. Cynober: I agree.
231
How Can We Modulate Cytokine Production and Action? Dr. Elia: A growing number of amino acids have been proposed for immunonutrition and you mentioned some of them. My question is rather the opposite: are there any amino acids that you would avoid or reduce for a treatment? Dr. Cynober: For cytokine activation and things like that? Dr. Elia: Yes, or in terms of either laboratory or clinical studies. Dr. Cynober: We can talk all night about protein requirement in the trauma situation and competition between amino acids. To answer your question, there is an amino acid I have almost not discussed during this presentation and that is arginine because there is a close relationship between arginine provided by food intake and the capacity to synthesize nitric oxide. This has been well demonstrated by several investigators [7]: the low part of de novo synthesis in the ability to generate nitric oxide. Of course in certain situations of overexpression of proinflammatory cytokines, overexpression of inducible nitric oxide synthesis, in patients with multiple organ failure, unstable hemodynamics, I think that large amounts of arginine may be detrimental. But in my opinion these are extreme situations, but it does not mean that I agree with Heyland et al. [8] who probably extrapolate too much data from the literature. But in certain situations, to answer to your question, yes, I think that extra arginine may be detrimental, overstimulating the immune system. Dr. Biesalski: Can I make a short comment on that because you mentioned vitamin C and the Takala et al. [4] study. I think burn injury is an absolute contraindication for arginine because there is a massive formation of superoxide anions and if nitric oxide production is increased you get peroxynitrite and neutralization of proteins. The action of vitamin C lowers nitric oxide production and is effective in preventing edema formation. So additional arginine might be not beneficial. Dr. Cynober: The comment on vitamin C is well taken. With regard to arginine I don’t know because specifically in burned rats there is no morbidity or mortality related to extra arginine provision, which is contrary to data obtained in peritonitis.
References 1 Grimble RF: Interaction between nutrients, pro-inflammatory cytokines and inflammation. Clin Sci (Lond) 1996;91:121–130. 2 Ockenga J, Rohde F, Suttmann U, et al: Ketotifen in HIV-infected patients: Effects on body weight and release of TNF-alpha. Eur J Clin Pharmacol 1996;50:167–170. 3 Fisher CJ Jr, Agosti JM, Opal SM, et al: Treatment of septic shock with the tumor necrosis factor receptor: Fc fusion protein. The Soluble TNF Receptor Sepsis Study Group. N Engl J Med 1996;334:1697–1702. 4 Takala J, Ruokonen E, Webster NR, et al: Increased mortality associated with growth hormone treatment in critically ill adults. N Engl J Med 1999;341:785–792. 5 Albina JE, Gladden P, Walsh WR: Detrimental effects of an omega-3 fatty acid-enriched diet on wound healing. J Parenter Enteral Nutr 1993;17:519–520. 6 Obled C, Papet I, Breuille D: Sulfur-containing amino acids and glutathione in diseases; in Cynober L (ed): Metabolic and Therapeutic Aspects of Amino Acids in Clinical Nutrition. Boca Raton, CRC Press, 2004, pp 667–688. 7 Morris SM: Regulation of arginine availability and its impact on NO synthesis; in Ignarro LJ (ed): Nitric Oxide: Biology and Pathobiology. San Diego, Academic Press, 2000, pp 187–197. 8 Heyland DK, Novak F, Drover JW, et al: Should immunonutrition become routine in critically ill patients? A systematic review of the evidence. JAMA 2001;286:944–953.
232
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, pp 233–247, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
How Can We Improve Functional Outcomes? Marinos Elia Institute of Human Nutrition, University of Southampton, Southampton General Hospital, Southampton, UK
Which Outcomes? Health has multiple dimensions. In a broad sense, health can be considered in terms of physical, mental and social well-being, but this may be difficult to measure, particularly in whole populations. The earliest population indices of health were based on readily available and unambiguous indicators, such as mortality; and these were applied to particular age groups, such as the first year of life (infant mortality). The legal requirement in many societies to record mortality meant that generally complete data could be obtained. However, although such information has been of considerable value, this may change over time as societies evolve. Health indicators are often specifically chosen to reflect current problems within societies, especially those requiring improvement. Mortality statistics may reveal new layers of concerns, which may lead to the development of new indicators of health outcome to reflect these new concerns. For example, in low income countries, where infant mortality is high (or mortality from specific treatable diseases are high) appropriate policies may produce improvements in outcome relatively easily. In developed countries, where this mortality has already been reduced to very low levels, the cost of reducing it to an even lower level may be disproportionately high and unacceptable to policy makers. Therefore, new policies are established so that resources can be redirected to address other more important health problems in the general population. For example, as survival from a particular condition or groups of conditions improves, a range of persisting disabilities may develop, e.g. increasing life expectancy in countries with well-developed economies is associated with greater disability and dependency in older life, and a poorer quality of life. As a consequence, indicators of these health outcomes have become increasingly more prominent in 233
How Can We Improve Functional Outcomes? developed countries. Furthermore, new indicators may not only be used as markers of health, but they may also become incorporated into clinical and public health policies, and even concepts and definitions of health. Societies in different parts of the world are in various stages of economic and industrial development, which means that health outcomes that are a high priority in one society may be much less important in another society. Global, national, regional, local, and individual perspectives of health outcomes may vary considerably, with consequent variation in the choice of health outcomes. A global perspective on nutritional problems [1] reveals that iron deficiency anemia (⬃2,000 million people worldwide) and iodine deficiency (⬃740 million) affects more individuals than protein-energy malnutrition (⬃150 million), but the latter has a greater impact on mortality. The trends also differ, with iodine deficiency decreasing, obesity rapidly increasing, and protein-energy malnutrition showing relatively little overall change in absolute prevalence. Therefore, health outcome measures used by the WHO to assess global health may be different from those used in particular countries where, for example, iodine deficiency is a trivial public health problem. Another example concerns malnutrition and mortality due to AIDS, which is continuing to increase in many parts of the developing world, and decreasing in developed parts of the world [2]. Health inequalities [3], including those involving malnutrition, also occur within developed countries. For example, a secondary analysis of the National Diet and Nutrition Survey (UK) on individuals aged 65 years and over [4] revealed that the risk of malnutrition (medium and high risk, using criteria similar to those used in the ‘Malnutrition Universal Screening Tool’ [5]) was greater in the north of England (19.4%) than central (12.3%) and south England (11.3%). There is a north–south divide with respect to other health inequalities which the government is keen to eliminate [6]. The commitment to reduce health inequalities (nutritional and non-nutritional) within and between countries varies according to geography. Individual health professionals may also have different perspectives about outcome measures. A clinical perspective involving treatment of individual patients with established disease may be very different from a public health perspective which involves prevention of disease in the entire population. Economic issues are relevant to both clinical and public health outcome measures. However, improved survival as a result of medical advances may leave large numbers of people debilitated and dependent on the community for financial and social support. Furthermore, in clinical practice improvement in economic outcome in one health care setting does not necessarily mean improved outcomes in another setting. For example, in some health care systems emphasis is placed on the length of hospital stay because of its high cost. A reduction in the length of hospital stay may be interpreted as a success for the hospital setting, but this could be detrimental to the community if hospitalized patients are discharged prematurely and ill advisedly, 234
How Can We Improve Functional Outcomes? Cumulative environmental stress (including nutritional stress)
Functional capability
Younger life Growth and development
Intermediate (adult) life Establish/maintain peak function
Acute illness
Older life Prevent disability and dependency
Variability in function
Modifying the environment
Threshold of disability
Age
Fig. 1. A lifecourse perspective of functional outcomes (see text for details).
so that a heavier burden of care and cost fall on the community and to society as a whole. At the same time, policies in the community may influence the type of patients admitted to hospital. Unified health and economic policies across all care settings may yield different results from those involving individual health care settings. Here I will consider both public health strategies during the lifecourse, which may have cumulative beneficial effects jin old age, and to clinical strategies which deal with acute and chronic conditions once they have arisen. A nutritional focus will be maintained throughout.
