Nutritional Management of Diabetes Mellitus And Dysmetabolic Syndrome (Nestle‚ Nutrition Workshop Series: Clinical & Performance Program)

Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome

Nestlé Nutrition Workshop Series Clinical & Performance Program, Vol. 11

Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome Editors J.P. Bantle, Minneapolis, MN, USA G. Slama, Paris, France

Nestec Ltd., 55 Avenue Nestlé, CH–1800 Vevey (Switzerland) S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland) www.karger.com © 2006 Nestec Ltd., Vevey (Switzerland) and S. Karger AG, Basel (Switzerland). All rights reserved. This book is protected by copyright. No part of it may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, or recording, or otherwise, without the written permission of the publisher. Printed in Switzerland on acid-free paper by Reinhardt Druck, Basel ISSN 1422–7584 ISBN-10: 3–8055–8095–9 ISBN-13: 978–3–8055–8095–3 Library of Congress Cataloging-in-Publication Data Nutritional management of diabetes mellitus and dysmetabolic syndrome / editors, J.P. Bantle, G. Slama. p. ; cm. – (Nestlé Nutrition Workshop series. Clinical & performance program ; v. 11) Includes bibliographical references and index. ISBN 3-8055-8095-9 (hard cover : alk. paper) 1. Diabetes mellitus–Diet therapy. 2. Diabetes mellitus–Nutritional aspects. 3. Metabolic syndrome–Diet therapy. 4. Metabolic syndrome –Nutritional aspects. I. Bantle, John P., 1947- . II. Slama, G. III. Nestlé Nutrition Institute. IV. Series: Nestlé Nutrition workshop series. Clinical & performance programme ; v. 11. [DNLM: 1. Diabetes Mellitus–diet therapy. 2. Metabolic Syndrome X–diet therapy. W1 NE228C v.11 2006 / WK 818 N9769 2006] RC662.N893 2006 616.4⬘620654–dc22 2006010085

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

The material contained in this volume was submitted as previously unpublished material, except in the instances in which credit has been given to the source from which some of the illustrative material was derived. Great care has been taken to maintain the accuracy of the information contained in the volume. However, neither Nestec Ltd. nor S. Karger AG can be held responsible for errors or for any consequences arising from the use of the information contained herein.

Contents

VII Foreword XI Contributors The Dysmetabolic Syndrome 1 The Dysmetabolic Syndrome: Epidemiology and Etiology Sauerwein, H.P. (The Netherlands) 15 Traditional Chinese Medicine in the Treatment of Diabetes Zhao, H.-L.; Tong, P.C.Y.; Chan, J.C.N. (Hong Kong, SAR, China) 31 Pharmacological and Surgical Intervention for the Prevention of Diabetes Chiasson, J.-L. (Canada) Glycemic Effect of Carbohydrates 43 The Glycemic Index: Methodology and Use Kendall, C.W.C.; Augustin, L.S.A.; Emam, A.; Josse, A.R.; Saxena, N.; Jenkins, D.J.A. (Canada) 57 The Argument against Glycemic Index: What Are the Other Options? Franz, M.J. (USA) 73 Low Glycemic Index Foods Should Play a Role in Improving Overall Glycemic Control in Type-1 and Type-2 Diabetic Patients and, More Specifically, in Correcting Excessive Postprandial Hyperglycemia Slama, G.; Elgrably, F.; Kabir, M.; Rizkalla, S. (France)

V

Contents 83 Is Fructose the Optimal Low Glycemic Index Sweetener? Bantle, J.P. (USA) Beyond Glycemic Control 97 Optimal Diet for Glycemia and Lipids Knowler, W.C. (USA) 107 Antioxidants and Diabetes Mooradian, A.D. (USA) 127 Dietary and Body Weight Control: Therapeutic Education, Motivational Interviewing and Cognitive-Behavioral Approaches for Long-Term Weight Loss Maintenance Golay, A. (Switzerland) Diabetes in the Life Cycle 139 The Accelerator Hypothesis: A Unifying Explanation for Type-1 and Type-2 Diabetes Wilkin, T.J. (United Kingdom) 155 Diet and Medical Therapy in the Optimal Management of Gestational Diabetes Mellitus Metzger, B.E. (USA) 171 Do Meal Replacement Drinks Have a Role in Diabetes Management? Ditschuneit, H.H. (Germany) The Role of Drugs and Diet Therapy – Alone and Together 183 Physical Activity in Prevention and Management of Obesity and Type-2 Diabetes Hill, J.O.; Stuht, J.; Wyatt, H.R.; Regensteiner, J.G. (USA) 197 The Role of Lifestyle Modification in Dysmetabolic Syndrome Management Foreyt, J.P. (USA) 207 Critical Review of the International Guidelines: What Is Agreed upon – What Is Not? Katsilambros, N.; Liatis, S.; Makrilakis, K. (Greece) 219 Subject Index

VI

Foreword

Globally, the number of persons with diabetes and at risk of diabetes and cardiovascular disease is reaching epidemic proportions. Over the next decade the number is expected to grow by 25%, largely driven by the rising prevalence of obesity and inactivity. The World Health Organization (WHO) estimates that 200 million persons worldwide will have diabetes by 2010, and that number will reach 330 million by 2025. The problem is especially serious in Asia where there are 90 million people with diabetes. This includes four of the world’s five largest populations with diabetes: India, 33 million people with diabetes; China, 23 million; Pakistan, 9 million, and Japan, 7 million. The WHO predicts that in less than a decade, 60% of the worldwide population with diabetes will be in Asia. It is with these facts in mind that Nestlé Nutrition chose the topic ‘Nutritional Management of Diabetes Mellitus and the Dysmetabolic Syndrome’ for the 11th Nestlé Nutrition Clinical and Performance Program Workshop Series, and the site, Hangzhou, China. Unless action is taken to change the predicted path of diabetes, the disease will become a huge economic burden – both from direct healthcare costs and indirect costs due to a decline in workplace productivity, as well as losses due to premature morbidity and mortality. Therefore, individuals at risk of diabetes must be identified, and prevention and suitable treatment interventions implemented. With Nestlé Nutrition’s interest in nutrition, and the superb expertise of our chairmen, Prof. John Bantle and Prof. Gerard Slama, a program was developed highlighting the scientific evidence examining the impact of lifestyle, which includes both nutritional management and physical activity, in the prevention and treatment of diabetes. Although the optimal diet for diabetes has not yet been defined, because the optimal study has not yet been done, there is general agreement that the nutritional recommendations of the various international diabetes organizations are reasonable. There was also general agreement that nutrition and lifestyle management must be individualized for each person to enhance the potential for a successful outcome.

VII

Foreword We thank our superb chairmen, our expert speakers, and especially our hosts from Nestlé China, specifically KeLan Liu and Kelly Gao and their team, for their wonderful organization and attention to so many logistical details. It is due to their efforts that participants from around the globe were able to participate in a stimulating workshop and some wonderful Chinese culture. Patricia S. Anthony, MS, RD Manager, Clinical Services HealthCare Nutrition Nestec Ltd., Vevey, Switzerland

VIII

11th Nestlé Nutrition Workshop Clinical & Performance Program Hangzhou, China, October 30–November 3, 2005

Contributors

Chairpersons and Speakers Prof. John P. Bantle Division of Endocrinology and Diabetes Department of Medicine University of Minnesota 420 Delaware Street SE Minneapolis, MN 55455 USA E-Mail [email protected]

Prof. Jean-Louis Chiasson Research Center – CHUM (Hôtel Dieu) 3850 St. Urbain Street Montréal, Qué. H2W 1T7 Canada E-Mail jean.louis.chiasson@ umontreal.ca

Marion J. Franz, MS, RD, CDE Nutrition Concepts by Franz, Inc. 6635 Limerick Drive Minneapolis, MN 55439 USA E-Mail [email protected]

Prof. Alain Golay Service of Therapeutic Education for Chronic Diseases Diabetes – Obesity Department of Community Medicine University Hospital of Geneva 24, rue Micheli-du-Crest CH–1211 Geneva 14 Switzerland E-Mail [email protected]

Dr. Herwig H. Ditschuneit Medizinische Universitätsklinik Universitätsklinikum Ulm Robert-Koch-Strasse 8 DE–89081 Ulm Germany E-Mail herwig.ditschuneit@ medizin.uni-ulm.de

Dr. James O. Hill Center for Human Nutrition University of Colorado School of Medicine Campus Box 263 Denver, CO 80262 USA E-Mail [email protected]

Prof. John P. Foreyt Behavioral Medicine Research Center Baylor College of Medicine 6655 Travis Street, Suite 320 Houston, TX 77030 USA E-Mail [email protected]

Prof. Nicholas Katsilambros Laiko General Hospital Agiou Thoma 17 Street (Goudi) GR–11527 Athens Greece E-Mail [email protected]

XI

Contributors Prof. Cyril W.C. Kendall Department of Nutritional Sciences Faculty of Medicine University of Toronto Clinical Nutrition and Risk Factor Modification Center St. Michael’s Hospital Toronto, Ont. M5S 3E2 Canada E-Mail [email protected]

Prof. William C. Knowler Diabetes Epidemiology and Clinical Research Section National Institute of Diabetes and Digestive and Kidney Diseases 1550 E Indian School Road Phoenix, AZ 85014 USA E-Mail [email protected]

Prof. Boyd E. Metzger Northwestern University 15-735 Tarry Building 303 East Chicago Avenue Chicago, IL 60611 USA E-Mail [email protected]

Prof. Hans Peter Sauerwein Department of Endocrinology and Metabolism F5-170 Meibergdreef 9 NL–1105 AZ Amsterdam The Netherlands E-Mail [email protected]

Prof. Gérard Slama Department of Diabetes Hôtel Dieu Hospital 1, place du Parvis Notre Dame FR–75004 Paris France E-Mail gerard.slama@ htd.ap-hop-paris.fr

Prof. Terence Wilkin Department of Medicine Postgraduate Medical School Level 7, Derriford Hospital Plymouth PL6 8DH UK E-Mail terence.wilkin@ phnt.swest.nhs.uk

Dr. Hailu Zhao Dr. Arshag D. Mooradian Division of Endocrinology Saint Louis University 1402 S. Grand Blvd St. Louis, MO 63104 USA E-Mail [email protected]

Department of Medicine and Therapeutics Prince of Wales Hospital Ngan Shing Street 30-32 Shatin, Hong Kong, SAR China E-Mail [email protected]

Moderators Dr. Peter C.Y. Tong

Shanghai 200025 China E-Mail [email protected]

Prof. Mingdao Chen

Medicine Sun Yet-Sen University No. 135, Xin Gang Xi Road Guangzhou 510275 China

Prince of Wales Hospital Shatin, Hong Kong, SAR China E-Mail [email protected]

Prof. Zuzhi Fu Shanghai Institute of Endocrine and Metabolic Diseases 197 Ruijin Road II

XII

Contributors Prof. Low-Thone Ho Taipei VHG: 201 Shih-Pai Road, Section 2 Taipei Taiwan E-Mail [email protected]

Prof. Hongding Xiang Department of Endocrine Peking Union Medical College Hospital No. 1, Shuai Fu Yuan Dongcheng District Beijing 100730 China E-Mail [email protected]

Invited attendees Dr. Daniel Giannella Neto / Brazil Prof. Catherine Field / Canada Prof. Donglian Cai / China Prof. Jialun Chen / China Prof. Xiafei Chen / China Prof. Zongyi Ding / China Prof. Sheng Ge / China Prof. Renming Hu / China Prof. Linong Ji / China Prof. Wiing Jia / China Prof. Zhimin Liu / China Prof. Rongli Qian / China Prof. Jianqin Sun / China Prof. Mengli Sun / China Prof. Haoming Tian / China Prof. Jianping Weng / China Prof. Manying Xu / China Prof. Huixia Yang / China Prof Yongnian Yang / China Prof. Demin Yu / China Prof. Serge Halimi / France Dr. Patrick Serog / France Prof. Elisabeth Steinhagen-Thiessen / Germany Ms. King Chi, June Chan / Hong Kong

Nestlé Nutrition participants Mrs Penelope Small / Australia Dr. Olivier Ballevre / China Mrs. Bénédicte Sentenac / France Ms. Mandy Ma / Hong Kong Mr. Satoru Okada / Japan Ms. Jean Ang / Malaysia Mrs. Amelita Valenzuela / Philippines Ms. Ai-joo, Alicia Ng / Singapore Mrs Patricia Anthony / Switzerland Dr. Denis Barclay / Switzerland Mr. Dominique Brassart / Switzerland

Mr. Yung Kind, David Chan / Hong Kong Ms. Tsui Fun, Lornea Cheung / Hong Kong Dr. Chun Chung Chow / Hong Kong Ms. Wai Shan, Wendy Tam / Hong Kong Dr. Johanes Casay Chandrawinata / Indonesia Prof. Khalid Abdul Khadir / Malaysia Dr. Francisco Lagrutta / Panama Dr. Carlos Velarde / Panama Dr. Roberto Mirasol / Philippines Dr. Rosa Allyn Sy / Philippines Dr. Saddah Eshki / Saudi Arabia Dr. Chee Fang Sum / Singapore Dr. Kaushik Ramaiya / South Africa Dr. Gabriel Olveira / Spain Prof. Lee-Ming Chuant / Taiwan Dr. Chao-Hung Wang / Taiwan Dr. Natapong Kosachunhaunun / Thailand Dr. Apussanee Boonyavarakul / Thailand Dr. Nattachet Plengvidhya / Thailand Dr. Linda Wilkin / UK

Nestlé Nutrition Participants Dr. Jason Chieh Chou / Switzerland Prof. Ferdinand Haschke / Switzerland Dr. Natalia Leonova / Switzerland Dr. Eduardo Schiffrin / Switzerland Dr. Thomas Schweizer / Switzerland Mr. Pierre Wuersch / Switzerland Ms. Patricia Lee / Taiwan Mr. Keith Colin-Thome / Thailand Ms. Wirudchada Suttayakom / Thailand Ms. Fabienne Le Tadic / UK

XIII

The Dysmetabolic Syndrome Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 1–13, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

The Dysmetabolic Syndrome: Epidemiology and Etiology H.P. Sauerwein Department of Endocrinology and Metabolism, Academic Medical Centre, Amsterdam, The Netherlands

Abstract The metabolic syndrome is a common metabolic disorder that results from the increasing prevalence of obesity. It also refers to a clustering of specific cardiovascular disease risk factors whose underlying pathophysiology is thought to be related to insulin resistance with an excessive flux of fatty acids implicated. Opinions have varied as to whether the metabolic syndrome should be defined to indicate mainly insulin resistance, the metabolic consequences of obesity, risk of cardiovascular disease, or simply a collection of statistically related factors. Based on these different viewpoints 4 definition sets of the metabolic syndrome are formulated. The pros and cons of each of them are extensively discussed. A major role in the etiology of the metabolic syndrome is ascribed to the occurrence of insulin resistance. Data are provided that insulin resistance can worsen the expression of this syndrome, but cannot have a primary role. Therefore, insulin resistance is not the main player of the metabolic syndrome, but central obesity is. Free fatty acid induced insulin resistance is found and induced by central obesity. The metabolic syndrome is a cluster of abnormalities in which each of them deserves its own (maximal) treatment to diminish the risk for cardiovascular disease. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

Definition, Epidemiology and Its Problems The metabolic syndrome is a common metabolic disorder that results from the increasing prevalence of obesity. It also refers to a clustering of specific cardiovascular disease (CVD) risk factors, whose underlying pathophysiology is thought to be related to insulin resistance with excessive flux of fatty acids implicated. A pro-inflammatory state probably contributes to the syndrome 1

Sauerwein Table 1. Comparison of definitions of the metabolic syndrome WHO

EGSIR

Diabetes or IFG or IGT or IR (clamp) Plus 2 of the following: BMI ⬎30 or WHR ⬎0.9 (male); ⬎0.85 (female) TG ⬎1.7 mmol/l or HDL ⬍0.9 mmol/l (male) ⬍1.0 mmol/l (female) Blood pressure ⬎140/90 mm Hg

Fasting insulin: top 25% non-diabetics Plus 2 of the 3 or more of the following: following: Waist ⬎94 cm (male) Waist ⬎102 cm (male) ⬎80 cm (female) ⬎88 cm (female)

Albumin excretion ⬎20 ␮g/min

TG ⬎2.0 mmol/l or HDL ⬍1.0 mmol/l Blood pressure ⬎140/90 mm Hg or antihypertensive medication FPG ⬎6.1 mmol/l

ATP III

TG ⬎1.7 mmol/l HDL ⬍1.0 mmol/l (male) ⬍1.3 mmol/l (female) Blood pressure ⬎140/90 mm Hg or antihypertensive medication FPG ⬎6.1 mmol/l

ATP III ⫽ National Cholesterol Education Program’s Adult Treatment Panel III (NCEP:ATP III); BMI ⫽ body mass index; EGSIR ⫽ European Group for the Study of Insulin Resistance; FPG ⫽ fasting plasma glucose concentration; IFG ⫽ impaired fasting plasma glucose; IGT ⫽ impaired glucose tolerance; IR ⫽ insulin resistance; TG ⫽ plasma triglyceride concentration; WHR ⫽ waist-hip ratio.

[1, 2]. Until 1998 an internationally recognized definition did not exist. In that year a WHO consultation proposed a set of criteria (diabetes mellitus or impaired fasting glycemia or insulin resistance plus 2 or more of obesity, dyslipidemia, hypertension, or microalbuminuria). Subsequently, the National Cholesterol Education Program’s Adult Treatment Panel III (NCEP:ATP III; 3 or more of obesity, dyslipidemia, hypertension or impaired fasting plasma glucose) and the European Group for the Study of Insulin Resistance (EGSIR; insulin resistance plus 2 or more of obesity, dyslipidemia, hypertension or impaired fasting plasma glucose) have formulated definitions. The WHO definition and that of EGSIR agree that they both include either glucose intolerance or insulin resistance as an essential component. However, for the NCEP:ATP III definition, this criterion is not included [1] (table 1). Despite these differences in definition, they also have major similarities. All include dyslipidemia, hypertension and a parameter for obesity. Confusion about the relevance of considering the metabolic syndrome as a separate disease entity relates to differences in opinion regarding its definition. Opinions have varied as to whether the metabolic syndrome should be defined to indicate: (1) mainly insulin resistance; (2) the metabolic consequences of 2

The Dysmetabolic Syndrome: Epidemiology and Etiology obesity; (3) the risk of CVD, or (4) simply as a collection of statistically related factors. In 1988, it was proposed that individuals displaying the cluster of abnormalities associated with insulin resistance/compensatory hyperinsulinemia (glucose intolerance, hypertriglyceridemia, low high-density lipoprotein and essential hypertension) were at a significantly increased risk of CVD. Because the importance of insulin resistance and the associated abnormalities were not widely appreciated as CVD risk factors at that time, the cluster of associated abnormalities was subsumed under the rubric of syndrome X. Since the introduction of the concept of syndrome X considerable information has evolved relevant to the role of insulin resistance in human disease. This has resulted in two somewhat disparate approaches to thinking about the clinical implications of insulin resistance and its consequences. One view recognizes that the abnormalities related to insulin resistance have broadened considerably, and the adverse clinical outcomes extend beyond type-2 diabetes mellitus and CVD. Because CVD is recognized to be just one of the multiple clinical syndromes, it seemed appropriate to replace the term syndrome X by the term ‘insulin resistance syndrome’. The cardiologic community (ATP III) recognized the importance of a ‘constellation of lipid and non-lipid risk factors of metabolic origin’ to be important as CVD risk factors, added abdominal obesity to the abnormalities initially proposed to comprise syndrome X, designated this cluster as the metabolic syndrome and stated ‘this syndrome is closely related to insulin resistance’. However, on the other hand, the stated purpose of ATP III was to provide criteria to make the clinical diagnosis of metabolic syndrome, not to provide a physiological construct to explain why insulin-resistant subjects are at increased CVD risk [3]. This divergent approach (simple criteria for the diagnosis of metabolic syndrome and emphasis on a physiological construct to explain why insulin-resistant subjects are at increased CVD risk) was greatly enhanced by their diagnostic criteria. As shown in table 1, only three of their criteria are necessary to diagnose the metabolic syndrome. This leaves the possibility that a lean glucose-tolerant subject can be labelled as being a patient with the metabolic syndrome. It is obvious that this divergent approach causes major problems for epidemiological studies and treatment. Because of its ease for use, the NCEP:ATP III criteria are used most frequently. In those studies the prevalence varies in urban populations from 8% (India) to 24% (USA) in men, and from 7% (France) to 43% (Iran) in women. A very consistent finding is that the prevalence of the metabolic syndrome is highly age-dependent. This pattern is clear in Iran where the prevalence is ⬍10% for both men and women in the 20- to 29-year age group, rising to 38 and 67%, respectively, in the 60- to 69-year age group. Similarly, in a French population, the prevalence rises from ⬍5.6% in the 30- to 39-year age group to 17.5% in the 60- to 64-year age group. The prevalence data for the USA are comparable to those for Iran [1]. A recent study in 15,540 Chinese adults confirmed the rather 3

Sauerwein high prevalence in developing countries [4]. The age-standardized prevalence was 9.8% in men and 17.8% in women. The age-related prevalence showed the same tendencies as anywhere else in the world, albeit less steep. It was 8.4% in men for the 35- to 44-year age group and 10.4% in the men aged 65–74 years. In women these figures were 9.4 and 28.6%, respectively [5]. Knowledge of the impact of the metabolic syndrome according to standard definitions on the cardiovascular and overall mortality in the general population is crucial for developing public health policy and clinical guidelines for its prevention and treatment [5]. However, comparisons of the published prevalence of the metabolic syndrome for different populations are difficult despite attempts to reach agreement on the definition of the metabolic syndrome. When different definitions are applied to the same study population, the prevalence can differ by ⬃60% [5]. It is even more troublesome that many studies compare prevalences using different criteria. It can be stated that the ultimate importance of the recognition of the metabolic syndrome as a separate entity is that it helps to identify individuals at high risk of both type-2 diabetes and CVD. However, the different definitions are not equal in this respect. For Finland and Italy the WHO/EGSIR criteria are better than the NCEP:ATP III definition, while in the San Antonio study the reverse was found [6]. To promote consistency in epidemiological research related to the metabolic syndrome, the American Diabetes Association and the European Association for the Study of Diabetes analyzed the existing literature (mainly based on ATP III criteria) and concluded that the metabolic syndrome has been imprecisely defined and existing diagnostic criteria did not consider many other related CVD risk factors [2]. In 2005 the International Diabetes Federation (IDF) released a consensus clinical definition of the metabolic syndrome for worldwide use that included central obesity as a prerequisite. The IDF definition varied from the ATP III criteria with different criteria for glucose intolerance (table 2). A major additional difference was the inclusion of criteria for obesity in different ethnic groups, as the risk of type-2 diabetes is apparent at much lower levels of adiposity in Asian populations than in the European population [6]. This stricter definition can create additional problems. Comparison of the ATP III definition and the new IDF criteria in a representative sample of ⬃4,000 subjects in South Australia demonstrated that the IDF definition categorized 15–20% more people as having the metabolic syndrome. The IDF recommends ‘aggressive and uncompromising’ management of those classified to reduce CVD and diabetes. If this definition regains widespread acceptance, then substantially more people will receive management, including drug therapy [7]. Time will tell whether this increased cost in monetary and other terms is justified. Time will also tell whether these new criteria will resolve the discussion about the diagnostic criteria for the metabolic syndrome. 4

The Dysmetabolic Syndrome: Epidemiology and Etiology Table 2. Comparison of definitions of the metabolic syndrome by IDF and ATP III Parameter

IDF criterion

Comparison to ATP III

Obesity

Waist circumference: ethnic specificity TG ⬎1.7 mmol/l or specific treatment Male: ⬍1.0 mmol/l female: ⬍1.3 mmol/l or specific treatment Blood pressure: ⬎130/85 mm Hg or specific treatment FPG: ⬎5.6 mmol/l or previously diagnosed diabetes mellitus

Different

Triglycerides HDL

Hypertension Glucose

Same Same

Same More strict

ATP III ⫽ National Cholesterol Education Program’s Adult Treatment Panel III (NCEP:ATP III); FPG ⫽ fasting plasma glucose concentration; IDF ⫽ International Diabetes Federation; IGT ⫽ impaired glucose tolerance; TG ⫽ plasma triglyceride concentration.

Etiology Medical science usually defines a syndrome as an ‘aggregate of symptoms and signs associated with any morbid process, and constituting together the picture of disease’. The specific signs and symptoms are usually caused by a unifying underlying pathology, and their components confer a risk that is different from the sum of its parts. This definition does not seem to be applied in the discussions about the pathophysiology of the metabolic syndrome. In this discussion sometimes cause and effect are reversed, especially about the relation between free fatty acids (FFAs) and insulin sensitivity [1]. It has been stated that accumulating evidence strongly indicates that insulin resistance is the common pathogenetic factor for the individual components of the metabolic syndrome and explains the trait cluster [2, 8]. In many of the studies on the relation between insulin resistance and the occurrence of metabolic syndrome, surrogate measures of insulin resistance were used and these measures often loaded on more than one of the underlying factors. Few studies have examined the associations between the metabolic syndrome and direct measures of insulin sensitivity as the euglycemichyperinsulinemic clamp or the frequently sampled intravenous glucose tolerance test. Applying those techniques, still strong associations have been found between low insulin sensitivity and the metabolic syndrome in non-diabetic subjects. Subjects with ATP III- or WHO-defined metabolic syndrome had 5

Sauerwein 5- to 10-fold increased risks of being in the lowest quartile of directly measured insulin sensitivity [9]. It has been concluded that these observations provide strong support for the notion that individuals with the metabolic syndrome are insulin-resistant and this disorder may be at the core of the cluster of metabolic abnormalities that characterizes the syndrome [9]. Although the conclusion that ‘insulin resistance may be at the core of cluster of abnormalities’ is carefully formulated and does not state that insulin resistance is the cause of the metabolic syndrome, the implicit suggestion is that it is. Is this conclusion justified? In other words, can the cluster of abnormalities together forming the metabolic syndrome be explained by diminished insulin action? A series of arguments will be provided to prove that this is not true. In order to fulfill its role as a unifying underlying pathology, insulin resistance should not only be strongly related to the metabolic syndrome, but should also be present in every patient with the metabolic syndrome. This requirement is not fulfilled. In the definition proposed by WHO and EGSIR a central role is given to insulin resistance; however, according to the ATP III criteria, the existence of glucose intolerance or insulin resistance is not a prerequisite for diagnosis of the metabolic syndrome and ATP III criteria have a low sensitivity for identifying insulin resistance with dyslipidemia in non-diabetic individuals at increased risk for CVD [8]. This suggests that it could be possible that subjects diagnosed as having the metabolic syndrome do not have this abnormality which is considered to be essential for development of this syndrome. This suggestion proved to be true. In a study of 443 healthy volunteers in the USA, in whom insulin sensitivity was measured with the gold standard, it was shown that ⬃21% of the subjects evaluated met the ATP III criteria for identification of the metabolic syndrome. Approximately two thirds of these subjects were insulin-resistant. This shows a high correlation between the metabolic syndrome and insulin resistance, but more importantly it also shows that ⬃30% were insulin-sensitive [10]. A recent study, reported in the Annals of Internal Medicine on 258 obese non-diabetic subjects, showed that 78% of those with the metabolic syndrome were insulin-resistant and 48% with insulin resistance met the criteria of the metabolic syndrome, again values too low for insulin resistance to be a cause of the syndrome [11]. Another approach could be to show that treatment of insulin resistance alone will cure or improve the abnormalities of the metabolic syndrome. There are no data showing this. Treatment of the metabolic syndrome is treatment of its different components. There are no data showing that treatment of all its components adds something extra above this. There are also no data showing that treatment of one component ‘cures’ the other components [1]. It has been suggested that peroxisome proliferator-activated receptor-␥ agonists could be ideal agents for managing the metabolic syndrome, as they reduce insulin resistance by influencing FFA flux. Systematic studies are lacking, but even if treatment with this agent alone improves or cures the metabolic syndrome, this cannot be used as an argument for insulin resistance being the core player 6

The Dysmetabolic Syndrome: Epidemiology and Etiology in the development of the metabolic syndrome. These agents have pleiotropic effects far beyond improving insulin resistance [12]. The third approach to explore the role of insulin resistance in the development of the metabolic syndrome is to look at studies on long-term insulin administration and the development of the metabolic abnormalities of the metabolic syndrome. Insulin inhibits glucose production and stimulates glucose uptake – oxidation and glycogen synthesis in the insulin-sensitive tissues, muscle and adipose tissue. Suppression of production and stimulation of oxidation of glucose require less insulin than stimulation of uptake. In the traditionally glucocentric view of insulin resistance, a defect in insulin action requires more insulin than usual to maintain normal glucose fluxes [13]. In this glucocentric view it is ignored that insulin has many more regulatory tasks than those related to glucose metabolism. Another important function of insulin is suppression of lipolysis. Less insulin is required for suppression of lipolysis (suppression of FFA flux) than for regulation of glucose metabolism [14]. Numerous data have shown that lipids and especially high FFA levels will induce insulin resistance. There is now a growing appreciation that a chronic elevation in FFA levels is an early event that contributes to the development of insulin resistance [1, 15]. Insulin resistance will further increase FFA levels, and this can worsen insulin resistance again. In this way insulin resistance can worsen the metabolic abnormalities, but is never the primary contributor. The relation between insulin resistance and hypertension seems to be well established [1]. Resistance to the metabolic effects of insulin and compensatory hyperinsulinemia have been postulated to mediate human essential hypertension, especially when associated with obesity. Evidence supporting this hypothesis has come mainly from epidemiological studies showing correlations between insulin resistance, hyperinsulinemia, and blood pressure, and from short-term studies suggesting that insulin has renal and sympathetic effects that could raise blood pressure if the effects were sustained. However, there have been no studies demonstrating a direct causal relationship between chronic hypertension and insulin resistance or hyperinsulinemia in humans. The few long-term studies that have been conducted in dogs and humans do not support the hypothesis that hyperinsulinemia causes hypertension or potentiates the hypertensive effects of other pressor agents such as angiotensin II or increased adrenergic tone. To the contrary, multiple studies in dogs and humans suggest that the vasodilator action of insulin tends to reduce blood pressure. Although resistance to insulin’s metabolic effects has been suggested to be essential for hyperinsulinemia to cause hypertension, chronic increases in plasma insulin concentrations do not cause hypertension in dogs or humans, even in the presence of insulin resistance. Recent studies have further shown that the blood pressure-lowering effects of anti-hyperglycemic agents, initially believed to lower blood pressure by decreasing insulin resistance, may be unrelated to their effects on insulin sensitivity. Obesity appears to be a key factor in accounting for correlations between insulin resistance, 7

Sauerwein hyperinsulinemia, and hypertension, but increased blood pressure in obesity does not appear to be mediated by insulin resistance and hyperinsulinemia [16]. Even in full-blown metabolic syndrome insulin resistance contributes only modestly to the increased prevalence of hypertension [17]. These data point to new upcoming pathophysiological findings about extra-adrenal cortisol production in adipose tissue. Within adipose tissue, the enzyme 11␤-hydroxysteroid dehydrogenase type-1 interconverts inactive glucocorticoid cortisone and cortisol. In vivo, it is the reductase activity that is believed to predominate, generating cortisol in an autocrine/paracrine manner within the adipocyte microenvironment. An increasing amount of data shows that cortisol production by adipose tissue is increased in obesity and contributes to insulin resistance. Induction of weight loss reverses these changes [18]. These data suggest that insulin resistance is the consequence of abnormalities induced by local cortisol overproduction, a consequence of obesity. Based on these data, the suggestion crops up that insulin resistance is therefore not the main player in the pathophysiology of the metabolic syndrome, but a consequence of obesity. An exception in this series of arguments against the primary role of insulin resistance in the pathogenesis of the components of the metabolic syndrome is hypertriglyceridemia. Literature data indicate that this abnormality is caused by overproduction and reduced clearance of very low-density lipoprotein, both processes regulated by insulin. Overproduction of very low-density lipoprotein will lower high-density lipoprotein [19]. Conclusion It is clear that more than one distinct pathophysiological process underlies the clinical expression of the metabolic syndrome, and insulin resistance can in some sense be related to them but seems to be less prominent than usually stated in literature. Adipose tissue is an active metabolic organ. An increase in the size of this organ with consequent changes in its metabolism can readily explain the features of the metabolic syndrome. Therefore, insulin resistance is not the main player in the metabolic syndrome, but central obesity is. FFA-induced insulin resistance is found and induced by central obesity. The same holds true for hypertension. References 1 Eckel RH, Grundy SM, Zimmet PZ: The metabolic syndrome. Lancet 2005;365:1415–1428. 2 Kahn R, Buse J, Ferrannini E, Stern M: The metabolic syndrome: time for a critical appraisal. Joint statement from the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care 2005;28:2289–2304. 3 Reaven GM: The insulin resistance syndrome: definition and dietary approaches to treatment. Annu Rev Nutr 2005;25:391–406.

8

The Dysmetabolic Syndrome: Epidemiology and Etiology 4 Gu D, Reynolds K, Wu X, et al: Prevalence of the metabolic syndrome and overweight among adults in China. Lancet 2005;365:1398–1405. 5 Lakka HM, Laaksonen DE, Lakka TA, et al: The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. JAMA 2002;228:2709–2716. 6 Alberti KGMM, Zimmet P, Shaw J: The metabolic syndrome-a new worldwide definition. Lancet 2005;366:1059–1062. 7 Adams RJ, Appleton S, Wilson DH, et al: Population comparison of two clinical approaches to the metabolic syndrome. Diabetes Care 2005;28:2777–2779. 8 Liao Y, Kwon S, Shaughnessy S, et al: Critical evaluation of adult treatment panel III in identifying insulin resistance with dyslipidemia. Diabetes Care 2004;27:978–983. 9 Hanley AJG, Wagenknecht LE, D’Agostino RB, et al: Identification of subjects with insulin resistance and ␤-cell dysfunction using alternative definitions of the metabolic syndrome. Diabetes 2003;52:2740–2747. 10 Cheal KL, Abbasi F, Lamendola C, et al: Relationship to insulin resistance of the adult treatment panel III diagnostic criteria for identification of the metabolic syndrome. Diabetes 2004;53:1195–1200. 11 McLaughlin T, Abbasi F, Cheal K, et al: Use of metabolic markers to identify overweight individuals who are insulin resistant. Ann Intern Med 2003;39:802–809. 12 Staels B, Fruchart JC: Therapeutic roles of peroxisome proliferators-activated receptor agonists. Diabetes 2005;54:2460–2470. 13 Rizza RA, Mandarino LJ, Gerich JE: Dose-response characteristics for effects of insulin on production and utilization of glucose in man. Am J Physiol 1981;240:E630–E639. 14 Nurjhan N, Campbell PJ, Kennedy FP, et al: Insulin dose-response characteristics for suppression of glycerol release and conversion to glucose in humans. Diabetes 1986;35:1326–1331. 15 Boden G, Laakso M: Lipids and glucose in type 2 diabetes. What is cause and effect. Diabetes Care 2004;27:2253–2259. 16 Hall JE, Brands MW, Zappe DH, Alonso Galicia M: Insulin resistance, hyperinsulinemia, and hypertension: causes, consequences, or merely correlations? Proc Soc Exp Biol Med 1995;208:317–329. 17 Hanley AJG, Karter AJ, Festa A, et al: Factor analysis of metabolic syndrome using directly measured insulin sensitivity. The insulin resistance atherosclerosis study. Diabetes 2002;51: 2642–2647. 18 Tomlinson JW, Moore JS, Clark PMS, et al: Weight loss increases 11␤-hydroxysteroid dehydrogenase type 1 expression in human adipose tissue. J Clin Endocrinol Metab 2004;89: 2711–2716. 19 Taskinen MR: Diabetic dyslipidaemia: from basic research to clinical practice. Diabetologia 2003;46:733–749.

Discussion Dr. Bantle: Could you say a little bit about the associations between the different definitions of metabolic syndrome and cardiovascular disease? It would seem to me that the definition to use is the one that is the best predictor of events. Dr. Sauerwein: In my presentation I showed data that in Finland the European standards are better, but that in Holland, despite being close to Finland, the American ones are better predictors of cardiovascular disease, indicating that we need to redefine our diagnostic criteria. For reliable prediction and comparison of published data we need one definition. That is why the International Diabetes Federation (IDF) came up with a lot of new data, but they include so many things that we have to wait for proof of their validity. Dr. Bantle: Is there any evidence that this definition works as a predictor of future events? Dr. Sauerwein: It was just promoted. There is always a debate between the people who are in charge of one definition and those defending the other one.

9

Sauerwein Dr. Katsilambros: If I understood properly, you said that bariatric surgery does not improve hypertension. There is the so-called SHO study, a Scandinavian obesity study, which is perhaps the largest in the world and with a long follow-up, in which different kinds of operations were performed. After 2, 6 and 8 years the hypertension rate was lower when compared to the starting point before operation. However, after 10 years the rate again increased back to the beginning [1]. But even 10 years after the operation these people were still considerably obese and any benefit was not very lasting, and also, at least in this population, aging was another factor adding to the prevalence and incidence of obesity. So in my opinion at least, your statement about there being no relation to hypertension with regard to bariatric surgery is perhaps not the best way to express that, but I may be wrong. Dr. Sauerwein: The data I showed are from that Swedish study. What the slide showed is that the incidence of hypertension has returned to baseline after 10 years. When insulin resistance is the common denominator, this dissociation cannot occur unless, as you suggested, a new mechanism for the induction of hypertension develops in this time period. This is a possibility, but pathophysiological data in favor of this are lacking. I think that other mechanisms like local cortisol production in abdominal fat deserve real consideration. Many people are considering insulin resistance and metabolic syndrome to be more or less synonymous. I am of the opinion that we need a more subtle approach. Dr. Chiasson: I thought that your discussion was very interesting, and certainly Kahn et al. [2] have gone through this questioning of the metabolic syndrome. I think it is good that we raise questions about whether this is really a metabolic syndrome or just factors that are in parallel but totally independent, or whether they have a common background. So if I understand correctly, insulin resistance may not be the common denominator but you believe that the free fatty acids (FFAs) could be. FFA requires obesity to increase the plasma level. I was just wondering if the increase in FFA in obesity or under any other circumstances was not in fact due to insulin resistance, because otherwise you would expect the physiological level of insulin to be able to maintain inhibition on lipolysis. How do you explain this discrepancy that you are trying to propose? Dr. Sauerwein: What I did not share with you is that fat in the diet has no influence on insulin resistance. I was part of a study were we consumed eucaloric diets with either 85% fat or 85% carbohydrate as the energy source for 14 days without hardly any induction of insulin resistance [3, 4]. The same holds true for diabetics [5]. So it is not just the fat or (FFA) fatty acids themselves, but the induction of obesity with excess fatty acids taken up by muscle, ␤-cells, etc., that induces insulin resistance. Induction of insulin resistance will induce a vicious cycle in which insulin resistance will stimulate FFA release, as you described. This will aggravate insulin resistance. However the starting point must be excess intake and obesity. Eucaloric fat intake is less of a problem. Dr. Chiasson: I am not sure I understand correctly. I was under the impression that deposition of fat and triglyceride in the muscle and other tissues was related to insulin resistance and the increase in FFA in the circulation. Dr. Sauerwein: This is true, but what I want to stress are the initial changes. Most of the studies are cross-sectional, ignoring the sequence of events. Confusion can also be ascribed to definition problems. A high fat diet is sometimes called a eucaloric diet with a high percentage of fat in it, but more frequently a hypercaloric diet has additional fat. This distinction is not always made, but is important as our data show. Dr. Hill: You emphasized the importance of adipose tissue. We used to think that adipose tissue was pretty uninteresting and now we know it secretes many interesting products. Do you think products coming from fat cells will be found to be helpful in understanding the metabolic syndrome?

