Prader-Willi Syndrome As a Model for Obesity: International Symposium, Zurich, October 18-19, 2002

Prader-Willi Syndrome as a Model for Obesity

This symposium was held in memoriam for Andrea Prader. Drawing of Andrea Prader by his friend and colleague, Emile Gautier, formerly Chairman of the Department of Paediatrics, University of Lausanne, Switzerland.

International Symposium, Zurich, October 18–19, 2002

Prader-Willi Syndrome as a Model for Obesity

Editors

Urs Eiholzer Zurich Dagmar l'Allemand Zurich William B. Zipf Columbus, Ohio

38 figures and 25 tables, 2003

Basel · Freiburg · Paris · London · New York · Bangalore · Bangkok · Singapore · Tokyo · Sydney

Urs Eiholzer

William B. Zipf

Foundation Growth Puberty Adolescence Zurich, Switzerland

Clinical Professor Pediatrics Department of Pediatrics The Ohio State University Columbus, Ohio, USA

Dagmar l'Allemand Foundation Growth Puberty Adolescence Zurich, Switzerland

This book was sponsored by Pharmacia Endocrine Care with an unrestricted educational grant and by the Foundation Growth Puberty Adolescence, Zurich.

Library of Congress Cataloging-in-Publication Data Prader-Willi syndrome as a model for obesity : international symposium, Zurich, October 18–19, 2002 / editors, Urs Eiholzer, Dagmar l’Allemand, William B. Zipf. p. ; cm. Includes bibliographical references and indexes. ISBN 3–8055–7574–2 (hard cover : alk. paper) 1. Prader-Willi Syndrome–Congresses. 2. Somatotropin–Therapeutic use–Congresses. 3. Obesity in children–Congresses. I. Eiholzer, Urs, 1951– II. l’ Allemand, Dagmar. III. Zipf, William B. (William Byron) [DNLM: 1. Prader-Willi Syndrome–Congresses. 2. Growth Hormone–therapeutic use– Child–Congresses. 3. Homeostasis–physiology–Congresses. 4. Hormone Replacement Therapy–Child–Congresses. 5. Obesity–Child–Congresses. QS 675 P8959 2003] RJ520.P7P734 2003 618.92⬘85884–dc21 2003047425 Bibliographic Indices. This publication is listed in bibliographic services. Drug Dosage. The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any change in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug. All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher. © Copyright 2003 by S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland) www.karger.com Printed in Switzerland on acid-free paper by Reinhardt Druck, Basel ISBN 3–8055–7574–2

Contents

Introduction 1 Prader-Willi Syndrome as a Model for Obesity Eiholzer, U. (Zurich) Central Nervous System and Body Weight Homeostasis 7 Obesity due to Mutations in the Anorexigenic Melanocortin Pathway: A Paradigm for Obesity in Prader-Willi Syndrome? Krude, H.; Grüters, A. (Berlin) 15 Signals that Control Central Appetite Regulation Woods, S.C.; Clegg, D.J. (Cincinnati, Ohio) 31 Hypothalamic Neuropeptides and Regulation of Fat Mass in Prader-Willi Syndrome Goldstone, A.P. (London/Amsterdam); Unmehopa, U.A. (Amsterdam); Thomas, E.L.; Brynes, A.E.; Bell, J.D.; Frost, G.; Ghatei, M.A. (London); Holland, A. (Cambridge); Bloom, S.R. (London); Swaab, D.F. (Amsterdam) 44 Discussion Energy Balance in Prader-Willi Syndrome Compared to Simple Obesity 49 Assessment of Body Composition in Children with Prader-Willi Syndrome or Simple Obesity Ellis, K.J. (Houston, Tex.) 61 Physical Activity and Obesity Westerterp, K.R. (Maastricht) 70 Model for a Peripheral Signaling Defect in Prader-Willi Syndrome Lee, P.D.K. (Los Angeles, Calif.) 82 Discussion

Comorbidities or Fundamental Defects of Obesity 86 Characterization of Hyperphagia in Prader-Willi Syndrome Lindgren, A.C. (Stockholm) 93 Consequences of Impaired Growth Hormone Secretion for Body Composition and Metabolism in Obesity and Prader-Willi Syndrome Wabitsch, M. (Ulm) 102 Glucose Homeostasis in Prader-Willi Syndrome Zipf, W.B.; Schuster, D.; Osei, K. (Columbus, Ohio) 119 Sleep-Disordered Breathing in Children with Common Obesity Trang, H. (Paris) 128 Dysregulation of Respiration and Sleep in Prader-Willi Syndrome Schlüter, B. (Datteln) 140 Gonadal Function and Its Disorders in Simple Obesity and in Prader-Willi Syndrome Grugni, G.; Morabito, F. (Verbania); Crinò, A. (Rome) 156 Children with Prader-Willi Syndrome and Primary Obesity: A Comparison of Appetite and Psychosocial Profiles. The Family Perspective Gilmour, J.; Skuse, D. (London) 166 Discussion Comprehensive Treatment Approaches 179 Does Growth Hormone Affect Morbidities Associated with Obesity in Prader-Willi Syndrome? Carrel, A.L.; Allen, D.B. (Madison, Wisc.) 190 Role of Diet and Upbringing in Young Children with Prader-Willi Syndrome l’Allemand, D.; Bachmann, S.; Grieser, J.; Eiholzer, U. (Zurich) 198 Prader-Willi Syndrome: A Pervasive Neurodevelopmental Disorder Requiring a Multidisciplinary Care Approach Whitman, B.Y. (St. Louis, Mo.) 211 A Comprehensive Approach to Limiting Weight Gain and to Normalizing Body Composition in Prader-Willi Syndrome Eiholzer, U. (Zurich) 222 Discussion Epilogue 228 Closing Remarks Leibel, R.L. (New York, N.Y.) 232 Author Index 233 Subject Index

Contents

VI

Introduction Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 1–6

Prader-Willi Syndrome as a Model for Obesity Urs Eiholzer Foundation Growth Puberty Adolescence, Zurich, Switzerland

In 1956 the Prader-Willi syndrome (PWS) was first described by Andrea Prader, Alexis Labhart and Heinrich Willi. To those families affected by the syndrome it was very important that the condition had finally been given a name and was diagnostically categorized. Even though the first description did not mention an efficacious therapy, having a name for the syndrome provided children with PWS and their parents with protection against misguided diagnostics, repeated hospitalizations in search for an explanation for the various symptoms and unjustified accusations from people around them. All parents of grown-up patients with PWS have dreadful stories to tell: in a first stage, they were usually suspected of not feeding their child adequately and neglecting him or her. These accusations subsequently turned into reproaches that they were stuffing their child with food. Many of the children affected also had to endure – from today’s perspective – unnecessary examinations, before somebody finally came up with the correct diagnosis. While the diagnosis of PWS came as a relief for the family and relatives (they knew at last the reason for the child’s condition) on the other hand, it was always devastating news for them, in particular, in self-help groups when parents of a young child saw older, massively obese individuals with PWS. This first description already mentioned a causal hypothalamic involvement. Fortunately, during the following years, great progress was made in the genetic research into PWS. First, in some 60% of all cases a deletion on the long arm of chromosome 15 was detected and it was shown subsequently that the deletion occurred in the chromosome inherited from the father. Later, the phenomenon of genomic imprinting was discovered and the maternal disomy of chromosome 15 was identified as a further cause of PWS. This entailed the conclusion that the lack of paternal information in the area of 15 q11–13 leads

to PWS. Unfortunately, the detailed knowledge of the genetic cause does not explain the pathogenesis of the syndrome. Even to date, the link between the genetic defect and the symptoms – in particular to the hypothalamic disorder – remains largely unknown. Genetics, however, has become indispensable in diagnosing PWS. Efficient methods allow the syndrome to be diagnosed with a certainty of 99% based on a simple blood sample. Thanks to the ease and safety of the method, all newborns with hypotonia of unclear origin are tested for PWS. It has been shown that up to 50% of the newborns with severe hypotonia indeed suffer from PWS. Apart from genetic research and the corresponding results, hardly any new insights were gained up to the early 1990s. Until then, treatment focused on the never-ending fight for adequate energy input or, in the case of infants or toddlers, the parents’ fear of the beginning of the eating problem. As a consequence of this fear, infants and toddlers are often insufficiently fed, which exacerbates the growth failure. In the past years, it has also been shown that the young, still underweight children with PWS have increased fat mass despite their underweight. In other words, the disorder in body composition is present probability already at birth or soon thereafter. Even these children would overeat, if they were physically capable of doing so. From this viewpoint, the onset of obesity represents the external manifestation of insufficient satiety and, at the same time, is an expression of increasing physical strength, health and the children’s ability to get their way. In the 1990s, research into the mechanisms of PWS finally made great progress. Early on, scientists had raised the hypothesis of a possible growth hormone deficiency in PWS. A first trial, however, brought only little success. Proving the growth hormone deficiency was difficult for methodological reasons: decreased growth hormone secretion is also found in normal overweight children. However, in healthy children this is a reversible phenomenon and normalizes after weight loss. Most researchers, therefore, assumed that growth hormone deficiency in PWS is of a functional nature and corresponds to the one seen in simple obesity. Only after it had been shown that the IGF-1 levels in PWS were lower than in normal obesity and that children with PWS had decreased muscle mass in absolute terms (as opposed to children with nonsyndromal obesity, who had an increased muscle mass) the question as to a growth hormone deficiency due to a hypothalamic regulatory disturbance once again became the topic of discussion. In the meantime, a number of lines of evidence have emerged, suggesting that several symptoms of PWS are the consequence of the growth hormone deficiency. Some studies proved the growth hormone deficiency and showed that a growth hormone therapy with physiological doses improves growth, body composition and physical capability, as is otherwise only seen in patients with a genuine growth hormone deficiency. In other words, the improvement of

Eiholzer

2

growth, body composition and physical capability is by far greater than the one seen in studies, in which children without growth hormone deficiency were treated with growth hormone. As a consequence, the FDA and the authorities of most European countries included growth hormone treatment in the standard insurance benefits. While it became clear that growth hormone substitution normalizes growth and greatly improves body composition, it did not suffice to cause normalization. Even after long-term growth hormone treatment, muscle mass remained distinctly decreased and fat mass was increased. Nevertheless, growth hormone treatment significantly contributed to improving the affected child’s circumstances of life. This, however, revealed other symptoms of the syndrome that had been allocated only minor importance previously, such as hypoactivity, psychological abnormalities and the distinct developmental disturbance in language acquisition. Language difficulties and behavioural abnormalities prevent many of the affected children to perform at school as they should going by their intelligence. The question about the function of the gonadal axis and its substitution has not yet been resolved. In 1998 in Jesolo the view was expressed that if adequate management was available, some patients with PWS would be able to have children, because the severity of the hypothalamic regulatory disorder differs among patients. Since then, two women with PWS have given birth to a baby. The appropriate contraception strategy and substitution of sexual hormones has not yet been investigated. There are in particular no studies on the quality of life with or without the substitution of sexual hormones. PWS has become a model for obesity research in general and energy metabolism, since it represents a monocausal genetic origin of adiposity. Energy input and output are regulated by redundant systems in the hypothalamus and it is assumed that in PWS, some of these systems are disturbed. This includes symptoms such as hypoactivity (insufficient muscle mass), lack of satiety (hyperphagia, increased fat mass), regulatory disorder of growth hormone and gonadotropin secretion and many others. The large interest of researchers of various disciplines will undoubtedly lead to a large-scale expansion of the knowledge on and the mechanisms in PWS. All obese people should in the medium and long term gain from the insights of this research – even though the first ones to benefit will most likely be the patients with PWS.

Current Status of PWS Research

The lack of paternal information on the long arm of chromosome 15 is undoubtedly the cause for PWS. It is assumed that many symptoms can be

Introduction: PWS as a Model for Obesity

3

attributed to a hypothalamic disorder. However, the central hypothalamic disorder has not yet been identified. It was shown that leptin is probably not causally involved; it is increased in line with the increased fat mass. The insulin levels, however, are low, at least in children and adolescents. In the periphery, insulin sensitivity is increased, in contrast to most individuals with non-syndromal obesity. Whether the low insulin levels centrally play a role in energy regulation is open. Recently, ghrelin has been shown to be high in PWS, and not low, as in normal obese persons. So far, the increase in ghrelin is the only reproducible hypothalamic abnormality of energy regulation. Its role and in particular the question as to whether a disorder in ghrelin secretion is in fact causally responsible for hyperphagia or whether it is merely an expression of a secondary disorder also remain totally open.

Current Problems Encountered in the Treatment of PWS

The main problem of the treatment of PWS lies in the variety of its symptoms. First of all, the correct diagnosis needs to be established and the necessary checks have to be made. Usually this step involves a geneticist, who then explains the diagnosis to the parents. In the following, the patients need a specialist who is familiar with the syndrome, who is able to guide the patient and his or her family and can explain to them the variety of the symptoms, necessary investigations and treatment options over and over again, creating a relationship of trust. He should also be able to anticipate the specific problems and to involve the various other specialists at the appropriate time or to ask for their assistance. Who among the specialists interested assumes this role is of minor importance. However, it is important for the quality of life of the child affected and its parents that a qualified and competent person takes on this key role. At this stage, the most pressing problem of the families is the treatment of their child. Every single complex of symptoms entails other problems. The questions arising in the context of growth hormone treatment concern the adequate dose, when to institute growth hormone therapy and when to stop it, whether to substitute sexual hormones, if yes, immediately at onset of puberty or wait, but for how long? Does the quality of life improve during the substitution of sexual hormones? Since it has become evident that growth hormone treatment does not normalize muscle mass, the question arises of whether use should be made of other treatment options. A new approach has recently been proposed by our group: a daily training programme. The downside of a daily training programme, however, is that it requires an additional time investment on part of the parents and other caretakers. Similar statements can be made about hyperphagia. It is obvious that growth hormone treatment probably does not have an

Eiholzer

4

effect on hyperphagia. The lack of satiety persists and with it also the necessity to lock fridges and to lock up food, and of a continuous control function to be exercised by parents and other caretakers. This too represents a great strain on the family resources and the question arises as to how these families can be supported. New psychological approaches are asked for. The same applies to the management of behavioural abnormalities and the psychiatric disorders that are seen rather often. Apart from that there are many major and minor problems in the treatment of children with PWS. What are the best measures for a competent treatment of the frequent language acquisition disorders? Where shall the children start school? How to find a suitable place in a home where adequate care of the adult with PWS is provided?

Objectives and Results of the Symposium

Over the past years the knowledge on PWS has rapidly increased. Many of the published papers compare findings in patients affected by the syndrome with those of healthy obese people. PWS is the most frequent clearly defined disorder causing obesity. As a result researchers who investigate energy regulation use PWS as a model for a well-defined monocausal disorder. This symposium was organized on the assumption that new insights into the pathophysiology in PWS also promote the understanding of the causes of normal obesity. It was our objective to bring together the Prader-Willi researchers and their most recent findings with the leading researchers of energy regulation systems. At the symposium, it was our aim that both groups should exchange the current state of their research and provide the research at the interface between PWS and energy regulation with new impetus – on the one hand, for the direct benefit of patients with PWS and, on the other hand, for the benefit of all obese people. In the first part of this book H. Krude and A. Grüters (Germany), S. Woods and D. Clegg (USA) and T. Goldstone et al. (UK) present the latest ideas on the different energy regulation systems of the CNS and of incoming signals from the periphery such as from insulin, leptin and ghrelin. The main topic of the second part of this book is the assessment of body composition [K. Ellis (USA)], as well as the influence of physical training on fat mass and muscle mass [K. Westerterp (The Netherlands)] and a new model for a peripheral signalling defect in PWS [P.D.K. Lee (USA)]. The subject of the third part are comorbidities and fundamental defects of obesity in persons with or without PWS. The phenomenon of hyperphagia is described by A. Lindgren (Sweden), metabolic consequences of impaired growth hormone secretion are covered by M. Wabitsch (Germany), abnormalities of insulin secretion by W.B. Zipf et al. (USA), sleep-related breathing

Introduction: PWS as a Model for Obesity

5

disorders by H. Trang (France) and B. Schlüter (Germany), gonadal function by G. Grugni et al. (Italy) and behaviour and appetite profiles by J. Gilmour and D. Skuse (UK). The last part is devoted to treatment approaches: the role of human GH in PWS [A. Carrel and D. Allen (USA)], the role of diet and education [D. l’Allemand et al. (Switzerland)], as well as the comprehensive approaches to care in PWS by B. Whitman (USA) and U. Eiholzer (Switzerland). Finally, in his closing remarks R. Leibel summarizes the principle insights gained during this symposium, which have left a lasting impression on everybody present. We are greatly indebted to all the contributors and participants of the symposium and to Pharmacia Endocrine Care for making this event possible with an unrestricted educational grant. PD Dr. med. Urs Eiholzer, Foundation Growth Puberty Adolescence, Möhrlistrasse 69, CH–8006 Zurich (Switzerland) Tel. ⫹41 1 364 37 05, Fax ⫹41 1 364 37 01, E-Mail [email protected]

Eiholzer

6

Central Nervous System and Body Weight Homeostasis Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 7–14

Obesity due to Mutations in the Anorexigenic Melanocortin Pathway: A Paradigm for Obesity in Prader-Willi Syndrome? Heiko Krude, Annette Grüters Pediatric Endocrinology, Otto Heubner Center for Pediatrics, Charité University Hospital, Humboldt University, Berlin, Germany

Abstract So far the pathogenesis of obesity in Prader-Willi syndrome (PWS) has not been elucidated, although a detailed description of the underlying genetic defect is now available, e.g. loss of paternally expressed genes in the PWS critical region on chromosome 15. Among those genes so far no direct candidate seems to be related to the occurrence of obesity in PWS. In contrast over the last few years it has been possible to describe several other monogenetic forms of obesity. Mutations were identified mainly in genes which code for components of the anorexigenic leptin-melanocortin pathway of hypothalamic weight regulation. Using the clinical information of obesity and related disorders in these monogenetic obesity cases, one might draw conclusions by comparing these phenotypes with the characteristic features of obesity in PWS. Taken together this comparison reveals that obesity in PWS does not seem to be related to defects in the leptin-melanocortin pathway. Copyright © 2003 S. Karger AG, Basel

Obesity and Prader-Willi Syndrome

Obesity represents one of the most serious symptoms of the Prader-Willi syndrome (PWS) because the extreme overweight during the later course of the disease is resistant to dietary efforts and is the major reason for the increased mortality in PWS [1]. Therefore, it seems mandatory to develop a tailored and pathogenesis-based pharmacological treatment for obesity in PWS.

Unfortunately, although the genetic changes leading to PWS have been described in great detail, it remains unsolved how the loss of paternally expressed genes within the PWS critical region on chromosome 15 results in the pathogenesis of obesity [2]. The function of the genes which are not expressed in PWS has so far been unknown and in particular no obvious candidate gene which might be expressed in the hypothalamic circuits of weight regulation is among those deficient PWS genes. Therefore, an indirect effect of the missing gene products of the PWS region on regulatory circuits of weight regulation is likely. To get more insight into the pathogenesis of obesity one might arrive at a new hypothesis by comparing the obesity phenotype of the PWS with already described monogenic forms of obesity. In defining parallel or different findings in PWS compared to other monogenic obesity syndromes it might be possible to exclude or include some known genes relevant for weight regulation in the pathogenesis of obesity in PWS. Therefore, the aim of this review is to summarize the various findings in monogenic obesity described so far, mainly caused by mutations in the leptinmelanocortin pathway, and to describe the differences to obesity in PWS.

Monogenetic Obesity in Rodents

The first genes determining normal body weight in rodents were only identified at the end of the 20th century although it has been known for a long time that weight maintenance in rodents is strongly influenced by the genome. Cloning of the ob/ob mutant gene leptin [3] revealed that it codes for a hormone secreted by the adipose tissue signalling the actual body fat mass to the CNS. Subsequently, genes were identified which are expressed in the hypothalamus and which respond to the peripheral hormone leptin [4]. These genes encode for components of a neuroendocrine feedback loop which translates the information of leptin into counterregulatory efferent responses to achieve a state of energy homeostasis defined as the balance of energy intake and energy expenditure [5]. Within the arcuate nucleus of the hypothalamus, binding of leptin to the leptin receptor activates the formation of anorexigenic (weight-reducing) peptide neurotransmitters. Posttranslational cleavage of the precursor protein proopiomelanocortin (POMC) results in the formation of small peptides especially ␣-MSH and ␤-MSH which gain their target receptors, the melanocortin receptors 3 and 4 (MC4, MC3R) via axonal guidance to the paraventricular nucleus. Activation of the leptin receptor also increases the formation of the second anorexigenic peptide CART in the same POMC-expressing ARC cells. The increase of POMC and CART expression is paralleled by the decrease of the orexigenic (weightgaining) peptides NPY and AGRP expressed in a second set of ARC cells in

Krude/Grüters

8

response to leptin. More recently, two additional intestinal hormones were identified which act upon the hypothalamic melanocortin system, e.g. orexigenic ghrelin [6] secreted from the stomach and anorexigenic PYY3–36 [7] produced in the gut. Natural loss of function mutants of the leptin and leptin receptor gene in the ob/ob mice and db/db mice, respectively, leads to severely obese mice as has been shown by targeted deletion of the Pomc, Cart and Mc4r gene [8]. Changes of body weight were also found in mice carrying heterozygous MC4R gene mutations [9]. More recently two transcription factor genes have been described, the Sim1 and Nhlh2 gene which are expressed in the PVN and which probably influence the functional level of Pomc and Mc4r. For Sim1 an obese phenotype has been described in heterozygous knockout mice and targeted deletion of the Nhlh2 gene resulted in late onset obesity [10].These more recent examples of monogenic obesity open the view that all components of the leptin-melanocortin pathway, including genes regulating the transcriptional level of the functional relevant genes, are key players for the regulation of normal body weight in rodents.

Human Monogenic Obesity

Following the initial description of the first rodent genes involved in weight regulation the human homologue genes became attractive candidates for human obesity. Because feeding is not consciously driven during infancy disorders leading to obesity in the first months of life are more indicative of genetic defects. Accordingly, the first mutations causing human obesity were identified in cases of early onset and extreme obesity in the first months of life. So far rare mutations were found in the human leptin gene [11], the leptin receptor gene [12], the POMC gene [13], the prohormone convertase 1 gene [14], the MC4R gene [15, 16] and the SIM1 gene [17]. Severe obesity was inherited as a recessive trait in most cases despite gene mutations in the case of MC4R and SIM1. Apart from severe and early onset obesity, additional symptoms were found which can be explained by the complex function of the respective gene products. Due to the additional function of the leptin pathway for the induction of puberty, patients with leptin and leptin receptor mutations were found to have an associated delay of puberty. Children with POMC deficiency are red-haired and suffer from hypocortisolism which can be explained by the additional role of POMC-derived peptides in the regulation of pigmentation and adrenal function, respectively [18]. The single published patient with PC1 deficiency also has an impaired glucose metabolism, which results most likely from the role of PC1 in proinsulin processing [14].

Obesity due to Mutations in the Anorexigenic Melanocortin Pathway

9

Mutations in the MC4R gene were found to cause an obese phenotype without additional defects and although they are still rare, mutations in the MC4R gene are the most common monogenetic defects of human obesity occurring in up to 5% of obese individuals [19, 20].

PWS: Does the Leptin-Melanocortin Pathway Play a Role?

The unexpected finding of a pubertal delay resulting from a hypogonadotropic hypogonadism in leptin- and leptin receptor-deficient patients [21] represents a feature parallel to patients with PWS, who are affected by severe hypogonadotropic hypogonadism, too. The relevance of hypothalamic leptin function during puberty was underlined by the occurrence of pubertal signs in patients with leptin gene loss of function mutations after initiating recombinant leptin therapy [22]. The association of severe obesity and hypogonadotropic hypogonadism in leptin and leptin receptor deficiency turns the leptin and leptin receptor genes into interesting candidates for the pathogenesis of obesity and hypogonadism in PWS. Nevertheless, the subsequent measurement of leptin in PWS patients has shown appropriate serum concentrations with respect to their level of overweight, which argues for a normal secretion of leptin in obese PWS patients [23]. An important observation in patients with genetic defects in the leptinmelanocortin pathway is the early onset of obesity and the extreme overweight during infancy. This feature was initially described in leptin- and leptin receptordeficient children [11, 12] and is also present in POMC deficiency [13]. Already during the very first months of life these children are extremely hyperphagic and rapidly gain weight after birth. The observation of early onset hyperphagia and obesity in patients with leptin and POMC gene mutations obviously contrasts to the later onset of obesity in PWS. Although it has been described recently that even normal weight PWS infant children seem to have altered metabolic rates they tend to become overweight at the age of 4–5 years [24]. Therefore, based alone on this obvious finding of early onset obesity in leptin and POMC deficiency and the later onset in PWS patients it seems unlikely that the primary defect in PWS related to obesity affects the function of leptin or POMC. Mutation screening studies of the MC4R gene were mainly focused on patients with early onset obesity within the first years of life, but subsequent studies in family members of index patients further revealed that obesity might also manifest later in life and that MC4R mutation carriers are even of normal weight [25]. Therefore, according to the onset of obesity an alteration of the MC4R function seems to be consistent with obesity in PWS. The most significant argument against a contribution of the leptinmelanocortin pathway to the pathogenesis of obesity in PWS has been found only

Krude/Grüters

10

very recently when it has been shown that PWS patients have significantly higher ghrelin levels compared to common obese and monogenic obese patients [26]. Ghrelin is a hormone produced in the stomach in response to starvation and activates orexigenic pathways in the hypothalamus. Administration of ghrelin to rats significantly increases the caloric intake and reduces the metabolic rate resulting in increased body weight [27]. Subsequent studies revealed that obese patients are characterized by reduced levels of ghrelin which seem to represent an anorexigenic counterregulatory effort in response to increased body weight. Interestingly the same reduction of ghrelin secretion could be shown for monogenic obesity patients carrying MC4R or leptin receptor mutations [26], which demonstrates that the counterregulatory decrease of ghrelin is not dependent on an intact leptinmelanocortin pathway. Therefore, the higher ghrelin levels in PWS patients implicate that the molecular defect leading to obesity in PWS is different compared to monogenic obesity caused by mutations in the leptin-melanocortin pathway. Taken together, the detailed description of obesity and associated medical problems of patients with the aforementioned monogenic obesity argues against the idea that one of the genes representing the leptin-melanocortin pathway might be involved in the pathogenesis of obesity in PWS.

PWS and Common Obesity

So far two genome screens were performed to identify chromosomal regions linked to common obesity. In a French obese family study three loci were identified on chromosome 10, 5 and 2, respectively, which gained significant lod scores for linkage with the severity of obesity and/or the level of elevated leptin [28]. While no candidate genes were shown to be located in the chromosome 5 and 10 regions, and which still remain to be elucidated, the third locus on chromosome 2 (2p21–23) was found to contain the POMC gene. This locus with positive linkage in the French population was identical to the only identified locus in a Mexican-American screening [29] suggesting that allelic differences in the POMC gene might exist which are responsible for the significant linkage data. This suggestion was further supported by a subsequent study in the Mexican-American population, which, by using intragenic polymorphic markers, increased the lod score up to the highly significant level of 7.46 [30]. However, in the French and Mexican-American genome screening studies the positive linkage data did not include the PWS locus on chromosome 15. Therefore, these genome screen data clearly demonstrate that the genomic regions identified so far which could have been linked to common obesity did not include the PWS critical region. Therefore, the pathogenesis of obesity in PWS is obviously different from the genetics of common obesity.

Obesity due to Mutations in the Anorexigenic Melanocortin Pathway

11

Treatment Perspective

So far almost all efforts to treat obesity in PWS by programs which try to change the characteristic hyperphagic behavior largely failed. Therefore, a tailored pharmacological treatment of obesity would have substantial benefit for PWS patients and their relatives. As outlined before one might expect that in PWS the leptin-melanocortin pathway is not substantially affected. This finding opens the view that substances acting as MC4R agonists and which are under investigation now [31] could have a therapeutic effect on PWS since the MC4R seems to be unaffected in PWS. However, very recently it has been shown that those compounds which activate the MC4R exert an additional stimulatory effect on sexual behavior [32]. These side effects would obviously limit the use of these compounds in the treatment of any form of obesity. The more recent detailed description of endocrine changes in PWS revealed that patients with PWS fail to downregulate the peripheral orexigenic hormone ghrelin [26]. This observation might have a significant impact on the efforts to develop a pharmacological treatment of obesity in PWS. Assuming that the lack in downregulation of ghrelin represents one important component of the pathogenesis of obesity in PWS, compounds with ghrelin-inhibiting effects will have a substantial therapeutic benefit in PWS-related obesity. Nevertheless, only the detailed understanding of the molecular pathogenesis of obesity in PWS, which is still missing, will open additional pharmacological options to treat obesity in PWS.

References 1 2 3 4 5 6 7

8 9

10

Cassidy SB: Prader-Willi syndrome. J Med Genet 1997;34:917–923. Nicholls RD, Saitoh S, Horsthemke B: Imprinting in Prader-Willi and Angelman syndromes. Trends Genet 1998;14:194–200. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM: Positional cloning of the mouse obese gene and its human homologue. Nature 1994;372:425–432. Spiegelman BM, Flier JS: Obesity and the regulation of energy balance. Cell 2001;104:531–543. Cone RD: The central melanocortin system and energy homeostasis. Trends Endocrinol Metab 1999;10:211–216. Tschop M, Smiley DL, Heiman ML: Ghrelin induces adiposity in rodents. Nature 2000;407: 908–913. Batterham RL, Cowley MA, Small CJ, Herzog H, Cohen MA, Dakin CL, Wren AM, Brynes AE, Low MJ, Ghatei MA, Cone RD, Bloom SR: Gut hormone PYY(3–36) physiologically inhibits food intake. Nature 2002;418:650–654. Barsh G, Farooqi S, O’Rahilly S: Genetics of body-weight regulation. Nature 2000;404:644–651. Huszar D, Lynch CA, Fairchild-Huntress V, Dunmore JH, Fang Q, Berkemeier LR, Gu W, Kesterson RA, Boston BA, Cone RD, Smith FJ, Campfield LA, Burn P, Lee F: Targeted disruption of the melanocortin-4 receptor results in obesity in mice. Cell 1997;88:131–141. Nilaweera KN, Ellis C, Barrett P, Mercer JG, Morgan PJ: Hypothalamic bHLH transcription factors are novel candidates in the regulation of energy balance. Eur J Neurosci 2002;15:644–650.

Krude/Grüters

12

11

12

13

14

15 16 17 18 19

20

21 22

23 24 25

26

27 28

29

30

31

Montague CT, Farooqi IS, Whitehead JP, Soos MA, Rau H, Wareham NJ, Sewter CP, Digby JE, Mohammed SN, Hurst JA, Cheetham CH, Earley AR, Barnett AH, Prins JB, O’Rahilly S: Congenital leptin deficiency is associated with severe early-onset obesity in humans. Nature 1997;387:903–908. Clement K, Vaisse C, Lahlou N, Cabrol S, Pelloux V, Cassuto D, Gourmelen M, Dina C, Chambaz J, Lacorte JM, Basdevant A, Bougneres P, Lebouc Y, Froguel P, Guy-Grand B: A mutation in the human leptin receptor gene causes obesity and pituitary dysfunction. Nature 1998;392:398–401. Krude H, Biebermann H, Luck W, Horn R, Brabant G, Gruters A: Severe early-onset obesity, adrenal insufficiency and red hair pigmentation caused by POMC mutations in humans. Nat Genet 1998;19:155–157. Jackson RS, Creemers JW, Ohagi S, Raffin-Sanson ML, Sanders L, Montague CT, Hutton JC, O’Rahilly S: Obesity and impaired prohormone processing associated with mutations in the human prohormone convertase 1 gene. Nat Genet 1997;16:303–306. Yeo GS, Farooqi IS, Aminian S, Halsall DJ, Stanhope RG, O’Rahilly S: A frameshift mutation in MC4R associated with dominantly inherited human obesity. Nat Genet 1998;20:111–112. Vaisse C, Clement K, Guy-Grand B, Froguel P: A frameshift mutation in human MC4R is associated with a dominant form of obesity. Nat Genet 1998;20:113–114. Holder JL Jr, Butte NF, Zinn AR: Profound obesity associated with a balanced translocation that disrupts the SIM1 gene. Hum Mol Genet 2000;9:101–108. Krude H, Gruters A: Implications of proopiomelanocortin (POMC) mutations in humans: The POMC deficiency syndrome. Trends Endocrinol Metab 2000;11:15–22. Vaisse C, Clement K, Durand E, Hercberg S, Guy-Grand B, Froguel P: Melanocortin-4 receptor mutations are a frequent and heterogeneous cause of morbid obesity. J Clin Invest 2000;106: 253–262. Farooqi IS, Yeo GS, Keogh JM, Aminian S, Jebb SA, Butler G, Cheetham T, O’Rahilly S: Dominant and recessive inheritance of morbid obesity associated with melanocortin 4 receptor deficiency. J Clin Invest 2000;106:271–279. Farooqi IS: Leptin and the onset of puberty: Insights from rodent and human genetics. Semin Reprod Med 2002;20:139–144. Farooqi IS, Jebb SA, Langmack G, Lawrence E, Cheetham CH, Prentice AM, Hughes IA, McCamish MA, O’Rahilly S: Effects of recombinant leptin therapy in a child with congenital leptin deficiency. N Engl J Med 1999;341:879–884. Eiholzer U, Blum WF, Molinari L: Body fat determined by skinfold measurements is elevated despite underweight in infants with Prader-Labhart-Willi syndrome. J Pediatr 1999;134:222–225. Eiholzer U, Schlumpf M, Nordmann Y, l’Allemand D: Early manifestations of Prader-Willi syndrome: Influence of growth hormone. J Pediatr Endocrinol Metab 2001;14:1441–1444. Sina M, Hinney A, Ziegler A, Neupert T, Mayer H, Siegfried W, Blum WF, Remschmidt H, Hebebrand J: Phenotypes in three pedigrees with autosomal dominant obesity caused by haploinsufficiency mutations in the melanocortin-4 receptor gene. Am J Hum Genet 1999;65:1501–1507. Cummings DE, Clement K, Purnell JQ, Vaisse C, Foster KE, Frayo RS, Schwartz MW, Basdevant A, Weigle DS: Elevated plasma ghrelin levels in Prader Willi syndrome. Nat Med 2002;8: 643–644. Tschop M, Weyer C, Tataranni PA, Devanarayan V, Ravussin E, Heiman ML: Circulating ghrelin levels are decreased in human obesity. Diabetes 2001;50:707–709. Hager J, Dina C, Francke S, Dubois S, Houari M, Vatin V, Vaillant E, Lorentz N, Basdevant A, Clement K, Guy-Grand B, Froguel P: A genome-wide scan for human obesity genes reveals a major susceptibility locus on chromosome 10. Nat Genet 1998;20:304–308. Comuzzie AG, Hixson JE, Almasy L, Mitchell BD, Mahaney MC, Dyer TD, Stern MP, MacCluer JW, Blangero J: A major quantitative trait locus determining serum leptin levels and fat mass is located on human chromosome 2. Nat Genet 1997;15:273–276. Hixson JE, Almasy L, Cole S, Birnbaum S, Mitchell BD, Mahaney MC, Stern MP, MacCluer JW, Blangero J, Comuzzie AG: Normal variation in leptin levels in associated with polymorphisms in the proopiomelanocortin gene, POMC. J Clin Endocrinol Metab 1999;84:3187–3191. MacNeil DJ, Howard AD, Guan X, Fong TM, Nargund RP, Bednarek MA, Goulet MT, Weinberg DH, Strack AM, Marsh DJ, Chen HY, Shen CP, Chen AS, Rosenblum CI, MacNeil T, Tota M,

Obesity due to Mutations in the Anorexigenic Melanocortin Pathway

13

32

MacIntyre ED, Van der Ploeg LH: The role of melanocortins in body weight regulation: Opportunities for the treatment of obesity. Eur J Pharmacol 2002;440:141–157. Van der Ploeg LH, Martin WJ, Howard AD, Nargund RP, Austin CP, Guan X, Drisko J, Cashen D, Sebhat I, Patchett AA, Figueroa DJ, DiLella AG, Connolly BM, Weinberg DH, Tan CP, Palyha OC, Pong SS, MacNeil T, Rosenblum C, Vongs A, Tang R, Yu H, Sailer AW, Fong TM, Huang C, Tota MR, Chang RS, Stearns R, Tamvakopoulos C, Christ G, Drazen DL, Spar BD, Nelson RJ, MacIntyre DE: A role for the melanocortin 4 receptor in sexual function. Proc Natl Acad Sci USA 2002;99:11381–11386.

Dr. Heiko Krude, Otto-Heubner-Centrum for Pediatrics, Charité, Humboldt University, Augustenburgerplatz 1, D–13353 Berlin (Germany) Tel. ⫹49 30 450 559828, Fax ⫹49 30 450 566926, E-Mail [email protected]

Krude/Grüters

14

Central Nervous System and Body Weight Homeostasis Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 15–30

Signals that Control Central Appetite Regulation Stephen C. Woods, Deborah J. Clegg Department of Psychiatry, University of Cincinnati, Cincinnati, Ohio, USA

Abstract Knowledge of the hormones, neurotransmitters and neural pathways that regulate energy homeostasis is increasing at a rapid pace. Although food intake was historically thought to occur in response to acute energy demands such as low blood sugar, current evidence suggests instead that most individuals initiate meals according to habit and convenience. Although meal patterns vary considerably among individuals, most individuals maintain stable body weights and levels of adiposity over long intervals. The explanation is that energy homeostasis is accomplished via control of meal size. This article reviews current understanding of signals that mediate the brain controls over energy intake (food intake) and expenditure (metabolism and exercise). One group of signals is related to the control of individual meals, with some initiating them (exemplified by the stomach hormone, ghrelin) and others terminating them (exemplified by the ‘satiety’ signals such as the duodenal peptide, cholecystokinin). Other signals reflect the current state of energy balance (the size of adipose stores or the total amount of fat in the body) as exemplified by the adipose hormone, leptin, and the pancreatic hormone, insulin. The integration and coordination among meal-related signals, body fat-related signals, and multiple intrinsic brain circuits determines normal eating behavior. The general principle is that when an individual is underweight, decreased levels of adiposity signals allow larger meals to be consumed, and the opposite occurs during states of positive energy balance. Discussion will focus on differences between insulin and leptin as adiposity signals, on how they interact with satiety signals, on where in the brain metabolic signals are detected and how they are integrated to maintain energy balance over long intervals, and on possible malfunction of this system in individuals with the Prader-Willi syndrome. Copyright © 2003 S. Karger AG, Basel

Introduction

The control of food intake and body weight is a complex process utilizing information in the form of signals from throughout the body and many areas of

the brain. It is also an area of intense current investigation, such that any summary is necessarily tentative and soon out of date. That said, there are several important principles that help in understanding the neurobiology of energy homeostasis, and it is on these that we focus in this review. The first is that the amount of fat in the body (often estimated as body weight) is stable over long intervals in adults. Importantly, body fat content, while genetically determined, has a wide range of latitude. One’s genetic heritage determines the upper and lower limits of acceptable weight, and environmental factors determine where an individual lies within that range. In a stable environment, the individual will achieve the level of body fat appropriate for that environment [1–3]. If weight deviates due to some situational event, it drifts back when the opportunity arises. Likewise, if weight is experimentally or voluntarily moved off the regulated level, compensatory systems become activated returning body weight toward the baseline [3–5]. The body can muster behavioral as well as metabolic adjustments to ingest and/or expend more or less energy [6]. Hence, in a stable environment, the individual will appear to have what has been termed a set point such that whenever its weight is displaced away from the regulated level, it returns when conditions allow [6], and this is the case independent of the level of weight that is being maintained [7]. An important consideration is that the body weight of different individuals living in comparable and stable environments varies considerably due to genetic variation [8–11]. The second principle of the control system that regulates energy homeostasis is that food intake is controlled by how much is eaten once a meal is underway rather than by when meals are initiated. When food is abundant, individuals could, in principle, eat whenever they choose. However, most individuals develop stable meal patterns, adopting a schedule that is relatively consistent from day to day, such as eating ‘three square meals a day’. Although individuals vary considerably in terms of the number and size of individual meals, most maintain stable daily food intake and body weight. The ability to adopt a feeding schedule that coordinates with other demands of the environment allows optimal use of time for social activities, exercise, sleep patterns, and other behaviors. The point is that the signals that determine the onset of eating are distinct from those that terminate meal offset, and it is the latter where physiological controls are exerted [12]. During meals, sensory signals providing information on the quantity and quality of available food impinge on the brain. This includes exteroceptive cues (seeing the food), the smell and taste and physicochemical properties of the food, the degree of stomach distension, and so on [5, 13–15]. The brain integrates this information and determines when the meal will end. As reviewed below, there is considerable evidence that as the gut (i.e., the stomach and small intestine) processes food, specialized cells lining the gut secrete a number of

Woods/Clegg

16

peptide hormones that stimulate distant organs such as the liver and pancreas to secrete the appropriate mix of juices to digest food. Some of the same peptides double as signals to the nervous system, providing information on what and how much has been eaten. When this peptidergic signal gets sufficiently high, individuals feel full or satiated, and they stop eating. Other signals are continuously informing the brain as to the amount of fat stored in the body. These ‘adiposity’ signals influence the sensitivity of the meal-generated ‘satiety’ signals. If an individual has lost weight, a smaller adiposity signal reaches the brain, rendering it less sensitive to the satiating effect of satiety signals. The individual consequently eats larger meals and regains lost weight. Analogously, an individual who has gained weight will be more sensitive to satiety signals. This mechanism allows body weight to be maintained over long intervals, and it is independent of when meals are initiated [see reviews in 5, 13–15]. Hence, individuals can adopt an eating schedule that best fits their environment.

Meal-Related Signals

Different types of signals determine when individuals start and end meals. Until recently, meal initiation was thought to be caused by reductions in available nutrients to critical tissues such as the brain. Mayer’s glucostatic hypothesis stated that ‘hunger’ and the onset of eating are stimulated by reduced glucose utilization in the brain, and that satiety and meal termination occur as glucose utilization is restored [16, 17]. Other hypotheses were based upon the utilization of lipids, or total energy available, or body temperature [18–21]. All such hypotheses assume that fluctuations of available energy underlie meal taking, yet most evidence suggests the opposite. Unless an individual is severely food- and weight-restricted, meals are initiated when there is ample energy available to the brain and other tissues [20, 22, 23]. Most evidence suggests that meals are initiated based upon learning, habits and convenience [12, 24, 25]. Individuals eat at the same time every day, or because they have a break in their schedule, or because a meal has been prepared. As discussed above, the control system allows such individual flexibility because meal termination, not meal initiation, is what is regulated. Considerable evidence suggests that signals generated as food interacts with the gut (stomach and intestine) are relayed to the hindbrain where their collective message ultimately causes eating to stop (see fig. 1). The best known of these satiety signals is the duodenal peptide, cholecystokinin or CCK, which is secreted as food enters the intestine from the stomach [26]. CCK, along with other signals such as stomach distension, stimulates receptors on the vagus

Signals that Control Central Appetite Regulation

17

Catabolic path way s LHA ⫹ Response to satiety signals ⫺ POMC Anabolic pathways

PVN

NPY

Fat mass

NTS

ARC Leptin

Adiposity signals

Insulin

GI tract Vagus nerve

Satiety signals

Liver

Mechanical Chemical

Energy metabolism

Superior cervical ganglion

Cervical spine SNS afferents

CCK release

Fig. 1. Model summarizing different levels of control over energy homeostasis. During meals, signals such as CCK or distension of the stomach that arise from the gut (stomach and intestine) trigger nerve impulses in sensory nerves traveling to the hindbrain. These satiety signals synapse with neurons in the nucleus of the solitary tract (NTS) where they influence meal size. Signals related to body fat content such as leptin and insulin, collectively called adiposity signals, circulate in the blood to the brain. They pass through the blood-brain barrier in the region of the arcuate nucleus (ARC) and interact with neurons that synthesize POMC or NPY and AgRP. These neurons in turn project to other hypothalamic areas including the PVN and the lateral hypothalamic area (LHA). The net output of the PVN is catabolic and enhances the potency of satiety signals in the hindbrain. The net output of the lateral hypothalamic area, on the other hand, is anabolic, suppressing the activity of the satiety signals. In this way body fat content tends to remain relatively constant over long intervals by means of changes of meal size [from 13]. SNS ⫽ Sympathetic nervous system.

nerves, and nerve impulses are passed to the brain [27–29]. If individuals are administered antagonists to CCK prior to a meal, the meal is larger, implying that their endogenous CCK normally limits meal size [30–33]. Analogously, when administered CCK prior to a meal, individuals eat smaller meals, with larger doses of CCK eliciting greater reductions [26, 34–37]. Importantly, humans administered CCK report feeling full or sated earlier in a meal without

Woods/Clegg

18

Table 1. Partial list of gut peptides purported to reduce meal size and be satiety signals

Ref. No. CCK Bombesin Gastrin-releasing peptide (GRP) Neuromedin B (NMB) Enterostatin Amylin Apolipoprotein A-IV PP Glucagon Somatostatin Glucagon-like peptide-1 (GLP-1)

34, 35, 37 116, 117 118 119, 120 121–123 124 125, 126 110 127, 128 129 130, 131

feelings of nausea or discomfort [26, 37, 38]. Based upon this, manipulating CCK might be considered a logical weight-loss strategy. However, when animals are administered CCK prior to every meal such that the size of every meal is reduced, they compensate perfectly by eating more meals and maintaining their body weight [39, 40]. Hence, strategies that mimic increased satiety are not likely to be efficacious for weight loss in and of themselves. CCK is but one of several signals that are generated by food in the gut and that reduce meal size. Most gut peptides share the property of stimulating the hindbrain to elicit satiety, and table 1 lists several of these. The recently discovered peptide ghrelin, which is made in the stomach [41], is an interesting exception. Ghrelin is secreted just prior to meals [42], and when exogenous ghrelin is administered, individuals eat larger meals [43–46]. The signals that cause the stomach to synthesize and release ghrelin are as yet unknown. As discussed below, an important recent observation is that patients with the Prader-Willi syndrome (PWS) have high levels of ghrelin [47].

Adiposity Signals

As depicted in figure 1, another group of signals is proportional to body fat. These adiposity signals are circulating hormones that pass through the blood-brain barrier and activate receptors in the hypothalamus of the forebrain. Both leptin from fat cells and insulin from pancreatic B cells are secreted in direct proportion to body fat, and receptors for both peptides are found on neurons in the arcuate nucleus [13, 14]. When leptin action in the brain is reduced

Signals that Control Central Appetite Regulation

19

(during a fast or diet, or in individuals who lack leptin or its receptor), food intake and body weight are elevated; when leptin action in the brain is increased (by administering exogenous leptin), food intake and body weight are decreased [13, 14, 48]. Analogously, when insulin action in the brain is decreased (as in type 1 diabetes mellitus), individuals are hyperphagic, but they cannot become obese because insulin is required for fat cells to store fat. When insulin action in the brain is increased, individuals eat less and lose weight [49, 50]. Insulin and leptin act on similar neural circuits originating in the arcuate nucleus [13, 14, 51, 52], and when they are simultaneously administered, their combined effects become additive [53]. In sum, two distinct types of signals that control energy balance arise in the body and influence the brain. Information related to the amount and content of the food being eaten is relayed to the hindbrain, whereas information related to body fat is relayed to the forebrain (fig. 1). Both types of signals are integrated with multiple other information related to such factors as the social setting, learning and habits, stress, time of day, and emotional factors, in the control of food intake and energy expenditure.

Hypothalamic Signals

Two distinct populations of neurons have been identified in the arcuate nucleus, and each is influenced by both leptin and insulin [see reviews in 13, 14]. The first synthesizes the compound, proopiomelanocorticotropin (POMC), which is a precursor molecule for many neurotransmitters. The neurons in the arcuate nucleus convert POMC into ␣-melanocyte-stimulating hormone (␣MSH) and secrete it as a neurotransmitter in several other hypothalamic areas including the paraventricular nuclei (PVN) and the lateral hypothalamic area (LHA, see fig. 1). Leptin and insulin stimulate POMC synthesis and ␣MSH secretion [51, 54], and ␣MSH in turn causes a net catabolic response by acting at melanocortin-4 receptors (MC4R) in the PVN and LHA [13, 14], i.e. decreased food intake, increased energy expenditure, and weight loss. Administering ␣MSH (or synthetic analogs) has the same effect [55, 56], and administering antagonists to ␣MSH causes a net anabolic response (increased food intake, decreased energy expenditure and weight gain) [57, 58]. Individuals who lack ␣MSH or have defective MC4R are very obese. Identifying compounds that mimic the hypothalamic melanocortin system (leptin or insulin to ␣MSH to MC4R) is a major goal for the treatment of obesity. The other population of neurons in the arcuate nucleus synthesizes two important neurotransmitters, neuropeptide Y (NPY) and agouti-related peptide (AgRP). As occurs with ␣MSH from POMC neurons, NPY and AgRP are

Woods/Clegg

20

Table 2. Hypothalamic neuropeptides that influence energy homeostasis Catabolic peptides (many found in the PVN) Oxytocin Corticotropin-releasing hormone (CRH) Growth hormone-releasing hormone (GRH) Cocaine-amphetamine-related transcript (CART) ␣MSH Thyrotropin-releasing hormone (TRH) Glucagon-like peptide-1 (GLP-1) Anabolic peptides (many found in the LH) Melanin-concentrating hormone (MCH) Orexin A and B ␤-Endorphin NPY AgRP Reviews can be found in Ref. No. 5 and 13–15.

released as neurotransmitters in the PVN and LHA (fig. 1). Unlike what occurs at POMC neurons, insulin and leptin inhibit the activity of NPY/AgRP neurons, causing decreased synthesis and release of these transmitters [59, 60]. NPY is a potent anabolic compound, eliciting increased eating, decreased energy expenditure and weight gain by acting at one or more Y receptors in the PVN, LH and elsewhere [61–64]. AgRP administration has a similar effect, but it acts by antagonizing activity at MC4R. Hence, the NPY/AgRP neurons stimulate food intake in two ways. NPY acts at Y receptors to stimulate eating while AgRP acts at MC4R to reduce the ability of ␣MSH to inhibit feeding. As a general rule, the PVN can be considered a catabolic area of the hypothalamus since stimulation there decreases food intake and body weight and since several neurotransmitters it releases are themselves catabolic (see table 2). The LHA, on the other hand, is an anabolic area since stimulation there increases food intake and since the neurotransmitters it makes and releases elicit increased eating (table 2). Understanding the complex interactions among the numerous neurotransmitters that control energy balance is currently a hot research topic as investigators try to develop new compounds to treat obesity. Integration of Different Types of Signals

Meal-generated satiety signals help determine when a meal ends. The sensitivity of the brain to these signals is modified by the levels of adiposity signals.

Signals that Control Central Appetite Regulation

21

If an individual diets or loses weight for some other reason, insulin and leptin are decreased and less of an adiposity signal reaches the arcuate nucleus. One result is that the sensitivity to CCK is reduced, resulting in more food being eaten before satiety occurs. Likewise, if an individual has gained some weight, the increased activity of leptin and insulin makes the brain more sensitive to CCK, resulting in smaller meals [65–70]. This mechanism helps to maintain a stable weight while allowing considerable flexibility as to when meals occur. This discussion is overly simplistic because many other factors determine both meal onset and meal size. Certainly cognitive factors can easily override the simple negative feedback control of body fat, and the abundance and ubiquity of highly palatable, calorically dense foods probably contribute to the increased food intake thought to be responsible for the current ‘epidemic’ of obesity [71, 72].

Meal-Related Signals and the PWS

PWS is a human genetic disorder characterized by mental and physical abnormalities. The genetic mutation results from a deletion of a segment of the paternally derived chromosome 15q [73] or else a maternal disomy involving the same region on 15q [74]. It is estimated that one child in every 10,000–25,000 live births suffers from the syndrome [75, 76]. The diagnosis of PWS is made according to a set of consensus clinical criteria [77] including short stature, muscular hypotonia, excessive appetite with progressive obesity, hypogonadism, mental retardation, behavioral abnormalities, sleep disturbances, and dismorphic features. While other reports in this volume review the syndrome in more detail, we focus here upon what is known of the food intake and obesity of individuals with PWS. Children with PWS usually become overweight by the age of 4 years as a consequence of their insatiable appetite and compulsive eating [78]. Obesity progresses with age, with many PWS patients being more than twice their ideal body weight [79, 80]. PWS patients have a higher percentage of body fat and a lower percentage of lean tissue [80–82] than weight-matched controls. Obesity in PWS, like obesity in general, is associated with increased morbidity and mortality [83, 84]. However, despite their morbid obesity, PWS subjects are relatively protected against the metabolic complications typically associated with obesity, including insulin resistance, glucose intolerance, hyperlipidemia, cardiovascular disease, and mortality [85]. As an example, PWS females have lower fasting insulin, a higher insulin/glucose ratio, and an increased C-peptide/ insulin ratio than weight-matched females [86]. Fasting insulin levels are lower in PWS children than BMI-matched children [87–89], and PWS children have increased insulin sensitivity compared with obese controls [75, 89].

Woods/Clegg

22

One possible reason for this is that adipose tissue is distributed subcutaneously in PWS, with relatively little visceral fat, and this is true for both genders [79]. This distribution is reminiscent of what occurs in females, whereas non-PWS obese males, with increased visceral adiposity, have an elevated risk for the metabolic complications of obesity [90, 91]. When leptin was discovered, it was hypothesized that defects in leptin signaling might explain the hyperphagia and obesity of PWS patients. However, plasma leptin levels in obese PWS patients are the same as in weight-matched controls [86]. Therefore, the elevated plasma leptin in PWS is appropriate for the high level of subcutaneous adipose tissue, and it has been concluded that defects in leptin secretion are not responsible for the obese phenotype in PWS [92–99]. In sum, inappropriate levels of insulin or leptin cannot account for the obese phenotype of PWS patients. If inappropriate signaling of either hormone is involved, it could be due to reduced penetration through the blood-brain barrier and/or to reduced sensitivity by neurons in the arcuate nucleus. These possibilities have not been assessed. As discussed above, ghrelin is a novel enteric hormone [100] that increases food intake and body weight [44, 100–102]. Ghrelin has been implicated in mealtime hunger [42] and body weight regulation in animals [44, 102] and humans [42, 45, 46]. Cummings et al. [47] recently found that fasting plasma ghrelin levels are 4.5-fold higher in PWS subjects than in weight-matched obese controls. Although the genes encoding ghrelin and its receptor are not within the 15q chromosome implicated in PWS, genes in that region could encode factors that indirectly affect ghrelin expression [103]. However, ghrelin levels in PWS subjects were unrelated to BMI and were uniformly above the regression line for non-PWS individuals. All of these data suggest that PWS is characterized by elevated plasma ghrelin and further suggest that ghrelin may be involved in the etiology of the voracious appetite characterizing the syndrome. In fact, ghrelin levels in PWS subjects are comparable to or higher than those reported to stimulate appetite and food intake following ghrelin administration in humans [46]. Interventions that lower plasma ghrelin levels, such as gastric bypass surgery [104], warrant consideration in the treatment of obesity in PWS. PWS subjects reach satiety considerably more slowly than obese controls [105], suggesting that defective satiety signals or their reception might contribute to the syndrome. Several reports have indicated that basal CCK levels are normal in PWS [106–108]. However, there is one report that CCK release in response to eating a mixed meal is elevated in PWS subjects compared with obese controls [107], but the interpretation is confounded by the fact that the PWS subjects consumed 3 times more food than the controls. There is also a report that the positive correlation typically observed between fasting plasma

Signals that Control Central Appetite Regulation

23

fatty acids and CCK levels [109] is lacking in PWS subjects [108]. Hence, PWS may be characterized by a sluggish CCK release to rising plasma fatty acid levels during a high-fat meal. This is also an area that requires further research. The only other satiety peptide investigated with regard to PWS is pancreatic polypeptide (PP). PP is secreted from the intestine during meals, and it has been suggested to reduce meal size [110]. PWS patients are deficient in basal and meal-stimulated PP secretion [106, 111, 112]. However, deficits in PP secretion have also been reported in non-PWS obesity [113, 114]. The extremely large meals characteristic of PWS warrant further investigation into all possible satiety factors. There are only scattered reports on hypothalamic neurotransmitter systems involved with energy balance in PWS. In a postmortem study, NPY, but not AgRP mRNA expression, was decreased in PWS adults [86]. If confirmed, this might imply that there is a compensatory decrease of NPY in PWS in response to their hyperphagia and/or obesity. In another postmortem study, a significant reduction in the volume of the PVN was observed in PWS patients [115]. As discussed above, the PVN is an area that tends to reduce food intake and body weight, such that a reduction of activity in this hypothalamic nucleus might result in hyperphagia and body weight dysregulation associated with the PWS syndrome. Conclusion

A complex, highly redundant regulatory system controls energy homeostasis. It relies upon information related to the amount of fat in the body (adiposity signals), information about the food being eaten (satiety signals), and information about the social situation, habits and many other factors to function effectively. Patients with PWS have extreme hyperphagia and obesity, and they are characterized as eating especially large meals. While this pattern suggests that defective signaling or feedback must exist somewhere in the control system in PWS, too little is presently known to make any conclusions. Acknowledgments Preparation of this report was supported by NIH grant DK17844 to SCW.

References 1 2

Bray GA: The Obese Patient. Philadelphia, Saunders, 1976. Pinel JPJ, Assanand S, Lehman DR: Hunger, eating and ill health. Am Psychol 2000;55: 1105–1116.

Woods/Clegg

24

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

23 24 25 26 27 28

29 30

31

Wirtshafter D, Davis JD: Set points, settling points, and the control of body weight. Physiol Behav 1977;19:75–78. Woods SC, Decke E, Vasselli JR: Metabolic hormones and regulation of body weight. Psychol Rev 1974;81/1:26–43. Woods SC, Schwartz MW, Baskin DG, Seeley RJ: Food intake and the regulation of body weight. Annu Rev Psychol 2000;51:255–277. Keesey RE, Hirvonen MD: Body weight set-points: Determination and adjustment. J Nutr 1997;127:1875S–1883S. Hoebel BG, Teitelbaum P: Weight regulation in normal and hypothalamic hyperphagic rats. J Comp Physiol Psychol 1966;61:189–193. Bouchard C: The genetics of obesity: From genetic epidemiology to molecular markers. Mol Med Today 1995;1:45–50. Bray G, Bouchard C: Genetics of human obesity: Research directions. FASEB J 1997;11:937–945. Reed DR, Bachmanov AA, Beauchamp GK, Tordoff MG, Price RA: Heritable variation in food preferences and their contribution to obesity. Behav Genet 1997;27:373–387. West DB: Genetics of obesity in humans and animal models. Endocrinol Metab Clin North Am 1996;25:801–813. Woods SC, Strubbe JH: The psychobiology of meals. Psychon Bull Rev 1994;1:141–155. Schwartz MW, Woods SC, Porte DJ, Seeley RJ, Baskin DG: Central nervous system control of food intake. Nature 2000;404:661–671. Woods SC, Seeley RJ, Porte DJ, Schwartz MW: Signals that regulate food intake and energy homeostasis. Science 1998;280:1378–1383. Woods SC, Seeley RJ: Adiposity signals and the control of energy homeostasis. Nutrition 2000; 16:894–902. Mayer J: Regulation of energy intake and the body weight: The glucostatic and lipostatic hypothesis. Ann NY Acad Sci 1955;63:14–42. Mayer J, Thomas DW: Regulation of food intake and obesity. Science 1967;156:328–337. Friedman MI: Body fat and the metabolic control of food intake. Int J Obes 1990;14(suppl 3):53–67. Friedman MI: Fuel partitioning and food intake. Am J Clin Nutr 1998;67(suppl 3):513S–518S. Langhans W: Metabolic and glucostatic control of feeding. Proc Nutr Soc 1996;55:497–515. Nicolaidis S, Even P: Mesure du métabolisme de fond en relation avec la prise alimentaire: Hypothèse ischymétrique. C R Acad Sci III 1984;298:295–300. Epstein AN, Nicolaidis S, Miselis R: The glucoprivic control of food intake and the glucostatic theory of feeding behavior; in Mogenson GJ, Calaresci FR (eds): Neural Integration of Physiological Mechanisms and Behavior. Toronto, Toronto University Press, 1975, pp 148–168. Grossman SP: The role of glucose, insulin and glucagon in the regulation of food intake and body weight. Neurosci Biobehav Rev 1986;10:295–315. Woods SC: The eating paradox: How we tolerate food. Psychol Rev 1991;98:488–505. Woods SC, Ramsay DS: Pavlovian influences over food and drug intake. Behav Brain Res 2000; 110:175–182. Smith GP, Gibbs J: The satiating effects of cholecystokinin and bombesin-like peptides; in Smith GP (ed): Satiation. From Gut to Brain. New York, Oxford Publishing Company, 1998, pp 97–125. Corp ES, McQuade J, Moran TH, Smith GP: Characterization of type A and type B CCK receptor binding sites in rat vagus nerve. Brain Res 1993;623:161–166. Mercer JG, Lawrence CB: Selectivity of cholecystokinin receptor antagonists, MK-329 and L-365,260 for axonally transported CCK binding sites in the rat vagus nerve. Neurosci Lett 1992;137:229–231. Reidelberger RD: Abdominal vagal mediation of the satiety effects of exogenous and endogenous cholecystokinin in rats. Am J Physiol 1992;263:R1354–R1358. Beglinger C, Degen L, Matzinger D, D’Amato M, Drewe J: Loxiglumide, a CCK-A receptor antagonist, stimulates calorie intake and hunger feelings in humans. Am J Physiol 2001;280: R1149–R1154. Hewson G, Leighton GE, Hill RG, Hughes J: The cholecystokinin receptor antagonist L364,718 increases food intake in the rat by attenuation of endogenous cholecystokinin. Br J Pharmacol 1988;93:79–84.

Signals that Control Central Appetite Regulation

25

32 33 34 35 36

37

38 39 40

41 42 43

44 45

46

47 48 49 50 51 52 53 54

55

Moran TH, Ameglio PJ, Peyton HJ, Schwartz GJ, McHugh PR: Blockade of type A, but not type B, CCK receptors postpones satiety in rhesus monkeys. Am J Physiol 1993;265:R620–R624. Reidelberger RD, O’Rourke MF: Potent cholecystokinin antagonist L-364,718 stimulates food intake in rats. Am J Physiol 1989;257:R1512–R1518. Gibbs J, Young RC, Smith GP: Cholecystokinin decreases food intake in rats. J Comp Physiol Psychol 1973;84:488–495. Gibbs J, Young RC, Smith GP: Cholecystokinin elicits satiety in rats with open gastric fistulas. Nature 1973;245:323–325. Kulkosky PJ, Breckenridge C, Krinsky R, Woods SC: Satiety elicited by the C-terminal octapeptide of cholecystokinin-pancreozymin in normal and VMH-lesioned rats. Behav Biol 1976; 18/2:227–234. Smith GP, Gibbs J: The development and proof of the cholecystokinin hypothesis of satiety; in Dourish CT, Cooper SJ, Iversen SD, Iversen LL (eds): Multiple Cholecystokinin Receptors in the CNS. Oxford, Oxford University Press, 1992, pp 166–182. Smith GP, Gibbs J: Satiating effect of cholecystokinin. Ann NY Acad Sci 1994;713:236–241. West DB, Fey D, Woods SC: Cholecystokinin persistently suppresses meal size but not food intake in free-feeding rats. Am J Physiol 1984;246:R776–R787. West DB, Greenwood MRC, Marshall KA, Woods SC: Lithium chloride, cholecystokinin and meal patterns: Evidence the cholecystokinin suppresses meal size in rats without causing malaise. Appetite 1987;8:221–227. Inui A: Ghrelin: An orexigenic and somatotrophic signal from the stomach. Nature Rev Neurosci 2001;2:551–560. Cummings DE, Purnell JQ, Frayo RS, Schmidova K, Wisse BE, Weigle DS: A preprandial rise in plasma ghrelin levels suggests a role in meal initiation in humans. Diabetes 2001;50:1714–1719. Asakawa A, Inui A, Kaga T, Yuzuriha H, Nagata T, Ueno N, Makino S, Fujimiya M, Niijima A, Fujino MA, Kasuga M: Ghrelin is an appetite-stimulatory signal from stomach with structural resemblance to motilin. Gastroenterology 2001;120:337–345. Tschop M, Smiley DL, Heiman ML: Ghrelin induces adiposity in rodents. Nature 2000;407: 908–913. Wren AM, Small CJ, Ward HL, Murphy KG, Dakin CL, Taheri S, Kennedy AR, Roberts GH, Morgan DG, Ghatei MA, Bloom SR: The novel hypothalamic peptide ghrelin stimulates food intake and growth hormone secretion. Endocrinology 2000;141:4325–4328. Wren AM, Seal LJ, Cohen MA, Brynes AE, Frost GS, Murphy KG, Dhillo WS, Ghatei MA, Bloom SR: Ghrelin enhances appetite and increases food intake in humans. J Clin Endocrinol Metab 2001;86:5992. Cummings DE, Clement K, Purnell JQ, Vaisse C, Foster KE, Frayo RS, Schwartz MW, Basdevant A, Weigle DS: Elevated plasma ghrelin levels in Prader Willi syndrome. Nat Med 2002;8:643–644. Ahima RS, Saper CB, Flier JS, Elmquist JK: Leptin regulation of neuroendocrine systems. Front Neuroendocrinol 2000;21:263–307. Woods SC, Seeley RJ: Insulin as an adiposity signal. Int J Obes Relat Metab Disord 2001; 25(suppl 5):S35–S38. Woods SC: Insulin and the brain: A mutual dependency. Prog Psychobiol Physiol Psychol 1996;16:53–81. Benoit SC, Air EL, Coolen LM, Strauss R, Jackman A, Clegg DJ, Seeley RJ, Woods SC: The catabolic action of insulin in the brain is mediated by melanocortins. J Neurosci 2002;22:9048–9052. Seeley R, Yagaloff K, Fisher S, Burn P, Thiele T, van DG, Baskin D, Schwartz M: Melanocortin receptors in leptin effects. Nature 1997;390:349. Air EL, Benoit SC, Clegg DJ, Seeley RJ, Woods SC: Insulin and leptin combine additively to reduce food intake and body weight in rats. Endocrinology 2002;143:2449–2452. Schwartz MW, Seeley RJ, Weigle DS, Burn P, Campfield LA, Baskin DG: Leptin increases hypothalamic proopiomelanocortin (POMC) mRNA expression in the rostral arcuate nucleus. Diabetes 1997;46:2119–2123. Benoit SC, Schwartz MW, Baskin DG, Woods SC, Seeley RJ: CNS melanocortin system involvement in the regulation of food intake and body weight. Horm Behav 2000;37:299–308.