Interventions during the Lifecourse The increase in mass and change in structure of tissues during growth are associated with improved functional capacity, such as the strength of muscular contraction, the strength of bones, ability to work, and development of reproductive competence. Such functional capabilities typically peak during early adult life and then gradually decline [7, 8] (fig. 1). If function falls below the threshold of disability, quality of life decreases and the person may become dependent on carers, either in their own home or long care facilities such as residential care homes. The fall below the threshold of disability may be permanent, and develop either acutely, for example following a stroke, or more slowly during the course of a chronic debilitating illness such as arthritis, motor neuron disease or Parkinson’s disease. The fall below this threshold may also occur transiently during the course of a treatable acute or subacute 235
How Can We Improve Functional Outcomes? illness. The period of time spent below the threshold of disability is greater for an individual already close to the threshold of disability than for a healthier fitter person with better initial functional capacity. No time may be spent below the threshold of disability for a fit young adult with a high initial functional capability (fig. 1). Therefore, one way of improving outcome is to implement policies that ensure that functional capability is maintained at a high level, well above the disability threshold. This can be achieved by: (1) establishing a high peak functional capacity in young adult life; (2) reducing the rate of loss of functional capacity after the peak level has been reached, and (3) lowering the threshold of disability through modification of the environment. There is biological variability in functional capability, which is partly genetic and partly environmental, with a contribution of nutritional factors to the environmental component. These nutritional factors influence the first two strategies. The example given below, involving bone mineralization, illustrates how all these three components might be modified during the lifecourse to prevent osteoporotic fractures and disability in later life. However, the impact of osteoporotic fractures on clinical outcome also depends on equitable access to quality health care. Therefore, following the lifecourse example, which is described below, consideration is given about how nutritional interventions in patients with fractures can affect clinical outcome during a journey of clinical care. This journey frequently spans more than one health care setting, and therefore an integrated strategy that takes all these settings into consideration would have obvious advantages. A Lifecourse Example Although the strength of bone and the likelihood of fracture depends on its microarchitecture, it is also strongly influenced by the extent of mineralization. Individuals with a low bone mineral (osteoporosis) are more likely to develop fractures either spontaneously or following accidents. Reducing the frequency and severity of trauma in the elderly is one way of avoiding fracture development. Another strategy is to slow down the rate of bone mineral loss in adult life. Yet another strategy, which has received less attention, is to increase the peak bone mass, so that loss of mineral in later life would take longer to produce osteoporosis and to predispose to fractures. Fetus. One of the earliest environmental factors that may influence bone mineralization in adults is the intrauterine environment [9–11]. The growth and mineral content of bones are related to birth weight, which is influenced by the nutritional status of the pregnant mother as well as the function of the placenta which delivers nutrients to the fetus. Low birth weight infants become shorter adults. In the case of monozygotic twins [12, 13], in which the same relationship applies, it is reasonable to suggest that the effect is due to environmental and not genetic factors. Of further interest is the relationship between birth weight and adult bone mineral content, which remains 236
How Can We Improve Functional Outcomes? statistically significant after adjustment for adult height and a variety of other variables, such as age, sex, cigarette smoking, intake of alcohol and calcium and physical inactivity [9]. These observations may be mediated through growth hormone/insulin like growth factor and the hypothalamic-pituitary-adrenal axes [10]. Twin studies have also shown an association between birth weight and intestinal calcium absorption in adults (using strontium as a marker), which is partly explained by a lower circulating 1,25-hydroxy-vitamin D concentration [14]. It has been suggested that the intrauterine environment programs the postnatal accretion of bone mineral, potentially influencing the peak bone mineral mass during early adult life. It has also been suggested that optimization of maternal nutrition during intrauterine growth is a potential strategy for preventing osteoporotic fractures in the future generation. However, direct evidence that interventions before or during pregnancy can alter peak bone mass and reduce the risk of fractures in the human offspring is lacking. This is an area that is currently being investigated. Childhood and Adolescence. The rate and extent of bone mineral accretion during childhood and adolescence, as well as loss during most of adult life depends on many factors including physical activity and intake of calcium and vitamin D. Adults. A series of comprehensive reviews and meta-analyses have been carried out to establish the role of nutrients and drugs in reducing the rate of bone demineralization and risk of fractures. Administration of calcium and vitamin D in combination to individuals with postmenopausal osteoporosis for a period of 1.5 years decreased femoral neck bone mineral density relative to a placebo [15]. The effect is less than that produced by other treatments, such as drug therapy for 3 years (e.g. risendrionate or etidronate), but hormone replacement therapy was greater. Calcium and vitamin D in combination were also found to reduce the risk of hip fractures. Since bone mineral density is an important determinant of bone strength, accounting for more than 60% of the variance in breaking strength according to some reports [16, 17], changes in bone mineral density is frequently used as an outcome variable of intervention studies. However, fluoride supplementation appears to produce a major increase in bone mineral density without an effect on the incidence of hip fracture [15], possibly because of effects on the microarchitecture of bone. Other reviews have concluded that large supplements of calcium are efficacious in populations with a low intake of calcium, and vitamin D with calcium is efficacious in deficient populations [18]. Yet another meta-analysis on the use of vitamin D, with or without calcium, for a period of at least 1 year (9 studies, 1,130 postmenopausal women) concluded that there was a reduction in the risk of vertebral fractures and possibly in non-vertebral fractures [19]. Another meta-analysis in postmenopausal women concluded that calcium supplementation alone has a small positive effect on bone mineral density, and a trend (which was not significant) in reducing vertebral fractures [20]. There have also been summaries of several 237
How Can We Improve Functional Outcomes? meta-analyses or systematic reviews. One such article [20], which summarized 7 systematic reviews, reported that calcium and vitamin D have positive effects on bone mineral density of the hip, forearm and total body, and that vitamin D significantly reduced the incidence of vertebral fractures. In summary, a variety of nutritional and non-nutritional interventions could potentially be used (some are already being used) to prevent osteoporotic fractures. These mainly aim to increase peak bone mineral mass and decrease the demineralization after the peak mass has been achieved (although some also aim to increase bone turnover). However, controversy still exists about the role of routine screening for osteopenia, the long-term administration of drugs that affect bone mineral density, and the optimal time of initiating prescription. Few women have major fractures before the age of 65, probably because they do not tend to fall. Therefore programs that aim to reduce the frequency of falls later in life are important (see below). In the meantime a lifecourse perspective which involves optimizing pregnancy outcomes, implementing a healthy diet, exposure to sunlight, and increasing physical activity in those who are inactive should be strongly encouraged. Decreasing the Disability Threshold An alternative strategy to improve outcome is to reduce the disability threshold through environmental modification. For example, introduction of special cutlery, plates, and cups to patients with arthritis, fractures/deformities, or certain neuromuscular problems, who would otherwise depend on others to feed them, may allow them to become independent and increase their self-esteem. Providing supportive rails may allow patients who are unstable on their feet to get around the house and prevent falls before, during or after showering or bathing. Removal of slippery, uneven surfaces and providing adequate light in certain areas (either inside or outside the house) can also help lower the disability threshold. Such modifications may also help to reduce falls and the risk of fractures in vulnerable individuals (table 1) [21].