10

The Dysmetabolic Syndrome: Epidemiology and Etiology Dr. Sauerwein: This is a confusing area. You have to realize that morbid obesity is not always synonymous with major metabolic abnormalities. About 20% of the morbidly obese subjects are metabolically healthy [6]. I am not aware of any study focusing on the differences in expression of all those factors in morbidly obese subjects with and without major metabolic abnormalities, but I am convinced that the hormones produced by adipose tissue will have a major influence. However, I think it is too early now to make a statement about this. Dr. Ditschuneit: Do you think that lipolysis and concentrations of FFAs may be a target for treating metabolic syndrome? Dr. Sauerwein: Inhibition of lipolysis improves insulin resistance [7]. However suppression of lipolysis is only part of the story. The main problem is excess intake. With excess intake FFA will be stored in adipose tissue and muscle, inducing insulin resistance [8]. Even storage in adipose tissue, beyond a certain amount, induces insulin resistance, as adipose tissue increasing in size will attract macrophages [9]. This will induce a kind of local inflammation, increasing the degree of insulin resistance [10]. Dr. Slama: I think that there is ambiguity about the metabolic syndrome explaining the controversy, which is where we are now. The point is that the metabolic syndrome was first recognized by clinicians as a cluster of signs and symptoms associated with more complications in the future, all the complications we know. Then it was decided that the definition needs to have thresholds so that it is easier to recognize such a symptom. Now we ask, is that the best way to predict diabetes? Is that the best way to predict cardiovascular disease? Of course not. There is a very good equation, the Framingham equation, to predict cardiovascular complications; then there is a better equation, the score put forward by Hafner, to predict diabetes but it was not intended for that. The proposal that metabolic syndrome causes complications does not mean that the reverse is true. We don’t want to say that the definition of metabolic syndrome is the universal way to diagnose or predict diabetes or cardiovascular disease; we say that a cluster of people or a subgroup of people affected by such and such signs will be at a higher risk of cardiovascular disease and diabetes. In other words, I think that in the natural history between a normal situation toward vascular complications or diabetes, there is something which is early on, which is a definition of the metabolic syndrome at the very beginning, then the disease progresses and then Framingham or other predictors are better indicators, but on the shorter run. On the long run perhaps when it is time to put preventive measures into action, it is the metabolic syndrome, and for Framingham and such this is the time not of prevention but of early treatment. What is your opinion? Dr. Sauerwein: I completely agree with you. It is a cluster of abnormalities, nothing less, but definitely nothing more. Dr. T. Wilkin: I think one of the difficulties is that we impose the problem upon ourselves. We apply categorization to what are continuous variables. If there are 4 or 5 continuous variables and we apply categorization then individuals are bound to have very different levels, some of which will satisfy the categories and others which won’t. As long as we try to categorize what is continuous, we are going to have this problem. Dr. Sauerwein: I agree. Dr. Chieh Chou: I have a question related to the role of obesity in metabolic syndrome because in the case of diabetes with obesity, in patients with moderate weight loss, an improvement in the disease is often seen. So perhaps removing fat from the liver or muscle would really improve insulin resistance. What do you think? Dr. Sauerwein: I agree. When people are storing their fat in the only place where it should be (in certain subcutaneous areas), they have no problems related to glucose intolerance. Translocation of fat by thiazolidediones to those areas improves glucose

11

Sauerwein tolerance even despite the well-known increase in body weight [11]. Another problem is the unexplained relationship between abdominal obesity and insulin resistance. It was always thought that FFAs flowing from the intra-abdominal cavity to the liver would induce insulin resistance, but quantitatively the contribution of the intraabdominal-based FFA to total FFA flux proved to be around 20% [12]. Dr. Mingdao Chen: The American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD) statement for ‘metabolic syndrome’ is quite different from that of the IDF. The ADA and EASD tried to say that the metabolic syndrome is only a cluster of a few factors or risks, it is not even a syndrome. However, a syndrome means several different signs and symptoms together, a situation surely suited to the metabolic syndrome. Which definition do you support? Dr. Sauerwein: I think it is a cluster of a few factors, which have proven in the past to be predictors of cardiovascular disease. In that sense it is alright, but we should not draw pathophysiological conclusions from epidemiological data. Dr. Mingdao Chen: Perhaps they just mean this is not a disease. Dr. Sauerwein: Yes, that is what I tried to say. Dr. Slama: One of the difficulties with the metabolic syndrome is blood pressure. The definition of metabolic syndrome gives a blood pressure of ⱖ130 mm Hg, but the situation is absolutely not the same if the blood pressure is 130 or 180 mm Hg. In both cases the definition is upheld, but 130 mm Hg would be normal for people other than those with a cluster, and 180 mm Hg is abnormal whether cluster or no cluster. So for me the real interest of the definition of the metabolic syndrome is those people who have all or most of the items of the cluster in the near normal range but become abnormal because they have a cluster. Of course if a patient has a blood pressure of 180 mm Hg, a blood glucose of 200, 5 g triglyceride, he doesn’t need to be labeled metabolic syndrome, he is badly sick. So the real interest of the metabolic syndrome is those who are just close to the thresholds and will not be considered as having a disease point by point, item by item, but are really affected because they have most of them. Dr. Sauerwein: That is absolutely correct, but the problem is that this is being ignored in the literature.

References 1 Sjostrom L, Lindroos AK, Peltonen M, et al, Swedish Obese Subjects Study Scientific Group: Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric surgery. N Engl J Med 2004;351:2683–2693. 2 Kahn R, Buse J, Ferrannini E, Stern M, American Diabetes Association, European Association for the Study of Diabetes: The metabolic syndrome: time for a critical appraisal: joint statement from the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care 2005;28:2289–2304. 3 Bisschop PH, Pereira Arias AM, Ackermans MT, et al: The effects of carbohydrate variation in isocaloric diets on glycogenolysis and gluconeogenesis in healthy men. J Clin Endocrinol Metab 2000;85:1963–1967. 4 Bisschop PH, de Metz J, Ackermans MT, et al: Dietary fat content alters insulin-mediated glucose metabolism in healthy men. Am J Clin Nutr 2001;73:554–559. 5 Allick G, Bisschop PH, Ackermans MT, et al: A low-carbohydrate/high-fat diet improves glucoregulation in type 2 diabetes mellitus by reducing postabsorptive glycogenolysis. J Clin Endocrinol Metab 2004;89:6193–6197. 6 Sims EA: Are there persons who are obese, but metabolically healthy? Metabolism 2001;50: 1499–1504. 7 Bajaj M, Suraamornkul S, Romanelli A, et al: Effect of a sustained reduction in plasma free fatty acid concentration on intramuscular long-chain fatty Acyl-CoAs and insulin action in type 2 diabetic patients. Diabetes 2005;54:3148–3153.

12

The Dysmetabolic Syndrome: Epidemiology and Etiology 8 Miles JM, Park YS, Walewicz D, et al: Systemic and forearm triglyceride metabolism: fate of lipoprotein lipase-generated glycerol and free fatty acids. Diabetes 2004;53:521–527. 9 Weisberg SP, McCann D, Desai M, et al: Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest 2003;112:1796–1808. 10 Xu H, Barnes GT, Yang Q, et al: Chronic inflammation in fat plays a crucial role in the development of obesity-related insulin resistance. J Clin Invest 2003;112:1821–1830. 11 Mayerson AB, Hundal RS, Dufour S, et al: The effects of rosiglitazone on insulin sensitivity, lipolysis, and hepatic and skeletal muscle triglyceride content in patients with type 2 diabetes. Diabetes 2002;51:797–802. 12 Miles JM, Jensen MD: Counterpoint: visceral adiposity is not causally related to insulin resistance. Diabetes Care 2005;28:2326–2328.

13

Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 15–29, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

Traditional Chinese Medicine in the Treatment of Diabetes Hai-Lu Zhao, Peter C.Y. Tong, Juliana C.N. Chan Department of Medicine and Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong, SAR, China

Abstract This review focuses on the efficacy and safety of Chinese medicine in the treatment of type-2 diabetes. Included were 84 controlled clinical studies of type-2 diabetes treated with Chinese medicine for at least 1 month. Reported outcomes were: symptom relief; improvement in glycemia, insulin resistance and secondary failure, and adverse events. Symptom relief was achieved in most (⬎80%) of the patients receiving Chinese medicine. Compared with orthodox drugs, Chinese medicine had a 1.2-fold (95% CI 1.2–1.3) increase in symptom relief. The relative risk of achieving a fasting blood glucose of ⬍7.3 mmol/l or a postprandial blood glucose of ⬍8.2 mmol/l was: 3.0 (95% CI 1.4–6.5) for Chinese medicine plus diet versus diet; 2.0 (95% CI 1.4–3.0) for Chinese medicine versus placebo; 1.8 (95% CI 1.4–2.3) for combined Chinese medicine and orthodox drugs versus Yuquan Wan (a classic Chinese herbal formula for diabetes), 1.5 (95% CI 1.4–1.7) for combined Chinese medicine and orthodox drugs vs. orthodox drugs, and 1.3 (95% CI 1.2–1.5) for Chinese medicine versus orthodox drugs. A fasting blood glucose of ⬍8.2 mmol/l plus symptom relief was observed in 71–100% of the patients with secondary failure to oral anti-diabetic drugs. Serious adverse events including hypoglycemic coma and death were caused by adulteration with orthodox drugs, erroneous substitution, self-meditation, overdoses, and improper preparation. Chinese herbal medicine should be used cautiously with doctors’ prescription and follow-up. Long-term clinical studies may disclose the effectiveness of Chinese medicine in reducing the mortality and morbidity of diabetic complications. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

Diabetes mellitus is a group of complex diseases characterized by chronic hyperglycemia due to defects in insulin secretion, insulin action, or both. The long-term damage of chronic hyperglycemia of diabetes occurs in various organs, especially the eyes, kidneys, nerves, heart, and blood vessels. Type-2 diabetes is the most prevalent form of diabetes that often coexists 15

Zhao/Tong/Chan with other metabolic components such as obesity, hypertension and dyslipidemia. In China, type-2 diabetes is increasing in epidemic dimensions. In 2002 the National Survey on the Status of Nutrition and Health of the Chinese People conducted in adults over the age of 18 years revealed an increased morbidity from overweight (22.8% and 200 million, prevalence and total number), obesity (7.1% and 60 million), hypertension (18.8% and 160 million), impaired fasting glucose (1.9% and 20 million), diabetes (2.6% and 27 million), and dyslipidemia (18.6% and 160 million) [1]. The prevalence of metabolic syndrome is approximately 13% in the Chinese population [1, 2]. In parallel to the epidemic trend of diabetes and metabolic syndrome, the integration of traditional Chinese medicine (TCM) into the management of diabetes and metabolic syndrome has also grown. In hospital clinics, over 70% of adult patients with type-2 diabetes used both TCM and Western medicine [3, 4]. Over a 25-year period between 1981 and 2005, 511 clinical trials of TCM included 33,274 patients with type-2 diabetes. The majority (96%) of the patients had integrated treatment with TCM and Western medicines [5]. This review expounds evidence from clinical and pharmacological studies of TCM in the treatment of diabetes. The clinical studies included in this review meet the following criteria [6]: (1) controlled trials; (2) adult participants (18 years or older) with type-2 diabetes; (3) documented diagnostic criteria of type-2 diabetes; (4) intervention of TCM including Chinese herbal medicines and Chinese proprietary medicines for a duration of at least 1 month; (5) outcomes of all-cause mortality, cardiocerebrovascular events, quality of life, symptom-relief rate and normalization of blood glucose, and (6) adverse events (death, stroke, hypoglycemia, liver toxicity, kidney damage). Major findings from relevant mechanistic and pharmacological studies were highlighted to provide insight explanations of clinical efficacy. Electronic databases on Wan Fang data (827 Chinese medical journals by July, Week 3, 2005), Cochrane Library (2nd Quarter, 2005), and MEDLINE (1966 to July, Week 2, 2005) were searched using the index terms for type-2 diabetes, clinical trials and Chinese medicine. Relative risk (RR) with 95% confidential interval (CI) was used to express data extracted from the controlled trials of TCM.

Traditional Chinese Medicine TCM is both an art and a science of patient-centered healing with combined attention to body, mind, and spirit. Knowledge of TCM has been enriched for over 4,000 years of observation, investigation and clinical experience. The philosophy of TCM is rooted in Chinese cultures of Taoism (to follow nature’s way) and Confucius (to nurture humanity and morality) and the religion of Buddhism (to free from suffering). Traditionally, TCM doctors are 16

Traditional Chinese Medicine in the Treatment of Diabetes usually pharmacists and pharmacologists who themselves identify and collect herbs, prepare formulation and follow up their patients. Most of the published clinical trials of TCM were conducted by TCM doctors. TCM includes Chinese herbal medicine (CHM), acupuncture and specialized disciplines of surgery, orthopedics, pediatrics, and obstetrics and gynecology. Qigong (energy practice), Tuina (massage), Chinese martial arts, and diet per se are not representatives of TCM, although diet therapy is an important modality in disease management. Chinese herbs and herbal products are not necessarily the same term of TCM. In literature, TCM is characterized by individualized treatments based on the differentiation of syndrome (Zheng). CHM is the major modality in TCM practice. A prescription for CHM usually refers to a compound recipe (Fu Fang) that consists of principal, assistant, adjuvant, and guiding herbs to maximize therapeutic effects and minimize toxic effects. Ingredients in a CHM prescription are individualized and changed on a weekly basis to tailor for the patient’s age, gender, symptoms, anthropological characters, geological location and living environment. From a literature review, approximately 1,200 recipes and 150 herbs for diabetes, metabolic syndrome and associated complications have been documented since 1980 [5, 7–9]. Table 1 lists the Chinese herbs and classic recipes commonly used in clinical trials in diabetes and diabetic complications [5, 7, 8]. Although the bioactive components of most medicinal herbs remain unknown, several kinds of chemical compounds have reported properties for lowering blood glucose, increasing insulin secretion, and improving insulin resistance [7, 10]. The compounds extracted from Chinese herbal medicines include flavonoids, xanthones, triterpenoids, alkaloids, glycosides, alkyldisulfides, aminobutyric acid derivatives, guanidine, polysaccharides, peptides, and minerals [7, 10, 11].

Symptom Relief TCM is particularly effective in symptom relief. A systemic review of 6 clinical trails by Liu et al. [6] reported that most of the type-2 diabetic patients receiving CHM experienced an improvement in symptoms of dry mouth, polyphagia, polydipsia, polyuria, fatigue, sweating, constipation, numb limbs, and low back pain (table 2). Improvement rates were higher in patients receiving CHM than those treated with antidiabetic drugs (table 2). After using either CHM alone or integrated CHM and Western medicine for at least 2 months, symptoms also substantially improved in most (⬎80%) of the patients with diabetic complications. Reported complications include diabetic gastroparesis [12–16], nephropathy [17–25], neuropathy [26–44], retinopathy [45–47], gangrene [48, 49], peripheral vascular disease [50–52], and myocardial infarction [53]. 17

Zhao/Tong/Chan Table 1. Traditional Chinese medicine for diabetes mellitus Single herbs – Chinese Pin Yin and English name Bai Zhu Ovate atractylodes Bai Shao White peony root Ban Xia Pinellia Bian Dou Hyacinth bean Can E Silkworm moth Cang Zhu Atractylodes root Chai Hu Bupleurum root Chi Shao Red Peony root Chi Xiao Dou Adsukibean Chuan Xiong Ligusticum root Da Huang Rhubarb Dan Pi Moutan bark Dan Shen Salvia root Dang Gui Angelica Dang Shen Codonopsis root Di Gu Pi Lycium root bark Du Zhong Eucommia bark E Zhu Zedoary Fu Ling Poria Gan Cao Liquorice root Ge Gen Pueraria root Gou Ji Cibotium root Huai Niu Xi Achyranthes root Huang Bai Phellodendron bark Huang Jing Polygonatum root Huang Lian Coptis root Huang Qi Astragalus root Huang Qin Scutellaria root Jiang Can Silkworm Jin Ying Zi Cherokee rose fruit Li He Litchee pit Mai Dong Ophiopogon tuber Niu Bang Zi Arctium seed Ren Shen Ginseng San Qi Notoginseng root Sang Pi Mulberry root bark Sang Piao Xiao Mantis egg-case powder Sang Shen Zi Mulberry Sang Ye Mulberry leaf Sha Ren Amomum fruit Shan Dou Gen Root of straight sophora Shan Yao Dioscorea root Shan Zha Crataegus fruit Shan Zhu Yu Asiatic cornelian cherry fruit She Chuang Zi Cnidium seed Sheng Di Huang Dried rehmannia root Sheng Shai Shen Panax ginseng Shi Gao Gypsum Shi Hu Dendrobium

18

Traditional Chinese Medicine in the Treatment of Diabetes Table 1. (continued) Shu Di Huang Cooked rehmannia root Tai Zi Shen Pseudostellaria root Tao Ren Peach kernel Tian Dong Arisaema tuber Tian Hua Fen Trichosanthes root Wu Wei Zi Schisandra berry Xi Yang Shen American ginseng Xian Ling Pi Epimedium herb Xuan Shen Scrophularia root Yi Mi Coix seed Yi Mu Cao Leonurus Yu Zhu Solomon’s seal root Ze Xie Alisma tuber Zhi Mu Anemarrhena root Zhu Ling Polyporus Classic recipes and ingredients – Chinese Pin Yin and English name Baihu Tang Shi Gao (gypsum) (white tiger decoction) Zhi Mu (wind-weed rhizome) Gan Cao (prepared licorice root) Geng Mi (polished round-grained nonglutious rice) Buyang Huanwu Tang 1) Huang Qi (astragalus root) (decoction invigorating 2) Dang Gui (Chinese angelica root) Yang for recuperation) 3) Chi Shao (red peony root) 4) Chuan Xiong (Chuanxiong rhizome) 5) Tao Ren (peach kernel) 6) Hong Hua (safflower) 7) Di Long (earthworm) Liuwei Dihuang Wan Shu Di Huang (prepared rhizome of rehmannia) (bolus of rehmannia six) Shan Zhu Yu (dogwood fruit) Shan Yao (dried Chinese yam) Ze Xie (oriental water plantain) Fu Ling (poria) Mu Dan Pi (mountain bark) Shen Qi Wan Di Huang (dried rehmannia) (bolus invigorating the Shan Yao (Chinese yam) kidney Qi) Shan Zhu Yu (dogwood fruit) Ze Xie (oriental water plantain) Fu Ling (poria) Mu Dan Pi (mountain bark) Gui Zhi (cinnamon twig) Fu Zi (prepared aconite root) Yu Quan Wan Ge Gen (pueraria root) (jade spring bolus) Tian Hua Fen (trichosanthes root) Mai Dong (ophiopogon tuber) Sheng Di Huang (dried rehmannia root) Geng Mi (polished round-grained nonglutinous rice) Gan Cao (prepared licorice root) Wu Wei Zi (schisandra berry)

19

Zhao/Tong/Chan Table 2. Improvement rates (%) of diabetic symptoms in 35 clinical trials of Chinese medicine

Dry mouth Polyphagia/ polydipsia/ Polyuria Fatigue Sweating Constipation Numb limbs Low back pain

Chinese medicine

Western medicine

RR (95% CI)

p value

91.7 (1,587/1,730)

74.6 (930/1,247)

1.2 (1.19–1.27)

⬍0.0001

91.2 (1,484/1,616) 89.2 (1,650/1,849) 89.5 (1,359/1,518) 91.1 (1,327/1,456) 91.8 (1,305/1,427) 89.9 (1,418/1,578)

74.2 (827/1,114) 70.0 (897/1,281) 70.1 (794/1,122) 71.3 (724/1,155) 72.9 (806/1,105) 72.9 (841/1,153)

1.2 (1.19–1.28) 1.3 (1.23–1.33) 1.3 (1.23–1.33) 1.3 (1.22–1.32) 1.3 (1.21–1.30) 1.2 (1.19–1.28)

⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001

Glycemic Control Most of the clinical trials showed that CHM recipes were effective in glycemic control [6]. Glycemic control was measured with surrogate parameters including fasting and postprandial blood glucose and glycosylated hemoglobin (HbA1c) [6]. Normalization of blood glucose was defined by a fasting blood glucose of ⬍7.3 mmol/l or a postprandial blood glucose of ⬍8.2 mmol/l [6]. Data from 78 controlled clinical trials showed the rate of normalizing blood glucose was higher in diabetic patients using CHM compared with control subjects (table 3) [6]. As a whole CHM appeared at least as effective as orthodox antidiabetic drugs in reducing blood glucose and HbA1c [6]. The blood glucose-lowering effects of CHM are probably related to enhanced insulin secretion [6] and improved insulin resistance [54].

Secondary Failure to Antidiabetic Drugs Secondary failure to antidiabetic drugs substantially limits the effectiveness of Western drugs in the management of diabetes. Clinical trails have revealed that CHM and acupuncture in combination with Western medicine are effective in rescuing the secondary failure in patients with type-2 diabetes. Table 4 shows rates of improvement in glycemic control, as defined by a fasting plasma glucose of ⬍8.2 mmol/l plus symptom relief. The improvement rate was higher in patients (total number ⫽ 300) treated with integrated Chinese and Western medicine than those (total number ⫽ 296) receiving Western medicine alone in 5 controlled trials (RR ⫽ 1.1, 95% CI 1.02–1.18, p ⫽ 0.01) [55–59]. 20

Traditional Chinese Medicine in the Treatment of Diabetes Table 3. Normalization of fasting blood glucose levels in 78 trials Number Normalization of patients % Chinese medicine ⫹ diet Diet Chinese medicine Placebo Chinese medicine Oral antidiabetic drugs Chinese medicine ⫹ oral antidiabetic drugs Oral antidiabetic drugs Chinese medicine ⫹ oral antidiabetic drugs Yu Quan Wan

30 30 226

60 (n ⫽ 18) 20 (n ⫽ 6) 28.3 (n ⫽ 64)

222 1,546

14.0 (n ⫽ 31) 52.9 (n ⫽ 813)

918

39.2 (n ⫽ 360)

Relative risk p value (95% CI)

Number of trials

3.0 (1.39–6.50)

0.0033

1

2.0 (1.38–2.99)

0.0002

6

1.3 (1.22–1.47)

⬍0.0001 19

2,735

48.3 (n ⫽ 1,321) 1.5 (1.36–1.73)

⬍0.0001 48

2,088

30.7 (n ⫽ 640)

1,045

41.0 (n ⫽ 428)

258

1.8 (1.44–2.31)

⬍0.0001

4

22.5 (n ⫽ 58)

Normalization of blood glucose is defined by a fasting blood glucose of ⬍7.3 mmol/l or a postprandial blood glucose of ⬍8.2 mmol/l [6].

Table 4. Efficacy of Chinese medicine in rescuing secondary failure to oral antidiabetic drugs Controlled trial

Efficacy rate % (n) Reference

Xiao Ke Wan Glibenclamide Yiqi Yangyin ⫹ tolbutamide ⫹ metformin Tolbutamide ⫹ metformin Acupucture ⫹ Berberine ⫹ yeast ⫹ glibenclamide ⫹ metformin Glibenclamide ⫹ metformin Insulin Jiaweitaohechengqitang Metformin Jiaweitaohechengqitang Rosiglitazone

82.7 (81/98) 64.1 (41/64) 93.3 (56/60)

Wang and Hu [57], 2001

67.9 (38/56) 100 (80/80)

Xue and Li [58], 2001

100 (80/80) 87.5 (35/40) 73.2 (30/41) 75.7 (28/37) 71.4 (15/21) 73.7 (14/19)

Shi [56], 2000

Zhu et al. [59], 2002 Chen et al. [55], 2004

Efficacy is defined by a fasting blood glucose of ⬍8.2 mmol/l plus symptoms relieving.

21

Zhao/Tong/Chan Adverse Effects CHM is well tolerated and relatively safe in clinical use. Adverse reactions related to CHM use are uncommon. Approximately 5% of the total adverse events and drug-related deaths are caused by the use of CHM. Most of the trials in diabetes and metabolic syndrome revealed no adverse effects during CHM treatment [6]. Nonserious adverse events associated with CHM treatment included diarrhea, abdominal pain, nausea, and loss of appetite [6]. The blood glucose-lowering effect of CHM may turn into adverse events such as hypoglycemia if misused. For example, Xiao Ke Wan is a widely used drug composed of six herbs (Di Huang, Ge Gen, Huang Qi, Shan Yao, Tian Hua Fen, Wu Wei Zi) and 0.25 mg glyburide. The drug is indicated for type-2 diabetic patients with a deficiency syndrome of both Yin and Qi. It is estimated that the incidence of hypoglycemia is 5% in Xiao Ke Wan users [60, 61]. The risk of hypoglycemia is even alarming in elderly subjects, individuals with impaired hepatic and renal function, patients with acute infection, and in patients who concomitantly use insulin or other antidiabetic drugs [62]. Among 311 incidental cases with hypoglycemia reported in 15 clinical studies, 92 (29.6%), 9 (2.9%), 76 (24.4%), and 113 (36.3%) were caused by using Xiao Ke Wan, CHM, insulin and oral antidiabetic drugs, respectively [60, 61]. Moreover, patients with Xiao Ke Waninduced hypoglycemia had a high risk of death (3.3%) and stroke (9.8%) [63]. Therefore, extreme caution should be taken to prevent hypoglycemia and other serious adverse events when the Chinese herbal medicine is concomitantly used with orthodox drugs [64, 65].

Conclusions TCM is an individualized treatment based on differentiation of the syndrome. CHM as a whole is effective and relatively safe in relieving symptoms, controlling hyperglycemia, and rescuing secondary failure in patients with diabetes. Long-term controlled clinical investigations will disclose the effectiveness of Chinese medicine in reducing the mortality and morbidity of chronic complications in patients with diabetes. The majority of Chinese patients with diabetes use both Chinese medicine and orthodox drugs. In parallel to the epidemic tend of diabetes and increased use of integrated Chinese and Western medicine, herb-drug interactions may substantially occur when herbs are misused. Overdoses, improper preparation, erroneous substitution, adulteration with Western drugs or heavy metals, and self-meditation in using CHM can cause serious problems including hypoglycemic coma and death. Chinese medicine should be used cautiously following doctors’ prescription and supervision. Doctors should always obtain a complete history on the use of both Chinese medicine and Western drugs in the clinical assessment and prescription. 22

Traditional Chinese Medicine in the Treatment of Diabetes Acknowledgements We thank Dr. Sui Yi for her assistance in the literature search and data analysis. This paper is dedicated to the late Prof. Julian A.J.H. Critchley.

References 1 Ministry of Health People’s Republic of China: The 4th National Survey of Nutrition and Health of the Chinese People. http://www.moh.gov.cn/statistics/year2004/. Accessed on 6 June 2005. 2 Li ZY, Xu GB, Xia TA: Prevalence rate of metabolic syndrome and dyslipidemia in a large professional population in Beijing. Atherosclerosis 2006;184:188–192. 3 Chen Q, Zhao HL, Tong PCY, et al: Chinese herbal medicine in diabetes management (P1962). Diabetes Metab 2003;29(suppl):S170. 4 Zhao HL, Chen Q, Hao AZ, et al: Prescription frequency of herbal medicines in Chinese patients with type 2 diabetes mellitus. Atherosclerosis 2003;181(suppl):337–338. 5 Chen DY, Ge JY, Zhou DS, et al: Review of 23139 patients with type 2 diabetes treated Chinese medicine (in Chinese). Chin Arch Chin Med 2003;21:1225–1228. 6 Liu JP, Zhang M, Wang WY, Grimsgaard S: Chinese herbal medicines for type 2 diabetes mellitus. Cochrane Database Syst Rev 2004, CD003642. 7 Bailey CJ, Day C: Traditional plant medicines as treatments for diabetes. Diabetes Care 1989;12:553–564. 8 Zhou L, Zhou XF, Fu C, Wang Q: Chinese herbs in 271 compound recipes for diabetes (in Chinese). N J Trad Chin Med 2004;36:40–41. 9 Lin L: Progress in diabetes research of traditional Chinese medicine; in Lin L (ed): Diabetology of Integrated Chinese and Western Medicine. Beijing, China Medical Science & Technology Publishing House, 1999, pp 15–33. 10 Li WL, Zheng HC, Bukuru J, De Kimpe N: Natural medicines used in the traditional Chinese medical system for therapy of diabetes mellitus. J Ethnopharmacol 2004;92:1–21. 11 Jia W, Gao WY, Xiao PG: Antidiabetic drugs of plant origin used in China: compositions, pharmacology, and hypoglycemic machanisms. China J Chin Mat Med 2003;28:108–113. 12 Cai W: Clinical observation of 82 patients with diabetic gastroparesis treated with Xuefuzhuyu decoction (in Chinese). Jiangxi J Trad Chin Med 2004;35:27. 13 Dai HL, Song GQ: Clinical observation of 25 patients with diabetic gastroparesis treated with combination of Chinese and western medicine (in Chinese). Shandong J Trad Chin Med 2004;23:351–352. 14 Liang H, Sun YK: Combined treatment with Chinese and western medicine for diabetic gastroparesis: a report of 106 cases (in Chinese). Chin J Folk Med 2004;12:4–5. 15 Qiu YM, Shan JW, Hu TC: Controlled study of Banxiaxiexin decoction in 35 patients with diabetic gastroparesis (in Chinese). Fujian J Trad Chin Med 2004;35:24. 16 Tian BP, Hu B: Treatment with Chinese medicine based on differentiation of syndrome in 48 patients with diabetic gastroparesis (in Chinese). Liaoning J Trad Chin Med 2004;31:725–726. 17 Du X, Jin MW: Diyutangshen tablet for diabetic nephropathy of deficiency of Yin and Qi with internal heat and blood stasis (in Chinese). Chin J Pract Chin Mod Med 2004;4:3584–3586. 18 Jha V, Chugh KS: Nephropathy associated with animal, plant, and chemical toxins in the tropics. Semin Nephrol 2003;23:49–65. 19 Zheng JQ, Li H, Lv SG: Integrated treatment with Chinese and western medicine in 86 type 2 diabetic patients with nephropathy (in Chinese). Fujian Med J 2004;26:123–124. 20 Sun WS, Wu XL, Qiao CL, Liu R: Clinical study on effect of Tongluo capsule in treating diabetic nephropathy caused chronic renal failure (in Chinese). Chin J Integr Chin West Med 2004;24:704–706. 21 Dong KL, Li LM, Li GC: Clinical observation of combined treatment with Chinese and western medicine in 60 patients with diabetic nephropathy (in Chinese). Hunan J Trad Chin Med 2004;20:25–27. 22 Fu XJ, Zhang HE, Liu JH, et al: Clinical study of Qi Zhi formula in early diabetic nephropathy (in Chinese). Chin J Integr Chin West Med Nephrol 2004;5:535–536.

23

Zhao/Tong/Chan 23 Meng HL: Clinical observation of combined treatment with Chinese and western medicine in 45 patients with diabetic nephropathy (in Chinese). Forum Trad Chin Med 2004;19:40. 24 Huang BT, Tian YQ: Clinical observation of Tangshenning II in diabetic nephropathy: a report of 43 cases. New J Trad Chin Med 2004;36:36–37. 25 Huang MH, Gan XB, Chen JS: Clinical observation of Shen Kang I and II in diabetic nephropathy (in Chinese). Chin J Integr Chin West Med Nephrol 2004;5:409–410. 26 Abuaisha BB, Costanzi JB, Boulton AJ: Acupuncture for the treatment of chronic painful peripheral diabetic neuropathy: a long-term study. Diabetes Res Clin Pract 1998;39:115–121. 27 Qian ZR, Zhong XL, Fang YA: Traditional Chinese medicine combined with Western medicine in the treatment of diabetic peripheral neuropathy (in Chinese). Zhong Xi Yi Jie He Za Zhi 1987;7:131, 140–142. 28 Xu XF, Xu W: Treating 42 diabetic peripheral neuropathy with Chinese – western method (in Chinese). Inner Mongolia Med J 2004;36:182–183. 29 Lv YH: Jiangtang Tongluo Tang in treatment of 35 diabetic patients with neuropathy (in Chinese). Trad Chin Med Res 2004;17:42. 30 Wang XZ: Treatment with Xianteng Huoluo Yin in 30 cases of diabetic patients with neuropathy – clinical report (in Chinese). Beijing J Trad Chin Med 2004;23:289–291. 31 Zhang ZH, Li P: Huangqi Guizhi Wuwu Tang in treatment of 34 cases with diabetic neuropathy (in Chinese). J Pract Trad Chin Med 2004;20:4. 32 Gao YS: Buyang Huanwu Tang in treatment of 70 cases with diabetic neuropathy (in Chinese). Hunan J Trad Chin Med 2004;20:19–20. 33 Li P: Yiqi Yangyin Huoxue Tang in treatment of 80 cases with diabetic neuropathy (in Chinese). J Sichuan Trad Chin Med 2004;22:44. 34 Mu JP: Clinical observation of integrated traditional Chinese and western medicine in treatment of 97 cases with diabetic neuropathy (in Chinese). J Sichuan Trad Chin Med 2004;22: 34–35. 35 Xu SS: Clinical observation of Wenyang Huayu meathod in treatment of old diabetic neuropathy (in Chinese). Liaoning J Trad Chin Med 2004;31:376. 36 Gao Z: Integrated traditional Chinese and western medicine in treatment of 34 cases with multiple diabetic neuropathy (in Chinese). Jiangxi J Trad Chin Med 2004;25–27:25. 37 Zhou J, Wu JL, Zhang YX, Wu ST: Yiqi Huoxue Tongmai Tang in treatment of 70 cases with diabetic peripheral neuropathy (in Chinese). J Chin Med Pharmacol Inform 2004;11:153–154. 38 Meng HL: Huoshen Buxue Tang in treatment of 39 cases with diabetic neuropathy (in Chinese). Forum Trad Chin Med 2004;19:31–32. 39 Zhang DF: Tangluotong in treatment of 100 cases with diabetic peripheral neuropathy (in Chinese). Trad Chin Med Res 2004;17:34. 40 Hu YH, Sun ZX, Li J, Wu ST: Clinical observation of integrated traditional Chinese and western medicine in treatment of diabetic peripheral neuropathy (in Chinese). Chin Sci Technol Chin Med 2004;11:111–112. 41 Yu HY: Integrated traditional Chinese western medicine in treatment of 55 cases with diabetic neuropathy (in Chinese). Ji Lin J Trad Chin Med 2004;25:35. 42 Xiao W: Treatment of diabetic neuropathy of lower limbs with acupuncture, cupping and hypoglycemic agents: a report of 38 cases (in Chinese). J Anhui Trad Chin Med Coll 2004;23:22–24. 43 Peng JS: Combination of Dengzhanhuasu and Nimodipine in treatment of 52 cases of diabetic multiple neuropathy (in Chinese). J Chin Phys 2004;6:130–131. 44 Dou ZX: Juanbi Tongluo capsules in treatment of 36 cases with diabetic peripheral neuropathy (in Chinese). Forum Trad Chin Med 2004;19:33. 45 Zhou XD, You ML, Luo LL: Clinical observation of combined treatment with Zhuyu Huoxue decoction and urokinase in diabetic patients with vitreous hematoma (in Chinese). Guangming J Trad Chin Med 2004;19:27–29. 46 Wei W, Wei CH: Clinical observation of integrated traditional Chinese and western medicine in treatment of diabetic retinopathy and macula edema (in Chinese). Jiangsu J Trad Chin Med Pharmacol 2004;25:35. 47 Wang YC, Mao QB, Li JH, et al: Clinical observation of traditional Chinese medicine based on differentiation of syndrome in diabetic retinopathy (in Chinese). Mod J Integr Trad Chin West Med 2004;13:1018–1019. 48 Li JS: Clinical observation on 45 cases of diabetic gangrene treated by the combination of Chinese medicine and west medicine (in Chinese). Forum Trad Chin Med 2004;19:36.

24

Traditional Chinese Medicine in the Treatment of Diabetes 49 Xia JC, Zhai S, Yang XZ: Clinical observation on 24 cases of diabetic gangrene treated by the combination of Chinese medicine and west medicine (in Chinese). Hunan Guid J Trad Chin Med 2004;10:24–25. 50 He X: Tongluo Huoxue decoction in treatment of 37 cases with diabetic vascular changes of lower limbs (in Chinese). J Zhejiang Univ Trad Chin Med 2004;28:53–54. 51 Wang YJ, Wang XK, Chen LP: Effects of Danggui Sini decoction and Lulutong injection on treating diabetes leg arteriosclerosis obliterans (in Chinese). Chin J Basic Med Trad Chin Med 2004;10:60–62. 52 Zhang GH, Lin XZ: Buyang Huanwu decoction in treatment of early diabetic peripheral vascular changes: clinical observation of 61 cases (in Chinese). J Henan Univ Chin Med 2004;19:61. 53 Wang RP, Zhou HP: Controlled study of Gegenyin in 60 diabetic patients with myocardial infarction (in Chinese). Chin Arch Chin Med 2004;22:361–362. 54 Cheng JT, Liu IM, Chi TC, et al: Metformin-like effects of Quei Fu Di Huang Wan, a Chinese herbal mixture, on streptozotocin-induced diabetic rat. Horm Metab Res 2001;33:727–732. 55 Chen P, Zhu ZZ, Lang JM, et al: Jiaweitaohechengqitang for secondary failure to sulfonylurea in type 2 diabetes (in Chinese). Chin J Integr Trad West Med 2004;24:585–588. 56 Shi MY: Xiao Ke Wan and glibenclamide for secondary failure to sulfonylurea in type 2 diabetes (in Chinese). Anhui J Trad Chin Med 2000;13:173–175. 57 Wang JS, Hu WJ: Yiqi Yangyin Recipe for diabetic patients with secondary failure to sulfonylurea (in Chinese). Chin J Clin Healthcare 2001;4:188–189. 58 Xue RJ, Li PZ: Integrated treatment withacupucture and Western medicine for secondary failure in type 2 diabetes (in Chinese). Liaoning J Trad Chin Med 2001;28:40–41. 59 Zhu ZZ, Xiong MQ, Lin AZ, et al: Jiaweitaohechengqitang improves insulin resistance in type 2 diabetic patients with secondary failure to sulfonylurea (in Chinese). Chin J Inform Trad Chin Med 2002;9:25–27. 60 Li XW, Tang SW, Yan YH, et al: Analysis on hypoglycemia in the subjects with diabetes (in Chinese). Acta Acad Med Zun Yi 2002;25:423–424. 61 Cheng GL, Zhang RF: Clinical analysis of 88 cases with hypoglycemia (in Chinese). J Jinzhou Med Coll 2004;25:58. 62 Zhang WH: Hypoglycemia in 17 cases treated with anti-diabetic agents (in Chinese). Clin Misdiagn Misther 2003;16:144–145. 63 Dai JP: Xiao Ke Wan induced hypoglycemic coma in 20 cases (in Chinese). J Xianning Med Coll 2001;15:67. 64 Chan TY, Chan JC, Tomlinson B, Critchley JA: Chinese herbal medicines revisited: a Hong Kong perspective. Lancet 1993;342:1532–1534. 65 Tomlinson B, Chan TY, Chan JC, et al: Toxicity of complementary therapies: an eastern perspective. J Clin Pharmacol 2000;40:451–456.