Woods/Clegg

26

56

57

58

59

60 61 62 63 64 65

66

67 68 69 70 71

72 73 74

75

76 77 78

McMinn JE, Wilkinson CW, Havel PJ, Woods SC, Schwartz MW: Effect of intracerebroventricular alpha-MSH on food intake, adiposity, c-Fos induction, and neuropeptide expression. Am J Physiol 2000;279:R695–R703. Hagan MM, Rushing PA, Pritchard LM, Schwartz MW, Strack AM, Van der Ploeg HT, Woods SC, Seeley RJ: Long-term orexigenic effects of AgRP-(83–132) involve mechanisms other than melanocortin receptor blockade. Am J Physiol 2000;279:R47–R52. Hagan MM, Benoit SC, Rushing PA, Pritchard LM, Woods SC, Seeley RJ: Immediate and prolonged patterns of agouti-related peptide-(83–132)-induced c-Fos activation in hypothalamic and extrahypothalamic sites. Endocrinology 2001;142:1050–1056. Schwartz MW, Marks J, Sipols AJ, Baskin DG, Woods SC, Kahn SE, Porte D Jr: Central insulin administration reduces neuropeptide Y mRNA expression in the arcuate nucleus of food-deprived lean (Fa/Fa) but not obese (fa/fa) Zucker rats. Endocrinology 1991;128:2645–2647. Schwartz MW, Seeley RJ, Campfield LA, Burn P, Baskin DG: Identification of targets of leptin action in rat hypothalamus. J Clin Invest 1996;98:1101–1106. Billington CJ, Briggs JE, Grace M, Levine AS: Effects of intracerebroventricular injection of neuropeptide Y on energy metabolism. Am J Physiol 1991;260:R321–R327. Clark JT, Kalra PS, Crowley WR, Kalra SP: Neuropeptide Y and human pancreatic polypeptide stimulate feeding behavior in rats. Endocrinology 1984;115:427–429. Morley JE, Levine AS, Gosnell BA, Kneip J, Grace M: Effect of neuropeptide Y on ingestive behaviors in the rat. Am J Physiol 1987;252:R599–R609. Stanley BG, Leibowitz SF: Neuropeptide Y injected into the paraventricular hypothalamus: A powerful stimulant of feeding behavior. Proc Natl Acad Sci USA 1984;82:3940–3943. Barrachina MD, Martinez V, Wang L, Wei JY, Tache Y: Synergistic interaction between leptin and cholecystokinin to reduce short-term food intake in lean mice. Proc Natl Acad Sci USA 1997;94: 10455–10460. Figlewicz DP, Sipols AJ, Seeley RJ, Chavez M, Woods SC, Porte DJ: Intraventricular insulin enhances the meal-suppressive efficacy of intraventricular cholecystokinin octapeptide in the baboon. Behav Neurosci 1995;109:567–569. Matson CA, Wiater MF, Kuijper JL, Weigle DS: Synergy between leptin and cholecystokinin (CCK) to control daily caloric intake. Peptides 1997;18:1275–1278. Matson CA, Ritter RC: Long-term CCK-leptin synergy suggests a role for CCK in the regulation of body weight. Am J Physiol 1999;276:R1038–R1045. Matson CA, Reid DF, Cannon TA, Ritter RC: Cholecystokinin and leptin act synergistically to reduce body weight. Am J Physiol 2000;278:R882–R890. Riedy CA, Chavez M, Figlewicz DP, Woods SC: Central insulin enhances sensitivity to cholecystokinin. Physiol Behav 1995;58:755–760. Flegal KM, Ogden CL, Wei R, Kuczmarski RL, Johnson CL: Prevalence of overweight in US children: Comparison of US growth charts from the Centers for Disease Control and Prevention with other reference values for body mass index. Am J Clin Nutr 2001;73:1086–1093. Hill JO, Peters JC: Environmental contributions to the obesity epidemic. Science 1998;280: 1371–1374. Ledbetter DH, Riccardi VM, Airhart SD, Strobel RJ, Keenan BS, Crawford JD: Deletions of chromosome 15 as a cause of the Prader-Willi syndrome. N Engl J Med 1981;304:325–329. Knoll JH, Nicholls RD, Magenis RE, Graham JM Jr, Lalande M, Latt SA: Angelman and PraderWilli syndromes share a common chromosome 15 deletion but differ in parental origin of the deletion. Am J Med Genet 1989;32:285–290. Bray GA, Dahms WT, Swerdloff RS, Fiser RH, Atkinson RL, Carrel RE: The Prader-Willi syndrome: A study of 40 patients and a review of the literature. Medicine (Baltimore) 1983;62/2: 59–80. Ehara H, Ohno K, Takeshita K: Frequency of the Prader-Willi syndrome in the San-in district, Japan. Brain Dev 1995;17:324–326. Burman P, Ritzen EM, Lindgren AC: Endocrine dysfunction in Prader-Willi syndrome: A review with special reference to GH. Endocr Rev 2001;22:787–799. Laurance BM: Feeding babies. Nutr Health 1993;9/2:73–74.

Signals that Control Central Appetite Regulation

27

79

80 81 82 83 84 85

86

87 88

89 90

91

92

93 94

95

96

97

98

Meaney FJ, Butler MG: The developing role of anthropologists in medical genetics: Anthropometric assessment of the Prader-Labhart-Willi syndrome as an illustration. Med Anthropol 1989; 10/4:247–253. Schoeller DA, Levitsky LL, Bandini LG, Dietz WW, Walczak A: Energy expenditure and body composition in Prader-Willi syndrome. Metabolism 1988;37/2:115–120. Brambilla P, Bosio L, Manzoni P, Pietrobelli A, Beccaria L, Chiumello G: Peculiar body composition in patients with Prader-Labhart-Willi syndrome. Am J Clin Nutr 1997;65:1369–1374. Davies PS, Joughin C: Assessment of body composition in the Prader-Willi syndrome using bioelectrical impedance. Am J Med Genet 1992;44/1:75–78. Greenswag LR: Adults with Prader-Willi syndrome: A survey of 232 cases. Dev Med Child Neurol 1987;29/2:145–152. Laurance BM, Brito A, Wilkinson J: Prader-Willi syndrome after age 15 years. Arch Dis Child 1981;56/3:181–186. Goldstone AP, Thomas EL, Brynes AE, Bell JD, Frost G, Saeed N, Hajnal JV, Howard JK, Holland A, Bloom SR: Visceral adipose tissue and metabolic complications of obesity are reduced in Prader-Willi syndrome female adults: Evidence for novel influences on body fat distribution. J Clin Endocrinol Metab 2001;86:4330–4338. Goldstone AP, Brynes AE, Thomas EL, Bell JD, Frost G, Holland A, Ghatei MA, Bloom SR: Resting metabolic rate, plasma leptin concentrations, leptin receptor expression, and adipose tissue measured by whole-body magnetic resonance imaging in women with Prader-Willi syndrome. Am J Clin Nutr 2002;75:468–475. Eiholzer U, Stutz K, Weinmann C, Torresani T, Molinari L, Prader A: Low insulin, IGF-I and IGFBP-3 levels in children with Prader-Labhart-Willi syndrome. Eur J Pediatr 1998;157:890–893. Lindgren AC, Hagenas L, Ritzen EM: Growth hormone treatment of children with Prader-Willi syndrome: Effects on glucose and insulin homeostasis. Swedish National Growth Hormone Advisory Group. Horm Res 1999;51/4:157–161. Schuster DP, Osei K, Zipf WB: Characterization of alterations in glucose and insulin metabolism in Prader-Willi subjects. Metabolism 1996;45:1514–1520. Hennes MM, Dua A, Maas DL, Sonnenberg GE, Krakower GR, Kissebah AH: Relationships of plasma leptin levels to changes in plasma free fatty acids in women who are lean and women who are abdominally obese. Obes Res 1997;5:442–446. Yamashita S, Nakamura T, Shimomura I, Nishida M, Yoshida S, Kotani K, Kameda-Takemuara K, Tokunaga K, Matsuzawa Y: Insulin resistance and body fat distribution. Diabetes Care 1996;19/3: 287–291. Bueno G, Moreno LA, Pineda I, Campos J, Ruibal JL, Juste MG, Casado E, Bueno M: Serum leptin concentrations in children with Prader-Willi syndrome and non-syndromal obesity. J Pediatr Endocrinol Metab 2000;13:425–430. Butler MG, Moore J, Morawiecki A, Nicolson M: Comparison of leptin protein levels in PraderWilli syndrome and control individuals. Am J Med Genet 1998;75/1:7–12. Carlson MG, Snead WL, Oeser AM, Butler MG: Plasma leptin concentrations in lean and obese human subjects and Prader-Willi syndrome: Comparison of RIA and ELISA methods. J Lab Clin Med 1999;133/1:75–80. Cento RM, Proto C, Spada RS, Ragusa L, Reitano S, Napolitano V, Lanzone A: Serum leptin concentrations in obese women with Down syndrome and Prader-Willi syndrome. Gynecol Endocrinol 1999;13/1:36–41. Lindgren AC, Marcus C, Skwirut C, Elimam A, Hagenas L, Schalling M, Anvret M, Lonnqvist F: Increased leptin messenger RNA and serum leptin levels in children with Prader-Willi syndrome and nonsyndromal obesity. Pediatr Res 1997;42:593–596. Myers SE, Davis A, Whitman BY, Santiago JV, Landt M: Leptin concentrations in Prader-Willi syndrome before and after growth hormone replacement. Clin Endocrinol (Oxf) 2000; 52/1:101–105. Wallace AM, Hunter I, Galloway P, Greene SA, Donaldson MD: Obesity in the Prader-Labhart-Willi syndrome is not due to leptin deficiency but is accentuated by hypogonadism in male patients. Clin Endocrinol (Oxf) 1999;51:816–817.

Woods/Clegg

28

99 Weigle DS, Ganter SL, Kuijper JL, Leonetti DL, Boyko EJ, Fujimoto WY: Effect of regional fat distribution and Prader-Willi syndrome on plasma leptin levels. J Clin Endocrinol Metab 1997; 82:566–570. 100 Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K: Ghrelin is a growthhormone-releasing acylated peptide from stomach. Nature 1999;402:656–660. 101 Horvath TL, Diano S, Sotonyi P, Heiman M, Tschop M: Minireview: Ghrelin and the regulation of energy balance – A hypothalamic perspective. Endocrinology 2001;142:4163–4169. 102 Nakazato M, Murakami N, Date Y, Kojima M, Matsuo H, Kangawa K, Matsukura S: A role for ghrelin in the central regulation of feeding. Nature 2001;409:194–198. 103 Nicolls MR, Coulombe M, Gill RG: The basis of immunogenicity of endocrine allografts. Crit Rev Immunol 2001;21/1–3:87–101. 104 Cummings DE, Weigle DS, Frayo RS, Breen PA, Ma MK, Dellinger EP, Purnell JQ: Plasma ghrelin levels after diet-induced weight loss or gastric bypass surgery. N Engl J Med 2002;346:1623–1630. 105 Zipf WB, O’Dorisio TM, Berntson GG: Short-term infusion of pancreatic polypeptide: Effect on children with Prader-Willi syndrome. Am J Clin Nutr 1990;51:162–166. 106 Tomita T, Greeley G Jr, Watt L, Doull V, Chance R: Protein meal-stimulated pancreatic polypeptide secretion in Prader-Willi syndrome of adults. Pancreas 1989;4:395–400. 107 Holland AJ, Treasure J, Coskeran P, Dallow J, Milton N, Hillhouse E: Measurement of excessive appetite and metabolic changes in Prader-Willi syndrome. Int J Obes Relat Metab Disord 1993;17:527–532. 108 Butler MG, Carlson MG, Schmidt DE, Feurer ID, Thompson T: Plasma cholecystokinin levels in Prader-Willi syndrome and obese subjects. Am J Med Genet 2000;95/1:67–70. 109 Marchini G, Simoni MR, Bartolini F, Linden A: The relationship of plasma cholecystokinin levels to different feeding routines in newborn infants. Early Hum Dev 1993;35/1:31–35. 110 Billington CJ, Levine AS, Morley JE: Are peptides truly satiety agents? A method of testing for neurohumoral satiety effects. Am J Physiol 1983;245:R920–R926. 111 Zipf WB, O’Dorisio TM, Cataland S, Sotos J: Blunted pancreatic polypeptide responses in children with obesity of Prader-Willi syndrome. J Clin Endocrinol Metab 1981;52:1264–1266. 112 Zipf WB, O’Dorisio TM, Cataland S, Dixon K: Pancreatic polypeptide responses to protein meal challenges in obese but otherwise normal children and obese children with Prader-Willi syndrome. J Clin Endocrinol Metab 1983;57:1074–1080. 113 Glaser B, Zoghlin G, Pienta K, Vinik AI: Pancreatic polypeptide response to secretin in obesity: Effects of glucose intolerance. Horm Metab Res 1988;20/5:288–292. 114 Holst JJ, Schwartz TW, Lovgreen NA, Pedersen O, Beck-Nielsen H: Diurnal profile of pancreatic polypeptide, pancreatic glucagon, gut glucagon and insulin in human morbid obesity. Int J Obes 1983;7:529–538. 115 Swaab DF, Purba JS, Hofman MA: Alterations in the hypothalamic paraventricular nucleus and its oxytocin neurons (putative satiety cells) in Prader-Willi syndrome: A study of five cases. J Clin Endocrinol Metab 1995;80:573–579. 116 Gibbs J, Fauser DJ, Rowe EA, Rolls ET, Maddison SP: Bombesin suppresses feeding in rats. Nature 1979;282:208–210. 117 Gibbs J, Guss JL: Bombesin-like peptides and satiety. Appetite 1995;24/3:257. 118 Stein LJ, Woods SC: Gastrin releasing peptide reduces meal size in rats. Peptides 1982;3:833–835. 119 Ladenheim EE, Taylor JE, Coy DH, Moran TH: Blockade of feeding inhibition by neuromedin B using a selective receptor antagonist. Eur J Pharmacol 1994;271/1:R7–9. 120 Ladenheim EE, Taylor JE, Coy DH, Carrigan TS, Wohn A, Moran TH: Caudal hindbrain neuromedin B-preferring receptors participate in the control of food intake. Am J Physiol 1997;272:R433–R437. 121 Erlanson-Albertsson C, York D: Enterostatin – A peptide regulating fat intake. Obes Res 1997;5: 360–372. 122 Okada S, York DA, Bray GA, Erlanson-Albertsson C: Enterostatin (Val-Pro-Asp-Pro-Arg), the activation peptide of procolipase, selectively reduces fat intake. Physiol Behav 1991;49:1185–1189. 123 Shargill NS, Tsuji S, Bray GA, Erlanson-Albertsson C: Enterostatin suppresses food intake following injection into the third ventricle of rats. Brain Res 1991;544:137–140. 124 Morley JE, Flood JF: Amylin decreases food intake in mice. Peptides 1991;12:865–869.

Signals that Control Central Appetite Regulation

29

125 Fujimoto K, Machidori H, Iwakiri R, Yamamoto K, Fujisaki J, Sakata T, Tso P: Effect of intravenous administration of apolipoprotein A-IV on patterns of feeding, drinking and ambulatory activity in rats. Brain Res 1993;608:233–237. 126 Fujimoto K, Fukagawa K, Sakata T, Tso P: Suppression of food intake by apolioprotein A-IV is mediated through the central nervous system in rats. J Clin Invest 1993;91:1830–1833. 127 Geary N: Glucagon and the control of meal size; in Smith GP (ed): Satiation. From Gut to Brain. New York, Oxford University Press, 1998, pp 164–197. 128 Salter JM: Metabolic effects of glucagon in the Wistar rat. Am J Clin Nutr 1960;8:535–539. 129 Lotter EC, Krinsky R, McKay JM, Treneer CM, Porte D Jr, Woods SC: Somatostatin decreases food intake of rats and baboons. J Comp Physiol Psychol 1981;95/2:278–287. 130 Tang-Christensen M, Larsen PJ, Goke R, Fink-Jensen A, Jessop DS, Moller M, Sheikh SP: Central administration of GLP-1-(7–36) amide inhibits food and water intake in rats. Am J Physiol 1996; 271:R848–R856. 131 Turton MD, O’Shea D, Gunn I, Beak SA, Edwards CMB, Meeran K, Choi SJ, Taylor GM, Heath MM, Lambert PD, Wilding JPH, Smith DM, Ghatei MA, Herbert J, Bloom SR: A role for glucagon-like peptide-1 in the central regulation of feeding. Nature 1996;379:69–72.

Stephen C. Woods, PhD, Professor of Psychiatry, Department of Psychiatry, Box 0559, University of Cincinnati, Cincinnati, OH 45267 (USA) Tel. ⫹1 513 558 6799, Fax ⫹1 513 558 8990, E-Mail [email protected]

Woods/Clegg

30

Central Nervous System and Body Weight Homeostasis Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 31–43

Hypothalamic Neuropeptides and Regulation of Fat Mass in Prader-Willi Syndrome Anthony P. Goldstonea, b, f, Unga A. Unmehopaf, E. Louise Thomasc, Audrey E. Brynesd, Jimmy D. Bellc, Gary Frostd, Mohammad A. Ghateib, Anthony Hollande, Stephen R. Bloomb, Dick F. Swaabf a

Department of Endocrinology, St. Bartholomew’s Hospital, bEndocrine Unit, MRI Unit, MRC Clinical Sciences Centre and dDepartment of Dietetics, Imperial College School of Medicine, Hammersmith Hospital, London, and eSection of Developmental Psychiatry, University of Cambridge, UK, fGraduate School Neurosciences Amsterdam, Netherlands Institute for Brain Research, Amsterdam, The Netherlands

c

Abstract Childhood hyperphagia and obesity, growth hormone (GH) deficiency and hypogonadism in Prader-Willi syndrome (PWS) are thought to arise from developmental abnormalities in the hypothalamus, or its signalling inputs, but their nature is unclear. Body composition studies, using whole body magnetic resonance imaging, have found PWS adults to have greater fat mass relative to fat-free mass, but to have significantly less visceral adiposity, compared to controls, which appears to protect against metabolic complications of obesity, such as insulin resistance and hypertriglyceridaemia. These findings may reflect hormonal, hypothalamic, developmental and/or genetic influences on body composition and fat distribution. Studies have revealed normal leptin secretion, but elevated circulating ghrelin, that may have a role in the hyperphagia, and perhaps GH deficiency in PWS. Postmortem studies of hypothalami from PWS subjects have not found any abnormalities in orexigenic neuropeptide Y (NPY) and agouti-related protein (AGRP) neurons, growth hormone-releasing hormone (GHRH) neurons, or, in preliminary studies, anorexigenic pro-opiomelanocortin (POMC) or cocaine- and amphetamine-regulated transcript (CART) neurons, but the total and oxytocin neuron numbers are reduced in the paraventricular nucleus, which is likely to have some phenotypic role. Copyright © 2003 S. Karger AG, Basel

Introduction

In 1929, Harvey Cushing described the hypothalamus as containing ‘the very main spring of primitive existence – vegetative, emotional and reproductive – on which with more or less success, man has come to superimpose a cortex of inhibitions’. Defects within such processes seen in Prader-Willi syndrome (PWS), especially gross hyperphagia and obesity, hypogonadism and growth hormone (GH) deficiency, thus suggest the presence of developmental abnormalities of the hypothalamus, or its signalling inputs (table 1) [1–3]. Identification of such abnormalities in PWS may reveal potential therapies for this disease, and the role of imprinting in the control of hypothalamic gene expression, and help elucidate crucial human metabolic pathways [4, 5].

Body Composition and Fat Distribution in PWS

Using whole body magnetic resonance imaging (MRI) in PWS adult females, we have found a marked increase in adipose tissue (AT), relative to lean tissue (fig. 1a) [6, 7], as also seen in PWS children and adults using dual energy X-ray absorptiometry and isotopic dilution [8, 9]. This is probably explicable by increased calorific intake, GH deficiency, hypogonadism and reduced physical activity in PWS [3, 9], and leads to a reduced metabolic rate in PWS [6, 10]. PWS adult females (n
13) had significantly reduced visceral AT volume, compared to controls [visceral AT/subcutaneous AT ratio: 0.067  0.017 vs. 0.108  0.021 for non-PWS obese females (n
14), p  0.001, correcting for total adiposity], while no reduction in abdominal subcutaneous AT was seen (fig. 1b, c) [7]. We have recently confirmed that this reduction in visceral adiposity is also seen in PWS adult males (n
6), compared to controls [visceral AT/subcutaneous AT ratio: 0.090  0.034 vs. 0.265  0.060 for non-PWS obese males (n
5), p
0.02 correcting for age and total adiposity] (fig. 1d). The lower visceral adiposity in PWS women was associated with less insulin resistance, as measured by fasting insulin and insulin/glucose ratio, and hypertriglyceridaemia, and greater hepatic insulin extraction, as measured by C-peptide/insulin ratio, compared to obese controls [7]. The reduced visceral obesity in PWS adults occurs despite the presence of many phenotypes that would be expected to increase visceral adiposity (table 2), and could be explained by subtle impairment of the hypothalamicpituitary-adrenal axis, long-term consequences of intra-uterine and postnatal failure to thrive, lack of expression of imprinted PWS candidate genes or haploinsufficiency for other genes in the PWS chromosomal region, and childhood onset GH deficiency [3, 7, 11, 12].

Goldstone/Unmehopa/Thomas/Brynes/Bell/Frost/Ghatei/Holland/Bloom/Swaab

32

Table 1. Metabolic, hormonal and hypothalamic neuropeptide phenotypes in PWS Metabolic phenotypes Increased AT and reduced lean tissue Reduced relative visceral adiposity and metabolic consequences of obesity Reduced physical activity Reduced resting metabolic rate explicable by body composition Hormonal phenotypes Reduced circulating basal GH and IGF-1 secretion Reduced GH response to exercise, insulin-induced hypoglycaemia, glucagon, arginine, clonidine, GHRH  pyridostigmine, or GH secretagogues Hypogonadism in males, oligo/amenorrhoea in females Low testosterone in males, oestradiol levels often in follicular range in females Usually non-elevated basal gonadotrophins (except some males, possibly related to cryptorchidism) Subnormal LH and FSH responses to GnRH, but normal after prolonged treatment with clomiphene or repeated GnRH Normal thyroid function, but response to TRH may be exaggerated Grossly normal hypothalamic-pituitary-adrenal axis Normal leptin secretion Long isoform leptin receptor mRNA expressed in lymphocytes Increased circulating ghrelin Hypothalamic neuropeptide phenotypes Normal neuroanatomy of NPY, AGRP and POMC neurons in INF Normal co-localization of NPY and AGRP in INF neurons Normal increase in NPY (ICC staining or mRNA expression) or AGRP (ICC staining) in INF during illness Reduced NPY (ICC staining or mRNA expression) in INF compared to control, but not non-PWS obese adults, correcting for premorbid illness duration Normal AGRP (ICC staining) in INF compared to control or non-PWS obese adults, correcting for premorbid illness duration Normal increase in GHRH neuron cell number in INF during illness Normal GHRH neuron cell number in INF compared to control or non-PWS obese adults, correcting for premorbid illness duration and sex Reduced GHRH neuron cell number in INF in PWS child receiving exogenous GH treatment Normal neuroanatomy of oxytocin and vasopressin neurons in PVN Reduced total and oxytocin neuron cell number in PVN Reduced oxytocin neuron cell number in PVN (ndn knockout mouse) Normal vasopressin neuron cell number in PVN Normal neuroanatomy of CART neurons in INF, PVN and LHA No complete deficiency of POMC neurons in INF, or CART neurons in INF, PVN or LHA Reduced LHRH neuron cell number in pre-optic area (ndn knockout mouse) Normal neuroanatomy of MCH neurons in LHA

Hypothalamic Neuropeptides and Fat Mass in PWS

33

PWS female: r
 0.81, P 0.001 Control female: r
 0.92, P 0.001

100

8 Visceral AT (l)

Fat mass (kg)

80 60 40

PWS vs. control: P  0.001

4

40

30

a

50

60

PWS female: r
0.91, P0.001 PWS vs. control: P0.001

0

0

0

70

20

Fat-free mass (kg)

30

40

60

80

100

120

b

Total AT (l)

10

Control female: r
0.98, P  0.001 PWS female: r
 0.96, P 0.001

35

Control male: r
0.87, P0.001

8

25

Visceral AT (l)

Abdominal SCAT (l)

6

2

20

20 15

6 4

10 2

PWS male: r
0.89, P0.02 PWS vs. control: P0.001

PWS vs. control: P
0.1

5

0

0 0

c

Control female: r
0.93, P0.001

10

20

40

60 Total AT (l)

80

100

120

0

20

40

60

80

100

d

Total AT (l)

Fig. 1. Body composition and fat distribution in PWS adults. Relationship between fat mass and fat-free mass (a) in control (䊊, solid regression line) and PWS (䊏, dashed regression line) adult females; total AT volume and visceral AT volume (b) or abdominal subcutaneous AT (SCAT) volume (c) in control (䊊, solid regression line) and PWS (䊏, dashed regression line) adult females; total AT volume and visceral AT (d) in control (䉭, solid regression line) and PWS (䉱, dashed regression line) adult males, as measured by whole body magnetic resonance imaging. r indicates Pearson regression coefficient. Note that PWS adults have greater fat mass relative to fat-free mass, and relatively less visceral adiposity, but not abdominal subcutaneous AT, compared to controls [adapted from 6, reproduced with permission by the American Journal of Clinical Nutrition, copyright Am J Clin Nutr American Society for Clinical Nutrition (a), and from 7 (b, c), copyright owner The Endocrine Society].

This protection against visceral adiposity, insulin resistance and other metabolic complications of obesity would reduce the relative risk of development of type 2 diabetes mellitus and cardiovascular complications in PWS, although absolute rates will still be increased. Although there is much evidence

Goldstone/Unmehopa/Thomas/Brynes/Bell/Frost/Ghatei/Holland/Bloom/Swaab

34

Table 2. Effect of hormonal and behavioral phenotypes on visceral adiposity in PWS Phenotype

Effect on visceral adiposity

Presence of phenotype in PWS

Obesity Reduced physical activity Hypogonadism (male or female) GH deficiency

Increases Increases Increases Increases (in adults) Effect of childhood onset? Increases Reduced risk of obesity Effect on visceral adiposity? No change Decreases? Decreases?

Present Present Present Childhood onset

Stress, depression and anxiety Late gestational or early postnatal malnutrition Incomplete or delayed puberty Glucocorticoid deficiency Androgen deficiency (females)

Present Present Present Absent Absent

that adult-onset GH deficiency increases visceral adiposity, and GH treatment to adults with either childhood-onset or adult-onset GH deficiency reduces visceral adiposity, there is no published data on the influence of childhood-onset GH deficiency on visceral adiposity [13]. Indeed rodent studies would support a hypothesis that GH deficiency during early development in PWS could reduce visceral adiposity [14], which might be related to contrasting effects of GH to increase lipolysis, but also to increase adipocyte differentiation and proliferation.

Leptin and PWS

Leptin, the adipocyte-derived plasma hormone, interacts with brain pathways, particularly in the hypothalamus, to reduce food intake and body weight, as well as stimulate the HP-gonadal axis [15]. The hyperphagia, obesity and hypogonadism in PWS mirrors that seen in leptin deficiency and leptin receptor defects [16]. However, we have found no evidence for leptin deficiency in PWS adult females, adjusting for total and subcutaneous adiposity measured by whole body MRI, necessary given the primary influence of this fat depot on circulating leptin (fig. 2) [6], in agreement with other studies in PWS children and adults, using body mass index or X-ray absorptiometry to assess body composition [17, 18]. We have demonstrated that common and long isoform leptin receptor mRNA are expressed in PWS lymphocytes using RT-PCR [6]. This suggests that PWS region imprinted genes are not essential for leptin receptor

Hypothalamic Neuropeptides and Fat Mass in PWS

35

PWS vs. control: P
1.0

200

200

Control: r
 0.78, P 0.001 PWS: r
 0.85, P 0.001

Plasma leptin (g/l)

Plasma leptin (g/l)

10

10 Control: r
0.78, P0.001 PWS: r
0.85, P0.001 1

1 0

a

PWS vs. control: P
0.8

100

100

20

40

60

80

Total AT (l)

100

120

0

20

40

60

80

100

120

b

Subcutaneous AT (l)

Fig. 2. Plasma leptin in control and PWS adult females. Relationship between plasma leptin and total (a) or subcutaneous AT (b) volume, as determined by whole body MRI, in adult control females (䊊, solid regression line) and PWS females (䊏, dashed regression line). r indicates Pearson correlation coefficient. Note that the relationship between plasma leptin and total or subcutaneous AT is similar in control and PWS subjects [reproduced with permission by the American Journal of Clinical Nutrition, copyright Am J Clin Nutr American Society for Clinical Nutrition, from 7].

alternative splicing or expression, at least in lymphocytes, but hypothalamic leptin receptor expression has not yet been reported in PWS.

Ghrelin and PWS

Ghrelin is an endogenous ligand at the GH secretagogue receptor, located in the hypothalamus and pituitary, to stimulate GH secretion from the pituitary [19]. Ghrelin circulates peripherally, being secreted by the stomach, particularly when fasting, peaking before meal initiation [20], and is also found in hypothalamic neurons and pituitary. Ghrelin acutely stimulates food intake in rodents and humans, and chronic administration to rodents causes obesity [19, 21, 22]. A recent study has found grossly elevated plasma ghrelin levels in PWS [23] that could contribute to their hyperphagia. This is not seen in other causes of obesity, including genetic leptin resistance [23, 24], and is not related to GH deficiency or incomplete puberty. It remains to be seen whether this has any pathophysiological role in GH deficiency in PWS, perhaps through GH secretagogue receptor desensitization or reduced receptor number.

Goldstone/Unmehopa/Thomas/Brynes/Bell/Frost/Ghatei/Holland/Bloom/Swaab

36

Hypothalamic Feeding Neuropeptides in PWS

The major stimulatory feeding neuropeptides are neuropeptide Y (NPY) and agouti-related protein (AGRP) [5]. The latter is an endogenous -melanocytestimulating hormone (-MSH) antagonist at the melanocortin-4 (MC4R) receptor, co-localized with NPY in arcuate nucleus (ARC) neurons, called the infundibular nucleus (INF) in humans. During starvation NPY/AGRP neurons are activated by the fall in plasma leptin [5], and the increase in plasma ghrelin [21], to increase food intake, and reduce energy expenditure. By contrast -MSH, produced by ARC pro-opiomelanocortin (POMC)-expressing neurons, inhibits feeding and increases energy expenditure via the MC4R, particularly in the paraventricular nucleus (PVN) and lateral hypothalamic area (LHA) [5]. Hypothalamic NPY and AGRP overactivity in rodents leads to hyperphagia and obesity, and for NPY, hypogonadism and GH deficiency, through actions on the hypothalamic-pituitary axis [5]. Mutations in POMC and MC4R genes result in childhood-onset obesity in humans [16]. The NPY and melanocortin pathways are therefore leading candidates as affected circuits in PWS. It was therefore hypothesized that increased hypothalamic NPY and AGRP activity, or decreased hypothalamic POMC/-MSH activity, might be part of the PWS pathological process. NPY and AGRP Neurons This hypothesis has been studied using postmortem PWS adult hypothalami in the Netherlands Brain Bank [25]. No abnormal hypothalamic neuroanatomy or co-localization of NPY and AGRP neurons was seen in PWS adults. Activity of NPY/AGRP neurons in the INF and median eminence (ME), as measured by NPY immunocytochemistry (ICC) staining or mRNA expression by in situ hybridization, and AGRP ICC staining, was, however, positively correlated with the duration of the premorbid illness, in both controls and obese PWS subjects (fig. 3a–c) [25]. There was no evidence for increased activity of these NPY/AGRP neurons in obese PWS adults. On the contrary, the results suggest decreased activity of NPY neurons, since NPY ICC staining was significantly decreased and there was a tendency for a similar decrease in NPY mRNA expression, correcting for illness duration (fig. 3a, b) [25]. Similar reductions in NPY were seen in non-PWS obesity. AGRP peptide ICC staining was unchanged in PWS and non-PWS obese adults, compared to controls (fig. 3c) [25]. These results suggest an appropriate response of NPY/AGRP neurons to changes in peripheral signals during illness and resulting from obesity in PWS, and support the conclusion that NPY/AGRP neurons in the INF/ME are not defective in PWS. The pathophysiology of obesity in PWS subjects may therefore lie in downstream or separate leptin-dependent or

Hypothalamic Neuropeptides and Fat Mass in PWS

37

1

0.1 Control: r
 0.30, P
0.18 PWS: r
 0.92, P
0.01

NPY mRNA expression (arbitrary units)

NPY ICC masking volume (mm3)

PWS vs. control: P  0.001

5

0.01

10

1 Control: r
 0.71, P0.001 PWS: r
 0.66, P
0.01 0.1

0.1

1

10

100

0.1

1

10

100

b 1

200

PWS vs. control: P
0.90

0.1

0.01 Control: r
 0.82, P 0.001 PWS: r
 0.91, P
0.01 0.001 0.1

1

10

Duration premorbid illness (days)

100

GHRH cell number (103)

AGRP ICC masking volume (mm3)

a

c

PWS vs. control: P
0.07

200 100

PWS vs. control: P
0.78

100

Control: r
0.62, P
0.002 PWS: r
0.90, P
0.02

10 7 0.1

1

10

100

d

Duration premorbid illness (days)

Fig. 3. Hypothalamic NPY, AGRP and GHRH neurons in PWS adults. Relationship between NPY ICC staining volumes (a), NPY mRNA expression (b), AGRP ICC staining volumes (c), and GHRH cell number (d) in the INF/ME, and duration of premorbid illness in control adults (䊊, solid regression line), and obese PWS adults (). Note that the y-axes have log10 scales. Note that NPY peptide and mRNA, AGRP peptide and GHRH cell number increase with illness duration in both control and obese PWS subjects. Note that correcting for illness duration, NPY ICC staining and mRNA expression, but not AGRP ICC staining or GHRH cell number, appear lower in PWS subjects compared to controls. r represents Pearson correlation coefficient [adapted from 25 with permission; copyright owner The Endocrine Society].

leptin-independent feeding circuits. Furthermore, since ghrelin is thought to act within the hypothalamus to stimulate feeding by increasing activity of NPY and AGRP neurons [21], it remains to be determined why activation of these neurons was not seen, which could question the potential role of elevated plasma ghrelin in causing PWS hyperphagia [23]. The confounding effect of premorbid illness is one possible explanation.

Goldstone/Unmehopa/Thomas/Brynes/Bell/Frost/Ghatei/Holland/Bloom/Swaab

38

POMC and Cocaine- and Amphetamine-Regulated Transcript Neurons Cocaine- and amphetamine-regulated transcript (CART) is an orexigenic peptide, co-localized to POMC neurons in the ARC that are activated by leptin and inhibited by fasting, and also present in PVN and LHA neurons [5]. Studies have demonstrated the presence of POMC neurons in one section of the INF from all the PWS adult subjects studied previously, and have found CART neurons throughout the INF, PVN or LHA of PWS hypothalami [Goldstone et al., unpubl. observations]. This suggests that complete absence of POMC and CART neurons is not a cause of their hyperphagia, but a quantitative analysis of their cell number in the INF will be needed. Oxytocin Neurons A reduction in total (38%) and oxytocin (42%) cell number in the PVN of PWS adults, but no change in the vasopressin cell number has been reported [26]. Rodent studies have indicated that oxytocin is an orexigenic neuropeptide, which may mediate the inhibitory effects of gastric distension, CCK, lithium and CRH on food intake, through brainstem projections [27]. A reduction in oxytocin neurons in PWS may, therefore, play a causative role in the hyperphagia. This hypothesis, and its possible role in other PWS phenotypes, should be viewed in the context of only a few studies reporting stimulation of spontaneous feeding with antagonism of endogenous oxytocin; the normal feeding behaviour, body weight and fertility of oxytocin knockout mice, and a 29% reduction in PVN oxytocin neurons (whose projection and function are unknown) seen in ndn knockout mice, despite a lack of infertility, or obesity, but presence of some other PWS phenotypes, including neonatal lethality, improved spatial learning and memory and increased skin scraping activity [28]. Melanin-Concentrating Hormone Neurons The LHA contains neurons expressing the orexigenic peptide, melaninconcentrating hormone (MCH) [5]. Our pilot studies have demonstrated no obvious deficiency or excess of MCH neurons in the LHA, perifornical area and zona incerta of PWS hypothalami, but quantitative studies have not been performed [Goldstone et al., unpubl. observations].

Hypothalamic-Pituitary Neuropeptides in PWS

GH-Releasing Hormone Neurons Using postmortem material, the distribution of GHRH neuronal cell bodies in the INF and fibre terminal projections to the ME were similar in PWS

Hypothalamic Neuropeptides and Fat Mass in PWS

39

adults and controls [29]. A significant positive correlation was found between GHRH cell number in the INF/ME and the duration of the premorbid illness in both control and PWS adults (fig. 3d) [29]. This would reflect appropriate loss of negative feedback of GH and insulin-like growth factor 1 (IGF-1) to the hypothalamus during illness [30], and suggests normal GHRH neuronal function in PWS. GHRH cell number was similar in PWS, and control or non-PWS obese adults, adjusting for sex and illness duration, giving no evidence that the GH deficiency in PWS results from developmental deficiency of GHRH cell number [29]. As with all postmortem human studies the confounding effect of premorbid illness may limit such interpretations, but there was no suggestion of any reduction in GHRH neuronal cell number in PWS subjects who died suddenly, including one PWS child. A low GHRH cell number was found in another PWS child, who had been receiving exogenous GH therapy [29], which again suggests a normal response of GHRH neurons to the negative feedback of GH and IGF-1 in PWS. GnRH Neurons Ndn knockout mice have a 25% reduction in LHRH neurons in the medial pre-optic area, with necdin mRNA being co-localized to LHRH neurons [28]. However since these mice are not infertile, it is unclear what effect this has on reproductive function. GnRH neurons have not yet been examined in postmortem human PWS hypothalami.

Conclusion

PWS is associated with many metabolic, hormonal, hypothalamic and neuropeptide abnormalities (table 1). Many of the phenotypes in PWS may be explicable by developmental abnormalities in the hypothalamus or its peripheral inputs. Elucidation of these abnormalities has involved measuring circulating hormones and examining postmortem human hypothalamic PWS material, but has been made difficult by the lack of suitable mouse models for many PWS phenotypes. Several candidate PWS genes have, however, been implicated in failure to thrive and neonatal lethality, and some behavioural features, but the neuropathological and neuroendocrine consequences are unclear at present. The peculiar body composition in PWS with increased fat, reduced lean tissue, and reduced visceral adiposity that protects against metabolic consequences of obesity, probably reflects behavioural, developmental, hormonal and genetic abnormalities. Clinical studies have revealed normal leptin secretion, but elevated circulating ghrelin, that may have a role in the hyperphagia or GH deficiency of PWS. Postmortem human hypothalamic studies have found

Goldstone/Unmehopa/Thomas/Brynes/Bell/Frost/Ghatei/Holland/Bloom/Swaab

40

normal responses of orexigenic NPY and AGRP neurons, and GHRH neurons, to premorbid illness in PWS, but have not found evidence for overactivity of the former or deficiency of the latter that might explain hyperphagia and GH deficiency (although the effects of illness may complicate these conclusions). Limited studies have not found complete deficiency of hypothalamic POMC or CART neurons in PWS. Total and oxytocin neuron numbers are, however, reduced in the PVN, which is likely to have some phenotypic role. The complexity of hypothalamic neuropeptide pathways and the PWS chromosomal region will make the identification of the hypothalamic defects, and their links with PWS genes, a difficult challenge. The possibility of treating and preventing the clinical problems in PWS awaits such progress. Acknowledgments R.S. Williams (USA), A. Schulze and M. Bojsen-Møller (Denmark), A. Holland and J. Xuereb (UK), M.E.J. Schipper, H.M. Evenhuis and R.A.C. Roos (The Netherlands), P.T. Botha and L. Thornton (New Zealand), U. Eiholzer and C. Markwalder (Switzerland) for provision of brain material and clinical information from PWS subjects; Netherlands Brain Bank staff (R. Ravid, W. Kamphorst, F.C. Stam, P. van der Valk, B. Fisser, A. Holtrop, J. Wouda, M. Kooreman, M. Kahlmann and A.C.E. de Nijs) for the provision and processing of hypothalamic material and clinical information; patients, their families and other clinicians for tissue donation; the UK PWS Association, PWS patients, their carers and families for their support and keen participation in clinical studies; J.J. van Heerikhuize and C.W. Pool for their technical assistance; N. Saeed and J. Hajnal for assistance with MRI scanning; J.K. Howard for assistance with patients; T. Gotoda for assistance with RT-PCR; B. Harding for performing EBV-lymphocyte transformations; M. Kuhar (Division of Neuroscience, Emory University, Atlanta, Ga., USA) and A. Baird (The Scripps Research Institute, San Diego, Calif., USA) for provision of CART(106–129) and preprosomatocrinin (GHRH) antibodies; financial support from Merck Research Laboratories, Rahway, N.J., USA, Pharmacia and Upjohn, the UK Medical Research Council, the Royal Society of London, the Royal College of Physicians (London), PAD 9607 and Marconi Medical Systems.

References 1 2 3 4 5 6

Holm VA, Cassidy SB, Butler MG, Hanchett JM, Greenswag LR, Whitman BY, Greenberg F: Prader-Willi syndrome: Consensus diagnostic criteria. Pediatrics 1993;91:398–402. Swaab DF: Prader-Willi syndrome and the hypothalamus. Acta Paediatr Suppl 1997;423:50–54. Burman P, Ritzén EM, Lindgren AC: Endocrine dysfunction in Prader-Willi syndrome: A review with special reference to GH. Endocr Rev 2001;22:787–799. Nicholls RD, Knepper JL: Genome organization, function, and imprinting in Prader-Willi and Angelman syndromes. Annu Rev Genomics Hum Genet 1902;2:153–175. Kalra SP, Dube MG, Pu S, Xu B, Horvath TL, Kalra PS: Interacting appetite-regulating pathways in the hypothalamic regulation of body weight. Endocr Rev 1999;20:68–100. Goldstone AP, Brynes AE, Thomas EL, Bell JD, Frost G, Holland A, Ghatei MA, Bloom SR: Resting metabolic rate, plasma leptin concentrations, leptin receptor expression, and adipose

Hypothalamic Neuropeptides and Fat Mass in PWS

41

7

8 9

10

11 12 13 14 15 16 17

18

19 20 21

22

23 24 25

26

27

tissue measured by whole-body magnetic resonance imaging in women with Prader-Willi syndrome. Am J Clin Nutr 2002;75:468–475. Goldstone AP, Thomas EL, Brynes AE, Bell JD, Frost G, Saeed N, Hajnal JV, Howard JK, Holland A, Bloom SR: Visceral adipose tissue and metabolic complications of obesity are reduced in Prader-Willi syndrome female adults: Evidence for novel influences on body fat distribution. J Clin Endocrinol Metab 2001;86:4330–4338. Brambilla P, Bosio L, Manzoni P, Pietrobelli A, Beccaria L, Chiumello G: Peculiar body composition in patients with Prader-Labhart-Willi syndrome. Am J Clin Nutr 1997;65:1369–1374. van Mil EG, Westerterp KR, Gerver WJ, van Marken L, Kester AD, Saris WH: Body composition in Prader-Willi syndrome compared with nonsyndromal obesity: Relationship to physical activity and growth hormone function. J Pediatr 2001;139:708–714. van Mil EA, Westerterp KR, Gerver WJ, Curfs LM , Schrander-Stumpel CT, Kester AD, Saris WH: Energy expenditure at rest and during sleep in children with Prader-Willi syndrome is explained by body composition. Am J Clin Nutr 2000;71:752–756. Jackson AA, Langley-Evans SC, McCarthy HD: Nutritional influences in early life upon obesity and body proportions. Ciba Found Symp 1996;201:118–129. Bouchard C, Despres JP, Mauriege P: Genetic and nongenetic determinants of regional fat distribution. Endocr Rev 1993;14:72–93. de Boer H, Blok GJ, van der Veen EA: Clinical aspects of growth hormone deficiency in adults. Endocr Rev 1995;16:63–86. Flint DJ, Gardner MJ: Influence of growth hormone deficiency on growth and body composition in rats: Site-specific effects upon adipose tissue development. J Endocrinol 1993;137:203–211. Friedman JM, Halaas JL: Leptin and the regulation of body weight in mammals. Nature 1998; 395:763–770. Barsh GS, Farooqi IS, O’Rahilly S: Genetics of body-weight regulation. Nature 2000;404: 644–651. Weigle DS, Ganter SL, Kuijper JL, Leonetti DL, Boyko EJ, Fujimoto WY: Effect of regional fat distribution and Prader Willi syndrome on plasma leptin levels. J Clin Endocrinol Metab 1997;82: 566–570. Pietrobelli A, Allison DB, Faith MS, Beccaria L, Bosio L, Chiumello G, Campfield LA, Heymsfield SB: Prader-Willi syndrome: Relationship of adiposity to plasma leptin levels. Obes Res 1998;6:196–201. Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K: Ghrelin is a growth-hormonereleasing acylated peptide from stomach. Nature 1999;402:656–660. Cummings DE, Purnell JQ, Frayo RS, Schmidova K, Wisse BE, Weigle DS: A preprandial rise in plasma ghrelin levels suggests a role in meal initiation in humans. Diabetes 2001;50:1714–1719. Kamegai J, Tamura H, Shimizu T, Ishii S, Sugihara H, Wakabayashi I: Chronic central infusion of ghrelin increases hypothalamic neuropeptide Y and Agouti-related protein mRNA levels and body weight in rats. Diabetes 2001;50:2438–2443. Wren AM, Seal LJ, Cohen MA, Brynes AE, Frost GS, Murphy KG, Dhillo WS, Ghatei MA, Bloom SR: Ghrelin enhances appetite and increases food intake in humans. J Clin Endocrinol Metab 2001;86:5992–5995. Cummings DE, Clement K, Purnell JQ, Vaisse C, Foster KE, Frayo RS, Schwartz MW, Basdevant A, Weigle DS: Elevated plasma ghrelin levels in Prader Willi syndrome. Nat Med 2002;8:643–644. Tschop M, Weyer C, Tataranni PA, Devanarayan V, Ravussin E, Heiman ML: Circulating ghrelin levels are decreased in human obesity. Diabetes 2001;50:707–709. Goldstone AP, Unmehopa UA, Bloom SR, Swaab DF: Hypothalamic NPY and agouti-related protein are increased in human illness but not in Prader-Willi syndrome and other obese subjects. J Clin Endocrinol Metab 2002;87:927–937. Swaab DF, Purba JS, Hofman MA: Alterations in the hypothalamic paraventricular nucleus and its oxytocin neurons (putative satiety cells) in Prader-Willi syndrome: A study of five cases. J Clin Endocrinol Metab 1995;80:573–579. Verbalis JG, Blackburn RE, Hoffman GE, Stricker EM: Establishing behavioral and physiological functions of central oxytocin: Insights from studies of oxytocin and ingestive behaviors. Adv Exp Med Biol 1995;395:209–225.

Goldstone/Unmehopa/Thomas/Brynes/Bell/Frost/Ghatei/Holland/Bloom/Swaab

42

28

29

30

Muscatelli F, Abrous DN, Massacrier A, Boccaccio I, Le Moal M, Cau P, Cremer H: Disruption of the mouse Necdin gene results in hypothalamic and behavioral alterations reminiscent of the human Prader-Willi syndrome. Hum Mol Genet 2000;9:3101–3110. Goldstone AP, Unmehopa UA, Swaab DF: Hypothalamic growth hormone-releasing hormone (GHRH) cell number is increased in human illness, but is not reduced in Prader-Willi syndrome or obesity. Clin Endocrinol 2003, in press. Van den Berghe G: Novel insights into the neuroendocrinology of critical illness. Eur J Endocrinol 2000;143:1–13.

Dr. A.P. Goldstone, Department of Endocrinology, St. Bartholomew’s Hospital, West Smithfield, London EC1A 7BE (UK) Tel. 44 20 7601 8343, Fax 44 20 7601 8505, E-Mail [email protected]

Hypothalamic Neuropeptides and Fat Mass in PWS

43

Central Nervous System and Body Weight Homeostasis Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 44–48

Discussion

Hypothalamic Energy Regulation: The Orexigenic Pathway (R.L. Leibel) M.B. Ranke, Tübingen: GH-deficient children have a higher amount of body fat in relation to other compartments. With GH treatment leptin levels decrease, basal metabolic rate goes up, muscle mass increases and fat mass decreases at a faster rate than the increase in muscle mass. In addition, appetite increases. Total body weight is typically unchanged. There is a slight shift towards T3 (tri-iodothyronine) and the balance between insulin and glucose is fairly steady. If we consider these observations, it seems that GH lowers the set point for leptin, as in anorexia and some other conditions. Do you know what types of central mechanisms may be involved in this situation? R.L. Leibel, New York, N.Y.: I would suggest that GH treatment lowers body fat by direct effects on adipose tissue and that the resultant decrease in leptin may account for some of the increase in food intake that you report. The rise in metabolic rate is due to the increase in muscle mass. As the child gains body mass, including fat mass, the leptin levels return to where they were before treatment. K.R. Westerterp, Maastricht: I am very appreciative of your leptin model. Obese subjects have very high leptin levels and when they are deprived of food for a time period, leptin doesn’t decrease very much, I wonder whether they might reach a critical level where leptin supplementation might be beneficial? Leibel: In the model of leptin action that I showed, the effects of leptin at high levels are not as striking as those when leptin is relatively deficient. At low levels of leptin, one gets the very striking changes in physiology and food intake that I described. Moreover, I think that low doses of leptin might be very helpful in enabling individuals to sustain weight loss over long periods of time. So, I agree with the premise of your question. Westerterp: So, you think that this threshold is not achieved in obese subjects since when you deprive them of food for a while leptin doesn’t decrease very much. Therefore, they cannot be helped with additional leptin. Leibel: This point gets to an other question, which is: How do you know where an individual is, in terms of the threshold, with regard to their circulating leptin concentration? It could be far above the threshold or very close to it. The experiments that I described are done in individuals who have maintained

the same high level of body weight for many years. So, we are not looking at people who have had acute excursions of body weight. If this model is correct, one might predict that those individuals, coming down from a very high leptin level above their threshold to a point perhaps a little bit above the threshold, won’t display striking physiologic responses to weight loss. We haven’t done those experiments. So, you may have two groups of weight-reduced individuals. An individual who shows hypometabolism as a result of weight reduction, would probably benefit from leptin replacement. F. Rohner-Jeanrenaud, Geneva: Your threshold hypothesis is very interesting and I think you are justified in correcting the way that we usually regard the role of leptin. My question is: What is the impact of your hypothesis on the amount of central leptin receptors? In other words, is the number of leptin receptors involved in the process of leptin resistance or increased leptin sensitivity? Leibel: This is a very interesting question. What is the effect of low leptin on leptin receptor activity in the CNS? The leptin receptor is hard to quantify using PCR techniques. However, there is evidence for some increase in the expression of leptin receptor in the hypothalamus of leptin-deficient animals. It’s not three- or four-/fivefold higher, it’s only slightly higher, and the response cannot compensate for the ligand deficiency. Obesity due to Mutations in the Anorexigenic Melanocortin Pathway: A Paradigm for Obesity in Prader-Willi Syndrome? (H. Krude) Rohner-Jeanrenaud: Your presentation was primarily about MC4 receptors, but it is known that there are other receptors, in white adipose tissue, for example. In rodents, MC2 and MC5 receptors seem to mediate an effect of melanocortins on lipolysis. Is this the case in humans? Are there peripheral effects of melanocortins in addition to their central effects? Could you also comment on the role of MC3 receptors? H. Krude, Berlin: There are five different MC receptors. I didn’t mention the MC5 receptor in my presentation. The MC5 receptor is expressed on exocrine glands and targeted deletion results in a phenotype affecting tears and, for example, hair lipids. Thus far, we have not found MC5 receptor abnormalities in our patients, but we have not looked for subtle differences. Concerning the question about adipose tissue, there are reports regarding MC1 and MC2 receptors on adipocytes. There have also been quite a number of patients with identified mutations in the MC2 receptor, leading to ACTH resistance and congenital hypocortisolism, but these patients are not obese. So, the MC2 receptor is probably not an important component for adipose tissue regulation. This is also true for the MC1 receptor. There have been a number of patients diagnosed with MC1 receptor mutations, primarily in Ireland but also in other countries and they are, to my knowledge, not obese. Concerning the

Discussion

45

MC3 receptor, the last one which I did not mention and which is also expressed in the hypothalamus, no convincing data have been published demonstrating a monogenetic obesity-related defect in patients with an MC3 receptor mutation. A.P. Goldstone, London: Do you have any indications to show the relative contributions of changes in appetite versus changes in energy expenditure in the obese phenotype of those with POMC mutations? Krude: POMC-deficient patients are extremely hyperphagic. This is very easy to see in a clinical way. Parents have told us that they realized from the first day of life that their children are excessively hungry every day. We have preliminary data that POMC-deficient patients do not have an increase in metabolic rate relative to the increase in body weight. But we don’t have a good control group since this extreme form of obesity is otherwise uncommon in this young age group. B.P. Hauffa, Essen: You also mentioned the role of ␤-endorphins in pain sensitivity in these patients. Have you noticed any changes in pain threshold in your POMC patients? Krude: No, we did not observe any differences in pain sensation, although we did not measure it quantitatively. There is a mouse model with selective ␤-endorphin deficiency that has normal pain sensitivity, as measured using a hot plate assay. These mice also have lower stress-induced pain responses compared to wild-type mice. This relatively mild phenotype is in accordance with the redundancy of the endorphin and enkephalin pathways which might compensate for one another. Signals that Control Central Appetite Regulation (S.C. Woods) D. l’Allemand, Zurich: I would like to make a comment on insulin and to raise a question. Children with PWS have rather low insulin levels in spite of their obesity. How would you reconcile that information with the model of appetite regulation? We had postulated that a central nervous system insulin deficiency might be responsible for the initiation of hyperphagia. S.C. Woods, Cincinnati, Ohio: That is a good question. When we segregate adiposity into visceral and subcutaneous, insulin correlates much better with visceral adiposity and leptin correlates much better with subcutaneous adiposity. Therefore, the high levels of insulin are associated with the visceral adiposity, leading to metabolic complications. I suspect that this might be the reason why children with PWS have disproportionately low insulin levels for their body weight; their body fat is primarily subcutaneous rather than visceral. I can’t explain how insulin is getting into the brain in PWS. There are theories that peptides such as insulin don’t pass through the blood-brain barrier easily, or that the receptors in the brain are altered, as Dr. Leibel pointed out. There are many, many places where there can be subtle mutations. I don’t think

Central Nervous System and Body Weight Homeostasis

46

anything is known about what happens in the CNS in relation to the interactions of insulin and leptin. Anon: Normally, you eat and then you secrete insulin. But in a diabetic patient, you first inject insulin and then you eat. Therefore, the insulin surge comes first. What is the effect of this on appetite regulation in diabetics? Woods: You raise the question that I’m always asking. First of all, why is it that an individual can be obese with hyperinsulinaemia or hyperleptinaemia? Rudy Leibel would agree that this is a Nobel Prize question. I don’t know the answer to that. I only discussed what happens in the brain. Normally, insulin only goes up during meals in most individuals and it goes up to keep the glucose from going up too high. Now, if you take a diabetic patient who is insulin deficient, these individuals are well known to be hyperphagic, they have extreme appetite, they cannot become obese, because insulin is required for fat storage in the adipose tissue. Nonetheless they have the same hyperphagia as individuals, who don’t have leptin. Now the other part of your question is if you take an individual and inject insulin peripherally, what happens if the blood glucose goes down and, in particular, if it goes down rapidly? If you remember Dr. Leibel’s presentation, the threshold for glucose, the place where it takes off is at the low end. When you lower glucose to that level, what goes up with appetite? What is important to know though is that all of us, if we are not insulin deficient or diabetic, eat meals when the blood glucose is perfectly normal, not when we are hyperglycemic. We can elicit appetite by giving insulin and lowering blood glucose, but that is artificial. Comments about Ghrelin (M. Tschöp) P.D.K. Lee, Los Angeles, Calif.: Beside insulin, are there any other peripheral signals that have been identified for ghrelin? M. Tschöp, Potsdam-Rehbrücke: To control ghrelin secretion? The only one that has been well demonstrated is caloric intake in general. Initial data from an in vitro rat stomach model indicates that perhaps glucose itself also has an effect on decreasing ghrelin secretion. The other signal, I am aware of, that is relatively certain, is somatostatin, which suppresses ghrelin secretion, but I wouldn’t necessarily consider that as a peripheral signal. Hypothalamic Neuropeptides and Regulation of Fat Mass in Prader-Willi Syndrome (T. Goldstone) Krude: I have a question or a comment. You mentioned in your conclusion that there is a ghrelin resistance in PWS. But if we put all the information together from this session, we learned that in monogenetic forms of obesity affecting the POMC, MC4, leptin, leptin receptor genes, patients are extremely

Discussion

47

obese. However, they have normal downregulation of ghrelin. This tells us that the defect in PWS is a different one. Since they cannot downregulate ghrelin, I wouldn’t say they are resistant, I would rather say that they have deficiency in the appropriate downregulation of ghrelin. Goldstone: I agree. From the observations we have, which are limited, we have, at least in the rodent, evidence that NPY and AGRP are the major mediators of ghrelin action and so I was surprised that we did not find high levels of NPY and AGRP. Of course there are the complicating factors of premorbid illness. Indeed a couple of papers have shown that ghrelin levels may go up during illness, and so may mask changes in these neuropeptides in PWS versus non-PWS subjects. But in those who died suddenly we still found that NPY peptide levels were lower in PWS than control subjects, and indeed the difference was greatest in those with sudden death. I have not shown the data, but we have also looked at NPY and AGRP in two Prader-Willi children who died suddenly. They had very high levels of NPY and AGRP, compared to 4 age- and sex-matched children who died from sudden infant death syndrome (SIDS). There are however a couple of problems with the interpretation of these findings, in addition to the small numbers. Firstly, our control comparison with SIDS may not be appropriate, since we do not know what is the cause of death in these SIDS cases, while the PWS children, although dying suddenly, may have had an underlying illness such that they were sicker than the SIDS children. Secondly, the PWS children were not yet hyperphagic, and it is recognized that the hyperphagia may take a while to develop during childhood. It will be interesting to see whether ghrelin levels are raised in PWS children even before they have developed hyperphagia. It will also be necessary to examine the effects of ghrelin administration, lowering or blockade on appetite in PWS, before we can ascertain whether there are normal hypothalamic consequences of the elevated ghrelin levels in PWS that lead to hyperphagia. Lee: Did your current studies, did you confirm the change in size of the paraventricular nucleus and did you notice any other structural changes in the brain? Goldstone: The patients that we examined in the NPY, AGRP and GHRH studies actually include the patients that were previously studied in the paper examining total and oxytocin neuron number in the paraventricular nucleus (PVN). We have obtained a few more PWS hypothalami since that first study but we have not looked at the PVN in the new material. We obviously need to look at the PVN some more. There are also other pathways to look at. We have attempted to look at the expression of the MC4 receptor in the PVN, but we have been unable to obtain any immunocytochemical staining for this receptor, as is the case for all seven-transmembrane receptors that I have examined in formalin-fixed paraffin-embedded human hypothalamic material.

Central Nervous System and Body Weight Homeostasis

48

Energy Balance in Prader-Willi Syndrome Compared to Simple Obesity Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 49–60

Assessment of Body Composition in Children with Prader-Willi Syndrome or Simple Obesity Kenneth J. Ellis Body Composition Laboratory, USDA/ARS Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Houston, Tex., USA

Abstract While childhood obesity may not have immediate adverse health effects for many children, it is often predictive of obesity in adulthood with its clearly associated risks for various chronic diseases. However, there are some clinical conditions in childhood for which excess adiposity is of concern, and one of these is a genetic condition called Prader-Willi syndrome. The purpose of this presentation is to present both common or traditional body composition assays as well as those that are under development for use in children. The accuracy, precision, and potential bias of these assays for body fat mass or the degree of body fatness will be presented. These body composition assays include weight-for-height indices, deuterium dilution for total body water, dual-energy x-ray absorptiometry, bioelectrical impedance analysis, air displacement plethysmography, photon reflectance, computer tomography, and magnetic resonance imaging. The advantages and disadvantages among these techniques will be presented. Reference ranges for normal body composition for children and Z score models for assessing body fatness in children with diseases will be examined. Copyright © 2003 S. Karger AG, Basel

This work is a publication of the USDA/ARS Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, and Texas Children’s Hospital, Houston, Tex., USA. The contents of this publication do not necessarily reflect the views or policies of the USDA, nor does mention of trade names, commercial products, or organizations imply endorsement.

Introduction

Chemical analyses of human tissue and organs have contributed significantly to our basic understanding of physiology and metabolism, scientific foundations for modern medicine. It is technically rather simple to remove small amounts of tissue from a living subject, although these procedures are not without some discomfort and risk. However, the findings from a small tissue sample may not necessarily be representative of the total organ, much less that of the whole body. Over the last 50 years a number of indirect measurement techniques have been developed that give basic body composition for the whole body [1]. The clinical usefulness of these techniques is governed not only by their accuracy and precision, but also by their physiological interpretation. In keeping with the theme of this conference, this paper will examine the noninvasive techniques currently available for the assessment of body composition in humans, with a particular focus on obesity assessment in children. The simplest body composition model, called the two-compartment (2-C) model, assumes that body weight (Wt) can be divided into fat mass (FM) and fat-free mass (FFM). It is relatively easy to obtain accurate and precise measurements of body Wt. The direct measurement of FM, however, has never been easy and remains a significant challenge for most body composition techniques. The alternative approach has been to obtain some measure of FFM, and then calculate FM as Wt minus FFM. Three body composition assays, each with its origins in the 1960s, have been used as the criterion or reference techniques. These are (1) body density based on underwater weighing (UWW), (2) counting of gamma rays emitted from the body to derive total body potassium (TBK), and (3) isotope dilution for the measurement of total body water (TBW) [1]. These assays do not measure FM directly, but instead measure a general property of the whole body, which can be attributed mainly to the FFM compartment. These assays assumed that the density of FFM is relative stable at a given age [2] and that the ratio of TBK and TBW to FFM are also relatively constant [3, 4]. Although these 2-C models work reasonably well for healthy subjects, it has been shown that these assumptions may be violated for some diseases. For this conference, we want to examine some of the more recent techniques that have been developed for the assessment of human body composition, with a particular focus on their application in pediatric populations. Some of these techniques use large sophisticated stationary instruments, while others are sufficiently portable that they can be considered for use in epidemiological studies or in clinical or ‘bedside’ settings.

Ellis

50

Anthropometric Indices

The body mass index (BMI; Wt/height (Ht)2) is probably the most commonly used weight-for-height index [5, 6]. There are many national and international BMI reference standards. The US Centers for Disease Control and Prevention (CDC), for example, has a web site (www.cdc.gov/growthcharts) that includes BMI percentile information as a function of age for children. Unfortunately, the relationship between BMI and body fatness (%FM ⫽ 100 ⫻ FM/Wt) is too general to recommend the use BMI as a reliable screening index for obesity of the individual child [7]. The second most commonly used anthropometric technique to assess body fatness is based on skinfold (SF) measurements. The major limitations with this technique are poor precision, operator-dependent results, and extrapolation to an accurate estimate the body’s total FM. The large numbers of SF-based prediction equations that have been published for estimating total FM may best illustrate the latter case. In recent years, several alternatives to the use of SF calipers have been developed. These include the use of ultrasound, infrared interactance, and photon backscatter [8–10]. When used with children, mixed results for these technologies, regarding accuracy and precision, have been reported [11]. The one advantage of these techniques, however, is that the measurement technique can be automated which has lessened the operator-dependent bias observed for SF measurements. One of these instruments (LIPOMETER) does hold promise as an accurate measure of the subcutaneous fat layer at multiple locations over the body [9, 10]. The assessment of multiple locations over the body (fig. 1) takes only a few minutes to complete and has good precision. Further developments need to include more reference data for children, testing of the instrument’s response using non-Caucasian populations, and a comparison of its accuracy for FM with other techniques. This approach, however, should be very useful for monitoring changes in the subcutaneous fat layer, independent of any extrapolation to whole body FM or %FM. It may be sufficient, for example, to monitor only changes in the subcutaneous fat layer of children with Prader-Willi syndrome (PWS). Alternately, this technique may be coupled with an independent assay of the body’s abdominal fat content to derive an estimate of the visceral fat component.

Air Displacement Plethysmography

The UWW technique is used to measure body density [12]. The major drawbacks with this method are that the subject needs to be totally submerged

Body Composition Methods

51

3

5 LIPOMETER

7 8 11

Thickness of subcutaneous adipose tissue (mm) 10

14

1 4 6 10

2

9

12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

20

30

40

50

Neck Triceps Biceps Upper back Front chest Lateral chest Upper abdomen Lower abdomen Lower back Hip Front thigh Lateral thigh Rear thigh Inner thigh Calf

13 15

Fig. 1. Sites used with LIPOMETER assay of subcutaneous fat layer [see 9, 10].

underwater and exhale all of the air in his/her lungs. This can be difficult even for healthy subjects, not to mention those with various diseases, breathing problems, or fear of water. An alternative technique for the measurement of body density, called air displacement plethysmography (ADP), has been developed [13, 14] that overcomes most of these limitations. For this measurement, the subject sits in an air-filled chamber, which is coupled to a reference chamber of known volume. A small systematic differential volume perturbation is introduced between the two chambers, and the pressure difference is recorded. Small corrections are needed for the isothermal and adiabatic effects of the subject’s exhaled breath, for air trapped by body hair, heating of the air close to the body’s surface. The subject still needs to wear a swimsuit and swim cap to cover the head. The subject’s residual lung volume can be measured while in the chamber. It is clear that the major advantage of ADP is that the subject does not have to be submerged underwater, and the exam takes only a few minutes to

Ellis

52

complete. At present, there is only one commercially available ADP instrument, called the BodPod (Life Measurements, Davis, Calif., USA), and its software is still based on the 2-C model. The ADP technique is rapidly replacing the UWW approach as the preferred method for the measurement of body density [15]. As noted previously, a limitation with the basic 2-C model is that the same value for the density of the FFM is used for all subjects. This can be somewhat overcome by expanding to a 3-C model which requires the additional measurement of TBW. This model can be expanded to the 4-C model, which requires another additional measurement, i.e. that of the body’s mineral content. The best measure of TBW is obtained using a deuterium dilution assay, while dual-energy x-ray absorptiometry (DXA) provides an assay for bone mineral content (BMC), which is the major component of the body’s mineral mass. Although the 4-C model is one of the most accurate for body composition analysis, its practical use is limited by the requirement of three separate assays.

Bioelectrical Impedance Methods

For healthy older children and adults, the water content of the FFM is relatively constant [16]. Thus, if TBW can be measured, FFM can be calculated, and when subtracted from body Wt, an indirect estimate of FM is obtained. The classical reference technique for the TBW assay is based on the collection of a fluid sample (blood, urine, or salvia) from the subject at 2–4 h after receiving an oral tracer dose of a nonradioactive isotope of water [17]. Several assumptions regarding the size of the tracer dilution space, relative to the true TBW space, are needed [18]. The most frequently used isotope is deuterium, and its concentration in the body fluid can be assayed using mass spectroscopy or Fouriertransformed infrared analysis [19]. Using this assay, the TBW estimates are reported to be accurate to ⫾2% [1]. An alternative assay, based on tissues’ conductive properties, has been developed. When an electrical current is passed through tissue, there is resistance to the current that is related to the size and electrolyte content of the tissue. This is the basic principle for the body composition techniques called bioelectrical impedance analysis (BIA) and bioelectrical impedance spectroscopy (BIS). For BIA measurements, a weak nondetectable alternating current (50 kHz, 800 ␮A) is used. The BIS measurement uses an equally low current, but the frequency is varied from a few kilohertz to 1 MHz, which results in differences in the body’s resistance values. The BIA and BIS measurement techniques are considered secondary measurements because they must

Body Composition Methods

53

be calibrated using a reference technique such as deuterium dilution for TBW. The increasing popularity of this technology can be attributed, in part, to the low cost of the instruments, their simplicity of operation, and the high reproducibility obtained for the individual. These advantages, however, must be weighted against the accuracy of the body composition results [20, 21]. There have been numerous BIA calibration equations published for the estimated TBW and FFM in healthy pediatric and adult populations, but none have been adequately tested or verified for use in PWS children. After attaching a pair of electrodes on the body, usually at the wrist and ankle, an alternating current is passed through the body, and resistance (R) and reactance (X) are measured. In an electrical circuit model, the body is assumed to be a cylindrical conductor with its length proportional to the subject’s Ht. The impedance index (defined as Ht2/R) is calculated and assumed proportional to TBW. Other investigators have used BIA to calculate a parameter called the phase angle, which is based on the ratio between the resistance (R) and reactance (X) values [22, 23]. Using this model, a phase angle less than 5–6° would be considered abnormal, and is presumed to indicate a low lean tissue mass. The clinical applicability of the phase angle model needs further testing including in children with abnormal lean and fat tissue mass content, such as is seen in PWS [24]. The clear advantage of BIS compared with BIA is that estimates of both TBW and extracellular water are obtained [25]. Although only the TBW value is needed to derive an estimate for FM, the difference between TBW and extracellular water can be used to estimate intracellular water space, which is similar to the use of TBK as a body compositional marker for the body cell mass and an individual’s nutritional status [26].

Imaging Methods (DXA, Computer Tomography, Magnetic Resonance Imaging)

DXA, computer tomography (CT), and magnetic resonance imaging (MRI) have been used to study body composition. Each of these techniques requires a significance capital investment, trained technical staff, and annual operating expenses such as maintenance and service. Only the DXA instruments have standard software for body composition analysis. MRI and CT are often considered as the reference standards for volume-based measurements of body composition, matched only by the chemical-based models using neutron activation technique [3, 4]. The main advantages of CT and MRI are that 3-dimensional anatomical images can be obtained, whereas DXA provides only

Ellis

54

a

b

c

d Fig. 2. Abdominal CT fat analysis: typical abdominal CT cross-sectional image (a), dashed area represents subcutaneous adipose tissue (b), dark line shows the outline of the intra-abdominal area (c), and white area within the intra-abdominal area that is identified as the vascular adipose tissue (d).

a planar-projected 2-dimensional image. For example, CT and MRI can separate the abdominal FM into its subcutaneous and visceral compartments (fig. 2), while DXA cannot. On the other hand, a whole body DXA scan (fig. 3) provides regional FM information that is not easily obtained using CT or MRI unless multiple images are obtained. Several investigators have proposed using whole-body DXA in combination with a subcutaneous fat measurement (such as provided by the LIPOMETER instrument) to indirectly derive visceral FM, but these studies are still in progress (Ellis, unpubl. data). If only one assay can be performed, a DXA scan probably provides the best overall information about body composition. A whole-body DXA scan partitions the body into a 3-C model of body composition: FM, BMC, and nonbone, nonfat lean tissue mass. Regional body composition information can also be obtained from this scan. The scan procedure is relatively easy to perform, has a very low risk (dose ⬍10 ␮Sv, total body). DXA scans, for example, have been used to establish body composition references for preterm and healthy infants [27].