Intervention by Health Care Services and Their Effects on the Patient Journey Community Interventions to Prevent Falls and Bone Fractures Here I deal with the theme of bone fractures in the elderly by considering interventions that may reduce the frequency of falls and fractures and damage produced by such events. Falls in the elderly are an important cause of morbidity, such as fractures (they may also lead to a variety of other injuries and to hypothermia). Fractures of the proximal femur (hip fractures) in the elderly also cause a high mortality, ranging from 12 to 37% in the first year and 11% during the first few months after fracture [22]. Interventions for reducing falls in the community can be divided into those that involve the 238
How Can We Improve Functional Outcomes? Table 1. Interventions for reducing falls and their outcome in the community Environment Hazards in the homea e.g. Irregular slippery surface No supportive hand rails Poor lighting Other features detrimental to the activities of daily living Hazards outside the home Health Muscle strength/balance retraininga Optimizing drug therapya Correction of poor sight/other disabilitiesa Malnutrition e.g. Poor muscle strength/coordination Loss of cushioning effect of subcutaneous fat Low bone mineral mass Environment ⫹ healtha aEvidence
of benefit from meta-analyses.
health of individuals/populations, the environment, or a combination of both. Table 1 provides examples of the types of interventions that have been assessed, and identifies those in which a significant benefit has been reported, often through meta-analyses. Muscle strength and balance-retraining programs, home hazard modification and drug therapy modification, such as withdrawal of psychotropic drugs, are the main interventions that have shown benefit. Malnutrition could potentially predispose to falls and worsen their outcome in a number of ways: poor strength and inability to adequately compensate for an abnormal movements, such as those induced by ‘accidents’; reduction in the amount of subcutaneous fat, resulting in a smaller cushioning effect, which increases the risk of fracture, such as hip fracture; osteoporosis resulting in weaker bones which are more likely to fracture. Despite these possibilities, two types of nutritional interventions have failed to demonstrate a reduction in the number of falls. The first type involved vitamin D supplementation, with or without calcium [for a meta-analysis of 5 such studies, see 21]. The second type, involving administration of oral energy-dense multinutrient supplementation over a period of 12 weeks, was examined by only one small randomized controlled trial involving 50 frail elderly women (low body mass index or a history of weight loss). However, the number of subjects was small and therefore larger studies are required [23]. Furthermore, since multiple factors predispose to falls and affect their outcome, interactions between these factors, including nutritional factors, could be important. Multifactorial and multidisciplinary intervention studies, 239
How Can We Improve Functional Outcomes? in which health and environmental hazard modification have been combined, have generally yielded positive results [21]. However, such studies with complex designs have not included a nutritional intervention. Another general finding is that studies involving unselected populations (chosen on the basis of age or risk factors) have been found to be less likely to show benefit than those that have been selected on the basis of a previous history of falls [21]. A larger number of subjects would normally be required if unselected populations are to be studied. By analogy, nutritional interventions are more likely to demonstrate benefit in patients who are already malnourished or likely to become malnourished than those who are not malnourished and unlikely to become malnourished. Hospital interventions with Nutrition: Effect on Longer Term Outcomes Several studies have been undertaken to examine the effect of nutritional supplementation in hospitalized patients with fractured femurs. Four randomized control trials also examined outcome at 6 or 7 months after the patients were discharged from the acute hospital. All these involved multinutrient supplements containing protein energy, trace elements and vitamins. Two of these trials reported a better overall clinical course [24, 25]. Another reported a significantly lower mortality at 6 months [26] and the fourth reported a higher femoral bone mineral density compared to control patients [27]. These studies suggest that nutritional treatment started in an acute hospital could have longer term effects, which are apparent after the transfer of patients to a rehabilitation hospital or the community. In one of the studies [27] in which supplementation was continued for 6 months, it is obviously not possible to ascribe the benefit entirely to the early period of supplementation. Targeting Nutritional Interventions in the Community The aim of nutritional support is to treat patient groups that are likely to show benefit and not to treat those that are unlikely to show benefit. In the community, response to oral nutritional supplements was found to depend on the initial body mass index (BMI). Randomized controlled trials in which the mean BMI was ⬍20 kg/m2 showed the following differences with those in which the BMI was ⬎20 kg/m2, involving a variety of patient groups including orthopedic patients [28]. Suppression of food intake was less in those with a BMI of ⬍20 kg/m2 so that the net effect was that 83% of the supplemental energy added to the food intake (11 studies) compared to only 45% in those with a BMI of ⬎20 kg/m2 (10 studies; Fisher’s exact test, p ⬍ 0.024). Changes in body weight in patients receiving oral nutritional supplements in randomized controlled trials was greater in those with a mean BMI of ⬍20 kg/m2 (⫹4.6%, 12 trials) than those with a BMI of ⬎20 kg/m2 (1.74%, 11 trials), although the difference did not quite reach statistical significance 240
How Can We Improve Functional Outcomes? (p ⬍ 0.054, unpaired t test). The difference between supplemented and unsupplemented patients was also greater in those in with a BMI of ⬍20 kg/m2 (3.1 vs. 1.3%; p ⬍0.09). Improvements in body function were more likely to occur in patient groups with a BMI of ⬍20 kg/m2 than ⬎20 kg/m2. Functional benefits were found in 13 of 16 (81%) of randomized controlled trials when the BMI was ⬍20 kg/m2 (81%), and in only 2 of 9 (22%) in those with a mean BMI of ⬎20 kg/m2 (Fisher’s exact test, p ⬍ 0.009). Similarly, in non-randomized trials functional improvements were likely to occur in those with a BMI of ⬍20 kg/m2 (12/16 trials; 75%). The mean changes in weight were variable with no overall difference between those showing benefit and those who did not. However, in underweight patients with chronic obstructive airways disease and in the elderly, significant functional benefits were only found in trials in which more than 2 kg in weight had been gained (⬃4% body weight).