Discussion Dr. Katsilambros: I enjoyed your lecture very much. As I don’t come from China I have no idea about rehmannia. What kind of substance is it? How long can rehmannia be administered for type-2 diabetes? What were the results with regard to glycemia, metabolic factors, complications, and other factors including mortality and survival? What was the greatest length of time that this medication could be given to type-2 diabetic people? Dr. Zhao: In contrast to chemical drugs synthesized with monomers, traditional Chinese medications contain lots of plant-derived substances and compounds including alkaloids, flavonoids, glycosides, peptides, saponins, triterpenoids and xanthones [1]. Compounds purified from Chinese herbal medicines are disappointing because of their decreased therapeutic effects and increased toxicity when compared to the traditional Chinese decoction. I speculate that all six herbs in the rehmannia recipe work synergistically to enhance the body’s self-healing capacity in correcting multiple

25

Zhao/Tong/Chan metabolic abnormalities. Animal studies have shown that Cornus officinalis is the major contributor to the hypoglycemic action of rehmannia six [2]. In type-2 diabetic rats, an alcohol extract of cornus improved glucose tolerance associated with an elevation in the blood insulin level [3] and enhanced muscle GLUT4 expression [4]. In streptozotocin-induced diabetic rats, iridoid glucoside of the cornus extract attenuated microalbuminuria and glomerular damage [5]. In the clinical practice of traditional Chinese medicine, a minimal 1-month duration of treatment is required for patients with type-2 diabetes to show the benefits of symptom relief and improved glycemic and other metabolic control [6, 7]. Intervention with rehmannia six and lifestyle modification for 18–24 months significantly prevent the development of type-2 diabetes from impaired glucose tolerance [8, 9]. In a 3-month controlled clinical trial of 68 type2 diabetic patients with microalbuminuria, combined treatment with rehmannia six and captopril was better than captopril alone in delaying renal treatment [10]. To my knowledge, no clinical data are available regarding the efficacy of Chinese medicines on mortality and survival. Treatment is not the most important concept of traditional Chinese medicine. The most important thing for our physicians is the right judgment. If the right judgment is made as well as a very early diagnosis, then with some medical prescription there is a better chance of successful treatment of the patient. Dr. Chiasson: I also enjoyed your presentation very much and I think that it shows how one can take advantage of these thousands of years of experience with traditional Chinese medicine and bring it to the clinical bedside with a better understanding of how it works. Is there any effort being made to try to isolate the active ingredients in rehmannia six? You said that the toxicity may be increased by purification, but I thought that toxicity was due mainly to the heavy metal in those plants. All plants will carry different ingredients depending on where they grow, so the substances within the plant can vary. How is the dosage of rehmannia six decided upon? Dr. Zhao: Traditional medicine is widely used in most counties and communities. Significant efforts have been directed toward the safe and effective practice of this history-proven medicine. Today it is relatively easy to combine Chinese with modern Western medicine. However, it is very difficult to validate these two medical systems in terms of one another. Firstly, traditional Chinese medicines are defined neither by chemical structures nor by indications of specific diseases. Instead, traditional Chinese medicines are characterized by four properties of cold, hot, cool or warm, five flavors of bitter, pungent, salty, sour or sweet, and four actions in the downward (lowering), upward (lifting), outward (floating) or inward (sinking) directions [11]. A classic prescription of traditional Chinese medications usually contains several herbs to help the body to restore homeostasis. In the case of the rehmannia six recipe, the dosages are 24 g rehmannia, 12 g of cornus and dioscorea, and 9 g of mutan bark, alisma and poria. All six raw herbs are prepared in decoction and are taken orally 2–3 times/day. The dosages may be modified individually, tailored to the patient’s clinical conditions. This time-honored antidiabetic remedy has proven efficacy and safety profiles. In contrast, chemical substances purified with all the modern technologies were more toxic and even less effective compared with the classic preparation of decoction. The development of new chemical drugs from Chinese medications is a great challenge. Secondly, Chinese medicines are mostly derived from natural products of plants, animals and minerals. Due to environmental pollution, widespread use of chemical fertilizers and pesticides, the population explosion, overspending and other reasons, large numbers of wild plants and animals are endangered, and the natural resources for Chinese medications have decreased. Moreover, heavy metal contaminations [12], adulteration with chemical drugs, and abuse of Chinese medications are the eminent problems associated with increased severe adverse events. Genetic heterogeneity was found in the marketing of Chinese herbs including rehmannia [13]. To assure the safety and effectiveness of Chinese medications, good agricultural,

26

Traditional Chinese Medicine in the Treatment of Diabetes manufacturing and clinical practices are implemented. Despite the shared therapeutic goals, traditional Chinese medicine and modern Western medicine use significantly different approaches in the treatment of diabetes. Traditional Chinese medicine is a holistic and individualized treatment in terms of diagnosis, prescription, medication, and dosages. Dr. T. Wilkin: Early in your presentation you showed us a slide suggesting that the insulin level rose when rehmannia six was given, and in the latter part of your presentation that insulin resistance fell. This is conceptually very important in the management of type-2 diabetes because the view is generally held that stimulation of insulin is perhaps not the appropriate approach but that insulin sensitization would rather be appropriate. Can you tell us which is happening, and which is the more dominant factor because the glucocentric view is probably not the one that we would tend to go with? Dr. Zhao: Thank you for your comments. You have just highlighted that rehmannia six improves glycemia by enhancing insulin secretion and decreasing insulin resistance. It is very true that improvement of insulin resistance is central to the management of the metabolic syndrome. Dr. Slama: I recall a study perhaps 15 years ago here in China in the region of Beijing and Shanghai. It was a study between France and China, trying to compare traditional medicine and a Western drug called glibenclamide in type-2 diabetic people. There were 3 groups, and in fact randomization was according to centers and not patients. At some centers patients were treated only with glibenclamide, other centers treated patients only with traditional Chinese medicine, and the third group of centers, most interestingly, combined the two approaches. We found that after 6 months glibenclamide was efficient in improving blood glucose control, and that the traditional treatment didn’t do anything significant in terms of blood glucose control but, most interestingly, the combination of the two treatments did better than glibenclamide alone. Dr. Zhao: Thank you for sharing your findings. Traditional Chinese medicine may have limited advantages in lowering blood glucose. Beyond the glucocentric view, Chinese medicine can contribute to the prevention of diabetes in high-risk individuals, prevention of diabetic complications, relieving symptoms, reducing drug resistance and toxicities, and rescuing secondary failure to chemical drugs [14]. In our survey of 3,904 patients with diabetes, 93% used both Chinese and Western medicines [15]. The advantage of integrative medicine is unclear [16]. Dr. Slama: Would you agree that we have a double task when treating diabetic patients. The first one is the prevention of macrovascular complications, but also we have to fight against blindness, neuropathy and nephropathy, and only by lowering the blood glucose level can we reach this goal. So really we should act on both sides to prevent macro- and microvascular complications and also focus on blood glucoselowering agents. Dr. Zhao: I agree with you that high blood glucose is a risk factor associated with the development of macro- and microvascular complications. In a systematic review of 66 randomized trials involving 8,302 subjects, some Chinese herbal medicines show hypoglycemic effects in type-2 diabetes [17]. Novel hypoglycemic agents might be identified from enriched Chinese medications. But the focus of traditional Chinese medicine is not glucose. The ultimate goal of Chinese medications is to help the body to restore balances for maintaining health. Dr. Halimi: I have a question about insulin secretion or insulin sensitivity in your studies with rehmannia six. You have shown that, in normal Wistar rats receiving this plant, insulin levels are multiplied about 5 times. Then you presented some data related to diabetes prevention in patients with impaired glucose tolerance, or dysmetabolic syndrome, all subjects having normal or subnormal fasting blood glucose

27

Zhao/Tong/Chan levels. Did they exhibit hypoglycemic episodes with rehmannia six or did they only normalize glycemia and insulinemia? Is this situation comparable to sulfonylureas or metformin? Dr. Zhao: The philosophical basis of traditional Chinese medicine is the dynamic balance of Yin-Yang. Eating too much and exercising too little cause imbalances reflected by syndromes of deficiency and excess in diabetes. In the rehmannia six recipe, the deficient syndrome of Yin is supplemented by the herbs rehmannia, dioscorea, and cornus, whereas the excessive syndrome of Yang is cleared by the other three herbs, mutan bark, alisma, and poria. The approaches of Chinese and Western medicine to healing are completely different. With our ignorance of the philosophical basis, it is impossible to make critical judgments of Chinese medicine using the Western approach. Dr. Gerasimidi-Vazeou: Is the dose of rehmannia six you tested in animal studies equivalent to the one you gave to humans? What is the toxicity of the ingredients? Have you performed any toxicity experiments? Dr. Zhao: Repeated oral administration of rehmannia six at dosages of 26 mg/kg for 3 days significantly improved insulin resistance in obese Zucker rats [18]. The corresponding clinical dose of each single herb is 8 g of rehmannia, 4 g of cornus and dioscorea, and 3 g of mutan bark, alisma and poria. Each ingredient herb is potentially toxic. However, when the six herbs are prepared as a decoction, adverse events are rare and minor. The major problem associated with rehmannia six is overuse. Dr. Gerasimidi-Vazeou: You said that the ingredients of the recipe change according to the physician’s estimate of his patients. In animal studies you used a standardized recipe but what about humans? What happens with the dosage of several ingredients you use to make the recipe for every single patient? How do you change the dosage of the ingredients, and how much is this standardized? Dr. Zhao: Documented in the China Pharmacopoeia 2000 are the standards for traditional Chinese medicines. One gram of rehmannia six should contain at least 1.0 mg of paeonol (Dan Pi Fen). Dr. Gerasimidi-Vazeou: What I am trying to ask you is whether there are consistent standardized ingredients in the recipe you use for humans because you said that you change the ratio between ingredients? What about the data related to the human studies you presented, have you tested a standard recipe or not? Dr. Zhao: Paeonol is the standard ingredient extracted from mutan bark. Dr. Bantle: I have a question about the rehmannia six supply. Where does it come from? Is it made by pharmaceutical companies? What other sort of preparation does it undergo? Dr. Zhao: There are over 10 manufacturers producing rehmannia six with different dosage forms. Rehmannia six may be prepared as a decoction by boiling the six herbs. Dr. Hill: You mentioned that almost 77% of the population use both traditional and Western medicine. Does this mean that a lot of the physicians are trained in both, or does it mean that people go both to a traditional physician and a Western medicine physician? Dr. Zhao: Doctors may prescribe both Chinese and Western medicine. Western medicines are available both by prescription and ‘over the counter’, as is also the case for Chinese medicines. Alternatively, patients can also see physicians who practice traditional Chinese medicine as well as Western medical doctors. Dr. Hill: Are there many physicians who are trained in both traditional and Western medicine? Dr. Zhao: In mainland China, most doctors have been trained in both Chinese and Western medicine. General hospitals in China provide services in both Western and traditional Chinese medicine. So the patients can benefit from both systems.

28

Traditional Chinese Medicine in the Treatment of Diabetes Dr. Mingdao Chen: Most doctors are trained in Western style medicine but they still use the traditional Chinese medicine to complement Western medicine. They learn traditional Chinese medicine at university, as I did, over several months. I use traditional Chinese medicine to supplement normal treatment, and on the other hand the traditional Chinese medical doctors also use Western medicine. This system is particularly important in China and the patients benefit from this system. Dr. Zhao: Indeed, two opinions are better than one. In the future I expect to see an increase in adverse events and drug interactions published in the literature because of the increased use of combined traditional Chinese and Western medicine.

References 1 Li WL, Zheng HC, Bukuru J, De Kimpe N: Natural medicines used in the traditional Chinese medical system for therapy of diabetes mellitus. J Ethnopharmacol 2004;92:1–21. 2 Liou SS, Liu IM, Hsu SF, Cheng JT: Corni fructus as the major herb of Die-Huang-Wan for lowering plasma glucose in Wistar rats. J Pharm Pharmacol 2004;56:1443–1447. 3 Qian DS, Luo L, He M, et al: Therapeutic effect of alcohol extract of Cornus officinalis Sieb. et Zucc on postmeal insulin and blood glucose concentrations in NIDDM rats. J Nantong Univ Med Sci 2000;20:337–339. 4 Qian DS, Zhu YF, Zhu Q: Effect of alcohol extract of Cornus officinalis Sieb. et Zucc on GLUT4 expression in skeletal muscle in type 2 (non-insulin-dependent) diabetic mellitus rats (in Chinese). Zhongguo Zhong Yao Za Zhi 2001;26:859–862. 5 Xu HQ, Zhu YF, Zhu Q: Protecting effect of iridoid glycoside in Fructus corni officinalis on experimental diabetic nephropathy. J Nangjing Univ Tradit Chin Med 2003;19:342–344. 6 Li Y: Rehmannia six and an anti-diabetic decoction in the treatment of 158 patients with type 2 diabetes. Liaoning J Tradit Chin Med 2000;27:252–253. 7 Xi YP: Combined treatment with rehmannia six and metformin in 48 patients with type 2 diabetes. Anhui J Tradit Chin Med 2002;14:5. 8 Zeng YH, Chen F, Wang YS, et al: Combined treatment with traditional Chinese and Western medicines in subjects with impaired glucose tolerance. Chin J Integr Med Gastroenterol 2000;8:196–198. 9 Wang H, Liang X, Yu XM, Zuo Y: Rehmannia six in the treatment of impaired glucose tolerance. Liao Ning J Tradit Chin Med 2002;29:58–59. 10 Chen JL, Ling FM: Combined treatment with rehmannia six and captopril in diabetic patients with microalbuminuria. New J Tradit Chin Med 2004;36:26–27. 11 Zhang E-Q: The Chinese Materia Medica. Shanghai, Shanghai College of Traditional Chinese Medicine Press, 1992. 12 Ma SM, Huang ZY, Wang QE, et al: Determination of harmful elements in rehmannia six by microwave digestion and inductively coupled plasma mass spectrometry. J Guangxi Norm Univ 2003;20:136–137. 13 Yuan M, Hong Y: Heterogeneity of Chinese medical herbs in Singapore assessed by fluorescence AFLP analysis. Am J Chin Med 2003;31:773–779. 14 Chen P, Zhu ZZ, Lang JM, et al: Clinical observation on effect of Yiqi Yangyin Huoxue Tongfu principle in treating diabetes mellitus type 2 of secondary failure to sulfonylurea agents (in Chinese). Zhongguo Zhong Xi Yi Jie He Za Zhi 2004;24:585–588. 15 Chen Q, Zhu ZZ, Lang JM, et al: Chinese herbal medicine in diabetes management (P1962). Diabetes Metab 2003;29(suppl):170. 16 Shi ZX: Peculiar clinical dominance of integrative Chinese and Western medicine (in Chinese). Zhongguo Zhong Xi Yi Jie He Za Zhi 2005;25:101–102. 17 Liu JP, Zhang M, Wang WY, Grimsgaard S: Chinese herbal medicines for type 2 diabetes mellitus. Cochrane Database Syst Rev 2005; CD003642; DOI 003610.001002/14651858; CD140036242. pub14651852. 18 Wu YC, Hsu JH, Liu IM, et al: Increase of insulin sensitivity in diabetic rats received diehuang-wan, a herbal mixture used in Chinese traditional medicine. Acta Pharmacol Sin 2002;23:1181–1187.

29

Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 31–42, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

Pharmacological and Surgical Intervention for the Prevention of Diabetes Jean-Louis Chiasson Research Group on Diabetes and Metabolic Regulation, Research Center, Centre hospitalier de l’Université de Montréal and Department of Medicine, Université de Montréal, Montreal, Que., Canada

Abstract The increasing prevalence of diabetes is reaching epidemic proportion worldwide. Because of the associated morbidity and mortality, it is exerting major pressure on the healthcare system. With a better understanding of the pathophysiology of type-2 diabetes, the concept of primary prevention has emerged. A number of studies have confirmed that intensive lifestyle modification was very effective in the prevention of diabetes in the impaired glucose tolerance (IGT) population. However, maintaining long-term lifestyle modification is a major challenge. It is, therefore, important to have other strategies, either pharmacological or surgical, that can be used as an adjunct or alternative to lifestyle modification. The Chinese study showed that metformin and acarbose could reduce the risk of diabetes by 65 and 83%, respectively, in IGT subjects. The efficacy of metformin was confirmed by the Diabetes Prevention Program (31% risk reduction) and that of acarbose by the STOP-NIDDM trial (36% risk reduction) in a similar high-risk population. The TRIPOD study showed that troglitazone could reduce the risk of diabetes by 55% in Hispanic women with a history of gestational diabetes. And more recently, the XENDOS study showed that orlistat could reduced the risk of diabetes by 37% in obese subjects when used as an adjunct to an intensive lifestyle program. Three studies have suggested that bariatric surgery in morbidly obese subjects could reduce the risk of diabetes to near zero. Furthermore, a number of studies have examined the effect of a renin angiotensin aldosterone system inhibitor, as well as statin and hormone replacement therapy on the prevention of type-2 diabetes in high-risk subjects as secondary outcomes and have suggested that they could be of potential benefit. The accumulating evidence is now overwhelming. Yes, diabetes can be prevented or delayed in high-risk populations. With this new information, we need to design new strategies to screen high-risk populations and to implement the new treatments that have proven effective in the prevention of type-2 diabetes. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

31

Chiasson We are facing a worldwide explosion in the prevalence of type-2 diabetes mellitus [1]. Because it is associated with high morbidity and excess mortality imposing a major burden on healthcare cost [2–4], it is definitely one of the major challenges of the 21st century. Though it is generally accepted that type-2 diabetes develops in genetically susceptible individuals, it is also recognized that it is usually precipitated by environmental factors such as sedentary lifestyle and obesity [5, 6]. These factors contribute to the development of insulin resistance and, in those subjects with limited ␤-cell mass capacity, probably genetically determined, to the development of glucose intolerance, initially impaired glucose tolerance (IGT), and eventually type-2 diabetes [7]. This understanding of the pathophysiology of glucose intolerance served as the rationale for the concept of the prevention of type-2 diabetes. This concept has now been confirmed by a number of prospective trials that have shown that both non-pharmacological and pharmacological treatments as well as bariatric surgery in a high-risk population with IGT could prevent, or at least delay, the progression to diabetes. The evidence for pharmacological and surgical intervention as a means to prevent or delay the progression of IGT to type-2 diabetes is reviewed here.

Pharmacological Interventions for the Prevention of Diabetes as Primary Endpoint Insulin resistance overstresses the ␤ cells and, in genetically susceptible individuals, results in a reduction in their capacity to secrete insulin. This combination of impaired insulin action and secretion will favor the development of IGT, a prediabetic state characterized by postprandial hyperglycemia. This moderate postprandial hyperglycemia is sufficient to induce glucose toxicity and further contribute to the progression of IGT to diabetes. Therefore, it was hypothesized that any pharmacological intervention that would decrease insulin resistance and/or the stress on the ␤ cells could potentially prevent or delay the progression of IGT to type-2 diabetes. Overall, 5 randomized controlled trials have examined the effect of drug interventions on the prevention of diabetes in IGT subjects as a primary outcome (table 1). The Chinese study published in 2001 examined the effect of acarbose and metformin on the incidence of type-2 diabetes in 321 subjects with IGT over a 3-year period [8]. Both acarbose and metformin were effective in reducing the risk of diabetes by 83% (p ⫽ 0.0001) and 65% (p ⫽ 0.0002), respectively, compared to an incidence of 11.6% in the control group (table 1). The efficacy of both drugs in preventing type-2 diabetes was very impressive. 32

Pharmacological and Surgical Intervention for the Prevention of Diabetes Table 1. Pharmacological interventions for the prevention of type-2 diabetes as a primary outcome Study

Number Intervention of subjects

Duration Incidence of Risk of study diabetes in reduction years control group % %/year

The Chinese study [8] The DPP study [9] The STOP-NIDDM trial [10] The TRIPOD study [12] The XENDOS study [13]

261

3.0 3.0 2.8 3.3

11.6

2,155 1,368

Acarbose Metformin Metformin Acarbose

11.0 12.1

83 65 31 36

236

Troglitazone 2.5

12.1

55

3,305

Orlistat

2.25a

37

4.0

aOnly 21% had IGT and both groups were submitted to an intensive lifestylemodification program.

The efficacy of metformin in reducing the risk of diabetes in IGT subjects was confirmed by the Diabetes Prevention Program published in 2002 [9]. In this study, 3,234 subjects with IGT were randomized to a lifestyle program, metformin 850 mg b.i.d. or placebo b.i.d., and followed for 2.8 years. Overall, metformin reduced the risk of diabetes by 31% (p ⫽ 0.001) based on 2 oral glucose tolerance tests (OGTTs) compared to an incidence of 11.0%/year in the placebo group (table 1). However, it was less effective in the older subjects (ⱖ60 years) and the less obese (body mass index ⬍35 kg/m2) [9]. The efficacy of acarbose was also confirmed by the Study to Prevent NIDDM trial, also published in 2002 [10]. In this international trial, 1,368 subjects with IGT were randomized to acarbose 100 mg or placebo t.i.d. with meals and followed for an average of 3.3 years. Based on 1 OGTT, acarbose reduced the risk of diabetes by 25% (p ⫽ 0.0015) compared to an incidence of 12.1%/year in the placebo group. If the diagnosis of diabetes was based on 2 OGTTs as now recommended by the American Diabetes Association [11], the drug was associated with a 36.4% reduction in the risk of diabetes (p ⫽ 0.0003; table 1). The beneficial effect was seen independent of age, gender or body mass index. Also published in 2002, the TRIPOD study examined the effect of troglitazone on the prevention of diabetes in Hispanic women (n ⫽ 236) with a history of gestational diabetes [12]. The overweight women were randomized to placebo or troglitazone 400 mg o.d. and followed for 2.5 years. Troglitazone reduced the risk of diabetes by 55% (p ⫽ 0.009) compared to the placebo group (incidence ⫽ 12.1%/year; table 1). Though troglitazone proved to be an effective drug for prevention of diabetes, it was discontinued because of liver toxicity. 33

Chiasson Metformin

Acarbose

Troglitazone

Orlistat

Relative risk reduction (%)

0

20

40

35

37

42 60

55

80

100

Fig. 1. Non-pharmacological and pharmacological interventions on the prevention of type-2 diabetes as a primary outcome. The histograms represent the populationadjusted mean risk reduction where there is more than one intervention trial.

More recently, the Xenical in the Prevention of Diabetes in Obese Subjects Study assessed the effect of orlistat as an adjunct to a lifestyle-modification program on the prevention of diabetes in obese non-diabetic subjects [13]. Overall, 3,305 subjects were submitted to an intensive lifestyle program; only 21% had IGT. They were also randomized to orlistat 120 mg or placebo t.i.d. with meals. Both groups lost weight, but it was more important in the orlistat group (5.8 versus 3.0 kg; p ⬍ 0.001). Orlistat resulted in a 37% reduction in the risk of diabetes compared to an incidence of 2.25%/year in the placebo group (p ⫽ 0.003; table 1). Despite methodological problems with the study, it does support a preventive effect of orlistat on the development of diabetes in a high-risk population. The evidence that pharmacological intervention can prevent or delay the progression of IGT to diabetes is overwhelming (fig. 1). This is definitely true for metformin and acarbose. It is also true for troglitazone but unfortunately it is liver toxic and has been taken off the market. It is hoped that the other thiazolidinediones will also be as effective in decreasing the incidence of diabetes in high-risk populations. Current studies are now testing rosiglitazone and pioglitazone [14]. Xenical is very likely effective in decreasing weight and thus reducing the risk of diabetes. However, methodological problems, such as the high drop out rate and the carry forward analysis, make the study difficult to interpret.

Bariatric Surgery We still have very few data on the use of bariatric surgery in massively obese subjects on the prevention of diabetes. Three studies, however, have sufficient 34

Pharmacological and Surgical Intervention for the Prevention of Diabetes Table 2. Bariatric surgery and the prevention of type-2 diabetes Study

Number of subjects (surgery/control)

Duration of follow-up years

Weight loss %

Incidence of diabetes, %/year after surgery

Pories et al. [15] Long et al. [16] SOS study [17]

152/– 109/27 517/539

7.6 5.8 10.0

33a 52b 16a

0.17 0.15 0.10

aPercent

bPercent

body weight. excess body weight.

data on the prevention of diabetes that are interesting and worth discussing (table 2). Pories et al. [15] submitted 608 morbidly obese patients to gastric bypass and followed them prospectively for 7.6 years. Among these patients, 152 had IGT at baseline and, interestingly, only 2 progressed to diabetes for an annual incidence of 0.17%. Unfortunately, there was no control group (table 2). Long et al. [16] also examined the effect of gastric bypass on the incidence of diabetes in 109 morbidly obese subjects with IGT. As controls, they used 27 subjects with IGT who also had morbid obesity but did not undergo surgery for personal or other non-medical reasons. After an average follow-up of 4.8 years, the annual incidence of diabetes in the experimental group was 0.15% compared to 4.6% in the control group (table 2). This remains a non-randomized study and selection bias cannot be ruled out. More recently, Sjöstrom et al. [17] published the Swedish Obese Subjects study on 4,047 morbidly obese subjects who underwent bariatric surgery. Though this is not a randomized study, all subjects were attributed matched controls. At the end of the 2-year follow-up, the annual incidence of diabetes in the subjects who underwent bariatric surgery was 0.10% compared to 3.2% in the controls (table 2). It is not clear how many subjects had IGT at baseline. Though these studies in morbidly obese subjects with or without IGT were not randomized, the data are convincing that bariatric surgery resulting in sustained weight loss is an acceptable and effective alternative in reducing the risk of type-2 diabetes.

Other Pharmacological Interventions for the Prevention of Diabetes as Secondary Outcome Other studies have also examined the effect of other drugs on the prevention of diabetes in high-risk populations as secondary outcome. At least 10 studies have assessed the effect of angiotensin-converting enzyme inhibitors and angiotensin receptors antagonists on the prevention of 35

Chiasson Table 3. Other pharmacological interventions and the prevention of type-2 diabetes as secondary outcome Study

Intervention

Number Duration of Comparator of follow-up subjects years

Renin angiotensin aldosterone system inhibitors CAPPP [18] Captonil 10,413 6.1 STOP-HTN2 [19] LIFE [20] HOPE [21] ALLHAT [26]

Enalapril/ lisinopril Losartan Ramipril Lisinopril

3,930

6.0

7,598 5,720 33,357

4.8 4.5 4.9

SOLVD [27] Enalapril 391 ALPINE Candesartan 393 [28] SCOPE Candesartan 4,964 [29] CHARM Candesartan 7,601 [30] VALUE Valsartan 15,245 [31] Statin WOSCOPS Pravastatin 5,974 [23] Hormone replacement therapy HERS [32] Estradiol 2,763 0.625 mg Progesterone 2.5 mg

2.9 1.0

Diuretic and/or ␤-blockers ␤-blocker

Risk p value reduction % 14

0.039

15

NS

3.7

␤-blocker 25 Placebo 34 Diuretic 30 Ca2⫹ channel blocker Placebo 74 Diuretic ⫾ 87 ␤-blocker Placebo 19

0.001 ⬍0.001 ⬍0.001

0.09

2.0

Placebo

19

⬍0.001

4.2

Diuretic/ ␤-blocker

23

⬍0.001

4.8

Placebo

30

0.036

4.1

Placebo

35

0.006

⬍0.001 0.03

diabetes as secondary objective (table 3) [18–21]. Eight of these studies found a significant reduction in the incidence of new cases of diabetes by the renin angiotensin aldosterone system inhibitors based on fasting plasma glucose. The relative risk reduction varied between 14 and 87% with an overall mean adjusted for population of 25.6% (fig. 2). Pravastatin and estrogen/progesterone replacement therapy are two other drugs that have also shown potential for the prevention of type-2 diabetes (fig. 2). The West of Scotland Coronary Prevention Study examined the effect of pravastatin on the development of diabetes as a secondary endpoint [22]. After a follow-up of 4.8 years, pravastatin was associated with a 30% risk reduction of diabetes (p ⫽ 0.036) [23]. The Heart and Estrogen/progestin Replacement Study [24] also examined the effect of estradiol 0.625 mg plus 36

Pharmacological and Surgical Intervention for the Prevention of Diabetes

ACEI ⫹ARA

Statin

Estrogen/ progestin

Relative risk reduction (%)

0

20 25.6 40

30

35

60

80

100

Fig. 2. Pharmacological interventions and the prevention of type-2 diabetes as a secondary outcome. The histograms represent the population-adjusted mean risk reduction where there is more than one intervention trial.

medroxyprogesterone acetate 2.5 mg versus placebo on the development of diabetes over a 4-year period based on fasting plasma glucose. The hormone replacement therapy was associated with a 35% risk reduction for diabetes (p ⫽ 0.006). All these studies are interesting and do suggest the potential benefit of these drugs in the prevention of diabetes. But as a secondary outcome, they can only be considered as generating a hypothesis and need to be confirmed in prospective studies in which the prevention of diabetes will be the primary outcome. Conclusion It is now established that type-2 diabetes can be prevented or delayed through a lifestyle-modification program or pharmacological intervention. Since weight reduction and exercise is difficult to maintain in the long-term, it is important to have pharmacological agents such as acarbose, metformin and probably orlistat that have been shown to be effective in reducing the risk of diabetes in high-risk populations. These can be used as an adjunct or an alternative to lifestyle modification. Bariatric surgery can also be used as an effective alternative in morbidly obese subjects to prevent diabetes. A number of studies have suggested that other drugs, such as renin angiotensin aldosterone system inhibitors, statins and hormone replacement could also be of potential benefit for the prevention of diabetes. These, however, need to be confirmed in prospective trials. Some are currently being tested in ongoing trials. The DREAM study is testing ramipril and rosiglitazone [14], the NAVIGATOR study nateglinide and valsartan [25], and finally, 37

Chiasson ACT NOW is assessing pioglitazone in preventing the progression of IGT to type-2 diabetes. With this new knowledge, strong recommendations have to be made to screen and treat IGT. These new strategies should help to attenuate the worldwide burden of diabetes.

Acknowledgement We are grateful to Susanne Bordeleau for preparing the manuscript and illustrations.

References 1 Wild S, Roglic G, Green A, et al: Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 2004;27:1047–1053. 2 Klein R, Klein BEK, Moss SE: Relation of glycemic control to diabetic microvascular complications in diabetes mellitus. Ann Intern Med 1996;124:90–96. 3 de Marco R, Locatelli F, Zoppini G, et al: Cause-specific mortality in type 2 diabetes. The Verona Diabetes Study. Diabetes Care 1999;22:756–761. 4 American Diabetes Association: Economic consequences of diabetes mellitus in the US in 1997. Diabetes Care 1998;21:296–309. 5 Zimmet PZ: The pathogenesis and prevention of diabetes in adults. Genes, autoimmunity, and demography. Diabetes Care 1995;18:1050–1064. 6 Kahn SE, Porte D Jr: Pathophysiology of type II diabetes mellitus; in Porte D Jr, Sherwin RS (eds): Diabetes Mellitus. Stamford, Appleton & Lange, 1996, pp 487–512. 7 Lillioja S, Bogardus C: Obesity and insulin resistance: lessons learned from the Pima Indians. Diabetes Metab Rev 1988;4:517–540. 8 Yang W, Lixiang L, Jinwu Q, et al: The preventive effect of acarbose and metformin on the progression to diabetes mellitus in the IGT population: a 3-year multicenter prospective study. Chin J Endocrinol Metab 2001;17:131–136. 9 Knowler WC, Barrett-Connor E, Fowler SE, et al: Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393–403. 10 Chiasson JL, Josse RG, Gomis R, et al: Acarbose for prevention of type 2 diabetes mellitus: the STOP-NIDDM randomised trial. Lancet 2002;359:2072–2077. 11 Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 1997;20:1183–1197. 12 Buchanan TA, Xiang AH, Peters RK, et al: Preservation of pancreatic beta-cell function and prevention of type 2 diabetes by pharmacological treatment of insulin resistance in high-risk Hispanic women. Diabetes 2002;51:2796–2803. 13 Torgerson JS, Hauptman J, Boldrin MN, Sjostrom L: XENical in the prevention of diabetes in obese subjects (XENDOS) study: a randomized study of orlistat as an adjunct to lifestyle changes for the prevention of type 2 diabetes in obese patients. Diabetes Care 2004;27:155–161. 14 Gerstein HC, Yusuf S, Holman R, et al: Rationale, design and recruitment characteristics of a large, simple international trial of diabetes prevention: the DREAM trial. Diabetologia 2004;47:1519–1527. 15 Pories WJ, Swanson MS, MacDonald KG, et al: Who would have thought it? An operation proves to be the most effective therapy for adult-onset diabetes mellitus. Ann Surg 1995;222: 339–352. 16 Long SD, O’Brien K, MacDodnald KG Jr, et al: Weight loss in severely obese subjects prevents the progression of impaired glucose tolerance to type II diabetes. Diabetes Care 1994;17: 372–375. 17 Sjöstrom L, Lindroos AK, Peltonen M, et al: Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric surgery. N Engl J Med 2004;351:2683–2693.

38

Pharmacological and Surgical Intervention for the Prevention of Diabetes 18 Niklason A, Hedner T, Niskanen L, Lanke J: Development of diabetes is retarded by ACE inhibition in hypertensive patients – a subanalysis of the Captopril Prevention Project (CAPPP). J Hypertens 2004;22:645–652. 19 Hansson L, Lindholm LH, Ekbom T, et al: Randomised trial of old and new antihypertensive drugs in elderly patients: cardiovascular mortality and morbidity the Swedish Trial in Old Patients with Hypertension-2 study. Lancet 1999;354:1751–1756. 20 Lindholm LH, Ibsen H, Borch-Johnsen K, et al: Risk of new-onset diabetes in the Losartan Intervention for Endpoint reduction in hypertension study. J Hypertens 2002;20:1879–1886. 21 Yusuf S, Gerstein H, Hoogwerf B, et al: Ramipril and the development of diabetes. JAMA 2001;286:1882–1885. 22 Shepherd J, Cobbe SM, Ford I, et al: Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. West of Scotland Coronary Prevention Study Group. N Engl J Med 1995;333:1301–1307. 23 Freeman DJ, Norrie J, Sattar N, et al: Pravastatin and the development of diabetes mellitus: evidence for a protective treatment effect in the West of Scotland Coronary Prevention Study. Circulation 2001;103:357–362. 24 Hulley S, Grady D, Bush T, et al: Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. Heart and Estrogen/progestin Replacement Study (HERS) Research Group. JAMA 1998;280:605–613. 25 Scheen AJ: Prevention of type 2 diabetes mellitus through inhibition of the renin-angiotensin system. Drugs 2004;64:2537–2565. 26 ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group. The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial: major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs. diuretic: The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA 2002;288:2981–2997. 27 Vermes E, Ducharme A, Bourassa MG, et al: Enalapril reduces the incidence of diabetes in patients with chronic heart failure: insight from the Studies Of Left Ventricular Dysfunction (SOLVD). Circulation 2003;107:1291–1296. 28 Lindholm LH, Persson M, Alaupovic P, et al: Metabolic outcome during 1 year in newly detected hypertensives: results of the Antihypertensive Treatment and Lipid Profile in a North of Sweden Efficacy Evaluation (ALPINE study). J Hypertens 2003;21:1563–1574. 29 Lithell H, Hansson L, Skoog I, et al: The Study on Cognition and Prognosis in the Elderly (SCOPE): principal results of a randomized double-blind intervention trial. J Hypertens 2003;21:875–886. 30 Pfeffer MA, Swedberg K, Granger CB, et al: Effects of candesartan on mortality and morbidity in patients with chronic heart failure: the CHARM-Overall programme. Lancet 2003;362:759–766. 31 Julius S, Kjeldsen SE, Weber M, et al: Outcomes in hypertensive patients at high cardiovascular risk treated with regimens based on valsartan or amlodipine: the VALUE randomised trial. Lancet 2004;363:2022–2031. 32 Kanaya AM, Herrington D, Vittinghoff E, et al: Glycemic effects of postmenopausal hormone therapy: the Heart and Estrogen/progestin Replacement Study. A randomized, double-blind, placebo-controlled trial. Ann Intern Med 2003;138:1–9.