Body Composition Methods

55

Fig. 3. Whole-body DXA scan image showing the full skeleton and the overlying soft tissue.

Pediatric Body Composition References

There are a number of publications that provide reference ranges for normal body composition in healthy pediatric populations [28–31]. Most of these studies have used DXA as the primary body composition assay. The mean FM, FFM, and BMC for a representative Caucasian pediatric population are provided in table 1. For the evaluation of bone mineral density of the individual, there are two interactive Internet websites (www.bcm.tmc.edu/bodycomplab; www-stat-class.Stanford.EDU/pediatric-bones). These sites give age-, gender-, and ethnic-specific Z scores for whole-body, hip, and spine bone mineral density measurements in children. Similar Z score calculators are being developed for lean tissue mass and FM (Ellis, unpubl. data).

Ellis

56

Table 1. Typical body composition values for US pediatric population (Caucasian) Age FFM, kg group, years females

males

females

males

females

males

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

15.1 (2.1) 16.8 (1.7) 18.6 (1.9) 20.7 (2.2) 22.7 (3.0) 24.8 (3.8) 30.1 (4.0) 36.1 (8.0) 39.7 (5.9) 41.8 (12.9) 53.4 (7.6) 55.3 (7.3) 57.2 (6.9) 58.0 (6.3) 61.9 (7.6)

4.4 (0.5) 5.4 (2.3) 5.1 (1.0) 6.7 (2.4) 9.0 (5.9) 6.9 (1.7) 11.3 (4.9) 12.7 (8.9) 7.1 (1.5) 14.8 (6.7) 16.2 (7.3) 13.5 (7.9) 23.5 (9.9) 22.8 (9.9) 22.1 (7.0)

3.6 (0.8) 3.9 (1.0) 4.3 (2.1) 4.4 (2.3) 5.5 (1.9) 6.7 (1.4) 9.7 (8.1) 11.8 (9.8) 11.7 (7.2) 20.2 (12.2) 12.7 (7.0) 13.0 (6.0) 10.8 (5.1) 10.6 (3.8) 10.9 (3.6)

0.437 (0.036) 0.526 (0.093) 0.548 (0.075) 0.636 (0.092) 0.818 (0.185) 0.933 (0.090) 1.303 (0.350) 1.253 (0.497) 1.443 (0.388) 2.294 (0.350) 1.645 (0.280) 2.212 (0.225) 2.280 (0.193) 2.245 (0.285) 2.261 (0.211)

0.48 (0.08) 0.60 (0.15) 0.62 (0.14) 0.78 (0.13) 0.90 (0.22) 0.99 (0.22) 1.20 (0.28) 1.43 (0.26) 1.59 (0.30) 1.81 (0.48) 2.05 (0.51) 2.23 (0.46) 2.63 (0.35) 2.58 (0.31) 2.71 (0.45)

13.1 (1.2) 14.9 (2.3) 16.0 (2.2) 17.7 (1.7) 21.3 (3.9) 22.1 (7.1) 30.7 (7.0) 31.9 (8.1) 33.1 (4.4) 36.0 (4.9) 39.1 (4.0) 39.3 (3.7) 41.4 (2.8) 45.1 (4.8) 42.1 (7.0)

Fat, kg

Bone mineral, kg

Values represent mean with SD in parentheses.

Conclusions

If only changes in FM are of interest then several body composition techniques can be used. If changes in bone are to be examined, then DXA offers the better choice. The performance of different body composition techniques is provided in table 2, including the minimum detectable change for an individual. For population studies the minimum detectable changes are dependent on the number of subjects in the study (sample size), the assay’s precision, and the biological variability of the parameter being measured among subjects in the population [32]. In practical terms, the size of the population is often limited and the biological variability is not easily manipulated, hence the selection of the more precise measurement technique is critical. A second consideration, especially for longitudinal studies, is the length of the time interval between repeat measurements. If the time interval is too short, the changes needed to reach statistical significance may not be physiologically possible to achieve. Alternately, if the time period between measurements is relatively long, then the normal physiological changes in body composition may mask the treatment effect being studied.

Body Composition Methods

57

Table 2. Body composition methods Instrument costa

Precisionb %

Accuracyc %

Minimum detectable changed

TBW D2O dilution BIA/BIS

M L

1–2 2–4

2–4 3–7

2 liters (5) 4 liters (10)

FFM UWW ADP DXA BIA/BIS

M M H L

1–2 1–2 1.5 2–4

2–3 2–3 1–4 2–8

2 kg (4) 2 kg (4) 1.5 kg (2) 4 kg (7)

FM UWW ADP DXA BIA/BIS CT/MRI

M M H L VH

2–3 2–3 2–3 ⬎5 ⬃1–2

⬎5 ⬎5 ⬃5 ⬎8 ⬃2

2 kg (11) 2 kg (11) 2 kg (11) 4 kg (22) 10 cm2 (5)e

Body composition: compartment and method

Values in parentheses represent percentage. a Initial instrument cost and annual operating cost combined: L ⬍ USD5,000; M ⬍ USD50,000; H ⬍ USD100,000; VH ⬎ USD100,000. b Reproducibility for repeat measurements. c Accuracy for absolute mass or volume estimate. d Sample size ⫽ 1 (% change for a 70-kg subject with 25% fat). e Cross-sectional area measurements only, single slice.

Acknowledgment The editorial assistance of L. Loddeke is gratefully acknowledged. This work was supported by the US Department of Agriculture, Agricultural Research Service under Cooperative Agreement #58-6250-6-001 with Baylor College of Medicine.

References 1 2 3 4

Ellis KJ: Human body composition: In vivo methods. Physiol Rev 2000;80:649–680. Lohman TG: Advances in Body Composition Assessment. Champaign, Human Kinetics, 1992. Wang ZM, Pierson RN Jr, Heymsfield SB: The five level model: A new approach to organizing body composition research. Am J Clin Nutr 1992;56:19–28. Wang ZM, Heshka S, Pierson RN Jr, Heymsfield SB: Systematic organization of body-composition methodology: An overview with emphasis on component-based models. Am J Clin Nutr 1995; 61:457–465.

Ellis

58

5 6 7 8 9 10

11

12 13 14

15 16

17 18 19 20 21

22 23 24

25

26 27 28

Cole TJ, Bellizzi MC, Flegal KM, Dietz WH: Establishing a standard definition for child overweight and obesity worldwide: International survey. Br Med J 2000;320:1240–1243. Dietz WH, Robinson TN: Use of the body mass index (BMI) as a measure of overweight in children and adolescents. J Pediatr 1998;132:191–193. Ellis KJ, Abrams SA, Wong WW: Monitoring childhood obesity: Assessment of the weight/ height2 index. Am J Epidemiol 1999;150:939–946. Conway JM, Norris KH, Bodwell CE: A new approach for estimation of body composition: Infrared interactance. Am J Clin Nutr 1984;40:1123–1130. Moller R, Tafeit E, Smolle KH, Kullnig P: Lipometer: Determining the thickness of a subcutaneous fatty layer. Biosens Bioelectron 1994;9:13–16. Moller R, Tafeit E, Pieber TR, Sudi K, Reibnegger G: Measurement of subcutaneous adipose tissue topography (SAT-top) by means of a new optical device, Lipometer, and the evaluation of standard factor coefficients in healthy subjects. Am J Hum Biol 2000;12:231– 239. Thomas DW, Ryde SJ, Ali PA, Birks JL, Evans CJ, Saunders NH, Al-Zeibak S, Dutton J, Hancock DA: The performance of an infra-red interactance instrument for assessing total body fat. Physiol Meas 1997;18:305–315. Lohman TG: Applicability of body composition techniques and constants for children and youths. Exerc Sport Sci Rev 1986;14:325–357. Dempster P, Aitkens S: A new air displacement method for the determination of human body composition. Med Sci Sports Exerc 1995;27:1692–1697. McCrory MA, Gomex TD, Bernauer EM, Mole PA: Evaluation of a new air displacement plethysmograph for measuring human body composition. Med Sci Sports Exerc 1995;27: 1686–1691. Fields DA, Goran MI, McCrory MA: Body-composition assessment via air-displacement plethysmography in adults and children: A review. Am J Clin Nutr 2002;75:453–467. Wang ZM, Deurenberg P, Wang W, Pietrobelli A, Baumgartner RN, Heymsfield SB: Hydration of fat-free mass: Review and critique of a classic body composition constant. Am J Clin Nutr 1999;69:833–841. Sheng HP, Huggins RA: A review of body composition studies with emphasis on total body water and fat. Am J Clin Nutr 1979;32:630–647. Schoeller DA: Isotope dilution methods; in Bjorntorp P, Brodoff BN (eds): Obesity. New York, Lippincott, 1991, pp 80–88. Aslani A, Hansen RD: Plasma sample preparation by ultrafiltration for total body water determination. Ann NY Acad Sci 2000;904:287–289. Ellis KJ, Shypailo RJ, Wong WW: Body water measurements in a multiethnic pediatric population using multifrequency bioelectrical impedance spectroscopy. Am J Clin Nutr 1999;70:847–853. Ellis KJ, Bell SJ, Chertow GM, Chumlea WC, Knox TA, Kotler DP, Lukaski HC, Schoeller DA: Bioelectrical impedance methods in clinical research: A follow-up to the NIH Technology Assessment Conference. Nutrition 1999;15:874–880. Piccoli A, Rossi B, Pillon L, Bucciante G: A new method for monitoring body fluid variation by bioimpedance analysis: The R-Xc graph. Kidney Int 1994;46:534–539. Piccoli A, Rossi B, Pillon L, Bucciante G: Body fluid overload and bioelectrical impedance analysis in renal patients. Miner Electrolyte Metab 1996;22:76–78. Eiholzer U, I’Allemand D, van der Sluis I, Steinert H, Gasser T, Ellis K: Body composition abnormalities in children with Prader-Willi syndrome and long-term effects of growth hormone therapy. Horm Res 2000;53:200–206. Matthie J, Zarowitz B, DeLorenzo A, Andreoli A, Katzarski K, Pan G, Withers P: Analytic assessment of the various bioimpedance methods used to estimate body water. J Appl Physiol 1998; 84:1801–1816. Moore FD, Olsen KH, McMurrey JD, Parker HV, Ball MR, Boyden CM: The Body Cell Mass and Its Supporting Environment. London, Saunders, 1963. Koo WW, Walters JC, Hockman EM: Body composition in human infants at birth and postnally. J Nutr 2000;130:2188–2194. Ellis KJ, Abrams SA, Wong WW: Body composition of a young, multiethnic female population. Am J Clin Nutr 1997;65:724–731.

Body Composition Methods

59

29 30 31

32

Ellis KJ: Body composition of a young, multiethnic, male population. Am J Clin Nutr 1997;66: 1323–1331. Ellis KJ, Shypailo RJ, Abrams SA, Wong WW: The reference child and adolescent models of body composition. A contemporary comparison. Ann NY Acad Sci 2000;904:374–382. Ellis KJ, Shypailo RJ, Hardin DS, Perez MD, Motil KJ, Wong WW, Abrams SA: Z score prediction model for assessment of bone mineral content in pediatric diseases. J Bone Miner Res 2001; 16:1658–1664. Hassager C, Christiansen C: Measurement of bone mineral density. Calcif Tissue Int 1995;57:1–5.

Kenneth J. Ellis, PhD, Professor, CNRC – Rm# 4020, 1100 Bates St., Houston, TX 77030-2600 (USA) Tel. ⫹1 713 798 7131, Fax ⫹1 713 798 7130, E-Mail [email protected]

Ellis

60

Energy Balance in Prader-Willi Syndrome Compared to Simple Obesity Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 61–69

Physical Activity and Obesity Klaas R. Westerterp Department of Human Biology, Maastricht University, Maastricht, The Netherlands

Abstract Obesity is the long-term result of a positive energy balance. It is a consequence of an increased energy intake or decreased energy expenditure. The latter can be caused by a reduction of resting energy expenditure or by a reduction of physical activity. With the application of the doubly labelled water method for the measurement of total energy expenditure in free-living conditions, in combination with the measurement of resting energy expenditure, it is possible to study physical activity in relation to obesity. On average, activity-related energy expenditure in obese and lean subjects is similar. However, obese subjects need more energy to do body movements than the nonobese subjects, especially as regards weight-bearing activities. Thus, the obese are generally less physically active than the nonobese although there is no difference in activity-related energy expenditure. An exercise intervention, in combination with energy restriction, only has a minor or no effect on the energy expenditure of a subject and consequently does not result in additional weight loss. However, subjects who maintain a normal body weight are characterized by sustained higher levels of freeliving physical activity than subjects who are unsuccessful at weight control. Activity energy expenditure increases from early childhood to adolescence. There is no evidence of a difference in activity energy expenditure between obese and nonobese children and adolescents. Prader-Willi patients, as a model for a genetic predisposition to obesity, are less physically active than matched obese control subjects. Copyright © 2003 S. Karger AG, Basel

Introduction

Since the application of doubly labelled water for the measurement of total energy expenditure in human subjects there have been three meta-analyses on the relation between physical activity and obesity. Schoeller and Fjeld [1]

Table 1. Subject characteristics

Age, years Height, m Body mass, kg BMI, kg/m2 Fat-free mass, kg BMR, MJ/day ADMR, MJ/day ADMR/BMR

Women (n ⫽ 215)

Men (n ⫽ 278)

48 ⫾ 22 (19–96) 1.65 ⫾ 0.07 (1.49–1.86) 70 ⫾ 19 (38–164) 25.8 ⫾ 6.5 (12.5–55.3) 45 ⫾ 7 (29–79) 5.9 ⫾ 1.0 (3.6–10.8) 9.8 ⫾ 2.3 (4.1–18.4) 1.65 ⫾ 0.28 (1.13–2.57)

49 ⫾ 19 (19–96) 1.77 ⫾ 0.07 (1.60–1.97) 83 ⫾ 19 (50–216) 26.3 ⫾ 5.4 (15.7–61.7) 60 ⫾ 9 (39–93) 7.4 ⫾ 1.2 (4.7–12.7) 12.9 ⫾ 2.9 (5.0–21.5) 1.74 ⫾ 0.29 (1.13–2.60)

Values represent mean ⫾ SD with the range in parentheses.

already concluded that obese individuals expend slightly more energy in physical activity than do nonobese individuals. Prentice et al. [2] suggested that, except in massive obesity, patterns of physical activity are quite similar at different levels of body mass index (BMI). In a later analysis it was confirmed that the majority of obese subjects is moderately active [3]. Here, an update is presented on doubly labelled water-assessed physical activity and obesity. Additionally, results are presented on simultaneous doubly labelled waterassessed physical activity and accelerometer-assessed body movement in relation to body weight. Subsequently, studies are reviewed on the effect of diet and exercise interventions on body weight, where activity is measured with doubly labelled water as well. Finally, evidence is presented for physical activity and obesity from infancy and early childhood to adolescence.

Doubly Labelled Water-Assessed Physical Activity and Obesity

In Maastricht, we evaluated 493 subjects, excluding those aged ⬍19, those with an intervention in energy intake, an intervention in physical activity including athletic performance, pregnancy, lactation and disease. All observations were performed over an interval of 2 weeks according to the Maastricht protocol [4]. Studies included a wide body mass range with a mean (SD) BMI of 25.8 kg/m2 (SD 6.5) for women and 26.3 kg/m2 (SD 5.4) for men (table 1). The physical activity level (PAL) of the subjects calculated by expressing the average daily metabolic rate as a multiple of basal metabolic rate (PAL ⫽ ADMR/BMR) was similar at different levels of BMI as reported by Prentice et al. [2]. The PAL

Westerterp

62

a

b

MJ/day 3.0

10

2.5

5

2.0

0

1.5

⫺5

⫺10

1.0 10

20

30

40

50

60

70 kg/m2

10

20

30

40

50

60

Fig. 1. PAL as a function of BMI in 493 subjects (table 1). a PAL calculated as ADMR divided by BMR. b PAL calculated as the residual of the ADMR-BMR relation.

calculated as the residual of the ADMR-BMR relation was also independent of BMI (fig. 1). Activity-induced energy expenditure (AEE) was calculated as [(0.9 ⫻ ADMR) – BMR)], assuming diet-induced energy expenditure is about 10% of ADMR in subjects consuming an average mixed diet that meets energy requirements. AEE was positively related to BMI, i.e. obese subjects had a higher AEE than the nonobese ones. Correcting AEE for differences in body size by expressing AEE per kilogram body mass resulted in a negative relation with BMI (fig. 2). The PAL of a subject as measured with doubly labelled water reflects the energy expenditure for physical activity but not the amount of physical activities, i.e. duration and intensity [2]. An alternative method many studies have used to adjust for differences in body size is by expressing nonbasal energy expenditure per kilogram body mass, assuming that energy expenditure associated with physical activity is weight-dependent. Prentice et al. [5] suggested that normalizing AEE by dividing by body mass to the exponent 1.0 (body mass 1.0) overcorrects for body size in heavier people, making them appear less active, as not all daily activities are weight-dependent. They suggested that an exponent close to 0.5 is more appropriate for sedentary lifestyles. The value of the exponent of 0.5 gave the best ‘normalization’ of the energy expenditure of various activities for differences in body weight as measured in a respiration

Physical Activity and Obesity

63

70 kg/m2

a

MJ/day

b

MJ/kg/day

10.0

0.15

7.5 0.10

5.0

0.05 2.5

0.0

0.00 10

20

30

40

50

60

70 kg/m2

10

20

30

40

50

60

70 kg/m2

Fig. 2. Activity-associated energy expenditure as a function of BMI in 493 subjects (table 1). a The absolute values in MJ/day. b The values in MJ/kg/day to correct for differences in body size.

chamber. Schoeller and Jefford [6] measured energy expenditure while subjects performed controlled light activities representative of activities of daily life. Allometric regression analysis indicated that the energy cost of these light activities was proportional to body weight and it was concluded that normalization of energy expenditure of physical activity by the division by body weight is an appropriate means for comparing the volume (intensity ⫻ time) of physical activity between individuals of different body sizes. Thus, while figure 1 shows no difference in physical activity with increasing BMI, figure 2 shows an increase in AEE with increasing BMI and, after correcting AEE for differences in body size by expressing AEE per kilogram body mass, a negative relation with BMI as shown before [7]. Ekelund et al. [8] compared physical activity as well as activities in the obese (BMI ⬎30) and matched nonobese adolescents where physical activity was measured with doubly labelled water simultaneously with physical activity measurements with an accelerometer. The obese were less physically active than the nonobese although there was no difference in PAL. In conclusion, duration and intensity of physical activities do not need to be identical with energy spent on activity. While activity energy expenditure increases with increasing body size, the obese are generally less physically active than the nonobese.

Westerterp

64

Effect of Doubly Labelled Water-Assessed Diet and Exercise Interventions on Body Weight

The addition of exercise to an energy-restricted diet results in little further weight loss. Exercise does not reverse the weight loss-induced depression of the resting metabolic rate and weight loss is not different for groups undergoing dietary restriction and dietary restriction plus exercise. The latter implicates that the direct cost of the exercise training is compensated by a reduction of activity-associated energy expenditure outside the training sessions. Two studies compared activity-associated energy expenditure before and after dietary restriction and dietary restriction plus exercise, as measured with doubly labelled water. Racette et al. [9] designed diets to promote a weight loss of 1 kg/week by prescribing a diet in the diet-only group and added the calculated energy costs of the exercise for the diet plus exercise group to create a comparable energy deficit. They observed a maintenance of ADMR in the exercise group while ADMR decreased in the diet-only group. Kempen et al. [10] provided all subjects with an identical low-energy formula diet. They observed a comparable decrease of ADMR in the diet plus exercise and the diet-only group. ADMR dropped significantly, and to a similar extent, with both treatments suggesting no net effect of the exercise training on the activity-associated energy expenditure. Explanations for the relatively minor or nonexistent effect of the addition of exercise to an energy-restricted diet are a low compliance to the exercise prescription and/or a negative effect of exercise training on dietary compliance. The overall conclusion was that the size of the exercise intervention only had a minor or no effect on the activity level of the subjects and consequently did not result in additional weight loss. Finally there are studies showing the effect of energy restriction on physical activity. Velthuis-te Wierik et al. [11] observed the effect of a moderately energy-restricted diet on the energy metabolism in nonobese men (BMI 24.9 ⫾ 1.9 kg/m2). For 10 weeks the subjects received a diet with 67% of their measured ADMR during weight maintenance. The consequent weight loss was 7.4 ⫾ 1.7 kg and the activity level (ADMR/BMR) went down from 1.85 ⫾ 0.37 to 1.65 ⫾ 0.29 (p ⫽ 0.06), i.e. there was a tendency for a reduction of physical activity by reducing energy intake. There is evidence that physical activity is of importance for weight maintenance, especially for the prevention of weight regains after weight loss. Schoeller et al. [12] assessed energy expenditure during physical activity in weight-reduced women and found that lower activity levels were associated with greater weight gains at follow-up. Weinsier et al. [13] compared total free-living activity energy expenditure and PAL in women successful and unsuccessful at

Physical Activity and Obesity

65

maintaining a normal body weight. Two groups were identified on the basis of extreme weight changes: maintainers had a weight gain ⱕ2 kg/year and gainers had a weight gain ⱖ6 kg/year. Gainers had a lower AEE, a lower PAL, and less muscle strength. A lower AEE in the gainers explained ⬃77% of their greater weight gain after 1 year. In conclusion, an exercise intervention, in combination with energy restriction, only has a minor or no effect on the energy expenditure of a subject and consequently does not result in additional weight loss. However, subjects who maintain a normal body weight are characterized by sustained higher levels of free-living physical activity than subjects who are unsuccessful at weight control.

Physical Activity and Obesity from Infancy and Early Childhood to Adolescence

The PAL increases from infancy and early childhood to adolescence. Hoos et al. [14] reviewed data of studies in children and adolescents where PAL was calculated from ADMR as measured with doubly labelled water. The age range covered was from 3 to 16 years. PAL increased from a value of 1.5 to 1.8 over the corresponding age range (PAL ⫽ 0.025 age ⫹ 1.40, r ⫽ 0.85, p ⬍ 0.001). There was no difference for girls and boys. The low PAL value of 1.5 in young children could be due to the fact than they sleep longer and thus spend less time on physical activities. As observed in adults, the prevalence of obesity in children is increasing. The etiology of the development of childhood obesity is poorly understood, but is a consequence of alterations in the regulation of the energy balance between energy expenditure and energy intake. It is not known whether obesity is caused by an increase in energy intake relative to energy needs, a decrease in energy expenditure relative to energy intake, or the effect of both [15]. In a longitudinal study done over 4 years in preadolescent children, changes in fatness were related to initial fatness and parental fatness, but not to reduced energy expenditure [16]. In an editorial the interpretation of the results were criticized because the original data were not presented with the estimated validity [17]. Indeed it is very difficult to trace the cause of a change in fat mass from changes in energy intake and energy expenditure due to the fact that small discrepancies between energy intake and energy expenditure have large consequences when they are persistent for a longer time. As an example, for the longitudinal study over 4 years in preadolescent children as mentioned above [16], the energy imbalance is related to energy turnover. The average energy expenditure at the second evaluation was 6.32 ⫾ 0.95 MJ/day and the rate of change in absolute fat mass of the children was 0.89 ⫾ 1.08 kg/year. The change in fat mass is equivalent to

Westerterp

66

0.095 MJ/day, assuming an energy equivalent of 39 MJ/kg fat or 1.5% of measured energy expenditure. Discrepancies between reported energy intake and measured energy expenditure of about 20–50% have been described and are mainly due to underreporting of food intake [18]. The analytical precision of the gold standard for the measurement of energy expenditure under free-living conditions, the doubly labelled water method, is estimated to be 3–6% [19]. Thus, in our sedentary society with abundant palatable energy-dense foods, there is an increase in the prevalence of obesity but we do not know whether the cause is an increase in energy intake or a decrease in energy expenditure. Further insight into the development of obesity might be gained from studies in subjects with a genetic predisposition to obesity. The Prader-Willi syndrome (PWS) is the most recognized form of genetic obesity in humans [20]. The disorder is characterized by perinatal and neonatal hypotonia, followed by an obese phase in childhood. The obesity is likely to be caused by a combination of low-energy expenditure and a high-energy intake [21, 22]. Van Mil et al. [23] measured PALs in PWS compared with matched obese controls. The subjects were 10 females and 7 males, age 7.5–19.8 years, with PWS and 17 obese controls matched for gender and bone age. The residuals of the ADMR-BMR relation were systematically lower for the PWS subjects compared to the controls (p ⬍ 0.001), pointing out a decrease in activity-related energy expenditure in PWS patients. Other measures for physical activity also demonstrated that PWS patients are less active during childhood and adolescence. In conclusion, activity energy expenditure increases from early childhood to adolescence. There is no evidence for a difference in activity energy expenditure between obese and nonobese children and adolescents. Prader-Willi patients, as a model for genetic predisposition to obesity, are less physically active than matched obese control subjects.

Discussion

What are the beneficial effects of physical activity for obese subjects, being on average moderately active and having a limited ability to perform exercise? The answer probably is that a high level of energy turnover, i.e. a high level of physical activity, promotes weight maintenance. Regulation of energy balance is likely to be better at a higher energy turnover. Obviously one can eat more when one is physically active, without getting into a positive energy balance. There are speculations that exercise reduces the perceived hunger in obese subjects and there is evidence that exercise increases fat oxidation in a high-fat diet. Obese subjects can only reach a significant weight loss with an energyrestricted diet. Mild energy restriction will already result in a very significant

Physical Activity and Obesity

67

weight loss when the subject complies with the diet. An increase in physical activity is necessary to compensate for the reduction in AEE and should be facilitated by the lower body mass. Indirect evidence suggests that modern inactive lifestyles are important in the etiology of obesity and possibly represent a dominant factor. People adopt increasingly sedentary lifestyles in which motorized transport, mechanized equipment, and energy-saving domestic appliances replace physically arduous tasks. In England, the average person now watches over 26 h of television a week. Proxy measures of physical inactivity, such as car ownership and television watching, seem closely related to changes in obesity [24]. It is suggested that the data are sufficiently strong to advocate promoting an increase in activities of daily living as a first strategy to prevent weight gain and regain [25]. Combined observations of the activity pattern with motion sensors and PAL derived from total energy expenditure showed that it is the distribution of time attributed to inactivity and moderate-intensity activity rather than exercise that determines our total activity level [26]. In the normal PAL range, the distribution of time spent on activities with low and moderate intensity, i.e. lying, sitting and standing, and walking and cycling, determines the activity level. High activity does not have much impact as a determinant of PAL in the normal population. Based on studies on the effect of diet and exercise this observation also appeared to induce weight loss in the obese. The addition of exercise to an energyrestricted diet does not result in further weight loss [9, 10] because the cost in terms of energy of exercise training is probably compensated by a reduction of energy spent on physical activity outside the training sessions. Subjects wanting to increase their activity level should exchange low-intensity activities such as watching television or sitting behind a computer for moderate-intensity activities such as walking or cycling. Moderate-intensity activities are better tolerated than high-intensity activities, especially by the obese. The ultimate effect on PAL might have important health benefits. Subjects with a higher PAL have a higher energy expenditure and consequently a lower risk for a positive energy balance. References 1 2

3 4 5

Schoeller DA, Fjeld CR: Human energy metabolism: What have we learned from the doubly labeled water method? Annu Rev Nutr 1991;11:355–373. Prentice AM, Black AE, Coward WA, Cole TJ: Energy expenditure in overweight and obese adults in affluent societies: An analysis of 319 doubly-labelled water measurements. Int J Obes 1996; 50:93–97. Westerterp KR: Obesity and physical activity. Int J Obes 1999;23/S1:59–64. Westerterp KR, Wouters L, Van Marken Lichtenbelt WD: The Maastricht protocol for the measurement of body composition and energy expenditure with labeled water. Obes Res 1995;3/S1:49–57. Prentice AM, Goldberg GR, Murgatroyd PR, Cole TJ: Physical activity and obesity: Problems in correcting expenditure for body size. Int J Obes 1996;20:688–691.

Westerterp

68

6 7 8

9

10 11

12 13

14 15

16

17 18

19 20 21 22 23

24 25 26

Schoeller DA, Jefford G: Determinants of the energy costs of light activities: Inferences for interpreting doubly labeled water data. Int J Obes 2002;26:97–101. Schulz LO, Schoeller DA: A compilation of total daily energy expenditures and body weights in healthy adults. Am J Clin Nutr 1994;60:676–681. Ekelund U, Aman J, Yngve A, Renman C, Westerterp K, Sjöström M: Physical activity but not energy expenditure is reduced in obese adolescents: A case-control study. Am J Clin Nutr 2002; 76:935–941. Racette SB, Schoeller DA, Kushner RF, Neil KM, Herling-Iaffaldano K: Effects of aerobic exercise and dietary carbohydrate on energy expenditure and body composition during weight reduction in obese women. Am J Clin Nutr 1995,61:486–494. Kempen KPG, Saris WHM, Westerterp KR: Energy balance during 8 weeks energy-restrictive diet with and without exercise in obese females. Am J Clin Nutr 1995;62:722–729. Velthuis-te Wierik EJM, Westerterp KR, Van den Berg H: Impact of a moderately energyrestricted diet on energy metabolism and body composition in non-obese men. Int J Obes 1995;19: 318–324. Schoeller DA, Shay K, Kushner RF: How much physical activity is needed to minimize weight gain in previously obese women? Am J Clin Nutr 1997;66:551–556. Weinsier RL, Hunter GR, Desmond RA, Byrne NM, Zuckerman PA, Darnell BE: Free-living activity energy expenditure in women successful and unsuccessful at maintaining normal body weight. Am J Clin Nutr 2002;75:499–504. Hoos MB, Gerver WJM, Kester AD, Westerterp KR: Physical activity levels in children and adolescents. Int J Obes, in press. Goran MI, Sun M: Total energy expenditure and physical activity in prepubertal children: Recent advances based on the application of the doubly labeled water method. Am J Clin Nutr 1998;68 (suppl):944S–949S. Goran MI, Shewchuk R, Gower BA, Nagy TR, Carpenter WH, Johnson RK: Longitudinal changes in fatness in white children: No effect of childhood energy expenditure. Am J Clin Nutr 1998;67: 309–316. Dietz WH: Does energy expenditure affect changes in body fat in children? Am J Clin Nutr 1998; 67:190–191. Goris AHC, Westerterp-Plantenga MS, Westerterp KR: Undereating and underrecording of habitual food intake in obese men: Selective underreporting of fat intake. Am J Clin Nutr 2000;71: 130–134. Schoeller DA, Hnilicka JM: Reliability of the doubly labeled water method for the measurement of total daily energy expenditure in free-living subjects. J Nutr 1996;126:348S–354S. Glenn CC, Driscol DJ, Yang TP, Nichols RD: Genomic imprinting: Potential function and mechanisms revealed by the Prader-Wili and Angelman syndromes. Mol Hum Reprod 1997;3:321–332. Schoeller DA, Levitsky LL, Bandini LG, Dietz WW, Walczak A: Energy expenditure and body composition in Prader-Willi syndrome. Metabolism 1988;37:115–120. Holm VA, Pipes PL: Food and children with Prader-Willi syndrome. Am J Dis Child 1976;130: 1063–1067. Van Mil EGAH, Westerterp KR, Kester ADM, Curfs LMG, Gerver WJM, Schrander-Stumpel CTRM, Saris WHM: Activity related energy expenditure in children and adolescents with PraderWilli syndrome. Int J Obes 2000;24:429–434. Prentice AM, Jebb SA: Obesity in Britain: Gluttony or sloth? BMJ 1995;31:437–439. Wyatt HR, Hill JO: Let’s get serious about promoting physical activity. Am J Clin Nutr 2002;75: 449–450. Westerterp KR: Pattern and intensity of physical activity. Nature 2001;410:539.

Prof. Dr. Klaas R. Westerterp, Department of Human Biology, Maastricht University, PO Box 616, NL–6200 MD Maastricht (The Netherlands) Tel. ⫹31 43 3881628, Fax ⫹31 43 3670976, E-Mail [email protected]

Physical Activity and Obesity

69

Energy Balance in Prader-Willi Syndrome Compared to Simple Obesity Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 70–81

Model for a Peripheral Signaling Defect in Prader-Willi Syndrome Phillip D.K. Lee Department of Pediatrics, David Geffen School of Medicine at UCLA, Los Angeles, Calif., USA

Abstract Traditional hypotheses of obesity pathogenesis in Prader-Willi syndrome (PWS) have emphasized a model of primary hypothalamic dysfunction, i.e. an imbalance between central orexigenic and anorexigenic signals leading to a central satiety defect. However, in PWS, there is a selective increase in body fat, with negative effects on bone and nonbone lean compartments. This is in contrast to most other conditions of centrally mediated hyperphagia where all body compartments are positively affected. Moreover, the deficiencies of growth hormone and gonadotropins, often cited in support of a hypothalamic defect, are not universally observed in cases of PWS and rarely occur as associated primary defects in other conditions in the absence of other central hormone deficiencies. These inconsistencies prompt consideration of an alternate model, involving a primary defect in peripheral substrate shuttling resulting in increased fat deposition and deficient muscle tissue. Theoretically, this could lead to a lack of a muscle-generated satiety signal and the starvation-like eating behavior observed in PWS. The peripheral signaling defect model ties together many features of PWS-associated obesity that are not easily explained by currently accepted hypotheses. Copyright © 2003 S. Karger AG, Basel

Introduction

Selection of an appropriate theoretical construct for modeling the pathogenesis of the overeating and body composition abnormalities in Prader-Willi syndrome (PWS) has important implications for determining future directions for research and therapy. Traditional hypotheses regarding the cause of obesity and polyphagia in PWS have emphasized a model of hypothalamic dysfunction.

In this central model, there is a postulated imbalance between central orexigenic and anorexigenic signals, coupled with a disordered response to peripheral signals, resulting in an insatiable appetite, weight gain and obesity. An alternate, less-tested hypothesis is that there is a defect in peripheral signaling, resulting in a lack of normal nutrient-dependent suppression of central appetite stimulation. Before reviewing the evidence for and against these two models, it may be helpful to briefly review the status of pertinent background information regarding the regulation of human appetite and body composition.

Normal Regulation of Human Appetite

The regulation of human appetite, food intake and consequent changes in body composition are obviously complex. Animal studies have given us important information about the mechanisms that control these processes. However, findings from animal studies are not consistently mirrored by human experience. Several points related to human eating behavior are apparent from observational and research data [reviewed in 1–4]: (1) Normal eating behavior can be separated into three phases: an initial ‘drive to eat’ or hunger, an immediate feeling of fullness or satiety following a meal and a longer-lasting period of satiation. Each of these phases is influenced by numerous genetic, physiologic and environmental factors that differ between individuals. (2) The upper limit of total body mass is poorly regulated, if at all. Few effective physiologic mechanisms exist to protect against excess energy intake and weight gain. (3) On the other hand, multiple protective mechanisms exist to maintain energy balance, prevent starvation and avoid weight loss. (4) Pedigree studies demonstrate that the risk for an obese phenotype is influenced by heritable factors [5]. These factors, not yet fully characterized except in rare pedigrees with monogenic obesity, appear to play a facilitating role for weight gain in the normal population. (5) Identified peripheral satiety and satiation signals appear to have minimal effects in humans, even when present in excess amounts [reviewed in 6–8]. (6) On the other hand, inherent deficiencies of such signals can lead to excessive food intake. The condition of congenital leptin deficiency due to leptin gene mutation, resulting in excessive appetite and morbid obesity [9], is the primary example identified thus far. However, secondary hypoleptinemia, such as

Peripheral Defect in PWS

71

in athletes and in anorectic conditions [10], does not seem to induce hyperphagia or increased fat mass. (7) Some investigations suggest that humans have evolved intrinsic taste preferences for calorie-dense foods, including sugar and fat. Therefore, in the presence of compatible environmental signals, humans apparently have a natural tendency to ingest excess calories, maintain positive energy balance and favor weight gain. Such environmental signals include large food portions, high fat and sugary foods, and ready availability of food. These factors may explain the rising prevalence of obesity in the industrialized world, particularly when coupled with the reduction of physical activity associated with urbanization [11].

Normal Regulation of Body Composition

The regulation of human body composition is at least as complex as appetite regulation, and our understanding of relevant physiology is beset with an even greater number of unknown factors. Based on observation and quantitative measurements, Forbes [reviewed in 3, 4] has made the following points: (1) In the presence of excess energy intake, body weight increases. The increase in total body weight is composed of increases in both fat mass and fat-free (bone and cell) mass (FFM). (2) Deficient energy intake is characterized by loss of fat and lean mass. Obese individuals will lose proportionately more fat mass, but all individuals lose FFM during weight loss. (3) These coordinated changes in fat and lean mass during weight gain/loss have been termed the ‘companionship rule’ [3]. (4) The companionship rule can be altered to some extent by several factors, e.g. high protein intake, steroids, or physical exertion. However, it is unusual for this rule to be completely ignored in nature. (5) During fasting, obese individuals lose weight (fat and lean mass, as a percentage of starting weight) at a slower rate than nonobese individuals. Although the physiologic control of total energy intake and total body weight regulation are the subject of considerable current investigation, the mechanisms that account for substrate partitioning, compartmentalization and determination of body composition are largely uncharacterized [8, 12]. The control mechanisms that permit appropriate coordinated fed-state substrate shuttling into fat, muscle and bone are not completely defined, although various endocrine, paracrine and autocrine influences have been characterized at the level of individual cells and tissues. For example, physical exertion and androgens increase muscle cell mass.

Lee

72

However, it is not clear how these influences affect substrate shuttling into muscle in vivo.

Abnormal Regulation of Appetite and Body Composition

Given the wide variability in normal human appetite and resultant body composition and our lack of understanding of these processes, defining abnormal conditions can often be difficult. Strictly speaking, overeating per se cannot be regarded as a physiologic abnormality even if it induces pathophysiology. Therefore, defining the limits of normal versus abnormal appetite can be difficult except in extreme cases. In terms of body composition, abnormality might be defined as a violation of the companionship rule, i.e. opposing changes in fat versus lean mass. It is apparent, for example, that obesity in the general population is a consequence of normal mechanisms that follow from excess energy intake and, in this sense, may be considered an extension of normal physiology. In the normal obese individual, the increased total body weight results from increases in all clinical body compartments: fat, bone and nonbone lean mass [3, 4]. Peripheral satiation signals, such as leptin, are elevated in normal obesity [13–15]; a central resistance to these signals has been postulated to account for the ongoing ingestion of excess calories. However, recent data suggest that central responses to peripheral signals are intact in normal obesity [16]. It seems likely that higher cortical functioning may supercede basic physiologic control mechanisms. The study of normal versus abnormal appetite regulation and application of the companionship rule is limited by lack of concurrent studies of energy intake and body composition. For example, abnormal appetite regulation and hyperinsulinemia have been identified in hypothalamic obesity, but detailed body composition studies have not been published [17, 18]. In the condition of ‘growth without growth hormone (GH)’, hyperphagia and hyperinsulinemia have been postulated to account for the normal linear growth and many of these individuals appear to have increased fat and lean mass; however, detailed body composition studies are also lacking [19]. Moreover, many studies rely on indirect measures of body composition, such as body mass index, that may not accurately reflect changes in body compartments. Two hormonal conditions, GH deficiency and cortisol excess, have relatively well-characterized abnormalities in body composition, consisting of poor linear growth, low bone mineral density, high relative or absolute body fat and low nonbone lean mass. In addition, increased lean and fat mass has been demonstrated in conditions associated with hyperinsulinemia and insulin resistance. However, appetite regulation has not been well characterized in these conditions.

Peripheral Defect in PWS

73

Body Composition and Appetite in PWS

Our understanding of hyperphagia and obesity in PWS has evolved against this background of limited information regarding normal and abnormal physiology. Clinical similarities and dissimilarities between PWS and other pathologic conditions might provide important corollary information, as has been recently attempted in a comparison of PWS and hyperphagic short stature [20]. However, complete data for comparative conditions is lacking. The theory that the overeating in PWS might be primarily a behavioral or psychiatric condition, rather than an intrinsic physiologic abnormality, has been for the most part discarded. Instead, the operative idea is that the overeating and resultant obesity in PWS represent a relatively pure form of physiologic dysregulation and that characterization of this phenomenon might give insights into normal physiology. This hypothesis is supported by the extreme disorder in appetite regulation, primarily involving failure of satiation [21], and the apparent violation of the companionship rule, as discussed below. Central Model (fig. 1) The central dysfunction model is by far the most accepted construct for the pathogenesis of overeating and obesity in PWS [22]: a defect in CNS orexigenic and/or anorexigenic signals leads to a net stimulation of orexigenic pathways. The consequent overeating then leads to increased body fat and obesity. This model mirrors the proposed mechanism of non-PWS obesity, in which central stimulation of energy intake is thought to be the causative mechanism. As pointed out above, this mechanism represents a natural extension of the apparent human tendency to favor orexigenic pathways and excess energy intake. Indirect supportive evidence for a central defect in PWS comes from the supposition that other intracerebral abnormalities occur in this condition. In particular, GH deficiency, temperature dysregulation, cognitive impairment and appetite dysregulation in PWS are all related in this model through a shared, as yet unidentified, CNS pathology. In this regard, reductions in the size of the paraventricular nuclei and the volume of oxytocin-containing neurons appear to be reduced in PWS [23]. Serum levels of known potential anorexigenic peptides are high in PWS. Therefore, the central model relies on the supposition that there is CNS resistance to these peptides, that their effects are overridden by orexigenic factors, or that the dysfunctional pathways bypass usual regulatory mechanisms. In the case of leptin, high levels are correlated with body fat in PWS as in normal obese individuals [24] and leptin suppression of NPY appears to be intact [16]; therefore, there is no current evidence of a resistance phenomenon per se. Serum insulin levels are low in many individuals with PWS, but high levels are

Lee

74

Central model Cognitive defect Thermoregulatory defect

Generalized CNS abnormality Excess orexigenic signal Resistance to satiation signal?

X GH deficiency GnRH deficiency

Insatiable appetite

Obesity Abnormal body composition Abnormal sexual development

Fig. 1. Central model. The central model starts with a generalized CNS dysfunction affecting cognition and hypothalamic function. Increased orexigenic signaling, perhaps due to resistance to satiety/satiation signals, leads to hyperphagia and obesity. Hypothalamic dysfunction causes GH and GnRH deficiencies, leading to the body composition abnormalities and hypogonadism.

observed in others [reviewed in 25], arguing against insulin deficiency as a cause of the satiation defect. Clinical studies of other centrally acting postulated appetite-regulatory factors have not consistently supported a central resistance hypothesis and studies in rodent models of PWS have not been revealing. Therefore, direct evidence of a specific hypothalamic or CNS dysregulation has not yet been shown. Unlike usual exogenous obesity, weight gain in PWS is characterized by a preferential gain of body fat, with relative paucity of bone mineral and nonbone lean mass [25–27]. The co-occurrence of a GH deficiency has been postulated to account for the unusual body composition in PWS since similar changes can be seen in GH deficiency [28, 29]. Furthermore, the body composition abnormalities are alleviated with GH therapy. However, recent studies indicate that visceral fat is decreased in PWS, whereas it is increased in GH deficiency [30]. In addition, personal clinical observation suggests that the severity and early onset of body composition abnormalities in PWS are strikingly dissimilar from typical congenital GH deficiency. Nonetheless, since increased appetite alone will not result in the abnormal phenotype observed in PWS, GH deficiency or an alternate mechanism is needed to account for the unusual body composition.

Peripheral Defect in PWS

75

Overall, a defect in central signaling is an attractive model for the appetite dysregulation in PWS. This model agrees with many clinical findings in PWS and, coupled with the postulated GH deficiency, provides a coherent picture of pathophysiology. However, essential elements of this model, including the specific central signals involved, their relationship to the genetic abnormalities in PWS, and the complete pathogenesis of the abnormal body composition, have not been identified. Peripheral Model An alternate model, which has received much less attention, involves peripheral signaling. In this model, a tissue-generated satiation signal is lacking. A second defect in substrate shuttling, perhaps related to the same signaling pathway, is necessary to explain the unusual body composition. This alternate model of a peripheral signaling defect is particularly interesting when one considers potential faults in the central model. The central model depends in part on the assumption that the GH and gonadotropin deficiencies provide corollary evidence for hypophyseal and CNS dysfunction. However, neither of these deficiencies are absolute in PWS. A proportion of subjects with PWS have a normal GH response to provocative stimulation [reviewed in 25, 31, 32]. Gonadotropin secretion in PWS can be stimulated by pharmacologic agents (e.g. serotoninergic agents) and by primary gonadal failure [reviewed in 31]. An argument has been made that the GH deficiency in PWS is at the hypothalamic level (i.e. GHRH deficiency); therefore, secretagogues acting at the pituitary level or through SRIF inhibition may result in a positive response. However, this theory is contradicted by the universally low GH response to the same secretagogues observed in other forms of congenital hypothalamic deficiency. In addition, deficiencies of ACTH, TSH and vasopressin do not occur with increased frequency in PWS [26, 31]. This clearly distinguishes PWS from the more common form of hypothalamic hypopituitarism which accounts for most cases of congenital hypopituitarism [reviewed in 33]. Anatomical studies have failed to show pituitary hypoplasia or deficient GHRH or GnRH neurons in PWS [34]. The severity of hypotonia, apparently beginning in utero, and the lack of GH deficiency-related neonatal hypoglycemia further distinguish PWS from typical congenital GH deficiency. The isolated association of GH/GHRH and gonadotropin/GnRH deficiencies is unusual in nature, as might be expected by the separate ontogenic pathways of these hormone systems [35]. Defects in primordial genes, such as the Gsh-1 homeobox protein gene, can lead to GHRH and GnRH deficiencies in rodents, but this is accompanied by small pituitary size and other hormonal deficiencies [36], neither of which are observed in PWS. Overall, it appears that the GH and GnRH deficiencies may not be intrinsic to the hypophysis.

Lee

76

The GH/GHRH and gonadotropin/GnRH deficiencies could be linked through a common central regulatory mechanism or, as has been suggested by animal studies [37], by a deficiency of one system affecting the other. However, a potential mechanism for either of these possibilities has not been demonstrated in humans. Alternatively, the paired deficiencies could be linked through peripheral signals, such as nutrition or environmental signals, to which both systems are known to be susceptible [38–40]. Another line of indirect evidence for a peripheral signaling defect is inferred from the ineffectiveness of central appetite suppressants on appetite in PWS [41]. On the other hand, limited data using metformin, a peripherally acting agent [42], has shown notable effectiveness on reducing body weight [43] (unpubl. data). Finally, the poorly characterized thermoregulation defect in PWS has often been attributed to a hypophyseal defect. However, current understanding of thermoregulation emphasizes the importance of peripheral pathways [44]. In this respect, skeletal muscle, which is deficient in PWS, is thought to be a primary source of heat generation in a cold environment, i.e. ‘shivering thermogenesis’. Constructing a Peripheral Model (fig. 2) A basic peripheral model for the pathogenesis of hyperphagia and obesity in PWS as follows: (1) Following a meal, substrate is abnormally and preferentially shuttled into fat, resulting in increased fat mass and an absolute deficit in muscle and bone mass, and a relative deficit in bone mineral. This element of the model is supported by the findings of increased fat mass in normal or underweight individuals with PWS [45]. However, the mechanisms of substrate shuttling and the potentially abnormal shuttling signal are not defined. (2) The lack of muscle tissue leads to deficiency of an as yet unidentified muscle-generated satiation signal, leading to excessive and continuous preponderance of orexigenic signals. Lack of a muscle-generated satiety signal could account for the clinical impression that the hyperphagia in PWS more closely resembles nutritional deprivation than normal hunger (pers. observations). The possibility that the abnormal shuttling and satiety signals are linked or identical is an attractive concept. (3) Decreased satiation then leads to uncontrolled need for energy intake, further augmenting the tendency to increased body fat and overweight. Ancillary points in the model include: (1) The IGF-1 deficiency is linked to the substrate shuttling defect and not to the deficiency in GH secretion per se. The association of IGF-1 deficiency with low muscle mass might be compared to similar findings in kwashiorkor [46, 47]. GH treatment presumably bypasses the IGF-1 synthetic defect causing an increased tissue production of IGF-1 which contributes to the subsequent

Peripheral Defect in PWS

77

Peripheral model

Cognitive defect

Unbalanced orexigenic signal Insatiable appetite

X Decreased muscle mass

Decreased satiation signal

?

GnRH Thermoregulatory defect

Substrate shuttling defect

Increased fat

IGF deficiency

GH

Obesity Osteoporosis Abnormal sexual development

Fig. 2. Peripheral model. The peripheral model starts with a defect in substrate shuttling, resulting in preferential partitioning of ingested nutrients into fat, a deficit in muscle tissue and deficient IGF synthesis. The increased fat mass leads to diminished GH secretion, further contributing to the body composition abnormalities. The decreased skeletal muscle mass contributes to the thermoregulatory defect. Either due to the decreased skeletal muscle mass or to an intrinsic defect in the myocytes, there is deficient satiety/satiation signaling to the CNS, leading to hyperphagia. Gonadotropin deficiency, perhaps related to substrate, muscle and/or GH/IGF deficiency, causes hypogonadism, which then contributes to body composition abnormalities.

improvement in bone mineral density and nonbone lean mass. In PWS individuals without GH therapy, IGF-1 levels have been correlated with measures of bone mineral density and FFM [48]. (2) The deficient GH secretion may be related to increased body fat (i.e. via increased free-fatty acids and/or hyperleptinemia), perhaps combined with the proposed lack of muscle peripheral signaling. This finding differs from the high GH levels observed in kwashiorkor, in which both lean and fat mass are decreased [46, 47]. Perhaps relevant are reports of a muscle afferent signal for pituitary secretion of bioactive GH [49]. (3) The possibility that deficient GnRH secretion may also be linked to muscle signaling should also be considered. Although GnRH secretion has been most closely linked to body fat and leptin levels [50], these are clearly not linked in PWS in which leptin levels are high and gonadotropin secretion is

Lee

78

impaired. However, virtually all studies correlating body fat and GnRH secretion have failed to take into account that lean mass is correlated with fat mass in normal individuals. The postulated lack of the putative muscle-generated satiation signal could be related either to the low muscle mass in PWS or to an intrinsic deficiency in the myocytes. The apparent lack of GH treatment effect on eating behavior, despite improvement in nonbone lean mass, tends to argue for the latter. The peripheral signaling model ties together many observed elements of pathophysiology in PWS. However, it shares a major problem with the central signaling model, i.e. lack of characterized defective signaling factors or pathways. However, in the absence of definitive evidence for a central mechanism, future studies of obesity in PWS may do well to consider a peripheral signaling defect as an alternate model for investigation.

Conclusions

The selection of an appropriate working model for investigations of obesity in PWS is essential and must take into account both the satiation defect and the violation of the companionship rule. The central model is an extension of our basic understanding of appetite regulation and has many positive attributes. However, attempts to test this model in PWS have not been fruitful. The peripheral model relies on disordered regulation of substrate- or tissue-generated signaling, an area which is even more poorly understood than central appetite regulation. Nonetheless, the peripheral model has unifying features that argue for further investigation. Whichever model is selected, it should take into account the possibility that the body composition abnormality, rather than appetite dysregulation, may be the primary abnormality in PWS.

References 1 2 3 4 5 6 7 8

Blundell JE, Gillett A: Control of food intake in the obese. Obes Res 2001;9(suppl 4):236S–270S. Woods SC, Schwartz MW, Baskin DG, Seeley RJ: Food intake and the regulation of body weight. Annu Rev Psychol 2000;51:255–277. Forbes GB: The companionship of lean and fat. Basic Life Sci 1993;60:1–14. Forbes B: Perspectives on body composition. Opin Clin Nutr Metab Care 2002;5:25–30. Echwald SM: Genetics of human obesity: Lessons from mouse models and candidate genes. J Intern Med 1999;245:653–666. Bray GA: Afferent signals regulating food intake. Proc Nutr Soc 2000;59:373–384. Broberger C, Hökfelt T: Hypothalamic and vagal neuropeptide circuitries regulating food intake. Physiol Behav 2001;74:669–682. Jeanrenaud B, Rohner-Jeanrenaud F: Effects of neuropeptides and leptin on nutrient partitioning: Dysregulations in obesity. Annu Rev Med 2001;52:339–351.

Peripheral Defect in PWS

79

9

10 11 12 13 14 15 16

17 18

19

20 21 22 23

24

25 26

27 28 29

30

Montague CT, Farooqi S, Whitehead JP, Soos MA, Rau H, Wareham NJ, Sewter CP, Digby JE, Mohammed SN, Hurst JA, Cheetham CH, Earley AR, Barnett AH, Prins JB, O’Rahilly SO: Congenital leptin deficiency is associated with severe early-onset obesity in humans. Nature 1997; 387:903–908. Weimann E, Blum WF, Witzel C, Schwidergall S, Bohles HJ: Hypoleptinemia in female and male elite gymnasts. Eur J Clin Invest 1999;29:853–860. Kiess W, Galler A, Reich G, Kapellen T, Deutscher J, Raile K, Kratsch J: Clinical aspects of obesity in childhood and adolescence. Obes Rev 2001;2:29–36. Bray GA: Obesity, a disorder of nutrient partitioning: The MONA LISA hypothesis. J Nutr 1991; 121:1146–1162. Gundlach AL: Galanin/GALP and galanin receptor: Role in central control of feeding, body weight/obesity and reproduction. Eur J Pharmacol 2002;440:255–268. Butler MG, Carlson MG, Schmidt DE, Feurer ID, Thompson T: Plasma cholecystokinin levels in Prader-Willi syndrome and obese subjects. Am J Med Genet 2000;95:67–70. Kalra SP: Circumventing leptin resistance for weight control. Proc Natl Acad Sci USA 2001;98: 4279–4281. Goldstone AP, Unmehopa UA, Bloom SR, Swaab DF: Hypothalamic NPY and agouti-related protein are increased in human illness but not in Prader-Willi syndrome and other obese subjects. J Clin Endocrinol Metab 2002;87:927–937. Pinkney J, Wilding J, Williams G, MacFarlane I: Hypothalamic obesity in humans: What do we know and what can be done? Obes Rev 2002;3:27–34. Lustig RH, Rose SR, Burghen GA, Velasquez-Mieyer P, Broome DC, Smith K, Li H, Hudson MM, Heideman RL, Kun LE: Hypothalamic obesity caused by cranial insulin in children: Altered glucose and insulin dynamics and reversal by a somatostatin agonist. J Pediatr 1999;135:162–168. Roth C, Wilken B, Hanefeld F, Schröter W, Leonhardt U: Hyperphagia in children with craniopharyngioma is associated with hyperleptinemia and a failure in the downregulation of appetite. Eur J Endocrinol 1998;138:89–91. Gilmour J, Skuse D, Pembrey M: Hyperphagic short stature and Prader-Willi syndrome: A comparison of behavioural phenotypes, genotypes and indices of stress. Br J Psychiatry 2001;179:129–137. Lindgren AC, Barkeling B, Hagg A, Ritzén EM, Marcus C, Rossner S: Eating behavior in PraderWilli syndrome, normal weight and control groups. J Pediatr 2000;137:50–55. Swaab DF: Prader-Willi syndrome and the hypothalamus. Acta Paediatr Suppl 1997;423:50–54. Swaab DF, Purba JS, Hofman MA: Alterations in the hypothalamic paraventricular nucleus and its oxytocin neurons (putative satiety cells) in Prader-Willi syndrome: A study of five cases. J Clin Endocrinol Metab 1995;80:573–579. Goldstone AP, Brynes AE, Thomas EL, Bell JD, Frost G, Holland A, Ghatei MA, Bloom SR: Resting metabolic rate, plasma leptin concentrations, leptin receptor expression, and adipose tissue measured by whole-body magnetic resonance imaging in women with Prader-Willi syndrome. Am J Clin Nutr 2002;75:468–475. Eiholzer U: Prader-Willi Syndrome. Effects of Human Growth Hormone Treatment. Basel, Karger, 2001. Lee PDK, Hwu K, Henson H, Brown BT, Bricker JT, LeBlanc AD, Fiorotto ML, Greenberg F, Klish WJ: Body composition studies in Prader-Willi syndrome (PWS): Effects of growth hormone (GH) therapy. Basic Life Sci 1993;60:201–206. Brambilla P, Bosio L, Manzoni P, Pietrobelli A, Beccaria L, Chiumello G: Peculiar body composition in patients with Prader-Labhart-Willi syndrome. Am J Clin Nutr 1997;65:1369–1374. Lee PDK: Effects of growth hormone treatment in children with Prader-Willi syndrome. Growth Horm IGF Res 2000;10(suppl B):S77–S81. Eiholzer U, Bachmann S, l’Allemand D: Is there growth hormone deficiency in Prader-Willi syndrome? Six arguments to support the presence of hypothalamic growth hormone deficiency in Prader-Willi syndrome. Horm Res 2000;53(suppl 3):44–52. Goldstone AP, Thomas EL, Brynes AE, Bell JD, Frost G, Saeed N, Hajnal JV, Howard JK, Holland A, Bloom SR: Visceral adipose tissue and metabolic complications of obesity are reduced in Prader-Willi syndrome female adults: Evidence for novel influences on body fat distribution. J Clin Endocrinol Metab 2001;86:4330–4338.

Lee

80

31 32 33 34

35 36 37 38 39 40 41 42 43 44 45 46

47

48

49 50

Lee PDK: Endocrine and metabolic aspects of Prader-Willi syndrome; in Greenswag LR, Alexander R (eds): Management of Prader-Willi Syndrome, ed 2. New York, Spinger, 2001, pp 32–57. Burman P, Ritzén EM, Lindgren AC: Endocrine dysfunction in Prader-Willi syndrome: A review with special reference to GH. Endocr Rev 2001;22:787–799. Reiter EO, Rosenfeld RG: Normal and abnormal growth; in Wilson JD, Foster DW, Kronenberg HM, Larsen PR (eds): Williams Textbook of Endocrinology, ed 9. Philadelphia, Saunders, 1998, p 1465. Goldstone AP, Unmehopa UA, Swaab DF: Hypothalamic growth hormone-releasing hormone (GHRH) cell number in human illness, obesity and Prader-Willi syndrome (abstract OR 38-5). 84th Annual Meeting of the Endocrine Society, San Francisco, 2002. Scully KM, Rosenfeld MG: Pituitary development: Regulatory codes in mammalian ontogenesis. Science 2002;295:2231–2235. Li H, Zeitler PS, Valerius MT, Small K, Potter SS: Gsh-1, an orphan Hox gene, is required for normal pituitary development. EMBO J 1996;15:714–724. Wilson ME: The impact of the GH-IGF-I axis on gonadotropin secretion: Inferences from animal models. J Pediatr Endocrinol Metab 2001;14:115–140. Bellastella A, Pisano G, Iorio S, Pasquali D, Orio F, Venditto T, Sinisi AA: Endocrine secretions under abnormal light-dark cycles and in the blind. Horm Res 1998;49:153–157. Cameron JL, Helmreich DL, Schreihofer DA: Modulation of reproductive hormone secretion by nutritional intake: Stress signals versus metabolic signals. Hum Reprod 1993;8(suppl 2):162–167. Rissman EF: Behavioral regulation of gonadotropin-releasing hormone. Biol Reprod 1996;54: 413–419. Yaryura-Tobias JA, Grunes MS, Bayles ME, Neziroglu F: Hyperphagia and self-mutilation in Prader-Willi syndrome: Psychopharmacological issues. Eat Weight Disord 1998;3:163–167. Kirpichnikov D, McFarlane SI, Sowers JR: Metformin: An update. Ann Intern Med 2002;137: 25–33. Chan NN, Feher MD, Brdiges NA: Metformin therapy for diabetes in Prader-Willi syndrome. J R Soc Med 1998;91:598. Argyropoulos G, Harper ME: Uncoupling proteins and thermoregulation. J Appl Physiol 2002;92: 2187–2198. Eiholzer U, Blum WF, Molinari L: Body fat determined by skinfold measurements is elevated despite underweight in infants with Prader-Labhart-Willi syndrome. J Pediatr 1999;134:222–225. Zamboni G, Dufillot D, Antoniazzi F, Valentini R, Gendrel D, Tato L: Growth hormone-binding proteins and insulin-like growth factor-binding proteins in protein-energy malnutrition, before and nutritional rehabilitation. Pediatr Res 1996;39:410–414. Soliman AT, Hassan AE, Aref MK, Hintz RL, Rosenfeld RG, Rogol AD: Serum insulin-like growth factors I and II concentrations and growth hormone and insulin responses to arginine infusion in children with protein-energy malnutrition before and after nutritional rehabilitation. Pediatr Res 1986;20:1122–1130. Van Mil EG, Westerterp KR, Gerver WJ, Van Marken Lichtenbelt WD, Kester AD, Saris WH: Body composition in Prader-Willi syndrome compared with nonsyndromal obesity: Relationship to physical activity and growth hormone function. J Pediatr 2001;139:708–714. McCall GE, Grindeland RE, Roy RR, Edgerton VR: Muscle afferent activity modulates bioassayable growth hormone in human plasma. J Appl Physiol 2000;89:1137–1141. Mann DR, Plant TM: Leptin and pubertal development. Semin Reprod Med 2002;20:93–102.

Phillip D.K. Lee, MD, Professor of Endocrinology, Mattel Children’s Hospital at UCLA, 10833 LeConte Avenue, Room MDCC 22-315, Los Angeles, CA 90095-1752 (USA) Tel. ⫹1 310 825 6244, Fax ⫹1 310 206 5843, E-Mail [email protected]

Peripheral Defect in PWS

81

Energy Balance in Prader-Willi Syndrome Compared to Simple Obesity Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 82–85

Discussion

Assessment of Body Composition in Children with Prader-Willi Syndrome or Simple Obesity (K.J. Ellis) A.L. Carrel, Madison, Wisc.: You mentioned that for DEXA, there are very good references for all different ages. And yet there is still continuous debate about pediatric data or references. Are the references just for bone, or are there accepted references for body composition as well? And do the references differ from one brand (of DEXA machine) to another? K.J. Ellis, Houston, Tex.: You have to be sure you use the references for the same machine, because there are significant differences between the machines. There is no international organization that has established pediatric references for body composition, like you have for BMI. But there is an effort going on right now with the NCHS (US National Child Health Survey) to do this work. These references, when established, will only start at an age of 5 years. Those of us who have obtained body composition data in infants are attempting to derive a common definition. Right now, for bone you can go on my website (www.bcm.tmc.edu/bodycomplab) and calculate Z-scores. However, you can’t get it for lean tissue. If you write to me, I can do the analysis unofficially. But right now, it is not on the website. Body Composition in Infants with and without Prader-Willi Syndrome: Principles and Pitfalls (C. Fusch) Anon: Is it common practice to measure body composition in your neonatal intensive care department, and which of the methods do you prefer in a practical setting? C. Fusch, Greifswald: We use DEXA to examine growth of and bone mineralization status in preterm infants. We use other methods for research. We use bioimpedance and skinfold thickness as you have seen, to validate the pattern of growth, particularly for children that are small or large for gestational age. However, for routine measurements, we use DEXA.

Physical Activity and Obesity (K. Westerterp) M.B. Ranke, Tübingen: I may just add a little anecdote: 25 years ago, I saw my first patient with PWS. The boy was 13 years old and the parents came to me and said the boy had just had his confirmation at the age of 13 and at that time children are usually given watches. That was before the time when they had electronic watches. So, this child was given a Swiss automatic watch, you had to move your arm to keep it wound up. The father said he had given this very expensive watch to his child, but it doesn’t work. And this just indicates that children with PWS don’t move. So part of the problem is low energy expenditure. Muscle Hypotonia and Effects of Physical Training (Y. Nordmann) K.R. Westerterp, Maastricht: How do your subjects feel during and after this program? Is it a sort of self-stimulating system, which might explain the later decline in performance? Is it something, they don’t feel happy with? You would think that they are fitter and they can move around easier due to the training. Y. Nordmann, Zurich: I think it was rather a problem of motivation. We saw it also in the training compliance protocols, after 2 months of training, they were a little bit bored with this exercise. I think we can avoid this problem if we do different muscle group training programs. R.L. Leibel, New York, N.Y.: It seems to me that the number of calories expended in this kind of exercise is quite small. So I am surprised that you get any change in body composition from this kind of activity. Have you tried to calculate the calories spent and the amount of tissue gained or lost? They don’t seem to jibe, unless it’s much more energy costly for these kids to do this exercise. Nordmann: We did not calculate caloric expenditure, but there are two possible explanations for the apparent discrepancy. First, there could be a geometric reason. If the calf circumference increases, it may be because the muscle mass is bigger. Secondly, there could be a short loop regulation system for body and fat mass, but that would be the first example of such a situation. Leibel: No change in total body weight or skinfold? Nordmann: No, no change. Model for a Peripheral Signalling Defect in Prader-Willi Syndrome (P.D.K. Lee) Leibel: The leptin-deficient and melanocortin receptor-deficient animals have rather striking differences in body composition that could be described as being the result of partitioning differences. The melanocortin 4-deficient mouse shows an increase in both lean body mass and adipose tissue. The ob mouse, on the other hand, has the phenotype that you and others described for Prader-Willi syndrome, a relative deficiency of lean body mass and an increase in fat mass. It remains unclear whether all of the ob body composition phenotype is due to

Discussion

83

central mechanisms or whether there are peripheral actions of leptin that may account for a part of it. But certainly these animals show this relative deficiency of lean mass to fat mass, as you are probably well aware. P.D.K. Lee, Los Angeles, Calif.: I am not that familiar with body composition studies in rodents. However, my impression is that while lean mass is not as affected, it does go up with the fat mass. Leibel: The lean mass of the ob mouse is so strikingly deficient that you can even see it in the gross size of the brain of the animal. And it is said that leptin replacement actually increases brain size of treated animals. So they have very striking deficit in lean body mass, relative to their fat mass. The ratios would be in the same directions, as for Prader-Willi syndrome. What I am saying is that leptin deficiency itself is capable of producing the phenotype that you are talking about, although clearly these children don’t appear to have leptin deficiency. D. l’Allemand, Zurich: I think, in order to understand the defect of a substrate partitioning, it would be important to know the fat content of the muscle mass, because that would have important implications for the regulation of insulin resistance. Do you know of such investigations in children with PWS, or does Professor Ellis know of such investigations and the suitable methods to study this question? Would a DEXA for example be sufficient to measure that? Lee: I don’t have any information regarding this question. l’Allemand: Since you suppose a defect in substrate shuttling or partitioning and I asked myself, if there is already any information, any evidence for that assumption. Lee: In PWS? What I mean by substrate partitioning is basically, when you eat a meal somehow your body has to decide how much of the nutrient goes into protein, fat and bone. And at least I could find very little information on this topic either in normal or in Prader-Willi subjects. However, if one looks at the body composition in Prader-Willi patients, there must be a defect. The more they eat, the more they just gain more and more fat mass. But I don’t know if I answered your question. Ranke: I think the question of Dagmar l’Allemand is a very important question in the context of the whole session of the afternoon, because we have been discussing body composition. But that is a rather static approach to the problems and I think one has to look, where the research in type II diabetes is headed. Muscle is actually one of the essential tissues and new MRI and SPECT techniques allow analysis of functional muscle and perhaps other tissues. In order to prove this very interesting hypothesis, most probably one has to include investigations with Prader-Willi patients. Some of these investigations are actually not invasive. The other approach would most probably include histological investigations to see if leucine and other substances are dealt with in another way in PWS compared to simple obesity.

Energy Balance in Prader-Willi Syndrome Compared to Simple Obesity

84

A.P. Goldstone, London: Two comments: Firstly, humans and rodents are very dissimilar, especially in terms of the hypothalamic-pituitary axis, because leptindeficient and resistant animals have activation of the hypothalamopituitary-adrenal axis, which probably contributes to a lot of the body composition changes and we do not see that in the leptin-deficient humans or in PWS. So I think we have to be careful using the rodent model of hypothalamic abnormalities in humans. Secondly, in defense of the hypothalamic central story, we still know so little about the expression of the PWS region genes in the brain and hypothalamus. We must also remember the Angelman syndrome story, where there turned out to be regional imprinting, where imprinting only occurred in certain brain regions that correlated with the phenotype selectivity. We also need a better idea of where and when the genes are expressed normally during different stages of development, and whether there are temporal changes in imprinting. These studies may help explain why the hypothalamic abnormalities in PWS are so selective. Finally, we also need to move away from just being hypothalamic in our thoughts. There are many other areas of the brain that we need to examine in explaining PWS phenotypes. Lee: I personally have not discarded the hypothalamic model either, but it just bothers me that with all of the works done in this area, there have been very few clinically relevant results. When we look at the clinical picture, it doesn’t really match the modelling in my opinion. Ranke: Just to come back to one of the former lectures. I would always think that one argument for the central origin is the mobility, the motor problem. But from child development of course we know that motor development is a reciprocal matter. You know, the child has to make the experience. So, there is the efferent pathway that is influencing central development. It is a little bit an issue, where the lack of experience during the early neonatal phase is actually the consequence of a persistent inability to move. And I question whether all the exercises that were presented here in Dr. Nordmann’s study, driving around in Switzerland, are actually going to have a lasting result, because they don’t want to move. I mean, they have some kind of an impairment in the drive to move. If you want to have some permanent effects, you have to inbuild some motivation to do so. And that is something that most probably needs to be learnt at a very early stage and not later on. Leibel: So another point in favor of hypothalamic centrality of the syndrome is that many of the genes that are in the minimum physical interval for the PWS are expressed only for the paternal allele in the brain. Lee: I don’t think that the function of genes have been identified. Leibel: That’s correct. But the ones that are known appear to be primarily centrally expressed. And the imprinting shows up in terms of the paternal allele being expressed in the hypothalamus.