Interventions to Improve Equity and Quality of Healthcare Services Health inequalities and inequitable access to health care has been of increasing concern to national and international agencies [29] and governments of both high and low income countries. Although many examples exist, the following concerns data on home artificial nutritional support produced by the British Artificial Nutrition Survey, which is the largest ongoing survey of its kind worldwide. Inequity in Home Artificial Nutrition Support A survey within the Trent Region (population 5.5. million) revealed that there was an up to threefold variation in the point prevalence of home tube feeding [30]. A larger survey in 1998 involving 38 health authorities (total population 17.3 million or ⬃0.46 million/health authority) revealed that the point prevalence of home parenteral nutrition varied from zero/million (several health authorities) to about 30/million [31]. Since this was unlikely to occur by chance as a result of random variation (p ⬍ 0.001; Poisson distribution), at least two alternative possibilities exist: large geographic variation in the conditions requiring home artificial nutrition, and inequity in access to nutritional support. Although there are some geographic variations in the prevalence of conditions requiring nutritional support, it is hard to envisage that they are responsible for the magnitude of observed variations. It is more likely that attitudes to nutritional support vary across geographic locations within the UK. Indeed there is evidence that regions close to the two national intestinal failure units are more likely to receive home artificial nutritional support than those who are not. Furthermore, in Scotland, where there is an organized national network for home parenteral nutrition, the prevalence of 241
How Can We Improve Functional Outcomes? home parenteral nutrition (⬃14/million) is generally greater than in England, but similar to the northwest region of England [30], which is served by one of the national intestinal failure units. Selection of patients for home nutritional support is a complex issue and depends on the availability of expertise (e.g. only about 40% of British hospitals have a nutrition support team), ethical issues, and nutritional training and education of health professionals, which needs improvement. Quality of Home Nutritional Support Annual surveys by the British Artificial Nutrition Survey involving reports from over 200 centers/year suggests that up to a fifth of the centers do not allocate sufficient time to training the patient and caregiver for home enteral tube feeding before discharge from hospital. Furthermore, patient records are not kept in 24–31% of centers. Financial arrangements are also a problem, although these have tended to decrease with time. Surprisingly, patients are not followed up in 13–20% of centers and this could have detrimental effects on outcome. A possible example concerns patients receiving home enteral tube feeding following a cerebrovascular accident [32]. At 1 year, 13% had returned to full oral feeding. One of the concerns is that if patients are instructed to continue on tube feeding as the sole source of nutritional support, without follow-up to check whether swallowing function has improved (a not uncommon situation), patients may be receiving tube feeding unnecessarily. Therefore, simple changes in the home care service could produce improvements in clinical outcome.
Interventions to Provide Better Nutrition Education and Training to Health Professionals Much malnutrition in developed countries is under-recognized and undertreated in hospitals (inpatients and outpatients), nursing homes and in the community [5]. Without recognition it is difficult to provide effective treatment. Among the factors responsible for poor recognition are the following: diffuseness of responsibility; lack of an integrated infrastructure for dealing with nutritional problems within and between different health care settings; lack of consistent criteria to identify and treat malnutrition, and inadequate education and training. Since there is a nutritional component to most diseases, education should benefit from multi-professional input. In the UK, the Nutrition Task Force Project Team on Nutrition Education and Training [33] has provided specific guidelines for areas of nutritional competence that should be acquired by various health professionals, including doctors, nurses, dentists, pharmacists, physiotherapists, and speech and language therapists. Health professionals should appreciate the importance of nutrition in both prevention and treatment of disease. Professional organizations, 242
How Can We Improve Functional Outcomes? the Department of Health and cross-governmental agencies, all have a role to play. Each initiative should be undertaken as part of a coordinated strategy. References 1 2 3 4 5
6 7 8 9 10 11
12 13 14 15 16 17 18 19
20
21 22 23
24 25
World Health Organization: Nutrition. http://www.who.who.int/nut/ 2003. Weiss R: AIDS: Unbeatable 20 years on. Lancet 2001;357:2073–2074. Graham H: Understanding Health Inequalities. Buckingham, Open University Press, 2000. Finch S, Doyle W, Lowe C, et al: National Diet and Nutrition Survey: People aged 65 years and over. London, Stationery Office, 1998. Elia M (ed): The ‘MUST’ Report. Nutritional Screening for Adults: A Multidisciplinary Responsibility. Development and Use of the ‘Malnutrition Universal Screening Tool’ (‘MUST’) for Adults. A Report by the Malnutrition Advisory Group of the British Association for Parenteral and Enteral Nutrition. Reddich, BAPEN, 2003, p 127. Department of Health: Tackling Health Inequalities: A Programme for Action. London, Department of Health Publications, 2003. World Health Organization: Active Aging: A Policy Framework. Geneva, WHO, 2002. Kalachea A, Kickbusch I: A Global Strategy for Healthy Ageing. Geneva, WHO, 1997, pp 4–5. Gale CR, Martyn CN, Kellingray S, et al: Intrauterine programming and adult body composition. J Clin Endocrinol Metab 2001;86:267–272. Javaid MK, Cooper C: Prenatal and childhood influences on osteoporosis. Best Pract Res Endocrinol Metab 2002;16:349–367. Godfrey K, Walker-Bone K, Robinson S, et al: Neonatal bone mass: Influence of parental birthweight, maternal smoking, body composition, and activity during pregnancy. J Bone Miner Res 2001;16:1694–1703. Loos RFJ, Beunen G, Fagard R, et al: Birth weight and body composition in young adult men – A prospective twin study. Int J Obes 2000;25:1537–1545. Allison DB, Paultre F, Heymsfield SB, Pi-Sunyer FX: Is the intra-uterine period really a critical period for the development of adiposity? Int J Obes 1995;19:397–402. Arden NK, Major P, Poole JR, et al: Size at birth, adult intestinal calcium absorption and 1,25(OH) (2) vitamin D. Q J Med 2002;95:15–21. Hauselmann HJ, Rizzoli R: A comprehensive review of treatments for postmenopausal osteoporosis. Osteoporos Int 2003;14:2–12. Turner CH, Burr DB: basic biomechanical measurements of bone: A tutorial. Bone 1993; 14:595–608. Amman P, Rizzoli R: Bone strength and its determinants. Osteoporos Int 2003;14(suppl 3): S12–S18. Vallecillo G, Diez A, Carbonell J, Gonzalez Marcias J: Treatment of osteoporosis with calcium and vitamin D. Systematic review. Med Clin (Barc) 2000;115:46–51. Papadimitropoulos E, Wells G, Shea B, et al: Meta-analyses of therapies for postmenopausal osteoporosis. VIII: Meta-analysis of the efficacy of vitamin D treatment in preventing osteoporosis in post-menopausal women. Endocr Rev 2002;23:560–569. Shea B, Wells G, Cranney A, et al: Meta-analyses of therapies for postmenopausal osteoporosis. VII. Meta-analysis of calcium supplementation for the prevention of postmenopausal osteoporosis. Endocr Rev 2002;23:552–559. Gillespie LD, Gillespie WJ, Robertson MC, et al: Interventions for preventing falls in elderly people. Cochrane Library. No 694. Chichester, Wiley, 2001. Lyons AR: Clinical outcomes and treatment of hip fractures. Am J Med 1997;103:51S–64S. Gray-Donald K, Payette H, Boutier V: Randomized clinical trial of nutritional supplementation shows little effect on functional status among free living elderly. J Nutr 1995;125: 2965–2971. Delmi M, Rapin C-H, Bengoa J-M, et al: Dietary supplementation in elderly patients with fractured neck of femur. Lancet 1990;335:1013–1016. Tkatch L, Rapin C-H, Rizzoli R, et al: Benefits of oral protein supplementation in elderly patients with fracture of the proximal femur. J Am Coll Nutr 1992;11:519–525.