Discussion Dr. Katsilambros: My compliments for this nice presentation. Please allow me a very specific question. If a close relative of yours had impaired glucose tolerance, was 50 years old, and had a body mass index (BMI) of 34, this person would then either have to change his lifestyle or take drugs. The question at this point, would you also give him drugs if he changed his lifestyle? But if this person was not willing at all to change his lifestyle towards the better, would you subscribe drugs straight from the beginning? Dr. Chiasson: That is an interesting question and certainly a problem that we are or will be faced with. After looking at the data of the Diabetes Prevention Program (DPP), the Diabetes Prevention Study (DPS) and the Dutchen study, there is no

39

Chiasson doubt that the lifestyle modification program should be first-line treatment of prediabetic states. Of course if people are open to that, want to lose weight, want to exercise, I think that has to be supported through prescription as well as through counseling. However, in this population, which I would follow over time, if I see that plasma glucose starts to increase gradually despite these measures, I would consider pharmacological intervention. In Canada those are the guidelines. We are not sure what the failings are of the lifestyle modification program. Of course if we follow plasma glucose, whether it be fasting or after 75 g glucose, and it keeps increasing, then we should add drugs. Metformin and acarbose have been shown to be effective. I think the postprandial glucose is a very strong predictor of progression to diabetes. Personally I don’t think that even in patients who are unwilling to change their lifestyle I would start with medication because I don’t think that medication and then overindulging would be of any benefit, at least in the long-term. We all have the responsibility to be very convincing, follow the patients and encourage them to comply to a well-designed lifestyle modification program. And on top of that I would do exactly as for cooperative patients and would most likely end up adding some drugs. I think the drugs are there as an alternative, certainly in combination with lifestyle modification, although there is no study showing that this is going to be more effective. One would suspect that, but again if they operate with the same mechanism, perhaps not, we don’t know. Dr. Ditschuneit: With bariatric surgery remarkable amounts of weight loss can be achieved. The weight loss is 30% in the Swedish Obese Subjects study, and after 10 years it is still 16%. How can we explain that some of these patients, although they remain obese, will not develop diabetes after being operated? Is there a range of BMI in which the risk of diabetes is particularly low or high? Dr. Chiasson: I will answer the second part of your question first. I don’t think that bariatric surgery cures diabetes. I don’t think that we have any cure for diabetes, though there is in most cases a normalization of the plasma glucose, certainly the fasting plasma glucose. However if the patients are challenged, the first phase insulin secretion still remains totally out of proportion; so the basic problem is still there. Why do they respond to this huge amount of weight loss, much more than with diet alone? My simple view on the subject is that the pancreas is still able to compensate for so much insulin resistance, and by doing that insulin resistance is decreased to a level where the ␤-cell can compensate for the insulin resistance. That is my simple explanation for why despite the fact that they are still obese there is a normalization of plasma glucose. Dr. Hill: I think we have to consider the economics of all this. For example, if you performed bariatric surgery on everybody in the US who qualifies with a BMI above 40, the cost would be more than the total health care dollars spent by the US. So it is probably one reason the surgeons won’t take over the world in this area. But even with drugs, xenical costs about USD 100/month. As diabetes is increasing, can you perhaps address how we can afford these kinds of approaches to its management? Dr. Chiasson: First regarding surgery, you are right, bariatric surgery costs a lot, but it is quite well known that in the US and in Canada there has been a huge increase over the last 5 years at least for bariatric surgery. There is a large population and a huge marker for this type of approach because obesity has become a major problem. But the government, at least in Canada where we have a national medicare system, would never pay for any bariatric surgery. Regarding the economics, I have not seen economic studies done with orlistat, I know that it has been done with metformin and acarbose, and I know that for those two drugs that are on the market at the present time the cost:benefit ratio is positive. Even the studies that have only looked at intervention have not taken into account the long-term complications. If you add that in, then it is even more cost-effective. So yes, I think it would be cost-effective to do some prevention study. I think it would pay for the government, at least in Canada, to pay for a year at a gym rather than paying for the costs of obesity and diabetic complications in the long-term.

40

Pharmacological and Surgical Intervention for the Prevention of Diabetes Dr. Hill: But then you have to get them to use the gym. Dr. Slama: I would like to challenge the answer you gave to Dr. Katsilambros who asked you if you would give drugs to those patients who are not willing to lose weight or modify their lifestyle. You answer was you do not want them to overindulge. Do you really think that those patients who do not lose weight overindulge; do you really think that those patients who are not able to exercise regularly overindulge? Could you tell us how much the mean weight loss is in your regular patients on a large cohort; how much exercise you obtain in your cohort of patients? I don’t have the results from my practice, but it is my opinion that very few patients really lose weight in the long-term. Should we not give them drug therapy just because they don’t lose weight? Dr. Chiasson: I think that is an over-interpretation of my statement. Like you, I think my success is very low, and that obesity comes from overindulging. I don’t think that obesity appeared in Europe during the Second World War, so it comes from overindulging. The second thing is that the group that Dr. Katsilambros mentioned, somebody who does not want to stick to a diet, and those are usually the people who do overindulge. That is a different thing. I am not saying that I would not give them drugs; what I am saying is that in cooperative patients it very easy, they want to lose weight, and despite the fact that they don’t lose much weight, they go on a diet even if they maintain their weight and do exercise a little bit more than they did, that will be beneficial. By following the plasma glucose I then decide when to start with pharmacological intervention. In the other patients I have to work harder to convince them that it is important, but I will do exactly the same thing. Dr. Slama: Work harder but without too much hope because we know what results we can obtain. I agree of course that they are overindulging once in their life. Dr. Chiasson: That is true, but if you put diabetics on a diet and follow them, even if they don’t lose weight the blood sugar will come down. Most of the time it is because they have better eating habits even if they don’t lose weight. That in itself, and I strongly believe in the postprandial plasma glucose. I strongly believe that if the diet is just modified, and I will not talk about glycemic index or fibers but I think they should be part of the diet, if the postprandial plasma glucose is reduced, I think just there a long-term beneficial effect should be seen, even if the patient doesn’t lose weight, but if the plasma glucose is followed it should tell you when to start drug intervention. Dr. Slama: So on this point we agree. But you cannot link your prescription to the results obtained beforehand on lifestyle modification. Of course you only need to prescribe drugs for those people exhibiting some adherence to your principles, trying their best whatever the results, and when you are convinced that they are trying their best. Dr. Chiasson: Yes, as I said how do you define diet? I think that is a problem. This being said, I think that the plasma glucose should dictate the prescription. Ms. Franz: Your earlier comment that people with diabetes have improved blood glucose levels even without weight loss because of better eating habits is of interest. It has been suggested that a reduced energy intake is more important than weight loss for improving insulin resistance [1]. Do you think the reduction in diabetes in subjects undergoing bariatric surgery is just as likely from their reduced food intake as it is from their weight loss? Dr. Chiasson: Definitely they eat a lot less because they have a small stomach. One of the problems is that in most of the studies the metabolic assessment has not been done properly, and that is a major problem. Very often we don’t have a good evaluation beforehand and we don’t have a good follow-up after bariatric surgery. But now there are some very interesting things being done using laparoscopy; the ileum is being changed and put right after the stomach. There are a lot of interesting things being done, but again that would still remain for a morbidly obese population. Dr. Halimi: I have two questions. First in contrast to the lifestyle intervention in the DPP study and the Study to Prevent NIDDM, do you think that metformin and

41

Chiasson acarbose do not prevent diabetes, but represent an early treatment of type 2? Second with regard to acarbose, do you think that it could be an indirect effect via gut factors? Dr. Chiasson: I didn’t want to get involved in this discussion as to whether it is treatment or prevention of diabetes. Of course it is a good question and I don’t think we can answer it. Certain drugs are being tested and it is going to be very difficult. Metformin and thiazolidinediones (TZDs) have long-term effects; it takes 2–3 months to have a maximum effect, and I presume it takes 2–3 months to see the effect disappear. So it will be difficult unless the patients are followed over time to tell whether the disease is being treated or prevented. With acarbose it is different because acarbose is not absorbed, it decreases the postprandial, that is the main effect. It does not have any impact on the oral glucose tolerance test. The oral glucose tolerance test still remains a valid diagnostic tool. We all know that acarbose and all the ␣-glucose inhibitors increase GLIT-1, but whether that has an impact on prevention, we don’t know; whether it has an impact on reducing the emptying of the stomach, we don’t know. It would be very difficult to study that with acarbose. I doubt very much that it has an effect on the regeneration of the ␤-cell. It is an interesting concept; I think it works in rats, but I don’t think it works in humans. I have seen data in rats with ␣-glucosidase inhibitors that show exactly the same regeneration of the ␤-cell as TZD. So one has to be very careful in interpreting the data from rat to human. There are still lots of open questions; there are other factors, gut factors, short chain fatty acid and all this kind of thing that could be involved, but we don’t have the answer yet. But independent of all that, whether we treat early diabetes or prevent diabetes, if we delay the complications by 5 or 10 years that will reduce the burden on the health care system. Dr. Jialun Chen: Thank you very much for your comprehensive review. You mentioned the TRIPOD study, and we now know that there are two other studies with troglitazone. The duration of the intervention is only 10 months. Compared with the placebo group there was a risk reduction of diabetes of 75%. The subgroup study from Denver and Philadelphia continued the DPP study. They have completed the study after 3 years and the risk reduction was 88%. It is very impressive, and I would like to know your opinion about this issue. Dr. Chiasson: I am not familiar with the TZD study. The only studies that I am aware of are the DREAM study, in which I am participating, and the Fronso study which is ongoing at the present time. But so far we don’t have any data from those studies. There was also the Proactive study which is a study in diabetes. In the DPP, and you will correct me if I am wrong, the troglitazone arm was very effective, it had a 75% risk reduction when the drug was discontinued. But when the drug was tested later on, it went back up to the incidence equivalent to the control group. The TRIPOD study, in which Buchanan tried to make a big thing about the prolonged effect of troglitazone, was based on 6 events and to me 6 events can fall anywhere over an 8month period. So I think that most of the drugs as well as the non-pharmacological interventions only work if they are followed them or if you take them.

Reference 1 Assali AR, Ganor A, Beigel Y, et al: Insulin resistance in obesity: body-weight or energy balance? J Endocrinol 2001;171:293–298.

42

Glycemic Effect of Carbohydrates Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 43–56, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

The Glycemic Index: Methodology and Use Cyril W.C. Kendall, Livia S.A. Augustin, Azadeh Emam, Andrea R. Josse, Nishta Saxena, David J.A. Jenkins Department of Nutritional Sciences, Faculty of Medicine, University of Toronto, and Clinical Nutrition and Risk Factor Modification Center, St. Michael’s Hospital; Toronto, Ont., Canada

Abstract The glycemic index concept owes much to the dietary fiber hypothesis that fiber would reduce the rate of nutrient absorption and increase the value of carbohydrate foods in the maintenance of health and treatment of disease. However, properties and components of food other than its fiber content contribute to the glycemic and endocrine responses postprandially. The aim of the glycemic index classification of foods was therefore to assist in the physiological classification of carbohydrate foods which, it was hoped, would be of relevance in the prevention and treatment of chronic diseases such as diabetes. Over the past two decades low glycemic index diets have been reported to improve glycemic control in diabetic subjects, to reduce serum lipids in hyperlipidemic subjects and possibly to aid in weight control. In large cohort studies, low glycemic index or glycemic load diets (glycemic index multiplied by total carbohydrate) have also been associated with higher levels of high-density lipoprotein cholesterol, reduced C-reactive protein concentrations and with a decreased risk of developing diabetes and cardiovascular disease. More recently, some case-control and cohort studies have also found positive associations between the dietary glycemic index and the risk of colon, breast and other cancers. While the glycemic index concept continues to be debated and there remain inconsistencies in the data, sufficient positive findings have emerged to suggest that the glycemic index is an aspect of diet of potential importance in the treatment and prevention of chronic diseases. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basell

The glycemic index concept combines the concepts of the dietary fiber hypothesis and the insulin-resistance syndrome in an attempt to find dietary 43

Kendall/Augustin/Emam/Josse/Saxena/Jenkins strategies to prevent and manage chronic disease (diabetes, coronary heart disease – CHD). The original dietary fiber hypothesis predicted that different carbohydrate foods would result in different physiological responses, notably, lower postprandial glycemic responses for the same amount of carbohydrate from high fiber, less processed foods. By the mid 1970s, viscous fibers were shown to effectively blunt both postprandial glucose and insulin responses [1]. Shortly thereafter, different low fiber starchy carbohydrate foods were shown to have different effects on the blood glucose response curve in both healthy and diabetic subjects [2]. In the early 1980s, it was considered important to start to systematically document differences between carbohydrate foods to allow a more rational form of carbohydrate exchange and to provide physiological information on foods rather than simply theoretical assessments based on macronutrient and fiber composition. The glycemic index was therefore developed to address these needs [3].

Glycemic Index Methodology The glycemic index assesses the blood glucose response of a fixed amount of available carbohydrate (generally 50 g) from a test food to the same amount of available carbohydrate from a standard food (glucose or white bread). Both test and standard foods were taken by the same subject in order to account for individual variation. The test food’s blood glucose response area was then expressed as a percentage of the standard’s. In general, the insulin responses observed for particular foods were found to be similar to the glycemic responses they elicited [4]. It also appeared that the rate of digestion of the food was a major determinant of the observed glycemic response [5]. Thus, when foods were digested in vitro, the rate of liberation of the carbohydrate products of digestion over 3–5 h reflected the blood glucose area in vivo [5]. A range of intrinsic and extrinsic factors which alter the rate of gastrointestinal motility, digestion and absorption of starchy foods was found to result in differences in glycemic index including the nature of the starch, cooking, particle size, the presence of fiber, fat and proteins [6]. Foods with lower glycemic indices were often the starchy staples of traditional cultures, such as pasta, whole grain pumpernickel breads, cracked wheat or barley, rice, dried peas, beans and lentils. In fact it appeared that the traditional use of low glycemic index carbohydrate foods in the diet was particularly true for those cultures which are now experiencing high rates of diabetes, such as the Pima Indians and the Australian Aborigines, and where the change to high glycemic index foods has been a recent phenomenon [7]. Obviously other lifestyle factors, including reduced physical activity and increased obesity also play a major role in increasing diabetes risk. Nevertheless, it was speculated that the desire for sweetness, resulting from rapid carbohydrate breakdown of refined starches in the mouth, has resulted in the selection of rapidly 44

The Glycemic Index: Methodology and Use digested and hence high glycemic index foods as cultures become more affluent. Thus high glycemic index foods are potentially a further dietary factor favoring the development of chronic disease.

Criticisms of the Glycemic Index Concept It has been argued that the glycemic index lacks clinical utility because differences in glycemic indices between foods are lost once these foods are taken in a mixed meal [8]. This observation may in part be explained by the fact that when a mixed meal consists of several carbohydrate sources, the effect of the lower glycemic index component is diluted in proportion to the amount of carbohydrate from the other foods. It is therefore essential that the mixed meal glycemic index be properly calculated [9]. It has also been said that the addition of small amounts of fat to the meal greatly alter the glycemic response. However studies where 8–24 g fat was fed in mixed meals containing 38–104 g carbohydrate had little effect on the predicted glycemic response [10]. Furthermore while large deviations in the dietary macronutrient profile may occur from time to time in the lives of individuals, these differences are likely to average out over time. It has also been argued that the glycemic index is a difficult concept to explain to the public and that it will add further needless complication and potential dietary restriction in management and prevention of diseases and that perhaps the modest gains that may be achieved are therefore not justified [11]. For the public, however, the glycemic index may simply be used as a tool for selecting better quality starchy foods. Over time it is hoped that the development of new low glycemic index foods will expand the range of choices to be selected not simply for the glycemic index but also for the fact that the foods in question may have a range of other health advantages. Obviously a certain amount of dietary understanding is required. Thus, carrots with a high glycemic index are not to be excluded from the diet. It is realized that there are other considerations relevant to the consumption of carrots and that the glycemic index is not significant in a low calorie food which contains high levels of other desirable factors (i.e. fiber, vitamins, minerals, etc.).

Metabolic Effects of Low Glycemic Index Foods It has been hypothesized that the health benefits of low glycemic index foods are due to their metabolic effects, specifically their ability to slow the rate at which glucose is absorbed from the small intestine (fig. 1). Studies in healthy men have demonstrated some of the metabolic effects of reducing the rate of absorption, for example, when glucose solution was sipped at an 45

Kendall/Augustin/Emam/Josse/Saxena/Jenkins

a

b

Fig. 1. Spreading the nutrient load. Hypothetical effect on of a low glycemic index meal (a) versus a high glycemic index meal (b) on gastrointestinal absorption of carbohydrate. Reproduced with permission by the American Journal of Clinical Nutrition. © Am Clin Nutr. American Society for Nutrition.

even rate over 180 min (sipping), as opposed to being taken as a bolus [12]. A marked economy in insulin secretion with sipping was seen along with lower serum free fatty acid (FFA) levels compared to the bolus (fig. 2). In part, this improvement, also observed after feeding low glycemic index meals, may be the result of sustained tissue insulinization, suppression of FFA release and the absence of a counter-regulatory endocrine response [12]. Other studies using low glycemic index meals have demonstrated an improved second meal effect reminiscent of the Staub-Traugott effect (where the first meal improves the glucose tolerance of the second meal) and related the improved postprandial glycemia of the second standard meal to prolonged suppression of FFA levels [13]. Increased food frequency has also been shown to reduce glycemic and insulinemic responses over the day in diabetic subjects and in longer term studies has been associated with reduced fasting blood lipid concentrations despite consumption of the same foods at the same 24-hour caloric intake [14]. However, spreading the nutrient load does not appear to increase the thermogenic effect of diets, which would favor weight reduction. It is therefore possible that some of the advantages in glycemia and fasting blood lipids seen after prolonging absorption by reducing the rate of absorption such as by sipping or by increasing the dietary meal frequency, may relate simply to reduced fluxes in nutrient uptake and less perturbation of the endocrine environment. 46

Serum FFA (µmol/l)

Blood glucose (mmol/l)

The Glycemic Index: Methodology and Use

8 6 4 2 0 60

0

Serum insulin (pmol/l) Serum C-peptide (pmol/l)

180

240

600 400 200 0 0

Plasma GIP (pmol/l)

120

60

120

180

240

400 200 0 0

60

120

180

240

1,500 1,000 500 0 0

60

120

180

240

100 50 0 0

60

120

180

240

Time (min)

Fig. 2. Mean ⫾ SE blood glucose, serum free fatty acid (FFA), insulin, and C-peptide; and plasma gastric inhibitory polypeptide (GIP) after taking glucose solution (50 g in 700 ml water) as a bolus over 5 min at time 0 (䊏) or sipping the same solution over 0–3.5 h at an even rate (䊐).

47

Kendall/Augustin/Emam/Josse/Saxena/Jenkins Table 1. Clinical trials assessing the effect of low glycemic index (GI) diets on glycosylated proteins in type-1 and 2 diabetes mellitus Diabetes type

n

Duration weeks

Change in diet GI valuea

Change in glycosylated proteins, %

Type of glycosylated protein

I I I I I II

7 8 9 54 104 8

6 3 2 24 52 2

⫺12 ⫺14 ⫺27 ⫺20 ⫺1.2 ⫺23

HbA1c Fructosamine Fructosamine HbA1c HbA1c HbA1c

II II II II II

16 6 15 25 20

12 6 2 12 3

⫺14 ⫺28 ⫺27 ⫺5 ⫺31

II I and II I and II

28 18 24

4 5 4

⫺20 ⫺26 ⫺5

⫺19*,b ⫺18*,b ⫺6.5c ⫺5.5c ⫺6.5*,c ⫺6.6*,b ⫺2.6c ⫺11*,b ⫺8*,c ⫺3.4*,c ⫺11*,b ⫺5.9*,b ⫺2.5*,c ⫺1.8c ⫺13*,c ⫺3c

HbA1c Fructosamine Fructosamine Fructosamine HbA1c Fructosamine Fructosamine Fructosamine HbA1c

*Significant effect: p ⬍ 0.05. aFrom high GI diet (reference food: white bread). bTreatment difference from baseline (within low GI treatment). cEnd-point difference (between treatments).

Clinical Effects of Low Glycemic Index Approaches In diabetes (types 1 and 2) the majority of studies (10 of 14; table 1) have shown improvements in glycosylated proteins [15]. Where it has been assessed, improvements in clotting factors and reductions in low-density lipoprotein cholesterol and triglycerides have also been reported, particularly in individuals with elevated blood lipids [15]. These reductions have been achieved despite no significant change in body weight. A recent meta-analysis examined the data from several randomized trials that assessed the efficacy of low glycemic index diets to control glycemia [16]. Seven of the 10 studies included in this analysis found improvements in glycemic control with a mean reduction in hemoglobin A1c (HbA1c) of 0.43% compared to the high glycemic index diet [16]. These data, though not in themselves definitive, are encouraging. Drug therapies which reduce the rate of glucose absorption have also been shown to be effective in the control of diabetes and its complications. ␣Glucosidase inhibitors such as acarbose, which reduce the rate of absorption of starch, sucrose, and to a lesser extent maltose, have been shown in large, multicenter trials to result in a significant reduction in HbA1c in type-2 48

The Glycemic Index: Methodology and Use diabetes [17, 18]. Furthermore, use of acarbose in patients with diabetes in the context of the UK Prospective Diabetes Study, improved glucose control, expressed as reduced HbA1c, to a degree similar to that achieved by current hypoglycemic therapy (e.g. metformin and sulfonylurea) [19]. Finally, in the STOP-NIDDM trial subjects with impaired glucose tolerance who received 100 mg acarbose three times daily showed a significantly reduced rate of conversion to diabetes versus the control group [20]. Findings of this nature provide additional encouragement that the principle of spreading the nutrient load by dietary means, in addition to altering the amount and nature of the macronutrients, may one day play a role in modifying glycemia in the management of diabetes.

Epidemiological Evidence for the Health Effects of Low Glycemic Index Diets Studies using the NHANES III database and also a British study have demonstrated a negative relationship between the glycemic index and highdensity lipoprotein cholesterol suggesting low glycemic index diets may preserve high-density lipoprotein cholesterol and have a positive effect in reducing CHD risk [21, 22]. In this respect both the Nurses Study and the Health Professionals Study have shown potential benefits in relation to diabetes incidence and CHD risk dependent on the dietary glycemic index [21, 23]. Of particular interest in terms of CHD was the observation that below a body mass index of 23 kg/m2 there was no association of the dietary glycemic index with CHD, suggesting that the effect of the dietary glycemic index may be increasingly important in those with a greater degree of insulin resistance. Situations where insulin resistance and insulin-like growth factors have been implicated are the so-called diet-related cancers – colon, breast and prostate [21]. It has been hypothesized that the risk of these cancers may be increased by insulin resistance [21]. Recently an Italian case-control study reported that the dietary glycemic index was related to colon cancer risk: the higher the glycemic index the greater the risk. Of particular interest was the finding that the association was only significant for white bread (high glycemic index food) which contrasted with pasta (low glycemic index food) where no association existed [21]. The epidemiological literature therefore provides further support for the role of low glycemic index diets in chronic disease prevention.

Diet, Lifestyle and Diabetes Control The value of diet and lifestyle change including the incidence of type-2 diabetes mellitus has been clearly demonstrated in three recent studies with 49

Kendall/Augustin/Emam/Josse/Saxena/Jenkins Table 2. Coronary heart disease risk factors associated in type-2 diabetes mellitus 1 2 3 4

Low high-density lipoproteins (high triglycerides) Modified low-density lipoproteins (glycosylated, oxidized, acetylated) Raised blood pressure Raised uric acid levels (a primary factor or simply an indicator of other abnormalities?) 5 Raised serum free fatty acid levels 6 Raised insulin levels 7 Raised factors VII and VIII, fibrinogen, and plasminogen activator inhibitor, and low tissue plasminogen activator levels

reductions of 30–50% in diabetes incidence in high-risk groups [24–26]. The Diabetes Prevention Program trial found that diet and lifestyle reduced the diabetes incidence by 58% and was more effective than metformin at 31% risk reduction [24]. Epidemiological studies, notably the Nurses Health Study, attributed 91% of the risk of type-2 diabetes to five major diet and lifestyle factors which included regular exercise, a body mass index of ⬍25 kg/m2, cereal fiber and consumption of low glycemic index diets [25].

Importance of Glycemic Control in Diabetes Based on the Diabetes Control and Complications Trial data in type-1 diabetes [27], it was considered that improved glycemic control would help in the prevention of diabetes-associated complications [28]. This view has been confirmed by the results of the UK Prospective Diabetes Study in type-2 diabetes [19]. Loss of excess body weight continues to be a goal for patients with type-2 diabetes to improve blood glucose control and serum lipid profiles. Furthermore, in view of the threefold to fourfold increased risk of CHD that accompanies diabetes mellitus, additional strategies are sought to normalize a constellation of CHD risk factors (table 2), including serum lipids. Recently the American Diabetes Association (ADA) stated that the ‘regulation of blood glucose to achieve near-normal levels is a primary goal in the management of diabetes, and, thus, dietary techniques that limit hyperglycemia following a meal are likely important in limiting the complications of diabetes’ [29]. The ADA added that ‘a recent analysis of the randomized controlled trials that have examined the efficacy of the glycemic index on overall blood glucose control indicates that the use of this technique can provide an additional benefit over that observed when total carbohydrate is considered alone’ [29]. However, with reference to the glycemic index, the ADA concluded that ‘the relationship between glycemic index and glycemic 50

The Glycemic Index: Methodology and Use load and the development of type-2 diabetes remains unclear at this time’ [29]. So while there is general agreement on the importance of glycemic control in controlling diabetes and preventing its associated complications and some recognition of the importance of controlling postprandial hyperglycemia, the glycemic index concept remains of interest but is at present not fully accepted.

Future Directions Within the last few years considerable interest has developed in the links between insulin resistance, the generation of reactive oxygen species, tissue damage and the liberation of pro-inflammatory cytokines and acute phase proteins, which appear to be involved in the development and progression of both diabetes and CHD. There is some research that indicates that the glycemic index may play a role in this sequence of events. In one study Creactive protein has been found to relate to high glycemic load diets [15]. Studies have also demonstrated that the postprandial rise in glucose is associated with depression of serum antioxidants [30] including lycopene and vitamin E. Data indicate that the higher the level of postprandial glycemia the greater the depression of serum antioxidants [30]. The concept is evolving that increased insulin resistance may result from oxidative stress. Supplementing subjects with the antioxidant vitamin E has been shown to improve glycemic control. Studies such as these suggest a possible beneficial role for low glycemic index diets through reducing oxidative damage. However, we are still in the early stages of this research area and longer term studies will be required to better define the relevance of these interesting new findings and what significance the glycemic index concept may have in it.

Conclusion The dietary glycemic index concept supports a role for the rate of carbohydrate digestion in the prevention and treatment of chronic disease, including those diseases which have been highlighted in the dietary fiber and insulin-resistance syndrome hypotheses. This concept should not be seen as particularly radical at a time when pharmacological approaches to slowing absorption, notably the ␣-glycoside hydrolase inhibitors, are now accepted in the management of diabetes. Further longer term efficacy studies as well as effectiveness studies are required to better determine the importance of the glycemic index in the regulation of blood glucose and the prevention of diabetic complications, particularly in relations to CHD risk factors. The possible role of the glycemic index in decreasing postprandial oxidative stress and pro-inflammatory processes also merits further investigation. 51

Kendall/Augustin/Emam/Josse/Saxena/Jenkins References 1 Jenkins DJ, Leeds AR, Gassull MA, et al: Decrease in postprandial insulin and glucose concentrations by guar and pectin. Ann Intern Med 1977;86:20–23. 2 Crapo PA, Reaven G, Olefsky J: Plasma glucose and insulin response to orally administered simple and complex carbohydrates. Diabetes 1976;25:741–747. 3 Jenkins DJ, Wolever TM, Taylor RH, et al: Glycemic index of foods: a physiological basis for carbohydrate exchange. Am J Clin Nutr 1981;34:362–366. 4 Wolever TM, Bolognesi C: Source and amount of carbohydrate affect postprandial glucose and insulin in normal subjects. J Nutr 1996;126:2798–2806. 5 Jenkins DJ, Ghafari H, Wolever TM, et al: Relationship between rate of digestion of foods and post-prandial glycaemia. Diabetologia 1982;22:450–455. 6 Thorne MJ, Thompson LU, Jenkins DJ: Factors affecting starch digestibility and the glycemic response with special reference to legumes. Am J Clin Nutr 1983;38:481–488. 7 Thorburn AW, Brand JC, Truswell AS: Slowly digested and absorbed carbohydrate in traditional bushfoods: a protective factor against diabetes? Am J Clin Nutr 1987;45:98–106. 8 Coulston, AM, Hollenbeck CB, Swislocki Al, Reaven GM: Effect of source of dietary carbohydrate on plasma glucose and insulin responses to mixed meals in subjects with NIDDM. Diabetes Care 1987;10:395–400. 9 Wolever TM, Nuttall FQ, Lee R, et al: Prediction of the relative blood glucose response of mixed meals using the white bread glycemic index. Diabetes Care 1985;8:418–428. 10 Wolever TM, Bolognesi C: Prediction of glucose and insulin responses of normal subjects after consuming mixed meals varying in energy, protein, fat, carbohydrate and glycemic index. J Nutr 1996;126:2807–2812. 11 Coulston AM, Reaven GM: Much ado about (almost) nothing. Diabetes Care 1997;20:241–243. 12 Jenkins DJ, Wolever TM, Ocana AM, et al: Metabolic effects of reducing rate of glucose ingestion by single bolus versus continuous sipping. Diabetes 1990;39:775–781. 13 Jenkins DJ, Wolever TM, Taylor RH, et al: Slow release dietary carbohydrate improves second meal tolerance. Am J Clin Nutr 1982;35:1339–1346. 14 Jenkins DJ, Wolever TM, Vuksan V, et al: Nibbling versus gorging: metabolic advantages of increased meal frequency. N Engl J Med 1989;321:929–934. 15 Jenkins DJ, Jenkins AL, Augustin LS, Kendall CW: Dietary therapy in type 2 diabetes mellitus: spreading the nutrient load; in LeRoith D, Taylor SI, Olefsky JM (eds): Diabetes Mellitus: A Fundatmental and Clinical Text, ed 3. Philadelphia, Lippincott Williams & Wilkins, 2004, pp 1085–1097. 16 Brand-Miller J, Hayne S, Petocz P, Colagiuri S: Low-glycemic index diets in the management of diabetes: a meta-analysis of randomized controlled trials. Diabetes Care 2003;26: 2261–2267. 17 Chiasson JL, Josse RG, Hunt JA, et al: Efficacy of acarbose in the treatment of patients with non-insulin-dependent diabetes mellitus. A multicenter controlled clinical trial. Ann Intern Med 1994;121:928–935. 18 Rodger NW, Chiasson JL, Josse RG, et al: Clinical experience with acarbose: results of a Canadian multicentre study. Clin Invest Med 1995;18:318–324. 19 Holman RR, Cull CA, Turner RC: A randomized double-blind trial of acarbose in type 2 diabetes shows improved glycemic control over 3 years (UK Prospective Diabetes Study 44). Diabetes Care 1999;22:960–964. 20 Chiasson JL, Josse RG, Gomis R, et al: Acarbose for prevention of type 2 diabetes mellitus: the STOP-NIDDM randomised trial. Lancet 2002;359:2072–2077. 21 Augustin LS, Franceschi S, Jenkins DJ, et al: Glycemic index in chronic disease: a review. Eur J Clin Nutr 2003;56:1049–1071. 22 Frost G, Leeds AA, Doré CJ, et al: Glycaemic index as a determinant of serum HDL-cholesterol concentration. Lancet 1999;353:1045–1048. 23 Salmeron J, Manson JE, Stampfer MJ, et al: Dietary fiber, glycemic load, and risk of noninsulin-dependent diabetes mellitus in women. JAMA 1997;277:472–477. 24 Knowler WC, Barrett-Connor E, Fowler SE, et al: Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393–403. 25 Hu FB, Manson JE, Stampfer MJ, et al: Diet, lifestyle, and the risk of type 2 diabetes mellitus in women. N Engl J Med 2001;345:790–797.

52

The Glycemic Index: Methodology and Use 26 Tuomilehto J, Lindstrom J, Eriksson JG, et al: Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344:1343–1350. 27 The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin treated diabetes mellitus. The Diabetes Control and Complications Trial Research Group. N Engl J Med 1993;329:977–986. 28 Franz MJ, Horton ES, Bantle JP, et al: Nutrition principles for the management of diabetes and related complications. Diabetes Care 1994;17:490–518. 29 Sheard NF, Clark NG, Brand-Miller JC, et al: Dietary carbohydrate (amount and type) in the prevention and management of diabetes: a statement by the American Diabetes Association. Diabetes Care 2004;27:2266–2271. 30 Ceriello A, Bortolotti N, Motz E, et al: Meal-generated oxidative stress in type 2 diabetic patients. Diabetes Care 1998;21:1529–1533.

Discussion Dr. Knowler: You started your talk by showing the effects of frequent small meals compared to larger meals in saying that the glycemic index (GI) concept evolved from this. Doesn’t that suggest that it might be just as easy to change meal frequency to frequent small meals rather than focusing on the GI for food? Might the same thing be accomplish more easily? Dr. Kendall: I think that this could be a viable approach to try to decrease the postprandial rise in blood glucose. So you can sip your drinks or eat a number of small meals a day and that would be an effective strategy but obviously you want to have a healthy diet if you are doing this. We want to give people as many choices as possible. However, trying to get people to eat 3 meals a day is pretty tough, so trying to get them to eat 17 meals a day would probably be a bit tougher. Dr. Mooradian: You started out defining the GI as essentially derived from the area under the curve of a test carbohydrate against a reference carbohydrate such as white bread or glucose, but you ended up focusing your discussion on the kinetics of glucose absorption and low GI of slowly absorbed food. How would this slow absorption change the area under the curve? Dr. Kendall: When carbohydrate is slowly absorbed then the glucose rise is not going to be as large, the insulin will be secreted and will clear the glucose from the blood as well. Again it is the very high glucose peaks which appear to be most damaging. This is when the oxidative reactions are occurring. Dr. Mooradian: Are you assuming that if the influx of glucose is slower then insulin will be more effective? Dr. Kendall: The glucose peak just won’t be the same and the level of insulin secretion won’t be the same either. I think that was demonstrated in the slide shown where glucose was given either as a bolus or sipped throughout the day. Not only was the peak glucose reduced, peak insulin was reduced and serum C peptide was also reduced, as was C peptide excretion [1]. Dr. Mooradian: That is not a fair comparison. It is a quantitative effect; if you are taking less carbohydrate you are going to have less glucose. Dr. Kendall: In that study the same amount of carbohydrate was provided, that was the point. So if the same amount of carbohydrate is given as a bolus or sipped throughout the day, the latter induces less insulin secretion [1]. Dr. Katsilambros: We know that acarbose also delays carbohydrate absorption. Do you know of any study comparing high GI with low acarbose versus low GI with placebo, or no acarbose? Dr. Kendall: It would be nice to do the low acarbose plus low GI study. I don’t believe it has been compared, at least not that I am aware of. But again if you reduce

53

Kendall/Augustin/Emam/Josse/Saxena/Jenkins the rate of glucose absorption with acarbose, you get a similar effect that hopefully is achieved with a low GI diet. Dr. Schiffrin: How do you assess the oxidative reaction after the hyperglycemic peak? What biomarkers are studied? Dr. Kendall: It can be assessed in a number of different ways. The study presented was conducted in Italy, by Ceriello [2]. He has been looking at different measures of oxidative stress, so it could be as simple as looking at some antioxidant, vitamins or carotinoids. We know that diabetic subjects tend to have lower levels of antioxidants in their blood, so there are a number of different measures that can be looked at. Dr. Halimi: How do you explain that despite a lower insulin response with low GI foods there is no benefit on body weight? Dr. Kendall: We are actually talking about two different things. In tightly controlled studies of the GI we are trying to maintain body weight, as this would otherwise be a confounder by affecting insulin sensitivity and fasting glucose. To address your question we need free-living studies, where subjects are allowed to lose weight, comparing high GI diets with low GI diets. We have seen this in a weight loss study conducted by Raatz et al. [3]. However, when you are achieving significant weight loss by controlling energy intake, often both study groups achieve similar results and it becomes difficult to detect differences in insulin resistance and fasting glucose. Dr. Slama: Can I add something to this question? First of all I don’t see any reason why people exhibiting less hyperglycemia and less hyperinsulinemia should lose weight because all the calories which are ingested are metabolized and the net energy balance is exactly the same whatever the blood glucose excursion. We have just finished a study done on a Weightwatcher population and we tried to see if people eating low GI food lose weight. In fact there was exactly the same weight loss in the two populations with high and with low GI; the only difference was satiety. The people on low GI said that it was much easier to follow the diet rather than the high GI because of satiety. So there is no reason to lose weight more than to equal the energy intake but it is an easier way to follow the diet. Dr. Kendall: I think that is a good point. Many low GI foods are whole foods, the energy density would tend to be lower. If you look at low GI food you actually do get an increased excretion or malabsorption of starch as well. There is a correlation between low GI foods and increased starch excretion. Dr. Jianqin Sun: I have a question regarding the low GI for the lower body mass index. I would like to know what the mechanism for the weight loss is. My second question is with regard to the GI classification in terms of low, median and high; it seems a little different from the reference described in the literature. Dr. Kendall: The second question first. I was using the bead index, so to use the glucose index you have to multiply those numbers by 0.7. The reason for weight loss with low GI? Again low GI diets tend to contain many whole foods; they tend to be higher in fiber as well, so this may be responsible for the increased weight loss. Dr. Jianqin Sun: Is there any evidence for the energy expenditure difference between the low GI diet and the high GI diet? Dr. Kendall: I don’t think it has been looked at very carefully. There have been some studies looking at postprandial association with appetite. If you look at a lunch intake after a low versus a high GI breakfast, the amount of food consumed after the low GI breakfast tends to be lower [4]. Dr. Katsilambros: I would like to come back to Dr. Slama’s comment about the same weight loss between low and high GI diets in humans. I recall a study published 3 years ago in rats fed high and low GI diets. These animals are a good model because exactly the same amount of energy can be given to them, and we are not sure what happens in humans. All the experimental animals on a low GI diet lost more weight.