Discussion

85

Comorbidities or Fundamental Defects of Obesity Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 86–92

Characterization of Hyperphagia in Prader-Willi Syndrome Ann Christin Lindgren Pediatric Endocrine Unit, Department of Woman and Child Health, Karolinska Hospital, Stockholm, Sweden

Abstract Prader-Willi syndrome (PWS) is a complex neurogenetic disorder characterized by muscular hypotonia and feeding difficulties with failure to thrive during the first months of life. After the age of 2 years excessive eating occurs resulting in morbid obesity if the caloric intake is not restricted. This hyperphagia found in subjects with PWS is hypothesized to be due to hypothalamic dysfunction, thus the eating behaviour of individuals with PWS might illustrate how hypothalamic dysfunction affects eating behaviour. Different hormones and neuropeptides are involved in the regulation of appetite but still the underlying mechanism is not fully understood. The eating behaviour in children and adults with PWS in comparison with the behaviour in obese subjects has been studied by several authors. In these studies, the eating behaviour observed has shown a decreased satiety rather than an increased hunger in individuals with PWS. This observation was confirmed by studying the microstructure of the eating behaviour in children with PWS compared with obese and normal weight control subjects of the same age group. In this study the subjects with PWS had a longer duration of eating and a slower initial eating rate compared to both obese and normal weight subjects. The majority of the eating curves of the subjects with PWS are non-decelerating (linear or accelerating) compared to 10% of the normal weight and 30% of the obese subjects. Consequently, the microstructure of the eating behaviour in subjects with PWS differs from obese and normal weight controls. Subjects with PWS have a long duration of consumption, a slow eating rate and a high prevalence of non-decelerating eating curves. Thus, the eating behaviour found in subjects with PWS might be due to decreased satiation rather than increased hunger but more research is warranted. Copyright © 2003 S. Karger AG, Basel

Introduction

Prader-Willi syndrome (PWS) is a complex neurogenetic disorder. The main clinical characteristic at the beginning of life is the feeding difficulties

due to severe muscular hypotonia. However, after the age of 2 years excessive eating develops resulting in severe obesity if the caloric intake is not restricted [1, 2]. It seems that these patients have an apparently insatiable appetite in combination with other characteristic features like hypogonadism with delayed secondary sexual development, short stature, learning disabilities and a typical phenotype [3, 4]. Obsessive-compulsive symptoms [5, 6], skin picking, stubbornness, temper tantrums, as well as ventilatory dysfunction have also been reported in subjects with PWS [7, 8]. The principal genetic defect associated with the condition is a loss of the active paternally derived chromosome 15q11-q13. This is due to a deletion occurring in 70–75% of affected individuals. Several other abnormalities have also been linked with the syndrome, 20–25% of patients exhibit maternal disomy of the same region of chromosome 15, 2–5% have imprinting centre mutations and 1% have translocations [9, 10]. The individual gene or genes from within 15q11-q13 that cause the condition have yet to be identified. The combination of several autonomic dysfunctions that affect appetite regulation, growth, pubertal development, high pain threshold, control of breathing and alertness is suggestive of a common underlying hypothalamic-pituitary dysfunction, although so far, no specific morphologic lesions have been located. However, magnetic resonance imaging has revealed an abnormal bright spot in the posterior pituitary lobe of some affected subjects, which is considered to be a sign of hypothalamic dysfunction [11], and pituitary hypoplasia is also frequently observed in this syndrome [12]. Autopsies of 5 patients with PWS indicated that the paraventricular nucleus was reduced in size and reduced by 42–54% in numbers of oxytocin-secreting neurones [13]. In addition, Martin et al. [14] found that CSF oxytocin was elevated in PWS compared with normal controls, especially female subjects. Further irregularities associated with PWS have been identified, including a 30% reduction in GHRH-releasing neurones in the nucleus arcuatus, a downregulation of neuropeptide Y (NPY) and a deficiency of vasopressin [15]. Both NPY and oxytocin are hormones involved in the regulation of appetite.

Possible Mechanisms Involved in Excessive Appetite in PWS

The regulation of food intake and satiety is complex. Different neuropeptides and hormones are involved in this regulation. Several lines of evidence suggest that both GABAergic, serotonergic and neuropeptidergic mechanisms play an important role in excessive eating. It is well known that serotonin influences eating both in animals and humans. A reduced food intake leading to weight loss has been demonstrated with drugs that increase serotonin [16–18].

Hyperphagia in Prader-Willi Syndrome

87

Recently, Åkefeldt et al. [19] reported an increased concentration of the serotonin metabolite 5-HIAA in CSF of individuals with PWS, which supports the hypothesis of an increased serotonin turnover in PWS and possibly decreased synaptic serotonin transmission. The genes for the subunits of ␤-3, ␣-5 and ␥-3 of the GABA-A receptor have been located on the proximal long arm of chromosome 15, which is deleted in most subjects with PWS [20, 21]. Elevated blood levels of GABA have been found in subjects with PWS compared to age- and weight-matched controls [22]. High densities of GABA-A receptors are found in the anterior and medial hypothalamus while in other parts of the region more modestly so [20]. Since the ventromedial hypothalamus is known to be a satiety centre and the lateral hypothalamus a feeding centre, it may be reasonable to suspect alterations of the GABA-A receptor distributions in these regions may result in elevated GABA levels in the presence of normal GABA-B receptors. It might be reasonable to postulate that such alterations could reduce satiety or increase feeding. Many other neuropeptides are involved in the regulation of hunger and satiation, including e.g. leptin, NPY, cholecystokinin (CCK), insulin, pancreatic polypeptide and the recently discovered ghrelin. The hormone leptin, secreted by fat tissue, is one of the actors in this regulation while NPY is its potential mediator on the hypothalamic level. Administration of leptin has been shown to inhibit the synthesis and release of NPY in hypothalamus, thereby providing a potential mechanism for the reduction in food intake and subsequent weight loss as has been shown in rodents [23]. Thus, NPY increases with starvation and leptin levels are decreased. In several studies leptin levels in subjects with PWS have been found to follow the pattern generally observed in normal individuals and correlate to the fat mass or body mass index [24–26]. CCK is another hormone involved in mediating the satiety response to meals. This hormone, found in high levels in the gut and brain, has been studied in subjects with PWS. Free fatty acids (FFA) are responsible for the stimulation of CCK release after a fatty meal and both CCK and FFA rise in tandem in normal individuals. In subjects with PWS, there was a lack of responsiveness of CCK release to rising plasma FFA levels following a high-fat meal or stomach distension during eating [27]. An impaired pancreatic polypeptide secretion in response to a meal has been reported in subjects with PWS [28]. All these studies indicate an abnormality of satiety in PWS. A recently discovered hormone, ghrelin secreted from the gut, has been demonstrated to play an important role in the appetite regulation. Ghrelin increases food intake, body weight and growth hormone secretion in humans. In subjects with PWS, plasma ghrelin has been reported to be 4.5-fold higher than in equally obese controls [29]. The underlying mechanism has not been fully explained.

Lindgren

88

Behavioural Studies of Excessive Appetite in PWS

The eating behaviour in PWS was the topic of several previous studies. In these studies the eating behaviour in subjects with PWS has been compared with the behaviour in controls. Zipf and Berntson [30] have studied the eating behaviour in children with PWS compared with obese children. Holland et al. [2] have studied adults with PWS compared with normal weight subjects, and Fieldstone et al. [31] compared children and young adults with PWS with obese and normal weight controls. In all these studies the test meal consisted of free access to sandwich quarters, and the participants were asked to sit at the table where the sandwiches were served for a whole hour, even if they had finished eating. In these studies, the major difference found between subjects with PWS and controls was the long duration of the consumption of sandwich quarters. The subjects with PWS in all these studies continued to eat for the full hour, whereas the obese children and normal weight controls in these studies generally had finished eating after 15 min. Similar observations with normal weight adults were found in the study by Holland et al. [2]. However, in these experimental settings where the subjects were asked to remain at the table with the sandwiches for a full hour this might have influenced the results. The long duration of consumption found in subjects with PWS resulted in a greater food intake compared with all control groups. All these studies demonstrated an impaired satiety response in subjects with PWS. The study by Holland et al. [2] further confirms this observation, as the reduction in feelings of hunger in subjects with PWS was found to correlate with the extent of changes in blood glucose levels which often were above the normal physiological range, suggesting an abnormality of satiety in PWS. In the studies by Barkeling et al. [32, 33] and Lindgren et al. [34] the microstructure of eating behaviour in obese and non-obese children with PWS compared to obese healthy and normal weight children has been measured by ‘Viktor’, a universal eating monitor. A large number of eating parameters were measured by the computerized eating monitor while the children were served a standardized hot meal in excess. The plate of food was placed on top of a hidden scale linked to the computer and when the food is eaten the change in food weight is registered continuously and an eating curve is displayed on-line. After the meal, the eating data is fitted to a polynom and the computer calculates the amount of food eaten, time of consumption, eating rate (initial and total) and rate of deceleration. This study showed that the subjects with PWS had a longer duration of consumption compared with both the obese and normal weight subjects, but all children with PWS finished eating within 45 min. After having finished eating all children were immediately asked to leave the room while the subjects in the previously reported studies were asked to remain

Hyperphagia in Prader-Willi Syndrome

89

in the room for one full hour. In this study the eating rates were slower in the subjects with PWS compared with obese and normal weight children. There was a different eating pattern in children with PWS, with a linear or accelerating curve, compared to healthy obese and normal weight children which had an accelerating and decelerating pattern, respectively [34]. This difference in the eating pattern found in individuals with PWS suggests that the eating behaviour might be due to a decreased satiation rather than increased hunger. Food preferences among individuals with PWS have been studied. In one study the subjects with PWS showed a pattern of food preferences that was similar to that of their nonaffected controls [35]. However, in a recent paper by Fieldstone et al. [36] the subjects with PWS had a preference for high carbohydrate foods over high protein foods and high protein foods over high fat foods, while normal weight and obese control subjects demonstrated no difference in food preferences.

Conclusion

Excessive eating, weight gain, blunted growth and abnormal body composition, characteristics of PWS, have been the focus of much research, as well as compulsivity and further maladaptive behaviour in this population. The compulsive symptoms, similar to the obsessive compulsive disorder, observed in PWS probably interact with the incessant hunger and lack of satiation to engender the intense preoccupation with food and food seeking characteristics. Still it is not obvious whether these subjects with PWS have an increased hunger or decreased satiation or both. A dysregulation of serotonergic, GABAergic and/or neuropeptidergic systems in the regulation of food intake and satiety is likely but additional research is warranted.

References 1 2 3 4 5 6

Cassidy SB: Prader-Willi syndrome. J Med Genet 1997;34:917–923. Holland AJ, Treasure J, Coskeran P, Dallow J, Milton N, Hillhouse E: Measurement of excessive appetite and metabolic changes in Prader-Willi syndrome. Int J Obes 1993;17:527–532. Cassidy SB: Prader-Willi syndrome. Curr Probl Pediatr 1984;14:1–55. Ehara H, Ono K, Takeshita K: Growth and development patterns in Prader-Willi syndrome. J Intellect Disabil Res 1993;37:479–485. Dykens EM, Leckman JF, Cassidy SB: Obsessions and compulsions in Prader-Willi syndrome. J Child Psychol Psychiatry 1996;37:995–1002. Dimitropoulos A, Feurer ID, Roof E, Stone W, Butler MG, Sutcliffe J, Thompson T: Appetite behaviour, compulsivity and neurochemistry in Prader-Willi syndrome. Ment Retard Dev Disabil Res Rev 2000;6:125–130.

Lindgren

90

7 8

9 10 11 12

13

14

15 16 17 18 19 20

21

22

23

24

25 26

27 28

Schulter B, Buscbatz D, Trowitzsch E, Aksu F, Andler W: Respiratory control in children with Prader-Willi syndrome. Eur J Pediatr 1997;156:65–68. Lindgren AC, Hellström LG, Ritzén EM, Milerad J: Growth hormone treatment increases CO2-response, ventilation, and central respiratory drive in children with Prader-Willi syndrome. Eur J Pediatr 1999;158:936–940. Wharton RH, Loechner KJ: Genetics and clinical advances in Prader-Willi syndrome. Curr Opin Pediatr 1996;8:618–624. Nicholls RD, Saitoh S, Horsthemke B: Imprinting in Prader-Willi and Angelman syndromes. Trends Genet 1998;14:194–200. Miller L, Angulo M, Price D, Taneja S: MR of the pituitary in patients with Prader-Willi syndrome: Size determinations and imaging findings. Pediatr Radiol 1996;26:43–47. Schmidt H, Bechold S, Schwarz HP: Prader-Labhart-Willi syndrome: Auxological response to a conventional dose of growth hormone in patients with classical growth hormone deficiency. Eur J Med Res 2000;5:307–310. Swaab DF, Purba JS, Hofman MA: Alterations in the hypothalamic paraventricular nucleus and its oxytocin neurones (putative satiety cells) in Prader-Willi syndrome. A study of five cases. J Clin Endocrinol Metab 1995;80:573–579. Martin A, State M, Anderson GM, Kaye WM, Hanchett JM, McConaha CW, North WG, Leckman JF: Cerebrospinal fluid levels of oxytocin in Prader-Willi syndrome: A preliminary report. Biol Psychiatry 1998;44:1349–1352. Swaab DF: Prader-Willi syndrome and the hypothalamus. Acta Paediatr Suppl 1997;423:50–54. Luo S, Li ETS: Effects of repeated administration of serotonergic agonists on diet selection and body weight in rats. Pharmacol Biochem Behav 1991;3:495–500. Karnarek RB, Glick AL, Marks-Kaufman R: Dietary influences on the acute effects of anorectic drugs. Physiol Behav 1991;49:149–152. Selikowitz M, Sunman J, Pendergast A, Wright S: Fenfluramine in Prader-Willi syndrome: A double-blind, placebo controlled trial. Arch Dis Child 1990;65:112–114. Åkefeldt A, Ekman R, Gillberg C, Mansson JE: Cerebral fluid monoamines in Prader-Willi syndrome. Biol Psychiatry 1998;44:1321–1328. Wagstaff J, Knoll JHM, Fleming J, Kirkness EF, Martin-Gallardo A, Greenberg F, Graham JM, Menninger J, Ward D, Venter JC: Localization of gene encoding the GABA-A receptor beta-3 unit to Angelman/Prader-Willi region of human chromosome 15. Am J Hum Genet 1991;49:330–337. Meguro M, Mitsuya K, Sui H, Shigenami K, Kugoh H, Nakao M, Oshimura M: Evidence for uniparental, paternal expression of the human GABA-A receptor subunit genes, using microcellmediated chromosome transfer. Hum Mol Genet 1997;6:2127–2133. Ebert MH, Schmidt DE, Thompson T, Butler MG: Elevated plasma gamma-aminobutyric acid (GABA) levels in individuals with either Prader-Willi syndrome or Angelman syndrome. J Neuropsychiatry Clin Neurosci 1997;9:75–80. Stephens TW, Basinski M, Bristow PK, Bue-Valleskey JM, Burgett SG, Craft L, Hale J, Hoffmann J, Hsiung HM, Kriauciunas A, MacKeller W, Rosteck PR, Schoner B, Smith D, Tinsley FC, Zang XY, Heinman M: The role of neuropeptide Y in the anti-obesity action of the obese gene product. Nature 1995;377:530–532. Lindgren AC, Marcus C, Skwirut C, Elimam A, Hagenäs L, Schalling M, Anvret M, Lönnqvist F: Increased leptin messenger RNA and serum leptin levels in children with Prader-Willi syndrome and non-syndromal obesity. Pediatr Res 1997;42:593–596. Butler MG, Moore J, Morawiecki A, Nicolson M: Comparison of leptin protein levels in PraderWilli syndrome and control individuals. Am J Med Genet 1998;75:7–12. Bueno G, Moreno LA, Pineda I, Campos J, Ruibal JL, Juste MG, Casado E, Bueno M: Serum leptin concentrations in children with Prader-Willi syndrome and non-syndromal obesity. J Pediatr Endocrinol Metab 2000;13:425–430. Butler MG, Carlson MG, Schmidt DE, Feuter ID, Thompson T: Plasma cholecystokinin levels in Prader-Willi syndrome and obese subjects. Am J Med Genet 2000;95:67–70. Zipf WB, O’Dorisio TM, Cataland S, Dixon K: Pancreatic polypeptide response to protein meal challenges in obese but otherwise normal children and obese children with Prader-Willi syndrome. J Clin Endocrinol Metab 1983;57:1074–1080.

Hyperphagia in Prader-Willi Syndrome

91

29 30 31

32

33 34 35 36

Cummings DE, Clement K, Purnell JQ, Vaisse C, Foster KE, Frayo RS, Schwartz MW, Basdevant A, Weigle DS: Elevated plasma ghrelin levels in Prader-Willi syndrome. Nat Med 2002;8: 643–644. Zipf WB, Berntson GG: Characteristics of abnormal food-intake patterns in children with PraderWilli syndrome and study of effects of naloxone. Am J Clin Nutr 1987;46:277–281. Fieldstone A, Zipf WB, Sarter MF, Berntson GG: Food intake in Prader-Willi syndrome and controls with obesity after administration of benzodiazepine receptor agonist. Obes Res1998; 6/1:29–33. Barkeling B, Rössner S, Björvell H: Effects of a high-protein meal (meat) and a high-carbohydrate meal (vegetarian) on satiety measured by automated computerized monitoring of subsequent food intake, motivation to eat and food preferences. Int J Obes 1990;14:743–751. Barkeling B, Rössner S, Sjöberg A: Methodological studies on single meal food intake characteristics in normal weight and obese men and women. Int J Obes 1995;19:284–290. Lindgren AC, Barkeling B, Hägg A, Ritzén EM, Marcus C, Rössner S: Eating behaviour in PraderWilli syndrome, normal weight and obese control groups. J Pediatr 2000;137:50–55. Rankin KM, Mattes RD: Role of food familiarity and taste quality in food preferences of individuals with Prader-Willi syndrome. Int J Obes Relat Metab Disord 1996;20:759–762. Fieldstone A, Zipf WB, Schwartz HC, Berntson GG: Food preferences in Prader-Willi syndrome, normal weight and obese controls. Int J Obes 1997;21:1046–1052.

Ann Christin Lindgren, MD, PhD, Pediatric Endocrine Unit Q 02:08, Astrid Lindgren’s Childrens Hospital, Karolinska Hospital, SE–171 76 Stockholm (Sweden) Tel. ⫹46 733 380 816, Fax ⫹46 851 775 128, E-Mail [email protected]

Lindgren

92

Comorbidities or Fundamental Defects of Obesity Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 93–101

Consequences of Impaired Growth Hormone Secretion for Body Composition and Metabolism in Obesity and Prader-Willi Syndrome Martin Wabitsch Endocrine Research Laboratory, Department of Pediatrics, University Hospital, Ulm, Germany

Abstract Simple obesity in children is associated with impaired growth hormone (GH) secretion but increased circulating insulin-like growth factor-1 (IGF-1) levels. In obese adults with an abdominal fat distribution pattern and an increase of visceral fat impaired GH secretion is associated with signs of the metabolic syndrome which are ameliorated with GH treatment. GH treatment in such patients also results in a reduction and a redistribution of body fat with a reduction of visceral fat. GH deficiency in patients with Prader-Willi syndrome (PWS) is not only explained by their obese state but also by decreased secretion due to hypothalamic dysfunction. GH deficiency in PWS patients is associated with low circulating IGF 1 levels and a central fat distribution pattern with an increase in truncal subcutaneous fat. However, visceral fat mass is reduced. Even in normal-weight patients with PWS fat mass is increased. The disturbances in lipid and carbohydrate metabolism seen in these patients are partly explained by GH deficiency. The clinical findings in simple obese individuals and in PWS patients indicate that adipose tissue is a target organ for the action of GH. Lack of the hormone results in an increase in adipose tissue mass due to an increase in adipocyte volume especially at central body sites. The effects of GH on adipose tissue mass can readily be studied in in vitro incubated fat cells. The net result of chronic exposure of fat cells to GH is an increase in lipolysis and a suppression of the metabolic effects of insulin-including glucose transport, glucose oxidation and lipid accumulation finally aiming at reducing the volume of the adipocytes. Copyright © 2003 S. Karger AG, Basel

Body Composition in Simple Obesity and Prader-Willi Syndrome

Simple obesity is associated with suppressed levels of circulating growth hormone (GH) characterized by a blunted secretion with fewer secretory events and a shorter half-life compared with that seen in normal-weight subjects [1]. GH deficiency is a central symptom of the Prader-Willi syndrome (PWS) which cannot only be explained by the sometimes severe form of obesity. It is suggested that the hypothalamic dysfunction is an additional cause of decreased GH secretion. An earlier report showing a 30% reduction in GHRH-releasing neurones in the nucleus arcuatus [2] has not been confirmed in a recent analysis [3]. Reduced volumes of the pituitary have also been found [4]. In simple obesity deficiency of circulating GH is not recognized as an additional disease but seen as an associated endocrine alteration. However, it can be assumed that the lack of circulating biologically active GH in persons with obesity results in a lack of the metabolic effects of the hormone. In this respect it is interesting to see that GH deficiency can frequently be diagnosed in obese individuals with abdominal body fat distribution and metabolic syndrome. Treatment of obese individuals with recombinant human GH even with a low dose [5] has several positive effects on the metabolic disturbances in these patients and also leads to a reduction of fat mass especially of visceral fat. Due to its metabolic effects on adipose tissue it can be suggested that biological GH deficiency in obese individuals is aggravating the obese state by the lack of the lipolytic activity of the hormone. Patients with PWS are characterized mostly by severe obesity, to which the primary GH deficiency also contributes. Children with PWS demonstrate a profoundly abnormal body composition with a phenotype that is at least in part comparable with that seen in classical GH deficiency. Interestingly, also in young underweight patients with PWS the fat mass is increased in relationship to lean body mass as compared to healthy, age- and sex-matched controls [6]. A special feature of PWS is a reduction of muscle mass and a central accumulation of subcutaneous fat [7] as determined by skinfold thickness measurements. Recent studies have investigated visceral fat mass in patients with PWS. Interestingly and in contrast to adult patients with GH deficiency and with simple obesity patients with PWS have reduced visceral fat [3]. The specific features of body composition in patients with simple obesity and PWS are shown in table 1.

Metabolic Alterations in Simple Obesity and PWS Related to GH Deficiency

Table 2 shows alterations of some selected metabolic parameters frequently found in simple obesity and PWS. It should be noted that in children

Wabitsch

94

Table 1. Characteristics of body composition in patients with simple obesity and PWS Simple obesity

PWS

Lean body mass Muscle mass Bone mass

increased increased increased

decreased decreased decreased

Fat mass Visceral fat Truncal fat

increased increased increased

increased decreased increased

Lean body mass/fat mass ratio

decreased

decreased

Table 2. Metabolic alterations in simple obesity and PWS related to GH deficiency Simple obesity

PWS

Children Fasting insulin Fasting glucose

increased increased

decreased normal

Adults Fasting insulin Fasting glucose

increased increased

increased (in obesity) increased (in obesity)

Triglycerides LDL cholesterol HDL cholesterol

increased increased decreased

(increased) increased decreased

with PWS in contrast to the situation in severe obese adults with PWS fasting insulin levels are decreased as compared to control individuals of the same weight. In addition, fasting glucose levels are normal. It can be suggested that individuals with PWS have a reduced secretory capacity of pancreatic ␤-cells, which also persists throughout GH substitution. With the increase in age and the amount of body fat both parameters change. After a critical cutoff of elevated fat mass, insulin secretion is slightly enhanced and modest signs of insulin resistance can be observed. In adult PWS patients with extreme obesity type 2 diabetes mellitus is very frequent and occurs in up to 40%. Simple obesity may be associated with several adverse metabolic effects. Especially in obese patients with an abdominal body fat distribution and an increase of visceral fat several metabolic disturbances can be observed such as

GH, Body Fat and Metabolism

95

Growth hormone

Indirect growth-promoting effects

Direct metabolic effects

Liver and other organs Insulin-like growth factors

Bone

Chondrogenesis↑ Bone growth↑

Adipose tissue

Glucose metabolism

Lipolysis↑ Lipogenesis↓

Blood glucose↑ Insulin sensitivity↓

Other organs

Protein synthesis↑ Cell proliferation↑

Fig. 1. Direct and indirect biological effects of GH.

dyslipoproteinemia, hypotriglyceridemia, hyperinsulinemia and type 2 diabetes. These metabolic changes carry a high risk of cardiovascular disease. Dyslipoproteinemia with an increase in LDL cholesterol is a main feature of the metabolic syndrome where the amount of free fatty acids in the portal vein is a central causal factor for the underlying pathophysiology. In patients with PWS there is a nonspecific form of this dyslipoproteinemia with hypercholesterinemia and sometimes hypertriglyceridemia depending on the severity of obesity. These changes are directly linked to obesity and high dietary fat intake. However, it could also be shown that GH is able to directly stimulate the activity of LDL receptors in the liver and might, therefore, have a direct effect on lipid metabolism. The lack of GH in PWS and simple obesity, therefore, may contribute to the increased LDL cholesterol levels seen in both conditions. It is also interesting to see that GH plays a direct role in the improvement of lipid metabolism independent of the concomitant fat loss as has been shown in adults with GH deficiency [8].

Biological Effects of GH in Man

GH has various kinds of effects in the human body (fig. 1). They all aim at creating an anabolic situation and support growth. The indirect effects of GH are the growth-promoting effects. The indirect growth-promoting effects of GH

Wabitsch

96

are mediated by insulin-like growth factors (IGFs). GH stimulates their production in liver and in other organs. IGFs show an endocrine and a paracrine action in different organs. In bone, IGFs stimulate the proliferation of chondrocytes in growth zones and promote bone growth. In addition, the effects of GH on bone include an increase of mineral density. GH also stimulates the remodelling process in bone. The effects of IGFs in other organs consist of a stimulation of protein synthesis and cell proliferation. In this way, GH is able to increase muscle protein synthesis and growth. The indirect growth-promoting effects of GH aim at increasing lean body mass. Besides the growth-promoting effects GH has direct metabolic effects mediated by the GH receptor. These effects can be demonstrated in many different cells of the body. They can be readily observed in white adipocytes. The metabolic effects of GH contribute to the complex regulation of glucose and lipid metabolism. GH leads to an increase in blood glucose and a decrease in insulin sensitivity. As mentioned above GH is able to influence LDL receptor activity in the liver. One part of the alterations in lipid metabolism seen in GH-deficient individuals, however, is indirectly associated with the lack of the hormone and is due to the increase in visceral adipose tissue mass. The metabolic effects of GH in children and adolescents in vivo have recently been summarized [9].

Adipose Tissue as a Target Tissue for the Effects of GH

The clinical observations in patients with GH deficiency, in patients with PWS and also the observations obtained when patients with simple obesity are treated with GH indicate that adipose tissue is a target organ for the action of GH. GH deficiency in children is associated with a mild increase of fat mass. At the cellular level the fat cells of GH-deficient children are enlarged but present in reduced numbers. After initiation of GH substitution, both size and number of fat cells are shifted towards normal [10]. GH induces two different effects in adipose tissue: a lipolytic one, resulting in a reduction of fat cell volume and a growth-promoting one increasing the number of fat cells. Figure 2 is a synopsis of the various effects of GH seen in preadipocytes and adipocytes. The biochemical pathways of most of the cellular responses of fat cells to GH are not fully understood and some findings even seem to be contradictory. As there are now appropriate biological systems available that exhibit well-defined responses to GH, it will be a challenge for future research to clarify the mechanisms that are responsible for the growthpromoting and metabolic effects of GH in human adipose tissue [11].

GH, Body Fat and Metabolism

97

Human preadipocyte

Human adipocyte

Increase in IGF-1 production Stimulation of cell proliferation Inhibition of differentiation

Acute effects Increase in glucose transport Increase in lipogenesis Inhibition of lipolysis Chronic effects Decrease in glucose transport Decrease in lipogenesis Increase of lipolysis Refractoriness to acute insulin-like effects

Fig. 2. Acute effect of GH on glucose uptake in human fat cells. In vitro differentiated human fat cells [13] cultured in serum-free, chemically defined medium were preincubated without insulin for 2 h. Thereafter, uptake of radioactively labelled 2-deoxy-D-glucose was measured in the presence of human GH.

As has been shown in cultures of human adipocyte precursor cells in a serum-free chemically defined medium, GH is able to stimulate IGF-1 production and secretion in the cells and thus is able to stimulate proliferation of the undifferentiated precursor cells [12, 13]. During differentiation GH has an inhibitory effect on adipocyte gene expression. This effect seems to be closely related to the decreasing glucose transport and lipogenesis induced by this hormone in differentiating cells. There are two different metabolic effects of GH in this in vitro system. Addition of GH for 1–3 h to adipocytes that have been preincubated without GH results in a transient stimulation of glucose uptake and lipogenesis and an inhibition of lipolysis (fig. 3). The physiological relevance of this insulin-like effect of GH is unclear since it is not clearly correlated with clinical observations. However, the pulsatile secretion of GH which induces constant periods without detectable GH concentrations in serum could suggest that insulin-like effects of GH may also occur in vivo. The chronic metabolic effects of GH comprise a reduction in glucose uptake, glucose oxidation and lipogenesis and the stimulation of lipolysis leading to an increased release of glycerol from adipocytes [11]. In 3T3-F442A cells it has been shown that GH reduces the number of GLUT1 transporters and GLUT1-mRNA with no change in GLUT4 protein or mRNA levels [11]. The chronic metabolic effects of GH generally lead to a reduction of body fat by

Wabitsch

98

200

3-[3H]-2-deoxy-D-glucose uptake (cpm/dish)

hGH (100ng/ml)

100

0 0

30

60

90

120

180

240

360

Incubation time (min)

Fig. 3. Acute in vitro ‘insulin-like’ effects of human GH on differentiated human adipocytes.

Table 3. Characteristics of the GH receptor in fat cells in rat and man

Rat Binding sites/cell Kd

10,000 0.16 nM

Human Binding sites/cell Kd

1,000 0.22 nM

decreasing the mean fat cell volume. Another long-term effect of GH administration in vitro is the induction of refractoriness to the acute metabolic effects reported above. These effects seem to be a postreceptor phenomenon and have not involved a change in either the availability of GH receptors or the affinity for GH. Interestingly the induction of refractoriness can be blocked by inhibiting RNA synthesis. Furthermore, refractoriness can be transferred from refractory cells to sensitive tissue fragments suggesting the release of an unknown factor that induces this state [11]. Recently, we have investigated the different effects of GH in cultured human preadipocytes and adipocytes [13]. We could also show that GH inhibits glucose uptake and lipogenesis in differentiated human adipocytes and leads to a stimulation of lipolysis. The metabolic effects of GH in human adipocytes are weaker than those observed in rat adipocytes. This might be due to the 10 times smaller dissociation constant of the receptor (table 3) [13, 14].

GH, Body Fat and Metabolism

99

Obesity and GH Deficiency in Patients with PWS: Conclusions

With respect to patients with PWS it can be concluded that obesity in PWS is caused by the high caloric intake and the low energy expenditure in these patients. GH deficiency which is due to a hypothalamic dysfunction is an additional factor increasing body fat mass especially at central subcutaneous fat depots due to the lack of the lipolytic activity of the hormone. The baseline metabolic situation in patients with PWS is characterized by metabolic alterations due to GH deficiency and an independent decrease in insulin secretion. Morbid obesity and high fat intake are additional factors leading to further impairment in the lipid and carbohydrate metabolism including diabetes. Very recently, the consequences of GH deficiency in patients with PWS have been extensively summarized by Eiholzer [15]. The long-term benefits of GH therapy in relation to the GH dose used have also been documented in reports by this group [15] and very recently by Carrel et al. [16] in a 4-year follow-up study.

References 1

2 3

4 5

6 7

8

9 10 11

Veldhuis JD, Iranmanesh A, Ho KY, et al: Dual defects in pulsatile growth hormone secretion and clearance subserve the hyposomatotropism of obesity in man. J Clin Endocrinol Metab 1991; 72:51–59. Swaab DF: Das Prader-Willi-Syndrome und der Hypothalamus. PWS-Schriften. Prader-WilliSyndrom-Vereinigung Deutschland, 1997. Goldstone AP, Unmehopa UA, Thomas EL, Brynes AE, Bell JD, Frost G, Ghatei MA, Holland A, Bloom SR, Swaab DF: Hypothalamic neuropeptides and regulation of fat mass in Prader-Willi syndrome; in Eiholzer U, I’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 31–43. Miller L, Angulao L, Price D, Taneja S: MR of the pituitary in patients with Prader-Willi syndrome: Size determination and imaging findings. Pediatr Radiol 1996;26:43–47. Lucidi P, Parlanti N, Piccioni F, Santeusanio F, de Feo P: Short-term treatment with low doses of recombinant human GH stimulates lipolysis in visceral obese men. J Clin Metab Endocrinol 2002;87:3105–3109. Eiholzer U, Blum WF, Molinari L: Body fat determined by skinfold measurements is elevated despite underweight in infants with Parder-Labhart-Willi syndrome. J Pediatr 1999;134:222–225. Eiholzer U, Gisin R, Weinmann C, et al: Treatment with human growth hormone in patients with Prader-Labhart-Willi syndrome reduces body fat and increases muscle mass and physical performance. Eur J Pediatr 1998;157:368–377. Snyder DK, Clemmons DR, Underwood LE: Treatment of obese, diet-restricted subjects with growth hormone for 11 weeks: Effects on anabolism, lipolysis, and body composition. J Clin Endocrinol Metab 1988;67:54–61. Shulman DI: Metabolic effects of growth hormone in the child and adolescent. Curr Opin Pediatr 2002;14:432–436. Wabitsch M, Heinze E: Body fat in GH-deficient children and the effect of treatment. Horm Res 1993;40:5–9. Wabitsch M, Hauner H, Heinze E, Teller W: In-vitro effects of growth hormone in adipose tissue. Acta Paediatr Suppl 1994;406:48–53.

Wabitsch

100

12 13

14

15 16

Wabitsch M, Hauner H, Heinze E, Teller W: The role of GH/IGF’s in adipocyte differentiation. Metabolism 1995;44(suppl):45–49. Wabitsch M, Braun S, Hauner H, Heinze E, Ilondo MM, Shymko R, De Meyts P, Teller WM: Mitogenic and antiadipogenic properties of human growth hormone in human adipocyte precursor cells in primary culture. Pediatr Res 1996;40:450–456. Wabitsch M, Ilondo MM, Heinze E, Hauner H, Shymko RM, De Meyts P: Biological effects of human growth hormone on rat adipocyte precursor cells in primary culture. Metabolism 1994;45:34–42. Eiholzer U: Prader-Willi syndrome. Effects of human growth hormone treatment. Endocr Dev. Basel, Karger, 2001, vol 3. Carrel AL, Myers SE, Whitman BY, Allen DB: Benefits of long-term GH therapy in Prader-Willi syndrome: A 4-year study. J Clin Endocrinol Metab 2002;87:1581–1585.

Martin Wabitsch, MD, Endocrine Research Laboratory, Department of Pediatrics, University Hospital, Prittwitzstrasse 43, D–89075 Ulm (Germany) Tel. ⫹49 731 5002 7790, Fax ⫹49 731 5002 7789, E-Mail [email protected]

GH, Body Fat and Metabolism

101

Comorbidities or Fundamental Defects of Obesity Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 102–118

Glucose Homeostasis in Prader-Willi Syndrome William B. Zipf, Dara Schuster, Kwame Osei Department of Pediatrics, The Ohio State University, Columbus, Ohio, USA

Abstract Our increasing understanding of the obese condition reveals that it is not a single disease or condition but the phenotypic expression resulting from the sum of many different genes as they interact with the environment. The sum of evolutionary pressures has given rise to a redundant system that ensures adequate energy intake and energy balance in a multitude of changing environmental challenges. Unique combinations of normal genes or combinations of abnormal genes can all result in excess weight, particularly in an energy-rich environment. Our frustration in the study of obesity and the difficulty in finding ‘the cause of obesity’ are easily understood as we are likely to find many causes. Each individual with obesity is thus a unique expression of an obese pattern of genes and his/her environment. Thus, the study of obese individuals as a group would be expected to result in diverse observations making it difficult to understand mechanisms. In contrast, PWS is a human genetic form of obesity characterized by a well-defined appetite and metabolic disorder that maintains its special characteristics across a broad band of different genetic backgrounds and environmental conditions. In other words, the obesity of PWS looks nearly the same in all individuals no matter who their parents are, and where or how they live. Thus the PWS condition allows for an opportunity to study an obese condition that is rather pure in its etiology. Central to the question of energy balance is glucose homeostasis. A clear understanding of this aspect of the PWS condition is essential to understanding PWS obesity and is an opportunity to examine the unique genetic abnormality of this central energy process. Our studies of glucose homeostasis in the obese PWS individual reveal striking differences from that seen in the obese non-PWS individual. PWS obesity, in spite of its high fat mass to lean mass ratios, is associated with less insulin resistance than seen in equally obese non-PWS individuals. Further, our studies are consistent with an abnormality in the autonomic nervous system controls of pancreatic insulin secretion as being at least part of the observed difference in PWS obesity versus non-PWS obesity. Copyright © 2003 S. Karger AG, Basel

Introduction

Obesity is a major risk factor for the development of type 2 diabetes mellitus (non-insulin-dependent diabetes mellitus). Together with the genetic background, obesity influences the age of onset and severity of the condition. It appears to be important because of its role in the development of insulin resistance and hyperinsulinemia [1–5]. The association of obesity and insulin resistance is frequently seen in obese individuals even without impaired glucose tolerance or diabetes. Insulin resistance and hyperinsulinemia represent early changes in glucose metabolism that may lead to the future development of type 2 diabetes mellitus or NIDDM [6, 7]. Prader-Willi syndrome (PWS) is a genetic disease that has been associated with morbid obesity and an increased propensity for early development of type 2 diabetes mellitus [8–10]. The true prevalence of the disease in this population is unknown. Previous studies have demonstrated a variable increased prevalence of 7–20% compared with the general population prevalence of 5% [7–9]. Although the etiology of type 2 diabetes mellitus in PWS is unknown, it has been assumed to be related to the morbid obesity found in these patients and the concomitant insulin resistance [9–13]. However, over the past 20 years, evidence has been accumulating that this relationship has both significant and subtle differences in PWS individuals as opposed to non-PWS obese individuals. Understanding these differences may lead to new insights into understanding the relationship of obesity to the development of diabetes.

Glucose Tolerance in PWS with Mixed Meal Stimulation

A careful review and study of the available data suggest that the relationship between the obese condition and the development of diabetes is not entirely clear and may be different than that seen with non-PWS obese individuals. Parra et al. [11] reported hyperinsulinemia following an oral glucose tolerance test (OGTT) in PWS and obese groups compared with normal-weight controls [11]. Likewise, using the intravenous glucose tolerance test (IVGTT), Bier et al. [12] demonstrated similar glucose assimilation coefficient (Kg) values between PWS and obese controls. However, other larger studies with mixed meal nutrients as well as standard OGTT and IVGTT show a different response pattern between obese PWS and obese non-PWS children and adults.

Glucose Homeostasis in PWS

103

Experimental Design, Materials and Methods

For these initial studies [14], 12 patients with PWS and 19 equally obese non-PWS children aged 3.2–14 years were studied. The control group consisted of 19 severely obese children. The two groups were matched for age, weight, sex ratios, and percent ideal body weight for height. As a test for meal-related hormone response, each child had two meal stimulation tests on sequential days. The first test was a fortified protein milk shake calculated to give 0.2 g protein/kg and 4 kcal/kg body weight. The high protein test consisted of administering a meal with 2.0 g of protein and 10 kcal/kg body weight given as the protein fortified milk shake and added hamburger. Blood samples were obtained at 0 time, 15, 30, 45, 90, 120, 180 min from the zero sample. For these studies, glucose was assayed by standard methods using an autoanalyzer with the ferricyanide method. Insulin determinations were made by RIA as described [14]. Pancreatic polypeptide (PP) was assayed using antibody supplied by Lilly Laboratories as described [14]. Responses were analyzed by comparing basal values, peak response, peak response minus basal value (maximum rise), and calculating the area under the curve (AUC) using triangulation.

Results for PP

There were no differences in the basal PP values in the two groups before either the low or high protein meal. The low protein meal was associated with almost no PP release in the PWS group and a significantly greater but still low response in the obese control group. With the high protein meal stimulation test, differences were more marked. The PWS group response pattern was significantly less than that in the obese group whether expressed as peak response (181 ⫾ 51 vs. 580 ⫾ 127 pg/ml), maximum rise above baseline (140 ⫾ 521 vs. 617 ⫾ 127 pg/ml), or response area (25,055 ⫾ 7,113 vs. 67,527 ⫾ 15,027 pg/ml ⭈ min).

Results for Glucose

The basal glucose and glucose responses to the low and high protein meals revealed no significant difference in the mean of the basal glucose concentration between the PWS and obese control groups. However, the peak concentration and the response area in the PWS group were significantly less than those in the obese group after the low protein meal. There was no significant difference between the two groups in the glucose responses after the high protein

Zipf/Schuster/Osei

104

meal. The maximum mean value occurred at 30 min for both tests in both groups. With this study, no child in either group had an abnormally elevated basal or postmeal glucose level with all glucose values for all subjects being less than 131 mg/dl.

Results for Insulin

In contrast to the lack of differences observed for the glucose levels, differences were seen with insulin values and insulin responses. The mean basal insulin values in the PWS group were slightly less than that observed in the control obese group (13.2 ⫾ 0.6 vs. 17.9 ⫾ 1.5 mIU/ml). In addition, the low protein diet elicited a significantly less insulin response in the PWS group as opposed to the obese group for maximum response level (57.4 ⫾ 2.8 vs. 123 ⫾ 14 pg/ml), maximum rise (44 ⫾ 9 vs. 106 ⫾ 13 pg/ml), and AUC (5,143 ⫾ 1,084 vs. 9,1914 ⫾ 231 pg/ml ⭈ min). This was consistent with what we observed with the glucose response patterns. The high protein meal elicited an insulin response in the obese group that was similar to the insulin response in the PWS group. These studies demonstrated that PP deficiency was a unique characteristic of children with PWS. In addition, the studies demonstrated an absence of the usual relationship between the insulin responses in obese PWS patients as compared to insulin responses in nonobese otherwise normal children. Following this initial set of studies, Tomita et al. [15] using a similar experimental design also demonstrated an absence of the expected exaggerated rise in insulin with mixed meal stimulation tests in adults with PWS as compared to normal and obese but otherwise normal adults.

Follow-Up Studies

Because these results were in contrast to those reported by Parra et al. [11] and Bier et al. [12] who had studied glucose homeostasis with OGTT and IVGTT, and to follow up on this initial observation, a series of more detailed studies in both children and adults with PWS was performed using these methods [16]. Further, we elected to also study various components of glucose homeostasis in greater detail to better unmask potential causes for differences between these groups. Peripheral insulin levels are determined not only by beta cell secretion but also by hepatic clearance of insulin. Therefore, the liver also plays a major potential role in the development of peripheral hyperinsulinemia. Insulin and

Glucose Homeostasis in PWS

105

C-peptide are secreted in equimolar amounts, but only insulin is metabolized by the liver. Thus, the molar ratios of C-peptide and insulin have been used as a noninvasive means of assessing hepatic insulin clearance (IC) in this and other studies [17–20]. In this regard, abnormalities in hepatic insulin extraction (HIE) and IC are seen in various states of hyperinsulinemia and insulin resistance [21]. Apart from the potential alterations in insulin metabolism, the differences in insulin secretion and action previously reported in nondiabetic PWS could be partly ascribed to incomplete or nonthorough evaluation of beta cell function in PWS. Therefore, we used two independent studies to examine glucose homeostasis. First, we performed the OGTT to examine changes in glucose, insulin, and C-peptide response to oral glucose. Second, because previous studies have not addressed the phases of insulin release, we determined the phases of insulin secretion using the IVGTT. This issue is important because an absent or blunted first-phase and/or a decrease in second-phase insulin secretion have been found to be early predictors of impaired glucose tolerance in both insulin-dependent diabetes mellitus and type 2 diabetes mellitus [17–21]. Based on this background, the objectives of these new studies were: (1) to examine glucose, insulin, and C-peptide response to OGTT and IVGTT in adult PWS subjects, (2) to characterize the phases of insulin release, (3) to determine the contribution of hepatic IC to peripheral insulin concentration, and (4) to evaluate the age dependency of the insulin response to an oral glucose challenge. These results in PWS were compared with data from age-, sex-, and body mass index (BMI)-matched non-PWS obese controls.

Subjects and Methods

Subjects The study subjects comprised three groups. Group 1 consisted of 9 pediatric PWS subjects under the age of 13 years and 22 age-, sex-, weight-, and puberty stage-matched obese subjects who were studied using the OGTT. All pediatric participants were at Tanner stage 2–3 for sexual maturation. Group 2 consisted of 14 adult PWS subjects and 10 age-, weight-, and BMI-matched obese adults who underwent an OGTT. Group 3 involved 9 adult PWS subjects and 8 age-, sex-, and weight-matched obese adults who underwent an IVGTT. All participants were nondiabetic as defined by National Diabetes Data Group criteria [22]. All PWS subjects met the established diagnostic criteria for PWS [23]. All participants had normal cardiac, thyroid, hepatic, and renal function as determined by a thorough history and physical examination and supportive laboratory data where necessary. All non-PWS subjects had stable body weight

Zipf/Schuster/Osei

106

Table 1. Group 1: clinical characteristics of children with normal weight, non-PWS obesity and with PWS undergoing an OGTT Characteristics

Normal control (n ⫽ 19)

Obese control (n ⫽ 22)

PWS subjects (n ⫽ 9)

Age, years Gender (M/F) Height, m Weight, kg BMI Tanner stage

14 ⫾ 0.3 7/12 1.6 ⫾ 0.02 56 ⫾ 2.1 21 ⫾ 0.6 3–4

10 ⫾ 0.5 14/22 1.5 ⫾ 0.03 79 ⫾ 4.8 35.1 ⫾ 1.5 2

11.5 ⫾ 1.2 3/6 1.4 ⫾ 0.05* 72 ⫾ 8.5 35.5 ⫾ 2.3 2–3

*p ⬍ 0.05 Table 2. Group 2: clinical characteristics of adult PWS and healthy obese subjects undergoing an OGTT Characteristics

Normal control (n ⫽ 11)

Obese control (n ⫽ 10)

PWS subjects (n ⫽ 14)

Age, years Gender (M/F) Height, m Weight, kg BMI Tanner stage

33 ⫾ 2.9 1/10 1.64 ⫾ 0.03 61 ⫾ 3.7 22 ⫾ 0.7 5

33 ⫾ 2.5 0/10 1.6 ⫾ 0.02 101 ⫾ 9.0 39.0 ⫾ 3.0 5

33 ⫾ 2.9 5/9 1.48 ⫾ 0.01* 92 ⫾ 6.5 42 ⫾ 2.7 4–5

*p ⬍ 0.05

during the 3–6 months before the study. All PWS subjects were weight-gaining or weight-stable. Clinical characteristics are shown in tables 1–3. An agematched normal-weight group was included in the tables and figures as references. No statistical comparisons were made between this normal-weight group and the two obese groups. A signed informed consent form approved by the institutional review board was obtained from each subject after the protocol was thoroughly explained. Study Design The subjects reported to the Clinical Research Center of the Ohio State University hospital at 08.00 the day of the study. All subjects ingested their usual diet for 3 consecutive days before the study day. They refrained from

Glucose Homeostasis in PWS

107

Table 3. Group 3: clinical characteristics of adult PWS and non-PWS obese subjects undergoing frequently sampled IVGTT Characteristics

Normal control (n ⫽ 11)

Obese control (n ⫽ 8)

PWS subjects (n ⫽ 9)

Age, years Gender (M/F) Height, m Weight, kg BMI Lean body mass, % Body fat, %

32 ⫾ 2.8 7/4 1.7 ⫾ 0.02 71 ⫾ 2.4 25 ⫾ 0.6 72 ⫾ 1.0 28 ⫾ 1.9

31 ⫾ 2.5 1/7 1.6 ⫾ 0.02 80.2 ⫾ 3.4 30.7 ⫾ 1.4 61.3 ⫾ 2.3 38.7 ⫾ 2.3

25 ⫾ 1.9 2/7 1.47 ⫾ 0.03* 98 ⫾ 9.2 45.5 ⫾ 4.5 56 ⫾ 3.0 44 ⫾ 3.0

*p ⬍ 0.05

strenuous exercise 48 h before the study day. The studies were performed after a 10- to 12-hour overnight fast, with subjects in the supine position. Anthropometric measurements were performed on each of the study subjects. These included body weight, height, and BMI. Fat-free mass was measured by a bioelectrical impedance analyzer technique [24]. Obesity was defined as a BMI greater than 27 kg/m2. OGTT The standard OGTT was performed using 1.75 mg/kg (maximum, 75 g) oral glucose. Biochemical measurements included serum glucose and insulin. Criteria for normal glucose tolerance were determined by the criteria previously established by the NIDDM group [22]. IVGTT The IVGTT was performed as previously described [25]. On the morning of the study day, two angiocatheters were placed in the antecubital veins, one for administration of glucose and insulin and the other for blood sampling from the contralateral arm. Baseline glucose, insulin, and C-peptide levels were obtained at ⫺20, ⫺5 and 0 min. At time 0 min, intravenous glucose 0.3 g/kg (50% dextrose water) was administered over a 1-min period. Both first- and second-phase insulin responses were studied. First-phase insulin secretion was defined as insulin secretion between time 0 and 5 min, and second-phase secretion as between time 8 and 19 min. Blood samples for assay of glucose, insulin, and C-peptides were drawn at frequent intervals during the 20-min period. The blood samples were centrifuged at 4°C and sera were stored at ⫺2°C until assayed.

Zipf/Schuster/Osei

108

Chemical Analyses Serum glucose concentrations were measured by the glucose oxidase method using a glucose analyzer (Beckman Instruments, Fullerton, Calif., USA). Serum immunoreactive insulin and C-peptide levels were measured by a doubleantibody radioimmunoassay technique [26]. The sensitivity of the insulin assay was 2 mU/l serum; intra-assay and interassay coefficients of variation (CVs) were 5 and 10%, respectively. The sensitivity of the C-peptide assay was 0.16 nmol/l serum, and intra-assay and interassay CVs were 6 and 13%, respectively. There were no significant titers of insulin antibodies in any of our subjects.

Calculations and Statistical Analysis

Results are expressed as the mean ⫾ SEM. The AUC were calculated by the trapezoidal rule. The rate of glucose disposal (Kg) was calculated from the natural log transformation of glucose levels from time 8 to 19 min during the IVGTT using linear regression analysis. Basal HIE was calculated as the molar ratio of steady-state C-peptide and insulin concentration [27]. In the basal (steady) state, the kinetics (production/ clearance) of insulin and C-peptide are in equilibrium and differences in the halflife values of the two hormones should not influence the estimation of basal HIE and IC. Note that the molar ratios of C-peptide and insulin at steady state reflect HIE and IC in healthy individuals without renal dysfunction. During the IVGTT, the differences in kinetics on insulin and C-peptide would play a role in the determination of HIE in this non-steady state. Therefore, HIE and IC were calculated as the molar ratios of the incremental integrated areas for C-peptide and insulin [18, 28] from time 0 to 180 min. This time frame includes the return to a new basal (steady) state. The incremental integrated molar ratios of insulin and C-peptide are independent of the differences in the half-life values for the two peptides. Although Polonsky and Rubenstein [21] have described the pitfalls and limitations in the use of molar ratios as a reflection of HIE during the nonsteady state, the use of incremental integrated areas appears to be valid [28]. Statistical comparison between groups was made by the unpaired t test and ANOVA where appropriate. Results were considered statistically significant at p values less than 0.05.

Results

Subjects in group 1 included 22 obese and 9 PWS children matched for age, weight and puberty stage. Weight tended to be greater in the obese groups, but

Glucose Homeostasis in PWS

109

because these subjects were also significantly taller than the PWS group, the BMIs were closely matched (35.1 ⫾ 1.5 vs. 35.5 ⫾ 2.3 kg/m2). Both groups were matched for puberty (as clinically defined by Tanner staging) to eliminate the insulin resistance of puberty as a contributing factor to the insulin resistance. Mean systolic (117 ⫾ 4.0 vs. 122 ⫾ 3.0 mm Hg) and mean diastolic (69 ⫾ 4.4 vs. 77 ⫾ 1.5 mm Hg) blood pressures were lower in the pediatric PWS group compared with the obese group, although this did not reach statistical significance. Subjects in group 2 were adult PWS subjects and obese controls. In this group, neither weight (92 ⫾ 6.5 vs. 101 ⫾ 9.0 kg) nor BMI (42 ⫾ 2.7 vs. 39 ⫾ 3.0 kg/m2) were significantly different in PWS subjects compared with obese control subjects, respectively. Subjects in group 3 were matched for age and weight. The PWS group was significantly shorter than the obese controls (1.47 ⫾ 0.03 vs. 1.6 ⫾ 0.02 m, p ⬍ 0.001). There was no difference in weight (98 ⫾ 9.0 vs. 80 ⫾ 3.0 kg), although BMI was significantly higher in PWS subjects compared with obese controls (46 ⫾ 4.3 vs. 31 ⫾ 1.4, p ⬍ 0.01). Lean body mass (56 ⫾ 3 vs. 61 ⫾ 2%) tended to be lower and body fat (44 ⫾ 3 vs. 39 ⫾ 2.3%) tended to be higher in PWS subjects and obese controls, respectively. OGTT in Group 1 (Pediatric Studies) During the OGTT in the pediatric group, fasting (86 ⫾ 2 vs. 89 ⫾ 2 mg/dl), peak (144 ⫾ 11 vs. 147 ⫾ 4 mg/dl), and total glucose as measured by the AUC of glucose responses (6,984 ⫾ 1,320 vs. 6,963 ⫾ 615 mg/dl ⭈ min) were not significantly different in PWS versus obese children, respectively. In contrast, fasting (20 ⫾ 6 vs. 37 ⫾ 4 mU/ml), peak (114 ⫾ 24 vs. 214 ⫾ 23 mU/ml), and AUC insulin levels (12,673 ⫾ 2,176 vs. 26,734 ⫾ 2,608 mU/ml ⭈ min) and insulin (10,664 ⫾ 1,955 vs. 11,523 ⫾ 1,564 mU/ml ⭈ min) were significantly lower in PWS versus obese controls (fig. 1). OGTT in Group 2 (Adult Studies) The OGTT in the adult groups demonstrated no differences in fasting glucose (93.8 ⫾ 9.2 vs. 79.3 ⫾ 2.7 mg/dl) or insulin (16.7 ⫾ 2.8 vs. 13.5 ⫾ 2.5 mU/ml) in PWS and obese adults, respectively. The respective AUC values were similar in PWS and obese groups for glucose (10,664 ⫾ 1,955 vs. 11,623 ⫾ 1,584 mU/ml). ANOVA demonstrated no significant differences in glucose or insulin responses (fig. 2). IVGTT in Group 3 (Adult Studies) During the IVGTT, fasting glucose (78.3 ⫾ 3.5 vs. 78.2 ⫾ 2.0 mg/dl), insulin (6.9 ⫾ 1.1 vs. 9.9 ⫾ 1.4 mU/ml), and C-peptide (2.1 ⫾ 0.7 vs. 2.1 ⫾ 0.3 ng/ml)

Zipf/Schuster/Osei

110

250

180 160

200

140

Insulin (mU/ml)

Glucose (mg/dl)

120 100 80

150

100

60 Nonobese (n⫽ 17) 40

50

PWS (n⫽ 9) Obese child (n ⫽22)

20 0

0 0

30

60 Time (min)

120

180

0

30

60 Time (min)

120

180

Fig. 1. Glucose and insulin levels during OGTT in nonobese, obese and PWS children.

250

180

Nonobese PWS

160

Obese adult 200

140

Insulin (mU/ml)

Glucose (mg/dl)

120 100 80

150

100

60 40

50

20 0

0 0

30

60 90 Time (min)

120

180

0

30

60 90 Time (min)

120

Fig. 2. Glucose and insulin levels during OGTT in nonobese, obese and PWS adults.

Glucose Homeostasis in PWS

111

180

250

12

180 160

10

200

150

100

C-peptide (ng/ml)

Insulin (mU/ml)

Glucose (mg/dl)

140 120 100 80 60 Obese

50

Nonobese

8

6

4

40 2 20

PWS 0

0

0 0 2 4 6

8 10 12 14 16 18 20 Time (min)

0 2 4 6

8 10 12 14 16 18 20 Time (min)

0 2 4 6

8 10 12 14 16 18 20 Time (min)

Fig. 3. Glucose, insulin, and C-peptide levels during IVGTT in nonobese, obese and PWS adults.

levels were not different in PWS and obese adult groups. Despite a similar glucose response curve, significant differences in insulin and C-peptide response to an intravenous glucose load were noted during the IVGTT. The rate of glucose disposal (Kg) tended to be higher in PWS, but differences were not significantly different between the PWS and obese groups (2.08 ⫾ 0.46 vs. 1.87 ⫾ 0.21%/min). Both first-phase (138 ⫾ 42 vs. 454 ⫾ 102 mU/ml ⭈ min) and second-phase (295 ⫾ 66 vs. 1,015 ⫾ 231 mU/ml ⭈ min) insulin release were significantly reduced in PWS subjects (Kg). Similarly, first-phase (8.6 ⫾ 2.3 vs. 21 ⫾ 4.6 ng/dl) and second-phase (47 ⫾ 4.6 vs. 75 ⫾ 14 ng/dl) C-peptide responses were also significantly reduced in PWS subjects. Using the AUC analysis of first- and second-phase glucose and insulin, the insulin-to-glucose ratios (I/G ratios) were determined. PWS subjects demonstrated reduced I/G ratios during both first-phase (0.23 vs. 0.9) and second-phase (0.13 vs. 0.53) insulin response compared with the obese group (fig. 3). HIE and IC – Basal and Poststimulation Mean HIE and IC were 33% higher in the PWS group compared with the obese control group (15.4 ⫾ 1.5 vs. 10.3 ⫾ 1.6). Following the intravenous glucose load, HIE and IC decreased 66% in the PWS group and 77% in the obese group (5.2 ⫾ 0.8 vs. 2.4 ⫾ 0.4). Poststimulation HIE and IC were also

Zipf/Schuster/Osei

112

PWS

PWS 18

16

16

12 10 8 6

Obese

14

* Poststimulation HIE

14

Basal HIE

Obese

18

12 10 8 6

4

4

2

2

0

0

*

Fig. 4. Basal and poststimulation HIE in PWS subjects and obese control subjects. *p ⬍ 0.01.

significantly increased in the PWS group compared with the obese group (p ⬍ 0.01) (fig. 4). Correlation Coefficients Using univariate regression analysis, weight was significantly correlated with basal insulin level in the normal pediatric obese group (r ⫽ 0.59, p ⬍ 0.003) but not in the pediatric PWS group. Neither weight nor BMI demonstrated a significant correlation with basal insulin (p ⬎ 0.05) in the adult PWS or adult obese group.

Discussion

PWS is characterized by marked obesity associated with an abnormal appetite [9], a tendency to gain weight even with normal caloric intake [9], and an increased prevalence of diabetes mellitus [8, 13]. PWS subjects often demonstrate markedly decreased satiety and hyperphagia as a cause of their morbid obesity [8–13]. The hyperphagia of PWS has been thought to be of hypothalamic origin [29]. To provide further insight into glucose homeostasis PWS, we have carefully and systematically examined beta cell function (insulin and C-peptide) and HIE in PWS.

Glucose Homeostasis in PWS

113

Our past and present data in PWS subjects are of interest in several respects. Despite a similar degree of obesity and glucose response, the insulin response to both a mixed meal (a more natural stimulus in the normal situation) and an oral glucose load was significantly lower in the pediatric PWS group compared with the pediatric obese group. Thus, the obese non-PWS pediatric group appeared to be more insulin-resistant than the age- and BMI-matched PWS pediatric group. Stated another way, the PWS group appeared to be ‘protected’ from developing obesity-associated insulin resistance. The responses we observed in the nonPWS obese group as compared to our normal-weight control group and the observation that insulin resistance and type 2 diabetes mellitus are a rising concern for our obese pediatric populations clearly demonstrate that being prepubertal and in this young age group by themselves do not protect or prevent these abnormalities. Thus, the young age and prepubertal status of the PWS subjects do not explain this observation. The adult PWS group demonstrated similar patterns during the OGTT, with intermediate mean glucose and lower insulin response when compared with both nonobese controls, although these differences were no longer statistically significant. This may be due to the smaller group size or to the effects of aging. Consistent for both the pediatric and adult PWS groups, however, was a delayed peak in glucose and insulin during the OGTT when compared with the obese controls. Although the reason is unclear, the delayed peak of glucose and insulin could be attributed to a delay in gastric emptying. An absent PP response as documented by us for PWS is also seen in longstanding diabetes mellitus with gastric neuropathy from damage to vagal input to the stomach and is associated with delayed gastric emptying. To further evaluate glucose metabolic differences and to eliminate the effects of incretins and differences in gastric emptying, adult PWS and obese groups underwent an IVGTT. During the IVGTT, we found that glucose responses in PWS and obese subjects were similar. However, first- and secondphase insulin and C-peptide secretion were significantly lower in response to intravenous glucose in PWS, consistent with the OGTT results. A deficiency in this initial phase of insulin release has been shown to affect glucose homeostasis and cause mild prandial hyperglycemia in type 2 diabetes mellitus patients [30], individuals with impaired glucose tolerance, and individuals at risks for type 2 diabetes mellitus [31, 32]. Because first-phase insulin release is blunted in pre-type 1 and pre-type 2 diabetic patients, we can infer that beta cell dysfunction that can predispose individuals to type 2 diabetes mellitus exists in adult PWS subjects. Serum insulin levels were significantly lower during the OGTT and IVGTT in PWS versus non-PWS subjects with identical glucose levels. Indeed, I/G ratios were reduced in PWS. Furthermore, Kg was slightly greater in PWS than in non-PWS subjects.

Zipf/Schuster/Osei

114

The reason for beta cell dysfunction in PWS remains unknown; however, several possibilities exist. One possibility is a decreased vagal parasympathetic efferent tone to the pancreas, an important component of normal insulin secretion [33]. PWS is characterized by a number of findings suggesting an abnormal vagal tone [9]. In this regard, we have previously shown that PP secretion, a marker of autonomic nervous system dysfunction, is markedly blunted in the PWS patient [14, 34]. In this study and our previous study [14], we found evidence of a delay in peak glucose in insulin responses, suggesting a possible delay in gastric emptying, another vagal-mediated event. Peripheral insulin levels are determined by beta cell secretion and/or HIE and IC. In this regard, alterations in HIE and IC have previously been demonstrated in individuals with obesity and a predisposition to diabetes [35–37]. We are unaware of any study that has examined HIE and IC in PWS. Thus the contribution of HIE to peripheral insulin in PWS is unknown. We found that the PWS group had significantly higher HIE and IC during the basal and poststimulation stages compared with the obese group. Although the limitations of the use of molar ratios as a reflection of HIE and IC during non-steady state have been examined by several investigators, including Polonsky and Rubenstein [21] and Polonsky et al. [35], our findings in the non-steady state were consistent with the HIE values during the basal (steady) state. To further validate our findings, we believe the use of two compartment C-peptide kinetics to determine the prehepatic and posthepatic insulin secretion rates and subsequent HIE in PWS subjects will be necessary. With the degree of obesity seen in the PWS group and their increased tendency for type 2 diabetes mellitus, it was unexpected to find a higher HIE in this group compared with both obese and nonobese healthy individuals [36]. In addition, we observed a blunted response of HIE and IC to glucose stimulation in PWS versus obese controls. Normally, HIE decreases with glucose stimulation in human and experimental animals [21, 36–38]. This decrease in HIE during a glucose load has been attributed partly to an incretin effect and the saturation of insulin-binding receptors at the level of the liver. In this study of PWS, we observed the blunted response of HIE to glucose. Although the mechanism is unknown, we believe this may reflect an adaptive mechanism for maintenance of normal hepatic glucose production. Whether differences in the clearance rates of both insulin and C-peptide contributed to the differences in HIE between PWS and non-PWS subjects is uncertain. In general, obesity in non-PWS subjects has been associated with insulin resistance, hyperinsulinemia, and a reduction in insulin sensitivity [1–7, 35]. Several mechanisms for the insulin resistance of obesity have been proposed, including receptor and postreceptor defects. Because morbid obesity is a common component of PWS, we expected an increase in insulin secretion and

Glucose Homeostasis in PWS

115

hyperinsulinemia in the PWS group similar to that seen in the non-PWS obese group, but this was not the case. Furthermore, adult PWS subjects in this study tended to have a greater body weight, BMI, and percent body fat than obese controls. PWS individuals are also characterized by very low muscle tone and decreased spontaneous movement [9, 39] which contributes to decreased insulin sensitivity. In this study, our finding of a dissociation between obesity and insulin/C-peptide responses, as well as HIE an IC, is thus remarkable. The etiology for this dissociation of obesity, insulin resistance, and IC in PWS is unclear. Previous studies have demonstrated an inverse relationship between insulin resistance/hyperinsulinemia and upper body fat distribution [27, 38]. The peripheral hyperinsulinemia in upper-body obesity has been ascribed to both beta cell hypersecretion and decreased HIE [27, 38]. Previous studies have found that fat distribution in PWS is typically peripheral [9]. Our pilot study in the adult PWS group demonstrated a waist-to-hip ratio of 0.86, consistent with moderate upper-body obesity. Therefore, moderate upper-body fat distribution alone found in our PWS subjects was not associated with hyperinsulinemia, insulin resistance, or reduced HIE and IC in PWS. The mechanism of the dissociation in PWS remains to be elucidated, possibly by using accurate imaging modalities for visceral adiposity measurements. In summary, we have characterized glucose homeostasis in PWS subjects. Our findings demonstrate that nondiabetic PWS subjects manifest (1) a reduced beta cell response to glucose stimulation, (2) a significantly increased HIE when compared with obese controls, and (3) an association between obesity and beta cell function and IC, in contrast to normal obese subjects. We conclude that glucoregulatory mechanisms are different in obese PWS versus non-PWS subjects.

References 1

2

3 4 5 6

7

Rizza R, Mandarino L, Gerich J: Mechanism of insulin resistance in man. Assessment using the insulin dose-response curve in conjunction with insulin-receptor binding. Am J Med 1981;70: 169–175. Doeden B, Rizza R: Use of a variable insulin infusion to assess insulin action in obesity: Defects in both the kinetics and amplitude of response. J Clin Endocrinol Metab 1987;64: 902–908. Reaven G, Brand R, Chen Y, et al: Insulin resistance and insulin secretion are determinants of oral glucose tolerance in normal individuals. Diabetes 1993;42:1324–1332. Beard J, Ward W, Halter J, et al: Relationship of islet function to insulin action in human obesity. J Clin Endocrinol Metab 1987;65:59–64. Shim ML, Geffner ME: Insulin resistence in children. The Endocrinologist 1999;9:270–276. Lillioja S, Nyomba B, Saad M, et al: Exaggerated early insulin release and insulin resistance in a diabetes-prone population: A metabolic comparison of Pima Indians and Caucasians. J Clin Endocrinol Metab 1991;73:866–876. Martin B, Warram J, Krolewshi A, et al: Role of glucose and insulin resistance in development of type 2 diabetes mellitus: Results of a 25 year follow-up study. Lancet 1992;340:925–929.