243
How Can We Improve Functional Outcomes? 26 Sullivan DH, Nelson CL, Bopp MM, et al: Nightly enteral nutrition support of elderly hip fracture patients: A phase I trial. J Am Coll Nutr 1998;17:155–161. 27 Schurch M-A, Rizzoli R, Slosman D, et al: Protein supplements increase serum insulin-like growth factor and attenuate proximal femur bone loss in patients with recent hip fracture. Ann Intern Med 1996;128:801–809. 28 Stratton RJ, Green CJ, Elia M: Disease-Related Malnutrition. An Evidence-Based Approach to Treatment. Oxford, CABI, 2003. 29 Robertson A, Tirado C, Lobstein T, et al: Food and Health for Europe: A New Basis for Action. Geneva, WHO, 2004. 30 Elia M, Stratton R, Russell C: Current Aspects of Nutritional Support in the UK. Reddich, BAPEN, 2002. 31 Elia M, Russell C, Shaffer A, et al: Report of the Biritish Artificial Nutrition Survey (BANS). Reddich, BAPEN, 1999. 32 Elia M, Stratton RJ, Holden C, et al. Home enteral tube feeding following cerebrovascular accident. Clin Nutr 2001;20:27–30. 33 Department of Health: Core Curriculum for Nutrition in the Education of Health Professionals. The Health of the Nation. London, Department of Health, 1994.
Discussion Dr. Powell-Tuck: At the beginning of this workshop we talked about indexes of undernutrition: on the one hand we talked about anthropometric measurements, and on the other hand we talked about measurements of the acute phase response and effect by looking at low albumin for example. Even though we know that low albumin and things like that in hospital may have prognostic significance, wouldn’t you agree that when you look over the patient’s whole journey, the pre-hospital period, that whole curve you showed rather than the acute illness blips, that it becomes very important that we assess nutrition anthropometrically or by direct nutritionally significant things rather than concentrating on the acute phase response? Wouldn’t you also agree therefore that screening tools such as MVST, which under your guidance has recently been introduced into the UK, become something that we all ought to be pressing for very much more than the use of these nutritional indexes which only have prognostic significance in the context of acute illness? Dr. Elia:I fully agree. One needs to have a simple procedure that can be readily undertaken by a wide range of health professionals. Screening is usually undertaken by nurses or health care assistants, but it could be done by other health care workers. By thinking about nutrition, screening is a first step in the management of nutritional problems. Most screening tools in hospital include an acute disease effect, which can take different forms. I am concerned about the use of some of the biochemical indexes. Albumin has been mentioned, but people dying of anorexia nervosa may have a near normal albumin concentration. People who starve for several weeks (water only) may have a normal albumin concentration. In the seminal Keys semi-starvation study, where individuals lost 25% of their body weight over 6 months, the albumin concentration was within the normal range. But as indicated earlier, hypoalbuminemia can develop rapidly as a result of disease. There are also practical points. Results of blood tests can take time to be processed and communicated, and routine blood tests may be impractical in all patients in all care settings. Therefore there is a need for a simple, reliable, reproducible and valid tool that can be readily used in the different health care settings. I full agree with your comments. Dr. Morley: I really enjoyed the talk. I think it was a wonderful global view and I particularly like your emphasis on neonatal and postnatal care because I think we
244
How Can We Improve Functional Outcomes? have really done a terrible job. When we look at old people we forget that muscle mass and bone mass develop by the age of 15–20 years, and if they haven’t developed until then there are going to be problems when the subject is 80, it is very simple. What I care most about are the screening questionnaires because I think that is the secret. In the United States we introduced the determined screen which was a very unsuccessful screen because it had neither sensitivity nor specificity. It was basically a disease screen and aging screen, and while it increased awareness of nutrition, it finished up having almost everybody thinking that every older person should take a caloric supplement because they were at risk. I think that is a terrible approach. Keller [1] has developed a screen which is slowly being introduced throughout Canada that is the final approach that they are tending to use and it is an excellent outpatient screen. The Council on Nutrition has come up with an appetite questionnaire in the United States which will very nicely predict weight loss 6 months afterwards, just based on people’s appetite at the time that they are screened. The nutritional assessment has been extraordinarily successful as a secondary screen, not as a primary screen. We have to have those different screens and put them in place, but all of this leaves us with virtually no really good screening in children. We tend to look at growth development and so on, but we don’t really look at malnutrition in children. So when we have done all of this we are still left with the fact that we will let a child eat inadequately and not exercise, which I think should be a part of the nutrition screening in a child, and in the end we will say now they are 70, let’s try and fix something that we should be fixing now for our grandchildren or children. Dr. Elia: I agree with all that you have said. When we first started to try and get a screening tool, we were a bit over ambitious in thinking we could also tackle children at the same time. We could not get too much agreement about what cutoff points should be used. Unless we have that agreement between professionals, it will be difficult to implement national policies. Dr. Lochs: You put some emphasis on screening of nutritional status, but this is only one part. Wouldn’t the documentation of intake be as important? I recently heard a talk by McFee in which he showed a patient chart from 1905 and one could clearly see what the bowel habits of these patients were in the hospital 100 years ago. But one could not see whether the patients eat or what they got to eat, and I think this situation has not changed yet. So do you have any proposals as to how we could improve on that? Dr. Elia: I can tell you what applies to the ‘Malnutrition Universal Screening Tool’ (‘MUST’), which is ‘an acute disease effect’ that can result in no dietary intake for more than 5 days. The problem is not just a matter of identifying individuals who have not been eating, but also those who are unlikely to eat, such as unconscious individuals or those who develop a severe swallowing difficulty after a stroke. A screening tool is likely to be of little value unless it is linked to a care plan. This plan may vary from one care setting to another, and also according to local resources and policies. High risk patients typically require treatment, low risk patients routine clinical care, and medium risk patients, observation. In hospitals or nursing homes, this observation includes keeping a food chart of dietary intake. The screening procedure can be repeated at intervals, which may be one week in hospital, one month in nursing homes, and even longer in the community. Dr. DeLegge: I have a story to tell you and I would appreciate your feedback. I live in South Carolina, it is one of the poorest states in the US; it is mostly rural. On a survey 5 years ago in various hospitals across the state, we realized there were close to 20 different nutrition assessment programs which made little sense to one another, and then the standardized patient follow-up varied from institution to institution. Most of the nutrition care depended upon how the clinicians felt that they should