54

The Glycemic Index: Methodology and Use We shouldn’t ignore these studies. Certainly with humans it does not seem to happen but again we don’t know how strictly humans adhere to the given instructions. Dr. Kendall: Again we tend to think that all carbohydrates are available or that all protein and fat are absorbed and I don’t think we are that efficient in absorbing all the nutrients that we take in. We have recently run a study looking at nut intake at 3 different doses and on the highest nut intake there is an excretion of about 60 kcal/day with the high nuts and approximately 15–20% of the energy from nuts was not absorbed [5]. This represents a significant proportion of the diet. This may be one of the reasons why whole foods, intact particles, are extremely healthy. Not only is there reduced absorption of some nutrients but they are then reaching the large intestine. What happens to that food system in the large intestine? We know it is quite good for the bacteria and it may have a myriad of other biochemical and physiological effects that are extremely healthy. Dr. Katsilambros: It might be a good explanation. Are there any observations concerning physical activity and energy and movements, automatic movements of the muscles on the one diet and on the other diet? This could at least explain the animal findings. Dr. Kendall: That is a good point. The GI has been looked at in terms of energy performance and with the low GI diet at least in some studies there seems to be increased performance [6]. I think this is an area that has not really been looked at very carefully. Dr. Slama: I would like to add something to what Dr. Katsilambros said, and later I will show some results from an animal study. You are right that rats gained less weight during the first weeks and then caught up, and at 5 weeks finished up with exactly the same body weight. Mr. Wuersch: This was one of our concerns 20 years ago when we started this project on the GI: What role does the availability of the nutrients play? When we take two extremes, for example potato or pulses, which have a GI ratio of roughly 1–2 or 1–2.5, we find that 10% of the carbohydrate is not absorbed in the pulses, which is confirmed by the resistant starch analysis. This means that on the diet side, it represents a maximum of 5% of the energy intake, so it comes back to your 80 calories roughly for a daily intake. Consequently this difference is probably not so significant and it has nothing to do or only marginally to do with the GI measured [7]. Dr. Kendall: I think that is also an important point. So while there is a loss of carbohydrate in some low GI foods, this does not account for the dramatic reduction in GI in the food. However that 5% loss in calories in terms of weight maintenance could be extremely important. We tend to gain 0.45 kg a year, 0.9 kg a year, it is a very slow progression in terms of weight gain. The balance between weight maintenance and weight gain is very finely tuned, so anything that can help us to excrete some energy could be extremely important. Dr. T. Wilkin: I am not sure this question is a sensible one. You are obliged to use the GI because of the individual variation you would otherwise get. So what you are left with are relative, comparative data. How big is the range in absolute terms that you get between different foods? Is that a meaningful question? There could be relative differences but they could be very narrow differences in absolute terms. Dr. Kendall: No, the absolute difference can be quite large. Nuts have an extremely low GI of around 20 or so, and pasta will have a GI of around 60, mash potatoes are around 100, boiled potatoes may be slightly less than 100. Dr. T. Wilkin: That tells me there is a ratio of 6 or 7 to 1 across the range, but is that what is that meant in absolute terms? Dr. Kendall: Yes, a range of about 6- to 7-fold which is quite large. So there is variation from day to day in someone’s response to particular foods. We are still dealing with a black box, and if you look at some of the studies in saturated fat reduction, the majority of subjects reduce their low-density lipoprotein (LDL) cholesterol but there

55

Kendall/Augustin/Emam/Josse/Saxena/Jenkins are still some subjects who raise their LDL. So I don’t think you would interpret those data as saturated fat is good for the majority but not for those two individuals. You expect differences in response between individuals at particular times. The person who had an increase in LDL with decreased saturated fat on one occasion will most likely have a decrease in LDL with decreased saturated fat on most occasions. Dr. Gerasimidi-Vazeou: I would like to ask you whether there are any long-term studies regarding the effect of GI diets on HbA1c in type-1 diabetic patients? Dr. Kendall: I don’t believe any very long-term studies have been undertaken in type-1 diabetic subjects. Most of the studies have been conducted in type-2 diabetes and most have not been of great length, typically around 12 weeks. I know with the increased interest in carbohydrates and GI, more longer-term studies are in progress. So I think we will begin to see data from these longer-term studies in the next year or two. Dr. Bantle: I think the GI is largely predictable based on three main determinants which are fiber, fat and fructose. I ask you to comment on whether you think that is true. I would also like to ask you about the effect on the GI when you combine nutrients in a meal. In fact, does this not substantially diminish the difference among foods of different GIs? Dr. Kendall: If there are large amounts of fat in a meal and large amounts of protein, that can affect the glycemic response, but typically these are not at a level that will have a significant effect. We have looked at the glycemic responses to nuts, which are high in fat. A standard white bread control was provided with 0, 28, 56, or 85 g of nuts, and there was always a significant reduction at the 56- and 85-gram level. But this is a dramatic level of nuts, 56 and 85 g [8]. So although you can see an effect, it is not typically seen for most meals. In terms of a fiber effect, foods that have high levels of soluble fiber will have a reduced glycemic response. Although the same is not true for non-viscous fiber. For example whole wheat bread, brown bread or grains that have been finely ground will have the same GI as white bread. So food form is important as is the nature of starch. Pasta, which is also made from wheat flour, has a much lower glycemic response than does brown or white bread.

References 1 Jenkins DJ, Wolever TM, Ocana AM, et al: Metabolic effects of reducing rate of glucose ingestion by single bolus versus continuous sipping. Diabetes 1990;39:775–781. 2 Ceriello A: Postprandial hyperglycemia and diabetes complications: is it time to treat? Diabetes 2005;54:1–7. 3 Raatz SK, Torkelson CJ, Redmon JB, et al: Reduced glycemic index and glycemic load diets do not increase the effects of energy restriction on weight loss and insulin sensitivity in obese men and women. J Nutr 2005;135:2387–2391. 4 Anderson GH, Woodend D: Effect of glycemic carbohydrates on short-term satiety and food intake. Nutr Rev 2003;61:S17–S26. 5 Kendall CW, Ellis PR, Marchie A, et al: Lipid bioavailability from almonds: implications for cardiovascular health and weight loss – a randomized controlled dose-response study (abstract). Ann Nutr Metab 2003;47:617. 6 Kirwan JP, Cyr-Campbell D, Campbell WW, et al: Effects of moderate and high glycemic index meals on metabolism and exercise performance. Metabolism 2001;50:849–855. 7 Schweizer TF, Andersson H, Langkilde AM, et al: Nutrients excreted in ileostomy effluents after consumption of mixed diets with beans or potatoes. II. Starch, dietary fibre and sugars. Eur J Clin Nutr 1990;44:567–575. 8 Kendall CW, Marchie A, Parker TL, et al: Effect of nut consumption on postprandial starch digestion – a dose-response study (abstract). Ann Nut Metab 2003;47:636.

56

Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 57–72, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

The Argument against Glycemic Index: What Are the Other Options? Marion J. Franz Nutrition Concepts by Franz, Inc., Minneapolis, MN, USA

Abstract There is debate among professionals regarding the use of the glycemic index (GI) for meal planning. In type-1 diabetes, there are 4 studies (average duration ⬃4 weeks) comparing high versus low GI diets; none reported improvements in HbA1c, and although 2 reported improvements in fructosamine, 2 reported no differences. In type-2 diabetes, there are 12 studies (average duration ⬃5 weeks); 3 reported improvements in HbA1c and fructosamine, 5 reported no differences in HBA1c, and 3 reported no differences in fructosamine. In adults, there is limited evidence that a low GI diet is beneficial for weight loss or satiety. Three epidemiologic studies reported that a low GI/glycemic load (GL) is associated with a reduced risk of developing diabetes or prevalence of insulin resistance; however, 5 studies report no association between GI/GL and the risk of developing diabetes, fasting insulin or insulin resistance, or adiposity. In general, the total amount of carbohydrate in a meal is the primary meal-planning strategy for people with diabetes. The GI can be used as an adjunct for the fine tuning of postprandial blood glucose responses. Other food/meal-planning interventions have been shown to be more effective than the use of the GI. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

Carbohydrates can be classified based on their chemical structure and/or based on their physiological effects. Based on chemical structure the major dietary carbohydrate groups are: sugars, starch, and fiber. Defining carbohydrates by chemical structure, however, does not take into account their physiologically differing responses, such as differences in satiety value, gastric emptying times, and effects on glucose and insulin levels. To better define physiological responses, the concept of a glycemic index (GI) was developed by David Jenkins and colleagues in the 1970s. The GI is defined as the relative area under the postprandial glucose curve comparing 50 g of digestible 57

Franz carbohydrate from a test food to 50 g of carbohydrate ingested from 50 g of a standard food – pure glucose or white bread. As defined, the GI takes into account only the type of carbohydrate in food and ignores the total amount of carbohydrate in a typical food serving, although both the type and amount of carbohydrate influence the postprandial and insulin responses of a given ingested food [1]. In an attempt to compare the glucose-raising effect of foods with their widely differing amounts of carbohydrate, the glycemic load (GL) was developed. The GL is defined as the GI multiplied by the grams of carbohydrate in a specific portion of a carbohydrate-containing food or meal. For example, although carrots have a high GI, because they contain relatively small amounts of carbohydrate they have a low GL. Perhaps the most widely held belief in regard to carbohydrates and diabetes has been the assumption that the response to food carbohydrates was based on the chemical composition, e.g. sugars versus starches. However, about 20 studies have reported that when subjects are allowed to choose from a variety of starches and sugars, the glycemic response is identical as long as the total amount of carbohydrate is kept constant [2]. Therefore, the American Diabetes Association concluded that with regard to the effects of carbohydrate on glucose concentrations, the total amount of carbohydrate in meals and snacks is a key strategy [3].

Definitions of a Low, Medium, or High GI Food or Diet In general, if the GI number of a food is higher than 70, the food is considered to have a ‘high’ GI value. If the number is lower than 55, it has a ‘low’ GI. Medium GI foods have GIs in the range of 55–70. But because the methods used to calculate GI are not always standard (e.g. glucose and bread are both used for comparison), knowing the exact GI number may not be as important as knowing if a food is a ‘low’, ‘moderate’, or ‘high’ GI food. In general, low GI foods include non-starchy vegetables, fruits, dairy products, lentils and sugars such as fructose and lactose. Moderate GI foods are unprocessed grains and mixed dishes. High GI foods are refined grains and potatoes. Table 1 lists examples of low, moderate, and high GI foods. The GI of foods can be modified by changing the nature of starch (e.g. increasing amylopectin or decreasing amylose or by combing starch with protein [gluten]); altering cooking methods (e.g. reducing the extent of gelatinization or cooling to prevent retrogradation); using larger particle or piece size, adding some acids such as those in vinegar and lemon juice; adding soluble fibers such as psyllium, or by adding or substituting lactose, fructose, or sucrose for starch or glucose. It is likely that most people already eat a moderate GI diet. In the Nurses Health Study, the lowest quintile of GI for women was 64 and in the highest quintile 77, a difference of only 13 units [4]. In the Health Professional 58

The Argument against Glycemic Index Table 1. Examples of low, moderate, and high glycemic index (GI) foods Low GI foods

Moderate GI foods

High GI foods

Whole grain breads Bran cereals Fruit Non-starchy vegetables Milk Ice cream premium Reduced fat yogurt Oatmeal (slow cook oats) Lentils/baked beans

Rye bread Frosted flakes Soft drinks Ice cream Sucrose Pasta Soups Pizza Candy bars

White bread Corn flakes Watermelon Instant mashed potatoes Sports drinks Baked potatoes Carrots Rice Bagels

Study, GI was 65 in the lowest quintile for men and 79 in the highest, again a difference of only 14 units [5]. In the Insulin Resistance and Atherosclerosis Study, the average caloric intake was reported to be 1,987 kcal/day with 220 g/day of digestible carbohydrate, 19 g/day of fiber, and an average GI of 58 and an average GL of 128 [6]. It is unknown if further lowering of the GI can be achieved long-term. Such small differences suggest that it may be both impractical and unreasonable to drive the GI down in the general population. In the only 1-year study published thus far, one group of individuals attempted to follow a low GI diet while the other group ate their usual foods [7]. At the end of the year there was no significant difference in the GI between groups.

Problems with the GI – Methodology and Variability If foods do have different glycemic responses, why the controversy? Several problems with the methodology used to determine GI have been cited [1]. As noted above, 50 g of digestible carbohydrate from foods is used to determine the GI. Although this would seem logical, in reality it does not reflect the actual amounts of carbohydrate contributed by individual foods in the usual diet. As mentioned, foods that in usual portion sizes contribute minimal amounts of carbohydrate, despite having a high GI, would not elicit much of a glycemic response. And conversely, foods such as pizza may have a lower GI number, but the usual portion size consumed would contribute a considerably greater amount of total carbohydrate resulting in a higher GL. The GI only measures glucose above the beginning fasting glucose. If it measured what occurs naturally, the fasting glucose value would decrease over time and the area under the curve (AUC) would be greater. Therefore, some researchers favor the use of the whole AUC as the real measure of glucose availability [1]. If the AUC is calculated in this manner, the differences in 59

Franz GIs between foods are greatly attenuated. For example, a person with a fasting glucose of 75 mg/dl (4.2 mmol/l) ingests two foods, one with a GI of 100 and the other a GI of 72. If the GI were calculated by using the whole glucose AUC instead of only the area above the fasting glucose, the values would be 100 and 92, respectively. The difference changes from 29 to 8 units. Furthermore, the GI is measured in the morning, after an overnight fast. Several studies have reported that if the GI is measured after lunch, the differences in GI would be considerably less than after breakfast. A major problem, however, is the reproducibility and variability of the glucose response. Reproducibility of the glucose response in the same subject has not been adequately studied, and the individual blood glucose response to any food or meal is highly variable, both within and between individuals – ranging from 23 to 54% [1, 8]. Values for the GI of foods are broad. For example, Australian potatoes have a GI of 87–101, placing them in the high GI group. In the United States and Canada potatoes have a GI ranging from 56 to 77, placing them in the moderate GI group [9]. GI values for boiled rice vary from 45 to 112. Bananas range from 30 to 70, partially depending on their degree of ripeness. The GI from different types of spaghetti varies even more widely. White, durum-wheat semolina spaghetti varies from 45 to 65, depending on the length of cooking time. Even prepared foods vary greatly. For example All-Bran cereal ranged from 30 in Australia to 51 in Canada, and corn chips varied from 72 in 1985 to 42 in 1998 [10]. Testing of blood glucose before and after eating a particular food by an individual may be the best way to determine an individual’s glycemic response. However, this may not be accurate either as rarely do individuals eat only one food at a meal. The glucose response to a meal is determined first of all by the total amount of carbohydrate in a meal and secondly by the combination of carbohydrate foods in a meal. If a high GI food is eaten in combination with a low GI food, the GI response will be moderate. For example, a high GI cereal eaten with milk will have a moderate GI response. Adding fat to potatoes can also change their GI response. The GI response to a particular food can also be lowered by eating less of the food. Protein in the meal does not affect the glycemic response [11], whereas large amounts of fat may. Another major problem is that the GI is not the best indicator of healthy food choices. Although many healthy foods have a low GI (whole grains, fruits, vegetables, legumes, dairy products), there are also foods of questionable value with low or moderate GI values. For example, soft drinks, candies, sugars, and high fat foods fall into this questionable category. The GI of foods can be lowered by adding or substituting sugars, especially fructose, sugar alcohols, or fat. In addition, the insulin response to a given food is not linear and is not consistently related to either the carbohydrate content or glycemic effect of food [12]. The insulin response to a 100-gram portion of a particular food is not 60

The Argument against Glycemic Index Table 2. Summary of high versus low glycemic index diets in diabetes Outcome

Type-1 diabetes (4 studies; 36 subjects; ⬃4 weeks duration)

Type-2 diabetes (12 studies; 175 subjects; ⬃5 weeks duration)

low GI diet significantly better

no difference between diets

low GI diet significantly better

no difference between diets

HbA1c

0

Fructosamine

2 (n ⫽ 15) [14, 15] 0 1 (n ⫽ 8) [15]

2 (n ⫽ 19) [14, 16] 2 (n ⫽ 21) [16, 17] 3 (n ⫽ 27) [14, 15, 16] 2 (n ⫽ 21) [16, 17]

3 (n ⫽ 42) [19, 27, 28] 3 (n ⫽ 41) [20, 21, 24] 1 (n ⫽ 12) [27] 1 (n ⫽ 10) [26]

5 (n ⫽ 67) [16, 18, 23, 25, 26] 3 (n ⫽ 54) [18, 22, 23] 10 (n ⫽ 153) [16, 18, 19–25, 28] 3 (n ⫽ 49) [19, 23, 27]

1 (n ⫽ 7) [14] 1 (n ⫽ 12) [16]

2 (n ⫽ 20) [15, 16] 2 (n ⫽ 15) [14, 15]

4 (n ⫽ 53) [20, 21, 24, 27] 2 (n ⫽ 31) [16, 22]

HDL

0

3 (n ⫽ 27) [14–16]

1 (n ⫽ 21) [23]

LDL

–

–

1 (n ⫽ 12) [27]

5 (n ⫽ 69) [16, 18, 19, 22, 28] 8 (n ⫽ 113) [18, 19–21, 24, 25, 27, 28] 7 (n ⫽ 91) [16, 18, 20, 22, 24, 27, 28] 7 (n ⫽ 116) [18, 20, 22–25, 28]

FPG Insulin requirements/ levels Cholesterol Triglycerides

double that of the GI standard of 50 g. Although GI is relatively constant across different populations (lean, obese, impaired glucose tolerance, diabetes), the insulin responses vary widely [13]. Postprandial insulin responses to isocaloric amounts of food are not closely related to either the carbohydrate content or the glycemic effects of food; the glycemic response accounts for only 23% of the variability in insulin [12]. Thus, GI may not be a good marker to predict insulin response.

GI and Diabetes – Treatment and Prevention Although different carbohydrates do produce differing glycemic responses, to be of benefit clinically, this benefit should translate into longterm improvements in glycemia or lipids. Table 2 summarizes the research comparing high versus low GI diets on glucose and lipid outcomes in studies with a minimum duration of 2 weeks [14–28]. Examining the data reveals no clear trend in outcome benefits. A meta-analysis of GI diets in persons with diabetes reported a reduction in HbA1c by 0.4% units (a 7.4% decrease) from 61

Franz Table 3. Epidemiologic studies on glycemic index (GI), glycemic load (GL), and fiber and effect on insulin resistance and risk of diabetes Positive association Nurses Health Study 1997 [4] Health Professional Study Framingham Offspring Cohort 2004 [33]

Negative association GI/GL associated Iowa Women’s with risk of Study 2000 [31] developing diabetes GI/GL associated Zutphen Elderly with risk Study 2000 [32]

No association GI/GL and risk developing diabetes

No association between GI and fasting insulin GI/GL associated Atherosclerosis Risk No association with prevalence in Communities between GI/GL with of insulin Study (ARIC) insulin-resistant resistance 2002 [34] diseases; fiber beneficial association Inter99 Study No association 2005 [35] between GI/GL and insulin resistance; fiber beneficially associated with insulin resistance Insulin No association Resistance and between GI/GL and Atherosclerosis insulin sensitivity Study (IRAS) and adiposity; fiber 2005 [6] beneficial association

low GI diets compared to high GI diets and a reduction in HbA1c from baseline by 0.35% units [29]. However, included in the meta-analysis are two studies not included in table 2. In the study by Gilbertson et al. [7], there were no significant differences in the GI of the diets at study end in the study arms. Thus, it is questionable if the lowering of the GI in one study arm can be attributed to differences in GI. In the study by Giacco et al. [30], 50 g of fiber are included in the low GI group. Fiber and GI are not necessarily the same and therefore it is unknown if the effect on glucose response is due to the low GI of the diet or to the fiber content. These two studies account for 47% of the total subjects in the meta-analysis. Early epidemiological studies suggest that a low GI/GL diet may play a role in the prevention of diabetes [4, 5]. Table 3 summarizes the outcomes from later studies and from countries other than the United States. Whereas, 3 studies report a positive association between low GI/GL diets and the risk of developing diabetes or insulin resistance, 5 do not. Interestingly, in 3 studies fiber was positively associated with insulin sensitivity, whereas GI/GL was not.

62

The Argument against Glycemic Index GI and Weight Loss and Appetite Although many of the popular diet books promote the use of low GI foods in their diets, there is minimal evidence to suggest that a low GI diet contributes to weight loss. The diet books claim that high GI foods are digested rapidly causing blood glucose to surge and an oversecretion of insulin, both contributing to insulin resistance, increased appetite, overeating, and weight gain. However, figure 1 illustrates the glucose response to carbohydrate loads containing 50 g of glucose from dextrose, rice, corn, potato, and bread [36]. Although the peak responses differ slightly they occur at approximately the same time. Furthermore, the actual change in glucose is only ⬃20 mg/dl (1 mmol/l). Reporting the results in percentage of change as is done with the GI often distorts the small to modest changes in glycemic responses. Actual changes in postprandial glucose levels would be more useful. In reviewing weight changes in the 12 studies in persons with type-2 diabetes in table 2, one study reported more weight loss on the high GI diet, 1 study reported more weight loss on the low GI diet although the calories were also less in that arm, and 10 studies reported no differences in weight loss. Heilbronn et al. [25] asked the question does reducing the GI of a high carbohydrate diet confer a benefit during energy restriction. After 4 weeks of an energy-restricted diet, subjects were randomized to a low or a high GI diet. There was no difference in weight loss between the groups. Raben [37] reviewed weight loss on high versus low GI diets and concluded that no clear pattern of difference between low and high GI in terms of decreased food intake or weight loss is shown. In 20 studies (6 days to 6 months in duration), the average weight loss in 4 studies was 1.5 kg on the low GI diet, in 2 studies 1.6 kg on a high GI diet, and in 14 studies there was no difference in weight loss between the high and the low GI diets. It was concluded that there was no evidence that low GI foods are superior to high GI foods for long-term body weight control. On the other hand, Pawlak et al. [38] concluded that based on epidemiological evidence and a small weight loss study in adolescents there was support for a role of the GI in disease prevention and treatment, weight loss and satiety. Their hypothesis is that recommended lower intakes of dietary fat for weight loss has actually contributed to the increase in obesity, and that reduced dietary fat results in compensatory increases in the consumption of high GI carbohydrate, principally refined starchy foods and concentrated sugar. Because they are rapidly digested such foods cause a large increase in postprandially glucose and insulin causing a decrease in satiety and weight gain. The recently released US Dietary Guidelines for Americans, however, strongly recommend calorie control to manage body weight. They state: ‘When it comes to weight control, calories do count – not the proportions of carbohydrate, fat and protein in the diet. . . . The healthiest way to reduce 63

Franz 150

Potato

Dextrose

Corn

Rice

Bread

Glucose (mg/dl)

140 130 120 110 100 90 80 0

15

30

45

60

75

90

105

120

135

150

165

180

Time (min)

Fig. 1. Glucose response to 50 g of digestible carbohydrate from dextrose, rice, corn, potato and bread. Reprinted with permission from Crapo et al. [31].

calorie intake is to reduce one’s intake of added sugars, solid fats, and alcohol – they all provide calories, but they do not provide essential nutrients.’

Potential Benefits and Problems to Food Companies Incorporating the GI to Foods The current low carbohydrate diet fad is phasing out and, therefore, food companies may be looking for a new marketing approach. With the current publicity regarding the GI in diet books and by many health providers, it would appear logical to think that knowing the GI of foods would be useful. The low fat diet approach stopped working when food companies flooded the market with low fat foods that were not necessarily lower in calories. The same can be said regarding the low carb approach. Instead of avoiding carbohydrate foods which would lead to a reduction in calories the market became flooded with low carbohydrate (net carbohydrate, low impact carbohydrate) foods that also were not necessarily lower in calories. This can potentially happen to low GI foods as well. The methodology for determining the GI has been criticized and each food must be tested individually. This requires food companies to invest financial resources into GI testing of their food products. The GI must be tested in humans and cannot be determined from data bases. Companies that have foods with a naturally low GI have an advantage if the low GI movement catches on. The type of food companies most likely to benefit are companies with whole grain or unprocessed starch foods and dairy products. 64

The Argument against Glycemic Index Table 4. Randomized controlled trials and outcome studies: outcomes from medical nutrition therapy (MNT) for type-2 and type-1 diabetes Type of study

HbA1c decreases

MNT studies in type-2 diabetes UK Prospective 1.9% Diabetes Study Group [40] Franz et al. [41] 0.9% (4-year duration) 1.7% (newly diagnosed) Goldhaber-Fiebert 1.8% et al. [42] Ziemer et al. [43] 1.9% Lemon et al. [44] 1.7%

Interventions Reduced energy/fat intake Reduced energy/fat intake Portion control and healthier food substitutions Healthy food choices Carbohydrate counting and simplified meal plans

MNT in diabetes self-management training studies (DSMT) in type-2 diabetes Sadur et al. [45] 1.3% Healthy food choices for improved glycemia Rickheim et al. [46] 2% (newly diagnosed) Carbohydrate counting Polonsky et al. [47] 2.3% intensive DSMT Reduced energy/fat/ 1.7% standard DSMT carbohydrate Banister et al. [48] 1.5% Basic nutrition concepts/ individualized meal planning strategies MNT in type-1 diabetes management Delahanty and 0.9% Halford [49]

Kulkarni et al. [50]

1% (newly diagnosed)

Pieber et al. [51]

1.2%

DAFNE Study Group [52]

1%

DCCT who followed meal plan (carbohydrate counting) ⬎90% of the time compared to those ⬍45% of the time Carbohydrate counting and exchange lists Carbohydrate counting/ adjusting insulin dose based on carbohydrate intake Carbohydrate counting/ adjusting insulin dose based on carbohydrate intake

The problem with the low GI approach will be similar to the problems that occurred with the development of low fat and low carbohydrate foods. Food companies can develop low GI foods. This can be done by adding or substituting sugars, especially fructose, and fat to foods. However, this may change the ‘healthy’ image of low GI foods and turn off health providers and potentially the public to low GI foods. 65

Franz What Might Be More Helpful? Randomized controlled trials and outcome studies of medical nutrition therapy (MNT) in the treatment of type-1 and type-2 diabetes have reported improved HbA1c levels of approximately 1–2% units (a 15–22% decrease); however, low GI diets compared to high GI diets have only been shown to lower HbA1c by ⬃0.4% (7% decrease) [29]. MNT in these studies is provided by dietitians/nutritionists as MNT only or as MNT in combination with diabetes self-management training [39]. Interventions for type-2 diabetes include reduced energy intake, reduced fat/carbohydrate intake, carbohydrate counting and basic nutrition and healthy food choices for improved glycemia. Interventions for type-1 diabetes include carbohydrate counting and matching insulin doses to planned carbohydrate intake. Outcomes of the intervention are known by 3 months. Table 4 summarizes randomized controlled trials and outcome studies of MNT in type-1 and type-2 diabetes. Therefore it has been recommended that interventions demonstrated to have the greatest effect on overall glycemic control be implemented first [53]. Individuals with diabetes who can understand and benefit from the concept of low GI foods may be able to use pre- and post-meal blood glucose monitoring to fine tune their food choices. Until research demonstrates long-term benefits for people with diabetes in the use of the GI, making food choices should be kept as easy and simple as possible. Understanding what foods are carbohydrates, knowing portion sizes, and knowing how many servings to select for meals, and, if desired, for snacks, will benefit the majority of the people with diabetes and can increase variety and flexibility in food choices. References 1 Pi-Sunyer FX: Glycemic index and disease. Am J Clin Nutr 2002;76:290S–298S. 2 Franz MJ, Bantle JP, Beebe CA, et al: Evidence-based nutrition principles and recommendations for the treatment and prevention of diabetes and related complications. Diabetes Care 2002;25:148–198. 3 Sheard NF, Clark NG, Brand-Miller JC, et al: Dietary carbohydrate (amount and type) in the prevention and management of diabetes. A statement by the American Diabetes Association. Diabetes Care 2004;27:2266–2271. 4 Salmeron J, Manson JE, Stampfer MJ, et al: Dietary fiber, glycemic load and risk of noninsulin-dependent diabetes mellitus in women. JAMA 1997;277:472–477. 5 Salmeron J, Ascherio A, Rimm EB, et al: Dietary fiber, glycemic load, and risk of NIDDM in men. Diabetes Care 1997;20:545–550. 6 Liese A, Schulz M, Fang F, et al: Dietary glycemic index and glycemic load, carbohydrate and fiber intake, and measures of insulin sensitivity, secretion, and adiposity in the Insulin Resistance Atherosclerosis Study. Diabetes Care 2005;28:2832–2838. 7 Gilbertson HR, Brand-Miller JC, Thorburn AW, et al: The effect of flexible low glycemic index dietary advice versus measured carbohydrate exchange diets on glycemic control in children with type 1 diabetes. Diabetes Care 2001;24:1137–1143. 8 Wolever TM, Vorster HH, Bjorck I, et al: Determination of the glycaemic index of foods: interlaboratory study. Eur J Clin Nutr 1998;52:924–928.

66

The Argument against Glycemic Index 9 Fernandes G, Velangi A, Wolever TMS: Glycemic index of potatoes commonly consumed in North America. J Am Diet Assoc 2005;105:557–562. 10 Foster-Powell K, Holt SHA, Brand-Miller JC: International table of glycemic index and glycemic load values: 2002. Am J Clin Nutr 2002;76:5–56. 11 Franz MJ: Protein and diabetes: much advice, little research. Curr Diab Rep 2002;2:457–464. 12 Holt SHA, Brand-Miller JC, Petocz P: An insulin index of foods: the insulin demand generated by 1000-kg portions of common foods. Am J Clin Nutr 1997;66:1264–1276. 13 Wolever TMS, Chiasson J-L, Hunt JA, et al: Similarity of relative glycaemic but not relative insulinaeic responses in norm, IGT, and diabetic subjects. Nutr Res 1998;18:1667–1676. 14 Collier GR, Giudici S, Kalmusky J, et al: Low glycaemic index starchy foods improve glucose control and lower serum cholesterol in diabetic children. Diabetes Nutr Metab 1988;1: 11–19. 15 Fontveille AM, Acosta M, Rizkalla SW, et al: A moderate switch from high to low glycaemicindex diets for 3 weeks improve metabolic control of type I (IDDM) diabaetic subjects. Diabetes Nutr Metab 1988;1:139–143. 16 Fontvieille AM, Rizkalla SW, Penformis A, et al: The use of low glycaemic index foods improves metabolic control of diabetic subjects over five weeks. Diabet Med 1992;9: 444–450. 17 Lafrance L, Rabasa-Lhoret R, Poisson D, et al: The effects of different glycaemic index foods and dietary fibre intake on glycaemic control in type 1 diabetic patients on intensive insulin therapy. Diabet Med 1998;15:972–978. 18 Jenkins DJA, Wolever TMS, Buckley G, et al: Low glycemic index starchy foods in the diabetic diet. Am J Clin Nutr 1988;48:248–254. 19 Brand JC, Colagiuri S, Crossman D, et al: Low glycemic index foods improve long-term glycemic control in NIDDM. Diabetes Care 1991;14:95–101. 20 Wolever TMS, Jenkins DJA, Vuksan V, et al: Beneficial effects of low-glycemic index diet in type 2 diabetes. Diabet Med 1992;9:451–458. 21 Wolever TMS, Jenkins DJA, Vuksan V, et al: Beneficial effect of low-glycaemic index diet in overweight NIDDM. Diabetes Care 1992;15:562–564. 22 Frost G, Wilding S, Beecham J: Dietary advice based on the glycaemic index improves dietary profiles and metabolic control in type 2 diabetic patients. Diabet Med 1994;11:397–401. 23 Luscome ND, Noakes M, Clifton PM: Diets high and low in glycemic index versus high monounsaturated fat diets: effects on glucose and lipid metabolism in NIDDM. Eur J Clin Nutr 1999;53:473–478. 24 Jarvi A, Karlstrom B, Grandfeldt Y, et al: Improved glycaemic control and lipid profile and normalized fibrinolytic activity on a low glycemic index diet in type 2 diabetic patients. Diabetes Care 1999;22:10–18. 25 Heilbronn LK, Noakes M, Clifton PM: The effect of high- and low glycemic index energy restricted diets on plasma lipid and glucose profiles in type 2 diabetic subjects with varying glycemic control. J Am Coll Nutr 202:21:120–127. 26 Komindr S, Lerdvutrisopon N, Ingsriswang S, et al: Effect of long-term intake of Asian food with different glycemic indices on diabetic control and protein conservation in type 2 diabetic patients. J Med Assoc Thai 2001;84:85–97. 27 Rizkalle SW, Taghrid L, laromiguiere M, et al: Improved plasma glucose control, whole-body glucose utilization, and lipid profile on a low-glycemic index diet in type 2 diabetic men. Diabetes Care 2004;27:1866–1872. 28 Jimenez-Cruz A, Bacardi-Gascon M, Turnbull WH, et al: A flexible, low-glycemic index Mexican-style diet in overweight and obese subjects with type 2 diabetes improves metabolic parameters during a 6-week treatment period. Diabetes Care 2003;26:1967–1970. 29 Brand-Miller J, Hayne S, Petocz P, Colagiuri S: Low-glycemic index diets in the management of diabetes: a meta-analysis of randomized controlled trials. Diabetes Care 2003;26: 2466–2468. 30 Giacco R, Pariool M, Rivellese AA, et al: Long-term dietary treatment with increased amounts of fiber-rich, low-glycemic index natural foods improves blood glucose control and reduces the number of hypoglycemic events in type 1 diabetic patients. Diabetes Care 2000;23: 1461–1466. 31 Meyer KA, Kushi LH, Jacobs DR Jr, et al: Carbohydrates, dietary fiber, and incident type 2 diabetes in older women. Am J Clin Nutr 2000;71:921–930.

67

Franz 32 Feskens EJ, Loeber JG, Kromhout D: Diet and physical activity as determinants of hyperinsulinemia: the Zutphen Elderly Study. Am J Epidemiol 1994;140:350–360. 33 McKeown NM, Meigs JB, Liu S, et al: Carbohydrate nutrition, insulin resistance, and the prevalence of the metabolic syndrome in the Framingham Offspring Cohort. Diabetes Care 2004;27:538–546. 34 Stevens J, Ahn K, Juhaeri, et al: Dietary fiber intake and glycemic index and incidence of diabetes in African-American and white adults: the ARIC study. Diabetes Care 2002;25: 1715–1721. 35 Lau C, Faerch K, Glumer C, et al: Dietary glycemic index, glycemic load, fiber, simple sugars, and insulin resistance: the Inter99 study. Diabetes Care 2005;28:1397–1403. 36 Crapo PA, Reaven G, Olefsky J: Postprandial plasma-glucose and -insulin responses to different complex carbohydrates. Diabetes 1977;26:1178–1183. 37 Raben A: Should obese patients be counseled to follow a low-glycaemic index diet? No. Obes Rev 2002;3:245–256. 38 Pawlak DB, Ebbeling CB, Ludwig DS: Should obese patients be counseled to follow a lowglycaemic index diet? Yes. Obes Rev 2002;3:235–243. 39 Pastors JG, Franz MJ, Warshaw H, et al: How effective is medical nutrition therapy in diabetes care? J Am Diet Assoc 2003;103:827–831. 40 UK Prospective Diabetes Study 7: Response of fasting plasma glucose to diet therapy in newly presenting type II diabetic patients, UKPDS Group. Metabolism 1990;39:905–912. 41 Franz MJ, Monk A, Barry B, et al: Effectiveness of medical nutrition therapy provided by dietitians in the management of non-insulin-dependent diabetes mellitus: a randomized, controlled clinical trial. J Am Diet Assoc 1995;95:1009–1017. 42 Goldhaber-Fiebert JD, Goldhaber-Fiebert SN, Tristan ML, Nathan DM: Randomized controlled community-based nutrition and exercise intervention improves glycemia and cardiovascular risk factors in type 2 diabetic patients in rural Costa Rica. Diabetes Care 2003;26: 24–29. 43 Ziemer DC, Berkowitz KJ, Panayioto RM, et al: A simple meal plan emphasizing healthy food choices is as effective as an exchange-based meal plan for urban African Americans with type 2 diabetes. Diabetes Care 2003;26:1719–1724. 44 Lemon CC, Lacey K, Lohse B, et al: Outcomes monitoring of health, behavior, and quality of life after nutrition intervention in adults with type 2 diabetes. J Am Diet Assoc 2004;104: 1805–1815. 45 Sadur CN, Moline N, Costa M, et al: Diabetes management in a health maintenance organization. Efficacy of care management using cluster visits. Diabetes Care 1999;22: 2011–2017. 46 Rickheim PL, Weaver TW, Flader JL, Kendall DM: Assessment of group versus individual diabetes education: a randomized study. Diabetes Care 2002;25:269–274. 47 Polonsky WH, Earles J, Smith S, et al: Integrating medical management with diabetes selfmanagement training: a randomized control trial of the Diabetes Outpatient Intensive Treatment program. Diabetes Care 2003;26:2048–2053. 48 Banister NA, Jastrow ST, Hodges V, et al: Diabetes self-management training program in a community clinic improves patient outcomes at modest cost. J Am Diet Assoc 2004;104: 807–810. 49 Delahanty LM, Halford BN: The role of diet behaviors in achieving improved glycemic control in intensively treated patients in the Diabetes Control and Complications Trial. Diabetes Care 1993;16:1453–1458. 50 Kulkarni K, Castle G, Gregory R, et al: Nutrition Practice Guidelines for Type 1 Diabetes Mellitus positively affect dietitian practices and patient outcomes. The Diabetes Care and Education Dietetic Practice Group. J Am Diet Assoc 1998;98:62–70. 51 Pieber TR, Brunner GA, Schnedl WJ, et al: Evaluation of a structured outpatient group education program for intensive insulin therapy. Diabetes Care 1995;18:625–630. 52 DAFNE Study Group: Training in flexible, intensive insulin management to enable dietary freedom in people with type 1 diabetes: dose adjustment for normal eating (DAFNE) randomised controlled trial. BMJ 2002;325:746. 53 Franz MJ: The glycemic index: not the most effective nutrition therapy intervention. Diabetes Care 2003;26:2466–2468.