Zipf/Schuster/Osei

116

8 9 10 11 12 13 14

15 16 17 18 19 20 21 22 23 24 25

26 27 28 29 30 31 32 33 34

Donaldson MDC, Chu CE, Cooke A, et al: The Prader-Willi syndrome. Arch Dis Child 1994; 70:58–63. Cassidy S: Prader-Willi syndrome. Curr Probl Pediartr 1984;14:5–55. Nozaki Y, Katoh K: Endocrinological abnormalities in Prader-Willi syndrome. Acta Paediatr Jpn 1981;23:301–306. Parra A, Cervantes C, Schultz R: Immunoreactive insulin and growth hormone responses in patients with Prader-Willi syndrome. J Pediatr 1983;83:587–593. Bier DM, Kaplan SL, Havel RJ: The Prader-Willi syndrome: Regulation of fat transport. Diabetes 1977;26:874–881. Hall B, Smith D: Prader-Willi syndrome. J Pediatr 1972;81:286–293. Zipf WB, Odorisio TM, Cataland S, et al: Pancreatic polypeptide responses to protein meal challenges in obese but otherwise normal children and obese children with Prader-Willi syndrome. J Clin Endocrinol Metab 1983;57:1074–1080. Tomita T, Greeley G, Watt L, et al: Protein meal stimulated pancreatic polypeptide secretion in Prader-Willi syndrome of adults. Pancreas 1989;4:395–400. Schuster DP, Osei K, Zipf WB: Characterization of alterations in glucose and insulin metabolism in Prader-Willi subjects. Metabolism 1996;45:1514–1520. Shapiro ET, Tillil H, Miller MA, et al: Insulin secretion and clearance: Comparison after oral and intravenous glucose. Diabetes 1987;36:1365–1371. Shuster LT, Go VLW, Rizza RA, et al: Incretin effect is due to increased secretion and decreased clearance of insulin in normal humans. Diabetes 1988;37:200–203. Nauck MA, Hornberger E, Siegel EG, et al: Incretin effects of increasing glucose loads in man calculated from venous insulin and C-peptide response. J Clin Endocrinol Metab 1986;63:492–498. Katz AL, Rubenstein AH: Metabolism of pro-insulin, insulin and C-peptide in rats. J Clin Invest 1983;52:1133–1121. Polonsky KS, Rubenstein AH: C-peptide as a measure of the secretion and hepatic extraction of insulin. Diabetes 1984;33:486–494. National Diabetes Criteria Data Groups: Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 1979;28:1–57. Holm VA, Cassidy SB, Butler MG, et al: Prader-Willi syndrome: Consensus diagnostic criteria. Pediatrics 1993;91:398–402. Segal KR, Loan MC, Fitzgerald PL, et al: Lean body mass estimation by bioelectrical impedance analysis: A four-site cross-validation study. Am J Clin Nutr 1988;47:4–17. Ader M, Pacini G, Yang YJ, et al: Importance of glucose per se to intravenous glucose tolerance. Comparison of the minimal model prediction with direct measurements. Diabetes 1985;34: 1092–1103. Kuzuya H, Blix PM, Horwitz DI, et al: Determination of free and total insulin and C-peptide in insulin-treated diabetics. Diabetes 1977;26:2226. Osei K, Cottrell DA, Orabella MM: Insulin sensitivity, glucose effectiveness and body fat distribution pattern in non-diabetic offspring of patients with NIDDM. Diabetes Care 1991;14:890–896. Radziuk J, Morishima T: Assessment of insulin kinetics in vivo; in Lanner J, Pohl SL (eds): Methods in Diabetes Research. New York, Wiley, 1986, pp 91–106. Fukushima M, Tokunaga K, Lupien J, et al: Dynamic and static phases of obesity following lesions of PVN and VMH. Part II. Am J Physiol 1987;253:523–529. Bruce DG, Chisholm DJ, Storlien LH, et al: Physiological importance of deficiency in early prandial insulin secretion in non-insulin-dependent diabetes. Diabetes 1988;37:736–744. Colwell JA, Lein A: Diminished insulin response to hyperglycemia in prediabetes and diabetes. Diabetes 1967;16:560–565. Luzi L, DeFronzo RA: Effect of loss of first-phase insulin secretion on hepatic glucose production and tissue glucose disposal in humans. Am J Physiol 1989;257:E241–E246. Berthoud H-R, Powley TL: Morphology and distribution of efferent vagal innervations of rat pancreas as revealed with anterograde transport of Dil. Brain Res 1991;553:336–341. Schwartz TW: Pancreatic polypeptides: A hormone under vagal control. Gastroenterology 1983; 85:1411–1419.

Glucose Homeostasis in PWS

117

35 36 37

38 39

Polonsky KS, Given BD, Hirsch L, et al: Quantitative study of insulin secretion and clearance in normal and obese subjects. J Clin Invest 1988;81:435–441. Osei K, Holland GC: Altered C-peptide/insulin molar ratios and glucose turnover rates after stimulation in non-diabetic offspring of type II diabetic patients. Metabolism 1987;36:122–127. Cozzolino D, Sessa G, Salvatore T, et al: Hyperinsulinemia in offspring of non-insulin-dependent diabetes mellitus patients: The role played by abnormal clearance of insulin. Metabolism 1995; 44:1278–1282. Freedman DS, Srinvasan SR, Burke GL, et al: Relation of body fat distribution to hyperinsulinemia in children and adolescents: The Bogalusa Heart Study. J Clin Nutr 1987;46:403–410. Schoeller DA, Levitsky LL, Bandini LG, Dietz WW, Walczak A: Energy expenditure and body composition in Prader-Willi syndrome. Metabolism 1988;37:115–120.

Prof. Dr. William Zipf, EndoDiab SVCS, 6353 #120 Presidential Gateway, Central Ohio Ped, Columbus, OH 43231-7691 (USA) Tel. ⫹1 614 839 3040, Fax ⫹1 614 839 3041, E-Mail [email protected]

Zipf/Schuster/Osei

118

Comorbidities or Fundamental Defects of Obesity Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 119–127

Sleep-Disordered Breathing in Children with Common Obesity H. Trang Service de Physiologie, Hôpital Robert Debré, Paris, France

Abstract Sleep is a physiological state during which the respiratory system (both respiratory mechanics and respiratory control) encounters challenging conditions. Respiratory disorders during sleep are of particular importance during childhood. This paper focuses on the clinical aspects of sleep-disordered breathing in children with common obesity and provides some hypotheses for the underlying mechanisms. Aspects regarding syndromic obesity are beyond the scope of this review. Copyright © 2003 S. Karger AG, Basel

Introduction

In the general population, about 8–12% of children snore, and sleepdisordered breathing (SDB) occurs in 0.7–2% of children based on epidemiological studies from Europe [1–3], USA [4, 5] and Asia [6]. SDB occurs equally in boys and girls at the peak age between 2 and 8 years. The leading cause is hypertrophy of the tonsils and adenoids. Unrecognized or untreated SDB may result in potential cardiovascular complications and cognitive and intellectual deficits [4]. The present paper focuses on SDB in a particular pediatric population, the children with common obesity, and provides hypotheses for mechanisms underlying SDB in these patients. Aspects regarding syndromic obesity are beyond the scope of this review.

Table 1. Main characteristics of the obese children Authors

Country

Mallory et al. [8] Silvestri et al. [9] Marcus et al. [10] Chay et al. [11] Trang et al. [12]

USA

41

USA

32

USA

22

Singapore France

Number of patients

146 38

Age, years

BMI, kg/m2

Abnormal polysomnography, %

10 ⫾ 4 (3–20) 9⫾3 (3–14) 10 ⫾ 5 (2–20)

208% of IBW 33 ⫾ 8 (21–56) 33 ⫾ 9

37

13 (4–17)b

39 (30–65)b

59a 45 13 66

Values represent mean with the range in parentheses. IBW ⫽ Ideal body weight. a Nap polysomnography. b Median (range).

SDB in Obese Children

In a large number of studies overweight and obesity in adults have been shown to be highly associated with SDB [7]. In contrast, the early pediatric studies rather focused on feeding difficulties and failure to thrive observed in up to 50% of children with SDB in some series [see references in 4]. Only recently have epidemiological data identified obesity as an independent predisposing factor for SDB in children and adolescents aged 2–18 years. Obesity [defined as body mass index (BMI) ⬎28 kg/m2] increases the risk of SDB by 4.59. The neck circumference is significantly larger in children with SDB than in those without it [5]. In comparison to obese adults, the data for obese children are much less numerous. Only five studies have evaluated SDB in obese children using polysomnography [8–12]. Demographic and main characteristics of the study populations are given in table 1. Most children had morbid obesity defined as a BMI ⬎30 kg/m2 or weight ⬎180% of the ideal weight [8–10, 12]. Clinical symptoms suggestive of SDB in obese children are not different from those in the nonobese children, but are found at a very high prevalence. They include nocturnal snoring, labored breathing and apneas, nocturnal sweating, restless sleep with frequent awakenings during the night, mouth breathing and daytime fatigue. It is common that clinical symptoms are denied or underestimated by children and their families. Scores calculated from sleep questionnaires are not predictive for the presence of SDB, neither do they correlate with its severity.

Trang

120

BA, girl, 15 years, BMI 51 C4A2 C3A1 O2A2 O1A1 REOG LEOG ECG NBF NF THO

(@)

(@)

(@)

(@)

ABD 100

RAD EtCO2

100 100 100 100 100 99

99

100 100 99

99

100 100 100 99

98

98

100 100

43 43 43 43 43 44 44 44 44 45 45 45 44 44 43 42 42 42 42 42 42 42 44 44 43 38 38 38 43 43 42 41 41 41 44 44 43 42 42 42 42 42 43 43 43 41 41 41 41 45 41 43 43 43 43 43 43 43 44 44

CO2g 10"

20"

30"

40"

50"

60"

Fig. 1. Polysomnographic recordings showing repetitive obstructive apneas during rapid eye movement sleep associated with normal oxygen saturation and end-tidal PCO2, but frequent EEG arousals. From top to bottom: C4A2/C3A1/O2A2/O1A1, electroencephalogram, REOG, LEOG (electrooculograms), ECG (electrocardiogram), NBF (nasobuccal flow), NF (nasal flow), THO, ABD (respiratory movements of the thorax and abdomen), RAD (oxygen saturation), EtCO2, end-tidal CO2.

Polysomnography is the gold standard examination for determining the presence and the severity of SDB [13]. The main findings in obese children are obstructive apneas or hypopneas that occur predominantly during rapid eye movement sleep and may induce hypoventilation and sleep fragmentation (fig. 1, 2). A flow limitation breathing pattern may be caused by partial upper airway obstruction and is associated with the recruitment of accessory respiratory muscles. Respiratory efforts are evidenced by increased esophageal pressure whenever an esophageal catheter is placed, or they are the result of an increased amplitude of out-of-phase respiratory movements of the thorax and the abdomen (fig. 3, 4). Repetitive obstructive apneas and hypopneas may occur with severe desaturation and sleep disturbances due to frequent arousals or awakenings throughout the night (fig. 5). Obstructive sleep apneas are found at a high prevalence (ranging from 37 to 59%) in obese children [8–12]. The prevalence of SDB is 66% in one study investigating obstructive apneas and hypopneas [12]. The degree of obesity correlates with the apnea index and inversely correlates with the lowest value of oxygen saturation during sleep [4]. Mean end-tidal PCO2 is found to be higher in children with a body weight higher than 200% of its ideal value than in the others [9]. In most studies, however, no significant correlations exist between

SDB in Childhood Obesity

121

RH, boy, 4 years, BMI 52 C4A2 O2A2 C3A1 REOG LEOG GG EKG RR NBFL THO (@) ABD SaO2

90

90

89

89

86

85

83

84

84

84

62.9

36.3

46.8

60.3

53.7

50.8

53.5

41.1

45.8

45.8

O2G EtCO2 CO2g 5"

10"

15"

20"

25"

30"

Fig. 2. Two obstructive apneas associated with out-of-phase respiratory movements of thorax and abdomen, desaturation, hypercapnia and movement arousals.

BA, girl, 15 years, BMI 51 C4A2 C3A1 O2A2 O1A1 REOG LEOG ECG NBF NF THO ABD 100

RAD EtCO2

100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100

43 42 42 42 43 43 41 40 43 46 46 43 42 40 40 40 40 42 45 43 42 41 41 43 45 37 32 37 41 42 40 40 41 42 42 42 43 42 42 42 43 44 40 38 41 44 44 44 44 45 45 45 46 46 45 45 45 45 45 45

CO2g 10"

20"

30"

40"

50"

60"

Fig. 3. Flow-limitation breathing pattern due to partial upper airway obstruction with frequent electroencephalogram arousals.

polysomnographic variables (respiratory event indices, percent of sleep time with oxygen saturation lower than 90%, percent of sleep time with end-tidal PCO2 higher than 50 mm Hg, arousal indices) and age, or gender, or BMI [8, 12]. Although awake sitting and supine respiratory resistance is increased in most

Trang

122

RH, boy, 4 years, BMI 52 C4A2 O2A2 C3A1 REOG LEOG GG EKG RR NBFL THO ABD SaO2

91

87

85

87

87

86

90

92

93

94

54.8

52.7

54.2

52.7

53.2

54.9

55.3

57.7

57.7

55.3

O2G EtCO2 CO2g 5"

10"

15"

20"

25"

30"

Fig. 4. Flow-limitation breathing pattern, as observed on nasal flow signal (NBFL), associated with desaturation and hypercapnia.

RH, boy, 4 years, BMI 52 SaO2 HR CA OA MA HYPO TRat

Stage

Tonus ACT Delta REM EtCO2 ⫹1h

⫹2h

⫹3h

⫹4h

⫹5h

⫹6h

⫹7h

⫹8h

⫹9h

Fig. 5. Sleep data of one night showing the occurrence of many obstructive apneashypopneas (OA), concomitant desaturation (SaO2), and frequent awakenings during sleep (stage).

obese children, they are predictive neither of the presence nor of the severity of SDB [12]. SDB in obese children may differ in many respects from that in obese adults. Excessive daytime somnolence is a hallmark of obese adults with SDB.

SDB in Childhood Obesity

123

In contrast, sleep latency objectively measured by the multiple sleep latency test is found to be normal in obese children with SDB [10]. The obesityhypoventilation syndrome with daytime hypercapnia is a clinical presentation found in a number of obese adults [14, 15], but has been reported in 2 obese children only [16, 17].

Mechanisms for SDB in the Obese

Most data evaluating the putative mechanisms underlying SDB in the obese derived from animal or adult human studies. Many physiological changes occur in association with the obese status. Whether these changes actually generate SDB or not remains to be established and requires further investigations. (1) Narrowing of the upper airway may be due to anatomic obstruction by enlarged tonsils or adenoids [4], to fat deposit around upper airway [18], to increased pharyngeal collapsibility shown in obese adults [19], and/or due to laryngeal dysfunction-produced expiratory braking during sleep [20]. (2) Alterations in respiratory mechanisms are associated with obesity. Lung volumes are decreased, mainly functional residual capacity and expiratory reserve volume [10, 21]. Subcutaneous fat deposit around the body decreases chest wall compliance, increases intra-abdominal pressure (both factors may limit diaphragmatic excursion), produces shallow tidal breathing and increased work of breathing [22]. However, there is no data showing respiratory muscle fatigue. (3) Abnormal central chemosensitivity has been demonstrated in mutant obese mice [23]. However, data remain conflicting in adult humans, with studies showing either decreased, unchanged or increased ventilatory responses to chemical stimuli. (4) Finally, one may suggest a genetic predisposition for SDB. Candidate genes may include those involved in the regulation of many pathways, metabolism, energy homeostasis, control of arousals and respiration [24–26].

Is Obesity Associated with Narcolepsy?

Case Report Joris was a 15-year-old boy with a BMI of 51 and a weight of 146 kg. He presented with heavy nocturnal snoring, breathing difficulties during sleep, morning headaches, and daytime sleepiness. Overnight polysomnography showed obstructive sleep apneas (index 10 events per hour of sleep) and nocturnal hypoventilation with desaturation (nadir oxygen saturation 85%) and

Trang

124

hypercapnia (maximal end-tidal PCO2 55 mm Hg). Treatment included a diet and nocturnal continuous positive airway pressure. Respiratory nocturnal symptoms resolved, but Joris complained of excessive daytime somnolence. No cataplexy episodes were noted. Multiple sleep latency tests showed sleep onset with rapid eye movement sleep and a sleep latency of 8 min. HLA DR15 was negative. Treatment with modafinil significantly improved daytime vigilance. Question: Narcolepsy or not narcolepsy? An association between obesity and narcolepsy has been suggested based on molecular findings [27, 28]. However, data in humans remain very scarce and further studies are required. Narcoleptic patients have been found to have a higher BMI [29, 30].

Treatment of SDB in Obese Children

Weight loss is the starting point for treatment in obese children and adolescents with obstructive SDB. Nasal nocturnal continuous positive airway pressure is efficient in normalizing gas exchange during sleep when weight loss is impossible or moderate [31]. Tonsillectomy and/or adenoidectomy may be recommended if tonsils and/or adenoids are hypertrophied, but may induce a certain increase in BMI [32].

Conclusion

Obese children are at high risk of SDB. Polysomnography is the key examination for identifying obstructive apneas and/or hypopneas and for rating SDB severity (based on respiratory event index, on the importance of nocturnal hypoventilation and on sleep fragmentation). Treatment includes a diet aimed at losing weight associated with nocturnal continuous positive airway pressure in the case of severe SDB.

References 1 2 3

4

Teculescu DB, Caillier I, Perrin P, Rebstock E, Rauch A: Snoring in French preschool children. Pediatr Pulmonol 1992;13:239–244. Ali NJ, Pitson DJ, Stradling JR: Snoring, sleep disturbance, and behaviour in 4–5 year olds. Arch Dis Childhood 1993;68:360–366. Brunetti L, Rana S, Lospalluti ML, Pietrafesa A, Francavilla R, Fanelli M, Armenio L: Prevalence of obstructive sleep apnea syndrome in a cohort of 1 027 children in southern Italy. Chest 2001; 120:1930–1935. Marcus CL: Sleep-disordered breathing in children. Am J Respir Crit Care Med 2001;164:16–30.

SDB in Childhood Obesity

125

5

6 7 8 9

10 11

12

13 14 15

16 17 18

19

20 21 22 23 24

25 26 27

Redline S, Tishler PV, Schluter M, Aylor J, Clark K, Graham G: Risk factors for sleep-disordered breathing in children. Associations with obesity, race, and respiratory problems. Am J Respir Crit Care Med 1999;159:1527–1532. Anuntaseree W, Rookkapan K, Kuasirikul S, Thongsuksai P: Snoring and obstructive sleep apnea in Thai school-age children: Prevalence and predisposing factors. Pediatr Pulmonol 2001;32:222–227. Young T, Peppard PE, Gottlieb DJ: Epidemiology of obstructive sleep apnea. A population health perspective. Am J Respir Crit Care Med 2002;165:1217–1239. Mallory GB, Fiser DH, Jackson R: Sleep-associated breathing disorders in morbidly obese children and adolescents. J Pediatr 1989;115:892–897. Silvestri JM, Weese-Mayer DE, Bass MT, Kenny AS, Hauptman SA, Pearsall SM: Polysomnography in obese children with a history of sleep-associated breathing disorders. Pediatr Pulmonol 1993;16:124–129. Marcus CL, Curtis S, Koerner CB, Joffe A, Serwint JR, Loughlin GM: Evaluation of pulmonary function and polysomnography in obese children and adolescents. Pediatr Pulmonol 1996;21:176–183. Chay OM, Goh A, Abisheganaden J, Tang J, Lim WH, Chan YH, Wee MK, Johan A, John AB, Cheng HK, Lin M, Chee T, Rajan U, Wang S, Machin D: Obstructive sleep apnea syndrome in obese Singapore children. Pediatr Pulmonol 2000;29:284–290. Trang H, Frelut ML, Navarro J, Gaultier C: Absence of correlation between clinical score, respiratory resistance, and obstructive sleep apnea in children with morbid obesity (abstract). 16th Congress of the European Sleep Research Society, Reykjavik, 2002. American Thoracic Society: Standards and indications for cardiopulmonary sleep studies in children. Am J Respir Crit Care Med 1996;153:866–878. Kessler R, Chaouat A, Schinkewitch P, Faller M, Casel S, Krieger J, Weitzenblum E: The obesityhypoventilation syndrome revisited. Chest 2001;120:369–376. Berger KI, Ayappa I, Chatr-Amontri B, Marfatia A, Sorkin IB, Rapoport DM, Goldring RM: Obesity hypoventilation syndrome as a spectrum of respiratory disturbances during sleep. Chest 2001;120:1231–1238. Riley DJ, Santiago TV, Edelman NH: Complications of obesity-hypoventilation syndrome in childhood. Am J Dis Child 1976;130:671–674. Bourne RA, Maltby CC, Donaldson JD: Obese hypoventilation syndrome of early childhood requiring ventilatory support. Int J Pediatr Otorhinolaryngol 1988;16:61–68. Whittle AT, Marshall I, Mortimore IL, Wraith PK, Sellar RJ, Douglas NJ: Neck soft tissue and fat distribution: Comparison between normal men and women by magnetic resonance imaging. Thorax 1999;54:323–328. Sforza E, Bacon W, Weiss T, Thibault A, Petiau C, Krieger J: Upper airway collapsibility and cephalometric variables in patients with obstructive sleep apnea. Am J Respir Crit Care Med 2000; 161:347–352. Tuck SA, Dort JC, Remmers JE: Braking of expiratory airflow in obese pigs during wakefulness and sleep. Respir Physiol 2001;128:241–245. Inselman LS, Milanese A, Deurloo A: Effect of obesity on pulmonary function in children. Pediatr Pulmonol 1993;16:130–137. Pankow W, Podszus T, Gutheil T, Penzel T, Peter JH, Von Wichert P: Expiratory flow limitation and intrinsic positive end-expiratory pressure in obesity. J Appl Physiol 1998;85:1236–1243. Tankersley C, Kleeberger S, Russ B, Schwartz A, Smith P: Modified control of breathing in genetically obese (ob/ob) mice. J Appl Physiol 1996;81:716–723. Fung SJ, Yamuy J, Sampogna S, Morales FR, Chase MH: Hypocretin (orexin) input to trigeminal and hypoglossal motoneurons in the cat: A double-labeling immunohistochemical study. Brain Res 2001;903:257–262. O’Donnell CP, Tanskerley CG, Polotsky VP, Schwartz AR, Smith PL: Leptin, obesity, and respiratory function. Respir Physiol 2000;119:173–180. Phipps PR, Starritt E, Caterson I, Grunstein RR: Association of serum leptin with hypoventilation in human obesity. Thorax 2002;57:75–76. Hara J, Beuckmann CT, Nambu T, Willie JT, Chemelli RM, Sinton CM, Sugiyama F, Yagami K, Goto K, Yanagisawa M, Sakurai T: Genetic ablation of orexin neurons in mice results in narcolepsy, hypophagia, and obesity. Neuron 2001;30:345–354.

Trang

126

28 29 30 31 32

Nishino S, Ripley B, Overeem S, Lammers GJ, Mignot E: Hypocretin (orexin) deficiency in human narcolepsy. Lancet 2000;355:39–40. Dahmen N, Bierbrauer J, Kasten M: Increased prevalence of obesity in narcoleptic patients and relatives. Eur Arch Psychiatry Clin Neurosci 2001:251:85–89. Schuld A, Hebebrand J, Geller F, Pollmächer T: Increased body-mass index in patients with narcolepsy. Lancet 2000;355:1274–1275. Soultan Z, Wadowski S, Rao M, Kravath RE: Effect of treating obstructive sleep apnea by tonsillectomy and/or adenoidectomy on obesity in children. Arch Pediatr Adolesc Med 1999;153:33–37. Marcus CL, Ward SL, Mallory GB, Rosen CL, Beckerman RC, Weese-Mayer DE, Brouillette RT, Trang H, Brooks LJ: Use of nasal continuous positive airway pressure as treatment of childhood obstructive apnea. J Pediatr 1995;127:88–94.

Ha Trang, MD, PhD, Service de Physiologie, Hôpital Robert Debré, 48 boulevard Serurier, F–75019 Paris (France) Tel. ⫹33 1 40 03 24 76, Fax ⫹33 1 40 03 47 70, E-Mail [email protected]

SDB in Childhood Obesity

127

Comorbidities or Fundamental Defects of Obesity Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 128–139

Dysregulation of Respiration and Sleep in Prader-Willi Syndrome Bernhard Schlüter Vestische Kinderklinik Datteln, University of Witten/Herdecke, Datteln, Germany

Abstract Is respiratory dysregulation in Prader-Willi syndrome (PWS) simply due to obesity or is it caused by underlying pathophysiological deficits? The literature on disturbed sleep and respiration in PWS was reviewed and discussed in comparison with prevalence rates of obesity, hypersomnolence and sleep-related respiratory disturbances in the general population. Cross-sectional observational studies reporting clinical case series of Prader-Willi patients included questionnaire studies without reference groups (n ⫽ 3) and physiological studies with (n ⫽ 10) and without (n ⫽ 11) reference groups. Only one population-based epidemiological study and a few intervention studies were found (n ⫽ 6). In Prader-Willi patients prevalence rates of hypersomnolence and snoring were markedly higher than in the general population. Both, hypersomnolence and snoring, were related to obesity as an important risk factor for sleep apnea in the general population. In contrast, excessive daytime sleepiness in Prader-Willi patients occurred independent of the degree of obesity and was more pronounced than could be expected from the amount of sleep apnea. Disorganization of REM sleep which was detected in Prader-Willi patients independent of the degree of obesity hinted at hypothalamic dysfunction as a pathophysiological mechanism for hypersomnolence. Obesity blunted central chemoreceptor function in Prader-Willi patients but not in controls, whereas peripheral chemoreceptor function was found to be absent in PraderWilli patients independent of the degree of obesity. To date, published data with respect to the dysregulation of respiration in PWS are limited. Disturbances of sleep and breathing in Prader-Willi patients seem to be multifactorial in origin. They are related to hypothalamic dysfunction, which represents the central pathophysiological deficit in PWS. Obesity is thought to be an important factor aggravating the clinical manifestation of respiratory dysfunction. Copyright © 2003 S. Karger AG, Basel

Introduction

This symposium focuses on the Prader-Willi syndrome (PWS) as a model for obesity, which means not merely overweight but an increase of body fat in relation to lean body mass [1]. Excessive or rapid weight gain on weight-for-length charts after 12 months but before 6 years of age resulting in central obesity in the absence of intervention is a major diagnostic criterion of PWS [2]. In the general adult population obesity is a factor known to increase the risk of sleep-related respiratory disturbances, which frequently are accompanied by snoring and obstructive sleep apnea at night as well as hypersomnolence throughout the day as a consequence of sleep fragmentation [3–7]. Excessive daytime sleepiness was not associated with PWS in the early reports [8]. One author [9] noted that somnolence sometimes occurred in association with PWS and one study [10] found that for 24 patients, somnolence was an almost universal accompaniment of the syndrome, especially in the grossly obese patients who slept excessively and often would fall asleep and become cyanosed within minutes of sitting. At that time the term ‘Pickwickian syndrome’ [11] was used in order to outline the hypothesis that excessive fat caused excessive sleep and cyanosis and unless weight loss could be achieved such patients were likely to die from cor pulmonale [10]. Meanwhile, the term ‘Pickwickian syndrome’ has been abandoned, because quite heterogeneous disease entities have been reported under this heading, and not all of them included obstructive apnea [12]. Sleep disturbances or sleep apnea were listed only as a minor diagnostic criterion for PWS [2], although sleepiness (not otherwise defined) was a problem in 231 out of 232 people surveyed in one North American study [13]. It could easily be hypothesized that hypersomnolence in Prader-Willi patients is the consequence of sleep apnea due to obesity. This simple hypothesis is challenged by the fact that obesity is evolving with age in Prader-Willi patients, and that sleep apnea in childhood is not identical with sleep apnea in adults [14]. In this article the published evidence with respect to disturbances of sleep and breathing in the Prader-Willi syndrome is reviewed. It will be shown that obstructive sleep apnea is found only in a proportion of Prader-Willi patients, that there are factors other than obesity which contribute to the manifestation of sleep-related respiratory disturbances in PWS, and that there are factors other than sleep apnea which contribute to the manifestation of excessive daytime sleepiness in Prader-Willi patients [15–24]. Consequences of the therapy for respiratory dysregulation in PWS will be addressed.

Respiration, Sleep and PWS

129

Table 1. Reference values: prevalence of overweight, obesity, habitual snoring, hypersomnolence and sleep-related respiratory disturbances in the general population Parameter

Population

Age years

Prevalence %

Ref. No.

Overweight (BMIⱖ85th percentile)

US 1988–1994

6–12 12–17 6 15

⬎10 ⬎20 22 31

25–27 25–27 25, 28 25, 28

UK 1996 Obesity

US 1988–1994

12–17

8–17

25

(BMIⱖ95th percentile)

UK 1996

6 15

10 17

25, 28 25, 28

Habitual snoring

Germany 1999

1–6

UK 1989–1990 UK 1992 Italy 1990–1991 Netherlands 1997

4–5 6–7 10–15 ⬎35

38.9 (sometimes) 9.1 (always when asleep) 12.1 11.4 5.6 8.9

30 30 32 33 7 31

Germany 1999

1–6

UK 1989–1990 UK 1992 Island 1995

4–5 6–7 15–16

45.6 (sometimes) 3.7 (every day) 20.7 10.2 16 (very often) 6.5 (always)

30 30 32 33 29 29

US 1993

30–60

24 (men) 9 (women)

4 4

AHIⱖ5 plus hypersomnolence

US 1993

30–60

4 (men) 2 (women)

4 4

SDB

UK 1989–1990

4–5

0.7–1.1

32

Hypersomnolence, daytime sleepiness

Sleep-related respiratory disturbances AHIⱖ5

AHI ⫽ Apnea hypopnea index; SDB ⫽ sleep-disordered breathing identified by overnight video and oximetry recordings.

Prevalence of Self-Reported Symptoms in PWS and the General Population

Hypersomnolence and snoring are cardinal symptoms and complaints which arouse suspicion of sleep-related respiratory disturbances. These symptoms can easily be asked about in a questionnaire study. But it is a problem to define the terms hypersomnolence and snoring and to quantify the intensity of these

Schlüter

130

Table 2. Self-reported symptoms of disturbed sleep and respiration in PWS Authors

Laurance et al. [10] Kaplan et al. [20] Fredrickson et al. [41] Bohr [34] Greenswag et al. [13] Helbing-Zwanenburg et al. [23] Clarke et al. [8] Vela-Bueno et al. [24] Lämmer and Desaga [19] Butler et al. [35]

Hyper somnolence

Snoring

n/n

%

n/n

24/24 5/5 4/4 4/4 231/232 20/21

100 100 100 100 99 95 93 89 86 73

56/60 8/9 12/14 47/64

Obesity

Age years

Gender (M/F)

not reported 5/5 100 4/4 100 3/4 75 not reported not reported

yes yes yes yes yes yes

15–41 3–23 3–23 12–19 16–64 12–54

13/11 2/3 1/3 2/2 115/117 13/8

not reported 6/9 67 10/14 71 40/63 63

yes yes yes yes

16–43 3–21 5–37 ⬎17 (n ⫽ 32) ⬍18 (n ⫽ 34) 16–39

29/32 3/6 7/7 20/12

1.7–42

12/13

%

Richards et al. [17]

8/14

57

7/14

50

Cassidy et al. [15]

13/25

52

11/25

44

yes: 9 no: 5 yes

symptoms. Thus, anamnestic data and questionnaire studies may be distorted by a recall bias. In the general population (table 1), an increasing prevalence of overweight and obesity among children and adolescents has been observed [25–28]. Estimates for the prevalence of hypersomnolence ranged from 6.5% (always) to 16% (very often) in adolescents [29] and from 3.7% (every day) to 45.6% (sometimes) in children [30] (table 1). Habitual snoring was estimated to be present in about 9% of adults [31] and in a range of 5.6–38.9% of children [7, 30, 32, 33] (table 1). In the general population prevalence rates of snoring and hypersomnolence were both positively correlated with body mass index (BMI), male sex and age [3–7]. In case series of adult and infant patients with PWS (all obese) the prevalence of hypersomnolence ranged from 52 to 100% and was markedly above the prevalence rates of the general population (table 2). Interestingly, in the case series reported by Bohr [34], Vela-Bueno et al. [24], Lämmer and Desaga [19], Richards et al. [17] and Cassidy et al. [15] the proportion of hypersomnolent patients was higher than the proportion of snoring patients. This is in accordance with the result of the only population-based epidemiological study published thus far with hypersomnolence in 73% and snoring in only 63% of

Respiration, Sleep and PWS

131

20/14 9/5

patients [35]. In Prader-Willi patients snoring was found to be positively correlated to the degree of obesity [35], whereas hypersomnolence was independent of BMI [8, 15, 22, 35]. In summary, subjective measures provide some evidence that hypersomnolence in PWS may not merely be attributed to sleep-disordered breathing.

Objective Measures of Sleep-Disordered Breathing and Daytime Function

Whole-night polysomnography and multiple sleep latency tests (MSLT) are objective methods used to quantify sleep-related respiratory disturbances and daytime function. MSLT criteria for excessive daytime sleepiness are a sleep latency of less than 5 min and/or the occurrence of rapid eye movement (REM) sleep at sleep onset [36]. In healthy men in the age range of 30–50 years a mean value of 8.5 min was found for sleep latency in MSLT [36]. In healthy prepubertal children sleep latencies usually are below 10 min duration [37]. Cessations of airflow of at least 10 s duration in adults [6] and at least 5 s duration in children older than 12 months [38] despite continuing ventilatory effort, usually associated with a decrease of at least 4% in oxyhemoglobin saturation, are classified as obstructive apneas. An apnea-hypopnea index of less than 5/h is regarded as normal [4], apnea-hypopnea index of more than 10/h is a clear indicator of the sleep apnea syndrome [4]. In the general population prevalence rates of relevant sleep-related respiratory disturbances with excessive daytime sleepiness and obstructive sleep apnea ranged from 2 to 4% in female and male adults [4] and 0.7 to 1.1% in children [32–33] (table 1). In case series of PWS excessive daytime sleepiness was found in 33.3–100% of patients (table 3). With respect to the sleep apnea syndrome the results of different studies were divergent. Cassidy et al. [15], Harris and Allen [16], Richards et al. [17], Friedman et al. [18], and Lämmer and Desaga [19] reported a high prevalence (50–100%), whereas Kaplan et al. [20], Hertz et al. [21], Cataletto et al. [22], Helbing-Zwanenburg et al. [23] and Vela-Bueno et al. [24] reported a low prevalence (0–22%) of sleep apnea in obese PWS patients with hypersomnolence (table 3). These divergent results may be due to selection bias. Although the database of polysomnographic recordings in PWS is limited, objective measures provide stronger evidence for the fact that apart from sleep apnea other pathophysiological mechanisms must be present to explain the amount of excessive daytime sleepiness in Prader-Willi patients.

Schlüter

132

Table 3. Objective measures of disturbed sleep and respiration in PWS Authors

EDS1

Sleep apnea syndrome2

Obesity3

Age years

Gender (M/F)

19–40 11–17 16–39

0/2 3/1 9/5

6–24 5–37

6/3 7/7

n/n

%

n/n

%

Cassidy et al. [15] Harris and Allen [16] Richards et al. [17]

2/2 4/4 8/14

100 100 57

2/2 3/4 12/14

100 75 86

Friedman et al.[18] Lämmer and Desaga [19]

not reported 5/14 36

6/9 7/14

67 50

yes yes yes: 9 no: 5 yes yes

Kaplan et al. [20] Hertz et al. [21] Adults Children Cataletto et al. [22] Vela-Bueno et al. [24] Helbing-Zwanenburg et al. [23]

not reported

1/5

20

yes

3–23

2/3

10/14 71 not reported 4/6 67 5/9 56 7/21 33.3

0/15 2/9 1/6 0/9 0/13

0 22 17 0 0

yes no/yes yes yes yes

18–47 2–11 8–40 3–21 12–54

7/8 4/5 6/0 3/6 13/8

Excessive daytime sleepiness: MSLT, sleep latency ⬍5 min and/or 1 REM sleep onset nap. Polysomnography, apnea hypopnea index ⬎10/h. 3 BMI ⬎25 kg/m2. 1 2

Disturbances of Sleep and Wakefulness Patterns in PWS

Several authors described features in the sleep-wake cycle of Prader-Willi patients which were significantly different from those in normal healthy persons, excessive daytime sleepiness together with REM sleep disorders [15, 16, 19–24, 34, 35, 39–41] and shortage of slow wave sleep [23, 24]. Especially in adult patients the REM sleep seemed disorganized, including the occurrence of sleep onset REM [15, 21, 23, 24, 34, 40], REM during daytime naps [16, 23], shortened REM sleep latency [15, 16, 21, 34], increased number of REM periods [21] but reduction in the total amount of REM sleep [23], increased arousal frequency during REM sleep [23], REM associated hypoventilation [24, 41] and decreases of oxyhemoglobin saturation during REM sleep [15, 16, 19, 21, 22, 34, 40]. To date, these phenomena have been documented only in a small number of patients. However, they provide some evidence that a disturbed regulation of the sleep and wakefulness cycle rather than respiratory control is the primary cause of hypersomnolence and excessive daytime sleepiness in PWS.

Respiration, Sleep and PWS

133

In this context several authors discussed the possibility that Prader-Willi patients suffer from a narcoleptic syndrome [23, 39]. Narcolepsy is a neurological disorder characterized by abnormal sleep tendencies, including excessive daytime sleepiness, disturbed nocturnal sleep, and pathological manifestations of REM sleep, including sleep onset REM periods and the dissociated REM sleepinhibitory processes of cataplexy, sleep paralysis and hypnagogic hallucinations [42]. Narcolepsy is associated with the human leukocyte antigen system in more than 95% of all Caucasian patients with cataplexy being positive for HLA subtypes DR2 and DQ1 [42]. Helbing-Zwanenburg et al. [23] examined HLA subtypes in Prader-Willi patients. In this study only 2 out of 8 Prader-Willi subjects with both excessive daytime sleepiness and cataplexy were DR2-positive; thus, no genetic overlap between PWS and narcolepsy was found [23]. Moreover, some authors described a paroxysmal loss of muscle tone resembling cataplectic episodes in Prader-Willi patients [15, 23, 34]. The significance of this phenomenon is very uncertain. It is even more difficult to prove that cataplexy occurs in Prader-Willi patients than it is in narcoleptic subjects, because it is based on clinical facts only, is not mentioned by the subjects and is often ignored by physicians [23]. The only population-based epidemiological study on PWS reported 1 out of 64 patients with a history of narcolepsy and 10 out of 65 patients with ‘atypical seizures’ [35]. Wharton et al. [39] postulated that an altered set point for sleep and arousal modulated by hypothalamic neurosecretory peptides or their receptors exists in individuals with PWS. The posterior, lateral and medial hypothalamus is the cephalad continuation of the brainstem reticular activating system. It is known from animal studies that transection of the brainstem at the level of the posterior hypothalamus completely abolishes wakefulness; posterior hypothalamic lesions generally produce lethargy and sleepiness [43]. Hypersomnolence, coma, and changes in circadian rhythmicity may be noted in humans, depending on the exact site and extent of hypothalamic disease or dysfunction [43].

Special Pathophysiological Aspects with Respect to Respiratory Dysregulation

The Upper Airway Richards et al. [17] investigated the role of anatomical factors in obstructive sleep apnea in the PWS. On clinical assessment, the nasopharynx, oropharynx and hypopharynx were small in only 1 out of 14 subjects, radiological studies showed a slight reduction in the cross-sectional area in 9 out of 14 subjects at the oropharyngeal level and in 4 subjects at the nasopharyngeal level. The authors concluded that sleep apnea and minor radiological evidence of

Schlüter

134

narrowing of the upper airway were common in PWS and that obstructive sleep apnea was one important factor related to sleepiness, but that an additional central disturbance of sleep mechanisms must be present to explain the occurrence of excessive daytime sleepiness in PWS patients [17]. Pulmonary Function The study of Hakonarson et al. [44] demonstrated that individuals of all ages with PWS had evidence of restrictive ventilatory impairment, which appeared to be the result of thoracic muscle weakness independent of obesity or chest wall abnormalities. Severe muscular hypotonia in PWS is a dominant feature in the neonatal period and persists during childhood and adolescence [2]. Both mild restrictive and obstructive pulmonary disease, but no evidence of decreased thoracic muscle strength was found in obese but otherwise clinically healthy individuals [44]. Autonomic Nervous System Function With respect to autonomic nervous system function findings in patients with PWS were markedly different from those in patients with obstructive sleep apnea. Whereas the acute and chronic answer to hypoxia and apnea was characterized by increased sympathetic activity [45], autonomic nervous system function in PWS was characterized by a reduction in parasympathetic activity [46]. Compared with controls, subjects with PWS exhibited both higher resting pulse rates and lower incremental pulse rates after standing; this was most notable in subjects with the highest BMI, for whom diastolic and mean arterial pressure were significantly low. This finding supported a faulty baroreceptor reflex principally affecting the parasympathetic branch of the reflex arc [46]. In nonsyndromal obesity depressions of sympathetic and parasympathetic activity were weakly correlated with an increasing percentage of body fat. Thus, the parasympathetic deficiency in PWS patients is believed to be more specifically related to PWS than to the individual’s degree of obesity per se [46]. Chemoreceptor Function Information on the actual blood gas values (pH, pO2, pCO2) is mediated via peripheral and central chemoreceptors to the central rhythm generator of respiration in the medulla oblongata [47]. Chemoreceptors represent an important part of the chemical feedback regulation loop of respiratory control, which is of special importance during sleep. In the awake state, the rhythm of inspiration and expiration is primarily maintained by a respiratory drive produced by the neuronal activity of the ascending reticular arousal system of the brain stem which is reduced during sleep [47]. The ventilatory and arousal responses to hypoxia, hyperoxia and hypercapnia during sleep and wakefulness are

Respiration, Sleep and PWS

135

influenced by chemoreceptor function [47]. Independent of obesity peripheral chemoreceptor activity was found to be absent in Prader-Willi patients since their minute ventilation in response to hyperoxia was not decreased and they did not show significant increases in minute ventilation with rapidly developing oxygen desaturation [48, 49]. Patients with PWS failed to arouse to hypoxic challenge and did not increase their respiratory rate in response to hypoxia. The fall in end-tidal CO2 during hypoxia was significantly lower than in control subjects. These findings indicated decreased ventilatory response to hypoxia in sleeping patients with PWS, and they corresponded to markedly blunted step and ramp hypoxic ventilatory responses in awake patients with PWS [48–50]. Peripheral chemoreceptor afferents may also play a role in the delayed arousal response to hypercapnia in PWS [51, 52]. Arens et al. [48] demonstrated a diminished response to hypercapnia in obese PWS patients suggesting that obesity in PWS drastically blunted central chemoreceptor responses.

Therapeutic Strategies Improving Sleep-Disordered Breathing in PWS

Reduction of Body Weight There are anecdotal reports that respiratory disturbances in obese patients with PWS improved after considerable weight reduction [19, 53]. In a group of 14 PWS patients with a mean BMI of 42.8 kg/m2 a mean reduction of body weight of 11% did not show any significant effect on ventilation during sleep [19]. Nasal Continuous/Bilevel Positive Airway Pressure Therapy In the case reports of Weinzheimer et al. [54], Müller et al. [55], and Sforza et al. [40] successful treatment of sleep-related respiratory disturbances by nasal continuous/bilevel positive airway pressure (NCPAP/BiPAP) therapy in obese PWS patients was documented. Lämmer and Desaga [19] treated 7 of 14 patients in their case series with NCPAP/BiPAP therapy. Some of these authors concluded that obstructive apnea had to be the main cause of hypersomnolence in PWS. A closer look at their results reveals that sleep onset REM in these patients did not change with NCPAP therapy. Thus, there is evidence even from these case reports that factors other than obstructive sleep apnea syndrome are present. Growth Hormone Therapy Recently it has been found that body composition of PWS patients resembles body composition of patients with growth hormone (GH) deficiency. Consecutively, the effect of GH therapy was systematically studied in PWS patients. GH was found to increase respiratory muscle strength, physical

Schlüter

136

strength and agility of PWS patients, which was thought to be in part the consequence of improved respiration [56–59]. It is of special interest that GH therapy seems to have a stimulatory effect on central respiratory structures [60]. In the study of Lindgren et al. [60] a significant improvement of resting ventilation, airway occlusion pressure and ventilatory response to CO2 was observed 6–9 months after the start of GH therapy in a case series of 9 children with PWS.

References 1

2 3 4 5 6 7

8 9 10 11 12 13 14

15 16 17 18 19 20 21

Van Mil EGAH, Westerterp KR, Gerver WJM, Van Marken Lichtenbelt WD, Kester ADM, Saris WHM: Body composition in Prader-Willi syndrome compared with nonsyndromal obesity: Relationship to physical activity and growth hormone function. J Pediatr 2001;139:708–714. Holm VA, Cassidy SB, Butler MG, Hanchett JM, Greenswag LR, Whitman BY, Greenberg F: Prader-Willi syndrome: Consensus diagnostic criteria. Pediatrics 1993;91:398–402. Hudgel DW: Treatment of obstructive sleep apnea: A review. Chest 1996;109:1346–1358. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S: The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993;328:1230–1235. Hoffstein V: Snoring. Chest 1996;109:201–222. Strollo PJ Jr, Rogers RM: Obstructive sleep apnea. N Engl J Med 1996;334:99–104. Corbo GM, Forastiere F, Agabiti N, Pistelli R, Dell’Orco V, Perucci CA, Valente S: Snoring in 9- to 15-year-old children: Risk factors and clinical relevance. Pediatrics 2001;108: 1149–1154. Clarke DJ, Waters J, Corbett JA: Adults with Prader-Willi syndrome: Abnormalities of sleep and behaviour. J R Soc Med 1989;82:21–24. Jancar J: Prader-Willi syndrome (hypotonia, obesity, hypogonadism, growth, and mental retardation). J Ment Defic Res 1971;15:20–29. Laurance BM, Brito A, Wilkinson J: Prader-Willi syndrome after age 15 years. Arch Dis Child 1981;56:181–186. Burwell CS, Robin ED, Whaley RD, Bickelmann A: Extreme obesity associated with alveolar hypoventilation – A Pickwickian syndrome. Am J Med 1956;21:811–818. Kuhl W: History of clinical research on the sleep apnea syndrome. The early days of polysomnography. Respiration 1997;64 (suppl 1):5–10. Greenswag LR: Adults with Prader-Willi syndrome: A survey of 232 cases. Dev Med Child Neurol 1987;29:145–152. Carroll JL, Loughlin GM: Obstructive sleep apnea syndrome in infants and children: Clinical features and pathophysiology; in Ferber R, Kryger M (eds): Principles and Practice of Sleep Medicine in the Child. Philadelphia, Saunders, 1995, pp 163–191. Cassidy SB, McKillop JA, Morgan WJ: Sleep disorders in Prader-Willi syndrome. Dysmorphol Clin Genet 1990;4:13–17. Harris JC, Allen RP: Sleep disordered breathing and circadian disturbances of REM in PraderWilli syndrome. Sleep Res 1985;14:235A. Richards A, Quaghebeur G, Clift S, Holland A, Dahlitz M, Parkes D: The upper airway and sleep apnoea in the Prader-Willi syndrome. Clin Otolaryngol 1994;19:193–197. Friedman E, Ferber R, Wharton R, Dietz W: Sleep apnea in the Prader-Willi syndrome. Sleep Res 1986;13:142. Lämmer C, Desaga J-F: Schlafapnoe und Prader-Willi-Syndrom (PWS); in Mayer G (ed): Jahrbuch Schlafmedizin in Deutschland 1994. München, MMV Medizin Verlag, 1995, pp 83–87. Kaplan J, Fredrickson PA, Richardson JW: Sleep and breathing in patients with the Prader-Willi syndrome. Mayo Clin Proc 1991;66:1124–1126. Hertz G, Cataletto M, Feinsilver SH, Angulo M: Sleep and breathing patterns in patients with Prader-Willi syndrome: Effects of age and gender. Sleep 1993;16:366–371.

Respiration, Sleep and PWS

137

22 23 24

25 26 27

28 29 30

31

32 33 34

35

36

37 38 39 40

41 42 43 44

Cataletto M, Hertz G, Feinsilver S, Angulo M: Sleep and breathing patterns in adult patients with Prader-Willi syndrome. Am J Genet 1992;42:260–261. Helbing-Zwanenburg B, Kamphuisen HAC, Mourtazaev MS: The origin of excessive daytime sleepiness in the Prader-Willi syndrome. J Intellect Disabil Res 1993;37:533–541. Vela-Bueno A, Kales A, Soldatos CR, Dobladez-Blanco B, Campos-Castello J, Espino-Hutardo P, Olivan-Palacios J: Sleep in the Prader-Willi syndrome. Clinical and polygraphic findings. Arch Neurol 1984;41:294–296. Sokol RJ: The chronic disease of childhood obesity: The sleeping giant has awakened. J Pediatr 2000;136:711–713. Jahns L, Siega-Riz AM, Popkin BM: The increasing prevalence of snacking among US children from 1977 to 1996. J Pediatr 2001;138:493–498. US Department of Health and Human Services: Update: Prevalence of overweight among children, adolescents, and adults – United States 1988–94. MMWR Morb Mortal Wkly Rep 1997; 46:199–202. Reilly JJ, Dorosty AR: Epidemic of obesity in UK children. Lancet 1999;354:1874–1875. Benediktsdottir B, Gislason T: Who is the sleepy teenager? Somnologie 1997;1(suppl 2): 6. Paditz E, Gräther M, Koch R, Erler T, Hoch B, Schäfer T, Stute H, Wiater A: Häufigkeit von OSAS-Symptomen im Kleinkindesalter – Vorstudie, Multizenterstudie der AG Pädiatrie DGSM. Somnologie 1999;3:313–318. Knuistingh Neven A, Middelkoop HAM, Kemp B, Kamphuisen HAC, Springer MP: The prevalence of a clinically significant sleep apnoea syndrome in the Netherlands. Somnologie 1997;1 (suppl 2):8. Ali NJ, Pitson DJ, Stradling JR: Snoring, sleep disturbance, and behaviour in 4–5 year olds. Arch Dis Child 1993;68:360–366. Ali NJ, Pitson D, Stradling JR: Natural history of snoring and related behaviour problems between the ages of 4 and 7 years. Arch Dis Child 1994;71:74–76. Bohr KC: Schlafuntersuchungen beim Prader-Willi-Syndrom (PWS) unter spezieller Berücksichtigung der Atmung; in Meier-Ewert K, Rüther E (eds): Schlafmedizin. Suttgart, Fischer, 1993, pp 264–268. Butler JV, Whittington JE, Holland AJ, Boer H, Clarke D, Webb T: Prevalence of, and risk factors for, physical ill-health in people with Prader-Willi syndrome: A population-based study. Dev Med Child Neurol 2002;44:248–255. Weess HG, Sauter C, Geisler P, Böhning W, Wilhelm B, Rotte M, Gresele C, Schneider C, Schulz H, Lund R, Steinberg R, Arbeitsgruppe Vigilanz der Deutschen Gesellschaft für Schlafforschung und Schlafmedizin: Vigilanz, Einschlafneigung, Daueraufmerksamkeit, Müdigkeit, Schläfrigkeit – Diagnostische Instrumentarien zur Messung müdigkeits- und schläfrigkeitsbezogener Prozesse und deren Gütekriterien. Somnologie 2000;4:20–38. Gozal D, Wang M, Pope DW Jr: Objective sleepiness measures in pediatric obstructive sleep apnea. Pediatrics 2001;108:693–697. Wiater A, Niewerth HJ: Polysomnographic standards for infants and children. Somnologie 2000; 4:39–42. Wharton RH, Levine K, Hobson AJ: Abnormalities of sleep and arousal in the Prader-Willi syndrome. Am J Med Genet 1992;42:261. Sforza E, Krieger J, Geisert J, Kurtz D: Sleep and breathing abnormalities in a case of Prader-Willi syndrome. The effects of acute continuous positive airway pressure treatment. Acta Paediatr Scand 1991;80:80–85. Fredrickson PA, Kaplan J, Richardson JW: Sleep disorders center evaluation of Prader-Willi patients. Sleep Res 1989;18:332A. Hungs M, Mignot E: Pathophysiological and clinical aspects of narcolepsy: A disorder associated with hypocretin abnormalities. Somnologie 2000;4:111–116. Sheldon SH, Spire JP, Levy HB: Anatomy of sleep; in Sheldon SH, Spire JP, Levy HB (eds): Pediatric Sleep Medicine. Philadelphia, Saunders, 1992, pp 37–45. Hakonarson H, Moskovitz J, Daigle KL, Cassidy SB, Cloutier MM: Pulmonary function abnormalities in Prader-Willi syndrome. J Pediatr 1995;126:565–570.

Schlüter

138

45 46 47

48 49 50

51 52 53

54 55 56

57

58

59

60

Fletcher EC: Sympathetic activity and blood pressure in the sleep apnea syndrome. Respiration 1997;64(suppl 1):22–28. DiMario FJ Jr, Dunham B, Burleson JA, Moskovitz J, Cassidy SB: An evaluation of autonomic nervous system function in patients with Prader-Willi syndrome. Pediatrics 1994;93:76–81. Schläfke ME, Hein H, Kotterba S, Rasche K, Rühle KH, Sanner B, Schäfer C, Schäfer D, Schäfer T: Zur Bedeutung des CO2-Partialdruckes als Messgrösse im Schlaflabor. Somnologie 1997;1: 184–196. Arens R, Gozal D, Omlin KJ, Livingston FR, Liu J, Keens TG, Davidson Ward SL: Hypoxic and hypercapnic ventilatory responses in Prader-Willi syndrome. J Appl Physiol 1994;77:2224–2230. Gozal D, Arens R, Omlin KJ, Davidson Ward SL, Keens TG: Absent peripheral chemosensitivity in Prader Willi syndrome. J Appl Physiol 1994;77:2231–2236. Arens R, Gozal D, Burrell BC, Bailey SL, Bautista DB, Keens TG, Davidson Ward SL: Arousal and cardiorespiratory responses to hypoxia in Prader-Willi syndrome. Am J Respir Crit Care Med 1996;153:283–287. Livingston FR, Arens R, Bailey SL, Keens TG, Davidson Ward SL: Hypercapnic arousal responses in Prader-Willi syndrome. Chest 1995;108:1627–1631. Schlüter B, Buschatz D, Trowitzsch E, Aksu F, Andler W: Respiratory control in children with Prader-Willi syndrome. Eur J Pediatr 1997;156:65–68. Orenstein DM, Boat TF, Owens RP, Horowitz JG, Primiano FP Jr, Germann K, Doershuk CF: The obesity hypoventilation syndrome in children with the Prader-Willi syndrome: A possible role for familial decreased response to carbon dioxide. J Pediatr 1980;97:765–767. Weinzheimer HR, Brack C, Haverkamp F, Lentze MJ, Loos L: Prader-Willi-Syndrom: Extreme obstruktive Apnoen und PEEP-Maskenbeatmung. Monatsschr Kinderheilkd 1994;142:S58. Müller T, Erler T, Oehlschläger: Adipöser Jugendlicher mit Vigilanzstörung. Monatsschr Kinderheilkd 1999;147:756–757. Carrel AL, Myers SE, Whitman BY, Allen DB: Growth hormone improves body composition, fat utilization, physical strength and agility, and growth in Prader-Willi syndrome: A controlled study. J Pediatr 1999;134:215–221. Eiholzer U, Gisin R, Weinmann C, Kriemler S, Steinert H, Torresani T, Zachmann M, Prader A: Treatment with human growth hormone in patients with Prader-Labhart-Willi syndrome reduces body fat and increases muscle mass and physical performance. Eur J Pediatr 1998;157:368–377. Davies PSW, Evans S, Broomhead S, Clough H, Day JME, Laidlaw A, Barnes ND: Effect of growth hormone on height, weight, and body composition in Prader-Willi syndrome. Arch Dis Child 1998;78:474–476. Myers SE, Carrel AL, Whitman BY, Allen DB: Sustained benefit after 2 years of growth hormone on body composition, fat utilization, physical strength and agility, and growth in Prader-Willi syndrome. J Pediatr 2000;137:42–49. Lindgren AC, Hellström LG, Ritzen EM, Milerad J: Growth hormone treatment increases CO2 response, ventilation and central inspiratory drive in children with Prader-Willi syndrome. Eur J Pediatr 1999;158:936–940.

Priv.-Doz. Dr. Bernhard Schlüter, Vestische Kinderklinik Datteln, University of Witten/Herdecke, Dr. Friedrich-Steiner Strasse 5, D–45711 Datteln (Germany) Tel. ⫹49 2363 9750, Fax ⫹49 2363 64211, E-Mail [email protected]

Respiration, Sleep and PWS

139

Comorbidities or Fundamental Defects of Obesity Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 140–155

Gonadal Function and Its Disorders in Simple Obesity and in Prader-Willi Syndrome Graziano Grugnia, Francesco Morabitoa, Antonino Crinòb a

Division of Auxology, IRCCS St. Giuseppe Hospital, Italian Auxological Institute Foundation, Verbania, and bUnit of Autoimmune Endocrine Diseases, IRCCS Bambino Gesù Hospital, Rome, Italy

Abstract Excessive body weight and reproductive dysfunction are common features of simple obesity and Prader-Willi syndrome (PWS). Overweight impacts on reproductive function early in life. It has been observed that the onset of menarche may occur at an earlier age in obese girls than in normal weight girls. Excessive body weight is commonly associated with a decreased reproductive potential. Furthermore, obesity may represent a significant factor in the pathogenesis of the polycystic ovary syndrome. Obesity in men is associated with low total testosterone levels, whereas free testosterone is normal until massive obesity occurs. Concerning the gonadotropin patterns, LH and FSH secretion in simple obesity is generally normal in both sexes. In PWS, hypogonadism is prenatal in onset and is evident in delayed and incomplete pubertal development. Nevertheless, pregnancy and birth have been reported, while fertility in males with PWS has never been demonstrated. As concerns endocrine aspects, we analyzed the gonadotropin response to GnRH in 38 PWS adults. In this group, FSH and LH levels are compatible with hypogonadotropic hypogonadism in a significant proportion of patients, while a large number of subjects (39.4%) have a normal gonadotropin response to GnRH. Furthermore, a small number of PWS shows signs of hypergonadotropic hypogonadism. Moreover, gonadotropin response to GnRH is significantly lower in subjects with uniparental maternal disomy than in patients with del15. In conclusion, reproductive alterations in simple obesity and in PWS are related to different mechanisms, with a prevalence of central dysfunction in the latter. Copyright © 2003 S. Karger AG, Basel

Introduction

It has long been known that nutritional status and reproduction are intimately related [1, 2]. Thus, the observation that food restriction may either delay sexual maturation or derange adult reproductive function led to the hypothesis that menarche is associated with a threshold body weight or body fat content [3, 4]. In fact, menarche typically occurs at a minimal body weight of 48 kg, independently from the age of sexual maturation [5]. Subsequent evidence has suggested that body fat distribution, rather than total body fat mass, is more closely related to sex steroid levels at the onset of puberty [6]. Moreover, it is well established that the reproductive function in adults is dependent on the availability of energy in the environment. In adult women, the attainment of a given amount of fat mass is required for the maintenance of normal reproductive capacity [7]. In male subjects, total testosterone (TT) is inversely correlated with body weight [8]. Indeed, extremes of body mass are associated with alterations of the pituitary-gonadal axis in humans. Women experiencing severe dietary restriction (e.g., anorexia nervosa) or who are high performance athletes (e.g., ballet dancers and long distance runners) have severely impaired reproductive systems [9]. On the opposite side of the spectrum, obesity produces a variety of alterations of the gonadal hormone system [10]. As overweight is an increasing problem for fertility [11], extensive research on the role of obesity in human reproduction has been published [12]. Excessive body weight and reproductive dysfunction are common features of simple obesity and Prader-Willi syndrome (PWS). Hypogonadism has been a recognized major feature of PWS since it was first described [13]. Nevertheless, the majority of the studies of the gonadal axis in PWS were performed in the 1970s. In this paper, our current understanding of reproductive alterations both in nonsyndromal obesity and in PWS will be briefly reviewed. In addition, our findings on the gonadotropin response to GnRH in 38 adults with PWS are reported.

Simple Obesity

Obesity impacts on reproductive function early in life. It has been observed that the onset of menarche may occur at an earlier age in obese girls than in normal weight girls [14], a fact consistent with the ‘critical mass’ hypothesis [15] and the hypothesis that the rise in leptin levels may be the earlier signal of the onset of puberty [16]. As leptin is produced mainly by adipose tissue [17], the larger the fat mass the more leptin is produced.

Hypogonadism in Prader-Willi Syndrome

141

Simple obesity may frequently be associated with hirsutism and menstrual disorders, particularly in women with a central distribution of body fat. Several studies have suggested a relationship between visceral obesity and an androgenic sex steroid profile in women [10]. Excessive body weight is commonly associated with an increased risk of anovulatory cycles [12], as well as with an increased number of atretic follicles [18]. These observations indicate that obesity has not only a direct influence on the start of menstruation, but it may also have an important role in the subsequent regulation of this process. A possible explanation of these alterations may be related to the negative effects exerted by high leptin levels on gonadal steroidogenesis, leading to ineffective follicular maturation [19]. Therefore, it is not surprising that obese women often have a decreased reproductive potential, as demonstrated by time to conception, poor response to infertility treatment, and increased risk of miscarriage [20, 21]. Obesity is very common among women with the polycystic ovary syndrome (PCOS) and may represent a significant factor in the pathogenesis of this syndrome [22]. Epidemiologic evidence suggests that the onset of obesity just before or during puberty could represent a risk factor for the subsequent development of PCOS. PCOS is a heterogeneous disorder characterized by hyperandrogenism and chronic anovulation [23]. Obese women with PCOS have a different hormonal environment from that of lean affected patients. The mechanism for the relationship between obesity and hyperandrogenism is multifactorial. Several studies have demonstrated an increased production of estrogens in obese PCOS subjects [10]. Obesity itself may be associated with high estrogen concentrations, mostly due to peripheral aromatization of circulating androgens to estrogen by adipose tissue. Obese PCOS women also have lower sex hormonebinding globulin (SHBG) levels than their nonobese counterparts [24]. As a result, an increased fraction of both free estrogens and testosterone may be available at the level of target tissues. Furthermore, obese PCOS patients have more severe hyperinsulinemia and insulin resistance than normal weight PCOS women [25]. The presence of both hyperinsulinemia and obesity seems to play a synergistic role in influencing androgen levels and adrenal steroidogenesis [26]. Higher concentration of androgens were demonstrated in obese PCOS subjects compared with lean patients [27]. The elevated levels of androgens and insulin increase the synthesis of IGF-1 and suppress SHBG and IGFBP1 production in the liver. Consequently, higher amounts of free forms of IGF-1 and insulin might influence ovarian steroidogenesis in obese PCOS women [28]. Concerning the gonadotropin pattern, where obesity and PCOS are present, a higher prevalence of insulin resistance together with disproportionately high LH secretion was reported [23]. Nevertheless, conflicting reports exist about the association of PCOS with LH hypersecretion in obese patients [25].

Grugni/Morabito/Crinò

142

Other studies have found that BMI negatively influences LH pulse amplitude and the peak increment of LH in response to GnRH stimulation in PCOS women [22]. Because LH and insulin appear to be the key hormones in the stimulation of androgen production by the ovary [29], it has been suggested that obese women with PCOS may be particularly susceptible to the effect of insulin on ovarian steroidogenesis [25]. In fact, insulin is able to stimulate the LH-mediated ovarian androgen synthesis and also may regulate the delivery of androgens in peripheral tissues [30]. Despite the evidence that leptin may regulate both gonadal steroidogenesis and gonadotropin secretion, the contribution of leptin to the pathogenesis of PCOS has not been established [16]. LH and FSH secretion in premenopausal women with simple obesity is normal [10]. It has been demonstrated that the onset of ovarian failure and increased secretion of FSH at menopause occur earlier in obese subjects than in women of normal weight [12]. However, obese menopausal patients have lower FSH levels than nonobese subjects, due to greater estrogen exposure [31]. Gonadotropin levels are normal in obese men, with a tendency to reduced LH basal concentration in severe obesity [32]. Furthermore, a normal pituitary reserve of LH and FSH following injection of GnRH is generally reported [12]. Obesity in men is associated with lower TT levels than normal weight controls [33]. Moreover, TT response to hCG stimulation is lower than in control subjects. This decrease in TT seems to be related to the fall in SHBG which occurs with increasing BMI, as well as to the reduction of LH pulse amplitude and hyperestrogenemia [34]. The reduction of SHBG appears to be connected to the inhibitory effect induced by hyperinsulinemia. In addition, hyperleptinemia is able to play an important role in the development of reduced androgens in male obesity [32]. Free testosterone levels are less affected than TT concentrations. Free testosterone is normal until massive obesity occurs, suggesting a functional impairment of the gonadotropic stimulation of the Leydig cells [35]. Despite these findings, there is no evidence of clinical androgen deficiency or abnormal semen analysis in morbidly obese men [36]. Thus, it has been suggested that the decrease in androgen levels is still within physiological concentrations, without any consequences as regards sexual activity [35].

Prader-Willi Syndrome

Hypogonadism is an almost invariable feature of PWS and is prenatal in onset [37]. At birth scrotal hypoplasia and cryptorchidism are common in males, and the penis is small or very small in the majority of the patients. Genital hypoplasia is the rule in females, and is manifested by hypoplasia of the labia minora and clitoris [38]. These features persist throughout life, although

Hypogonadism in Prader-Willi Syndrome

143

cryptorchid testes may still descend spontaneously in some patients up to adolescence. In both boys and girls with PWS, the development of secondary sexual characteristics, such as pubic and axillary hair, may progress early or normally. This is related to the activation of the adrenal gland that occurs at adrenarche. However, it is of note that in patients with precocious pubarche, sexual hair failed to develop beyond a certain stage [39]. Puberty is characteristically delayed and usually incomplete, although precocious puberty has been reported [40]. The degree of hypogonadism is variable and is more severe in males than in females. Boys with PWS rarely progress past Tanner genital stage 2–3. Testes are generally very small with a volume of less than 6 ml, and sometimes they are impalpable [41]. The scrotum is typically flat, and the penis is infantile. Nevertheless, single cases with normal adult testicular volume and spontaneous erections and ejaculations have been reported [42]. Adult males only occasionally have voice change, or substantial facial or body hair [38]. Gynecomastia may be present in subjects with PWS-associated obesity, as in patients with simple obesity. In females with PWS, pubic hair is normal in approximately 40% and scanty in 60% [43]. The onset of breast development may be at the normal age, and is presumably due to estrogens derived from adrenal androgen conversion in peripheral tissues [44]. However, breast development is variable and ranged from Tanner stages 1–4, regardless of age. Precocious thelarche among the female patients has not yet been found in PWS. Primary amenorrhea is reported in two thirds of the patients, and oligomenorrhea in the remaining subjects. Menarche may occur between the ages of 10 and 38 years [43, 45]. Bleeding is mostly irregular and its frequency varied from monthly to 1–2 times a year. Nevertheless, precocious menarche has been also described in girls with PWS [39, 46]. Moreover, pregnancy has recently been reported in two females with molecular genetically confirmed PWS. In the first case, a woman with maternal disomy gave birth to an unaffected child [47]. In another report, a woman with deletion 15q11-q13 gave birth to a girl with Angelman syndrome [48]. In contrast, fertility in males with PWS has never been documented. Hypogonadism associated with PWS is generally linked to a hypothalamic dysfunction in the regulation of gonadotropin secretion. Most authors have proposed that hypogonadism in PWS is caused by a gonadotropin deficiency of hypothalamic origin. Thus, genital hypoplasia at birth is probably due to an intrauterine hypogonadotropic hypogonadism [44]. In confirmation of this diagnosis, decreased basal levels of LH and FSH [41, 49] and a blunted response to GnRH stimulation [41, 50], that improves after repeated subcutaneous injections of GnRH [51], have been demonstrated. Furthermore, clomiphene administration has been reported to induce pubertal development [49, 52] as

Grugni/Morabito/Crinò

144

well as an increase in gonadotropins with improved GnRH responsiveness [49, 53]. The restoration of LH and FSH responsiveness by either repeated GnRH or clomiphene treatment led to the confirmation of the fact that the defect is located in the hypothalamus. Moreover, since clomiphene is able to restore gonadotropin responsiveness, the basic defect seems to be related to a disordered hypothalamic control of GnRH synthesis and/or release [41]. Finally, a selective hypothalamic dysfunction has been demonstrated in one PWS patient with elevated FSH levels, successfully treated with hCG [54]. In this regard, high FSH and low LH levels in less frequent cases of PWS were shown [55, 56]. However, other findings are not in accordance with the suggestion that hypogonadism in PWS is always of central origin [57], as demonstrated by either the low testicular responsiveness to hCG in males [49, 51, 58, 59] or the subnormal estradiol response to administration of hMG in a female [41]. Furthermore, other authors have found an exaggerated gonadotropin response to GnRH [57, 58, 60], an increased excretion of urinary gonadotropins [13], and elevated plasma gonadotropin levels [57, 61]. It has, therefore, been suggested that in addition to a hypothalamic defect, in patients with PWS hypogonadism may be related to a primary gonadal disorder [41]. In this respect, an absence of spermatogonia [62] and/or Leydig cells [58] in testicular biopsies, as well as interstitial cell defects [49], thickened basement membranes and atrophy of seminiferous tubules with some hyalinization [58, 61] have been reported. Nevertheless, the possibility that cryptorchidism as a consequence of hypothalamic dysfunction may cause the assumed primary testicular damage cannot be ruled out [44, 63]. Therefore, it may be speculated that some findings, which were interpreted by many authors as the consequence of a primary testicular failure, are rather the consequence of a maturational arrest of the gonadal axis due to hypothalamic dysfunction in an early stage of maturation [64, 65]. However, PWS appears to be a heterogeneous disorder in respect to hypogonadism. In female patients, both absence of follicle development [66] and normal ovary histology [48] have been reported. There is also evidence for delayed normalization of testosterone levels in men [63] as well as a normal gonadotropin response to GnRH in some individuals of both sexes [64]. An exaggerated gonadotropin responsiveness to GnRH was also found in a female with PWS [60]. Because almost all studies on the gonadal axis in PWS were performed before 1980, this heterogeneity may in part be explained by a poor selection of the patients. In these papers the diagnosis of PWS was based on clinical criteria, before the genetics of PWS had been delineated. Thus, our current understanding of hypothalamus-pituitary-gonadal axis in PWS may be misled by an inadequate characterization of the patients.