245
How Can We Improve Functional Outcomes? treat their particular patients. In response to this we put together a formalized nutrition-based survey at 5 sites. We were able to correlate the patient’s nutrition status to days of re-hospitalization in the next year. We caught the attention of the hospitals because suddenly we were talking dollars. We were also able to focus on patients with chronic disease. This included a lot of patients who were uninsured and used the emergency room as their primary physician. The hospital provided free care. The financial burden of the care of this group of patients falls directly onto the hospitals. The hospitals are searching for any method to prevent re-hospitalization. The interesting part is that we found that current nutrition screening surveys done by nurses were incorrect for identifying at-risk patients 30% of the time. This was even worse when left up to the physicians. We finally realized that we needed to have someone whose total job was to do nothing but nutrition screening in order to impact outcomes. This could not be the regular dieticians as they are responsible for many other tasks, including following up on the patients who are at nutritional risk. Do you think that in order for us to correctly address this problem we have to put the finances behind what we are talking about and not make another duty for the nurse, not make another duty for the physician, but in fact have the personnel specifically trained and designed to do nothing but identify these patients? Dr. Elia: The more resources that are available for implementing a program, the better. However, resources vary of course from place to place and from one health care setting to another. But the extent to which screening tools and procedures agree or disagree depends on the criteria used. The more objective the criteria, usually the less likelihood of disagreement. Let me give you an example. With the ‘MUST’ we undertook 10 reproducibility studies on the same patients. These were between nurses/nurses, doctors/nurses, students/nurses, and other combinations of health workers. The agreement was between 92 and 100%, and in 5 of the 10 studies the agreement was almost perfect (98–100%). This is because the observers were appropriately trained and the ideal tool involves simple and objective criteria. The more complex and subjective the criteria, the less likely the consistency, especially if there has been no training. I don’t know details of your specific tool, but these general comments might be relevant. Dr. Labadarios: In relation to the point being discussed, in South Africa we are actually developing middle level workers in nutrition for the purpose of doing exactly what you said, screening on understanding that the screen has to be simple in order to be reproducible. We are interested in the ‘MUST’ screening tool that you were talking about. Thank you for elegantly pointing out the tremendous complexity of interventions and looking at one parameter only is really a problem. The first question from Dr. Powell-Tuck was on the necessity for simplicity in assessing nutritional status. Assuming that one does that as step one if you like, do you see the need for routinely available or in selective cases more specialized chemical investigations being done among those patients who are actually identified as being at risk, or would you use the ‘MUST’ alone? Dr. Elia: I think the ‘MUST’ is an aid to clinical judgment, it is not a replacement, and it aims to detect mainly protein energy malnutrition. But we know very well that other specific nutrient deficiencies such as iron deficiency may occur in subjects with a normal protein-energy status and even in those who are overweight or obese. However, there is a strong relationship between ‘MUST’ scores and the status of many nutrients. Dr. Morley: Coming back to screening and who should do it. I may not get this totally right because I have only heard Keller [1] who I think has tried harder than anyone to actually look at what happens when you screen and understanding the process and outcomes. She found that if a physician screened there was far more chance that the patient would actually get follow through, which sort of makes some
246
How Can We Improve Functional Outcomes? sense, but in addition to that the patient would follow through themselves with what they were told to do. As you went through the health professionals the outcome was a less good, so that when you got to the volunteers who screened, they actually appear in many cases to do a little better than the physician at doing the screening, but the outcomes were not nearly as good because then you had to refer to the physician, the physician had to do something, the physician didn’t know what to do. I think until we can actually manage to teach physicians about nutrition we are wasting our time. There are some of us who clearly care about it but we are a minority of physicians. The average physician cardiologists would care about cholesterol but would be quite happy if the cholesterol was very low even though that most probably suggests severe cytokine disease, as Rauchhaus et al. [2] have shown, and impeding death in the patients. They would say, well that is great, your cholesterol is nothing we have really succeeded. I think adequate treatment about nutrition and education is absolutely the key. We have been trying to do this for years with medical students and about the only people who pay any attention are the doctors in the nursing home because they see a lot of it and understand, but the rest say it is not important. So what is your solution to make it important? Dr. Elia: It is a difficult problem and I think it is going to take years to overcome because at the heart of all this lack of appreciation of nutrition as a problem, and lack of education, which needs to come more and more to the forefront. I fully agree with you that there is little point in identifying someone who is malnourished unless there is a pathway at the other end, preferably a consistent pathway, to manage the patients. This is the difference between a screening test, which is the procedure in the first initial step and may subsequently lead to assessment, and a screening program, which involves the remainder of the management, including treatment, and may operate between health care settings. There is little point in having a test without a program. At the same time it is necessary to take into account local policies, initiatives and resources. A screening test or program that is absolutely rigid probably won’t work.
References 1 Keller E: Screening for a growth disorder using the Dr. Keller precision measuring instrument and system (in German). Anthropol Anz 1994;52:321–326. 2 Rauchhaus M, Koloczek V, Volk H, et al: Inflammatory cytokines and the possible immunological role for lipoproteins in chronic heart failure. Int J Cardiol 2000;76:125–133.
247
Lochs H, Thomas DR (eds): Home Care Enteral Feeding. Nestlé Nutrition Workshop Series Clinical & Performance Program, vol 10, p 249, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2005.