68

The Argument against Glycemic Index Discussion Dr. Katsilambros: Ms. Franz and Dr. Slama, I understand that existing studies in humans comparing low versus high glycemic index (GI) diets do not show any real weight change. However, during the previous discussion I mentioned that there are experimental animal studies that do show a weight change between low and high GI diets. I have found the slide from the study and with your permission I would like to show it. Ms. Franz: While Dr. Katsilambros gets ready to show the slide, I would like to make a comment on satiety studies. Although subjects may report feeling more satisfied after consuming different meals, this doesn’t always translate into eating fewer calories. For example, Stubbs et al. [1] in a 1-day study reported that although subjective hunger was less after a high protein breakfast compared to a high fat or high carbohydrate breakfast, lunch time intake 5 h later and energy intake for the rest of the day were similar after all three breakfasts. Currently the majority of research on satiety is very short-term and the effect of satiety on future calorie intake is rarely studied. While we wait for the slide does anyone else have a comment or questions? Dr. Slama: I would like to make three brief comments and ask you one question. To be fair you should quote Razels and Crapo and for European people Otto and Spater who published these results in 1973. They were first published in German and then in French, but they can still be found in Medline. The second thing is you said something which seems wrong to me: you say that adding milk to a portion of cereals will decrease the blood glucose. If you add a portion of skin milk to 50 g cereal, this will increase but not decrease blood glucose, unless you are using caribou milk which is very rich in lipids. Ms. Franz: Thank you for your comment. What I meant to say is that if you combine a high GI cereal with milk, a low GI food, you lower the GI of the meal. The high GI of the cereal or the low GI of milk is not maintained and instead you have a moderate GI response. Dr. Slama: My third remark is that we proponents of the low GI do not claim that the GI concept should be substituted by another concept, but it should be added to another concept. So my question is: how as a clinician can you make this claim if you have not tried it yourself? Ms. Franz: I have used the concept of differing glycemic responses from foods with many patients. For example, when examining food and glucose records you often find glucose responses that do not seem to make sense by just looking at the total amount of carbohydrate eaten. A possible explanation may be that different carbohydrate foods do have differing glucose responses even when the total amount of carbohydrate is the same. If persons with diabetes observe that some foods consistently cause their post-meal glucose response to be higher than other foods, even when keeping the carbohydrate amount consistent, the next time they eat that food they can try eating less of it or, if on insulin, increase their insulin dose. So I agree with the concept that foods with equal amounts of carbohydrate may have differing post-meal glucose responses which can be helpful to patients. They can use their food and glucose test records to make food choices that can beneficially lower their post-meal glucose responses. Dr. Slama: You don’t need to explain your concept to the patient. You only have to say eat that rather than that when you have the occasion or the choice. You don’t have to explain the area under the curve and the ratio and to multiply by the age. You only have to give some very simple advice. Ms. Franz: As a clinician, I am reluctant to suggest to patients that they should not eat potatoes because they have a high GI when I am not confident this is true. For example, while potatoes from Australia appear to have a high GI of 87–101, potatoes

69

Franz tested in the US and Canada had a moderate GI of 56–77 [2]. Therefore, my problem with the GI is with the variability and reliability of the published GI values. Dr. Slama: Try it yourself and you will be convinced. Dr. Katsilambros: This is a slide published in 2002 and these are rats studied over 32 weeks, not for 1 week. You see an enormous difference in weight, about 75 g at the end of 30 weeks. The upper line is high GI, the lower line is low GI. 75 g in a rat is something like 20 kg in a man. Still I absolutely agree and I expect that in humans this is not the case. But again in humans we are not absolutely sure how strictly the diet can be kept unless it is made in the laboratory and under close supervision. Finally I absolutely agree with you and Dr. Slama that this is an additive dietary treatment, it is not an alternative. Dr. Mooradian: What was the strain of rats used in the study? Dr. Katsilambros: I don’t remember. Dr. Mooradian: Because the strain makes a difference in terms of growth. One has to interpret these data as favorable in terms of weight reduction versus stunting of growth. Ms. Franz: Were the diets comparable in digestible or available carbohydrate and calories? Dr. Mooradian: I guess to be able to interpret that slide you need to know whether the rats were pair-fed and had exactly the same amount of calorie intake. Dr. Katsilambros: Yes, I show isoenergetic. Dr. Eshki: When looking at nutrition tools for diabetes, such as the carbohydrate count, GI and other nutrition tools, I believe all these methods are very effective. When providing consultation to patients, their diet must be customized. For instance, when an individual visits a tailor to have a suit made, the tailor is going to take the measurements and all other factors before customizing the suit. With my patients, I look at several factors, such as culture, lifestyle conditions, the patient’s onset, peak and duration of medication and the dietary history, and based on that I customize the proper diet. I see all nutrition tools as useful and don’t deny the effectiveness of one over the other. My question is, do you think the GI can be helpful in some way? Ms. Franz: Yes, as I mentioned previously the GI can be a helpful tool for patients who keep food and blood glucose records. However, as a clinician my first priority is to provide a framework that patients can use to plan food choices and meals and that is based on what the individual with diabetes feels is feasible and realistic for them to implement. In managing diabetes, how much a person eats is more important than what he/she eats. Just because a food has a low GI does not mean people with diabetes can eat unlimited portions of that food without affecting their blood glucose levels. Portions do count! I have found it helpful if patients understand what foods contain carbohydrate; what are average carbohydrate portion sizes, and how many carbohydrate servings to eat for meals or snacks, if desired. To plan food choices and meals people with diabetes can use carbohydrate counting, exchange lists, or experience. People with diabetes use their own experience to determine what works for them and what doesn’t, and the GI concept may help explain some of their experiences. If target blood glucose goals are not being met, decisions need to be made to determine if changes in food intake or in medical therapy (i.e. medications) are needed. Dr. Hill: It is very interesting to have these two talks back to back. While we can change people’s diets in the short-term, it is difficult to do this over the long-term. The question in my mind is whether the GI is a useful tool for long-term improvement of diets. Are people following the GI diet eating healthily? Do people know what to do with the message given by the GI? Are people able to stick with this concept more than other concepts? The best example is the recent Atkins diet craze which in the short-term is great at producing weight loss but in the long-term it doesn’t seem to be able to do this.

70

The Argument against Glycemic Index Ms. Franz: There is little, if any, evidence that people with diabetes can in the long-term change the GI of their usual diet. In a 1-year study, children in the low GI group did have significantly better HbA1c levels than the group using a carbohydrate exchange diet [3]. However, the study reported no differences in mean GI between the 2 groups at study end and even the authors stated it was difficult to attribute the difference in HbA1c to diet when there was no apparent difference in the mean GI. The majority of studies comparing low and high GI diets have been short-term. Furthermore, it is likely that most people already eat a moderate GI diet and it is not known if this can be changed long-term to a low GI diet. Dr. Hill: Some people have suggested putting the GI on food labels. As an educator, how do you think the population would handle that? Ms. Franz: You would have to ask somebody from a country where GI values are included on a food label as to the usefulness of this information. As an educator, I find the most useful information on the food label to be the serving size expressed in portions that patients can understand and the total grams of carbohydrate in that serving size. Total calories and grams of fat, saturated fat and protein in the serving size are also useful information. Dr. Slama: I would like to make some comments on what has been said. First of all we do not promote chocolate, ice cream, pizza, because they have a low GI. We do not leave people with a list of low GI food, but with a list of low GI healthy food. If time is short one must do what one can, but for me nutritional education is a long running process. I do not tell my patient’s right away that now they have to learn what the GI is. I come with it later on when I want to tune the results. So really it is a process which we have to take with precaution and time. Ms. Franz: I would agree with you. Most patients need support long-term to make and maintain lifestyle changes and part of that support over time may be fine tuning some of their food choices using the GI concept. My concern is when low GI foods are promoted to the public as being healthy food choices, and low GI food become ‘good’ foods and high GI foods become ‘bad’ foods. Dr. Slama: I agree that you have to take care of the amount of carbohydrate. But already diagnosed diabetic people are eating less carbohydrate than they should, and not too much. Ms. Franz: That is an interesting comment about how much carbohydrate people with diabetes should be and are eating. In this regard, I find the report from the United Kingdom Prospective Diabetes Study on estimated dietary intakes of interest. The intent of the nutrition intervention was to encourage patients to eat 50–55% of their energy intake as carbohydrate, protein 10–15%, and fat 30–35%. Despite the intensive intervention, patients reported a similar proportion of their energy intake as carbohydrate (43%) as the general population, protein intake 21%, and fat 37% [4]. Males reported an estimated energy intake of ⬃1,800 kcal/day and females ⬃1,450 kcal/day. This suggests that people with diabetes eat a moderate carbohydrate diet and do not eat either a high or a low carbohydrate diet. Dr. Gerasimidi-Vazeou: Regarding type-1 diabetic patients, taking only carbohydrate measurement and GI into consideration and omitting fat consumption is not sufficient to improve glycemic control. This is also true judging from the experience at our center. We live in a real world and our patients do not always follow our suggestions. Perhaps we should also take fat consumption into consideration because we know that in Western countries the latter is over 30% of the daily calorie intake. Ms. Franz: I certainly agree with you. Patients with both type-1 and type-2 diabetes need also to pay attention to total caloric intake as well as the total meat and fat servings they eat. You can’t just focus on carbohydrates.

71

Franz References 1 Stubbs RJ, van Wyk MC, Johnstone AM, Harbron CG: Breakfasts high in protein, fat or carbohydrate: effect on within-day appetite and energy balance. Eur J Clin Nutr 1996;50:409–417. 2 Fernandes G, Velangi A, Wolever TM: Glycemic index of potatoes commonly consumed in North America. J Am Diet Assoc 2005;105:557–562. 3 Gilbertson HR, Brand-Miller JC, Thorburn AW, et al: The effect of flexible low glycemic index dietary advice versus measured carbohydrate exchange diets on glycemic control in children with type 1 diabetes. Diabetes Care 2001;24:1137–1143. 4 Eeley EA, Stratton IM, Hadden DR, et al: UKPDS 18: estimated dietary intake in type 2 diabetic patients randomly allocated to diet, sulphonylurea or insulin therapy. UK Prospective Diabetes Study Group. Diabet Med 1996;13:656–662.

72

Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 73–81, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

Low Glycemic Index Foods Should Play a Role in Improving Overall Glycemic Control in Type-1 and Type-2 Diabetic Patients and, More Specifically, in Correcting Excessive Postprandial Hyperglycemia Gérard Slama, Fabienne Elgrably, Morvarid Kabir, Salwa Rizkalla Department of Diabetes, Hôtel Dieu Hospital, University René Descartes, Paris V and Assistance Publique, Hôpitaux de Paris, Paris, France

Abstract There is a large bulk of evidence that using low glycemic index (GI) foods has a very significant impact on the amelioration of metabolic disturbances observed in diabetic and/or hyperlipidemic patients and in subjects affected by the metabolic syndrome. Studies bringing convincing evidence against this concept are very rare if any. Improvement is observed not only in postprandial blood glucose and insulin variations but also in circulating plasma lipid levels and the morphology and function of adipocytes. Using the concept of low GI foods in diet counseling of diabetic patients is not exclusive of other measures to improve postprandial and overall blood glucose control. On the contrary, the use of low GI foods should be considered as one of other means and tools available to improve diabetes control (such as other dietary modifications, use of specific and nonspecific drug therapy altering postprandial blood glucose). Among these therapies, the most promising ones are ␣-glucosidase inhibitors, glynides, rapid insulin analogues and in the near future the GLP1 analogue. Again, all these classes of drugs could be associated with one another in order to obtain a postprandial delta excursion target of not below 20 and not above 40–50 mg/dl blood glucose. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

73

Slama/Elgrably/Kabir/Rizkalla Introduction Increasing postprandial plasma glucose and insulin excursions are assumed to increase the severity of diabetes and to be independent predictors of the risk of atherosclerotic diseases and adiposity. Many possible connections have been found between postprandial events and the development of diabetes complications [1]. Lowering postprandial plasma glucose and insulin responses are relevant in preventing and managing diabetes mellitus [2, 3]. Therefore, interventions to reduce postprandial plasma glucose and insulin spikes are one of the essential goals in the therapeutic strategy in diabetic patients and could reduce the risk of developing cardiovascular complications. There is growing recognition that the postprandial glycemic [4, 5] and insulinemic [6] responses to different foods may vary despite equal amounts of total absorbable carbohydrates. This concept is in favor of using low glycemic index (GI) carbohydrates. The notion of GI was proposed more than 20 years ago by Jenkins et al. [5] as a practical way to classify carbohydrate-containing foods according to their effect on postprandial blood glucose rather than according to the mere carbohydrate content. The GI is defined as the incremental area under the blood glucose response curve of a 50-gram carbohydrate portion of a test food expressed as a percent of the response to the same amount of carbohydrate from a standard food (glucose or white bread) taken by the same subject [7]. In practice the actual carbohydrate load from a normal portion varies considerably. It is well known now that both type and amount of carbohydrate influence the glycemic response [8, 9]. In order to address this problem, the concept of glycemic load was introduced. Glycemic load, calculated as the amount of carbohydrate in one serving multiplied by the GI of the food, allows comparisons of the likely glycemic effects of realistic portions of different foods [10]. All the centers worldwide, where the experimental and clinical use of low GI foods has been tested, enthusiastically use diet counseling of diabetic or hyperlipidemic patients and even normal subjects at risk of cardiovascular diseases. A large number of studies has in fact demonstrated the efficiency of diet counseling as regards the use of low GI foods in these patients. More striking effects were noted in the improvement of postprandial blood glucose excursions and, consequently, in glycated hemoglobin, in fasting plasma lipids, particularly triglyceride levels and, marginally, total and LDL cholesterol. We will give below some insights into a few of our clinical and experimental studies on this topic. At this stage, we would like to point out that the use of low GI foods is only part of a more general strategy to improve postprandial blood glucose hyperglycemia and thus overall blood glucose control. 74

Low Glycemic Index Foods in Diabetic Patients Impact of Using Low GI Foods on the Metabolic Disturbances Observed in Certain Categories of Patients Many papers have been published on the use of low GI foods in type-2 diabetic patients and a smaller number in type-1 subjects. Recently, Brandt-Miller et al. [11] looked at 14 studies altogether in a meta-analysis, comprising a total of 356 subjects (203 with type-1 diabetes and 153 with type-2 diabetes). They demonstrated that low GI diets reduced HbA1c by 0.43% points over and above that produced by high GI diets. Our findings are perfectly corroborated by these observations. Low GI diets were also found to have beneficial effects on plasma lipids. In two well-controlled studies in type-2 diabetic subjects [12, 13], there was a reduction of LDL cholesterol and an improvement of the capacity for fibrinolysis (PAI-1) with a low GI diet compared with a high GI diet. Plasma free fatty acids were also found to be lowered by low GI diet [13]. These metabolic effects would be predicted to promote insulin sensitivity and to reduce the risk of CVD. Very significant results have also been observed using low GI foods in hyperlipidemic subjects. Some papers are now published studying the use of low GI diets in subjects affected by the metabolic syndrome with very encouraging results. It should be firmly stressed that none of the clinicians convinced by the practical utility of the above concept are claiming that it is the only way to improve metabolic control. Rather, most of us consider the use of low GI foods as only part of a more general strategy to correct abnormal postprandial hyperglycemia where all the means should be considered as not excluding each other.

Means and Tools to Improve Excessive Postprandial Blood Glucose Excursions The means and tools to improve excessive postprandial blood glucose excursions belong to one of three categories: dietary manipulations, nonspecific and specific drug interventions. Dietary Manipulations The dietary manipulations to improve postprandial blood glucose levels include: (1) global limitations of carbohydrate intake but not below 45% of the daily caloric needs, (2) dividing the carbohydrate-rich foods into different snacks and meals, and (3) limitation of the carbohydrate intake in meals known to be particularly hyperglycemic, like breakfasts, with a shift towards other meals known to be less hyperglycemic, very often lunches and afternoon 75

Slama/Elgrably/Kabir/Rizkalla snacks. In the same line, foods rich in dietary fibers and/or with low GI should be considered. Drugs Not Specifically Affecting Postprandial Blood Glucose Excursions In this category, we find metformin, thiazolidinediones, short- or long-acting sulfonylureas and long-acting insulin. These drugs mainly affect the fasting and interprandial blood glucose levels with a minimal or no effect on the relative elevation of postprandial blood glucose levels compared to the preprandial blood glucose levels; in other words, the above drugs modify the set point of the daily blood glucose curve without significantly affecting the postprandial incremental levels. Drugs Specifically Affecting Postprandial Blood Glucose Excursions The three main therapies used to improve specifically postprandial blood glucose excursions are ␣-glucosidase inhibitors, glinides and bolus preprandial administration of rapid insulin. • ␣-glucosidase inhibitors (Glucor®) are very effective drugs. They are wrongly regarded as drugs difficult to use due to undesirable gastrointestinal side effects. A strategy of a very progressive introduction of low dose tablets with a slowly progressive increase over weeks most of the time permits a very good tolerance profile. We recommend to start with as low as 25 mg once a day for 2 or 3 weeks with an increase step by step, meal by meal, and to stop the increase for one given meal if unbearable flatulence occurs or if the target postprandial delta (blood glucose) excursion is attained, i.e. between 30 and 50 mg/dl. This strategy often leads to something like 100 mg Glucor being prescribed before breakfast, 25 mg before lunch and 50 mg before dinner. • Glinides (Novonorm®, Starlix®) are also often considered as weak oral hyoglycemic agents but seem to us compounds which are very easy to handle and which are perfect as a second-line treatment in addition to ␣glucosidase inhibitors or as first-line therapy. • Bolus preprandial administration of rapid insulin: The most frequent use of preprandial bolus insulin is, nowadays, given using rapid insulin analogues (Humalog®, NovoRapid®, Apidra®). These kinds of insulin are administered either with a syringe, a pen or a pump. Less common ways of administering preprandial insulin bolus are the intranasal or the pulmonary routes of administration. These methods have yet to be fully investigated to become widely used. Other drugs are very promising in specifically modifying postprandial blood glucose levels, particularly GLP-1 and amyline analogues.

76

Low Glycemic Index Foods in Diabetic Patients Personal Clinical and Experimental Data Using the Concept of Low GI Foods Clinical Data Utilization of Low GI Foods in Type-1 and/or Type-2 Diabetic Patients At least three of our studies have addressed this topic: • The first one published in 1988 by Fontvieille et al. [14] was done on a small group of type-1 diabetic patients. This study demonstrated that 3 weeks of a low GI food-enriched diet compared to a high GI diet, in a crossover design, induced a significant improvement in plasma fructoseamine levels in spite of a small but significant decrease in daily insulin needs. • Another study by the same authors [15] was conducted with the same crossover methodology in a mixed type-1 and type-2 population of diabetic subjects over a period of 5 weeks. This study showed a reduction of 11% in fasting glycemia, 13% in fructoseamine and 20% in triglycerides. • Our most recent study by Rizkalla et al. [13] was conducted in a group of 11 type-2 diabetic subjects and demonstrated that a 4-week low GI diet resulted in: (1) an improvement of plasma glucose control, (2) a decrease of whole body glucose utilization (measured by the euglycemic hyperinsulinemic clamp), (3) a decrease of some lipid profiles and (4) a reduction in the capacity for fibrinolysis. In slightly overweight subjects, however, we demonstrated that 5 weeks of low GI foods were able to decrease total fat mass and to improve the plasma lipid profile rather than to improve plasma glucose control [16].

Experimental Data Our first experimental work in this field consisted of designing a method to test the concept in animals (i.e. in rats: to validate the GI concept in acute conditions). We, thus, conditioned the rats to consume any presented meal within the next 5 min if they wanted to eat after an overnight fast. Different foods were then tested, i.e. glucose powder, different foods rich in amylose or resistant starch (like mung beans), and different types of corn starch rich in amylopectin (like waxy corn starch). We demonstrated that the classification of starchy foods utilized in human nutrition was also done experimentally [17]. Consequently, we aimed to evaluate the chronic consequences of consuming low and high GI foods in rats. We used diabetic (STZ-n2: low streptozotocin dose injected on the 2nd day after birth) and nondiabetic male SpragueDawley rats. They were fed either a low or a high GI diet for 3 and 12 weeks. At the end of the nutritional periods, the activity and gene expression of different proteins, in different tissues, implicated in glucose and lipid metabolism were evaluated. Four of our studies all produced the same type of results: each study gave a more mechanistic explanation than the preceding one. 77

Slama/Elgrably/Kabir/Rizkalla Glucose metabolism was improved in diabetic and nondiabetic rats; the postprandial insulin profiles were lower with low GI foods. Circulating lipid levels: The most reproducible result observed with low GI was a significant reduction of plasma cholesterol and triglyceride levels. Size of adipocytes, metabolism and gene expression: Again, the most reproducible result observed in all types of experimental animals was a reduction in the diameter of the adipocytes in rats fed a low GI diet compared to those fed a high GI diet. These adipocytes during the low GI period were more metabolically active with an ex vivo increase in insulinstimulated 14C-glucose transport and oxidation but a decrease in lipogenesis [18]. A decrease in the activity and gene expression of lipogenic enzymes was also found [19]. Leptin levels as well as its gene expression in epididymal adipose tissue were found to be decreased just before the increase in adiposity in the high GI fed rats [20].

• • •

Conclusion (1)

(2)

(3)

There is a large bulk of evidence that using low GI foods has a greatly significant impact on the amelioration of the metabolic disturbances observed in diabetic and/or hyperlipidemic patients and in subjects affected by the metabolic syndrome. Studies showing convincing evidence against this concept are very rare if any. Improvement is observed not only of postprandial blood glucose and insulin variations but also of circulating plasma lipid levels and of the morphology and function of adipocytes. Using the concept of low GI foods in the diet counseling of diabetic patients does not exclude other measures to improve postprandial and overall blood glucose control. On the contrary, the use of low GI foods should be considered as one among other means and tools available to improve diabetes control (such as other dietary modifications, and use of specific and nonspecific drug therapies altering postprandial blood glucose). Among these therapies, the most promising ones are ␣glucosidase inhibitors, glinides, rapid insulin analogues and in the near future GLP-1 analogues. Again, all these classes of drugs could be associated with one another in order to obtain a postprandial delta excursion target not below 20 and not above 40–50 mg/dl blood glucose.

References 1 Ceriello A: The possible role of postprandial hyperglycaemia in the pathogenesis of diabetic complications. Diabetologia 2003;46(suppl 1):M9–M16. 2 Liu S, Willet W, Stampfer M, et al: Prospective study of dietary glycemic load, carbohydrate intake, and risk of coronary heart disease in US women. Am J Clin Nutr 2000;71:1455–1461.

78

Low Glycemic Index Foods in Diabetic Patients 3 Bonora E, Muggeo M: Postprandial blood glucose as a risk factor for cardiovascular disease in type II diabetes: the epidemiological evidence. Diabetologia 2001;44:2107–2114. 4 Crapo A, Insel J, Sperling M, Kolterman G: Comparison of serum glucose, insulin and glucagon responses to different types of complex carbohydrate in noninsulin-dependent diabetic patients. Am J Clin Nutr 1981;34:184–190. 5 Jenkins D, Wolever T, Taylor H, et al: Glycemic index of foods: a physiological basis for carbohydrate exchange. Am J Clin Nutr 1981;34:362–366. 6 Bornet FRJ, Costagliola D, Rizkalla SW, et al: Insulinemic and glycemic indexes of six starch-rich foods taken alone and in a mixed meal by type 2 diabetes. Am J Clin Nutr 1987;45:588–595. 7 FAO/WHO: Carbohydrates in human nutrition. Report of a joint FAO/World Health Organization Expert Consultation. FAO Food and Nutrition, 1998, paper 66. 8 Wolever T, Mehling C: Long-term effect of varying the source or amount of dietary carbohydrate on postprandial plasma glucose, insulin, triacylglycerol, and free fatty acid concentrations in subjects with impaired glucose tolerance. Am J Clin Nutr 2003;77:612–621. 9 Barclay A, Brand-Miller J, Wolever T: Glycemic index, glycemic load, and glycemic response are not the same. Diabetes 2005;28:1839–1840. 10 Schulze M, Liu S, Rimm E, et al: Glycemic index, glycemic load, and dietary fiber intake and incidence of type 2 diabetes in younger and middle-aged women. Am J Clin Nutr 2004;80:348–356. 11 Brand-Miller J, Hayne S, Petocz P, Colagiuri S: Low-glycemic index diets in the management of diabetes: a meta-analysis of randomized controlled trials. Diabetes Care 2003;26:2261–2267. 12 Jarvi A, Karlstrom B, Granfeldt Y, et al: Improved glycemic control and lipid profile and normalized fibrinolytic activity on a low-glycemic index diet in type 2 diabetic patients. Diabetes Care 1999;22:10–18. 13 Rizkalla S, Taghrid L, Laromiguiere M, et al: Improved plasma glucose control, whole-body glucose utilization, and lipid profile on a low-glycemic index diet in type 2 diabetic men: a randomized controlled trial. Diabetes Care 2004;27:1866–1872. 14 Fontvieille A, Acosta M, Rizkalla S, et al: A moderate switch from high to low glycemic index foods for 3 weeks improves the metabolic control of type 1 (IDDM) diabetic subjects. Diabetes Nutr Metab 1988;1:139–143. 15 Fontvieille A, Rizkalla S, Penfornis A, et al: The use of low glycemic index foods improves metabolic control of diabetic patients over five weeks. Diabet Med 1992;9:1–7. 16 Bouché C, Rizkalla S, Luo J, et al: Five week, low-glycemic index diet decreases total fat mass and improves plasma lipid profile in moderately overweight nondiabetic subjects. Diabetes Care 2002;25:822–828. 17 Lerer-Metzger M, Rizkalla S, Luo J, et al: Effect of long-term low-glycemic index starchy food on plasma glucose and lipid concentrations and adipose tissue cellularity in normal and diabetic rats. Br J Nutr 1996;75:723–732. 18 Kabir M, Rizkalla S, Champ M, et al: Dietary amylose-amylopectin starch content affects glucose and lipid metabolism in adipocytes of normal and diabetic rats. J Nutr 1998;128: 35–43. 19 Kabir M, Rizkalla S, Quignard-Boulangé A, et al: A high glycemic index starch diet affects lipid storage-related enzymes in normal and to a lesser extent in diabetic rats. J Nutr 1998;128: 1878–1883. 20 Kabir M, Guerre-Millo M, Laromigiere M, et al: Negative regulation of leptin by chronic high glycemic index starch diet. Metabolism 2000;49:764–769.

Discussion Dr. Katsilambros: Congratulations, I think that especially the study on adipocytes is very important. I would like to ask you if the low or high index diets in the rat were isoenergetic or were the rats fed ad libitum. Dr. Slama: They were fed ad libitum. Dr. Katsilambros: That is the difference then. Dr. Slama: Yes, and the amount of carbohydrate was higher, 70% than the regular diet in the normal rats.

79

Slama/Elgrably/Kabir/Rizkalla Dr. Katsilambros: It can be hypothesized that those rats on a low carbohydrate diet finally ate more calories than those on a high carbohydrate diet. In this sense this might explain the difference in your findings and the findings which I presented in the previous talk. This is 100% confirmed, it was in the past, it is in the present, and it will be in the future. Dr. Schiffrin: Regarding the changes that you described in adipose tissue, was there more or less macrophage infiltration following the diet? Dr. Slama: The macrophage story in adipose tissue is very recent, at least to my knowledge. When we did the study we didn’t look at that. Perhaps in the future we will dissect, but for adipose tissue volume, adipose activity, tissue activity, it is certain, the adipocytes; but the macrophage I don’t know. Dr. Chiasson: I was just wondering are you interpreting the decrease in the size of the adipocyte due to a decrease in the insulin level? Dr. Slama: No, because it has also been observed in insulinopenic rats at least, but I don’t know why it is so. Perhaps the disposition of the free fatty acids plays a major role. The free fatty levels are really much less after the low glycemic index diet. I think that you know better than I the role which is supposed to be played by free fatty acids. We, as doctors and also as physiologists, have been used to studying life at a steady state, at a basal level, but probably the most important aspect of nutrition is postprandial. It seems obvious that nutrition does things also postprandially, and the hypothesis of Ceriello [1] and others on oxidative stress and many other aspects should also play a role. I don’t know exactly what the major determinant of the phenomenon is, but in one study on animals it has been observed every time, and this is probably the most striking effect of a low glycemic index, the effect on adipocytes. Dr. Chiasson: In the postprandial profile between carbohydrate alone or within a mixed meal, you showed that the mixed meal had a lower rise in postprandial plasma glucose. Do you think that this is due only to the difference in the insulin profile? Dr. Slama: It could be due to dilution of the calories; it could be due more certainly to increased fat intake, and perhaps also to protein intake and insulin secretion. So it is probably a mixed phenomenon which I can’t fully explain, but the fact is here and it is well documented in the literature. Dr. Chiasson: What is well documented? Are you saying that it modifies the absorption? Dr. Slama: It is well demonstrated that when you eat a mixed meal, you obtain a lower glucose increase for the same amount of carbohydrate. Dr. Chiasson: Yes, but usually that is mainly due to the disposition of glucose rather than to the absorption. Dr. Slama: No idea; I have not seen that. If you say so I will look at the results, but I have not measured that. It is said that fat decreases stomach emptying and it plays a role, but perhaps it is also absorbed in the gut. Dr. Chiasson: We have looked at absorption, labeling the carbohydrate, and whether it was given alone or within a mixed meal. We had exactly the same absorption profile, suggesting that the differences would be more in the disposition of the glucose absorbed and the insulin response. If you have fat in your meal then you would expect that you are creating insulin resistance and it will go higher up and so that could explain the lower profile within the mixed meal. Dr. Halimi: I have a comment regarding the possible limits of the glycemic index which is based on the measure of peripheral glycemia and insulinemia, i.e. posthepatic glycemia and post-hepatic insulinemia. However, we know that the liver plays a major role in normal people, which is quite different in the metabolic syndrome, in steatosis and in type-2 diabetes. Yet the glycemic index cannot take into account the respective roles of the different tissues (gut, liver, peripheral) involved in this complex process.

80

Low Glycemic Index Foods in Diabetic Patients Dr. Slama: There are two things. First, in the first part of this morning’s session the question was raised on peak value, raising time and negative part, etc. I think that the postprandial blood glucose excursion should be seen as a disturbance in a very finely tuned system, and the postprandial excursion is perhaps an overflow of the capability of the body to really maintain blood glucose around 1 g. It is an overflow. So the less brutal the disturbance is, the easier it will be for the body to maintain. To sustain what I am saying I refer to work we have done and published in a confidential review of nutrition in small laboratory pigs with Lang et al. [2, 3]. I don’t remember all the details but these pigs had a catheter in the carotid artery and another in the portal vein. These pigs were put on glucose clamp and they had a tube in the stomach to really know what we were addressing. With the low glycemic index, the blood glucose excursion was lower and all the glucose was absorbed, as seen by the glucose clamp but not by blood glucose excursion. It is also the same on the insulin level in the portal vein but not in the peripheral vein. Dr. Jianquin Sun: I have a question regarding some of the inconsistent results from different clinical trials in terms of glycemic control or lipid profile. Could you explain what is the reason for that result? Dr. Slama: I think that we can also discuss paper by paper but I have not read all the papers so I cannot really answer your question. As I already said there are low glycemic index foods and low glycemic index foods. For example, as we already said, pizza and chocolate bars have very low glycemic index but in my opinion they are not to be recommended as daily foods. So if in one study they use such foods and in others they do not, of course the results will be totally different. So we have to look very carefully at every methodological aspect: the way the glycemic index of foods was calculated, the way the food was administered, and so on. So I cannot give a general answer to your important question. Because of course these critics are addressed to the glycemic index, not reproducibility, variability. For me this is not really convincing. Why should the glycemic food, starchy food, give exactly the same value? But day after day it makes a difference, and in all the studies I have seen and done myself, I have never seen a discrepancy.

References 1 Ceriello A: Postprandial hyperglycemia and diabetes complications: is it time to treat? Diabetes 2005;54:1–7. 2 Lang V, Vaugelade P, Bernard F, et al: Euglycemic hyperinsulinemic clamp to assess posthepatic glucose appearance after carbohydrate loading. 1. Validation in pigs. Am J Clin Nutr 1999;69:1174–1182. 3 Lang V, Bornet FR, Vaugelade P, et al: Euglycemic hyperinsulinemic clamp to assess posthepatic glucose appearance after carbohydrate loading. 2. Evaluation of corn and mung bean starches in healthy men. Am J Clin Nutr 1999;69:1183–1188.

81

Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 83–95, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

Is Fructose the Optimal Low Glycemic Index Sweetener? John P. Bantle Division of Endocrinology and Diabetes, Department of Medicine, University of Minnesota, MN, USA

Abstract Fructose is a monosaccharide which is abundant in nature. It is the sweetest naturally occurring carbohydrate. The availability of fructose increased substantially when it became possible in the 1960s to economically produce high fructose syrups from corn starch and other starches. Such high fructose syrups are now used to sweeten soft drinks, fruit drinks, baked goods, jams, syrups and candies. The most recent data available suggest that fructose consumption is increasing worldwide. Fructose presently accounts for about 10% of average total energy intake in the United States. Studies in both healthy and diabetic subjects demonstrated that fructose produced a smaller postprandial rise in plasma glucose and serum insulin than other common carbohydrates. Substitution of dietary fructose for other carbohydrates produced a 13% reduction in mean plasma glucose in a study of type-1 and type-2 diabetic subjects. However, there is concern that fructose may aggravate lipemia, particularly in men. In one study, daylong plasma triglycerides (estimated by determining the area under response curves) in healthy men was 32% greater during a high fructose diet than during a high glucose diet. There is also concern that fructose may be a factor contributing to the growing worldwide prevalence of obesity. Increasing fructose consumption is temporally associated with the increase in obesity. Moreover, on theoretical grounds, dietary fructose might increase energy intake. Fructose stimulates insulin secretion less than does glucose and glucose-containing carbohydrates. Since insulin increases leptin release, lower circulating insulin and leptin after fructose ingestion might inhibit appetite less than consumption of other carbohydrates and lead to increased energy intake. However, there is not yet any convincing experimental evidence that dietary fructose does increase energy intake. Although evidence that fructose has adverse effects is limited, adding fructose in large amounts to the diet may be undesirable, particularly for men. Fructose that occurs naturally in fruits and vegetables is a modest component of energy intake and should not be of concern. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

83

Bantle Fructose is a six-carbon monosaccharide which is abundant in nature. Free fructose is present in honey, dates, figs, apples, grapes and most berries. An additional, important quantitative natural source of fructose is the disaccharide sucrose which is composed of equimolar quantities of fructose and glucose. When ingested by humans, fructose is absorbed by an active transport system but at a slower rate than is glucose [1]. Co-ingestion of glucose increases intestinal absorptive capacity for fructose. In the absence of glucose, human intestinal capacity to absorb fructose appears to be quite variable with some people unable to completely absorb 30- to 40-gram quantities [1]. Those individuals unable to completely absorb ingested fructose are at risk for diarrhea and other gastrointestinal side effects. The first several steps in fructose and glucose metabolism differ significantly. Fructose stimulates only modest insulin secretion and does not require the presence of insulin to enter cells [2]. Avidly taken up by hepatic cells, fructose is rapidly converted to fructose-1-phosphate and bypasses the early rate-limiting steps of glucose metabolism. Fructose-1-phosphate is mainly converted to lactate, glucose and glycogen [3]. Gluconeogenesis from fructose is increased by starvation and poorly controlled diabetes. Fructose may also form acetyl CoA which is used in fatty acid synthesis. Enhanced activity of lipogenic enzymes with chronic fructose feeding may promote hepatic triglyceride production and output of VLDL particles. Presumably, energy intake must be excessive for fructose to stimulate lipogenesis. Fructose is the sweetest tasting naturally occurring carbohydrate (table 1). Advances in technology in the 1960s made possible the production of inexpensive high fructose syrups from corn starch [4]. Corn is an abundant worldwide source of starch. To make high fructose syrups, corn starch is first separated from other corn by-products by wet milling. Next the starch is digested with mineral acid and amylase to form glucose. The enzyme glucose isomerase is then used to convert glucose to fructose. A syrup containing 42% fructose is the first product of this process. Through chromatographic enrichment, 55 and 90% high fructose syrups can then be produced. The 55% high fructose syrup has taste and sweetness equivalent to sucrose. Because of sweetness and low cost, high fructose syrups found commercial application. In the mid 1980s, 55% high fructose syrup was adopted by the carbonated-beverage industry and became the predominant sweetener in soft drinks. The United States has from the beginning been the world’s largest producer of high fructose corn syrups but Japan, Canada, South Korea, China, Argentina, and other countries are also significant producers [5]. In Asia, tapioca starch and broken rice are used in production of high fructose syrups. High fructose syrups are widely used in soft drinks, fruit drinks, baked goods, jams, syrups and candies. In 1977–1978, average fructose intake was estimated to be 37 g per day, accounting for ⬃8% of total energy intake in the 84

Fructose: Optimal Sweetener? Table 1. Sweetness relative to sucrose [4] Substance

Relative sweetness

Fructose High fructose syrup – 90% Sucrose High fructose syrup – 55% High fructose syrup – 42% Glucose

117 106 100 99 92 67

United States [6]. In 1987–1988, fructose intake had increased to 39 g per day, accounting for ⬃9% of energy intake [7] and in 1988–1994, it had further increased to 55 g per day, accounting for ⬃10% of energy intake [8]. Approximately one third of fructose came from fruits, vegetables and other natural sources and two thirds were added to beverages and foods in the diet. A similar trend toward substantial caloric sweetener and fructose consumption is occurring worldwide [9]. There has long been interest in the metabolic effects of fructose, particularly in people with diabetes. Studies in diabetic subjects done in the 1970s and 1980s demonstrated that fructose-containing test meals produced smaller postprandial increases in plasma glucose than test meals containing isocaloric amounts of sucrose, glucose and starch [10–12]. Jenkins et al. [13] greatly expanded our knowledge about the differences in response to dietary carbohydrates with the development of the glycemic index of foods. Glycemic index was defined as the increase in plasma glucose area from 0 to 120 min after ingestion of 50 g of available carbohydrate from a test food compared to 50 g of carbohydrate from a reference food such as glucose. The glycemic indices of carbohydrate-containing foods vary substantially, with fructose having a particularly low glycemic index (table 2). In an effort to further evaluate the potential for fructose to lower postprandial plasma glucose, we developed five test meals containing different carbohydrates and fed the meals to healthy and diabetic volunteers [14]. The meals contained nearly identical amounts of carbohydrate, protein and fat but a different test carbohydrate which accounted for 24 or 25% of total calories. The test carbohydrates were glucose, fructose, sucrose, potato starch and wheat starch. Plasma glucose and serum insulin were measured before and at intervals for 240 min after the meals. In healthy volunteers, type-1 diabetic volunteers and type-2 diabetic volunteers, the fructose meal produced the smallest postprandial increment in plasma glucose and the smallest increment in postprandial glucose area (fig. 1). The fructose meals also produced the smallest increment in serum insulin in healthy and type-2 diabetic volunteers but the differences among meals were not significant. 85

Bantle Table 2. Glycemic indices of selected carbohydrate-containing foods [from 13] Food

Glycemic index

Food

Glycemic index

Instant potato White rice White bread Frozen peas Sweet corn Carrots Lentils Kidney beans

80 72 69 51 59 92 29 29

Banana Apple Orange juice Glucose Sucrose Fructose Skim milk Ice cream

62 39 46 100 59 20 32 36

Reference food was 50 g glucose.

Area increments in plasma glucose (min·mg/dl)

30,000

25,000

20,000

15,000

10,000

5,000

0

F S PWG

F S PWG

F S PWG

Healthy subjects

Type-1 diabetics

Type-2 diabetics

Fig. 1. Area increments in plasma glucose (mean ⫾ SEM) after test meals indicated as follows: fructose (F), sucrose (S), potato (P), wheat (W) and glucose (G) [from 14].