Hypogonadism in Prader-Willi Syndrome

145

To further clarify the pathophysiology of the gonadal dysfunction in PWS, we have analyzed the gonadotropin response to GnRH in a large group of adults with genetically confirmed PWS.

Experimental Data

Subjects and Methods Thirty-eight PWS patients, 11 males and 27 females, aged 15.9–29.1 years (mean ⫾ SE: 21.2 ⫾ 0.6 years), were included in the study (table 1). All subjects showed the typical clinical PWS features [67]. Cytogenetic studies were performed in all patients, and 25 of them had interstitial deletion of the proximal long arm of chromosome 15 (15q11-q13) (del15). Moreover, uniparental maternal disomy (UPD) for chromosome 15 was found in 13 individuals by DNA analysis. Thirty-five PWS subjects were obese and 3 were nonobese (cases 3, 26, 31). Mean BMI (⫾SE) was 40.8 ⫾ 1.4. In both genders, incomplete development of secondary sexual characteristics according to Tanner standards was present [68]. In males testes were palpable with a volume of less than 6 ml. One male (case 38) had a unilateral orchidectomy at the age 12. Seven females had spontaneous menarche between the ages of 14.1 and 25.3 years (25.9%). Irregular menstruation was present in 3 patients (11.1%; cases 4, 19, 26). No PWS subject was undergoing hormonal replacement therapy. Eighteen patients with simple obesity (BMI 39.3 ⫾ 1.3; 7 males and 11 females aged 16.3–26.9 years, mean ⫾ SE: 22.1 ⫾ 0.4) were enrolled as a control group. Obese women were studied in the follicular phase of the cycle, and none of them had PCOS. The entire study protocol was approved by the local Ethical Committees. After an overnight fast, all the subjects underwent a standard GnRH bolus test (100 ␮g; Lutrelef, Ferring, Kiel, Germany). Blood samples for LH, FSH, estradiol or TT determination were drawn from an indwelling catheter inserted in the antecubital vein 30 min before and at 0, 30, 60, 90 and 120 min after the intravenous bolus. Serum LH and FSH were measured with the immunochemiluminescent assay (Immulite, Diagnostic Product, USA). The analytical sensitivity of LH and FSH was 0.1 IU/l. The intra- and interassay CVs at 3.1 IU/l were 5.3 and 11.9% for LH, and 4.4 and 10.5% at 8.0 IU/l for FSH. The normal range for LH basal values was 1.4–7.7 IU/l (male) and 1.6–8.3 IU/l (female), and 1.5–14 IU/l (male) and 3.4–10.0 IU/l (female) for FSH. Estradiol and TT were determined by chemiluminescent immunoassay (Immulite, Diagnostic Product). The assay detection limits were 10 pg/ml and 0.1 ng/ml, respectively. The intra- and interassay CVs were 8 and 11.4% for estradiol and 4.9 and 9.0% for TT. Normal

Grugni/Morabito/Crinò

146

values for estradiol were ⱖ30 pg/ml (follicular phase), 120–370 pg/ml (peak) and 60–260 pg/ml (luteal phase) and 2.7–17.3 ng/ml for TT. Gonadotropin response to the provocative test was evaluated either as the difference between the basal level and the peak value (delta response) or as the area under the curve (AUC IU/l/h) calculated applying the trapezoidal method. LH and FSH responses to GnRH administration were analyzed according to the literature [69]. Statistical analysis was performed by the t test for unpaired data, and using analysis of variance for parametric or nonparametric (Mann-Whitney test) data, where appropriate. The relationship between pairs of variables was assessed by Pearson’s correlation. p value less than 0.05 was considered significant. Results PWS patients and the obese control group were well matched for sex, age and BMI. In the PWS group, the basal levels of both LH and FSH were within the normal range in 8 patients (2 females, 6 males), elevated in 2 males, and reduced in 18 subjects (16 females, 2 males) (table 1). Ten PWS patients showed low FSH or LH basal values (9 females, 1 male). Basal serum estradiol was ⱖ30 pg/ml in 11 females and TT was normal in 2 males, while the remaining PWS subjects had reduced levels of circulating sex hormones. All subjects with simple obesity showed normal gonadotropin (females ⫽ LH 4.4 ⫾ 0.3 IU/l vs. PWS females 1.0 ⫾ 0.2 IU/l, p ⬍ 0.001; FSH 5.4 ⫾ 0.5 IU/l vs. PWS females 2.3 ⫾ 0.3 IU/l, p ⬍ 0.001; males ⫽ LH 4.7 ⫾ 0.3 IU/l vs. PWS males 4.5 ⫾ 1.2 IU/l, p ⫽ 0.9; FSH 8.4 ⫾ 0.7 IU/l vs. PWS males 8.9 ⫾ 3.0 IU/l, p ⫽ 0.8) and TT (8.7 ⫾ 1.2 ng/ml vs. PWS 2.3 ⫾ 1.9 ng/ml, p ⬍ 0.03) or estradiol (58 ⫾ 6 pg/ml vs. PWS 32.6 ⫾ 2.7 pg/ml, p ⬍ 0.001) basal levels. The individual values of the gonadotropin response to GnRH stimulation of patients with PWS were reported in figures 1 and 2. The following patterns of secretion were found: (1) PWSa: 14 subjects (11 females, 3 males) showed a normal gonadotropin response [LH: 14.1 ⫾ 0.9 IU/l (delta), 1,259 ⫾ 119 IU/l/h (AUC); FSH: 10.5 ⫾ 0.6 IU/l (delta), 1,528 ⫾ 88 IU/l/h (AUC)]; (2) PWSb: 15 patients (11 females, 4 males) demonstrated a blunted gonadotropin response [LH: 1.1 ⫾ 0.3 IU/l (delta), 139 ⫾ 37 IU/l/h (AUC); FSH: 1.2 ⫾ 0.3 IU/l (delta), 158 ⫾ 34 IU/l/h (AUC)]; (3) PWSc: 2 females (cases 23, 24) had a reduced LH response with a normal FSH level [LH: 2.1 ⫾ 2.1 IU/l (delta), 186 ⫾ 126 IU/l/h (AUC); FSH: 11.8 ⫾ 1.6 IU/l (delta), 1,345 ⫾ 180 IU/l/h (AUC)]; (4) PWSd: 2 females (cases 25, 26) had a reduced FSH response with a normal LH level [LH: 14.3 ⫾ 0.2 IU/l (delta), 1,304 ⫾ 58 IU/l/h (AUC); FSH: 1.7 ⫾ 0.3 IU/l (delta), 286 ⫾ 24 IU/l/h (AUC)]; (5) PWSe: 3 males (cases 35, 36, 37) showed an exaggerated gonadotropin response

Hypogonadism in Prader-Willi Syndrome

147

Grugni/Morabito/Crinò

Table 1. Clinical and laboratory data of patients with PWS Patient No.

148

Sex Age, years

Karyotype BMI

TS

Menarche LHb

LHd

LH AUC

FSHb

FSHd

FSH AUC

Estradiol /TT

1 2 3 4

F F F F

17.9 20.9 15.9 27.2

del15 del15 del15 UPD

31.0 42.3 25.3 47.6

B2PH3 B3PH3 B2PH2 B4PH3

0.6 0.9 0.1 2.0

11.4 15.0 11.1 17.0

859 1,422 1,015 1,500

5.0 3.9 4.8 3.0

7.5 10.4 12.0 8.0

1,290 1,525 1,680 1,110

25 72 26 22

5 6 7 8 9

F F F F F

16.0 19.0 24.9 22.6 18.2

UPD UPD del15 del15 del15

37.8 34.7 45.1 53.6 46.8

B4PH3 B2PH3 B2PH2 B4PH4 B3PH3

0.3 0.5 0.5 0.2 0.4

10.0 11.7 10.0 11.8 11.9

862 834 802 903 955

6.2 2.3 0.9 4.0 2.3

10.3 10.0 12.3 12.0 8.1

1,678 1,290 1,306 1,515 1,051

28 31 27 54 34

10 11 12 13 14 15 16 17 18

F F F F F F F F F

28.8 20.1 21.9 17.4 22.0 19.4 15.9 17.8 25.7

del15 del15 del15 del15 UPD UPD UPD UPD UPD

39.3 39.1 56.9 40.4 49.2 48.3 44.3 50.6 53.8

B4PH4 B3PH3 B3PH4 B3PH3 B3PH3 B3PH2 B2PH3 B2PH3 B2PH3

0.5 5.0 0.1 0.3 0.1 0.5 0.1 0.1 0.1

15.0 19.0 0.0 0.1 0.0 0.0 1.0 0.6 0.9

1,119 1,920 12 39 12 60 93 62 76

4.2 3.7 0.4 0.2 0.1 0.1 1.4 0.1 0.6

15.6 9.3 0.0 0.2 0.5 0.6 1.9 1.2 1.5

1,836 1,210 39 37 57 61 354 106 211

24 52 18 52 15 36 27 30 29

19

F

22.7

UPD

42.5

B3PH3

0.6

3.5

364

2.0

3.0

331

28

20 21

F F

20.4 27.9

UPD del15

45.6 53.8

B4PH3 B3PH3

0.5 0.1

0.0 0.4

60 40

0.1 0.1

0.0 1.2

12 133

25 62

22 23

F F

21.5 18.3

del15 del15

51.0 34.2

B4PH3 B2PH3

absent absent absent 15.5 years absent absent absent absent 15.9 years absent absent absent absent absent absent absent absent 23.8 years 17.4 years absent 25.3 years absent absent

1.3 0.1

3.6 4.2

357 313

2.5 2.5

3.0 13.5

304 1,525

21 49

Hypogonadism in Prader-Willi Syndrome

24

F

16.5

del15

46.2

25 26

F F

21.1 26.0

del15 del15

45.0 23.0

27 28 29 30 31 32 33 34 35 36 37 38

F M M M M M M M M M M M

22.1 17.5 19.2 29.1 16.9 17.0 16.1 22.8 21.1 28.1 26.4 23.0

del15 UPD del15 del15 UPD UPD del15 del15 del15 del15 del15 del15

32.8 39.5 43.6 42.1 21.0 31.0 41.6 40.0 29.0 32.1 38.4 34.6

Mean ⫾ SE

21.2 ⫾ 0.6

40.8 ⫾ 1.4

B2PH2 14.1 years B4PH3 absent B3PH3 15.3 years B2PH3 absent G2PH3 G2PH3 G2PH2 G2PH4 G2PH2 G2PH3 G2PH2 G2PH3 G2PH3 G2PH3 G2PH3

0.5

0.0

60

2.7

10.2

1,165

22

4.5 4.1

14.6 14.1

1,363 1,246

2.9 1.2

2.0 1.4

310 262

18 24

4.3 3.1 3.0 6.2 2.6 0.4 1.7 0.7 15.1 3.8 9.0 4.5

81.7 16.9 18.2 19.6 1.7 1.1 2.9 1.0 46.7 39.0 48.1 7.5

5,824 1,575 1,764 2,103 225 96 463 127 5,454 4,413 4,927 1,173

5.1 7.0 6.4 11.3 0.6 1.4 2.5 0.1 30.6 9.9 24.3 4.0

14.7 10.1 13.4 8.6 0.1 3.2 2.5 0.0 23.8 30.9 30.9 27.0

1,806 1,815 2,022 2,066 82 378 252 12 5,910 3,931 5,889 3,635

30 1.4 6.9 2.1 1.8 1.3 0.1 0.5 4.7 3.1 1.6 2.0

2.0 ⫾ 0.4 12.4 ⫾ 2.8

1,170 ⫾ 246 4.2 ⫾ 1.0

8.4 ⫾ 1.3 1,268 ⫾ 237

32.6 ⫾ 2.7 /2.3 ⫾ 1.9

Gonadotropin secretion is expressed either as basal levels (LHb, FSHb: IU/l), delta values (LHd, FSHd: IU/l) or area under the curve (LH AUC, FSH AUC: IU/l/h) to GnRH. BMI ⫽ Body mass index (kg/m2); TS ⫽ Tanner pubertal stages; del15 ⫽ interstitial deletion of the proximal long arm of chromosome 15 (15q11-q13); UPD ⫽ uniparental maternal disomy of chromosome 15.

149

100 90 80

LH (IU/l)

70 60 50 40 30 20 10 0

0

30

a

60

90

120

0

30

Time (min)

60

90

120

b

Time (min)

Fig. 1. PWS patients with normal (a) and altered (b) LH response to GnRH.

80 70 60

FSH (IU/l)

50 40 30 20 10 0

a

0

30

60 Time (min)

90

120

0

30

60 Time (min)

90

120

b

Fig. 2. PWS patients with normal (a) and altered (b) FSH response to GnRH.

[LH: 44.6 ⫾ 2.8 IU/l (delta), 4,931 ⫾ 300 IU/l/h (AUC); FSH: 28.5 ⫾ 2.3 IU/l (delta), 5,243 ⫾ 656 IU/l/h (AUC)]; (6) PWSf: 1 female (case 27) demonstrated an exaggerated LH response with a normal FSH level, and (7) PWSg: 1 male (case 38) had an exaggerated FSH response with a normal LH level.

Grugni/Morabito/Crinò

150

In obese controls, the gonadotropin response to GnRH was similar to that observed in the PWSa group: LH: 14.6 ⫾ 0.4 IU/l (delta), 1,301 ⫾ 128 IU/l/h (AUC); FSH: 11.3 ⫾ 0.5 IU/l (delta), 1,589 ⫾ 98 IU/l/h (AUC). On the other hand, the values of LH differed significantly from PWSb ⫹ PWSc (delta LH ⫽ 1.2 ⫾ 0.3 IU/l, p ⬍ 0.0005; LH AUC ⫽ 144 ⫾ 34 IU/l/h, p ⬍ 0.0005). Furthermore, LH response to GnRH was lower than in PWSe ⫹ PWSf (delta LH ⫽ 53.8 ⫾ 9.4 IU/l, p ⬍ 0.0001; LH AUC ⫽ 5,154 ⫾ 308 IU/l/h, p ⬍ 0.0001). In the group with simple obesity, FSH values were higher than those found in PWSb ⫹ PWSd (delta FSH ⫽ 1.3 ⫾ 0.2 IU/l, p ⬍ 0.0001; FSH AUC 173 ⫾ 32 IU/l/h, p ⬍ 0.0005), and significantly lower if compared with PWSe ⫹ PWSg (delta FSH ⫽ 28.1 ⫾ 1.7 IU/l, p ⬍ 0.0001; FSH AUC ⫽ 4,841 ⫾ 613 IU/l/h, p ⬍ 0.0001). The same results were obtained when PWSa was compared with the remaining PWS subgroups. Finally, in PWS subjects no significant correlation was found between gonadotropin secretory pattern, age and BMI. On the contrary, gonadotropin response to GnRH was significantly higher in PWS patients with del15 than that observed in disomic subjects: delta LH: 16.2 ⫾ 3.8 IU/l (del15) versus 4.9 ⫾ 1.8 IU/l (UPD), p ⬍ 0.05; LH AUC: 1,546 ⫾ 345 IU/l/h (del15) versus 447 ⫾ 155 IU/l/h (UPD), p ⬍ 0.04; delta FSH: 10.8 ⫾ 1.8 IU/l (del15) versus 3.8 ⫾ 1.1 IU/l (UPD), p ⬍ 0.02; FSH AUC: 1,628 ⫾ 327 IU/l/h (del15) versus 575 ⫾ 181 IU/l/h (UPD), p ⬍ 0.04.

Conclusions

Excessive body weight produces a variety of alterations in the reproductive system. The relationship between essential obesity and gonadal dysfunction is multifactorial, and both central and peripheral factors may be involved. However, the impairment of gonadal function associated with simple obesity seems to be related to different mechanisms from those observed in patients with PWS. PWS appears to be a heterogenic disorder in respect to sexual alterations. The variation in hypogonadism between subjects with PWS is demonstrated by the variability in pubertal development as well as by the different patterns of gonadotropin secretion. These findings are supported by our results in a series of 38 young adults with genetically confirmed PWS. In this group, LH and FSH levels are compatible with hypogonadotropic hypogonadism in a significant proportion of patients. Furthermore, a small number of subjects, particularly males, shows signs of hypergonadotropic hypogonadism, probably due to cryptorchidism, as a consequence of early central hypogonadism, and its treatment. Nevertheless, an exaggerated LH responsiveness to GnRH in one female

Hypogonadism in Prader-Willi Syndrome

151

was also observed, as previously reported [60]. This draws attention to questions of whether cryptorchidism, as such, causes the supposed primary gonadal failure. Moreover, we have found a normal gonadotropin response to GnRH in a large number of patients (39.4%), notwithstanding the presence of incomplete sexual development. In this regard, additional studies are necessary to explain this discrepancy between phenotype and normal gonadal function since an altered gonadotropin bioactivity cannot be ruled out, at least in some patients. Finally, gonadotropin response to GnRH was significantly lower in subjects with UPD than in patients with del15. Our current understanding of the hypothalamic-pituitary-gonadal axis in PWS is still incomplete, and many aspects of the pathophysiology of the reproductive axis in this syndrome remain to be explained. Gathering more knowledge on the functioning of the reproductive system in PWS might be used not only to determine more appropriate treatment, including contraception, but also to normalize the life of the patients to the greatest possible extent. Therefore, further studies are required to better define the role of other factors potentially involved in the altered reproductive function, such as insulin, leptin and inhibin B levels. Moreover, most subjects with PWS have a reduced GH secretory capacity. Consequently, it will be of great interest to determine to what extent alterations in the GH/IGF-1 axis in these patients may contribute to the impaired gonadal function.

References 1 2 3 4 5 6

7 8 9 10 11 12

Frisch RE, Revelle R: Height and weight at menarche and a hypothesis of critical body weights and adolescent events. Science 1970;169:397–399. Frisch RE, Revelle R: Components of weight at menarche and the initiation of the adolescent growth spurt in girls: Estimated total body water, lean body weight, and fat. Hum Biol 1973;45:469–483. Frisch RE, McArthur J: Menstrual cycles: Fatness as a determinant of minimum weight or height necessary for maintenance or onset. Science 1974;185:949–951. Schwartz MW, Seeley RJ: Neuroendocrine responses to starvation and weight loss. N Engl J Med 1997;336:1802–1811. Frisch RE: Pubertal adipose tissue: Is it necessary for normal sexual maturation? Evidence from the rat and human female. Fed Proc 1980;39:2395–2400. deRidder CM, Thijssen JH, Bruning PF, Van der Brande JL, Zonderland ML, Erich WB: Body fat mass, body fat distribution, and pubertal development: A longitudinal study of physical and hormonal sexual maturation of girls. J Clin Endocrinol Metab 1992;75:442–446. Stewart DE: Reproductive function in eating disorders. Ann Med 1992;24:287–291. Tchernof A, Després JP, Belanger A, Dupont A, Prud’homme D, Moorjani S, Lupien PJ, Labrie F: Reduced testosterone and adrenal C19 steroid levels in obese men. Metabolism 1995;44:513–519. De Souza M, Metzger D: Reproductive dysfunction in amenorrheic athletes and anorexic patients: A review. Med Sci Sports Exerc 1991;23:995–1007. Smith SR: The endocrinology of obesity. Endocrinol Metab Clin North Am 1996;25:921–942. Wickelgren I: Obesity: How big a problem? Science 1998;280:1364–1367. Bray GA: Obesity and reproduction. Hum Reprod 1997;12(suppl 1):26–32.

Grugni/Morabito/Crinò

152

13

14 15 16 17 18 19 20 21

22 23 24

25 26

27

28

29

30

31 32

33 34

35

Prader A, Labhart A, Willi H: Ein Syndrom von Adipositas, Kleinwuchs, Kryptorchismus und Oligophrenie nach myotonieartigem Zustand im Neugeborenenalter. Schweiz Med Wochenschr 1956;86:1260–1261. Jaruratanasirikul S, Mosuwan L, Lebel L: Growth pattern and age at menarche of obese girls in a transitional society. J Pediatr Endocrinol Metab 1997;10:487–490. Frisch RE, Revelle R: Height and weight at menarche and a hypothesis of menarche. Arch Dis Child 1971;46:695–701. Moschos S, Chan JL, Mantzoros CS: Leptin and reproduction: A review. Fertil Steril 2002;77: 433–443. Zhang Y, Proenca R, Maffei M, Baroni M, Leopold L, Friedman JM: Positional cloning of the mouse obese gene and its human homologue. Nature 1994;372:425–432. Fischer ER, Gregorio R, Stephan T, Nolan S, Danowski TS: Ovarian changes in women with morbid obesity. Obstet Gynecol 1974;44:839–844. Caprio M, Fabbrini E, Isidori AM, Aversa A, Fabbri A: Leptin in reproduction. Trends Endocrinol Metab 2001;12:65–72. Norman RJ, Clark AM: Obesity and reproductive disorders: A review. Reprod Fertil Dev 1998; 10:55–63. Fedorcsak P, Dale PO, Storeng R, Tanbo T, Abyholm T: The impact of obesity and insulin resistance on the outcome of IVF or ICSI in women with polycystic ovarian syndrome. Hum Reprod 2001;16:1086–1091. Arroyo A, Laughlin GA, Morales AJ, Yen SSC: Inappropriate gonadotropin secretion in polycystic ovary syndrome: Influence of adiposity. J Clin Endocrinol Metab 1997;82:3728–3733. Franks S: Polycystic ovary syndrome. N Engl J Med 1995;333:853–861. Kiddy DS, Sharp PS, White DM, Scanlon MF, Mason HD, Bray CS, Polson DW, Reed MJ, Franks S: Differences in clinical and endocrine features between obese and non-obese subjects with polycystic ovary syndrome: An analysis of 263 consecutive cases. Clin Endocrinol 1990;32:213–220. Pasquali R, Casimirri F: The impact of obesity on hyperandrogenism and polycystic ovary syndrome in premenopausal women. Clin Endocrinol 1993;39:1–16. Ciampelli M, Fulghesu AM, Cucinelli F, Pavone V, Ronsisvalle E, Guido M, Caruso A, Lanzone A: Impact of insulin and body mass index on metabolic and endocrine variables in polycystic ovary syndrome. Metabolism 1999;48:167–172. Acien P, Quereda F, Matallin P, Villaroya E, Lopez-Fernandez JA, Acien M, Mauri M, Alfayale R: Insulin, androgens, and obesity in women with and without polycystic ovary syndrome: A heterogeneous group of disorders. Fertil Steril 1999;72:32–40. Morales AJ, Laughlin GA, Butzow T, Maheshwari H, Baumann G, Yen SS: Insulin, somatotropic, and luteinizing hormone axes in lean and obese women with polycystic ovary syndrome: Common and distinct features. J Clin Endocrinol Metab 1996;81:2854–2864. Huber-Buchholz MM, Carey DGP, Norman RJ: Restoration of reproductive potential by lifestyle modification in obese polycystic ovary syndrome: Role of insulin sensitivity and luteinizing hormone. J Clin Endocrinol Metab 1999;84:1470–1474. Nader S, Charles MA, Saad MF, Berkowitz AS, Bogardus C: Serum androgens in hyperinsulinemic Pima Indian and obese Caucasian women and their response to short-term insulin infusion. J Endocrinol Invest 1993;16:403–406. Malacara JM, Fajardo ME, Nava LE: Gonadotropins at menopause: The influence of obesity, insulin resistance, and estrogens. Steroids 2001;66:559–567. Isidori AM, Caprio M, Strollo F, Moretti C, Frajese G, Isidori A, Fabbri A: Leptin and androgens in male obesity: Evidence for leptin contribution to reduced androgen levels. J Clin Endocrinol Metab 1999;84:3673–3680. Vermeulen A: Decreased androgen levels and obesity in men. Ann Med 1996;28:13–15. Vermeulen A, Kaufman JM, Deslypere JP, Thomas G: Attenuated luteinizing hormone (LH) pulse amplitude but normal LH pulse frequency, and its relation to plasma androgens in hypogonadism of obese men. J Clin Endocrinol Metab 1993;76:1140–1146. Lima N, Cavaliere H, Knobel M, Halpern A, Medeiros-Neto G: Decreased androgen levels in massively obese men may be associated with impaired function of the gonadostat. Int J Obes 2000;24: 1433–1437.

Hypogonadism in Prader-Willi Syndrome

153

36 37 38 39 40 41 42 43 44 45 46 47 48

49 50 51

52 53 54 55 56 57 58 59

60 61 62

Gordon Baker H: Reproductive effects of nontesticular illness. Endocrinol Metab Clin North Am 1998;27:831–850. Lee PDK: Endocrine and metabolic aspects of Prader-Willi syndrome; in Greenswag LR, Alexander R (eds): Management of the Prader-Willi Syndrome. New York, Springer,1995, pp 32–57. Cassidy SB: Prader-Willi syndrome. J Med Genet 1997;34:917–923. Kauli R, Prager-Levin R, Laron Z: Pubertal development in the Prader-Labhart-Willi syndrome. Acta Paediatr Scand 1978;67:763–767. Linnemann K, Schroder C, Mix M, Kruger G, Fusch C: Prader-Labhart-Willi syndrome with central precocious puberty and empty sella syndrome. Acta Paediatr 1999;88:1295–1297. Jeffcoate WJ, Laurance BM, Edwards CRW, Besser GM: Endocrine function in the Prader-Willi syndrome. Clin Endocrinol 1980;12:81–89. Vanelli M, Bernasconi S, Caronna N, Virdis R, Terzi C, Giovannelli G: Precocious puberty in a male with Prader-Labhart-Willi syndrome. Helv Paediatr Acta 1984;39:373–377. Cassidy SB, Rubin KG, Mukaida CS: Genital abnormalities and hypogonadism in 105 patients with Prader-Willi syndrome. Am J Med Genet 1987;28:922–923. Lindgren AC: Prader-Willi Syndrome – Diagnosis and Effects of Growth Hormone Treatment. Stockholm, Repro Print, 1998. Greenswag LR: Adults with Prader-Willi syndrome: A survey of 232 cases. Dev Med Child Neurol 1987;29:145–152. Walterspiel JN, Wolff J, Heinze E: Pubertas praecox bei Prader-Labhart-Willi Syndrom. Klin Padiatr 1981;193:120–121. Akefeldt A, Tornhage CJ, Gillberg C: A woman with Prader-Willi syndrome gives birth to a healthy baby girl. Dev Med Child Neurol 1999;41:789–790. Schulze A, Mogensen H, Hamborg-Petersen B, Graem N, Ostergaard JR, Brondum-Nielsen K: Fertility in Prader-Willi syndrome: A case report with Angelman syndrome in the offspring. Acta Paediatr 2001;90:455–459. Hamilton CR, Scully R, Kliman B: Hypogonadotropism in Prader-Willi syndrome: Induction of puberty and spermatogenesis by clomiphene citrate. Am J Med 1972;52:322–329. Ritzen EM, Bome P, Hall K: Endocrine physiology and therapy in Prader-Willi syndrome; in Cassidy SB (ed): Prader-Willi Syndrome. Berlin, Springer, 1992, vol 61, pp 153–169. Tolis G, Lewis W, Verdy M, Friesen H, Solomon S, Pagalis G, Pavlatos F, Fessas P, Rochefort JG: Anterior pituitary function in the Prader-Labhart-Willi (PLW) syndrome. J Clin Endocrinol Metab 1974;39:1061–1066. Linde R, McNeil L, Rabin D: Induction of menarche by clomiphene citrate in a fifteen-year-old girl with the Prader-Labhart-Willi syndrome. Fertil Steril 1982;37:118–120. McGuffin WL, Rogol AD: Response to LHRH and clomiphene citrate in two women with the Prader-Labhart-Willi syndrome. J Clin Endocrinol Metab 1975;41:325–331. Jaskulsky SR, Stone NN: Hypogonadism in Prader-Willi syndrome. Urology 1987;29:207–208. Nagai T, Mimura N, Tomizawa T, Monden T, Mori M: Prader-Willi syndrome with elevated follicle stimulating hormone levels and diabetes mellitus. Intern Med 1998;37:1039–1041. Saad S, Paterson WF, Donaldson MDC: Gonadal function and pubertal development in boys with Prader-Willi syndrome. Horm Res 1996;46(suppl 2):1–135. Garty B, Shuper A, Mimouni M, Varsano I, Kauli R: Primary gonadal failure and precocious adrenarche in a boy with Prader-Labhart-Willi syndrome. Eur J Pediatr 1982;139:201–203. Seyler LE, Arulanatham K, O’Connor CF: Hypergonadotropic hypogonadism in the PraderLabhart-Willi syndrome. J Pediatr 1979;94:435–437. Zarate A, Soria J, Canales ES, Kastin AJ, Schally AV, Toledano RG: Pituitary response to synthetic luteinizing hormone-releasing hormone in Prader-Willi syndrome, prepubertal and pubertal children. Neuroendocrinology 1973;13:321–326. Endo M, Tasaka Y, Matsuura N, Matsuda I: Laurence-Moon-Biedl syndrome and Prader-Willi syndrome in a single family. Eur J Pediatr 1976;123:269–276. Wannarachue N, Ruvalcaba HA, Kelley V: Hypogonadism in Prader-Willi syndrome. Am J Ment Defic 1975;79:592–603. Katcher ML, Bargman G, Gilbert EF, Opitz JM: Absence of spermatogonia in the Prader-Willi syndrome. Eur J Pediatr 1977;124:257–260.

Grugni/Morabito/Crinò

154

63 64 65 66 67 68 69

Burman P, Ritzen EM, Lindgren AC: Endocrine dysfunction in Prader-Willi syndrome: A review with special reference to GH. Endocr Rev 2001;22:787–799. Muller J: Hypogonadism and endocrine metabolic disorders in Prader-Willi syndrome. Acta Paediatr Suppl 1997;423:58–59. Eiholzer U: Prader-Willi Syndrome. Effects of Human Growth Hormone Treatment. Basel, Karger, 2001. Bray G, Dahms W, Swerdloss R, Fiser R, Atkinson R, Carrel X: The Prader-Willi syndrome. A study of 40 patients and a review of the literature. Medicine 1983;62:59–80. Holm VA, Cassidy SB, Butler MG, Hanchett JM, Greenswag LR, Whitman BY, Greenberg F: Prader-Willi syndrome: Consensus diagnostic criteria. Pediatrics 1993;91:398–402. Tanner JM: Growth and Adolescence. Oxford, Blackwell Scientific Publications, 1962. Oerter KE, Uriarte MM, Rose SR, Barnes KM, Cutler GB: Gonadotropin secretory dynamics during puberty in normal girls and boys. J Clin Endocrinol Metab 1990;71:1251–1258.

G. Grugni, MD, Division of Auxology, St. Giuseppe Hospital, IRCCS Italian Auxological Institute Foundation, PO Box 1, I–28921 Verbania Intra (Italia) Tel. ⫹39 0323 514247, Fax ⫹39 0323 587694, E-Mail [email protected]

Hypogonadism in Prader-Willi Syndrome

155

Comorbidities or Fundamental Defects of Obesity Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 156–165

Children with Prader-Willi Syndrome and Primary Obesity: A Comparison of Appetite and Psychosocial Profiles The Family Perspective

Jane Gilmour, David Skuse Behavioural and Brain Sciences Unit, Institute of Child Health, University College London, London, UK

Abstract There are few data comparing children with Prader-Willi syndrome (PWS) and primary obesity (PO): two groups who show broad similarities in their response to food. Those data that do exist describe one-off observations of eating in controlled conditions or describe the physiology of eating. Comparisons from the family perspective between the groups have been neglected in the literature, but have implications for the utility of PWS as a model for obesity. For individuals with both PO and PWS, intervention in childhood with parental involvement has greatest efficacy. Children with a diagnosis of PWS (n ⫽ 50, mean age 10.5 years old, SD ⫽ 3.6 years, 57% male) were compared with a group of children with PO (n ⫽ 24, mean age 9.8 years old, SD ⫽ 3.2 years, 54% male) matched for BMI and socioeconomic status. All families in the PWS group were aware of their child’s diagnosis. While cognitive assessments were not completed in the course of the study, we must assume children with PWS had a lower level of cognitive ability (IQ) than children with PO. Parents, recruited through an advert in a women’s magazine, completed questionnaires including the Family Environment Scale regarding a number of different domains such as family functioning, typical eating patterns and parental attitudes about their child’s appetite. While there were statistical differences in family characteristics (PO families reported less cohesion, active recreation, expressiveness and more conflict than the PWS group) both groups’ mean scores on all dimensions were well within the average range and so are unlikely to constitute a meaningful clinical difference. The quality and severity of hyperphagia were similar in the PWS and PO groups. Nor were there differences regarding parental attributions about their child’s excessive appetite, despite a distinct aetiology in each group. Parental perceptions about the most challenging aspect of their child’s behaviour were also comparable. Copyright © 2003 S. Karger AG, Basel

Introduction

We aim to explore the parallels between Prader-Willi syndrome (PWS) and primary obesity (PO) from the family perspective. Understanding the differences and similarities between the conditions from this viewpoint is likely to have major implications for treatment models. For individuals with who are obese, early intervention has greatest efficacy [1, 2]. For children and young people, parental involvement in treatment increases the likelihood of a successful outcome [3]. Obesity is defined according to convention as a body mass index (BMI) of 30 kg/m2 which is equivalent to the 97th centile in UK populations [4]. For the purposes of this paper we refer to individuals with PO as those who are obese and who have no known condition or syndrome that underlies their obesity. It is important to be cautious about the use of the term PO. There are a number of specific biological factors that increase the likelihood of obesity such as a rare genetic mutation associated with leptin deficiency [5]. Further, there are almost certainly as yet unidentified factors that may have an influence on BMI. While it is important to keep these qualifications in mind, in the course of the study we refer to PO, with an understanding of the limitations of the meaning of the label. Obesity is associated with a multitude of diverse physical conditions such as non-insulin diabetes, endocrinological problems, or hypertension [6]. Recent data [7] suggest that psychosocial problems such as depression or low selfesteem may be evident in clinical obese populations, specifically female clinical populations, but are not characteristic of obese individuals in the general population. Longitudinal data are sparse but there are a number of risk factors associated with obesity, some of which are almost certainly causative rather than associative or secondary to obesity. Having obese parents increases the risk of obesity in children via two routes. First there is a genetic predisposition to obesity for at least a proportion of individuals who have so-called PO [8] which has been described using a number of different methodologies, including twin studies. There are probably many different simple genetic and polygenetic influences on obesity. In addition, obese parents present a risk to their children in terms of environmental exposure. This might include the types of food offered or life-style choices such as activity levels [9]. Families who have lower socioeconomic status are more likely to have members that are obese [10]. The mechanism for this effect may be both causative and secondary. Families of low socioeconomic status consume proportionately more fats, sugars, potatoes and cereals but fewer fresh vegetables, fruit and high-fibre products [11]. Foods that are traditionally healthy and low in fat such as highquality meat, fruit and salads are relatively speaking more expensive to buy,

Children with PWS and PO

157

certainly in industrialized western society [11]. It may also be that living in the stressful circumstances often associated with low socioeconomic status may mean that individuals are more likely to gravitate to high calorie ‘short-term’ fix foods. However there are no data as far as the author is aware that can substantiate this hypothesis. PWS, first described in 1956 [12], is caused by one of three different genetic anomalies, all of paternal origin on the long arm of chromosome 15 [13]. Seventy percent of affected individuals have a deletion at 15q 11-13, a further 28% have uniparental disomy and the final 2% have an imprinting mutation [14]. The key phenotypic feature is hyperphagia, leading almost certainly to the characteristic obesity [15]. The hyperphagia has an onset around the early primary school years [16] – though in our experience hyperphagia is not invariable in individuals who have a confirmed PWS genotype. Affected individuals often fail to thrive in infancy [15], in some cases secondary to functional feeding difficulties. They also have delayed motor milestones [15] and general learning difficulties [17]. There are additional endocrinological [18], neurological [19], behavioural features [20] and anthropometric features such as short stature [17] associated with the condition. A hypothalamic dysfunction is assumed because the key phenotypic features are known to be associated with hypothalamic nuclei such as appetite and growth. Very few studies directly compare PWS and PO. The studies that exist largely comprise a direct comparison of the microstructure of eating behaviour in laboratory-controlled conditions. Data indicate similar rates of eating [21] but there are indications from one study of an eating episode that the groups differ in their food group preferences – the PWS group preferred high carbohydrate foods [22]. PWS groups eat more calories in an hour-long observation than PO comparison groups [23]. Observational data [24] and data from questionnaires presented during the observation period [23] suggest that the PWS group has delayed satiety and experience hunger more quickly after eating as compared to the PO groups [23]. These findings are important but it is equally essential to describe habitual eating patterns – not least because the eating patterns of many individuals with PWS are rigorously controlled as they have an identified medical condition. One might hypothesize that control from parents, teachers, carers and so on is less rigorous in individuals who are ‘just obese’. We must consider the effect that having unlimited access to food (as is the case in the laboratory observation experiments) may have on the PWS group as compared to PO comparison samples. A number of studies directly comparing PO and PWS have examined the physiology of appetite response. For example, Zipf and Bernston [21] examined the effect of the opioid inhibitor, nalaxone, on appetite. Administration of the drug had no significant effect on the number of calories eaten by individuals

Gilmour/Skuse

158

with PWS in a laboratory session though it did in obese comparisons. Studies investigating metabolic changes in the general population following eating suggest cholecystokinin is a key part of the appetite regulation system [25]. Evidence indicates that while the PWS populations show an impairment in cholecystokinin levels, PO groups do not [23]. Similarly PWS groups have blood glucose levels above the expected range in the general population and above the range measured in the obese groups [26]. In sum, existing data suggest that distinct differences in physiological systems underlie observed eating behaviour in PO and PWS groups. We must acknowledge that there may be a subgroup of PO that has similarities to the PWS group in terms of eating behaviour and physiological mechanisms. Until there is the expertise to partial out subgroups within PO, such differences and similarities will remain obscured. Aims In this study we aimed to answer three questions. First, we asked how the family characteristics compare between families with a child who has PWS and PO. This is a key preliminary question. Without an exploration of family characteristics, other family perspective comparisons may be confounded. Second, how do families of hyperphagic children describe the severity and abnormality of typical appetite responses? We considered this important as no other data of which we are aware describe habitual patterns according to the family as opposed to one-off observations in laboratory conditions. Finally, how do families understand their child’s overeating? We hypothesized that the PWS group – who have a specific explanation for their child’s eating behaviour – would describe a different set of attributions and explanations as compared to the PO group who had no such information.

Methods Twenty-four children with PO were recruited (mean age 9.8 years old, SD ⫽ 3.2 years, 54% boys). We asked families about their contact with medical services. We excluded any child who had a condition that may have some influence on BMI. Therefore so far as we can ascertain these children have PO. Fifty children with a confirmed PWS diagnosis (mean age 10.5 years old, SD ⫽ 3.6 years, 57% boys) participated in the study. Both groups were recruited through a women’s magazine advert. Once the families made contact with the team, questionnaires were sent to the family to complete. The groups were balanced in terms of socioeconomic status and BMI. The mean standardized BMI was 1.9 (SD ⫽ 1.5) and 2.0 (SD ⫽ 1.5) for the PO and PWS groups, respectively. The PO group had a standardized height group mean of ⫺0.22 (SD ⫽ 2.1) which was, as expected, significantly greater (p ⬍ 0.01) than the PWS group mean of ⫺1.7 (SD ⫽ 1.6).

Children with PWS and PO

159

Table 1. Family characteristics (FES T scores) FES dimension

PO

PWS

Cohesion* Expressiveness* Conflict* Independence Achievement Intellectual/cultural Active recreation* Moral religious Organized Control

48.7 (16.6) 48.3 (9.5) 55.4 (12.6) 44.4 (15.0) 43.4 (13.3) 44.7 (12.0) 43.3 (12.1) 43.7 (9.3) 49.8 (9.2) 53.1 (11.9)

56.8 (14.2) 54.4 (10.7) 47.9 (12.0) 47.3 (13.8) 42.3 (11.7) 50.1 (13.5) 50.9 (14.2) 47.1 (10.4) 51.8 (10.8) 50.5 (9.6)

Values represent mean with the SD in parentheses. *p ⬍ 0.05.

The Family Characteristics The Family Environment Scale (FES) [1] is a well-standardized instrument used to characterize family characteristics along ten dimensions (see table 1). The instrument is calibrated as a T score, with a mean of 50 and an SD of 10. Any score between 40 and 60 should be considered well within normal limits. The FES has been used in previous studies examining response to treatment in PO families. Those PO families that are categorized as organized (families who had a T score of 65 or more on the organized dimension) are more likely to respond well to treatment [27]. Table 1 compares the family characteristics of the PO and PWS families. The PWS families reported more cohesion (t ⫽ ⫺2.2, p ⬍ 0.05), expressiveness (t ⫽ ⫺2.4, p ⬍ 0.05), and active recreation (t ⫽ ⫺2.3, p ⬍ 0.05) but less conflict (t ⫽ ⫺2.4, p ⬍ 0.05). All the group mean scores were well within the average range and therefore it is difficult to argue that such differences would constitute a clinical, meaningful difference in family characteristics. Family Reports of the Severity and Abnormality of Appetite The questionnaire comprised 23 items relating to eating behaviour and appetite. The questionnaire is based on an interview, which has good discriminate validity in identifying children with hyperphagia in the clinical range [28]. There are three domains of appetite response. Thirteen items were summed describing ‘hyperphagia’. This includes items such as gorging on food until vomiting or getting up in the night to look for food. The maximum score on this item is 13. The second domain describes ‘pica’. It comprises four items (with a maximum score of 4) such as eating non-food items such as toothpaste or eating raw or frozen food. The final domain describes ‘foraging behaviour’. It includes five items such as eating food from the floor or foraging in a dustbin for food. The range of scores on this item is 0 to 5. Figure 1 shows the family-reported habitual appetite behaviours on the three domains described. There were no statistically significant differences in the data using a Student t test for independent samples on any of the three domains. The means (SD) were as

Gilmour/Skuse

160

7

PO PWS

6 5 4 3 2 1 0 Hyperphagia

Foraging

Pica

Fig. 1. Typical eating patterns – three domains of appetite (group means are shown).

follows: 7.0 (3.3) for the PO and 4.2 (3.1) for the PWS group, respectively, on the hyperphagia domain, and 1.0 (1.3) versus.5 (1.6) for the pica domain as compared to 2.5 (1.2) versus 1.8 (1.4) for the foraging behaviour domain. Family Explanations for Appetite Disturbance Given that there were systematic differences in the information concerning aetiology in the PWS as compared to the PO groups, we fully expected resultant group differences in explanations for their child’s disturbed pattern of eating behaviour. In order to explore this issue we asked families a series of open questions concerning their child. Families’ responses were categorized into nominal categories by a team member who was unaware of group status. Comparisons were made using ␹2 tests in all cases. First we asked what would happen if the child had unlimited access to food. Nine percent of the PO and 2% of the PWS groups (non-significant) indicated there would be no significant difficulty – their child would demonstrate self-control and stop eating independently. Nine percent in the PO group as compared to 25% in the PWS group (non-significant) of families reported that their child would demonstrate some control, with prompting. For example, the children may need a gentle verbal reminder to ask if they had had enough to eat. The majority of both groups (82% of the PO and 73% of PWS groups; non-significant) reported that their child would have no control at all. These families suggested that their child would eat until all the food was gone or would need to be removed from the room, away from the food in order to stop eating. Second, we asked families if their child had taken food without permission. All families said that their child had done so at some point. We then asked why they thought their child had taken food without permission. As before all ␹2 comparisons were not significant. Twentyeight percent of the PO as compared to 16% of the PWS groups suggested that the reason their child had taken food was hunger. Other explanations included greed (5% in the PO and 6% in the PWS groups) or opportunity (22% in the PO vs. 56% in the PWS groups). In the remaining proportion of the groups (44 and 22% in the PO and PWS groups, respectively), the explanations were miscellaneous such as ‘he copied his brother’. Finally we asked families why their child overate on the last occasion. Remarkably around half of both groups (58% in the PO and 46% in the PWS group) reported that they did not know why their child had overeaten on the last occasion. Of the remaining families who provided an explanation 20% of the PO and 8% of the PWS groups said hunger (non-significant) was the reason for

Children with PWS and PO

161

their child’s last episode of overeating. Thirty percent of the PO versus 9% of the PWS group (non-significant) suggested the reason was greed; 30 versus 70% (non-significant) said that opportunity was the reason and 20 versus 13% (non-significant) attributed it to an emotional state such as boredom or sadness. All the families were aware of their child’s diagnosis in the PWS group. It is striking that none of the group referred to PWS for any of the eating behaviours about which we asked. Finally we asked families what aspect of behaviour was most difficult to manage. We did not test cognitive ability but we must assume the PWS group have a significantly lower IQ than the PO group as the former characteristically have general learning difficulties and the latter are likely to have a normally distributed IQ. In addition, it is important to note that while we did not ask families to document specific problems, PWS is associated with a number of different physical conditions and behavioural difficulties. Despite these presumed differences, the families in both groups gave comparable explanations for those aspects of their child’s behaviour they found most challenging. Learning difficulties (3% in the PO vs. 3.5% in the PWS group), behavioural difficulties (35 vs. 42%), eating (30 vs. 23.5%), physical problems (6 vs. 9%) social difficulties (23 vs. 20%) and miscellaneous difficulties (3 vs. 2%) were listed as areas of greatest concern. The categorizations were mutually exclusive and all comparisons were non-significant.

Discussion

Despite former evidence indicating significant differences in the microstructure [e.g. 23] and physiological mechanisms of appetite [e.g. 21], as well as in the information families receive about the underlying cause of hyperphagia in PWS as compared to PO, we found broad similarities in family characteristics and family perception about the characterization of and attributions about typical eating patterns. There are well-established social factors that are associated with PO. For example, the active recreational Sub-Scale in the FES [1] is a family attribute that has been previously established to be associated with PO, with families tending to be less active than the general population. While there were statistical group differences on a number of the subscales, both groups’ scores were well within normal limits on all dimensions, making it difficult to argue there is a meaningful difference between samples. It was an important preliminary exploration to establish that family characteristics were broadly similar. This indicates that other extraneous factors that may have influenced family perspectives to some small degree were controlled. We have raised the possibility that within the PO group there may be a number of different unidentified subtypes. While it is important to acknowledge this caveat, it is also fair to conclude that it does not confound the findings in the current paper. The aim of this paper is to explore the family perspectives of a group that have a clear diagnosis underlying the disruptive

Gilmour/Skuse

162

appetite patterns (PWS) as compared to those that do not (PO). Interestingly, we found that despite having a clearly different understanding of aetiology, family perspectives were largely comparable. In other words, while characterizing the subgroups within PO is theoretically interesting, it has little bearing on the conclusions in the paper. One might have predicted that the PWS group would show a greater degree of abnormality in their appetite response – an effect that may have been evident in the pica domain scores, for instance. No such effect was evident, as families described analogous habitual eating patterns in terms of severity and abnormality in both groups. Nor was there any difference in terms of severity on any of the three domains of appetite we assessed. Therefore, it seems that while there are subtle differences in the microstructure of the groups’ eating, the family experiences on a day to day basis are indistinguishable. There are a number of limitations to consider. First we acknowledge that families responding to the advert may have responded precisely because they felt their child’s eating was out of control. In other words, the study may have attracted an unusually severely affected sample. This may be the case but one must assume that the bias would apply equally to both the PWS and the PO groups. Further, we know from the literature that parents of a child with PO are more likely to be obese themselves, while this effect is not reported in the PWS parent population. We must consider how this may have influenced parental responses. We might hypothesize that obese parents may actually minimize their child’s overeating, if it was a pattern they themselves also showed. The data from this study are striking, indicating broad similarities from the family perspective between these conditions. From this we might consider the implications for treatment and support models for families who have a child with either PWS or so-called PO. We know that behavioural management strategies involving parents are most likely to be successful in PO. It might be useful to consider extending this approach to the treatment of PWS groups. A secondary benefit from such involvement might be to increase awareness in the PWS families in a day to day context about the effects of PWS as the findings from this study suggest that having a diagnosis in and of itself does not necessarily influence attributions about eating. From the PO family perspective, it is clear that family reports of habitual eating patterns in this self-selected group are similar to a clinically identified group. This has implications for the level of support that families who have a child with PO may need. There is an acknowledgement that it is stressful and at times distressing to monitor and control food intake in PWS households. Data from this study suggest that PO families may be subject to the same stresses as the PWS groups. Indeed one might hypothesize that the stresses in the PO families are greater as there is no obvious explanation for their child’s behaviour.

Children with PWS and PO

163

Acknowledgments We are grateful to the families who took part in the study, Joanne Newbolt and the Wellcome Trust for their financial support.

References 1 2 3 4 5 6 7 8 9 10 11 12

13 14 15 16 17 18

19 20 21 22

Moos RH, Moos BS: Family Environment Scale Manual. Palo Alto, Consulting Psychologists Press, 1986. Epstein LH, Wing RR, Valoski A: Childhood obesity. Pediatr Clin North Am 1985;32:363–379. Coates TJ, Killen JD, Slinkard LA: Parent participation in a treatment program for overweight adolescents. Int J Eat Disord 1982;1:37–48. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH: Establishing a standard definition for child overweight and obesity worldwide: International survey. Br Med J 2000;320:1240. O’Rahilly S: Leptin: Defining its role in humans by the clinical study of genetic disorders. Nutr Rev 2002;60:S68–S87. Slyper AH: Childhood obesity, adipose tissue distribution and the pediatric practitioner. Pediatrics 1998;102:1–9. Braet C, Mervielde I, Vandereycken W: Psychological aspects of childhood obesity. J Pediatr Psychol 1997;22:59–71. Stunkard AJ: Genetic contributions to human obesity. Res Publ Assoc Res Nerv Ment Dis 1991;69:205–218. Epstein LA, Valoski R, Wing RR, McCurley J: Ten-year follow-up of behavioural, family-based treatment for obese children. JAMA 1990;264:2519–2523. Gortmaker SL, Must A, Perrin JM, Sobol AM, Dietz WH: Social and economic consequences of overweight in adolescence and young adulthood. N Engl J Med 1993;329:1008–1012. James WP, Nelson M, Ralph A, Leather S: Socio-economic determinants of health: The contribution of nutrition to inequalities in health. Br Med J 1997;314:1545. Prader A, Labhart A, Willi H: Ein Syndrom von Adipositas, Kleinwuchs, Kryptorchismus und Oligophrenie nach myatonieartigem Zustand im Neugeborenenalter. Schweiz Med Wochenschr 1956;86:1260–1261. Ledbetter DH, Riccardi VM, Airhardt S, Strobel RJ, Keene BS, Crawford JD: Deletions of chromosome 15 as a cause of Prader-Willi syndrome. N Engl J Med 1981;304:325–329. Nicholls RD, Ohta T, Gray TA: Genetic abnormalities in Prader-Willi syndrome and lessons from mouse models. Acta Paediatr Suppl 1999;88:99–104. Cassidy S: Introduction and overview of Prader-Willi syndrome; in Cassidy SB (ed): Prader-Willi Syndrome. NATO Series ASI. Berlin, Springer, 1992, vol 61, pp 1–11. Holm VA, Cassidy SB, Butler MG, Hanchett JM, Greenswag LR, Whitman BY, Greenberg F: Prader-Willi syndrome: Consensus diagnostic criteria. Pediatrics 1993;91:398–402. Butler MG: Prader-Willi syndrome current understanding and diagnosis. Am J Genet 1989;35:319–332. Angulo M, Castro-Magana M, Uy J, Rosenfeld W: Growth hormone evaluation and treatment in Prader-Willi syndrome; in Cassidy SB (ed): Prader-Willi Syndrome. NATO Series ASI. Berlin, Springer, 1992, vol 61, pp 172–187. Helbing-Zwanenburg B, Kamphuisen HAC, Mourtazev MS: The origin of excessive day-time sleepiness in the Prader-Willi syndrome. J Intellect Disabil Res 1993;37:533–541. Curfs LM, Wiegers AM, Sommers JR, Borghgraef F, Fryns FP: Strengths and weakness in the cognitive profile of youngsters with Prader-Willi syndrome. Clin Genet 1991;40:430–434. Zipf WB, Bernston GB: Characteristics of abnormal food intake patterns in children with PraderWilli and a study of the effects of nalaxone. Am J Clin Nutr 1987;46:277–281. Fieldstone A, Zipf WB, Schwartz HC, Berston GG: Food preferences in Prader-Willi syndrome, normal weight and obese control. Int J Obes Relat Metab Disord 1997;21:1046–1052.

Gilmour/Skuse

164

23

24 25 26

27 28

Holland AJ, Treasure J, Coskeran P, Dallow J, Milton M, Hillhouse E: Measurement of excessive appetite and metabolic changes in Prader-Willi syndrome. Int J Intellect Disabil Res 1993;17: 527–532. Lindgren AC, Barkeling B, Hagg A, Ritzen EM, Marcus C, Rossner S: Eating behaviour in PraderWilli syndrome, normal weight and obese control groups. J Pediatr 2000;137:50–55. Morely JE: Neuropeptide regulation of appetite and weight. Endocr Rev 1987;8:256–287. Zipf WB, O’Dorisio TM, Cataland S, Dixon K: Pancreatic polypeptide responses to protein meal challenges in obese but otherwise normal children and obese children with Prader-Willi syndrome. J Clin Endocrinol 1982;57:1074–1080. Kirschenbaum DS, Harris ES, Tomarken AJ: Effects of parental involvement in behavioural weight loss therapy for preadolescents. Behav Ther 1984;15:485–500. Skuse D, Albanese A, Stanhope R, Gilmour J, Voss L: A new stress-related syndrome of growth failure and hyperphagia in children, associated with reversibility of growth-hormone insufficiency. Lancet 1996;348:353–357.

Jane Gilmour, PhD, Behavioural and Brain Sciences Unit, Institute of Child Health, University College London, 30 Guilford Street, London WC1N 1EH (UK) Tel. ⫹44 207 905 2394, Fax ⫹44 20 7829 8657, E-Mail [email protected]

Children with PWS and PO

165

Comorbidities or Fundamental Defects of Obesity Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 166–178

Discussion

Characterization of Hyperphagia in Prader-Willi Syndrome (A.C. Lindgren) Anon: How much impact do you think the behavior modifications that we give to our families, e.g. making sure that the children and adolescents are eating limited amounts, plays a role in what we see in this eating task? Is there an overlay of what they have been taught, that they should only eat about so much, that’s enough etc. in those individuals? And the second question I have is, how do you interpret the obesity study, that you have done, in relationship to the normal? In other words, it looks like the obese individuals were consuming less during that meal, even though they maintain the eating rate that was accelerated for the 8 to 10 minutes. They actually seem to stop eating sooner. How do you interpret those results? A.C. Lindgren, Stockholm: I think the behavior the children have learned might have a great influence. We tried to ask about this via a questionnaire and we asked the parents of the patients with PWS. The parents denied that they told the children that they shouldn’t eat so much. Nevertheless, I think it’s a big influence. Regarding the obese subjects, the results might also reflect that they have learned a behavior that they shouldn’t eat so much but they had a higher, but not significant, consumption of food. I think, yes, that learned behavior is very important. S.C. Woods, Cincinnati, Ohio: As I saw your graph, in fact as you just said, the obese patients actually ate more and it looked like the PWS kids and the normal kids ate exactly the same amount – it just took longer time to get there. That is interesting and it is just, unless there was some conditioning that went on with these kids, that they must eat more often with normal size meals. This implies a quite different ideology. My question is: What instructions did you give these people, when they sat down to eat? Lindgren: We told them to eat their breakfast as usual in the morning and nothing to eat until noon when they come up to the hospital and then we told them to eat as much as they want to, as they usually eat. They were seated in a room. The normal weight and the obese children and youngsters were seated alone in the room. However, the subjects with PWS had their parents or

caretakers with them in the room. This is because we noticed that many PWS children, they started to look at the table that was hiding the scale, lifted the plate up and everything went wrong. They were trying to find out what we have under the table. We had to have some adult in the room, because we had several patients that destroyed the results. K.R. Westerterp, Maastricht: I think the slow eating rate in your PraderWilli patients is very intriguing and I wonder about what you will do in further research, whether you think of maybe offering them another meal after a certain time and therefore to see, whether the eating rate is different in normal subjects as compared to Prader-Willi syndrome because in one way or another, you would like to explain why they eat so much. Lindgren: Yes, that would be interesting to do. But that’s what you have done, Tony (Holland), in your study. They had free access to sandwiches for a quarter of an hour and they continued to eat, as I understand. And after this hour you stopped the experiment and… D.J. Holland, Cambridge, UK: Can I just follow up on that? Because I think it is a very interesting study you have done. Did you actually ask them to rate the hunger during the course? Lindgren: Yes, they rated if they were hungry before lunch (when they sat down at the table) and then they rated if they were feeling full after the meal. Holland: What we found is that the Prader-Willi group did eventually rate themselves feeling less hungry or full. But only after eating three times more food or calories than the control group and then as you say, what happened is that not long afterwards their ratings of hunger started bouncing up again, much sooner than the control group. The one thing that seems to correlate with them feeling full was when levels of blood glucose, often in the diabetic range, were high. So it seemed to us that what brought about satiation was this change in glucose. As that presumably came down, they started rating themselves as feeling hungry. Lindgren: That’s interesting. Unfortunately, we didn’t measure that. Woods: If you look at our data too, when you look at the mean, you have this very high food consumption, but if you look at the actual data, there really were multiple groups. At least there were a hiding group and a lowering group and the hiding group dominated statistics, they really brought the mean up. There were still a large number of the PWS patients, who consumed, as you have shown in your data. In other words, they didn’t consume that much more than our obese controls. Our obese controls consumed more than our normals. However, the PWS subjects didn’t consume more than the obese, they just consumed it over the whole hour, as you had shown. So within our data, which includes about 25 children with PWS, we did see different patterns. We haven’t separated by age and sex, which I think you have to do ultimately.

Discussion

167

Lindgren: Yes, in older children, when they are about 15 years, there is a difference of consumption, depending on the gender. U. Eiholzer, Zurich: Do you know or do you think that there are different eating behaviors between UPDs and deletions? Lindgren: I have no idea. These subjects all had deletions. We only had one case with UPD in this group. So I cannot tell you. Eiholzer: Because my personal impression is that Prader-Willi children with darker hair eat faster than the others. Lindgren: I don’t know. These were all light-haired Swedish PWS. So I cannot tell you. Anon: In the slow eating, is hypotonia playing a role? Lindgren: We know that in infants, the hypotonia is playing a role. These children (the study subjects) didn’t have a normal muscular strength, but they were able to walk and to talk and to behave rather normally … and they didn’t have any problems with eating. However, hypotonia might have an influence, because they do not have a normal muscular strength. B.Y. Whitman, St. Louis, Mo.: I am also interested in that link of time in eating. Over the many years, when I have sat down to have a meal with kids with PWS, they have eaten their meal and cleaned their plate before I sat down and lifted my fork. It’s almost like they inhale the food. So what you are describing with that continuous eating is very different from what I have experienced in living with them over the years and I am wondering how the experimental situation might have changed the pattern of eating. Lindgren: I was expecting to find that situation, that they were going to throw the food into their mouths and eat very quickly. So we were really surprised, when we found this type of eating curve. D. l’Allemand, Zurich: I think we all have the impression that the young children with PWS don’t have the feeding problem in the same way as the older children. Are you aware of studies that have given evidence for that observation? Could you differentiate and relate the type of satiation in young and older children? Lindgren: That’s so intriguing in children with PWS, that you start with this failure to thrive and then you have this excessive appetite during to rest of your life. No, I am not aware of studies. But it is very intriguing, so I hope there will be some studies in the future. Anon: I think in infants it would be interesting to see how they drink the bottle or they suckle the mother’s breast. It may be lower than the usual babies. Lindgren: But usually they have a slow eating pattern. That’s what has been reported. And they can’t suck because of the muscular hypotonia. We don’t really know if they are feeling hungry, as compared with normal babies. They have a weak cry and they are sleeping during the whole 24 hours. Many parents are telling us ‘we have to wake him up to give him food’.

Comorbidities or Fundamental Defects of Obesity

168

F. Horber, Zurich: I am intrigued by your curve of the obese children. Why did they stop eating? Lindgren: I think, they have learned they shouldn’t eat. And they were eating so fast also that they felt full. And as I told you, when they slowed down and put down their fork, they were instructed then to leave the room when they were feeling this sensation of fullness/upset. So they could not sit in the room. Horber: A follow-up question: Did you ask them afterwards, the three groups, when they were hungry again? Lindgren: No, we didn’t. Horber: And did you do any interventions, for instance, in the PWS subjects? Did you investigate whether this slow eating pattern in PWS had any previous educational sessions? Lindgren: We asked the parents in a questionnaire, if the children have been taught to eat in a slow way or to a certain limit of food and so on and the parents denied it. However, all these children were on restricted caloric intake. So there must have been some education. Horber: There is the Swedish obese studies in adults. Is there anything like this in children? Lindgren: Yes there is a study going on in healthy obese children. Horber: Did you perform any genetic studies in the three groups? Lindgren: Only in the Prader-Willi group, not in the obese. The obese subjects belong to obesity clinics in Sweden. I know most of these patients and none of them has an MC4 receptor defect or anything. So I think, most of them had been genetically tested, but I cannot really tell you. Horber: When I look at your data about Prader-Willi, we did gastric bands on three Prader-Willi patients and we were horrified by the results. You know that’s what we are actually teaching our adult patients, to eat slowly. Because, when they eat slowly, they tolerate the band very nicely. When I look at your data, I would think that PWS children would do wonderfully having a gastric band. However in my experience in three cases, this is not what I see. They get upset, if they cannot eat, what they want to eat quickly. Lindgren: Interesting. Yes we were surprised, when we saw that eating curve in PWS. We thought it would be the contrary. F. Rohner-Jeanrenaud, Geneva: Is there any known defect in gastrointestinal tract mobility and emptying in PWS children? Horber: We looked at it in one patient and we couldn’t find anything. Consequences of Impaired Growth Hormone Secretion for Body Composition and Metabolism in Obesity and Prader-Willi Syndrome (M. Wabitsch) Anon: Are you sure there was an insulin-like effect at the very beginning? Could it be IGF-1? Because when you inject IGF-1 in someone, there is an

Discussion

169

initial hypoglycemic response. And I wonder whether, since you showed that there was IGF-1 made by the adipocyte, the initial transient effect could be an insulin-like effect of the IGF-1. M. Wabitsch, Ulm: It’s not possible to block this effect with IGF-1 antibodies. I don’t think the effect is due to the IGF that is produced and secreted by these cells. l’Allemand: Wouldn’t it be dangerous to treat obese people with GH, if this has a cell multiplying effect or if this has an IGF-like effect on the preadipocytes? Is it only the acute effect? Why doesn’t the fat mass increase finally? Wabitsch: The first and the main effect we see, when we treat obese patients – this has been published in the literature – it’s the lipolytic effect, which is more pronounced than the effect of increase in cell numbers of adipocytes. I don’t know the data on the long-term effect of GH treatment. Of course we could hypothesize that there might be an increase in the cell number and when we stop treatment after a certain time, maybe the adipose tissue will be even increased after the treatment period. l’Allemand: There has been observed a dose difference in adults treated with GH with low doses being more efficient. Can you comment on that? Wabitsch: Already at low doses of GH we see this lipolytic effect, and low doses of GH have a small increase in IGF-1. Maybe this is not sufficient to increase the cell number. But this is only what we can think. The lipolytic effect is very well demonstrated also at low concentrations. l’Allemand: Thank you, I think, that’s important information. A.P. Goldstone, London: You very nicely showed this proliferative effect of GH on, what I think, was subcutaneously derived preadipocytes. Given our finding of reduced visceral adiposity in PWS adults, do you know of any data showing a proliferative effect of GH on visceral preadipocytes? Wabitsch: We did such studies only in a few patients, because it’s very difficult to get this tissue and to culture these cells; it’s much more difficult than subcutaneous cells. We have the impression that the lipolytic effect of the hormone was increased in the visceral adipocytes. Goldstone: Did you also demonstrate a proliferative effect? Wabitsch: I cannot comment on that, we do not have enough data. However, maybe that would be an explanation why patients with PWS have reduced visceral fat, as we have discussed yesterday. Anon: Just a comment, concerning the lipolytic effect of GH. This can be demonstrated in children who are not GH-deficient; for example, if you investigate fat mass in SGA (small for gestational age) children, who are normally slim from outside. You can also demonstrate that this fat mass is further reduced in the first year of GH treatment. This has been demonstrated by Michael Ranke’s group recently.