Concluding Remarks
I think we have done very well and the speakers have been highly committed. Some of you got very tough subjects assigned to you to speak on, and we deeply appreciate your expertise and cooperation. Two things impress me, one is that from all over the world experts in this field have assembled, and I am delighted by the degree of agreement that there is among us. I sometimes think that some of us who look at these issues feel somewhat alone, and so it is very nice to see that we are much more in agreement than we are in disagreement when we get together and discuss these very difficult subjects. The second thing that I am impressed by is the absolute need for education of other health care providers and physicians, both in our home countries and generally worldwide. I commend Nestlé for providing a conference like this and also for the next step which is going to be the dissemination because I think that the educational objectives of this will only be met when we disseminate the results of the discussions to the rest of our colleagues. David Thomas
There is not much to add. For me, and I hope I am speaking for the majority of the audience, it was a pleasure as well as a learning experience to be at this meeting, and I also would like to thank everybody. Prof. Thomas has already raised the point several times and said it will be interesting to see what consequences are drawn. For me also it seems that education of our medical community is the first consequence of such a symposium. Thank you very much again. Herbert Lochs
249
Subject Index
N-acetyl-leucinyl-leucinyl-norleucinal (ALLN), proinflammatory cytokine reduction 221 Aging anorexia of the elderly, see Anorexia muscle loss, see Sarcopenia nutritional status changes over lifetime body composition 31, 32 determinants 32 heritability 36–40 metabolic impact 33–36 methodological issues in characterization 32, 33, 41–43 weight loss versus weight gain 31 Albumin malnutrition assessment 85, 86 predictive value in cancer patients 131 Alcohol acute effects in brain 194 chronic effects on immune system in stress 194, 195 Alcoholism, weight loss 170 Anorexia elderly pathological causes 170, 171 physiological causes 17, 167, 168 sociological causes 16, 17 nutritional interventions 19 pharmacological interventions 20
starvation versus cachexia 18, 19 Anthropometry malnutrition assessment 77, 78, 85 oral protein and energy supplementation trial metaanalysis, elderly 114, 117 Anti-chymotrypsin, inflammation marker 9 Appetite exercise effects 173, 174 food quantity versus quality 177 Arginine, supplementation precautions in burn patients 232 Aspiration, high-risk patients 100, 101 Bioelectrical impedance analysis (BIA) body composition change measurement over lifetime 32, 33, 41 precision of measurements 42 principles 78 B lymphocytes adaptive immune response to commensal intestinal bacteria 209, 210 immune response 190 Body cell mass (BCM) clinical consequences of loss 76, 77 cutoff points for health-related risks 73, 74, 80, 81
251
Subject Index Body mass index (BMI) cutoff points for health-related risks 81 heritability 39, 40 targeting in fracture prevention 240, 241 Cachexia cytokine roles 27, 170 definition 167, 169 distinguishing from starvation 18, 19 pathogenesis 170, 171 Calcium birth weight effects on levels 237 fracture risk reduction 237, 238 supplementation in the elderly 175 Cancer enteral nutrition versus total parenteral nutrition comparison trials advantages and disadvantages 135, 136 complications 130, 131, 132 home enteral nutrition 132–135 mortality 130 nutritional status 130–132 patient preferences 136 tumor growth rate effects 141, 142 weight loss in the elderly 171 Cannabinoids, undernutrition intervention in the elderly 20 Celiac disease, HLA DR2 phenotype 187 Child growth food insecurity relationship 2–4 impairment consequences 5, 6 inflammation relationship 4, 5 Cholecystokinin, aging effects on levels 168 Cimetidine, weight loss 170 Cloricromene, proinflammatory cytokine reduction 221 Computed tomography (CT), body composition measurements 43 Coping, stress effects on immune response studies 202 Corticotropin-releasing factor, exercise induction 174 Cortisol immune suppression by stress 193, 198, 199, 201 proinflammatory cytokine inhibition strategies 223 C-reactive protein (CRP) depression association 201, 202
252
environmental factors in reduction 12, 13 heart disease risks 28 inflammation marker 9, 87 periodontitis 28 stress effects on levels 200 Curcumin, proinflammatory cytokine reduction 220 Dehydration, mortality 165, 166 Delayed skin hypersensitivity response, malnutrition assessment 86 Dementia home enteral nutrition 144 weight loss 170, 176, 177 Dendritic cell (DC), modulation of intestinal flora effects on immune system 204, 205, 207, 209, 210 Depression stress association 201, 202 weight loss 170 Desferrioxamine, proinflammatory cytokine reduction 225 Digoxin, weight loss 170 Dual-energy X-ray absorptiometry (DEXA) body composition measurements 41–43 principles 78, 79 Economics cost analysis parameters 156, 157 ethical considerations 156–159 home enteral nutrition Europe 64, 66–71 United States 45, 46, 52, 54, 71 Elderly energy intake 15 nutritional status 15, 16 oral protein and energy supplementation trial meta-analysis anthropometry 114, 117 benefits 117, 118 complications 105, 114 description of studies 105–112 design of trials 121, 122 hospital stay length 105, 114, 115 micronutrient supplementation as confounding factor 124 mortality outcomes and subgroup analysis 105, 113, 122 selection of studies 104, 105 weight change 114, 116
Subject Index undernutrition frequency in acute hospitals 20 medical illness causes 17, 18, 170, 171 nutritional interventions 19, 25, 26 pharmacological interventions 20 physiological causes 17, 167, 168 sociological causes 16, 17 vitamin supplementation 174, 175 Energy intake, nursing home diets 174, 175 Enteral nutrition (EN), see also Home enteral nutrition cancer patient studies of enteral nutrition versus total parenteral nutrition comparison trials advantages and disadvantages 135, 136 complications 130, 131, 132 home enteral nutrition 132–135 mortality 130 nutritional status 130–132 patient preferences 136 tumor growth rate effects 141, 142 goals in terminal illness 152, 153 metabolic effects compared with total parenteral nutrition 128, 129 termination ethics 153, 154 Etanercept, undernutrition intervention in the elderly 20 Ethics economic considerations 156–159 home enteral nutrition in the United States 68, 69 informed consent 163, 164 legal considerations 154, 155, 161, 162 morality 148 nutritional support termination 153, 154, 161 pediatric nutritional support 155, 156 percutaneous endoscopic gastrostomy in terminal illness 68, 69, 99, 146, 147 religion 148–150 self-determination and medical decision making 150–152, 161, 162 ‘vegetative’ patient nutritional support withdrawal 153 Exercise appetite effects 173, 174 corticotropin-releasing factor induction 174 elderly patients 25, 26
energy intake versus exercise insufficiency in obesity 173 Fat-free mass (FFM) heritability 37, 38 measurement 32, 33, 73, 77–79 metabolic impact of age-related changes 33, 34 nutritional status changes over lifetime 31 physiological mechanisms of loss 74, 75 Fat mass (FM) heritability 37, 38 measurement 32, 33, 77–79 metabolic impact of age-related changes 33, 34 nutritional status changes over lifetime 31 Fluoxetine, weight loss 170 Folate, supplementation 175 Food insecurity child growth relationship 2–4 poverty relationship 1 Fractures birth weight effects on bone health 237 calcium and vitamin D in risk reduction 237, 238 health care service interventions for prevention community interventions 238–240 hospital interventions 240 targeting of interventions 240, 241 Glutamine, proinflammatory cytokine reduction 224, 225, 229 Gut-associated lymphoid tissue (GALT), intestinal microbiota interactions 205 Health outcomes dimensions 233–235 health care service interventions community interventions 238–240 hospital interventions 240 targeting of interventions 240, 241 health professionals attitudes 234, 235 training 242, 243 interventions across lifecourse adults 237, 238 children and adolescents 237 disability threshold reduction 238