Since fructose has an agreeable taste similar to that of sucrose and since fructose produces a smaller postprandial rise in plasma glucose than other common carbohydrates, fructose seemed to be an excellent candidate for a sweetening agent in the diabetic diet. To test this possibility, we studied 12 type-1 and 12 type-2 diabetic subjects who were fed three isocaloric diets for 8 days each using a randomized, crossover design [15]. The three diets provided, 86

Fructose: Optimal Sweetener? respectively, 21% of energy as fructose, 23% of energy as sucrose, and almost all carbohydrate energy as starch with less than 5% of energy derived from fructose and sucrose. All meals were prepared in a metabolic kitchen and provided to subjects. The fructose diet resulted in significantly lower 1- and 2-hour postprandial plasma glucose levels, overall mean plasma glucose, and urinary glucose excretion than did the starch diet. The reductions in mean plasma glucose with the fructose diet were 24% in type-1 diabetic subjects and 7% in type-2 diabetic subjects. Of note, the fructose diet increased postprandial lactate. There were no differences between the sucrose and starch diets in any of the measure of glycemic control in either subject group. It next seemed important to extend the period of dietary intervention with fructose to see if beneficial effects on glycemia persisted and to look for potential adverse effects. Accordingly, we compared isocaloric high fructose (20% of energy derived from fructose) and high starch diets (less than 3% of energy derived from fructose) in 6 type-1 and 12 type-2 diabetic subjects using a crossover design [16]. Both study diets were composed of common foods. All meals were prepared in a metabolic kitchen and provided to subjects for 28 days. The diets were well received by all subjects. Mean plasma glucose, urine glucose and serum glycosylated albumin were all lower during the fructose diet than during the starch diet. On day 28 of the fructose diet, mean plasma glucose was 13% lower than on day 28 of the starch diet. However, of concern, fasting serum LDL cholesterol on day 28 of the fructose diet was 11% higher than the corresponding value for LDL cholesterol on day 28 of the starch diet. Thus, a diet in which fructose was substituted for other carbohydrates was pleasant to eat and resulted in reduced glycemia in people with diabetes but appeared to have an adverse effect on serum LDL cholesterol. This raised concern about fructose as a sweetening agent in the diabetic diet. This finding also raised concern about the potential effects of dietary fructose in the general population since, in the United States and many other countries, fructose is a significant source of dietary energy [9]. Several studies did not find adverse effects of dietary fructose on serum lipids in healthy subjects [17–19]. However, these studies either compared fructose to sucrose or were outpatient studies that did not provide meals to subjects. Since sucrose is composed of 50% fructose, it is not an optimal reference. Moreover, rigorous control of nutrient intake requires the provision of meals. Thus, these studies may not be reliable in assessing the effects of dietary fructose on serum lipids. Two studies which compared a high fructose diet to a diet nearly devoid of fructose and established rigorous control of nutrient intake by providing all food to subjects both reported adverse effects of fructose on serum lipids [20, 21]. Hallfrisch et al. [20] reported that high fructose diets consumed for 5 weeks increased fasting plasma LDL cholesterol in healthy and hyperinsulinemic men and increased fasting plasma triglycerides in hyperinsulinemic men. Reiser et al. [21] found that a high fructose diet consumed for 5 weeks increased fasting plasma LDL cholesterol in healthy men and fasting plasma 87

Bantle triglycerides in both healthy and hyperinsulinemic men. These two well-done studies suggested that dietary fructose does adversely affect serum lipids, at least in men. Women were not included in either study. In an effort to gain additional insight into the effects of fructose on plasma lipids, we compared high and low fructose diets in 24 healthy volunteers (12 men and 12 women; 6 of each gender age ⬍40 years and 6 of each gender age ⬎40 years) [22]. All subjects consumed two isocaloric diets for 6 weeks. One diet provided 17% of energy as fructose. The other diet was sweetened with glucose and was nearly devoid of fructose. Diet order was assigned randomly using a balanced, crossover design. Both diets were composed of common foods and contained nearly identical amounts of carbohydrate, protein, fat, fiber, cholesterol and saturated, monounsaturated and polyunsaturated fatty acids. All meals were prepared in the metabolic kitchen of the University of Minnesota General Clinical Research Center. The fructose diet resulted in higher fasting total and LDL plasma cholesterol at day 28 but this effect did not persist at day 42 (table 3). The plasma triglyceride responses to the diets differed by gender. The fructose diet had no significant effect on fasting or postprandial plasma triglycerides in women (table 3, fig. 2). However, in men, the fructose diet produced significantly higher fasting and postprandial plasma triglycerides. This effect persisted through day 42. On day 42 of the fructose diet, daylong plasma triglycerides (estimated by determining the area under the response curves) in men was 32% greater than during the glucose diet. We concluded that diets high in added fructose may be undesirable, particularly for men. Glucose may be a suitable replacement sugar. Another potential concern about fructose is its association with increased energy intake and obesity. Worldwide trends in per capita consumption of caloric sweeteners (of which high fructose syrups are a major component) demonstrated an increase from 232 kcal/day in the year 1962 to 306 kcal/day in the year 2000 [9]. In the United States, caloric sweeteners accounted for 16% of energy intake in the year 1996 [9]. About 43% of the caloric sweeteners came from soft drinks and fruit drinks. Several authors have suggested that dietary fructose may play a role in the worldwide increase in obesity prevalence [23, 24]. Their reasons for implicating fructose are principally two. The first is the association, mentioned above, between increasing consumption of fructose and increasing obesity. The second is the theoretical possibility that dietary fructose increases energy intake. Clearly dietary fructose stimulates insulin secretion less than glucose and glucosecontaining carbohydrates. Insulin stimulates leptin release from adipocytes [25] and circulating insulin and leptin concentrations were thus lower after ingestion of fructose-containing meals than after ingestion of glucose-containing meals in healthy women [26]. However, energy intake by the women was not greater during the fructose-containing meals. Nevertheless, lower circulating insulin and leptin after fructose consumption might inhibit appetite less than consumption of other carbohydrates and lead to an increase in food intake. 88

Fructose: Optimal Sweetener? Table 3. Effects of the two study diets on mean fasting plasma lipids [22] Day

Plasma cholesterol, mmol/l Fructose diet Glucose diet pa Plasma LDL cholesterol, mmol/l Fructose diet Glucose diet pa Plasma HDL cholesterol, mmol/l Fructose diet Glucose diet pa Plasma triglycerides, mmol/l Women Fructose diet Glucose diet pa Men Fructose diet Glucose diet pa

14

28

42

4.53 4.43 0.154

4.61 4.30 ⬍0.001

4.30 4.22 0.169

2.67 2.59 0.256

2.69 2.49 ⬍0.001

2.49 2.49 0.756

1.35 1.40 0.077

1.37 1.37 0.897

1.30 1.30 0.965

0.97 0.88 0.298

1.02 0.99 0.810

0.93 0.97 0.631

1.32 1.12 0.018

1.30 1.03 0.001

1.25 0.95 ⬍0.001

The means for each endpoint have a common SE based on the appropriate repeated-measures ANOVA error term; to convert cholesterol to mg/dl, multiply by 38.6; to convert triglycerides to mg/dl, multiply by 88.5. aBecause 6 paired comparisons of this endpoint were made (all data not shown), only p ⬍ 0.008 (0.05/6) should be considered significant at the 0.05 level.

Consistent with the idea that dietary fructose might increase energy intake are data from Ludwig et al. [27] which demonstrated an association between consumption of sugar-sweetened drinks and obesity in children. Further evidence is provided by Raben et al. [28] who fed overweight subjects supplements of either sucrose or artificial sweeteners for 10 weeks. The subjects who consumed sucrose demonstrated increases in energy intake, body weight, fat mass and blood pressure. However, it is important to point out that subjects in the sucrose group were ‘instructed’ to consume 2 g sucrose/kg body weight daily (⬃23% of energy intake) and were provided with the necessary sucrose-sweetened beverages and foods to do so. Thus, the increased sucrose intake was not spontaneous. These two studies were the main evidence cited by the World Health Organization when implicating sugars as a cause of obesity and to justify their recommendation that free sugar consumption be less than 10% of total daily energy intake [29]. 89

1.5

b

*

1.0

00 5:

:0

2:

0

00

0.5 7:

00

00

Clock time (h), day 42

5:

0

2:

:0 23

9:

7:

a

3 11 0 :3 0 14 :0 16 0 :0 0 19 :0 0

0.5

*

23

1.0

* *

3 11 0 :3 0 14 :0 16 0 :0 0 19 :0 0

1.5

* 2.0

9:

2.0

2.5

30

Mean plasma triacylglycerol (mmol/l)

2.5

30

Mean plasma triacylglycerol (mmol/l)

Bantle

Clock time (h), day 42

a) and Fig. 2. Mean plasma triacylglycerol (triglyceride) concentrations in women (a b) during the 24-hour metabolic profiles on day 42 of the fructose ( men (b ) and glucose (------) diets. Significant difference (*) between the two points is shown, p ⬍ 0.006 (0.05/9, Bonferroni adjustment for multiple comparisons). To convert triglycerides to mg/dl, multiply by 88.5 [from 22].

Although increasing fructose consumption is temporally associated with the increasing worldwide prevalence of obesity, there is little or no evidence proving cause and effect. In the US, increasing energy intake was associated with increased restaurant and fast-food meals and increased consumption of salty snacks and pizza as well as soft drinks [30]. Decreased physical activity is also almost certainly a factor in the increasing prevalence of obesity. To demonstrate that dietary fructose is important in causing obesity, it would be necessary to conduct a clinical trial demonstrating that fructose caused a spontaneous increase in energy intake. Given fructose’s availability, low cost and pleasant taste, such a clinical trial might provide important new information. In summary, fructose is a naturally occurring sugar with a pleasant taste. It produces a smaller postprandial rise in plasma glucose than other common carbohydrates and thus may be a useful sweetening agent in the diabetic diet. However, dietary fructose appears to have adverse effects on plasma lipids. Moreover, there is concern that dietary fructose may stimulate energy intake and promote weight gain and obesity. Thus, adding large amounts of fructose to the diet may be undesirable. Nevertheless, concern about fructose should not extend to the naturally occurring fructose in fruits and vegetables. These are healthy foods which provide only a modest amount of fructose in most diets.

References 1 Riby JE, Fujisawa T, Kretchmer N: Fructose absorption. Am J Clin Nutr 1993;58:748S–753S. 2 Henry RR, Crapo PA: Current issues in fructose metabolism. Annu Rev Nutr 1991;11:21–39.

90

Fructose: Optimal Sweetener? 3 Mayes PA: Intermediary metabolism of fructose. Am J Clin Nutr 1993;58:744S–765S. 4 Hanover LM, White JS: Manufacturing, composition and applications of fructose. Am J Clin Nutr 1993;58:724S–732S. 5 Vuilleumier S: Worldwide production of high-fructose syrup and crystalline fructose. Am J Clin Nutr 1993;58:733S–736S. 6 Park YK, Yetley EA: Intake and food sources of fructose in the United States. Am J Clin Nutr 1993;58:737S–747S. 7 Sigman-Grant M, Keast DR: Addendum to Am J Clin Nutr 1995;62(suppl):178S–194S. Am J Clin Nutr 1997;65:1572–1574. 8 Bialostosky K, et al: Dietary intake of macronutrients, micronutrients and other dietary constituents: United States 1988–94. National Center for Health Statistics. Vital Health Stat 2002;11(245):1–156. 9 Popkin BM, Nielsen SJ: The sweetening of the world’s diet. Obes Res 2003;11:1325–1332. 10 Akerblom HK, Siltanen I, Kallio AK: Does dietary fructose affect the control of diabetes in children? Acta Med Scand Suppl 1972;548:195–202. 11 Crapo PA, Kolterman OG, Olefsky JM: Effects of oral fructose in normal, diabetic, and impaired glucose tolerance subjects. Diabetes Care 1980;3:575–581. 12 Akgun S, Ertel NH: A comparison of carbohydrate metabolism after sucrose, sorbitol, and fructose meals in normal and diabetic subjects. Diabetes Care 1980;3:582–585. 13 Jenkins DJA, Wolever TMS, Tayler RH, et al: Glycemic index of foods: a physiologic basis for carbohydrate exchange. Am J Clin Nutr 1981;34:362–366. 14 Bantle JP, Laine DC, Castle GW, et al: Postprandial glucose and insulin responses to meals containing different carbohydrates in normal and diabetic subjects. N Engl J Med 1983;309:7–12. 15 Bantle JP, Laine DC, Thomas W: Metabolic effects of dietary fructose and sucrose in types 1 and 2 diabetic subjects. JAMA 1986;256:3241–3246. 16 Bantle JP, Swanson JE, Thomas W, Laine DC: Metabolic effects of dietary fructose in diabetic subjects. Diabetes Care 1992;15:1468–1476. 17 Crapo PA, Kolterman OG: The metabolic effects of 2-wk fructose feeding in normal subjects. Am J Clin Nutr 1984;39:525–534. 18 Bossetti BM, Kocher LM, Moranz JF, Falko JM: The effects of physiologic amounts of simple sugars on lipoprotein, glucose and insulin levels in normal subjects. Diabetes Care 1984;7: 309–312. 19 Koh ET, Ard NF, Mendoza F: Effects of fructose feeding on blood parameters and blood pressure in impaired glucose-tolerant subjects. J Am Diet Assoc 1988;88:932–938. 20 Hallfrisch J, Reiser S, Prather ES: Blood lipid distribution of hyperinsulinemic men consuming three levels of fructose. Am J Clin Nutr 1983;37:740–748. 21 Reiser S, Powell AS, Scholfield DJ, et al: Blood lipids, lipoproteins, apoproteins, and uric acid in men fed diets containing fructose or high-amylase cornstarch. Am J Clin Nutr 1989;49: 832–839. 22 Bantle JP, Raatz SK, Thomas W, Georgopoulos A: Effects of dietary fructose on plasma lipids in healthy subjects. Am J Clin Nutr 2000;72:1128–1134. 23 Elliott SS, Keim NL, Stern JS, et al: Fructose, weight gain and the insulin resistance syndrome. Am J Clin Nutr 2002;76:911–922. 24 Bray GA, Nielsen SJ, Popkin BM: Consumption of high-fructose corn syrup in beverages may play a role in the epidemic of obesity. Am J Clin Nutr 2004;79:537–543. 25 Saad MF, Khan A, Sharma A, et al: Physiological insulinemia acutely modulates plasma leptin. Diabetes 1998;47:544–549. 26 Teff KL, Elliott SS, Tschop M, et al: Dietary fructose reduces circulating insulin and leptin, attenuates postprandial suppression of ghrelin, and increases triglycerides in women. J Clin Endocrinol Metab 2004;89:2963–2972. 27 Ludwig DS, Peterson KE, Gortmaker SL: Relation between consumption of sugar-sweetened drinks and childhood obesity: a prospective, observational analysis. Lancet 2001;357:505–508. 28 Raben A, Vasilaras TH, Moller AC, Astrup A: Sucrose compared with artificial sweeteners: different effects on ad libitum food intake and body weight after 10 wk of supplementation in overweight subjects. Am J Clin Nutr 2002;76:721–729. 29 Waxman A: The WHO global strategy on diet, physical activity and health: the controversy on sugar. Development 2004;47:75–82. 30 Nielsen SJ, Siega-Riz AM, Popkin BM: Trends in energy intake in U.S. between 1977 and 1996: similar shifts seen across age groups. Obes Res 2002;10:370–378.

91

Bantle Discussion Dr. Katsilambros: A long time ago I learned personally from Prof. Alexander Margot that when patients come in at the beginning in a state of uncontrolled diabetes, it is quite undesirable and contraindicated to provide fructose because it is much more readily converted to glucose in uncontrolled diabetes. Is this also your opinion? The second comment refers to personal studies done many years ago when I was an assistant. Patients with hepatitis B were given fructose orally, once in the acute stage and again in the recovery stage. Glucose and fructose were measured separately. In the first acute phase there was fructose intolerance; the curve was high and delayed. Then in the recovery phase a considerable improvement in this curve was observed. This means that fructose intolerance was due to the diminished capacity of the liver to transform fructose to glucose. The related question now refers to your observation that men and women have large differences with regard to the triglyceride increase; that in men triglycerides were much higher than in women. Are you sure that these men were not drinkers? Dr. Bantle: Let me answer the first question which is about the effect of fructose in poorly controlled diabetes. I think you are absolutely correct. If fructose is fed to someone with poorly controlled diabetes, there is a greater rise in plasma glucose than in people who have well-controlled diabetes. But I think if you look at the response of such a person to ingested glucose, it is also greatly increased and in fact the glycemic response to fructose is somewhat less than glycemic response to glucose in that situation. It is true, however, that more fructose is converted to glucose in poorly controlled diabetes. Your second point about the use of fructose as a test for liver disease, I cannot answer. I don’t have experience with that. Please, what was the last question? Dr. Katsilambros: The fact that triglycerides were higher in men. Usually these men are volunteers and in many countries including yours these volunteers are paid, and certainly some of them are drinkers. With women this is not the case and it might possibly make a difference. Dr. Bantle: This might make a difference but we screened subjects carefully for a history of alcohol consumption. They were to avoid alcohol during the period of study but I cannot guarantee they did. We were quite certain they ate what we asked them to eat, but whether or not they used alcohol I can’t say for certain. They were solid Minnesotan citizens who pretty much do what you ask them to if they say they will. I will tell you a story about our study. One day I was called by one of the nurses to come and talk to one of the subjects. When I came into the room, the woman was crying. She said she had to tell me that she had eaten something that wasn’t in the study diet. She confessed to have eaten 2 Cherios, and then she started crying again. I was fairly confident the data from that woman were reliable. But I can’t verify that some of our subjects didn’t consume alcohol. One would hope that any effect of alcohol was evenly distributed between both treatments so that the randomization process would have reduced or eliminated any effect. Dr. T. Wilkin: If I recall a review I read some months ago now, the average American child is consuming something between 60 and 80 g of fructose from soda drinks alone in the course of the day, whereas if left to the natural sources of fruit and vegetables it might be between 10 and 15 g. Even if your evidence is such that you observe nothing special about the action of fructose, there was a huge increase in the number of calories that have been consumed as a result of fructose ingestion and the link between obesity and fructose can surely be argued still to be there. Dr. Bantle: Yes, and perhaps the problem with fructose is it tastes good and is pleasant to consume. The overall problem may be that we are victims of our own success with abundant food that tastes good and is inexpensive. Several years ago

92

Fructose: Optimal Sweetener? McDonalds was implicated in the outbreak of obesity and they may have a similar sort of problem. They make food that tastes good and is inexpensive. That may be as much the problem as the specific food type. I would not implicate fructose until we have better evidence. For instance, sucrose might do the same thing, and if we decided to sweeten drinks with glucose, it would potentially do the same thing. Dr. T. Wilkin: In your data of the 42-day study you showed us I think 14, 28 and 42 days, but I don’t think you showed us the time zero, and I wondered if the levels of time zero or levels of day 42 had returned to what they were at time zero. Dr. Bantle: The lipid values? Dr. T. Wilkin: Cholesterol values. Dr. Bantle: There were baseline values obtained. The main thing that happened was that lipids got better on the glucose diet which would suggest that the baseline diet was high in fructose. Dr. Golay: Do you have more details concerning the group of women: age, hormone replacement, etc.? Dr. Bantle: The women were picked so that 6 were under 40 and 6 were over 40. We looked at age in both men and women and could find no effect, although the subject sample size was small. Both in younger men and older men, the triglyceride values were higher on the fructose diet. In older and younger women there was no effect of the fructose diet on triglycerides. I don’t recall the hormonal status but I think most of the older women were postmenopausal. We did not stop any replacement therapy they were receiving so we cannot speak much on the effect of estrogens based on this study. Dr. Metzger: I am also interested in the differences between the men and women. Is there any difference in the handling of fructose between men and women? Dr. Bantle: Not that I know of and I am not sure there was a real difference in our study. In the study from Elliott et al. [1] that I showed, the women did have a rise in triglycerides when given a high fructose diet. However, there was more fructose, 30% of energy for 1 day, and I think that is an important issue. Dr. Slama: I had a similar experience as you many years ago and I came to the same conclusion. But for women I have not looked at that carefully; perhaps I missed something but I could go back to my data. Having a large experience with animal models I can say that when you feed rats large amounts of sucrose or fructose then a massive infiltration of fat is seen everywhere. What is striking is that the liver is twice the size of that of the control rat, and this liver is half fat and half glycogen. The same but less striking observation can be made in the muscle; there is a lot of fat in the muscle and a lot of glycogen which could be utilized by sportsmen. This is so true that not only we but also others are routinely using an insulin-resistant rat model: fructose fat rats. Dr. Bantle: I think you made a good point. Fructose is very lipogenic in animal models although many of the animal studies have employed very high fructose intakes, between 30 and 70% of energy. I think a key point that is not commonly reported is the energy balance of the animals. That is, with excess energy, fructose is much more lipogenic and there are more adverse effects on lipids than with an energy-deficient diet where most of the fructose is burned as it is consumed. But that hasn’t been studied. Dr. Slama: I can add that we observed this even pair-fed rats; they have the same body weight but much more fat and glycogen in the liver than control rats. So it is not only a question of energy and free access to palatable food, it is a question of metabolism and you have shown that the fructose is prone to conduct rather to glycogen and/or to lipids rather than going back to glucose. Dr. Bantle: I don’t doubt the observation but I have difficulty to explain how, if they are truly pair-fed and getting isocaloric diets, they would produce and store more

93

Bantle fat on one diet than on the other, no matter what the source of energy. Perhaps there is something about fructose that affects metabolism. Dr. Hill: I have a comment and a question. First I think it is very dangerous to correlate specific things in the environment with changes in obesity rates. There are so many things that correlate. Along those lines, it is disturbing to hear everybody blaming the obesity epidemic on fructose with such little data. My question relates to Dr. Slama’s question. One of the concerns is that fructose may be producing insulin resistance in the liver and that could lead to weight gain and obesity. Are there data available that this may be a pathway to link fructose with obesity? Dr. Bantle: This is a good question and I am afraid I am not aware of any data. Perhaps someone in the audience is. Is anyone aware of such data? Dr. Slama: No, because really in animals you can feed them with 70% calories coming from fructose. Dr. Bantle: Regarding your comment Dr. Hill, I think that the liquid sweeteners may in fact be more a problem than solid sweeteners. That is, all the calories that we get through soft drinks may in fact be highly undesirable. Dr. Chiasson: I remember that Dr. Reaven was using that specific animal model to create hypertension and also insulin resistance. I was just wondering whether in your experience the subjects did develop any change in the blood pressure? Dr. Bantle: We did monitor blood pressure carefully and could not demonstrate any difference between the two diets in blood pressure. I too read Dr. Reaven’s studies with interest. I think he used 70% fructose in his rat models. Dr. Halimi: As Dr. Slama’s group with the same animal model, we found the same things: a huge amount of visceral fat, a large liver with steatosis, and hypertension. Fructose increases plasma triglycerides in men but not in women, and not to the same extent in all populations, more subjects with metabolic syndrome. When plasma triglyceride increases, what happens to the low-density lipoprotein particle size in the studies mentioned? Dr. Bantle: We did not measure the particle size in our studies and I am not aware of anyone who has. So again, I am afraid I can’t answer the question based on any experience or data. Dr. Slama: I find Dr. Halimi’s comment very interesting. Could you not apply the definition of metabolic syndrome to both populations in your population and see if there is a trend towards more frequent metabolic syndrome in men than in women? Dr. Bantle: You mean retrospectively go back and look at the data and define those in the groups who had metabolic syndrome. This is an interesting idea. I think we did have a range of body mass indexes. So it may be possible for us to do that and see if it was those with higher body mass indexes and metabolic syndrome who had the more significant response to fructose. It is a good idea, thank you. Dr. T. Wilkin: Given the data on hepatic infiltration in the animals, did you have the opportunity to look at inflammatory markers? Dr. Bantle: We had the opportunity but we didn’t take it. In retrospect, I wish we had. Dr. Mooradian: Just a caution to those of us who do a lot of cell culture studies: fructose is a very nasty substance and it has a lot of cell toxicity. I have concerns about this fructose intake even though the effect on glycemia and lipids seems to be modest within the range of fructose consumed day to day. But at least in the cell culture field fructose is by far a much more toxic substrate than glucose is. I wonder if we are missing some additional effects of fructose when we just focus on glucose and the lipid effect of this sugar. Dr. Bantle: That is a good point and there may be other effects of fructose, as yet not defined, that are adverse. I would use the opportunity to point out that we were looking at 17% of energy as fructose which is something like the 90th percentile of intakes in the United States. It was quite a high fructose intake but 10% of the people

94

Fructose: Optimal Sweetener? in the United States consume that much or more. I am of the opinion that putting this much fructose into the energy stream is perhaps not a good idea. We suggested glucose would be a suitable replacement but it turns out that glucose is much less sweet than fructose. The amount you have to put in to make a soft drink taste like it does now would give 50% more calories. Recipes that employ sucrose or high fructose corn syrup would have to be changed in terms of preparation, amounts, baking time, caramelization and all sorts of factors. These would become problems that would need to be solved. Dr. Halimi: Just another comment about the high fructose diet in rats. It is not confirmed in human studies but in rats there is also a huge increase in many oxidative stress parameters and this parallels insulin resistance [2, 3]. It would be very interesting to examine the situation in humans, with a high fructose intake as the current intake in the US, to measure the oxidative stress in these populations. Dr. Bantle: I agree, more study in fructose is clearly needed to see if what happens in animals also happens in humans. Dr. Mooradian: Just to follow up on Dr. Halimi’s point: tomorrow I am going to show data specifically on the effect of fructose in oxidation. Fructose is 7- to 8-fold more pro-oxidant compared to glucose. Dr. Slama: Have you seen convincing data that 50:50% mixing of glucose and fructose is the same as sucrose in terms of the metabolic effect? Dr. Bantle: No, I don’t know of any such data. My assumption is that, since 55% fructose corn syrup contains nearly equal amounts of glucose and fructose, it would have the same effect as sucrose. Dr. Slama: I am not sure really but there might be some data proving the reverse. I think that the problem comes from your country because you are not producing sucrose or only very small amounts, and so you are producing your natural sweeteners, corn syrup, whereas in other countries sucrose is widely used. Dr. Bantle: I agree, we are part of the problem. Where I come from, as soon as you leave the city, there is corn as far as the eye can see in all directions. Dr. Slama: So you may have some trouble in your country to pick up all. Dr. Bantle: Corn is a good thing in most respects. Dr. Halimi: Just some information, very interesting in my opinion. In cooperation with a center specialized in animal feeding, we have confirmed that a high fructose diet is able to induce insulin resistance in male rats. But this diet does not induce insulin resistance in females [4]. Second, when the same diet, very rich in fructose, is reproduced using honey, then there is no insulin resistance. This could be due to antioxidative substances in the honey [5].

References 1 Elliott SS, Keim NL, Stern JS, et al: Fructose, weight gain and the insulin resistance syndrome. Am J Clin Nutr 2002;76:911–922. 2 Faure P, Rossini E, Wiernsperger N, et al: An insulin sensitizer improves the free radical defense system potential and insulin sensitivity in high fructose-fed rats. Diabetes 1999;48:353–357. 3 Faure P, Roussel A, Coudray C, et al: Zinc and insulin sensitivity. Biol Trace Elem Res 1992;32:305–310. 4 Busserolles J, Mazur A, Gueux E, et al: Metabolic syndrome in the rat: females are protected against the pro-oxidant effect of a high sucrose diet. Exp Biol Med (Maywood) 2002;227: 837–842. 5 Busserolles J, Gueux E, Rock E, et al: Substituting honey for refined carbohydrates protects rats from hypertriglyceridemic and prooxidative effects of fructose. J Nutr 2002;132: 3379–3382.

95

Beyond Glycemic Control Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 97–105, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

Optimal Diet for Glycemia and Lipids William C. Knowler National Institute of Diabetes and Digestive and Kidney Diseases, Phoenix, AZ, USA

Abstract Concentrations of glucose and lipids in the blood have important health implications and are influenced by dietary intake. Dietary intake and energy expenditure regulate body weight, which is also an important determinant of health. Thus it would be important to determine the optimal diet for affecting blood glucose and lipids and body weight. Many professional health organizations in different countries have made dietary recommendations that include caloric restriction when needed to prevent or reverse overweight or obesity, limitation of saturated and trans fat, and emphasis on fruits and vegetables. These professional groups have not recommended extremely low carbohydrate or extremely low fat diets, despite much popular interest and recent research in these approaches. In several clinical trials, diet and exercise interventions prevented or delayed the development of type-2 diabetes. These trials showed the value of diet interventions, but did not attempt to determine which dietary approach was optimal. Clinical trials attempting to determine the optimal diet suffer from small sample sizes, short follow-up, and poor follow-up of participants. Therefore, the optimal balance between the total fat and carbohydrate contents and the optimal types of fats and carbohydrates remain unknown. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

Introduction Dietary intake influences concentrations of glucose and lipids in the blood, both of which have important health implications. Hyperglycemia in persons with diabetes often leads to symptoms that can be relieved by lowering the blood glucose concentrations. Thus, dietary changes (caloric restriction or changing the composition) may help to relieve symptoms, although drug therapy is often also needed, especially with extreme hyperglycemia. Diabetes, a disease that carries substantial morbidity from microvascular, macrovascular and nerve damage as well as higher mortality rates, is defined by hyperglycemia. Among 97

Knowler people with diabetes, the degree of hyperglycemia accelerates the development of these complications, especially microvascular complications such as retinopathy and nephropathy. Among non-diabetic persons, the degree of hyperglycemia predicts subsequent development of diabetes and cardiovascular disease. Dyslipidemia is rarely symptomatic, so the motivation for changing serum lipids by dietary or other means is for long-term prevention of diseases associated with dyslipidemia. Serum concentrations of lipids, particularly cholesterol and its low-density lipoprotein and high-density lipoprotein fractions, are related to development of cardiovascular diseases. Because caloric intake and energy expenditure jointly determine change in body weight, dietary intake is one of the key components in the prevention or treatment of overweight or obesity, which in turn are associated with many chronic diseases, including type-2 diabetes and cardiovascular diseases. Therefore, the consumption of diets with favorable effects on glycemia, lipids, and body weight has the potential of improving health. Because of the increasing importance of cardiovascular disease as the leading cause of death in China [1], the dietary effects on cardiovascular disease are important to this nation.

Current Dietary Guidelines A number of national or professional health organizations have recommended guidelines for healthy diets, either for general health or focused on specific aims such as controlling obesity or diabetes, or preventing heart disease. Most current guidelines emphasize limitation of caloric consumption from fats. For example, the Chinese Ministry of Health guidelines for prevention and control of overweight and obesity in adults included moderate caloric restriction plus physical activity with an emphasis on diets with low fat content, complex carbohydrates (including cereals), and fresh fruits and vegetables [2]. In the setting of dyslipidemia, they also recommend limitation of saturated fat and cholesterol. The American Diabetes Association published a technical review of dietary guidelines for people with or at high risk of developing diabetes [3]. Recommendations for people with either type-1 or type-2 diabetes were to include carbohydrates from whole grains, fruit, vegetables, and low-fat milk. Less than 10% of calories should be derived from saturated fat, and about 10% should come from polyunsaturated fat. In the typical American diet that contains more than the recommended amount of saturated fat, the calories from saturated fat should be replaced by calories from carbohydrate and monounsaturated fat, with the combination of the two making up 60–70% of dietary intake. Interestingly, there was insufficient evidence to recommend how much of this combination should be carbohydrate and monounsaturated fat. Another controversial point for which they found insufficient evidence was the glycemic index of foods. Thus, they made no recommendation 98

Optimal Diet for Glycemia and Lipids regarding choosing foods based on the glycemic index. This technical review also contained recommendations concerning energy balance and obesity [3]. The reviewers concluded that prescription of weight-loss diets alone was unlikely to produce sustained weight loss, but structured programs including exercise and behavioral modification could produce sustained weight loss of 5–7%. The effectiveness of this type of program was, in fact, shown in the Diabetes Prevention Program clinical trial [4]. Similar, though not identical, dietary recommendations were made by the Diabetes and Nutrition Study Group of the European Association for the Study of Diabetes [5] and the American Heart Association [6]. Notably, none of these guidelines from professional health organizations recommended high-fat low-carbohydrate diets, or extremely low-fat diets (⬍10% of calories), despite the recent interest in such diets (see below).

Dietary Effects on Mortality Despite the abundance of dietary guidelines, it is hard to find evidence that dietary composition has major effects on longevity. In the Malmö Diet and Cancer Study, all-cause mortality rates were not significantly related to the fat content of the diet in either men or women [7]. The rate of death attributable to cardiovascular disease, however, was significantly inversely related to dietary fat intake in men, but not in women. The EPIC study evaluated dietary intake as a predictor of mortality among over 70,000 people at least 60 years of age in 10 European countries [8]. Instead of evaluating simple dietary composition (i.e. percent of calories from fat, carbohydrate, protein, and alcohol), they constructed a ‘Mediterranean diet score’ reflecting the dietary composition previously shown in Greece to be associated with longevity. The score was derived from points for answering dietary questions on consumption of various types of foods. Higher scores, indicating dietary composition more like that of the traditional ‘Mediterranean diet’, were derived from high consumption of vegetables, legumes, fruits, cereals, fish, and unsaturated fat relative to saturated fat; low consumption of meat and dairy products, and moderate alcohol consumption. The score was inversely and approximately linearly related to subsequent mortality. This suggested that throughout Europe, mortality rates were lower in those eating the ‘Mediterranean diet’. This favorable diet is described not specifically by fat and carbohydrate content, but by types of foods. A higher score may depend more on type of fat consumed rather than total quantity.

Diabetes Prevention Clinical Trials The degree of body weight relative to height, commonly expressed as the body mass index, is a strong predictor of the development of type-2 diabetes [9]. 99

Knowler This observation led to the obvious hypothesis that weight loss might lower a person’s risk of developing type-2 diabetes. Three large randomized clinical trials have tested the hypothesis that dietary change in persons at high risk of type-2 diabetes can reduce the incidence of diabetes during a treatment period of several years. The first two employed diet and exercise interventions compared with each other [10] or combined [11]. The third, the US Diabetes Prevention Program, used a structured program of diet and exercise modification designed to produce weight loss compared with a program of diet and exercise advice only [4, 12]. It also included two pharmacologic treatment arms that will not be discussed here. Lifestyle modification lowered the incidence rate of diabetes in all three of these clinical trials [4, 10, 11]. The diet intervention in the Diabetes Prevention Program started with a reduction of fat to less than 25% of caloric intake [4, 12]. This was followed, when necessary, by total caloric restriction. Emphasis was put on self-monitoring of diet, exercise, and weight, and case managers were used to help each individual make the appropriate changes. The goal was gradual weight loss of 0.5–1 kg/week with a longterm goal of loss of 7% of body weight. This intervention resulted in a 58% reduction of the incidence rate of diabetes during a treatment and follow-up period averaging 2.8 years/participant [4]. The weight-loss intervention also resulted in a significant fall in serum triglycerides compared with the adviceonly group [13]. Total and low-density lipoprotein cholesterol were also lowered, but not to an extent significantly different from the advice-only group. The Diabetes Prevention Program followed only one diet and exercise approach to weight loss. Although it was successful compared with the adviceonly group, this study did not evaluate different dietary approaches. Thus it is not known whether the weight loss induced in the Diabetes Prevention Program was due to the fat reduction, or whether other approaches to limiting caloric intake and increasing energy expenditure would have been more or less successful. It is likely that weight loss per se has more important effects on glycemia and lipids than does the type of diet used to achieve weight loss.

Other Evidence for Dietary Effects Although the technical review published by the American Diabetes Association found insufficient evidence for recommending food selection based on the glycemic index [3], there is some evidence suggesting that a lower glycemic index (effect of a food on raising the blood glucose) or glycemic load (a product of the amounts of food with their glycemic indices) may reduce the incidence of type-2 diabetes. The Health Professional Followup Study of Men reported that neither glycemic load nor cereal fiber intake alone was strongly related to subsequent diabetes. The combination of low cereal fiber intake and high glycemic load, however, was associated with an approximate doubling of the risk of developing diabetes [14]. 100

Optimal Diet for Glycemia and Lipids There has also been substantial controversy regarding the potential benefits of several recently popular diets, especially those with extremes of fat and carbohydrate content. A recent randomized controlled clinical trial attempted to evaluate four such diet plans ranging from the lowest carbohydrate and highest fat content (the Atkins diet) to the lowest fat and highest carbohydrate content (the Ornish diet) [15]. Forty persons were randomized to each of the four diets and followed for 1 year. As is typical in many published diet clinical trials, the dropout rate was high, making interpretation difficult, because dropping out of a weight loss trial may not be random but related to success with weight loss. Perhaps because of the dropout problem and the limited sample sizes, there were no significant differences between the treatment groups in weight loss or changes in lipids or other cardiovascular risk factors. Such studies, directly comparing different diet regimens for longer periods of time, but with larger sample sizes and more complete follow-up of all participants regardless of weight loss, are needed to provide firm evidence for recommending the optimal diet. Concern has been raised recently that the rapidly increasing consumption of high fructose corn syrup in the United States, due to its becoming the major sweetener used in beverages since the 1970s, is a major contributor to the increasing prevalence of obesity in the United States [16]. Reported consumption of sweetened beverages predicted the incidence of diabetes in the Nurses Health Study, in which persons reporting consumption of at least 1 beverage/day had about double the incidence rate of diabetes as those reporting such consumption less than once per month [17]. Conclusions Currently the answer to the question of what is the optimal diet for glycemia and lipids is unknown. The best way to answer this question will be from randomized clinical trials comparing different diet plans, along the lines of the clinical trial of the four diets described above [15]. Such studies should be larger and have at least several years of follow-up. Most important (and most difficult) will be assuring that almost all participants remain under followup throughout the study. In the opinion of the author, it is best to avoid overweight or obesity, and to limit saturated and trans fat, and high-fructose sweeteners. There is insufficient evidence to recommend the optimal balance between total fat and carbohydrate contents of the diet. References 1 He J, Gu D, Wu X, et al: Major causes of death among men and women in China. N Engl J Med 2005;353:1123–1134.