Comorbidities or Fundamental Defects of Obesity

170

Glucose Homeostasis in Prader-Willi Syndrome (W.B. Zipf) Woods: I believe that there is a strong relationship between insulin and visceral adiposity and I know most of your Prader-Willi are males and most of your control obese were females. We might guess that there is more subcutaneous and visceral obesity in your control obese women and that might have something to do with the lower insulin levels in the Prader-Willi group. The comment I want to make is that I agree that everything points to decreased parasympathetic activity. One other observation is that if you had decreased parasympathetic activity, you would expect a very slow way of eating, which is exactly what we saw in an earlier paper. I think that the one would predict the other. W.B. Zipf, Columbus, Ohio: That’s true. I hadn’t time about that. Anon: Is anything known about glucose transporters in Prader-Willi? Zipf: GLUT4 has been studied and it’s thought to be normal. We haven’t done that ourselves, but there are some published papers looking at that. Anon: If you reduce weight in normal obese, GLUT4 levels go down. Zipf: But that is thought to be, I believe, a consequence of hyperinsulinemia and again, if there are normal GLUT4 concentrations, this would go along with the fact that the Prader-Willi group seems not to be hyperinsulinemic. Do you have any idea why Prader-Willi individuals are so insulin sensitive. This information could help us in future treatment. Zipf: Well, I think it is an interesting topic and it has been brought up at a couple of round table discussions where we have wondered if Prader-Willi syndrome is a model of obesity if you don’t become insulin resistant. The insulin resistance is thought to be partially genetic, you have to have the genes for it. Secondly, in terms of the hyperinsulinemia, the visceral adiposity and free fatty acid levels both contribute. Free fatty acid levels were studied by Denis Bier in Prader-Willi syndrome and were thought to be fairly normal, which goes along with the normal visceral adiposity. If you aren’t insulin resistant, insulin is a potent anabolic hormone. Amino acid transport is not downregulated by hyperinsulinemia and some of the increased muscle tissue that is observed in the normal obese individuals may be in part contributed by the anabolic effects of the extra insulin. So a soft correction factor in normal obesity might be the hyperinsulinemia, inducing further muscle tissue growth. And then the increased muscle tissue increases, along with the total basal metabolic rate, not per gram of tissue, but per individual, correlating with the degree of obesity and fat tissue. If one does not become insulin-resistant, one cannot get the hyperinsulinemia to improve muscle tissue and one might end up with partitioning a greater amount of calories into fat mass. This is going three four five six steps out, but one does have to ask if

Discussion

171

insulin resistance might not have some protective value to an individual – it has serious side effects – but if one didn’t become that, maybe we would all look like Prader-Willi. Rohner-Jeanrenaud: Along the same line, are there any data on plasma adiponectin levels? Zipf: I’ve not seen that looked at yet. Anon: I have a follow-up question. For me it’s not very conceivable that people are very insulin sensitive and have a low lean body mass. How would you explain that in Prader-Willi? Because normally, if you are insulin sensitive and you get high protein, you would think that lean body mass would be normal. Zipf: Well again, if one would speculate that the tonic control, if the autonomic nervous system was downregulated, both sympathetic and parasympathetic, as I think they are from our data and others, there would be decreased tonic sympathetic activity to muscle tissue. We know that in obesity, as you increase your weight by as much as 10–15%, sympathetic tone goes up, and it appears that this increased sympathetic tone may counter obesity, increasing caloric intake and increasing the metabolic rate in the use of these calories. So, it is important to look at the autonomic nervous system in this condition, because it might possibly explain the observation you describe. l’Allemand: I would like to comment on this based on a study in children with PWS. Part of the story is GHD, which would explain both the low lean mass and the extreme insulin sensitivity. We observed that with GH treatment, the insulin sensitivity normalizes; they don’t become insulin resistant. But we know that there must be another factor, such as decreased parasympathetic tone. My question was, did you observe this in young adults under GH treatment? For example, what is happening with the hepatic insulin extraction with GH therapy? Zipf: That’s an excellent question. We don’t have the answer to that. Anon: Sweating is under the influence of the parasympathetic autonomic system and also growth hormone. I wonder whether sweating might be another clue to look at in Prader-Willi syndrome, to see whether they sweat normally or not. Zipf: I don’t know. I’ve not seen that one way or the other. We don’t see the axillary changes, because there must be different hormones involved. They don’t going through puberty, so we don’t see the adult body odor, but I don’t know about sweating. We get this constant complaint that they are always cold. Anon: That would go along with the body temperature control? Zipf: I don’t know. Does anyone else have a comment on just normal sweating with exercise and temperature? Anon: In a way, considering the fact that increased heat and sweating can be a problem in individuals who have obesity, perhaps the absence of that ever

Comorbidities or Fundamental Defects of Obesity

172

having been brought to my attention as an issue in these children with PWS might be an indication that they don’t sweat as much. I don’t know. It’s a good question. Anon: So, you don’t need puberty to sweat? Anon: No, you don’t need puberty to sweat. Sleep-Disordered Breathing in Children with Common Obesity (H. Trang) Dysregulation of Respiration and Sleep in Prader-Willi Syndrome (B. Schlüter) R. Stanhope, London, UK: I was very interested to hear the first speaker mention modafinil. I have been using modafinil in children with hypothalamic disorders and obesity. One of the side effects of modafinil is anorexia. I have been using it predominantly in children with postsurgical craniopharyngioma and in some patients with Prader-Willi syndrome and it seems to have a very profound effect in about 50% of cases. In the cases where it works to induce weight loss, it orders sleep-awake cycles first; so that within the first week, they have a change in sleep cycles, they are more awake during the day and they sleep better at night. Do you have any experience with this? H. Trang, Paris: Are you talking only about children with PWS? Stanhope: Only children with hypothalamic disorders with obesity. My experience is predominantly those with craniopharyngiomas, but I’ve treated about five children with PWS with a very similar effect and I have tried it in a few patients with simple obesity and it doesn’t seem to work. Trang: I have no experience with use of modafinil in children with hypothalamic disorders. Our main indications are children with primary obesity and narcolepsy. They were already on a diet to lose weight before treatment with modafinil; it was difficult to know the extent to which modafinil may have increased the weight loss. We did not have the opportunity to perform measurements of body composition before and during treatment. Anon: I have a question about the sleep, about the disturbances of the REM cycles. How is stage 4 sleep in these patients? We know that GH secretion is related to stage 4 sleep. B. Schlüter, Datteln: I think, this has to be studied. I can’t give you a detailed answer at the moment. Eiholzer: Possibly, respiration problems are underestimated in PWS. We have lost two children in the last ten years and in the histories of some children in our group we found situations with long-term apnea. We have begun to do polysomnographies and echocardiographies in all our patients. I think these procedures should be done, because until recently, medical doctors didn’t notice

Discussion

173

this problem. It is possible that more very young children die without diagnosis of this apnea problem. C. Sizonenko, Meyrin: Before we break, I would like to say there is one word that has not been said: This is the Pickwickian syndrome. Remember that in Pickwickian syndrome with severe obesity, we sleep after a meal and I hope you are not going to sleep after our next meal… Gonadal Function and Its Disorders in Simple Obesity and in Prader-Willi Syndrome (G. Grugni) Eiholzer: You have shown three patients with exaggerated LH and FSH response. Are these three men or is there also a woman? G. Grugni, Verbania: This is a very good question. There are three males with exaggerated LH and FSH response, one male had FSH-only hyperresponsivity and there is one female, as reported by Endo in 1975, with exaggerated LH-only hyperresponsivity – so this points out the problem, if cryptorchidism is responsible for gonadal damage. Eiholzer: Your 3 patients are male. I would guess that the one woman with exaggerated LH response was a midcycle peak. This is, I think, the only explanation, when we think that the exaggerated FSH response is due to cryptorchidism. Grugni: In males, the possibility exists that cryptorchidism causes damage, such as an alteration in the tubulus seminiferus, in Leydig cells and so on. But in females it is difficult to explain, why there is the LH hyperresponsiveness: I think probably the major cause is the hypothalamic dysfunction in the early stage of activation. But I really don’t know the pathogenesis in women. Anon: At what age do you give the adolescents estrogens or testosterone in your clinic? Grugni: There is no standard. I think that we must select the patients. In females after 13 or 14 years of age, near similar to normal subjects and in males after 16 years. But this is not a standard. Some patients need another approach. Also for mental disorders, this is not the same from one patient to the other. B.P. Hauffa, Essen: A hypothalamic dysfunction goes in both directions and we have seen a few female patients with premature onset of puberty. These patients later had an early pubertal arrest. Could you say something with respect to this subgroup? And the second question is, I can very well go along with your notion of arrest of pubertal development on a hypothalamic level. Could you speculate on the mechanisms that lead to this arrest? Grugni: The mechanisms of the arrest of the hypothalamic function. The real mechanism I don’t know. I know, but when we studied the hypothalamic region with positron emission tomography we found a reduced deoxyglucose capacity in this area. I think this is a demonstration of damage during the

Comorbidities or Fundamental Defects of Obesity

174

evolution of this structure. To the first question: In our group, we have not observed precocious puberty and we have only found LH response exaggerated for gender and age, but only seven females had menarche and three of them had irregular spotting. Our patients with exaggerated LH response had a primary amenorrhea. Eiholzer: I like to make a comment on testosterone substitution: I think, it is important to do testosterone substitution in the right biological situation. Because we have spoken today of psychotic problems and we will speak about that afterwards. Testosterone is known to possibly provoke psychotic problems. We have more psychological problems in puberty in normal boys. I think most of medical doctors give testosterone too late in young boys at the age of 20–25 years. And this is a biological situation, in which they are probably not able to deal with pubertal symptoms. And this is the reason I think, if pubertal development is induced or treated with testosterone substitution, it has to be done at 13–15 years. Grugni: I think there is a problem with the age of testosterone treatment. One problem is the peak of bone mineralization, which is at about 18–23 years in normal subjects. I think that in the decision to treat or not to treat male subjects with PWS, we have to take into consideration this fact regarding mineralization. If we treat at 25 or 30 years, it is too late to prevent from osteoporosis. Anon: I think this is an interesting issue. We are studying arguments for replacement therapy in boys with Klinefelter’s syndrome. There is accumulating information in studies showing that if we don’t treat them when they are 11 or 12, we miss the opportunity for improvements in cognitive functions. Now, I would imagine that some of this data might be applicable to Prader-Willi boys too. I don’t know if any studies have been done to look at that issue in PraderWilli syndrome, but there are data in Klinefelter’s. Sizonenko: I would like to make two comments. First is that although we see some patients with FSH and/or LH hyperresponses, perhaps due to cryptorchidism, that appears to me to be due to a hypothalamic defect. However, we haven’t studied this in more detail, or at least I don’t know of any work. Maybe you do. The second point I want to make is: We’re bringing here in this symposium the serotonergic pathway. And I think we should not neglect the serotonergic pathway, because it also has a lot of impact on hunger and so therefore maybe there is also a big disturbance of the serotonergic pathway. And I think that’s maybe for the future to try and pursue that point. Children with Prader-Willi Syndrome and Primary Obesity: A Comparison of Appetite and Psychosocial Profiles. The Family Perspective (J. Gilmour) M. Gellatly, Essex, UK: Thank you for your very interesting paper. I found the results fascinating, but there are some differences from surveys I have

Discussion

175

undertaken with parental members of PWS Association (UK) in 1989 and 1999. A main issue is that we found age of diagnosis is extremely important. Generally, an early diagnosis is consistent with a better ability to keep weight under control. I have also found that pica is grossly underreported by parents. Understandably, the vast majority will not admit that their child/adult is eating nonfood items and it tends to be when they move away from the parental home that other caretakers will report it. Diagnosis can also affect the decision whether to lock cupboards or not. A lot of our families, with a child or adult with PWS, find that once cupboards are locked, the food-seeking behavior actually stops. Individuals have said they feel secure, knowing food is inaccessible, so they stop looking for food. I think there are differences in single-parent families and two-parent families. I don’t know whether you have actually looked at that? Also I’d like to make the observation, also relevant to Dr. Lindgren’s paper this morning, that both of you referred to ‘simple’ obesity in normal people – no obesity is ‘simple’. Finally, I believe there is a need to compare with other learning disability cohorts, rather than with those with normal intellect, because learning disability per se often leads to abnormal behaviors including eating; plus there may be issues in respect of family dynamics. I don’t know whether you have got any comments on these – sorry there are so many points. J. Gilmour, London: I take your points. I think that particular study would benefit from other comparison groups. I think that’s a valid point and I take that point. I think your comments about pica being underreported, I think that’s true, but I think that bias would apply across both groups. So while we would have an underreported level of pica across both groups, I think the fact that the patterns of eating according to the parents is an interesting aspect, as far as the parental description was concerned. Just remind me of another… Gellatly: …Locking cupboards from an early age. The security this provides. Gilmour: That maybe certainly hadn’t been the experience I’ve had with some families. We have to do a different sort of study and look at their family environment in a bit more detail. The child has been motivated to find food – we’ve heard all sorts of interesting stories to get at the food – whether locks are on the cupboard or not. But that may well be a phenomenon which just exists in the current sample that I described. The business of early diagnosis, I think, is a valid point, too. We were not able to look at differences between early and later diagnosis; that may well have an influence on the way the families understand their child’s behavior. But I think it’s interesting that all of the PWS group families did have a diagnosis and clearly understood that the child had PWS and, yet, didn’t necessarily have an understanding that there might be an increased likelihood to feel hungry or at least have a different experience towards food, as compared to the comparison group. I think you raised some interesting points, thank you.

Comorbidities or Fundamental Defects of Obesity

176

Anon: That’s a fascinating approach and I think, more work needs to be done. I wonder what reference point the families are coming from. We get children of course sent to us for validation of short stature and that’s not uncommon. It might be a child who’s on the 25th percentile and the family perceives this as short stature, not because they are comparing them to the charts, but because they are comparing to the brothers and sisters, who are tall. So the reference by which your questions have been asked, I think, would be relevant if they have other children who are normal weight and have more normal eating patterns. That might be a different situation than if there were no other children in the family or if the parents or siblings themselves are obese. Have you made an effort to cross-compare, to look at, single child homes versus multiple child homes and if the parents or other children are overweight or normal? Gilmour: I have actually just collected data where I have asked for a nearest age sibling, to ask exactly that question, because I think that is a key point, making reference to an unaffected sibling in the family. Family characteristics are absolutely important. I hope to be able to answer that question, when I do my data analysis. Horber: You looked at PWS children, which you have genotyped, but the so-called simple obese were probably not genotyped. So if there was in that group some frameshift mutation of leptin, leptin receptor or MCR that would then probably make the comparison difficult. Gilmour: Absolutely. I am very happy to acknowledge that possibility in the data. We asked families to report the number of investigations. If there was anything that indicated to us some sort of syndrome or underlying specific explanation, we excluded them from the sample. But, absolutely, the possibility remains that there are a number of different underlying factors in the simple obesity group. Horber: I think that Prader-Willi may be difficult and complicated, but obesity is certainly not simple… Anon: Could you please comment on the possibilities of bias that would be introduced by recruiting the patients through newspapers in UK? Could you lay out the possibilities for us, what could have happened? Gilmour: I think the bias is most likely to be, and most certainly judging from having telephone contact with these families is simply that they were desperate. They wanted to let someone know about how things were. So I think the bias is more likely to be motivation to tell someone about their circumstances at home. There is a bias clearly of that sort in the recruitment policy. I think it’s potentially a different bias from the clinical sample. Anon: Would you comment a little bit about the fact that these families with PWS were informed of everything? That looks to me like the situation with petit patients, the parents forget with time in terms of the information they

Discussion

177

have been provided and in terms of behavior their children could have. Do you have some information you could give to clinicians? Gilmour: In terms of reminding families? Anon: Support, renewing the information, because you know, families are very often selective and they tend to remember only what they want to remember. Gilmour: I think that’s a fair point. I think there is are a number of things. The families who are part of support groups are often experts. I don’t think that groups of families are likely to need reminding about the features of PWS. In theory, if I asked anyone of the families there in the study, what does PWS mean, I think they would be able to tell me exactly, what it meant. However, from a day by day perspective, I think, they separate out, what they are seeing and what they know in their mind. I think that it could be useful intervention, when families have a medical follow-up, to just try and think about the last time a difficult situation happened and why they think that might be relevant to the Prader-Willi etc. Whitman: I’d like to comment about what you just said: A group of some younger parents of people with PWS a couple of years ago and I gathered together. I later heard back from one of them saying you’re telling me that now is the best there is and it’s downhill all the way… and it’s not so good now. I think when we see the younger parents and they see what’s ahead of them, they do very much remember, but the comparisons that they make are to another Prader-Willi child: My child will not be as bad as that older child now is. And all the way along, you can see them trying to convince themselves that their child, who is in the natural progression for PWS, is gonna be the one that’s different from the rest that are reported. So they really do try to deny what’s in front of them a lot.

Comorbidities or Fundamental Defects of Obesity

178

Comprehensive Treatment Approaches Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 179–189

Does Growth Hormone Affect Morbidities Associated with Obesity in Prader-Willi Syndrome? Aaron L. Carrel, David B. Allen Department of Pediatrics, University of Wisconsin Children’s Hospital, Madison, Wisc., USA

Abstract Children with Prader-Willi syndrome (PWS) display reduced muscle mass (lean body mass) and increased adipose tissue, body composition abnormalities resembling those seen in growth hormone (GH) deficiency. Body composition abnormalities, diminished GH responses to provocative testing, low insulin-like growth factor-1 levels, poor growth velocities, and the presence of other types of hypothalamic dysfunction support the presence of true GH deficiency (GHD) in many children with PWS. GH treatment in these children decreases body fat, and increases muscle mass, fat utilization and energy expenditure. Strength and agility performance are also improved. The metabolic effects, including changes in physical strength and agility, may be the most important features for this particular pediatric population. These improvements are most dramatic during the first year of GH therapy, however prolonged treatment still does not ‘normalize’ these parameters. These observations support a contribution of GHD to disabilities of children with PWS, and a clinically significant benefit of GH treatment. Copyright © 2003 S. Karger AG, Basel

Introduction

Prader-Willi syndrome (PWS) was first described in 1956 [1], with features suggesting hypothalamic dysfunction: hyperphagia, deficient growth hormone (GH) secretion, and hypogonadism. It is now known that PWS is

associated with absent expression of paternal alleles in the PWS region of chromosome 15q11-13 [2]. In ⬃70–75% of cases, this is due to deletion on the paternal copy of chromosome 15. Approximately 25% of cases involve uniparental maternal disomy, in which an individual inherits 2 copies of the maternal chromosome 15 and none of the paternal copy. Rare cases involve translocations, other rearrangements or other molecular defects lacking expression of paternal alleles. Thus, PWS is a manifestation of genomic imprinting; the critical region of chromosome 15 is active only in the paternally inherited chromosome. Although several genes and gene products in the ‘PWS region’ of chromosome 15q11-13 have been identified, the specific genes involved in the pathogenesis of the syndrome are not completely known. Current data suggests a population prevalence of 1 in 10,000–25,000 with no specific sex or race association. Affected children are characterized by obesity, marked hypotonia, short stature, hypogonadism and behavioral abnormalities. Many features of PWS suggest hypothalamic dysfunction, with some, such as hyperphagia, sleep disorders, deficient GH secretion and hypogonadism, having endocrine implications [3]. This article reviews current knowledge regarding GH therapy for morbidities associated with obesity, abnormal body composition, and impaired physical function in children with PWS. Individuals with PWS experience a variety of lifelong morbidities including obesity, hypotonia, osteoporosis, risk for diabetes, and behavioral abnormalities. One of the more debilitating is the severe hypotonia beginning in utero, as evidenced by observations of decreased fetal movement. Severe neonatal hypotonia contributes to feeding problems, failure to thrive and developmental delay. Muscle function may improve with age, but hypotonia is usually a lifelong feature. The best-known feature of PWS is the insatiable appetite and subsequent obesity. It is often stated that PWS is the most frequently occuring identified genetic obesity in humans [4]. Although obesity is usually defined by comparison of weight or weight-derived standards to population norms, it is important to realize that individuals with PWS have increased percentage body fat even with normal or decreased weight [5]. The natural history of individuals with PWS has not yet been comprehensively studied. However, current information documents high rates of morbidity due to obesity-related illnesses: type 2 diabetes, respirator disorders, and cardio-respiratory failure related to obesity and hypotonia. The past decade has seen important developments in the understanding of PWS and the treatment of its morbidities associated with obesity. Much of this work has focused on the role of underlying GH deficiency (GHD) in the pathophysiology, and the effects of GH therapy [6–12]. This disorder has highlighted a broader view of the anabolic ‘nongrowth’ effects of GH therapy in children. Specifically, improvements in physical function, body composition, lipid metabolism, and

Carrel/Allen

180

energy expenditure appear more consequential to these children than changes in height. PWS also represents a phenotype to evaluate many morbidities associated with obesity including insulin sensitivity, strength, agility, and fitness changes, as well as hypothalamic function and energy (both carbohydrate and lipid) metabolism. Evidence of abnormal regulation of energy and body composition is evident during infancy. Neonatal hypotonia in infants with PWS often leads to failure to thrive, yet percent body fat measurements are nonetheless elevated, suggesting early alterations in regulation of body composition even in the absence of obesity. Between the 2nd and 4th year of life in PWS, progressive obesity usually commences primarily due to excessive caloric intake, but also due to decreased energy expenditure and reduced physical activity.

What Is the Role of GH in the Regulation of Obesity and Body Composition?

GH has profound effects on body composition through its anabolic and lipolytic actions. In 1959, Raben and Hollenberg [13] first demonstrated that GH induces lipolysis, and opposed the action of insulin on adipocytes. GH increases fat mobilization by hydrolysis of triglycerides into glycerol and free fatty acids, stimulates fatty acid transportation from adipose tissue to the liver, and inhibits free fatty acid reesterification by adipocytes. These influences of GH on fuel metabolism promote protein conservation at the expense of lipid utilization. This finding provides a cornerstone for the anabolic effects of GH upon body composition. In addition to promoting longitudinal bone growth, GH has important effects on body composition in children. The relationship between GH secretion and body composition is intriguing. GH secretion is suppressed in obese individuals; both fewer secretory events and a shorter half-life of GH compared to that in normal-weight subjects have been demonstrated. The etiology of this finding remains unclear, but these alterations are partly reversed by weight loss [14, 15]. GH secretion is low in PWS, and whether these suppressed GH levels are a cause or result of the obesity in PWS patients is still debated. Support for a role of true GHD includes low insulin-like growth factor-1 (IGF-1) levels, lack of bone age acceleration normally seen with obesity, the body composition resembling classic GHD rather than simple overnutrition, and the exaggerated growth deceleration seen in caloric restriction in PWS. Physiologic effects of GH extend beyond the stimulation of linear growth, and include important metabolic effects upon adipose tissue. GH affects both proliferation and differentiation of preadipocytes, although this varies between

Does GH Affect Morbidities Associated with Obesity in PWS?

181

clonal cell lines and preadipocyte cultures. Both preadipocytes and mature adipocytes possess specific GH receptors. GH may mediate its actions via these receptors, but some effects are indirectly mediated through IGF-1 within adipose tissue. IGF-1 may then act back upon adipose tissue in an autocrine-like manner. GH promotes lipolysis via affecting the balance between lipoprotein lipase (LPL) and hormone-sensitive lipase. GH inhibits LPL, the enzyme that hydrolyzes triglycerides in the circulation to make them available for triglyceride accumulation in adipose tissue. GH also stimulates hormone-sensitive lipase, the rate-limiting step for release of stored triglyceride in adipocytes (lipolysis). PWS is a genetic disorder with predictable features characterized by obesity, hypotonia, short stature, hypogonadism, and behavioral abnormalities. Obesity occurs early in childhood, at a mean age of 2–3 years, resulting from hyperphagia with decreased perception of satiety, and amplified by decreased energy expenditure associated with lower fat-free body mass and reduced physical activity [16]. Hypothalamic dysfunction resulting in impaired GH secretion is also thought to contribute to abnormal body composition, energy expenditure, muscle strength, pulmonary function, and carbohydrate and lipid metabolism seen in PWS patients. The obesity in PWS more closely resembles that seen in GHD [in which total lean body mass (LBM) is decreased, and IGF-1 is relatively decreased] rather than typical ‘exogenous’ obesity (in which both LBM and fat mass are increased and IGF-1 is relatively increased). Thus, PWS patients exhibit a genetically determined propensity for obesity and body composition abnormalities closely resembling GHD. Further, PWS individuals demonstrate other abnormalities that may also be affected by abnormal body composition and GH secretion. Impaired glucose tolerance has been reported in 20% of adults with PWS, while triglyceride and cholesterol levels are reportedly normal. The body composition of PWS resembles that of severely GHD individuals (i.e. reduced LBM and increased fat mass, and reduced resting energy expenditure, REE). This phenotype is clearly distinguishable from the parallel increase in fat and LBM observed in overnourished obese but otherwise healthy individuals. The distinctive replacement of LBM by fat mass in PWS suggests that diminished GH secretion is a primary result of hypothalamic dysfunction rather than suppressed by obesity. Children and adults with GHD demonstrate marked increases in percent body fat, inappropriate reductions in LBM, and subnormal bone mineral density. In particular, excessive fat tissue is distributed in visceral and truncal areas, locations associated with increased cardiovascular risk. Replacement of GH attenuates these abnormalities, though it remains unknown whether it can do so completely. Children with PWS have a high percentage of body fat resembling

Carrel/Allen

182

the GHD state, with a decrease in lean tissue accompanied by decreased strength and agility, and also appear to respond favorably to administration of GH. In these patients, GH has been shown to decrease adipose tissue [i.e. lower respiratory quotient (RQ)], and increase lean body (muscle) mass, fat utilization and oxidation and increase energy expenditure. Effects of GH on adipose tissue metabolism appear to be important throughout life.

The Hypothalamus, PWS, and Obesity

The hypothalamus is at the center of the neuroendocrine hormonal system, including energy balance. Previously, hormonal regulation via classic negative feedback loops acting via pituitary hormone release included only (1) the GHreleasing factor/GH/IGF-1, (2) thyrotropin/thyroid hormone, (3) gonadotropin, and (4) corticotropin-releasing factor. However, the last 5–10 years has witnessed the description of a new type of hypothalamic feedback system including the adipocyte hormone leptin, the ␤-cell hormone insulin, and the newly discovered gastric hormone, ghrelin, which bind to specific receptors within the ventromedial hypothalamus to form the afferent arm of an axis that regulates energy expenditure [17]. PWS infants historically present with failure to thrive and hypotonia for the first 2 years of life, but subsequently demonstrate hyperphagia and progressive obesity. During these not quite distinct phases of development, some changes in hypothalamic regulation appear evident. Our data has revealed evidence of GHD or neurosecretory defect of GH in older children, with low provocative GH values and low IGF-1, while the infants demonstrate more ‘sufficient’ secretion of GH in response to provocative testing. This change in hypothalamic control may also be related to the changes seen in appetite, feeding, and hormonal regulation of energy balance.

Effects of GH Treatment on Body Composition

Administration of GH to GHD children not only restores linear growth, but also promotes growth of LBM, decreases fat mass by increasing fat oxidation and total body energy expenditure, increases bone mineral density following an initial period of increased bone resorption, and improves cardiovascular risk factors. Similarly, children with PWS respond to GH therapy with improvements in body composition. Improvements in body composition and muscle endurance and power were demonstrated in previous uncontrolled short-term GH treatment trials in children with PWS. Recent controlled studies of GH therapy in children with PWS have confirmed improvements in body composition following GH treatment.

Does GH Affect Morbidities Associated with Obesity in PWS?

183

80 70

Body fat (%)

60 50 40 30 20 10 Total controls

Total PWS

Female controls

Female PWS

Male controls

Male PWS

0

Fig. 1. Body composition in PWS (n ⫽ 54).

Pretreatment body composition studies of children with PWS revealed markedly increased percentage body fat [45.2 ⫾ 8.3% compared with healthy children of the same age and sex (16.7 ⫾ 3.5%); p ⬍ 0.0001] (fig. 1). LBM was low (20.5 ⫾ 6.1 kg, 50%) compared with a normal mean LBM of ⬃80% in age-matched healthy non-PWS controls. Over a period of 12 months, mean percentage body fat decreased by 8% overall (46.3 ⫾ 5.8 to 38.3 ⫾ 10.7%, p ⬍ 0.01) in GH-treated children, whereas no change was seen in the nontreated control PWS patients (42.6 ⫾ 8.1 to 45.8 ⫾ 8.8%; p ⫽ nonsignificant). LBM increased with GH treatment (to 25.6 ⫾ 4.3 kg, p ⬍ 0.01), but remained unchanged in control children (21.7 ⫾ 5.0 kg, p ⫽ nonsignificant). Similar body composition changes have been reported in response to 12–48 months of GH treatment by other investigators [12, 18, 19]. These remarkable changes in body composition attenuate, but do not regress, during more prolonged GH therapy at a dose of 1 mg/m2/day [19]. Recently reported data show that, in contrast to the marked reductions observed during the first 12 months of GH treatment, percentage body fat remains stable during 24–48 months of GH treatment (40.3 ⫾ 10.0%, p ⫽ nonsignificant versus 12-month measurement; p ⬍ 0.001 vs. baseline). Importantly, LBM, which increased significantly after 12 months of GH (25.2 ⫾ 6.9 vs. 22.9 ⫾ 15.7 kg, p ⬍ 0.01) increased further during months 24–48 (28.5 ⫾ 7.2 kg at 24 months, p ⬍ 0.01). Also demonstrated by figures 2 and 3 is the dose-response relationship seen with GH upon body composition. Using doses of GH comparable to

Carrel/Allen

184

Body fat by DEXA (%)

50

*

* *

40

*

30 Pre-GH therapy

1

0.3 mg/m2/day 1.0 mg/m2/day

*

*

2 Years of GH therapy

3

1.5 mg/m2/day Controls

4

Baseline

Fig. 2. Long-term GH therapy in PWS. Change in body fat. *p ⬍ 0.05 vs. baseline. DEXA ⫽ Dual-energy X-ray absorptiometry.

40

*

LBM by DEXA (kg)

35

* 30

* *

25

20 Pre-GH therapy 0.3 mg/m2/day 1.0 mg/m2/day

1

2 Years of GH therapy 1.5 mg/m2/day Controls

3

4

Baseline

Fig. 3. Long-term GH therapy in PWS. Change in LBM. *p ⬍ 0.05 vs. baseline. DEXA ⫽ Dual-energy X-ray absorptiometry.

Does GH Affect Morbidities Associated with Obesity in PWS?

185

doses used in adult GHD do not retain the beneficial changes in body fat (fig. 2) nor in accrual of LBM (fig. 3). Doses of 1 or 1.5 mg/m2/day, however, demonstrate a sustained loss of adipose tissue (percent body fat) and continued increases of LBM with prolonged treatment at 4 years.

Effect of GH on Energy Expenditure and Fat Utilization

Given their reduced LBM, children with PWS would be expected to demonstrate markedly reduced REE. Before GH treatment, children with PWS showed reduced REE compared with predicted values for healthy children matched for surface area (22.4 ⫾ 4.4 vs. 43.6 ⫾ 3.2 kcal/m2/h; p ⬍ 0.0001). A GH-induced increase in LBM would be expected to increase REE. A trend toward increased REE was observed after 12 months of GH therapy, and changes in REE reached statistical significance after 24 months, compared with patients’ own baseline measurements. While it is probable that GH effects on LBM accretion had positive effects on REE, it should also be noted that REE normally increases as children grow. However, no change in REE was demonstrated during the 12 months of control period without GH treatment [12]. GHD is associated with lipogenesis and fat storage predominating over the accretion of lean mass, even in the absence of overt obesity. Preference for fat utilization as an energy source is reflected in a reduction of RQ. The RQ normally ranges from 0.7 (strong predominance of fatty acid oxidation) to 1.0 (exclusive oxidation of carbohydrate) to ⬎1.0 (indicating lipogenesis from carbohydrate). GH treatment in children with PWS was associated with a decrease in RQ values (p ⬍ 0.05), indicating increased utilization of fat for energy, while RQ increased in the untreated controls. Thus, compared with non-GH-treated PWS controls, GH-treated PWS patients demonstrated a shift in energy use towards oxidation of fat, coinciding with reductions in fat mass shown by dualenergy X-ray absorptiometry scanning.

Effects of GH on Strength and Agility

Substantial evidence is accumulating which indicates improvements in body composition and linear growth in children with PWS following 12–48 months of GH therapy. Perhaps of greatest importance to patients and their families, however, is the hope that GH therapy could improve the child’s physical strength, activity and ability. Early reports included anecdotal accounts of dramatic gains in physical activity, and many parents also claimed striking improvements in physical stamina, strength and agility. Specifically, these

Carrel/Allen

186

included new gross motor skills (e.g. independently climbing up the steps of the school bus, carrying a gallon carton of milk at the grocery store, participating in a normal gym class without restrictions, and being able to join a karate class). The authors’ research has included objective measures of changes in physical function during GH treatment, including a timed run, sit-ups, and weight lifting. Improvements in running speed, long jump, sit-ups and arm curls after 12 months of GH treatment compared with controls were documented [5]. Following 48 months of GH treatment, improvements in long jumping and situps were maintained, while further improvement was seen in running speed and arm curls. In spite of these gains in physical function, children with PWS still scored well below 2 SDS compared with healthy children for all parameters studied. While the lack of a blinded, placebo-controlled study design admittedly weakens the scientific validity of these findings, they do suggest that measured improvements in strength and agility were associated with ‘real-life’ functional benefit to the children and their families. The clinical effects of GH in children with PWS have been carefully documented over the past few years. While the dramatic effects on physical agility and strength correlate with changes seen in body composition, however, the agility achieved remains far from standard for age-matched children. Whether earlier intervention with GH may improve these measurements is currently being studied.

Safety of GH Treatment in PWS

Theoretical concerns of GH therapy particularly applicable to PWS include the risk of scoliosis, an already increased concern in this population, and insulin resistance. No significant adverse side effects have been observed by the present authors or others during 48 months of GH treatment [12, 18, 19]. Serum glucose and insulin levels obtained during an oral glucose tolerance test were slightly, but nonsignificantly, higher than baseline levels during GH treatment. Two children of 54 original subjects treated with PWS did develop mild, reversible pseudotumor cerebri. Of esthetic concern, the lower facial height changes with an SDS of 1.2 at 24 months, supporting recent findings of accentuated growth of the high midface during GH therapy in PWS.

Conclusion

This past decade has seen significant advances in the understanding of PWS and in treatments to alleviate its distressing complications. Multiple

Does GH Affect Morbidities Associated with Obesity in PWS?

187

controlled randomized trials have documented the benefits of GH therapy in children with PWS upon body composition, strength and agility, energy expenditure, and linear growth. Arguably, these responses have more clinical impact in these children than changes in linear growth. The response of children with PWS to GH therapy is greatest during the first 12 months for all effects, including growth rate, decrease in body fat, increase in fat utilization, improvement in physical function, and laboratory alterations in carbohydrate and lipid metabolism. This reduction in response to GH is well documented in virtually all disorders treated with prolonged GH therapy. In spite of these encouraging results, GH therapy thus far has not ‘normalized’ body composition in PWS. This could reflect the influence of (1) non-GH factors regulating body composition affected by the genetic mutation causing PWS, or (2) the relatively late initiation of GH therapy following a critical period of abnormal adipose and muscle deposition in infancy. This relatively late initiation of GH replacement (i.e. 4–5 years of age) may occur after a ‘body composition template’ has occurred. The main function of mature adipocytes is accumulation of lipid in the fed state, and release of energy during exercise or fasting states. Since adipose tissue has a limited ability for de novo synthesis of free fatty acids, accumulation of triglycerides is dependent upon LPL, the enzyme primarily responsible for hydrolyzing triglycerides to free fatty acids, which can be transported into adipose tissue. GH produces a pronounced inhibition of adipose tissue LPL, while insulin and glucocorticoids stimulate LPL activity [20]. GH also inhibits the proliferation and differentiation of preadipocytes (immature cells that do not store triglycerides) into mature adipocytes. After the age of 4–5 years in children with PWS, the presence of increased numbers of mature adipocytes may provide a template less susceptible to the effects of GH on preadipocytes, inhibiting adipose accumulation. Studies of earlier initiation of GH therapy are in progress to address these possibilities. Acknowledgments This work was supported in part by NIH Grant M01 RR03186-13S1, as well as funding from Pharmacia and the Genentech Foundation for Growth and Development.

References 1 2 3

Prader A, Labhart A, Willi H: Ein Syndrom von Adipositas, Kleinwuchs, Kryptorchismus und Oligophrenie. Schweiz Med Wochenschr 1956;86:1260–1261. Butler MG, Thompson T: Prader-Willi syndrome: Clinical and genetic findings. Endocrinologist 2000;10(suppl 1):3S–16S. Ritzen EM, Bolme P, Hall K: Endocrine physiology and therapy in Prader-Willi syndrome; in Cassidy SB (ed): Prader-Willi Syndrome and Other 15q Deletion Disorders. NATO ASI Series. Berlin, Springer, 1992, vol H61, pp 153–169.

Carrel/Allen

188

4 5 6

7

8 9 10

11 12 13 14 15 16

17 18 19

20

Cassidy SB: Prader-Willi syndrome. J Med Genet 1997;34:917–923. Brambilla P, Bosio L, Manzoni P: Peculiar body composition in patients with Prader-LabhartWilli syndrome. Am J Clin Nutr 1997;65:1369–1374. Lindgren AC, Ritzen EM: Five years of growth hormone treatment in children with Prader-Willi syndrome. Swedish National Growth Hormone Advisory Group. Acta Paediatr Suppl 1999; 88/433:109–111. Eiholzer U, Bachmann S, l’Allemand D: Is there growth hormone deficiency in Prader-Willi syndrome? Six arguments to support the presence of hypothalamic growth hormone deficiency in PWS. Horm Res 2000;53(suppl 3):44–52. Davies PS: Growth hormone therapy in Prader-Willi syndrome. Int J Obes Relat Metab Disord 2001;25/1:2–7. Burman P, Ritzen EM, Lindgren AC: Endocrine dysfunction in Prader-Willi syndrome: A review with special reference to GH. Endocr Rev 2001;22:787–799. Carrel AL, Myers SE, Whitman BY, Allen DB: Growth hormone improves body composition, fat utilization, physical strength and agility, and growth in Prader-Willi syndrome: A controlled study. J Pediatr 1999;134:215–221. Carrel AL, Myers SE, Whitman BY, Allen DB: Sustained benefits of growth hormone in children with Prader-Willi syndrome are dose-dependent. J Pediatr Endocrinol Metab 2001;14:1096–1105. Carrel AL, Myers SE, Whitman BY, Allen DB: Benefits of long-term growth hormone therapy in Prader-Willi syndrome: A 4-year study. J Clin Endocrinol Metab 2002;87:1581–1585. Raben M, Hollenberg CH: Effect of growth hormone on plasma fatty acids. J Clin Invest 1959; 38:484–488. Dieguez C, Page MD, Scanlon MF: Growth hormone neuroregulation and its alterations in disease states. Clin Endocrinol (Oxf) 1988;28:109–143. Dieguez C, Casanueva FF: Influence of metabolic substrates and obesity on growth hormone secretion. Trends Endocrinol Metab 1995;6:55–59. van Mil EAG, Westerterp KR, Gerver WJ, et al: Energy expenditure at rest and during sleep in children with Prader-Willi syndrome is explained by body composition. Am J Clin Nutr 2000;71: 752–756. Lustig RL: Hypothalamic obesity: The sixth cranial endocrinopathy. Endocrinologist 2002;12: 210–217. Lindgren AC, Hagenas L, Ritzen EM: Growth hormone treatment of children with Prader-Willi syndrome: Effects on glucose and insulin homeostasis. Horm Res 1999;51/4:157–161. Eiholzer U, l’Allemand D: Growth hormone normalizes height, prediction of final height, hand length in children with Prader-Willi syndrome after 4 years of therapy. Horm Res 2000; 53/4:185–192. Richelson B: Actions of growth hormone in adipose tissue. Horm Res 1997;48/5:105–110.

Dr. A.L. Carrel, Department of Pediatrics, University of Wisconsin Hospital and Clinics, 600 Highland Avenue, Madison, WI 53792 (USA) Tel. ⫹1 608 265 8182, Fax ⫹1 608 265 7957, E-Mail [email protected]

Does GH Affect Morbidities Associated with Obesity in PWS?

189

Comprehensive Treatment Approaches Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 190–197

Role of Diet and Upbringing in Young Children with Prader-Willi Syndrome Dagmar l’Allemand, Sara Bachmann, Jürgen Grieser, Urs Eiholzer Foundation Growth Puberty Adolescence, Zurich, Switzerland

Abstract While growth hormone (GH) therapy improves growth and body composition in children with Prader-Willi syndrome (PWS), the basic disturbance of appetite regulation still persists. We analyzed factors to prevent obesity and examined genetic disposition for overweight, nutritional intake and style of upbringing in families of normal weight children with PWS and healthy controls. In a cross-sectional study, 18 young, nonobese children (age 4.4 ⫾ 2.2 years) with PWS were investigated for the BMI of family members as well as for nutrition and activity after 3 years of GH therapy. Nutritional intake and time spent on physical activity were assessed by the parents by means of 3-day reports. The style of upbringing was analyzed by means of modified standardized questionnaires (DEF, HAMEL) with questions on consistency in general upbringing and on control of nutrition. All results were compared to a control group of 19 age-, height- and weight-matched healthy children, and the influence of the examined factors on weight for height (WfH) was analyzed. Scores indicating the hereditary risk of obesity and time spent on physical activities were similar in both groups. Total energy and fat intake were significantly lower in children with PWS in comparison to the controls. Parents of children with PWS were more consistent both in their control of nutrition and in their upbringing in general. WfH of children with PWS was negatively correlated with consistency in general, while eating discipline was not correlated with WfH in PWS. Among healthy controls, only hereditary disposition to obesity was correlated with actual WfH. Our results indicate that in healthy children with normal weight, energy intake and energy expenditure are well balanced without exogenous intervention, but the set point of weight control is genetically determined. In PWS, however, due to the impaired regulation of appetite, normal weight and adequate nutrition intake can only be achieved through external control. Instead of simple food restriction, a consistent upbringing appears to be the most promising way forward. Such counselling for educational behavior might also be an efficient way to prevent obesity in families with a genetic risk of nonsyndromal obesity. Copyright © 2003 S. Karger AG, Basel

Introduction

The beneficial effects of growth hormone (GH) therapy on body composition and metabolism in Prader-Willi syndrome (PWS) have been clearly demonstrated [1–5]. The basic disturbance of appetite regulation, however, persists [2]. We wanted to identify factors other than GH therapy, such as diet and upbringing, that might be implicated in weight control in this form of genetic obesity. Recently, it has been suggested that prevention of simple obesity might be possible in children and their families [6] through counselling for a healthy lifestyle. The present study aims at analyzing factors in the environment of children with PWS, which slow down the development of the genetically caused obesity. Therefore, we examined young children with PWS during their phase of normal weight. We addressed the following issues: (1) the disposition for overweight in the families, (2) the nutritional intake of the patients and (3) the parents’ style of bringing up their child. The most important aim was to identify factors which protect against weight gain. Ultimately, these findings should be useful to work out strategies for obesity prevention, both in healthy children and in children with genetically caused obesity.

Methods Therefore, we designed a controlled cross-sectional study, involving 18 young, nonobese children with PWS and 19 healthy children matched for age, height and weight (table 1). These controls were recruited from families and friends of the investigators. In all individuals with PWS, the diagnosis had been confirmed genetically, documented by deletion or uniparental disomy of chromosome 15. Children with PWS were in a steady state of growth and metabolism during about 3 years of GH therapy, and they were examined and treated according to the PWS study design published beforehand [1]. For all comparisons, weight for height (WfH) was expressed as standard deviation score (SDS), adjusting for gender and age [7]. WfH SDS was slightly higher in the PWS group than in the controls, but not significantly so. To assess the hereditary disposition to obesity, we asked parents and grandparents to report their BMI and their tendency to gain weight as well as comorbidities such as arterial hypertension and type 2 diabetes. All these data were transferred into a score ranging from 2 to 12 points. Nutritional intake and the levels of physical activity were assessed by the parents by means of 3-day reports. The average nutrient composition of diets was calculated according to Swiss [8] and German nutrient tables [9, 10]. Energy intake was related to height and not to weight. This is recommended in children with PWS [11], because they have an increased fat mass, and height better reflects lean mass, the metabolically active compartment. Physical activity over 3 days was rated and expressed as a score, ranging from 144 to 576 points, as previously described [12].

Diet and Upbringing in PWS

191

Table 1. Patients

Total number, n Females, n Age, years Height, cm Weight, kg WfH (SDS) Duration of GH therapy, years

PWS

Controls

18 9 4.42 ⫾ 2.20 103.02 ⫾ 15.93 18.02 ⫾ 5.84 0.73 ⫾ 1.07 2.97 ⫾ 1.75

19 8 4.09 ⫾ 1.91 104.6 ⫾ 14.62 17.47 ⫾ 4.96 0.27 ⫾ 1.03 –

p

n.s. n.s. n.s. n.s.

Values represent means and standard deviation. The Mann-Whitney test was used for statistical evaluation. n.s. ⫽ Not significant.

The parents’ style of bringing up their child was analyzed by means of a selfconstructed questionnaire which contained items of two German standardized questionnaires (DEF ⫽ Diagnostischer Elternfragebogen [13], HAMEL ⫽ Hamburger Erziehungsverhaltensliste für Mütter [14]). Some questions were modified and expanded to assess nutritional control and eating habits. There were also questions about the perception of eating as a problem. For each group of items there was a specific score, expressed as percent of the maximum score that could be attained.The main evaluation parameter was WfH SDS. Both groups were compared by the Mann-Whitney U test. The influence of the observed factors on WfH SDS was separately examined in each group by linear regression analyses (Spearman’s rank correlation test). p values below 0.05 were considered significant.

Results and Discussion

The scores indicating the risk of obesity in family members were similar in both groups (fig. 1). Moreover, they were in the range of healthy subjects without an increased risk of comorbidities. Only in healthy controls was the hereditary disposition to obesity significantly correlated with the current WfH of the child (r ⫽ 0.47, p ⬍ 0.05). In fact, the set point of weight control is genetically determined [15]. The dietary reports in the healthy Swiss children showed that their energy and nutrient intake was within the normal range [16], but protein intake was high (fig. 2). In children with PWS, the caloric intake was in the recommended range for these patients, on average 10.7 kcal/cm/day [11]. In this way, the energy intake was significantly reduced by about 25% compared to healthy control children: The proportion of nutrient fat was impressively low with a mean of 25% compared to 33% in the controls.

l’Allemand/Bachmann/Grieser/Eiholzer

192

PWS Controls

Fig. 1. Hereditary disposition to obesity (mean and SD) displayed in percent of the maximum score in parents and grandparents of 18 children with PWS and of 19 controls matched for age, height and weight. n.s. ⫽ Not significant by Mann-Whitney U test. See Methods for further explanations.

n.s.

0

20

40

60

80

100

% of maximum score

4

45

16 3.5

**

40

*

35

3

10 8 6

2.5 2

20 15

4 2

0.5

10 5

0 PWS Controls

25

1.5 1

0

**

30 Fat (%)

12 Protein (g/kg/day)

Energy/height (kcal/cm/day)

14

0 PWS Controls

PWS Controls

Fig. 2. Energy and macronutrient intake (means and SD) of 18 children with PWS and of 19 control children. Significant differences by Mann-Whitney U test: *p ⬍ 0.05; **p ⬍ 0.01. The recommended intake for healthy Central European children [16] is indicated by a line.

From these results, we conclude that – even during GH treatment – children with PWS can only keep their weight stable if they reduce their energy intake to about 75% of what is recommended for healthy children. It might be suggested that food intake was underreported by the families at risk of obesity. Underreporting is more frequent in obese subjects, as identified by a comparison with the doubly labelled water method [17–19]. Nevertheless, the stable weight course of the children with PWS, which remains within the normal range [20], substantiates the accuracy of recording by the parents. In both groups, the intake of energy, related to height, or fat in percent of total energy was not correlated to WfH SDS (table 2). Thus, WfH is

Diet and Upbringing in PWS

193

Table 2. Correlations between nutritional intake or upbringing and WfH WfH SDS PWS

Controls

Energy/height Protein Fat

n.s. r ⫽ ⫺0.52* n.s.

n.s. n.s. n.s.

Eating as a problem Eating discipline Consistency of upbringing Rigorous and punitive upbringing

n.s. n.s. r ⫽ ⫺0.49* n.s.

n.s. n.s. n.s. n.s.

p ⬍ 0.05, Spearman’s rank correlation test. * ⫽ Significant.

PWS Controls

***

Eating as a problem

*

Eating discipline

*

Consistency of upbringing

**

Rigorous and punitive upbringing 0

20

40

60

80

100

% of maximum score

Fig. 3. Eating and style of upbringing (means and SD) displayed in percent of the maximum score of each group of items in families of 18 children with PWS and of 19 controls. Significant differences by Mann-Whitney U test: *p ⬍ 0.05; **p ⬍ 0.01; ***p ⬍ 0.001. See Methods for further explanations.

independent of the reported nutrition. Only in children with PWS did we observe that a low protein content of nutrition was correlated to a higher WfH. A low protein content of the diet leads to relatively higher proportions of fat and carbohydrate intake, these substrates being more efficient to supply energy and to accumulate fat. After having seen the limited role of diets in weight control, the influence of upbringing is analyzed (fig. 3). First, eating is perceived to be an important problem in families affected by PWS, because the abnormal appetite control

l’Allemand/Bachmann/Grieser/Eiholzer

194

causes hyperphagia in this syndrome and potentially leads to disabling obesity [21–23]. Moreover, motivation always plays a key role in a successful weight control management [24]. The parents of children with PWS are more consistent both in their control of eating and in the general style of bringing up their children (fig. 3). A restrictive diet as previously demonstrated can only succeed with consistency and reliability in the upbringing in general. In contrast, parents of children with PWS stated that they resort to rigorous disciplinary measures and punishment less often than parents of healthy children. Concerning the relationship between the style of upbringing and weight on an individual basis (table 2), there is a negative correlation between the consistency in the general upbringing and weight control in PWS: a less consistent upbringing leads to heavier children. For weight control, eating discipline and punishment do not seem to play a major role. In our opinion, these are the most important results of our the study. As also described by other authors, a successful weight control is related to self-efficacy and parenting skills in families of children with simple obesity [25–27]. Because of disturbed satiation, children with PWS seem to be helpless victims of their appetite drive. Therefore, external control of nutrition by parents and other caregivers is indispensable [28]. In contrast to PWS, there is no relationship at all between upbringing and weight in the healthy control children. The energy balance in healthy normal-weight children does not require outside control. In both groups, a similar amount of time was spent on sports activities (249 vs. 253 points per 3 days, nonsignificant). In older children with PWS, however, hypoactivity was clearly demonstrated [12, 29, 30]. In fact, these findings confirm that parents of young children with PWS may still be able to promote a healthy lifestyle as long as the children are young and compliant. In this line, it has been demonstrated that early intervention reduces the risk of nonsyndromal obesity not only in animals [31], but also in children [6].

Conclusions

In healthy children, weight is only associated with the hereditary disposition. The present results confirm that energy intake and energy expenditure are well balanced without exogenous intervention, but within a genetic set point of weight control. In young children with PWS, however, normal weight is maintained through external control, due to the impaired regulation of appetite. These families have achieved a reduction of energy intake in their children and adopted a

Diet and Upbringing in PWS

195

healthy lifestyle. It has to be pointed out that a consistent upbringing correlates better with lower weight in children with PWS than food restriction. The most successful strategy of obesity prevention in this form of genetically determined obesity seems to be the counselling in general issues of upbringing, but not a restrictive attitude towards food. Perhaps this also offers a benefit for the prevention of simple obesity. These general suggestions for efficient weight management may sound simple, but they have got lost in the last decades. During this time, the prevalence of nonsyndromal obesity has dramatically increased. To promote a healthy lifestyle and education is a trivial job, but it has to be done!

Acknowledgment This study was supported by the Swiss National Science Foundation grant No. 32.056063.98 and the Swiss Academy of Medical Sciences grant No. BK 286/98.

References 1

2

3

4

5

6 7 8 9 10 11 12

Eiholzer U, l’Allemand D: Growth hormone normalises height, prediction of final height and hand length in children with Prader-Willi syndrome after four years of therapy. Horm Res 2000; 53:185–192. Eiholzer U, l’Allemand D, van der Sluis I, Steinert H, Ellis K: Body composition abnormalities in children with Prader-Willi syndrome and long-term effects of growth hormone therapy. Horm Res 2000;53:200–206. l’Allemand D, Schlumpf M, Torresani T, Girard J, Eiholzer U: Insulin secretion before and under 3 years of growth hormone (GH) therapy in Prader-Willi syndrome (PWS). Exp Clin Endocrinol Diabetes 2000;108:127. l’Allemand D, Eiholzer U, Schlumpf M, Steinert H, Riesen W: Cardiovascular risk factors improve under 3 years of growth hormone therapy in Prader-Willi syndrome. Eur J Pediatr 2000; 159:835–842. Myers SE, Carrel AL, Whitman BY, Allen DB: Sustained benefit after 2 years of growth hormone on body composition, fat utilization, physical strength and agility, and growth in Prader-Willi syndrome. J Pediatr 2000;137:42–49. Müller M, Mast M, Asbeck K, Langnäse K, Grund A: Prevention of obesity – Is it possible? Obes Rev 2001;2:15–28. Prader A, Largo R, Molinari L, Issler C: Physical growth of Swiss children from birth to 20 years of age. Helv Paediatr Acta 1989(suppl 52):1–125. Hinsen H, Rechsteiner M, de Rham O, Schmocker B: Nährwerttabellen für Konsumentinnen und Konsumenten. Bern, Schweizerische Vereinigung für Ernährung, 1993. Ernährungsberatung Nestlé D: Kalorien mundgerecht. Frankfurt, Umschau Verlag, 1995. Elmadfa I, Aign W, Muskat E, Fritzsche D: Die grosse GU Nährwert-Tabelle. München, GU, 1999. Stadler DD: Nutritional management; in Greenswag LR, Alexander R (eds): Management of Prader-Willi Syndrome. New York, Springer, 1995, pp 88–114. Eiholzer U, Nordmann Y, l’Allemand D, Schlumpf M, Schmid S, Kromeyer-Hauschild K: Improving body composition and physical activity in Prader-Willi syndrome. J Pediatr 2003;142: 73–78.

l’Allemand/Bachmann/Grieser/Eiholzer

196

13 14 15 16 17 18 19 20 21

22 23 24 25

26

27 28 29 30

31

Dehmelt P, Kuhnert W, Zinn A: Diagnostischer Elternfragebogen. Weinheim, BeltzTest, 1989. Baumgärtel F: Hamburger Erziehungsverhaltensliste für Mütter. Göttingen, Verlag für Psychologie, 1979. Barsh GS, Farooqi IS, O’Rahilly S: Genetics of body weight regulation. Nature 2000;406: 644–651. DACH Empfehlungen: Referenzwerte für die Nährstoffzufuhr. Bern, Umschau-Braus, 2000. Westerterp KR, Verboeket-van de Venne WPHG, Meijer GAL, Hoorten F: Self-reported intake as a measure for energy intake. A validation against doubly labelled water. Obes Eur 1991;91:17–22. Goris AH, Westerterp MS, Westerterp KR: Undereating and underrecording of habitual food intake in obese men: Selective underreporting of fat intake. Am J Clin Nutr 2000;71:130–134. Bandini LG, Schoeller DA, Cyr HN, Dietz WH: Validity of reported energy intake in obese and nonobese adolescents. Am J Clin Nutr 1990;52:421–425. Eiholzer U, Schlumpf M, Nordmann Y, l’Allemand D: Early manifestations of Prader-Willi syndrome: Influence of growth hormone. J Pediatr Endocrinol Metab 2001;14:1441–1445. Holland AJ, Treasure J, Coskeran P, Dallow J, Milton N, Hillhouse E: Measurement of excessive appetite and metabolic changes in Prader-Willi syndrome. Int J Obes Relat Metab Disord 1993;17:527–532. Lindgren AC, Barkeling B, Hagg A, Ritzen EM, Marcus C, Rossner S: Eating behavior in PraderWilli syndrome, normal weight, and obese control groups. J Pediatr 2000;137:50–55. Cassidy SB: Prader-Willi syndrome. J Med Genet 1997;34:917–923. Williams GC, Grow VM, Freedman ZR, Ryan RM, Deci EL: Motivational predictors of weight loss and weight-loss maintenance. J Pers Soc Psychol 1996;70:115–126. Vahabzadeh Z: Familien übergewichtiger Kinder: Probleme und Bedürfnisse; in Herrmann A, Schürmann I, Zaumseil M (eds): Chronische Krankheit als Aufgabe. Fortschritte der Gemeindepsychologie und Gesundheitsförderung. Tübingen, dgvt-Verlag, 2001, vol 7, pp 211–242. Gilmour J, Skuse D: Children with Prader-Willi syndrome and primary obesity: A comparison of appetite and psychosocial profiles – The family perspective; in Eiholzer U, I’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 156–165. Robinson TN: Behavioural treatment of childhood and adolescent obesity. Int J Obes Relat Metab Disord 1999;23(suppl 2):S52–S57. Eiholzer U: Prader-Willi Syndrome. Effects of Human Growth Hormone Treatment. Endocr Dev. Basel, Karger, 2001, vol 3. Davies PS, Joughin C: Using stable isotopes to assess reduced physical activity of individuals with Prader-Willi syndrome. Am J Ment Retard 1993;98:349–353. van Mil EG, Westerterp KR, Kester AD, Curfs LM, Gerver WJ, Schrander-Stumpel CT, Saris WH: Activity related energy expenditure in children and adolescents with Prader-Willi syndrome. Int J Obes Relat Metab Disord 2000;24:429–434. Rosenbaum M, Leibel R: The physiology of body weight regulation: Relevance to the etiology of obesity in children. Pediatrics 1998;101:525–539.

Dr. med. Dagmar l’Allemand, Foundation Growth Puberty Adolescence, Möhrlistrasse 69, CH–8006 Zürich (Switzerland) Tel. ⫹41 1 3643700, Fax ⫹41 1 3643701, E-Mail [email protected]

Diet and Upbringing in PWS

197

Comprehensive Treatment Approaches Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 198–210

Prader-Willi Syndrome: A Pervasive Neurodevelopmental Disorder Requiring a Multidisciplinary Care Approach Barbara Y. Whitman Department of Pediatrics, St. Louis University School of Medicine, St. Louis, Mo., USA

Abstract Following the original description of Prader-Willi syndrome (PWS) in 1956, this complex disorder has baffled the most seasoned clinician and in the year 2002 continues to unfold as one of the most complex syndromes genetically, physiologically, and behaviorally. At the same time, PWS remains one of the most challenging syndromes to manage. The cardinal central hyperphagia, the accompanying early and chronic morbid obesity combined with pathologic short stature and decreased metabolic rates remain completely refractory to pharmacotherapy. Age-related, increasingly severe, behavior difficulties present similar pharmacotherapy challenges; most affected individuals prove either unresponsive or have significant negative responses to most psychotropic medications. The multidimensional problems of patients with PWS require a multidisciplinary team approach to management. The primary goal of medical treatment is normalization of weight for height. Comprehensive medical care may include growth hormone therapy for children with a documented growth hormone/IGF axis deficiency and short stature and in adults who meet current criteria for growth hormone deficiency; planned weight management programs that include dietary supervision, caloric restrictions, and nutritional planning and education for parents and caregivers; physical therapy for muscle hypotonia; and appropriate behavioral and psychotropic interventions. The multiple cognitive and behavioral dysfunctions associated with PWS shift and escalate over time, requiring lifelong management. This paper will review the agerelated cognitive and behavioral management needs and the use of a multidisciplinary team management approach as a response to those needs. Copyright © 2003 S. Karger AG, Basel

Introduction

First described in 1956, Prader-Willi syndrome (PWS) continues to unfold as one of the most complex and puzzling disorders genetically, physiologically, and behaviorally. Management of the syndrome is equally puzzling and complex. This complexity is driven by two issues: (1) hyperphagia – the cardinal defining feature – remains unresponsive to any pharmacotherapy and (2) the accompanying, age-related, increasing severe, behavioral component similarly presents unique management and pharmacologic challenges. PWS is characterized in most medical texts as an obesity syndrome of genetic origin that also includes some level of cognitive impairment. Clearly, lifelong morbidities and reduced life expectancy can result if the hyperphagia is poorly managed resulting in early, chronic, morbid obesity. And, clearly the accompanying pathologic short stature and decreased metabolic rates further complicate obesity management. Additionally, most medical texts state that the observed abnormalities result from a primary disturbance of the brain above the spinal cord, with particular impact on the hypothalamic-pituitary axis. A comparison of the nonhormonal role of the hypothalamus and some characteristics of persons with PWS highlights that relationship (table 1). This description, however, fails to recognize the central role of the accompanying behavioral components in defining the syndrome and in the management and adjustment of affected individuals. Current evidence suggests that PWS is best characterized as a pervasive neurodevelopmental disorder whose behaviors reflect a distributed central nervous system dysfunction that has yet to be fully described either anatomically or biochemically [1, 2]. In this view, the centrally driven food-related behavioral constellation, albeit dramatic, is just one of a list of neurobehavioral abnormalities attendant to this disorder – and when appropriately addressed, perhaps the easiest of the behavioral abnormalities to manage. This can be seen in the evolution of ‘the central issues’ facing families with an affected child and the primary interventions needed. Four distinct ‘eras’ of primary concern are identified: (1) hypotonia, (2) speech and language, (3) food-related behavior constellation, and (4) behavior. These eras are not mutually exclusive; hypotonia remains an issue throughout the life span of the affected individual, but its centrality as an issue is primarily in the infant and toddler ages. The management needs in each era must additionally consider and support the immediate and long-term role of the family in management and be ready to respond to the reflexive impact of poor management on family functioning.

PWS: A Neurobehavioral Disorder

199

Table 1. Comparison of normal hypothalamic functions with apparent hypothalamic dysfunction found in PWS Hypothalamus

PWS characteristics

Regulates appetite Sensitivity to pain

Hyperphagia Altered pain sensitivity, skin picking Temperature instability Altered sleep/wake cycles Emotional excesses Central apnea Short-term memory deficits

Regulates temperature Regulates day/night cycle Regulates emotions Regulates breathing Impacts memory

[Adapted from J. Hanchette by personal submission.]

Hypotonia

Most infants with PWS present with profound hypotonia, currently thought to result both from central abnormalities and an abnormal body composition that includes a significantly reduced lean body mass even in the most underweight infant. This hypotonia significantly impacts an infant’s ability to suck and thrive. Oral motor difficulties including difficulty swallowing are frequent. Many infants require gavage feeding or other special feeding techniques. Extreme lethargy, with decreased arousal and weak cry, are also prominent symptoms, often leading to the need to awaken the child to feed. Reflexes may be decreased or absent. This picture is frequently complicated by aspiration pneumonias resulting both from the difficulty swallowing, a muscular inability to handle normal infantile reflux, and occasionally induced by poor feeding techniques. Immediate management issues may include special instruction in feeding methods, superconcentrated nutritional input, and occupational therapy to address the oral motor concerns. Pictures of infants in this stage are striking in the degree of exhaustion noted in the eyes of many of these youngsters. Musculoskeletal problems including delayed motor milestones, scoliosis and kyphosis accompany the hypotonia. The average age of sitting is 12 months and for walking 24 months. Scoliosis, once thought to develop in later childhood, is now known to be a problem for many children as early as 12 months of age. The combination of reduced and hypotonic muscular tissue necessitates the need for early and continuing physical therapy. Additionally, scoliosis, even as significant as 25–30°, is often undetectable by routine examination, prompting a recommendation for yearly spine films.

Whitman

200

The Family Crisis support, genetic counseling, and education for the family in understanding and coping with the diagnosis is critical during these early months. In addition, two areas of lifestyle need to be addressed. The first concerns exercise. Since exercise will need to be a major part of the ongoing care for the child with PWS, having the family develop a habit of 30 min of exercise prior to the evening meal is critical. While at this point, this exercise may be the family walking and pushing a stroller with the baby in it, by the time the child is mobile and walking with the family a habit has been established. As the youngster becomes more ‘stubborn’ with age, the tool of holding dinner until the exercise is complete will already be a part of the daily routine. Secondly, family decision making and parenting processes need to be addressed. An example will serve to illustrate. Let us assume the family has gone out to dinner. The youngster with PWS wants a menu item that is not allowed, and one parent (often mother) says no, following which the youngster starts to tantrum. If the other parent says ‘just this once, don’t cause a scene’, the youngster has begun to learn that parents can be ‘divided and conquered’ and that having a tantrum gets the desired item. While behavior problems are an inherent part of the syndrome, previous studies indicate a strong correlation between more severe behavior problems and the level of parental conflict over child-rearing techniques [Whitman, unpubl. data; 3]. Thus, if this ability to ‘split’ parents becomes a set pattern, the youngster can begin to ‘rule’ the family through bad behavior. A short course of family therapy during these early stages can be helpful as a preventive strategy.

Speech and Language

Around age 3, the central issues switch to understanding and dealing with the speech and language delays, and for most, the emergence of the food-related behavioral constellation. Early difficulties with speech and language reflect the continuing impact of reduced oral motor skills as well as other anatomical deficits including altered growth of the larynx, a narrow over-jet, a narrow palatal arch, micrognathia, inadequate velopharyngeal movement and closure [4–6]. These deficits are reflected in receptive/expressive language delays, poor speech sound development, abnormal pitch, and hypernasality. Speech delays are usually the first concern raised by parents, with first words and vocabulary acquisition emerging anywhere from 18 months to 6 years of age. For many, emerging speech remains highly unintelligible for several years. A number of parents have noted, ‘I was afraid I would be the only one to ever understand him/her’. Many aged 6–12 years continue to display significant

PWS: A Neurobehavioral Disorder

201

residual articulation errors, hypernasality, pitch and voice quality difficulties, and are less intelligible in connected speech. Further, receptive and expressive language lags continue and there is emerging difficulty with processing language. Receptive and expressive nonverbal language deficits are also noted. Poor eye contact and an inability to perceive and perform routine nonverbal cues such as turn taking, volume modulation, prosidy, and proxemics are routinely observed at all ages. Conversational and social skill deficits beyond that expected for their IQ and developmental level have been documented by a number of investigators. Speech and language therapy, including the use of signs and supplemental communication boards, should be instituted as early as possible, no later than 12 months of age. Minimum goals of speech therapy include improved oral motor skills, reduced hypernasality and nasal emission, improved intelligibility of speech, better conversational skills, and increased receptive and expressive language abilities. Following the initial concerns with speech and language development, the issues evolve over time to the impact of speech and language deficits on education, social skills and long-term adjustment. Academic difficulties in the areas of reading, writing, and comprehension are frequently noted. Supportive educational services for children with language disorders may be necessary throughout school years. Recent work suggests that these communication deficits along with other previously described characteristics of the syndrome may reflect the presence of an autistic spectrum disorder as one aspect of the syndrome. Whitman [7] noted a number of autistic features in children aged 7 and under, with approximately 35% meeting criteria for frank autism; more importantly, by adolescence, 85% meet criteria for Asperger syndrome.

Food-Related Behavioral Constellation

A dramatic shift from the hypotonia and need for special feeding techniques in infancy occurs with the emergence of the food-related behavioral constellation which usually occurs between the ages of 2–4 (although ranges of 1–6 years are reported). This constellation includes an increased consumption that quickly escalates to chronic hyperphagia, a constant preoccupation with food, meal times, menus, and food seeking with a growing capacity to sneak food for immediate consumption, and the creation of clever hiding places for hoarding ‘emergency stashes’. At the same time, an alteration in sleep and behavior patterns is described [8]. Many children begin getting up at night to eat, often

Whitman

202

seeming to know without being told that they need to conceal this behavior from the family. Most children also display a growing behavioral irritability and an increasing tendency to tantrum when food requests are denied. As the child gets older, she/he becomes increasingly clever at illicit methods of obtaining food, increasingly clever at disguising their food-related activities, and may add some unpleasant and potentially dangerous food-related activities such as foraging through garbage cans, ingesting ‘food-appearing’ substances such as apple-scented shampoos, and stealing money to buy food or directly stealing food. Not infrequently, older children have picked or destroyed locks on cabinets and refrigerators to obtain the contents, have quietly left their homes at night to seek food, or have shoplifted food – not uncommonly with legal consequences. Because of their nature, both school and church environments frequently become unwelcome and unwelcoming environments to the child and his/her family, and public dining an impossible treat. A blunted or absent gag reflex and an inability to vomit further complicates an already difficult picture. Unlike ‘normal’ individuals, for those with PWS, extreme overeating or even ingesting spoiled food fails to induce vomiting. Without appropriate management, affected individuals sustain a rapid weight gain leading to early and chronic morbid obesity. Associated morbidities include gastrointestinal problems (constipation, diarrhea, gall bladder disease, intestinal blockages, fatty liver disease), metabolic disorders (e.g. type 2 diabetes), cardiopulmonary complications, increased sleep disturbance and sleep-related breathing difficulties, and further increased body composition abnormalities. Despite major advances in our understanding of obesity and its treatment, management of those with PWS remains complicated by the continued refractivity of the hyperphagia to pharmacotherapy. In the absence of effective pharmacotherapy, management of this issue remains environmental and behavioral, including dietary alterations addressing reduced energy requirements with strict control of caloric intake, an intense exercise regimen, medical support for growth, metabolic, hormonal, pulmonary, cardiac, sleep, gastrointestinal and body composition concerns. As adults, because of the need for constant supervision to prevent life-threatening obesity, most affected individuals require more restrictive care and closer supervision than others of similar intellectual abilities.