253
Subject Index Health outcomes (continued) interventions across lifecourse (continued) fetus 236, 237 overview 235, 236 societal differences in prioritization 234 Home enteral nutrition (HEN) cancer patients 132–135 definition 59, 60 delivery options 51, 52, 62, 63 dementia patients 144 enteral access options and devices 49, 50 Europe frequency of use 60, 61, 66, 67 funding 64, 66–71 indications 62, 67 legislation 63, 64, 69, 70 patient characteristics 61 formulas 51, 62, 63 historical perspective 45 monitoring 51–53 nutritional status outcomes 91–93 pediatric outcomes 49, 67 quality of life 47, 48, 56, 93, 94 recommended dietary allowances in disease 99 risk/benefit ratio analysis 89, 90 safety 46, 55, 90, 91 support in United Kingdom inequity 241, 242 quality 242 survival 48, 49, 55–57, 90, 91, 94–97 United States demographics 46 economics and reimbursement 45, 46, 52, 54, 71 ethics 68, 69 indications 46 nursing home use 54, 56 Human immunodeficiency virus (HIV), malnutrition impact in developing countries 1 Ibuprofen, proinflammatory cytokine inhibition 223 Inalopret, proinflammatory cytokine reduction 230 Inflammation, chronic child growth relationship 4, 5, 9 degenerative diseases 5, 6
254
environmental factors in reduction 12, 13 malnutrition pathways 6, 7, 75 proinflammatory cytokines, see also specific cytokines manipulation anti-inflammatory cytokines 223, 224 production inhibitors 220–222, 230 prospects 225, 226 target cell interaction and signaling interference 222, 223 nutrient studies amino acids 231, 232 glutamine 224, 225 omega-3 fatty acids 225 wasting disease role 219, 220 stress role 199, 200 Informed consent, nutritional support 163, 164 Insulin-like growth factor-1 (IGF-1), aging decline and sarcopenia 169, 219 Interleukin-1 (IL-1), inhibition strategies 223 Interleukin-6 (IL-6) catabolism role 75 sarcopenia role 169 Interleukin-10 (IL-10), proinflammatory cytokine counteraction 223, 224 Intestine barrier bacteria and substrate effects on function 181–183 structure 179, 180 microbiota adaptive immune response to commensals 209, 210 commensal versus pathogenic bacteria 187, 188 dendritic cell interactions 204, 205, 207, 209, 210 differential innate response to bacteria 207–209 enteral nutrition effects 183 fecal short-chain fatty acid studies 214, 215 health and disease 180, 181 host interaction biology 205–207 hygiene hypothesis 210 immune function interactions 204, 205
Subject Index infant diet effects 215, 216 intestinal epithelial cell sensing of bacterial signals 205, 206 modulation with prebiotics and probiotics 203, 211, 213–216 neonatal colonization 215 probiotic supplementation effects 183, 184 permeability antibiotic effects 187 assays 180, 185 diabetics 188 inflammation and malnutrition 11, 185, 186 Irritable bowel syndrome (IBS) intestinal microbiota 214 stress role 186 Isohelenin, proinflammatory cytokine reduction 222 Leptin sympathetic nervous system link 200 testosterone relationship 168 Living will durable power of attorney 164, 165 self-determination and medical decision making 150–152, 161, 162 Lymphocyte count, malnutrition assessment 86 Macrophages, immune response 189, 190 Magnetic resonance imaging (MRI), body composition measurements 43 Malnutrition assessment 77–79, 244–247 causes 73 clinical consequences 76, 77 definition 73, 101 health risks 80 immune response suppression mechanisms 195, 196 outpatient correction 82 pathways in inflammation 6, 7, 75 physiological mechanisms 74, 75 prevalence in hospitalized patients 79, 80 Megestrol acetate, undernutrition intervention in the elderly 20, 24 Melatonin, undernutrition intervention in the elderly 20 Membrane potential decrease in stress 88
measurement 88 Morality, considerations in nutritional support initiation 148 Muscle aging changes 24 loss, see Sarcopenia Nasogastric tube, reimbursement rate for feedings 54 Natural killer cells, immune response 189 Nitrogen balance, protein turnover assessment 87, 88 Nuclear factor-B (NF-B) antioxidant inhibition of activation 230 curcumin inhibition 220 isohelenin inhibition 222 toll-like receptor signal transduction 206, 208, 209 Nutritional support initiation, ethical concerns 144–148 Obesity, energy intake versus exercise insufficiency 173 Omega-3 fatty acids proinflammatory cytokine reduction 225 wound healing studies in rats 231 Pediatric nutritional support ethics 155, 156 home enteral nutrition outcomes 49, 67 Pentoxifylline, proinflammatory cytokine reduction 27, 220 Percutaneous endoscopic gastrostomy (PEG) aspiration high-risk patients 100, 101 complications 90, 91 ethics in terminal illness 68, 69, 99, 146, 147 indications 146, 147 reimbursement rate for feedings 54, 56, 57 survival 57, 90, 91, 94–97 Prealbumin, malnutrition assessment 85 Probiotics immunocompromised patient risks 216 intestinal microbiota modulation 203, 211, 213–216 live organism versus DNA effects 187 supplementation effects in enteral nutrition 183, 184
255
Subject Index Psychoneuroimmunobiology definition 189 historical perspective 199 immune response overview 189–192 mechanisms of stress modulation 193–196 psychological stress effects on immune response 192, 193 Religion, considerations in nutritional support initiation 148–150 Resting energy expenditure (REE) heritability 37, 38 impact of age-related changes in body composition 33–36 Sarcopenia cytokine induction 24, 219 definition 167 pathogenesis 24, 169 Selenium, proinflammatory cytokine reduction 225 Self-determination, considerations in nutritional support initiation 150–152, 161, 162 Skeletal muscle mass index (SMI), cutoff points for health-related risks 81, 82 Starvation definition 167 distinguishing from cachexia 18, 19 Statins, cytokine reduction 28, 230 Stress, see also Psychoneuroimmunobiology classification 201, 202 gastric effects 200, 201 inflammation 199, 200 irritable bowel syndrome role 186 membrane potential effects 88 Testosterone leptin relationship 168 undernutrition intervention in the elderly 23, 24 Thalidomide, tumor necrosis factor-␣ reduction 27 Theophylline, weight loss 170 T lymphocytes immune response 190, 191 intestinal flora effects on immune system 204, 205 proliferation 192
256
stress response 192–194 T helper subsets 191, 192, 213, 219 Toll-like receptors (TLRs) differential innate immune response to intestinal bacteria 207–209 intestinal epithelial cell sensing of bacterial signals 205, 206 signal transduction 206, 208, 209 soluble receptors 207 Total parenteral nutrition (TPN) cancer patient studies of enteral nutrition versus total parenteral nutrition comparison trials advantages and disadvantages 135, 136 complications 130, 131, 132 home enteral nutrition 132–135 mortality 130 nutritional status 130–132 patient preferences 136 tumor growth rate effects 141, 142 goals in terminal illness 152, 153 historical perspective 143 home parenteral nutrition 139, 140, 144, 163, 164 metabolic effects compared with enteral nutrition 128, 129 survival rates in cancer 143, 144 termination ethics 153, 154 Tumor necrosis factor-␣ (TNF-␣) antibody neutralization 222 cytokine interactions 228 ketotifen inhibition 230 polymorphisms and overexpression 228, 229 thalidomide reduction 27 ‘Vegetative’ patients, nutritional support withdrawal ethics 153 Vitamin C, proinflammatory cytokine reduction 225 Vitamin D birth weight effects on levels 237 fracture risk reduction 237, 238 supplementation in the elderly 175 Vitamin E, proinflammatory cytokine reduction 225 Zinc deficiency and hypogeusia 168 proinflammatory cytokine reduction 225