Behavior

Behaviorally, the younger child with PWS is described as happy, personable, affectionate, compliant and cooperative despite the emerging hyperphagia

PWS: A Neurobehavioral Disorder

203

Table 2. Behavioral characteristics of 100 persons with PWS aged 4–46 years of age [adapted from 10]

Behavior

%

Overeats Skin picking Stubborn Obsessions Tantrums Impulsive Labile Excessive sleep Talks too much Compulsions Anxious, worried Gets teased a lot Hoards (nonfood) Steals (food, money for food) Withdrawn Unhappy, sad

98 97 95 94 88 76 76 75 74 71 70 65 55 54 53 51

that occurs between 3 and 6 years of age. By contrast, adolescent age persons with PWS are described as extraordinarily stubborn, clever, manipulative, irritable, mood labile, angry, perseverative, ego-centered, demanding, and prone to temper tantrums and rage attacks that may include aggressiveness toward themselves and others. Lying and stealing are noted as ‘not uncommon’, as are emotional difficulties in the form of depression, obsessions and frank psychoses (table 2) [9, 10]. Indeed, unless obesity is life threatening, behavioral difficulties are the prime concern of caregivers [11]. When poorly managed, these behaviors can result in family turmoil, community placement failure, employment failure, referral for mental health consultation and even institutional placement. What happens to the previously happy, compliant personable child? A number of reports describe a noticeable, frequently acute shift and intensification in behavior and personality occurring between the ages of 6 and 8 [12]. This shift includes a decreasing ability to ‘go with the flow’, an increasing need for routine, increasing stubbornness, a decreased ability to tolerate frustration with an accompanying increase in behavior severity in response to frustration, increased difficulty with transitions, increased perseveration, more severe day/night sleep pattern alterations, and an increase in ‘skin picking’. In sum, there appears to be an increase in neurological and cognitive rigidity. At least part of these agerelated behavioral difficulties can be attributed to the increasing impact of multiple cognitive deficits on daily functioning. At least four separate sources of cognitive deficits can be identified: (1) global mental retardation, (2) language processing problems, (3) an independent pattern of learning disabilities in the

Whitman

204

Table 3. Cognitive characteristics [adapted from 20] % Normal Borderline IQ (70–85) Mild mental retardation (55–69) Moderate mental retardation (40–55) Severe to profound mental retardation (⬍40)

5 27 34 27 6

area of short-term memory and sequencing deficits [13], and (4) a failure to develop metacognitive abilities [14]. A number of reports indicated that ‘typical’ intellectual abilities for this population are in the mild to moderate range of mental retardation (table 3). Those with maternal disomy average about 10 points higher on verbal IQ scores than those with a paternal deletion, but this slight advantage is not reflected in functional differences. While approximately 5% will test in the normal range of intellectual ability, it is uniformly reported that functional capacities even for this group are significantly below that expected for nonaffected individuals with similar cognitive capacity. This may be in part due to the added and interactive impact of language, sequential learning, and metacognitive deficits on social skills, interpersonal relationships and adjustment (table 4). Sequential processing deficits impact the ability to put things in order, either a linear (1, 2, 3; a, b, c) or temporal (yesterday, today, tomorrow) order. Persons with sequential deficits often have trouble remembering a schedule, following a schedule, remembering what to do and when to do it, and figuring out the order in which they should tackle the components of a task. Since most academic tasks, and indeed, life tasks require a certain ability to say, ‘First I do this, then I do that’, etc., persons with sequential deficits are often overwhelmed and lost regarding what they are supposed to do, how they are supposed to do it, and when they are supposed to do it. This inability to ‘figure things out’ leads to constant anxiety, frustration and frequently acting out such as the behavioral difficulties we see in persons with PWS. Life seems constantly confusing and they seem inordinately unable to sort out the confusion. Thus many individuals with sequencing deficits are found to be excessively rigid and unable to alter the course of their actions when a particular technique is failing them. This inflexibility and ‘stubbornness’ of perspective is one of the most frequently described ‘behavioral difficulties’ of persons with PWS. Most often, however, it is not so much seen as a problem in academic areas where such deficits are accommodated; instead the effects are seen in the behavioral and interpersonal arena where a sudden shift in plans or activities is occasion for a major behavioral response with an assertion that ‘But you said

PWS: A Neurobehavioral Disorder

205

Table 4. Behavioral impact of sequential processing and metacognitive weakness Sequential deficits

Behavioral impact Abnormal intake, process and production of response to information Problems decoding and reading Reading comprehension deficits Trouble understanding, remembering, and executing multistep directions Trouble with calculations Confusion over time concepts and temporal meanings (before, after, until; yesterday, today, tomorrow) Delayed mastery of experiential sequences (days of week, months of year) Problems with demand language (poor sequencing of events and ideas) Poor use of time Trouble with cause-effect relationships

Metacognition deficits

Undeveloped or underutilized problem-solving strategies Ineffective abstraction Deficient inferential reasoning Tenuous formation of concepts Poor appreciation and application of rules Failure to learn from experience Inability to alter ineffective strategies Poor development of empathy

were going to…’ and an inability to flexibly respond. At that point the person with PWS appears completely unable to see what the rest of the environment view as ‘just common sense’, and frequently escalates into tantrums and aggression. Metacognition is defined as the ability to reflect on one’s own thinking, to abstract, and to step back and conceptualize the learning processes needed to meet task demands. Most descriptions of the older child with PWS underscore their ‘ego-centered’ nature, a general lack of empathy and ability to understand the needs of others, and difficulty in grasping the full meaning of a topic and the logic behind it. Even those persons with PWS whose cognitive ability tests in a normal range exhibit a concrete, black and white thinking pattern that is more characteristic of those with more severe cognitive limitations. A tenuous formation of concepts and poor appreciation and/or application of rules are common. Thus, they may carte blanche apply a single rule to all situations or conversely not generalize at all. As a result, the fit between a ‘rule’ and the ‘situation’ frequently appears self-serving and contrived to the casual observer.

Whitman

206

Behavioral Management Like the hyperphagia, atypical, unpredictable, and frequently toxic responses to most psychotropic agents complicate behavioral management [15, 16]. Thus, the primary tool for behavior management at all ages is an extremely structured, highly routine environment, with clear and invariant ‘rules and regulations’. While easy to describe, such environments are difficult to achieve. Families, particularly those with other children, have multiple and competing demands on their time and resources. One child’s need may be in direct conflict with the need of the child with PWS. The escalating need to minimize access to food often complicates engaging in such routine family activities such as attending a sibling’s ball game or allowing a sibling to invite friends for a ‘sleep-over’. Routine family tasks such as ‘running to the grocery for a loaf of bread’ become nonroutine and a source of stress if there is no one with which to leave the child with PWS while completing this errand. Good behavior management for the affected child often hinges on the availability of sufficient supports for the family to maintain some degree of normal functioning with a minimum of stress. Similarly, for adults in alternative care, consistency among and across staff is critical; unclear or conflicting expectations, or inconsistent routine can lead to serious and chronic behavioral difficulties. Most individuals with PWS who frequently exhibit problem behaviors are able to alter these behaviors when environmental changes are instituted. However, identifying the environmental precipitants and making appropriate changes may require considerable creativity, patience and effort. When problem behaviors fail to respond to well-planned behavioral interventions, or when a serious psychiatric need is apparent, pharmacologic interventions must be considered. Those with maternal disomy appear to be more vulnerable to serious psychiatric need than those with paternal deletions. Unfortunately, data document that most psychopharmacologic agents are either ineffective or worsen the behavior. In the past few years, for many the SSRIs (selective serotonin reuptake inhibitors) have provided a nonspecific behaviorstabilizing effect including fewer outbursts, a reduction in irritability, and less perseveration, but with no specific antidepressant effect. Some of the newer atypical neuroleptics have also proven useful in specific patients. However, even among those medications that have provided benefit, the response is inconsistent across patients. Prozac®, for instance, may prove highly beneficial for one patient and have a dramatic adverse response in the next. In addition to highly idiosyncratic responses to psychotropic medications, the abnormal body composition appears to contribute to atypical dosing requirements. Many individuals benefit from dosages one quarter to half that normally prescribed, but become toxic with an exacerbation of behaviors at normal dosage levels. For these reasons, psychotropic medication should be used for behavior management only when all

PWS: A Neurobehavioral Disorder

207

other interventions have failed, or when psychiatric needs such as psychoses dictate their use. Medical Issues and Intervention A number of medical issues have been mentioned throughout the previous sections, and there are other syndrome-related medical issues not addressed in this paper, such as the cryptorchidism and small genitalia. Discussion regarding these issues can be found in comprehensive descriptions of the syndrome. Recently this author and others have investigated the use of growth hormone replacement therapy (GHRT) to address both the growth failure and the abnormal body composition abnormalities for persons with PWS [17–19]. While we were hypothesizing improvements in linear growth, body composition, increased energy expenditure and fat utilization, improved muscle strength, physical ability and pulmonary functioning, a major question regarding efficacy hinged on the possible impact on behavior. In a non-Prader-Willi population, GHRT has been shown to improve alertness, increase activity level, reduce irritability, reduce tendencies for worrying, and improve interpersonal relationships. We hypothesized similar outcomes for those with PWS, but maintained an awareness of the equal possibility of worsened behavior as an idiosyncratic response to GHRT in this population. We were, however, able to show similar mood and behavior improvements in 54 children between the ages of 4 and 16 at the time of initiation of GHRT. As a result of those studies, a study of the use of GHRT in infancy with a primary focus on preventing the development of abnormal body composition is in progress.

Multidisciplinary Management

The complexity of the syndrome as well as the intensity of support needs for affected individuals and their families are best served through a coordinated multidisciplinary management approach. Many centers utilize a multidisciplinary clinic model with a primary case coordinator and a set cadre of professionals who attend each clinic session, and a secondary set of professionals for episodic referral. Other centers have a smaller clinic-based team with a set cadre for secondary referral. If the first model is utilized, each clinic should have available the primary medical provider (pediatrics, internal medicine), nursing, a dietician, physical therapist, occupational therapist, speech and language therapist, social work/family therapist, psychologist and endocrinologist. The secondary cadre includes orthopedics, cardiology/pulmonary, gastrointestinal, ophthalmology, neurology, psychiatry, dentist, and special educator. The case manager should be available to the family at the beginning of each clinic to

Whitman

208

direct the flow of appointments, and at the end of the clinic to help pull together the recommendations, answer questions, and provide a continuity of support. In addition, there should be a designated liaison between the clinic and external service providers. Teachers and school personnel, including cooks, janitors and bus drivers, should be contacted at the beginning of each school year and education regarding the syndrome and their part in helping maintain good physical and mental health for the child and family emphasized. A meeting with the special educator to review the educational program for scope and appropriateness should also occur at the beginning of each school year. Home visits should be made at least twice yearly, both to support the family in their own environment, and also to help with environmental modifications regarding food storage and garbage disposal. Finally, where at all possible, family gatherings for all families served should be convened on at least a semiannual schedule, more frequently if possible. Parents and children should meet separately to be able to compare experiences, air concerns, learn advocacy strategies, and offer each other support. Clearly, these children and their families require many and often intensive services. However, failure to provide these services can only result in the need for even more intense services, with poorer prognosis.

References 1 2 3

4 5 6 7 8 9

10 11

Whitman B: Understanding and managing the behavioral and psychological components of Prader-Willi syndrome. Prader-Willi Perspect 1995;3/3:3–11. Akefeldt A, Gillberg C: Behavior and personality characteristics of children and young adults with Prader-Willi syndrome: A controlled study. J Am Acad Child Adolesc Psychiatry 1999;38:761–769. Funahashi M, Harada T, Murakami M, Azumi M, Hattori K: Relationship between parental attitudes and behavior characteristics of people with Prader-Willi syndrome. Proceedings 4th Triennial IPWSO Scientific Conference, St. Paul, 2001, p 51. Kleppe SA, Katayama KM, Shipley KG, Foushee DR: The speech and language characteristics of children with Prader-Willi syndrome. J Speech Hear Disord 1990;55:300–309. Akefeldt A, Akefeldt B, Gillberg C: Voice, speech and language characteristics of children with Prader-Willi syndrome. J Intellect Disabil Res 1997;41:302–311. Lewis BA, Freebairn LA, Heeger S, Cassidy SB: Speech and language skills of individuals with Prader-Willi syndrome. Am J Speech Lang Pathol 2002;11/3:285–294. Whitman B: Prader-Willi syndrome and autistic spectrum disorder: A chance or an ‘intimate’ relationship. Proceedings 4th Triennial IPWSO Scientific Conference, St. Paul, 2001, p 40. Dimitropoulos A, Feurer ID, Butler MG, Thompson T: Emergence of compulsive behavior and tantrums in children with Prader-Willi syndrome. Am J Ment Retard 2001;106/1:39–51. Whitman B, Greenswag L: Psychological issues in Prader-Willi syndrome; in Greenswag L, Alexander R (eds): Management of Prader-Willi Syndrome, ed 2. New York, Springer, 1995, pp 125–141. Dykens E, Hodapp R, Finucane B: Prader-Willi syndrome; in Dykens E, Hodapp R, Finucane B (eds): Genetics and Mental Retardation Syndromes. Brookes, Baltimore, 2000, pp 169–205. Greenswag LR: Adults with Prader-Willi syndrome: A survey of 232 cases. Dev Med Child Neurol 1987;29:145–152.

PWS: A Neurobehavioral Disorder

209

12 13 14 15

16

17

18 19

20

Curfs LMG, Verhulst FC, Fryns JP: Behavioral and emotional problems in youngsters with Prader-Willi syndrome. Genet Couns 1991;2/1:33–41. Dykens EM, Hodapp RM, Walsh K, Nash LJ : Profiles, correlates, and trajectories of intelligence in Prader-Willi syndrome. J Am Acad Child Adolesc Psychiatry 1992;31:1125–1130. Gross-Tsur V, Landau Y, Benarroch F, Wertman-Elad R, Shalev R: Cognition, attention, and behavior in Prader-Willi syndrome. J Child Neurol 2001;16/4:288–290. Whitman B, Greenswag L, Boyt M: The use and impact of psychotropic medications for managing behavior in persons with Prader-Willi syndrome. Proceedings 13th Annual Prader-Willi Association Scientific Conference, Columbus,1998. Stein D, Keating J, Zar H, Hollander E: A survey of the phenomenology and pharmacotherapy of compulsive and impulsive-aggressive symptoms in Prader-Willi syndrome. J Neuropsychiatry Clin Neurosci 1994;6:23–29. Carrel A, Allen D, Myers S, Whitman B: Growth hormone improves body composition, fat utilization, physical strength and agility, and growth in Prader-Willi syndrome: A controlled study. J Pediatr 1999;134:215–221. Carrel A, Myers SE, Whitman BY, Allen DB: Benefits of long-term growth hormone therapy in Prader-Willi syndrome: A four year study. J Clin Endocrinol Metab 2002;874:1581–1585. Whitman B, Myers S, Carrel A, Allen D: The behavioral impact of growth hormone treatment for children and adolescents with Prader-Willi syndrome: A two-year controlled study. Pediatrics 2002;109:e35. Curfs LMG, Fryns JP: Prader-Willi syndrome: A review with special attention to the cognitive and behavioral profile. Birth Defects Original Article Ser 1992;28/1:99–104.

Barbara Y. Whitman, PhD, Associate Professor of Pediatrics, Department of Pediatrics, St. Louis University School of Medicine, 1465 S. Grand, St. Louis, MO 63104 (USA) Tel. ⫹1 314 268 4027, Fax ⫹1 314 268 4194, E-Mail [email protected], [email protected]

Whitman

210

Comprehensive Treatment Approaches Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 211–221

A Comprehensive Approach to Limiting Weight Gain and to Normalizing Body Composition in Prader-Willi Syndrome Urs Eiholzer Foundation Growth Puberty Adolescence, Zurich, Switzerland

Abstract The Prader-Willi syndrome (PWS) remains a complex disorder in many respects. The pathogenesis of the disturbed energy regulation with hyperphagia and hypoactivity and the disturbed body composition are still largely unknown. Limitation of nutritional input was the first and is until today the most important type of intervention in limiting weight gain and body disturbance. Growth hormone (GH) therapy – the second possible intervention in limiting weight gain and body composition disturbance – has greatly changed the phenotype of PWS. Height und weight are nearly normalized in prepubertal children. Experience has shown, however, that hyperphagia persists even during GH therapy and weight and body fat will decrease only if energy input is not increased. However, even if weight is nearly normalized, there is no normalization of body composition. The initial deficit in lean body mass is counteracted by GH only during the first year of therapy and there is no long-term additional gain of muscle mass beyond the first year. Recently it has been shown that in PWS a short daily training programme is sufficient to improve local body composition. In addition, such training increases spontaneous physical activity and leads to a significant and sustained improvement of physical capacity. Thus, today a third option of intervention exists to limit weight gain and body composition disturbance in PWS: a structured daily exercise programme. Moreover, it was shown that consistent upbringing behaviour is also negatively correlated with weight gain. Hence, the fourth intervention option to limit weight gain and body composition disturbance is to influence the style of upbringing via psychological counselling. Copyright © 2003 S. Karger AG, Basel

Introduction

The Prader-Willi syndrome (PWS) remains a complex disorder in many respects despite major advances in recent research. The pathogenesis of the disturbed energy regulation with hyperphagia and hypoactivity and the disturbed body composition are still largely unknown. For decades, most physicians and researchers as well as the authors of many publications focused primarily on hyperphagia and the increased fat mass. First Intervention: Limitation of Nutritional Input

Limitation of nutritional input was the first and has until today been the most important type of intervention in limiting weight gain and body disturbance. Prior to the early 1990s, growth, metabolism and body composition in PWS were assumed not to differ from non-syndromal obesity [1, 2]. Low growth hormone (GH) secretion was explained to be the consequence of obesity [3–5]. Furthermore, since growth was not as retarded as in GH deficiency, GH-independent growth as in simple obesity was assumed [6]. In addition, IGF-1 levels in PWS were not as decreased as in GH deficiency. Metabolic changes, e.g. of carbohydrate and lipid metabolism, were also thought to occur in parallel with obesity [7]. The general opinion of paediatric endocrinologists was that the use of GH in PWS is unwarranted [1]. Obesity is the main cause of morbidity and mortality in PWS [8, 9]. The huge fat accumulation in PWS is caused by an imbalance between energy intake and energy expenditure, which results in an increase in energy storage. Between 1 and 4 years of age, the feeding difficulties are generally replaced by the PWS-specific insatiable hunger drive, which develops insidiously and causes life-threatening obesity [10, 11]. For a long time, treatment was limited to non-specific measures, mainly to the reduction of energy intake and fat content, e.g. to less than 8–11 kcal/cm length and about 25% fat [12]. This was only possible by means of strict supervision and permanent locking up of food. Yet, even close supervision of nutrient intake only contributed to limiting obesity, but was not sufficient to avoid it. Limitation of nutritional input, however, was the first and has until these days been the most important type of intervention in limiting weight gain and body disturbance. Second Intervention: GH Treatment

When the first papers demonstrated that lean body mass in PWS was reduced [13–16], a new light was shed on the aetiology of obesity in this

Eiholzer

212

4

7 6

3

5 2

4 3

SDS

1

2

0

1 ⫺1

0

⫺2

⫺1 ⫺2

⫺3

⫺3

⫺4

*

⫺5 0

a

*

*

6 12 18 24 30 36 42 48 54 60

⫺4

*

⫺5

*

*

0 6 12 18 24 30 36 42 48 54 60

Months of GH therapy

b

Fig. 1. Height SDS (a) and WfH SDS (b) of children with PWS, referring to normative data of the Zurich Longitudinal Study [27], before and during up to 5.5 years of GHT. Values are shown for young, initially underweight children (䉱, n ⫽ 10) and overweight prepubertal children (䊏, n ⫽ 8). Medians are shown as thick lines; minimum and maximum values of the combined group are shown as thin lines. Significant differences, tested at 6, 48 and 60 months by the Wilcoxon test in each group vs. the value before therapy, are indicated as *p ⬍ 0.05 [reproduced from 18].

syndrome. As a matter of fact, short stature and increased fat mass in the presence of decreased muscle mass are typical symptoms of a GH deficiency. Subsequently it was thought that body composition in PWS, being abnormal as a consequence of GH deficiency, could be normalized by substitution with GH. Several studies on GH treatment in PWS children were initiated [review in 17]. In general the observed changes of growth document the efficacy of GH treatment in PWS (fig. 1). In our study, mean growth velocity in the first year increased in the prepubertal obese children from ⫺1.4 to 8.5 SD and remained above 2 SD throughout 3 years of therapy [19]. After 4 years of human GH treatment, height gain reached 1.8 SD, yielding an average height of 0 SD and a normalization of hand and, less markedly, foot length. Furthermore, the prediction of final height markedly improved in the prepubertal children and reached the range of their parental target height, despite the average adult height of untreated PWS reported as being reduced to ⫺2.5 SD of reference values [20]. The results on growth of all other studies are similar.

Comprehensive Approach in PWS Therapy

213

*

**

*

2 4

1

3

⫺1

SDS for height

SDS for height

0

⫺2 ⫺3

2

1

⫺4 ⫺5

0 ⫺6 ⫺7

n ⫽11 11 10 0

a

6

8

12 18 24

7

6 ⫺1

30 36 42

Months of GH therapy

n⫽11 11 10 0

6

8

7

6

12 18 24 30 36 42

Months of GH therapy

b

Fig. 2. Body composition measured by DEXA in 16 PWS children expressed as height-related SDS of a Dutch reference population [28] in those taller than 100 cm: lean mass (a) and fat mass (b). Medians (䊉, thick lines) and individual courses of young underweight (䉱, n ⫽ 4), prepubertal overweight (䊏, n ⫽ 8) and pubertal PWS children (---, n ⫽ 4). Significant differences vs. baseline are calculated as indicated in figure 1: *p ⬍ 0.05; **p ⬍ 0.01 [reproduced from 18].

Body weight, too, became normal (fig. 1). In the prepubertal obese children, weight for height (WfH) dropped during the first 2 years of growth hormone therapy (GHT), the decrease in WfH being greatest during the first 6 months of therapy; after 2 years WfH SD scores (SDS) stabilized at ⫹0.8 SD. In contrast, WfH in the very young children was reduced before therapy and continuously increased up to 0.4 SD at 36 months of treatment. Different to other studies, we adjusted lean body mass not only for age and sex, but also for height (fig. 2), in order to correct for the growth-related increase. In this way we were able to show that the initial deficit in lean body mass (⫺3.0 SD in overweight children) is counteracted by GH only during the first year of therapy (increase to ⫺1.79 SD) [21], and that no significant additional gain of muscle mass beyond the first year was achieved by GHT (⫺1.33 SD, at 42 months). In contrast to what could be concluded from basal studies [14] or 1- and 2-year GH treatment [16, 22–24], GHT could not, even in the long term, compensate for the initial deficit of lean mass [21]. This finding was corroborated by another study on resting energy expenditure. It was found that resting

Eiholzer

214

16

Energy intake height (kcal/cm/day)

14

**

12 10 8 6 4 2 0 PWS

Controls

Fig. 3. Energy intake, means ⫾ SD, of 18 children with PWS and of 19 control children. Significant difference by MannWhitney U test is indicated by asterisks: **p ⬍ 0.01.

energy expenditure, which is dependent on muscle mass, did only increase to a minor extent during GHT [25]. GHT – the second possible intervention in limiting weight gain and body composition disturbance – has greatly changed the phenotype of PWS. Height and weight are nearly normalized in prepubertal children. Therefore a new treatment option is now available to physicians treating these children. For the first time ever, a large number of children with PWS no longer becomes grossly obese – provided a restriction of calorie intake is maintained. Experience has shown that hyperphagia persists during therapy with exogenous GH. For this reason, and because weight and body fat will decrease only if energy input is not increased during GHT [21, 26], energy input has to be continuously monitored. Figure 3 shows the evaluation of nutrition protocols 3 years after the institution of GH treatment during a phase of stable weight. The energy intake in children with PWS was significantly reduced by about 25% compared to healthy control children. From these results, it is concluded that – even during GH treatment – children with PWS can keep their weight stable only if they reduce their energy intake to about 75% of what is recommended for healthy children. Although LBM is improved, it still remains far away from being normalized by GH treatment. Normalizing, or at least improving body composition, however, continues to be the main objective in the treatment of children with PWS.

Comprehensive Approach in PWS Therapy

215

PWS

*

Walking distance (km/3 days)

35 30

Controls

o

25 20 15 10 5 0 Before training

After training

Fig. 4. Physical activity before the start and at the end of the training programme of children with PWS (n ⫽ 17) and controls (n ⫽ 18) measured by means of pedometer registrations (Mechanical Pedometer Eschenbach, Germany; kindly provided by Dr. Karsegard, Department of Nutrition, University Hospital Geneva). Results expressed as means ⫾ SEM. *p ⬍ 0.05, significant difference within the PWS group tested by the Wilcoxon test; °p ⬍ 0.05, significant difference between the PWS and the control group tested by the Mann-Whitney U test.

Third Intervention: Carrying out an Exercise Programme on a Daily Basis

Even though GHT had sustained beneficial effects on physical strength and agility, the reports of parents and caregivers that children with PWS are more active while being administered GH remained unproven. As physical activity is the most important stimulus for muscle growth, we hypothesized that physical hypoactivity persists despite improvement of body composition and muscle strength. Therefore, we have recently measured spontaneous physical activity during GHT by pedometers. A pedometer converts impulses of movement into distance, the results are read in kilometres. We found that even while being administered GH, children with PWS show considerably less activity than the controls (11 km during 3 days vs. 26 km of the controls, fig. 4) [27]. Thus, in terms of energy, hypoactivity and a marked dislike of physical activity seem to be very essential symptoms of PWS. They are similarly important like the lack of satiety and the ensuing hyperphagia. In order to see whether hypoactivity is the cause or the consequence of the reduced muscle mass, it was investigated whether muscles of children with PWS can be trained in the same way as those of healthy controls with a training

Eiholzer

216

180

PWS

˚ ˚

160

˚˚

Controls ##

Number of repetitions

140 120 100

**

80

**

60 40 20 0 Start

End

3 off

Start

End

3 off

Fig. 5. Physical capacity before the start, at the end and 3 months after the end of the training programme of children with PWS (n ⫽ 17) and controls (n ⫽ 18). Results expressed as means ⫾ SEM. Significant differences within the PWS or the control group, tested by the Wilcoxon test, are indicated as **p ⬍ 0.01 and ##p ⬍ 0.01. Significant differences between the PWS and the control group, tested by the Mann-Whitney U test, are indicated as °p ⬍ 0.05 and °°p ⬍ 0.01 [reproduced with permission from Elsevier, 27].

programme for the calf muscles. Total daily training time was around 3–4 min. After 3 months of training it was found that children with PWS and controls increased their local muscle mass to a similar and significant extent. After 3 months off training local mass decreased again significantly in both groups close to the level before the start of the training. In conclusion, it seems possible to increase muscle mass in PWS to roughly the same extent as in healthy children. From this, it was concluded that reduced muscle mass even under GHT is a direct consequence of hypoactivity [27]. Moreover, and most promising, a significant increase in spontaneous physical activity was observed at the end of the training programme in the children with PWS. Daily walking distance was augmented from 45 to 70% compared to the baseline data of the controls (fig. 4). This means that it is possible to improve the sedentary behaviour pattern in PWS through an easy daily 3-min physical training programme [27]. Furthermore, the training programme led to a significant increase in physical capacity (fig. 5). Children with PWS had nearly trebled their performance, whereas the controls had at least doubled theirs at the end of the training programme. Without special training, the physical capacity decreased significantly in both groups but nevertheless remained

Comprehensive Approach in PWS Therapy

217

higher than at the beginning of the study, indicating a possible long-term training effect. In summary, a short training programme is sufficient to significantly improve local body composition, to significantly increase spontaneous physical activity and to lead to a significant and sustained improvement of the physical capacity [27]. Therefore, today there is a third option of intervention to limit weight gain and body composition disturbance: a structured and daily exercise programme. The great importance of physical activity should be clearly communicated to parents, carers and individuals with PWS. We suggest a personal and regular physical training programme for individuals with PWS, including a workout of a variety of different muscles. This new approach in the treatment of PWS opens up a complementary therapeutic option in addition to dietary control and GH treatment.

Fourth Intervention: Psychological Counselling of the Families

Because hyperphagia and hypoactivity persist even during GHT, limiting energy input and enhancing energy output continue to be the most important aims in the care of children with PWS. In most instances, the parents have to assume the responsibility for these tasks. They have to impose these measures on the child with PWS, who is mentally retarded and often behaviourally disturbed. Parents and families need so much time and strength that most of them risk to exhaust their resources. In order to allow for a better counselling, the behaviour in affected families as regards bringing up these children was studied and compared with that in control families. It was found that the parents’ way of bringing up their child with PWS may be an important factor in determining his or her body composition (fig. 6). Parents of PWS children presented a distinctly more consistent style of bringing up their child than the parents of the control group – also in very general terms, not only related to food. In particular, a significant negative correlation between the consistency in the general upbringing and weight control in PWS was found: in other words a less consistent style of upbringing leads to heavier children. Furthermore, eating discipline and punishment did not play a major role in weight control. In contrast to PWS, there is no relationship at all between the style of upbringing and weight in the healthy control children. The energy balance in healthy normal-weight children does not require outside control. This shows that consistent upbringing behaviour and possibly also the family structure have an influence on weight development. The same may be true for other problems encountered in PWS. This leads to the fourth intervention

Eiholzer

218

90

PWS

80

Controls

*

Maximum score (%)

70

**

60

*

50 40 30 20 10 0 Consistency: food

Consistency: upbringing

Punishments

Fig. 6. Eating and style of upbringing, means and SD, displayed in percent of the maximum score of each group of items in the families of 18 children with PWS and of 19 controls. Significant differences by Mann-Whitney U test are specified by asterisks: *p ⬍ 0.05, **p ⬍ 0.01.

option to limit weight gain and body composition disturbance: psychological counselling may influence the style of upbringing and the family structure, taking as a model the family structure, in which the children become less obese. It is hard work for the parents and the whole family to bring up a child with PWS. Therefore, these families have to be given assistance. They have to learn that admitting that one’s resources are exhausted and accepting professional help are nothing to be ashamed of.

Conclusion

Thus, GHT, limitation of nutritional input as well as carrying out a welldefined exercise programme on a daily basis and psychological counselling of the parents and families are today the most important tools to improve body composition and well-being of children with PWS.

Acknowledgments This study was supported by the Swiss National Science Foundation (grant No. 32.056063.98) and the Swiss Academy of Medical Sciences (grant No. BK 286/98).

Comprehensive Approach in PWS Therapy

219

References 1

2

3 4

5

6

7 8 9 10 11 12 13 14 15 16

17 18

19

20 21

Van Vliet G, Deal C: Growth hormone and body fat: An overview of studies in hypopituitarism and in exogenous obesity; in Prader A, Rappaport R (eds): Serono Colloquia Europe Series. Clinical Issues in Growth Disorders: Evaluation, Diagnosis and Therapy. London, Freund Publishing House, 1994, pp 119–129. Frederich R, Hamann A, Anderson S, Lollmann B, Lowell BB, Flier JS: Leptin levels reflect body lipid content in mice: Evidence for diet-induced resistance to leptin action. Nat Med 1995;1: 1311–1314. Copinschi G, Wegienka LC, Hane S, Forsham PH: Effect of arginine on serum levels of insulin and growth hormone in obese subjects. Metabolism 1967;16:485–491. Quabbe HJ, Helge H, Kubicki S: Nocturnal growth hormone secretion: Correlation with sleeping EEG in adults and pattern in children and adolescents with non-pituitary dwarfism, overgrowth and with obesity. Acta Endocrinol (Copenh) 1971;67:767–783. Martha PM Jr., Gorman KM, Blizzard RM, Rogol AD, Veldhuis JD: Endogenous growth hormone secretion and clearance rates in normal boys, as determined by deconvolution analysis: Relationship to age, pubertal status, and body mass. J Clin Endocrinol Metab 1992;74: 336–344. Martha PM Jr, Reiter EO, Davila N, Shaw MA, Holcombe JH, Baumann G: Serum growth hormone (GH)-binding protein/receptor: An important determinant of GH responsiveness. J Clin Endocrinol Metab 1992;75:1464–1469. Bray G: Genetic, hypothalamic and endocrine features of clinical and experimental obesity. Prog Brain Res 1992;93:333–341. Cassidy SB: Prader-Willi syndrome. J Med Genet 1997;34:917–923. Laurance BM, Brito A, Wilkinson J: Prader-Willi syndrome after age 15 years. Arch Dis Child 1981;56:181–186. Zellweger H: Diagnosis and therapy in the first phase of PWS; in Holm V, Sulzbacher SJ, Pipes PL (eds): The Prader-Willi Syndrome. Baltimore, University Park Press, 1981, pp 55–68. Greenswag LR: Adults with Prader-Willi syndrome: A survey of 232 cases. Dev Med Child Neurol 1987;29:145–152. Stadler DD: Nutritional management; in Greenswag LR, Alexander R (eds): Management of Prader-Willi Syndrome. New York, Springer, 1995, pp 88–114. Schoeller D, Levitsky L, Bandini L, Dietz W, Walczak A: Energy expenditure and body composition in Prader-Willi syndrome. Metabolism 1988;37:115–120. Brambilla P, Bosio L, Manzoni P, Pietrobelli A, Beccaria L, Chiumello G: Peculiar body composition in patients with Prader-Labhart-Willi syndrome. Am J Clin Nutr 1997;65:1369–1374. Eiholzer U, Blum WF, Molinari L: Body fat determined by skinfold measurements is elevated despite underweight in infants with Prader-Labhart-Willi syndrome. J Pediatr 1999;134:222–225. Eiholzer U, Gisin R, Weinmann C, Kriemler S, Steinert H, Torresani T, Zachmann M, Prader A: Treatment with human growth hormone in patients with Prader-Labhart-Willi syndrome reduces body fat and increases muscle mass and physical performance. Eur J Pediatr 1998;157:368–377. Eiholzer U: Prader-Willi Syndrome. Effects of Human Growth Hormone Treatment. Endocr Dev. Basel, Karger, 2001. Eiholzer U, Bachmann S, I’Allemand D: Is there a growth hormone deficiency in PWS? Six arguments to support the presence of a hypothalamic GHD in PWS. Horm Res 2000; 53(suppl 3):44–52. Eiholzer U, l’Allemand D: Growth hormone normalises height, prediction of final height and hand length in children with Prader-Willi syndrome after four years of therapy. Horm Res 2000;53: 185–192. Wollmann HA, Schultz U, Grauer M, Ranke M: Reference values for height and weight in PraderWilli syndrome based on 315 patients. Eur J Pediatr 1998;157:634–642. Eiholzer U, l’Allemand D, van der Sluis I, Steinert H, Ellis K: Body composition abnormalities in children with Prader-Willi syndrome and long-term effects of growth hormone therapy. Horm Res 2000;53:200–206.

Eiholzer

220

22

23

24

25

26 27

28 29

Carrel A, Myers S, Whitman B, Allen D: Growth hormone improves body composition, fat utilization, physical strength and agility in Prader-Willi syndrome: A controlled study. J Pediatr 1999;134:215–221. Lindgren AC, Hagenas L, Muller J, Blichfeldt S, Rosenborg M, Brismar T, Ritzen EM: Growth hormone treatment of children with Prader-Willi syndrome affects linear growth and body composition favourably. Acta Paediatr 1998;87:28–31. Davies HA, Evans S, Broomhead S, Clough H, Day JL, Laidlaw A, Barnes N: Effect of growth hormone on height, weight, and body composition in Prader-Willi syndrome. Arch Dis Child 1998;78:474–476. Myers SE, Carrel AL, Whitman BY, Allen DB: Sustained benefit after 2 years of growth hormone on body composition, fat utilization, physical strength and agility, and growth in Prader-Willi syndrome. J Pediatr 2000;137:42–49. Eiholzer U, Bachmann S, l’Allemand D: GH treatment as part of a comprehensive therapy design for children with PWS. Int Growth Monit 2000;10:2–8. Eiholzer U, Nordmann Y, l’Allemand D, Schlumpf M, Schmid S, Kromeyer-Hauschild K: Improving body composition and physical activity in Prader-Willi syndrome. J Pediatr 2003;142: 73–78. Prader A, Largo R, Molinari L, Issler C: Physical growth of Swiss children form birth to 20 years of age. Helv Paediatr Acta 198943(suppl 52):1–125. Boot AM, Bouquet J, de Ridder MA, Krenning EP, De Muinck K: Determinants of body composition measured by dual-energy X-ray absorptiometry in Dutch children and adolescents. Am J Clin Nutr 1997;66:232–238.

PD Dr. med. Urs Eiholzer, Foundation Growth Puberty Adolescence, Möhrlistrasse 69, CH–8006 Zurich (Switzerland) Tel. ⫹41 1 3643700, Fax ⫹41 1 3643701, E-Mail [email protected]

Comprehensive Approach in PWS Therapy

221

Comprehensive Treatment Approaches Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 222–227

Discussion

Does Growth Hormone Affect Morbidities Associated with Obesity in Prader-Willi Syndrome? (A.L. Carrel) Anon: You showed an increase in absolute energy expenditure throughout the trial in absolute terms, kcal/day. How does that change when you adjust for the change in fat-free mass that obviously has occurred? A.L. Carrel, Madison, Wisc.: I think adjusting for fat-free mass is really the best way of asking about energy expenditure. Our data supports that resting energy expenditure, if we accept our indirect calorimetry, and I think there may be questions about that, is lower than we would expect based on weight. Clearly, correcting for fat-free mass or evaluating per lean body mass, they seem to be the same as in non-PWS children. I think that’s really an important point. M. Angulo, Mineola, N.Y.: As you know, the dose of GH in the adult is very low. So I was wondering, if you expect the same results in terms of the body composition, bone mineral density and cholesterol in the adult using the higher dose. Carrel: I don’t have a lot of this data, but if you define children with PWS who have stopped growing as adults, we do have some data on this. If we looked at the response in body composition with the different doses, I think the data looks similar. The low dose or the adult dose does not seem to give these patients much benefit. But they do seem to continue to have benefit on body composition with the other GH doses. B.P. Hauffa, Essen: Aaron, I would like to ask you a question regarding the dose ranges, you were talking about. From what you said, it appears that you do not advocate the lower dose, but recommend using the so-called standard or higher doses. How about any side effects related to dosage? Have you seen any very high IGF concentrations in the high dose range and what were the consequences? Did you use this information to lower the dose in these children and if yes, what happened to the metabolic changes? Carrel: To start the answer, yes. We did think to follow IGF-1 levels and indeed did see a dose response with that, as you might predict. So there was a difference in terms of the high or standard dose, low dose, in terms of IGF-1 response. For the most part, the mean of all the kids in the high dose was not

outside the normal for IGF-1 ranges for kids. And I think there is starting to be a little bit of some correlation with lean body mass with IGF-1 levels. In terms of side effects, we did have some side effects that we thought could be related to growth hormone. Two children had pseudotumor cerebri, with intracranial hypertension in the first few weeks of dosing. Those were both in the early part of the study, when we used our standard 1 mg/m2/day. After reducing the dose we did not have continuing problems. Otherwise, we really did not have any other side effects. We have looked at glucose, insulin, scoliosis, behavior, without documenting any side effects. U. Eiholzer, Zurich: What about hyperphagia under GH treatment? Carrel: If there was any change in hyperphagia? That’s a great question. We didn’t specifically think to follow that; we have thought of dietary questionnaires that we have followed pretty regularly. We had asked families not to change the diet, just to try to keep one more variable consistent. But many families confessed that when their kids were on treatment, they could be a little bit more lenient in terms of some other restrictions. But I can’t say that I have data that says they have really affected the hyperphagia. A.P. Goldstone, London: In terms of reaching final height, do you have any figures as how many still show GH deficiency on provocative testing? Carrel: That’s a great point. No. We have very few children, who have reached final height for whom we have gone back and done provocative GH testing. Of the few we have done, we continue to see low provocative levels, even when there is less body fat. Anon: What is your opinion about treatment of these patients at early age before one or two years of age? My second question is, how were the IGF-1 levels during the long-term treatment in these patients? Carrel: How were they during treatment? When we started, IGF-1 levels were low, close to 2 SDS below for age IGF-1 Z-score. As we looked at treatment there was a dose response between the different doses. There was the highest response of the high dose. But still they were not outside the normal range that we would see or expect baseline age and gender. Anon: Do you think, it’s better to start GH therapy at one year of age? Carrel: Right now, we are collecting that data, so I don’t have all the data to present. But I think, we are showing that there are some benefits and especially if we are kind of focusing on the abnormalities of body composition with the emphasis on hypotonia. We have data on almost 30 infants, as young as 4 months with a mean of as little as a year, that says that there is low muscle, but there is already elevated fat mass. As we look at very preliminary data, there is improvement in motor milestones and motor function; clinically, hypotonia is still the biggest problem for the children at this age. I’d like to have the data to show you, but I think we sort of slowly are getting into it.

Discussion

223

G. Gillessen-Kaesbach, Essen: Did you measure the hand and feet? Do you have any data on this? Carrel: We did, but I do not have it to present. D. l’Allemand, Zurich: I would like to make a comment to the question before, i.e. if it’s useful to treat the children below the age of two years. I think we have to be very careful, and we have very contradictory data. The lean mass is increasing, but the children are more mobile and the activity is increasing. We think that they are more able to go to and search for food, and fat mass is increasing in the same range as lean mass is increasing. And so there is no clear benefit up to now. We did construct guidelines for Germany, saying that below the age of two years we must be very careful. Anyway, we spoke about that yesterday, before you use GH, it’s important to check the respiration and the sleep pattern in order to not have children with a breathing disorder. Final Height of Patients after Growth Hormone Treatment (M. Angulo) Anon: I would like to know about the two patients with type 1 diabetes, I don’t understand why type 1 diabetes, because I know in these patients type 2 diabetes is more common with long-term GH therapy. Angulo: Well the theory is that they are genetically predisposed to have diabetes type 1 like anybody else; having Prader-Willi doesn’t make them immune to have another disease. They were going to have diabetes type 1 regardless of the PWS. Anon: In those patients did you find positive antibodies for type… Angulo: …Everything was positive… Anon: Ah yes, with the ketoacidosis… Angulo: Everything. The typical case… Anon: I know there is a paper published in the past of about one case… Angulo: As a matter of fact I was very excited, because first thing – I have the slide, but I didn’t have enough time to show you – there was a significant decrease in the weight gain and I thought, wow the GH is working beautiful, more than I expected, but the truth was that he was developing diabetes type 1. Eiholzer: You didn’t mention if you have replaced testosterone or estradiol. Angulo: None of them were getting testosterone treatment when they were getting GH. I just have to mention that I retested three of them because they are gaining a lot of weight and the result was, indeed, that they have still GHD by provocative testing. Anon: What would be your recommendations for GH treatment in adulthood? Do you know of any studies that have extended our findings into adulthood? Angulo: We are in the middle of working with the adults now. The data is not clear yet. But at least I can tell you that when we started the GH in the adult,

Comprehensive Treatment Approaches

224

we have started to see the decreasing body weight and also the improvement in body composition. But that is what I was asking Dr. Carrel. The dose that is approved for adult GHD in the United States, I still believe is very very low. I was wondering how much anabolic effect we’re going to get.

Role of Diet and Upbringing in Young Children with Prader-Willi Syndrome (D. l’Allemand) Hauffa: What would you tell the parents that come to the foundation and ask you, how can I increase the consistency of upbringing in my child, in my family? What would the items of counselling be? What would you in fact tell them? l’Allemand: I think the problem is that there are no careful studies on what these items really are. You have to consider in each family the specific problems these families have and I think the most important thing is that, we heard it before, that you don’t really need a happy family, but you must have a stable situation, with transparent situations and you must have constancy and very consistent information. That’s a general message. But I think research in this field, how to manage families with obesity, has to be continued and has to put an effort on looking at this subject. P.S.W. Davies, Sydney: You use the three-day report to get your food intake and hence your energy intake data. One of the big problems with three-day reports, as I am sure you are aware, is that people either change to make the reporting easier or report to you what they think you want to hear. I wonder if you did any checking of your energy intake data to see if it was feasible by maybe looking at it as multiple of their predicted or measured resting metabolic rate to get some idea of whether the data was accurate? l’Allemand: Yes, we considered the problem, but we didn’t have the ability to use objective methods to measure the caloric intake. I think a factor that speaks for the low intake of calories in these children is that the parental information regarding food is very good, as they are very well educated on the feeding behavior. The second point is the weights of the children; they were normal. Hauffa: How would your results have changed if you had performed the study in twelve-year olds? l’Allemand: That’s a crucial question. I think the behavioral problems and the hyperphagia augment with age. The problem of managing weight in children with PWS becomes more and more difficult towards puberty. Before those stages, you have to create a stable situation in the family and then you can give growth hormone and both together will bring the child into a stable state. But when you have a family that is coming with a child already twelve or thirteen years old, it will be very difficult to start then with an antiobesity management.

Discussion

225

Prader-Willi Syndrome: A Pervasive Neurodevelopmental Disorder Requiring a Multidisciplinary Care Approach (B.Y. Whitman) Anon: Can I ask you about the use of topiramate or risperidone in these patients for behavioral problems that they have? There are a few papers about the use of topiramate and risperidone. B.Y. Whitman, St. Louis, Mo.: There is a study now going on in Florida regarding the use of topiramate for behavioral problems and the data suggests that it doesn’t do a lot for general behavioral problems. It does have an anorexic effect and it seems to stop skin picking. But there are problems. It causes sodium depletion and seizures, it causes liver toxicity and these folks are so exquisitely sensitive to any medication that I personally would not advise using it until we have much more data. Risperidone, a neuroleptic agent, is another agent recently used for PWS for which we have limited data. The serotonin reuptake inhibitors (SSRIs), such as Prozac® (fluoxetine HCl) are very idiosyncratic; one child may respond to Prozac, another would have no impact from Prozac and might benefit from Zoloft® (sertraline HCl) and you could go down the list of the SSRIs. And in the last data set that I collected there were a number of kids, who have been through each one of the SSRIs. It is preferable not to continue SSRI therapy long term, if you can avoid it, and you have to weigh the risks and benefits up front. My advice is that medication is the last thing you should consider. The first intervention is investigate the environment, because 95% of behavioral problems are environmentally induced. Differences in management, not having the food put away, not anticipating transitions, not having a structure and maintaining it; 95% of the problems can be handled by environmental modification. But it may take time. Tony (Holland), do you have an comments on the medication issue? D.J. Holland, Cambridge, UK: I think the main point to make is when faced with a child or an adult PWS, the task is to understand why they are behaving in the way that they are. And to do that often requires the skills of many different disciplines. You should base your intervention upon your understanding of that particular child’s behavior at that particular point in time. If you look at any children or adults with what we call learning disabilities or what you call mental retardation, often there are, as you quite rightly point out, very important environmental and other issues; and the approach is fundamentally a psychological one. However, if you find in your assessment evidence of a serious mental illness in exactly the same way as if any of us had a serious mental illness, then your task is to ask, what’s the best treatment of that? … and the treatment of that is initially going to be medication. If we are talking about serious depressive illness, it’s going to be an antidepressant. In terms of the SSRIs, you would tend to use those that also are effective in anxiety and, possibly, obsessions. What we are seeing in

Comprehensive Treatment Approaches

226

psychiatry is that medications, such as antidepressants, are becoming much more specific. So, you just don’t throw any antidepressants at them. You decide on the basis of your evaluation which antidepressant or other medication is appropriate. If the patient is psychotic, you will often need both an antidepressant and a neuroleptic medication, such as risperidone. But you must base that upon your assessment and your understanding of that particular person. Whitman: And I can’t underscore that enough. I was called by a school last year for a young man that was having something like four or five major aggressive episodes a day. I had worked with the school six years ago and he has been stable for six years and they were telling me, oh my goodness, he has got to go back on medication, we have got to get him on this, he’s just out of control. Well it didn’t ring clear. I spent about ten hours in this school observing. This was a confined school for children with learning issues, mental retardation, Tourette’s, behavior difficulties. I followed this young man around and the first thing that I discovered is that they had changed the school policy. Teachers were now allowed to keep a Coke on their desk. In every classroom I went into, there was a two quart Coke bottle and I said, ‘What do you expect? You have got a Coke bottle there!’ How rude it is, if the teacher could have a bit the kids couldn’t … and they said that none of his tantrums are around food. Yes they are! The entire time he was sitting in that classroom, every waking fiber of his being is concentrated on how to get that Coke. So any little thing would set off a behavior. He happened to also have a teacher, who was in the power struggles with him and didn’t realize, you should never argue with a person with PWS, because you have already lost the arguement before you start. One of the classic examples was he had to keep the journal and one day he laid the journal on the corner of her desk. The teacher said, ‘You have to get that over there into your cupboard’ or whatever they call it. He says, ‘No, I always leave it here.’ Now anybody with an understanding of PWS would realize that something was going on, that he needed to do that and to let it go. Instead, the teacher got into a fight with him and that child, he was 17, wound up being hauled off to jail and restrained. And they are saying he has the problem. I am sorry, but they have the problem. So we changed teachers. I got the rules reinstalled where his teachers could not have food in the classroom – there were also trays of brownies sitting around, which I found kind of fascinating. And I was really amazed that he was not having any more behavioral problems. Since we cleaned up the environment, he has not had an episode. That was last March (about 7 months ago). For 4 months prior to this, he had up to four or five major aggressive episodes a day and they had jailed him three times. This type of situation doesn’t need medication. It needs attention to what has setting him off and changing it. That is our responsibility, not his.

Discussion

227

Epilogue Eiholzer U, l’Allemand D, Zipf WB (eds): Prader-Willi Syndrome as a Model for Obesity. International Symposium, Zurich, 2002. Basel, Karger, 2003, pp 228–231

Closing Remarks Rudolph L. Leibel Division of Molecular Genetics, Department of Pediatrics, Columbia University College of Physicians and Surgeons, New York, N.Y., USA

On behalf of all the attendees I would like to thank Dr. Eiholzer and the other organizers for a spectacular conference offering a good mix of very basic considerations and also extremely practical insights into the Prader-Willi syndrome (PWS). I, for one, have learned an enormous amount from listening to all of these experts. There are many things that could be said about this meeting and what has been presented. I thought I would discuss the meeting in the context that I am most easy and familiar with: What can be learned from what has been said here about this disorder with respect to the general issue of the regulation of body weight. A Model for Energy Homeostasis

I think that the PWS can provide important insights into general issues related to the mechanisms of energy homeostasis in humans. Thus, I regard the molecular genetics of PWS as being as potentially as revealing as some of the monogenic mouse models of obesity. When work on the molecular genetics of those mice began nearly 20 years ago, many people feared that the mice were not going to teach us anything about energy homeostasis in general. Now we can say that virtually every one of the mouse models has taught us a great deal – and I think that the PWS can do the same. Genetic Issues Before exploring further how this disorder may be relevant to general energy homeostasis, let me first comment about the genetics. I think the best way to interpret what is known about the genetics of the PWS is that it represents a very complex disorder caused by functional loss of imprinted gene(s). Whether it is one gene or a series of genes, whether it is a small nuclear RNA or

something in the vicinity – I think PWS can be regarded as a gene loss disorder that causes the body to sense that it does not have adequate fat mass. What is missing here, what is lost, is a gene that plays a very important role in the defense of body fat. While investigating the genes that affect body composition in mice, we have found relatively few that are designed to suppress body fat. This is probably a result of evolutionary pressures. For our animal forbearers the problem was how to get and keep sufficient body fat. The problem of excess body fat probably was not a major problem due to environmental conditions. Thus, what we are looking for, when scrutinizing the genetics of the PWS, are genes that have a role in defending body fat. Somehow in these patients the brain is tricked into thinking that this very critical store of potential chemical energy is deficient. There is clearly a primary defect that involves the function and/or structure of the hypothalamus and reflects itself in the changes in growth hormone, in gonadotropins and the autonomic nervous system in very striking ways. Probably the effects on insulin are being conveyed by the autonomic nervous system and some of the growth problems may be secondary to the suppression of insulin as a result of autonomic changes. Additionally, there are clearly peripheral structural somatic effects of this gene or group of genes. What about the relevance of PWS to other forms of obesity? What the Prader-Willi genetics might show us is a new pathway in the regulation of body weight. Anthony Goldstone discussed whether or not apparent changes in the conventional neuropeptides in the brains of these individuals suggest that we may be looking at a new mechanism or pathway. I would say that the final convergent pathway – however these neuropeptides or other structures are affected – is likely to be the serotonergic components of the nervous system. The serotonergic pathway can account for changes in energy expenditure, energy intake, the gonadotropins, maybe even the carbohydrate preference that was reported here. The relevant gene may turn out to be a small nuclear RNA that is acting through the serotonergic pathway to make these patients resistant to current pharmacological and other approaches. This phenomenon by itself might teach us an enormous amount about normal physiology. Abnormalities of Partitioning Next, PWS can provide mechanistic insights into the partitioning of stored energy between fat and lean body mass. Several speakers touched upon the issue of partitioning. Prader-Willi is one of the most striking examples of a human disorder in which we see very profound abnormalities of partitioning, a relative reduction of lean body mass and an increase in fat mass. This disorder of body composition appears to be due at least in part, or perhaps totally, to derangements in hormones (growth hormone, insulin) that are related to the hypothalamic dysfunction.

Closing Remarks

229

Autonomic Physiology Understanding autonomic physiology, in the control of energy homeostasis, will clearly benefit from research on the PWS. There appear to be reductions in vagal parasympathetic ‘tone’ in patients with PWS. Additional attention should be paid to the other side of the autonomic nervous system, the sympathetic nervous system, which may account for some of the instabilities in thermal regulation in these individuals. Prader-Willi is a striking example of a central autonomic derangement that is clearly due to one or two genes. If we understood autonomic physiology better, we would surely understand how to take better care of patients, not only with this disorder, but with obesity in general. Neuropsychiatric Disorders This syndrome clearly has promise with regard to the molecular physiology of a number of neuropsychiatric disorders, including obsessive behavior and certain types of psychoses. Again, PWS represents either a single gene or certainly a single locus abnormality with very striking, characteristic neuropsychiatric consequences. From understanding the molecular basis of the PWS, we might learn a great deal about gene-specific effects in other neuropsychiatric disorders. Gene Therapy The genetics of PWS are unusual in the following sense: Generally, when we think about genetic insufficiency syndromes, we are dealing with genes that either have alterations of their coding sequences or changes in the regulatory sequences that drive the expression of the gene. In PWS, one copy of the relevant genes, which is all that is needed to have a normal phenotype, is actually intact in these individuals. However, that copy is ‘quiet’ in terms of expression. This disorder may turn out to be an opportunity for ‘gene therapy’ in which the therapy is not to introduce the gene, but simply to turn on the normal allele, that we know is there, but which is quiet as a result of imprinting. Obviously, some of the developmental changes could be reversed unless we are able to make the diagnosis sufficiently early in life. Nonetheless, we might be able to rescue a great deal of the PWS phenotype. Range of Phenotypes And finally, one of the things that is particularly fascinating about this disorder is the very wide range of phenotypes that one sees, even in individuals who have apparently the same genetic aberrations. Or if one takes all of the individuals with deletions or all of the individuals with the other mechanisms for this disorder, the disomies, you do see a fairly wide range of phenotypes with regard to somatic growth and onset of gonadal maturation. These differences are also

Leibel

230

susceptible of genetic analysis and could shed light on the molecular genetics of several important aspects of development. It would be fascinating, for example, to understand in patients with PWS the genetic differences that separate the individuals who actually grow better and have normal gonadal axis function from those who do not, assuming that they have the same underlying genetic disorder for PWS. There must be other modifier genes that are affecting these phenotypes. These represent further ground for study to understand what the modifiers are, particularly in PWS individuals who have an underlying derangement of these systems. This sort of work is underway in mouse models of obesity and in some monogenic human diseases, and I think it might be fruitful in PWS. Obviously, it will take the cooperation of many of the people in this room to be able to get the numbers of individuals to be able to look at this. In summary, I really think PWS is an important model for the understanding of energy homeostasis in humans. Obviously, anything that we can learn about the molecular physiology of PWS will be of immediate relevance and interest to children and parents of children with this disorder, but I think the ripple effects could be significant, and I encourage all of you to continue your excellent efforts because they will help all of us. Prof. Rudolph L. Leibel, Division of Molecular Genetics, Department of Pediatrics, Columbia University College of Physicians and Surgeons, New York, NY 10032 (USA) Tel. ⫹1 212 851 5315, Fax ⫹1 212 851 5306, E-Mail [email protected]

Closing Remarks

231

Author Index

Allen, D.B. 179 Bachmann, S. 190 Bell, J.D. 31 Bloom, S.R. 31 Brynes, A.E. 31 Carrel, A.L. 179 Clegg, D.J. 15 Crinò, A. 140

Goldstone, A.P. 31 Grieser, J. 190 Grugni, G. 140 Grüters, A. 7

Schlüter, B. 128 Schuster, D. 102 Skuse, D. 156 Swaab, D.F. 31

Holland, A. 31

Thomas, E.L. 31 Trang, H. 119

Krude, H. 7 Unmehopa, U.A. 31

Eiholzer, U. 1, 190, 211 Ellis, K.J. 49

l’Allemand, D. 190 Lee, P.D.K. 70 Leibel, R.L. 228 Lindgren, A.C. 86

Frost, G. 31

Morabito, F. 140

Ghatei, M.A. 31 Gilmour, J. 156

Osei, K. 102

Wabitsch, M. 93 Westerterp, K.R. 61 Whitman, B.Y. 198 Woods, S.C. 15 Zipf, W.B. 102

232

Subject Index

Agouti-related peptide (AgRP) appetite control 20, 21, 24, 37, 38 function 37 levels, sudden death of children 48 obesity, role in Prader-Willi syndrome 37, 38 Air displacement plethysmography, body composition assessment 51–53 ␥-Aminobutyric acid (GABA), abnormalities in Prader-Willi syndrome 88 Appetite control adiposity signals 19, 20 central signaling defects, Prader-Willi syndrome 74–76, 85 diabetics 47 hyperphagia, see Hyperphagia hypothalamic signals 20, 21 integration of signals 21, 22 meal patterns 16 meal-related signals 17–19, 22–24 normal regulation 71, 72 overview 15–17 peripheral signaling defects, Prader-Willi syndrome 76–79, 83–85 stable environment 16 Behavior, Prader-Willi syndrome patients adolescents 204, 225, 227 cognitive deficit onset and impact 204–206 eating behavior, see Hyperphagia

environmental management 207, 227 growth hormone replacement therapy, impact 208 neuropsychiatric disorders 230 pharmacological intervention 207, 208, 226 prevalence of traits 204 young children 203, 204 Bioelectric impedance analysis (BIA), body composition assessment 53, 54 Body composition assessment air displacement plethysmography 51–53 bioelectric impedance analysis 53, 54 body mass index 51 computed tomography 54, 55 dual-energy X-ray absorptiometry 54, 55, 57, 82 infants 82 interval between measurements 57 magnetic resonance imaging 54, 55 skinfold measurements 51 growth hormone effects 44, 73, 181–184, 186, 188, 191, 214, 215 insulin regulation 171, 172 leptin defect effects 35, 36 normal regulation 72, 73 pediatric body composition references 56, 57 Prader-Willi syndrome overview 32, 40, 182, 183

233

Body composition (continued) Prader-Willi syndrome (continued) partitioning abnormalities 229 simple obesity versus Prader-Willi syndrome 94 two-compartment model 50 Body mass index (BMI) body composition assessment 51 obesity definition 157 physical activity effects 64 Central model, obesity and hyperphagia in Prader-Willi syndrome 74–76, 85 Cholecystokinin (CCK) appetite control 17–19, 23, 24, 88 interhormonal influences on sensitivity 22 levels, Prader-Willi syndrome 159 Chromosome 15q, abnormalities in PraderWilli syndrome 1, 2, 8, 22, 87, 158, 180, 228, 229 Cocaine- and amphetamine-regulated transcript (CART), obesity role 8, 39 Cognitive deficits, Prader-Willi syndrome 204–206 Computed tomography (CT), body composition assessment 54, 55 C-peptide, Prader-Willi syndrome levels 109, 115 Cryptorchidism, Prader-Willi syndrome 143, 144, 152, 174 db/db mouse 9 Diabetes type 2, see also Glucose tolerance testing insulin injection, effects on appetite 47 pathogenesis 103 polycystic ovary syndrome 142 Prader-Willi syndrome, association 95 Diagnosis, Prader-Willi syndrome criteria 22 difficulty 1 early diagnosis and weight control 176 genetics 2 Dietary restriction, see also Weight control exercise effects combined, Prader-Willi syndrome 65, 66

Subject Index

first intervention, Prader-Willi syndrome 212 Dual-energy X-ray absorptiometry (DXA), body composition assessment 54, 55, 57, 82 Estrogen replacement therapy 174 Exercise, see Physical activity Family psychosocial profiling, primary obese versus Prader-Willi syndrome children appetite disturbance explanations, family 161, 162 appetite questionnaire 160, 161 denial, parents 177, 178 Family Environment Scale scores 160, 162 food preferences 157, 158 limitations of study 163, 177 similarities between groups 162, 163 subjects 159, 177 Fat distribution hormonal control 46 Prader-Willi syndrome 32, 34, 35, 116 sex steroid secretion effects, puberty 141 Fertility female Prader-Willi syndrome patients 3, 144 male Prader-Willi syndrome patients 144 Follicle-stimulating hormone (FSH) obesity effects on levels 143 Prader-Willi syndrome levels 144, 145 releasing hormone response, Prader-Willi syndrome 147–151, 174, 175 Food-related behavioral constellation features, Prader-Willi syndrome 202, 203 management 203 Gastric band, Prader-Willi syndrome children 169 Gene therapy, Prader-Willi syndrome prospects 230 Genital hypoplasia, Prader-Willi syndrome 143

234

Ghrelin functions 23, 36, 88 levels, Prader-Willi syndrome 4, 11, 19, 23, 36 Prader-Willi syndrome obesity, role 11, 36, 47, 48, 88 regulation of secretion 47 Glucose tolerance testing comparison of normal, obese, and Prader-Willi syndrome subjects calculations and statistical analysis 109 chemical analyses 109 correlation coefficients 113 C-peptide levels 109, 115 hepatic insulin extraction 112, 113, 115 insulin clearance 112, 113, 115 intravenous glucose tolerance test 108, 110, 112, 114 oral glucose tolerance test 108, 110, 114 overview 105, 106 study design 107, 108 subjects 106, 107, 109, 110 mixed-meal stimulation, Prader-Willi syndrome glucose response 104, 105 insulin values and response 105 oral versus intravenous glucose tolerance testing 103 pancreatic polypeptide response 104 study design 104 GLUT1, growth hormone effects on levels 98 GLUT4, Prader-Willi syndrome levels 171 Gonadotropin-releasing hormone (GnRH) deficiency and body composition effects 76–79 obesity, role 40, 76–79 polycystic ovary syndrome response 143 Prader-Willi syndrome levels 144, 145 response follicle-stimulating hormone 147–151, 174, 175

Subject Index

luteinizing hormone 147–151, 174, 175 study design 146, 147 subjects 146 Growth hormone (GH) adipose tissue effects GLUT1 levels 98 growth promotion 97, 170 in vitro studies 98, 170 lipolysis 97, 170, 182 receptors 99 biological activity 96, 97, 181, 182 body composition effects 44, 73, 181, 182 deficiency Prader-Willi syndrome 2, 75, 76, 100, 181–183 visceral adiposity 35 gonadal function, role 152 low-density lipoprotein, effects on receptors 96, 97 therapy adults 224, 225 age of initiation 223, 224 behavior impact 208 body composition effects 183, 184, 186, 188, 191, 214, 215 dosing 222 energy expenditure effects 186 fat utilization 186 linear growth response 213, 214, 224 overview 2–4, 100 reduction in response 188 safety 187 sleep disorder management 136, 137 strength and agility response 186, 187 Growth hormone releasing hormone (GHRH) deficiency and body composition effects 76, 77 obesity, role in Prader-Willi syndrome 39, 40, 76–78 Hepatic insulin extraction (HIE), Prader-Willi syndrome 112, 113, 115

235

Hyperphagia age of onset, Prader-Willi syndrome 203, 204 behavioral studies 89, 90, 166–169 eating rate, Prader-Willi syndrome 166–169 family psychosocial profiling, see Family psychosocial profiling, primary obese versus Prader-Willi syndrome children growth hormone therapy response 223 hypotonia effects 168 management 4, 5 physiological control, see Appetite control pro-opiomelanocortin mutations 46 Hypersomnolence Prader-Willi syndrome 129, 131, 132 self-reporting 130–132 Hypogonadism gonadotropin-releasing hormone response, Prader-Willi syndrome 147–151 nutritional status, impact 141 Prader-Willi syndrome features 144–146, 174, 175 Hypothalamus, see also specific hormones appetite control signals 20, 21 defects, Prader-Willi syndrome 199, 200, 229 Hypotonia, Prader-Willi syndrome 168, 180, 181, 183, 200, 201 Infants body composition assessment 82 developmental delays 200 hypotonia 168, 183, 200, 201 physical activity 66 suckling patterns, Prader-Willi syndrome 168 Insulin, see also Glucose tolerance testing abnormalities, Prader-Willi syndrome 4, 22 anabolic actions 171 appetite control 20, 46, 47 clearance, Prader-Willi syndrome 112, 113, 115 diabetes, injection effects on appetite 47

Subject Index

interhormonal influences on sensitivity 22 sensitivity, Prader-Willi syndrome 115, 116, 171 simple obesity versus Prader-Willi syndrome response 94, 95 Insulin-like growth factor-1 (IGF-1) deficiency, Prader-Willi syndrome 2, 77, 78, 181 hypoglycemic response 169, 170 Leptin appetite control 19, 20 growth hormone therapy response 222, 223 interhormonal influences on sensitivity 22, 88 mouse models, obesity 8, 9, 83, 84 Prader-Willi syndrome obesity, role 10, 23, 35, 36, 44, 45 puberty onset, role 141 therapy 44 Low-density lipoprotein (LDL) growth hormone effects on receptors 96, 97 simple obesity versus Prader-Willi syndrome levels 95, 96 Luteinizing hormone (LH) obesity effects on levels 143 Prader-Willi syndrome levels 144, 145 releasing hormone response, Prader-Willi syndrome 147–151, 174, 175 Magnetic resonance imaging (MRI), body composition assessment 54, 55, 84 MCR4 agonists 12 human mutations, obesity 9–11, 37 Prader-Willi syndrome obesity, role 10 MC receptors, see also MCR4 functions 45, 46 types 45 Melanin-concentrating hormone (MCH), Prader-Willi syndrome obesity 39 Melanocyte-stimulating hormone (MSH), appetite control 20, 37

236

Metacognition, Prader-Willi syndrome deficits 206 Modafinil, Prader-Willi syndrome children response 173 Monogenetic obesity db/db mouse 9 human gene mutations 9 Nhlh2 knockout mouse 9 ob/ob mouse 8, 9 Sim1 knockout mouse 9 Multidisciplinary management, Prader-Willi syndrome 208, 209 Naloxone, appetite response in Prader-Willi syndrome 158, 159 Narcolepsy cataplectic episodes 134 obesity association 124, 125 Nasal continuous positive airway pressure, sleep disorder management 125, 136 Neuropeptide Y (NPY) appetite control 20, 21, 24, 37, 38 levels, sudden death of children 48 obesity, role in Prader-Willi syndrome 37, 38, 87 Nhlh2 knockout mouse 9 Obesity age at onset, Prader-Willi syndrome 22, 87 common obesity genetic screening 12 definition 157 food intake, see Appetite control; Hyperphagia metabolic defects 95, 96 monogenetic, see Monogenetic obesity narcolepsy association 124, 125 prevention, see Weight control risk factors 157, 158 sleep-disordered breathing, see Sleepdisordered breathing ob/ob mouse 8, 9, 83 Obstructive sleep apnea, see Sleepdisordered breathing Overeating, see Hyperphagia; Obesity Oxytocin, Prader-Willi syndrome obesity 39, 41, 87

Subject Index

Pancreatic polypeptide appetite control 24, 88 deficiency, Prader-Willi syndrome 115 response, Prader-Willi syndrome 104 Paraventricular nucleus (PVN) appetite control 20, 21, 37–39 morphology, Prader-Willi syndrome 48, 87 Peripheral model, obesity and hyperphagia in Prader-Willi syndrome 76–79, 83–85 Phenotypic range, Prader-Willi syndrome 230, 231 Physical activity capacity, Prader-Willi syndrome children 217, 218 course from infancy and early childhood to adolescence, Prader-Willi syndrome 66, 67 daily exercise program, Prader-Willi syndrome 216–218 doubly labeled water assessed physical activity studies activity-induced energy expenditure calculation 63 body mass index changes 64 dietary restriction plus exercise effects 65, 66 physical activity level calculation 62, 63 subject characteristics 62 family education 201 muscle hypotonia 83 obesity control benefits 67, 68 overview, obesity impact studies 61, 62 Pickwickian syndrome 174 Pituitary gland, hypoplasia in Prader-Willi syndrome 87 Polycystic ovary syndrome (PCOS) gonadotropin-releasing hormone response 143 hormone levels 142 insulin resistance 142, 143 obesity association 142 Polysomnography, sleep-disordered breathing 121–123, 125, 132 Prevalence, Prader-Willi syndrome 103, 180

237

Pro-opiomelanocortin (POMC) appetite control 20, 21 human mutations, obesity 9, 37, 46 Prader-Willi syndrome obesity, role 10, 11, 39 processing 8 Psychological counseling, families 218, 219 Puberty, Prader-Willi syndrome features 144 Risperidone, behavioral problem management 226 Scoliosis 200 Selective serotonin reuptake inhibitors (SSRIs), behavioral problem management 226 Serotonin, turnover in Prader-Willi syndrome 87, 88 Sim1 human mutations, obesity 9 knockout mouse 9 Skinfold measurements, body composition assessment 51 Sleep, see also Hypersomnolence; Narcolepsy; Sleep-disordered breathing REM sleep, Prader-Willi syndrome 133, 134 sleep-wake cycle, Prader-Willi syndrome 133, 134 Sleep-disordered breathing (SDB) children versus adults 123, 124, 132 management, children growth hormone therapy 136, 137 nasal continuous positive airway pressure 125, 136 tonsillectomy/adenoidectomy 125 weight loss 125, 136 mechanisms, obesity autonomic nervous system dysfunction 135 chemoreceptor function 124, 135, 136 genetic predisposition 124

Subject Index

pulmonary function 124, 135 upper airway narrowing 124, 134, 135 mortality 173, 174 obesity association 120 polysomnography 121–123, 125, 132 prevalence, children all children 119, 130, 131 obese children 121, 131–133 self-reporting 130–132 symptoms 120 Snoring, see Sleep-disordered breathing Speech and language, Prader-Willi syndrome deficits and management 201, 202 Sweating, Prader-Willi syndrome features 172, 173 Testosterone Obesity, effects on levels 143 replacement therapy 174, 175 Topiramate, behavioral problem management 226 Visceral adiposity insulin modulation 171 Prader-Willi syndrome 32, 34, 35 sex steroid profile 142 Weight control, see also Family psychosocial profiling, primary obese versus Prader-Willi syndrome children dietary restriction 212 disposition to overweight, families 191, 192 exercise, see Physical activity growth hormone therapy, see Growth hormone nutritional intake, Prader-Willi syndrome children 191–196 parenting style influences 191, 192, 194, 195 psychological counseling, families 218, 219

238