101

Knowler 2 Chen C, Lu FC, Department of Disease Control, Ministry of Health, PR China: The guidelines for prevention and control of overweight and obesity in Chinese adults. Biomed Environ Sci 2004;17(suppl):1–36. 3 Franz MJ, Bantle JP, Beebe CA, et al: Evidence-based nutrition principles and recommendations for the treatment and prevention of diabetes and related complications. Diabetes Care 2002;25:148–198. 4 Knowler WC, Barrett-Connor E, Fowler SE, Diabetes Prevention Program Research Group: Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393–403. 5 Mann JI, De Leeuw I, Hermansen K, et al: Evidence-based nutritional approaches to the treatment and prevention of diabetes mellitus. Nutr Metab Cardiovasc Dis 2004;14:373–394. 6 Kraus RM, Eckel RH, Howard B, et al: American Heart Association Scientific Statement: AHA Dietary Guidelines. Circulation 2000;102:2284–2299. 7 Leosdottir M, Nilsson PM, Nilsson J-Å, et al: Dietary fat intake and early mortality patterns – data from the Malmö Diet and Cancer Study. J Intern Med 2005;258:153–165. 8 Trichopoulou A, Orfanos P, Norat T, et al: Modified Mediterranean diet and survival: EPICelderly prospective cohort study. BMJ 2005;330:991. 9 Knowler WC, Pettitt DJ, Savage PJ, Bennett PH: Diabetes incidence in Pima Indians: contributions of obesity and parental diabetes. Am J Epidemiol 1981;113:144–156. 10 Pan X-R, Li G-W, Wang J-X, et al: Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance: the Da Qing IGT and Diabetes Study. Diabetes Care 1997;20: 537–544. 11 Tuomilehto J, Lindstrom J, Eriksson JG, et al: Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344: 1343–1350. 12 Diabetes Prevention Program Research Group: The Diabetes Prevention Program (DPP): description of the lifestyle intervention. Diabetes Care 2002;25:2165–2171. 13 Ratner R, Goldberg R, Haffner S, Diabetes Prevention Program Research Group: Impact of intensive lifestyle and metformin therapy on cardiovascular disease risk factors in the diabetes prevention program. Diabetes Care 2005;28:888–894. 14 Salmeron J, Ascherio A, Rimm EB, et al: Dietary fiber, glycemic load, and risk of NIDDM in men. Diabetes Care 1997;20:545–550. 15 Dansinger ML, Gleason JA, Griffith JL, et al: Comparison of the Atkins, Ornish, Weight Watchers, and Zone diets for weight loss and heart disease risk reduction. JAMA 2005;293:43–53. 16 Bray GA, Nielsen SJ, Popkin BM: Consumption of high-fructose corn syrup in beverages may play a role in the epidemic of obesity. Am J Clin Nutr 2004;79:537–543. 17 Schultze MB, Manson JE, Ludwig DS, et al: Sugar-sweetened beverages, weight gain, and incidence of type 2 diabetes in young and middle-aged women. JAMA 2004;292:927–934.

Discussion Dr. T. Wilkin: The difficulty you had with your conclusions may reflect the initial slides which are very challenging. You showed us the exponential increase in type-2 diabetes with body mass index and the lack of effect of various amounts of fat in the diet, at least in one study. If you are looking at the problem as to whether it is dietary content or body mass, then the issue is surely one of tolerance. If you can tolerate the fats and if you can tolerate the glucose, and that by and large is related to body mass and adipocyte capacity, then you could probably get away with a healthy and a varied diet or what some might consider an unhealthy diet, whereas if you have gained the weight and lost the adipocyte capacity and lost the glucose tolerance, then your ability to tolerate becomes narrower and narrower. So what I took from your talk was a far stronger emphasis on body mass than on what the diet actually contains. Dr. Knowler: That is also my feeling, but I might add that this is in terms of type-2 diabetes, not necessarily of dyslipidemia. In terms of risk of diabetes, the body weight

102

Optimal Diet for Glycemia and Lipids or body mass index is by far the most important. As long as people can limit their weight or overweight people can reduce their weight then the specific dietary components may be less important. An important question for diabetes prevention is: what is the best approach to loosing weight? Unfortunately there is no easy answer for that question. Dr. Hill: One of the mistakes I think we make in the US is basing general recommendations on results from weight loss studies. For example Dr. Golay demonstrated that if you fix calories during food restriction the dietary composition is not an important determinant of the amount of weight loss. However, in most cases calorie intake is not fixed when people try to lose weight. Many people who go on the Atkins diet lose a lot of weight probably because they eat less. For dietary fat, lots of data suggest that the higher fat in the diet the more energy you are likely to consume. So it is dangerous to base population recommendations on short-term weight loss results. My question is whether or not you think that the optimum diet should vary with physical activity? Is the optimum diet for a sedentary person different than for someone who gets regular physical activity? Dr. Knowler: That is very likely but I actually don’t know how I would change the recommendation for the very active person. A person can be very active comsuming either a high fat or high carbohydrate diet. Dr. Hill: If you are very physically active and are burning a lot of fat, the optimum amount of fat in the diet could be higher than for a sedentary person. Dr. Knowler: It could be, but I don’t know what it is. Dr. Field: The Diabetes Prevention Program (DPP) trial is our best evidence that diet influences outcome or prevention. An argument that is raised for the implementation of intensive lifestyle intervention for the prevention of type-2 diabetes is the considerable cost. Do you think that this approach is feasible based on the number of people at risk? Dr. Knowler: When you are asking about feasibility and cost, are you referring to effort to prevent diabetes or treat diabetes, because the answer might be different? Dr. Field: To prevent. Dr. Knowler: Again I can’t give you the answer for that for sure. Two economic analysis of the DPP have been published this year in the Annals of Internal Medicine, one by the DPP itself, one by an outside group, and they came to very different conclusions [1, 2]. The DPP analysis concluded that the DPP style intervention, specifically the intensive lifestyle intervention, did give good health value for money spent and, in terms of quality of life years improved, was comparable or favorable to many things that are standard medical practices. The other paper [2] came to very different conclusions that it was just not economically feasible at all. And the major differences I think, these were both exercises in modeling disease progression and cost. They projected well beyond the data in the DPP because they were projecting what would happen over 30 years, and whenever you are projecting out 30 years, obviously what you get depends on what you assume is going to happen in those 30 years. Dr. Bantle: As we discuss diet, perhaps I can ask your thoughts about the growing body of evidence that suggests that energy intake, energy expenditure and body weight are all centrally regulated by the hypothalamus and under rigorous control. I am bothered by the thought that when we ask patients to override this control system by force of will, we are asking them to do something that is very difficult. Dr. Knowler: I think that central control is also subject to external influences. The fact that there have been huge changes in obesity in virtually all countries of the world means that things are changing due to outside influences. Dr. Huixia Yang: Some studies from Prof. Barker mentioned that birth weight is most important for diabetes and other chronic diseases. In the DPP study we don’t

103

Knowler know the patients’ relation to birth weight. In the patients with low birth weight, meaning intrauterine growth retardation, perhaps diabetes also increases. I just want to know the relation of lifestyle changes in diabetes to the incidence of other things. Dr. Knowler: Unfortunately I can’t comment about that because we didn’t have those data in the DPP. Dr. Huixia Yang: I just wanted to confirm the Barker hypothesis because Prof. Barker mentioned that a lot of chronic diseases may be of fetal origin. Dr. Knowler: I can’t comment on that, regarding the DPP. In our studies of the Pima Indians, we do know the birth weight of many people and we could confirm that hypothesis in the people with very low birth weights had a higher risk of diabetes, but so did the people with a high birth weight, but for very different reasons. The infants who had very high birth weight tended to be born from mothers who had diabetes during the pregnancy, and that it puts them at very high risk. The infants with very low birth weight tended to be the offspring of diabetic fathers. Some of that effect may also be genetic, the people in the extremes of birth weight are inheriting different genetic susceptibility factors. But I think a very important thing is that the extremes of birth weight associated with greater diabetes risk included only about 10% of the population. Over about 90% of the range of birth weight there was no effect on subsequent risk of diabetes. Therefore the extremes of birth weight are very important for the individuals in those extremems, at least in our population but they are not frequent enough to explain diabetes in that population. Dr. T. Wilkin: Could I just return to Dr. Bantle’s question about the hypothalamus and setting, which I think is a very interesting one. I wonder whether the way in which our hypothalami are set relates to a far distant evolutionary period in which the exposure to food was very different than it is now. Perhaps the setting was far more related to carbohydrate content than to fat content, and as far as I understand, fat has very little anorexic effect in any case. So I wonder if our setting points are simply unrelated to the food type exposure that we have nowadays and therefore unresponsive to. Dr. Ditschuneit: You mentioned the guidelines for an optimal diet and that two servings fish per week is optimal. I think that it has something to do with n-3 and n-6 fatty acids and that they may play a role in cardiovascular disease. I would like you to comment on the role of n-3 and n-6 fatty acids and the relation between these two fatty acids. Dr. Knowler: As you point out I think the predominant evidence is that these fatty acids would be beneficial in terms of cardiovascular disease. I am not aware of data that they would have direct effects on either body weight or risk of diabetes. Certainly from the standpoint of cardiovascular health I think the evidence would favor the n-3 and n-6 fatty acids from fish and some vegetable sources, and that is why they are put on several groups’ recommendations. An anecdotal observation, which is not very scientific is that two of the countries in the world that generally have the world’s lowest mortality rates, are Japan and Sweden. Both have very high fish intakes, but very different diets in many other respects. The commonality of fish consumption is an interesting observation. Dr. Gerasimidi-Vazeou: Is there any evidence about the impact of high fat-high protein diets on factors related to endothelial function like tumor necrosis factor-␣, interleukin-6, C-reactive proteins, as the weight and the lipids have always been the endpoints of studies on the Atkins diet? Dr. Knowler: Unfortunately I don’t know the answer to that question. Dr. Hill: The studies that have been published on the Atkins diet have not looked at that. They have looked at total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein and triglycerides. We are doing a study right now collecting those data but we won’t have any results for a while.

104

Optimal Diet for Glycemia and Lipids Dr. Mingdao Chen: From your first several slides you put the carbohydrate plus monounsaturated fatty acids together, two substances which are not falling into the same category. What is the ratio between carbohydrate and monounsaturated fatty acids? Dr. Knowler: Ms. Franz or Dr. Bantle might want to respond to that question, because I took this from their paper. There has been a lot of agreement that saturated and trans-saturated fat are unhealthy and should be limited. The question is if you are going to limit them, what do you replace them with? The two obvious choices are carbohydrate or polyunsaturated and monounsaturated fats. The monounsaturated fat tend to be neutral in terms of their effects on dyslipidemia. There is some evidence that in some people very high carbohydrate diets may contribute to hypertriglyceridemia. So I think the view is that if you are going to reduce saturated fats, the monounsaturated fats are may be a good thing to replace them with. A particular recommendation was one that the sum of carbohydrate and monounsaturated fats should be 60 or 70%. Might I just ask Ms. Franz to comment more specifically what she recommends? Ms. Franz: The American Diabetes Association is in the process of updating their nutrition recommendations. In regard to percentages for macronutrient distribution we could find no evidence to support ideal percentages from macronutrients. Support is given to the Dietary Reference Intakes report that recommends meeting the body’s nutritional needs while minimizing the risk of chronic diseases. Adults should consume 45–65% of total energy from carbohydrate, 20–35% from fat, and 10–35% from protein [3]. Saturated fat intake should be ⬍7% of total calories and intake of trans fat should be minimized. Either mono- or polyunsaturated fats can be used to replace saturated fats in the diet. Dr. L. Wilkin: No one as mentioned dietary salt. I am wondering if there is not a relationship between salty snacks and the consumption of sweetened drinks, especially among young people? So if we could cut the salt, would we not reduce the amount of soda that they drink? Dr. Knowler: That is a good question. Specifically I didn’t address dietary salt because I am not aware of data of it having direct effects on glycemia or lipids, which was my charge. But you are right, salt in the diet is very often correlated to fat in these salty snacks so it is an example of changing one aspect of diet, it is hard to isolate dietary components because they are also correlated.

References 1 Herman WH, Hoerger TJ, Brandle M, et al, Diabetes Prevention Program Research Group: The cost-effectiveness of lifestyle modification or metformin in preventing type 2 diabetes in adults with impaired glucose tolerance. Ann Intern Med 2005;142:323–332. 2 Eddy DM, Schlessinger L, Kahn R: Clinical outcomes and cost-effectiveness of strategies for managing people at high risk for diabetes. Ann Intern Med 2005;143:251–264. 3 Institute of Medicine: Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. Washington, National Academies Press, 2002.

105

Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 107–125, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

Antioxidants and Diabetes Arshag D. Mooradian Division of Endocrinology, Diabetes, and Metabolism, Department of Internal Medicine, St. Louis University School of Medicine, St. Louis, MO, USA

Abstract Overproduction of superoxide by the mitochondrial electron transport chain is the common link between the various pathways of glucotoxicity. The increased oxidative byproducts in diabetes are the result of a glucose-induced increase in the production of reactive oxygen species and decreased antioxidant defense capacity. Several epidemiologic observations indicate an inverse association between vitamin E intake and coronary heart disease (CHD). There are several limitations in such studies including the fact that they rely on food questionnaires and dietary recalls. Large interventional trials have yielded inconsistent results. Of concern is that, in some of these studies there was a greater incidence of lung cancer or CHD. These observations underscore the potential hazards of consuming large amounts of antioxidants. At the present time, given the inconsistencies of the studies available, the widespread supplementation with pharmacological doses of antioxidants should be discouraged. Future studies should focus on identifying reliable markers of oxidation to incorporate these measurements in the clinical interventional trial. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

Introduction Despite the advances made in the management of diabetes, people with uncontrolled hyperglycemia continue to suffer from dire complications. One of the most serious complications of diabetes is cardiovascular disease (CVD) [1]. Although there are multiple potential causes for increased risk of premature CVD in diabetes, increased oxidative stress has emerged as an important therapeutic target to reduce the burden of this complication. Indeed, it is possible that overproduction of superoxide by the mitochondrial electron transport chain is the common link between the various pathways of glucotoxicity [2]. Thus, it is plausible that identifying agents that can reduce oxida107

Mooradian tive load would reduce the risk of developing diabetic complications including CVD. Hence, the continued interest in antioxidant therapy in diabetes has generated a voluminous scientific literature. In this communication the potential mechanisms of increased oxidative load in diabetes will be reviewed and the currently available large trials of antioxidants in reducing disease burden will be critically evaluated. Finally the arguments in favor or against antioxidant supplementation will be discussed.

Increased Oxidative Load in Diabetes: the Role of Hyperglycemia Many but not all studies have found increased accumulation of oxidation byproducts in diabetes. Lipid peroxidation byproducts such as exhaled ethane or pentane, malondialdehyde, conjugated dienes and F2-isoprostanes are increased in diabetic subjects and in animal models of diabetes [for review see 3, 4]. Similarly, oxidation byproducts of proteins and deoxynucleic acids (DNA) are increased. The mechanisms underlying the increased oxidative byproducts in diabetes are multiple. These mechanisms can be generally classified as either a consequence of a glucose-induced increase in the production of reactive oxygen species (ROS), decreased antioxidant defense capacity, or the inability to eliminate the oxidized byproducts efficiently. The latter is not supported by as much experimental evidence as the former two mechanisms. Through glycolysis, glucose is converted to pyruvate that enters the tricarboxylic acid cycle, increasing the electron transport chain and, in the process, ROS is produced. This process has been shown to occur in several studies of endothelial as well as other cell cultures. Glucose-induced ROS production can be ameliorated with both conventional, such as vitamin C and E [5], and unconventional antioxidants, such as statins [6] and carvedelol [7]. It is noteworthy that in these experimental conditions the rate of ROS production decreases after the second or third hour of treatment suggesting that adaptive mechanisms are activated to ameliorate the initial surge of oxidative activity. These adaptations may originate at nuclear and extranuclear sites of action. Several genes are now identified that have antioxidant response elements (AREs) in the promoter region. The antioxidant response gene family involves at least six genes on at least two different chromosomes. Examples include nicotinamide adenine dinucleotide phosphate (NADPH) quinone oxidoreductase I and II (hNQO1 and hNQO2) genes that can mount a coordinated response to oxidant stress [8]. Other AREs have been identified and been shown to respond to conventional antioxidants sometimes downregulating gene products that have cardioprotective properties [9]. Thus the genomic response to oxidants and antioxidants could have both protective and deleterious consequences. 108

Antioxidants and Diabetes In addition to being a substrate for cellular respiration, glucose can induce ROS generation through multiple pathways including the promotion of glycation of proteins and activation of protein kinase C (PKC) activity [2]. Furthermore, glucose has auto-oxidative potential that has been demonstrated in cell-free systems [10]. In comparison to other simple sugars, fructose has the most potent auto-oxidative potential while deoxyribose has the least capacity of auto-oxidation [10]. Chronic hyperglycemia can also promote oxidative stress through interference with antioxidant defense systems. Diabetic individuals, especially those who are poorly controlled, are likely to develop multiple micronutrient deficiencies some of which have antioxidant activities [3, 11]. Of all the potential antioxidant deficiencies, depletion of the intracellular content of ascorbate can occur because of direct inhibition of the cellular uptake of dehydroascorbate by glucose [12]. Nevertheless, the effect of uncontrolled hyperglycemia on antioxidant defense capacity extends beyond individually known micronutrients [13]. The precise metabolic pathway by which hyperglycemia reduces the antioxidative defense capacity is not completely clear, but appears to be at least partly secondary to overconsumption of antioxidants in the presence of increased production of ROS.

Antioxidants and Disease Outcome: Observational Studies Observational studies correlating potential health benefits with the consumption of antioxidants are summarized in table 1. Several epidemiologic observations indicate an inverse association between vitamin E intake and CHD. For example, the Iowa Women’s Health Study (n ⫽ 34,486) [14], Nurses’ Health Study (n ⫽ 87,245) [15], and Health Professionals Follow-Up Study (men only; n ⫽ 39,910) [16] revealed that subjects in the highest quintile of vitamin E consumption from food and supplements had relative CHD risks of 0.78, 0.59, and 0.59, respectively, after adjustment for confounding variables such as age. The Nurses’ Health study [15] showed a decreased risk of coronary artery disease in women who took vitamin E supplements for more than 2 years. However, in this study women who took the supplements had a slightly better cardiovascular risk profile at baseline. It is also probable that people who choose the vitamin supplements had healthier lifestyle. The National Health and Nutrition Examination Survey (NHANES) I [17] concluded that there was an inverse association of vitamin C intake in males and all causes of death and CVDs. In this study diabetes and hypertension were not included as potential confounding variables. The Rotterdam study [18] and the Bruneck study [19] have observed the beneficial effects of carotenoid intake. However in the Rotterdam study, lifestyle variables were not taken into consideration and in the Bruneck study 109

Study

Study population

Study design and follow-up time

Antioxidants assessed

Dwyer et al. [20] Los Angeles Atherosclerosis Study

573 persons aged 40–60 years

18-month follow-up

Vitamin C and vitamin E intake determined by 24-hour recall, and plasma levels of antioxidants

D’Odorico et al. [19] Bruneck study

392 randomly selected men and women aged 45–65 years

5-year follow-up

Ford et al. (23)a NHANES III

1,010 controls; 277 persons with impaired glucose tolerance 230 diabetic patients aged 40–74 years

Cross-sectional

Results of study

Vitamin C supplement ⱖ857 mg/day was associated with increased intima media thickness (IMT) of the common carotid arteries; whereas vitamin E supplement of ⬎443 IU/day was associated with reduced progression. 18-month change in IMT was inversely related to the 3 measured oxygenated carotenoids (lutein, ␤cryptoxanthin, zeaxanthin) and one hydrocarbon carotenoid, ␣-carotene Plasma levels of carotenoids, ␣- and ␤-carotene plasma levels vitamins A and E were were inversely associated determined. Atherosclerosis with the prevalence of was assessed by duplex atherosclerosis of carotid ultrasound and femoral arteries Plasma level of ␤-Carotene and to lesser degree carotenoids measured cryptoxanthin and lycopene were inversely correlated with the degree of glucose tolerance abnormality. All the carotenoids were inversely related to fasting insulin concentration

Mooradian

110

Table 1. A select list of observational studies of the role of antioxidants

4,802 participants of Rotterdam aged 55–95 years

4-year follow-up

Will et al. [21]a NHANES III

237 diabetics 1,803 nondiabetics aged 40–74 years

Cross-sectional

Mayer-Davis et al. [22]a, SLVDS

387 type-2 diabetics aged 20–74 years

Cross-sectional and longitudinal 4-year follow-up

24-hour dietary recall

Coudray et al. [30]a EVA study

1,389 persons aged 59–71 years

Cross-sectional with a 4-year follow-up

Laboratory values of selenium, vitamin E and carotenoids were analyzed

Mayer-Davis et al. [31]a, IRAS and SLVD

520 persons from IRAS and 422 from SLVDS

Sanchez-Lugo et al. [32]a, IRAS

1,151 persons from IRAS group aged 40–69 years 247 women aged 56–71 years

IRAS: cross-sectional; SLVDS: crosssectional and longitudinal design Cross-sectional

Vitamin C assessed by food frequency interview in IRAS and 24-hour dietary recall interview in SLVDS Intake of vitamins E and C estimated from food frequency questionnaire Intake of vitamin C was assessed by food frequency questionnaire and vitamin C supplement consumption data

Jacques et al. [33] Nurses’ Health Study

10- to 12-year follow-up

Food frequency questionnaire for assessment of dietary antioxidant intake Vitamin C serum levels were assayed

High ␤-carotene intake but not vitamin C or E was associated with decreased incidence of myocardial infarction After adjustment for covariates, mean concentration of serum vitamin C did not differ according to diabetes status No protective effect of antioxidant intake on diabetic retinopathy. Potential harmful effect of vitamin E and carotenoid in certain subset of diabetics Selenium and vitamin E levels were increased in those subjects with lipemia. In diabetics and hypertensive subjects the carotenoids were reduced Vitamin C intake did not seem to be associated with CVD risk among diabetics Intake of vitamins E and C had no correlation with insulin sensitivity

111

Use of vitamin C supplements for more than 10 years was associated with a 77% lower prevalence of early lens opacities and 83% lower prevalence of moderate lens opacities

Antioxidants and Diabetes

Klipstein-Grobusch et al. [18]a Rotterdam Study

Study

Study population

Study design and follow-up time

Antioxidants assessed

Results of study

Kushi et al. [14] Iowa Women’s Health Study

34,486 Post-menopausal women aged 55–69 years

7-year follow-up

Vitamin E intake was found to be protective against CAD. Vitamin C and A intake had no impact on CAD

Losonczy et al. [24]a, EPESE

11,178 aged 67–105 years

⬃8- to 9-year follow-up

Dietary intake of vitamins A, E and C from food sources and supplements was evaluated with a questionnaire Vitamin E and C supplements taken was determined by medication and supplementation history

Gaziano et al. [25] Massachusetts Health Care Panel Study

1,299 elderly Mean of 4.75 Massachusetts years follow-up residents who were 66 years or older

Dietary carotene intake was assessed with a food diary

Salonen et al. [26]a Kuopio Ischaemic Heart Disease Risk Factor study

944 men aged 42–60 years

4-year follow-up

Plasma concentration of ␣-tocopherol was measured

Knekt et al. [27]

5,133 Finnish men and women aged 30–60 years

Longitudinal study with a 14-year follow-up

Carotene, vitamins C and and E consumption were estimated by the dietary history

All cause mortality and CAD mortality was reduced in those taking ␣-tocopherol vs. those who were not. Simultaneous use of vitamin C and E was associated with lower risk of mortality Fatal myocardial infarction was lower among those who were in the highest quartile for consumption of carotenecontaining food There was an inverse association observed between the ␣-tocopherol and the risk of developing diabetes mellitus Inverse relationship between dietary vitamin E and coronary mortality was observed. Some inverse correlation of coronary

Mooradian

112

Table 1. (continued)

39,910 US male 4-year follow-up health professionals 40–75 years of age

Carotene, vitamins C and E assessed with questionnaires

Shoff et al. [28]a Beaver Dam Eye Study

2,152 participants aged 43–84 years

Cross-sectional

Intake of vitamins E, C and carotene was determined with dietary questionnaires

Stampfer et al. [15] Nurses’ Health Study

87,245 female nurses aged 34–59 years 11,348 noninstitutionalized US adults aged 25–74 years

8-year follow-up

Vitamins E, C and carotene intake assessed with questionnaires Vitamin C intake measured with food frequency and supplement questionnaire

Groups of about 100 apparently healthy males aged 40–49 years from different regions of Europe

Cross-sectional

Enstrom et al. [17] NHANES I

Gey and Puska [29] Ischemic heart disease in crosscultural epidemiology

10-year follow-up

Plasma levels of vitamins A, C, E, ␤-carotene, and selenium

113

Reproduced with some modifications from Hasanain and Mooradian [3]. CAD ⫽ Coronary artery disease; CVD ⫽ cardiovascular disease; EPESE ⫽ Established Population for Epidemiologic Studies of the Elderly; EVA ⫽ Etude du Vieillissement Arterial; HbA1c ⫽ hemoglobin A1c; IRAS ⫽ Insulin Resistance Atherosclerosis Study; NHANES ⫽ National Health and Nutrition Examination Survey; SLVD ⫽ San Luis Valley Diabetes Study. aThese studies included a significant number of subjects with diabetes or insulin resistance.

Antioxidants and Diabetes

Rimm et al. [16] Health Professionals Study

mortality for vitamin C and carotenoids was observed only in women Higher intake of supplemental and possibly dietary vitamin E was associated with a significantly lower risk of CAD In people with diabetes no association between HbA1c and vitamins E, C and ␤-carotene intake was found Women who took vitamin E for more than 2 years were at decreased risk of CAD Inverse association between vitamin C intake and all cause mortality and CVD in males. No definite relation between vitamin C consumption and individual cancers Plasma levels of vitamins E and A were inversely related to the risk of ischemic heart disease

Mooradian the subjects with lower ␣- and ␤-carotene levels tended to be older and were smoking more in comparison to those with higher plasma levels of ␣- and ␤carotenes. Other observational trials [20–33] are also listed in table 1.There are several well-known limitations of the epidemiological studies. Often the confounding variables are difficult to identify to be able to make the necessary statistical adjustments. Vitamin intake in these studies is usually determined through notoriously unreliable food questionnaires and dietary recalls. Furthermore, it is also difficult to isolate the exact component of the diet that might have had an impact on a particular outcome.

Antioxidants and Disease Outcome: Interventional Trials There are many small short-term interventional trials showing a highly variable effect of antioxidant supplementation on markers of oxidation, glucose disposal and vascular reactivity. Differences in study design, the type and preparation of antioxidants used and lack of information as to the oxidative state of the population studied may explain some of the discrepancies in the outcome of these studies. However, the true clinical role of antioxidant therapy has to be evaluated in large randomized, double-blind and placebocontrolled studies. Some of the larger interventional studies that have been published are summarized in table 2. The outcomes of these studies are not always consistent. The Cambridge Heart Antioxidant Study (CHAOS) [34] and Antioxidant Supplementation in Atherosclerosis Prevention (ASAP) study [35] found a beneficial effect of vitamin E intake. The randomization of subjects in the CHAOS trial was fraught with problems, the number of events was small and the follow-up interval was short. The study population in the ASAP trial may have been deficient in vitamin E at baseline, and therefore the results of this study cannot be generalized to people with adequate vitamin E stores. In this study, the progression of carotid atherosclerosis was reduced only in the group of men who were smokers and consumed both vitamin E (136 IU) and vitamin C (250 mg) daily [35]. It is likely that the smokers are under greater oxidative stress hence they benefited the most from the antioxidants. The validity of the randomization and consumption of the vitamin supplements in each group was confirmed by showing that the plasma levels of ␣-tocopherol and ascorbate were appropriately increased by approximately 50–90% in those subjects who were randomized to receive the supplements [35]. Another trial showing the potential benefits of antioxidants was the Nutrition intervention trials in Linxian, China [36]. In this trial 29,548 Chinese participants were assigned to one of following combinations of supplements: (a) retinol 5,000 IU ⫹ zinc 22.5 mg; (b) riboflavin 3.2 mg ⫹ niacin 40 mg; (c) ascorbic acid 120 mg ⫹ molybdenum 30 ␮g, or (d) ␤-carotene 15 mg ⫹ selenium 114

Table 2. Large clinical trials conducted to delineate the role of antioxidant vitamins in preventing disease states Study

Study population

Study design and follow-up

Antioxidants assessed

Results of study

de Gaetano et al. [43]a Primary Prevention Project Yusuf et al. [37]a HOPE trial

4,495 Italians at risk of CVD

Randomized, placebo-controlled 3.6-year follow-up

Vitamin E 300 mg/day

Vitamin E showed no benefit

2,545 women 6,996 men 55 years or older at high risk of CVD 520 men and postmenopausal women aged 45–69 years

Double-blinded, randomized trial with a two by two factorial design Randomized, double-blind, placebo-controlled, 3-year follow-up

Vitamin E 400 IU/day for a mean of 4.5 years

Vitamin E showed no benefit

Salonen et al. [35] ASAP

GISSI trial [38]a GISSI-P

115

Hennekens et al. [45]

The progression of atherosclerosis was reduced in male participants who were smokers and were assigned to the combined treatment of vitamin E and C There was no harm or benefit demonstrated in the subjects receiving the ␤-carotene

Vitamin E conferred no significant benefit

␤-carotene had no adverse or beneficial effect even after

Antioxidants and Diabetes

Lee et al. [44] Women’s Health Study

Subjects were given one of the following treatments (a) 91 mg d-␣-tocopherol (b) 250 mg slow-release vitamin C (c) combination of above, or (d) placebo 39,876 women Randomized, ␤-carotene 50 mg, aged 45 years or double-blind, vitamin E 600 IU, and older placebo-controlled. aspirin 100 mg given on Treated for 2.1 years alternate days either and followed for 2 alone or in various additional years combinations of the three 11,324 Italian Randomized, Patients were assigned to patients surviving placebo-controlled, receive n-3 polyunsaturated recent (ⱕ3 months) with a 3.5-year fatty acid 1 g daily, MI follow-up vitamin E 300 mg daily, both or none 22,071 male Randomized, ␤-carotene 50-mg tablets physicians double-blind, given on alternate days

Study

Study population

Study design and follow-up

Physician Health Study I Omenn et al. [41] CARET

aged 40–84 years

placebo-controlled, 12-year follow-up Randomized, double-blind, placebo-controlled trial

18,314 men and women at high risk of lung cancer

Stephens et al. [34]a CHAOS

2002 patients with angiographically proven coronary atherosclerosis.

ATBC trial [39]

29,133 male smokers aged 50–69 years

Blot et al. [36] 29,584 Chinese Nutrition intervention participants trials in Linxian, aged 40–69 years China

Randomized, double-blind, placebo-controlled with a 1.4-year follow-up Randomized, double-blind, placebo-controlled with 5–8 years follow-up

Double-blind, randomized, placebo-controlled, 5.4-year follow-up

Antioxidants assessed

Results of study 12 years of supplementation

Patients either received 30 mg of ␤-carotene and 25,000 IU retinal palmitate taken daily or placebo

Patients either received 400 or 800 IU of vitamin E or placebo

␣-tocopherol 50 mg/day and ␤-carotene 20 mg/day, given alone or in combination

Daily supplementation with one of the following: (a) Retinol 5,000 IU ⫹ zinc 22.5 mg

There was no cardioprotective effect of combination of ␤-carotene and vitamin A. Increased incidence of lung cancer was observed in patients on supplementation Participants receiving 400 IU but not 800 IU vitamin E had a reduced rate of non-fatal MI

There was an increased incidence of lung cancer and cardiovascular events observed in patients taking carotene. There was slightly more incidence of hemorrhagic strokes in patients on vitamin E Reduction in total mortality, in cancer death and incidence especially for stomach cancer was seen in patients on the

Mooradian

116

Table 2. (continued)

Lee et al. [46] Women’s Health Study

39,876 healthy women aged 45 years or older

Randomized, double blind, placebo-controlled, 2 ⫻ 2 factorial design, followed up for an average of 10.1 years

(b) Riboflavin 3.2 mg ⫹ niacin 40 mg (c) Ascorbic acid 120 mg ⫹ molybdenum 30 ␮g (d) ␤-carotene 15 mg ⫹ selenium 50 ␮g ⫹ ␣-tocopherol 30 mg Vitamin E 600 IU and aspirin 100 mg given on alternate days

combination regimen of ␤-carotene, selenium and ␣-tocopherol

Vitamin E did not decrease risk of CHD

Reproduced with some modifications from Hasanain and Mooradian [3]. ATBC ⫽ Alpha-Tocopherol Beta Carotene; ASAP ⫽ Antioxidant Supplementation in Atherosclerosis Prevention; CARET ⫽ Beta Carotene and Retinol Efficacy Trial; CVD ⫽ cardiovascular disease; GISSI-P ⫽ Gruppo Italiano per lo Studio della Sopravivenza nell’Infarto miocardico-Prevenzione; MI ⫽ myocardial infarction. a These studies included a significant number of subjects with diabetes or insulin resistance.

Antioxidants and Diabetes

117

Mooradian 50 ␮g ⫹ ␣-tocopherol 30 mg. Supplementation with ␤-carotene, vitamin E, and selenium was associated with a significant reduction in mortality, mostly secondary to the reduced rate of gastric cancers [36]. However, the subjects in this study may have had some nutritional deficiencies at baseline. Another limitation of the study is that the precise component of the antioxidant supplement that was responsible for the beneficial outcome could not be ascertained. It is possible that the combination of the three antioxidants synergized the antioxidant properties of each other. Several other large randomized trials, such as the Heart Outcomes Prevention Evaluation (HOPE) study [37], the GISSI-P (Gruppo Italiano per lo Studio della Sopravivenza nell’Infarto miocardico-Prevenzione) trial [38] and the Alpha-Tocopherol, Beta Carotene Cancer Prevention (ATBC) study [39] and the Heart Protection trial (HPS) [40] failed to demonstrate any significant cardioprotective effect of vitamin E either alone or in combination with vitamin C and ␤-carotene. Of concern is that in the ATBC trial [39], there was a greater incidence of lung cancer observed in a subset of men who were smokers and who were taking vitamin A supplements. Similar observations were made in the ␤-Carotene and Retinol Efficacy Trial (CARET) [41] where the incidence of lung cancer was increased in patients who were smokers or who had a history of asbestos exposure and were on vitamin A supplementation. These observations underscore the potential hazards of consuming large amounts of antioxidants. It is also disconcerting that antioxidants such as vitamin C, vitamin E, ␤-carotene and selenium may hamper the beneficial effects of simvastatin and niacin on lipid profile [42]. Additional interventional trials [43–46] are listed in table 2. The most recently published study is that of Women’s Health Study [46]. In this study vitamin E supplements (600 mg every other day) did not protect healthy women against heart attacks, stroke or cancer. The preponderance of evidence in these interventional trials, unlike that of the observational trials, suggests that supplementation with conventional antioxidants, notably vitamin E, does not reduce the risk of CVD. However, the outcome of interventional studies may vary depending on the baseline nutritional status of the study population. Furthermore increased consumption of vitamins does not always translate into effective absorption and availability at critical cellular targets.

Antioxidant Supplementation: Pros and Cons There are several misconceptions regarding the benefit of supplementing diet with pharmacologic doses of antioxidants. A common misconception is that overweight people are well fed and are not at risk for micronutrient deficiencies. The lay public is under the impression that since supplements are natural products, they must be safe, and since they are available over the 118

Antioxidants and Diabetes Table 3. The arguments for and against the use of antioxidant supplementation For

Against

(a) Many do not consume a balanced diet

(a) Supplementation will encourage consumption of a poor diet (b) It may prevent heart disease, infection, (b) Body regulates micronutrient diabetes and cancer absorption and excretion (c) In small amounts it is harmless (c) The evidence for health benefits is small if any, and often contradictory (e.g. ␤-carotene and lung cancer) (d) No long-term toxicity data

counter, they must have the approval of the government agencies. Many also erroneously believe that antioxidant supplementation will achieve health benefits when conventional therapies fail. The arguments in favor of and against use of antioxidant supplementation are summarized in table 3. The fact that many, both in the industrialized world as well as in developing countries, do not consume a balanced diet that contains fresh fruits and vegetables, and the fact that antioxidants in small amounts appear to be safe, supplementation of the diet with modest amounts of conventional antioxidants may be justifiable. On the other hand, the fact that the body regulates the absorption and excretion of micronutrients, the concern about encouraging poor dietary habits and potential long-term toxicity and teratogenicity [3, 4], especially in certain groups of people, are arguments against routine supplementation of the diet with pharmacological doses of antioxidants. In a recent meta-analysis of 7 randomized trials of vitamin E and 8 trials of ␤-carotene treatment concluded that at the present time there is no convincing evidence for any beneficial effects of vitamin E or ␤-carotene on cardiovascular morbidity and mortality [47]. In this analysis, ␤-carotene led to a small but significant increase in all-cause mortality (7.4 vs. 7.0%, p ⫽ 0.003) and a slight increase in cardiovascular death (3.4 vs. 3.1%, p ⫽ 0.003) underscoring the concerns about the indiscriminate use of antioxidants in health maintenance.

Conclusions and Recommendations Diabetes mellitus is associated with increased tissue content of oxidation byproducts and reduced antioxidant defense system. Hyperglycemia appears to be a necessary and sufficient cause for inducing excessive production of ROS. Epidemiological studies have shown a correlation between dietary or supplemental intake of antioxidants and the reduced incidence of CVD. However interventional studies using select antioxidant supplements have failed to 119

Mooradian show significant benefits from supplementation and in some studies the potential for adverse outcomes has emerged. A common limitation in all the currently available interventional trials is the lack of measurements of the oxidative load of the study population, baseline micronutrient status and the lack of a marker of oxidation that can be monitored throughout the study to ascertain the bioavailability and efficacy of the supplements tested. At the present time given the inconsistencies in the studies available, the widespread supplementation with pharmacological doses of antioxidants should be discouraged. Future studies should focus on identifying reliable markers of oxidation to incorporate these measurements in clinical interventional trials, and to further develop novel antioxidants that reduce oxidative burst instead of functioning like conventional antioxidants that scavenge ROS after they have been produced.

References 1 Mooradian AD: Cardiovascular disease in type 2 diabetes mellitus: current management guidelines. Arch Intern Med 2003;163:33–40. 2 Brownlee M: The pathobiology of diabetic complications: a unifying mechanism. Diabetes 2005;54:1615–1625. 3 Hasanain B, Mooradian AD: Antioxidants and their influence in diabetes. Curr Diab Rep 2002;2:448–456. 4 Hasnain B, Mooradian AD: Recent trials of antioxidant therapy: what should we be telling our patients? Cleve Clin J Med 2004;71:327–334. 5 Horani MH, Haas MJ, Mooradian AD: Rapid adaptive down regulation of oxidative burst induced by high dextrose in human umbilical vein endothelial cells. Diabetes Res Clin Pract 2004;66:7–12. 6 Haas MJ, Horani MH, Parseghian SA, Mooradian AD: Statins prevent dextrose-induced endothelial barrier dysfunction possibly through inhibition of superoxide formation. Diabetes 2006;55: 474–479. 7 Horani MH, Haas MJ, Mooradian AD: Suppression of hyperglycemia-induced superoxide formation and endothelin-1 gene expression by carvedilol. Am J Ther 2006;213:2–7. 8 Jaiswal AK: Regulation of genes encoding NAD(P)H: quinone oxidoreductases. Free Radic Biol Med 2000;29:254–262. 9 Mooradian AD, Haas MJ, Wadud K: Ascorbic acid and ␣-tocopherol down regulate apolipoprotein AI gene expression in HepG2 and Caco–2 cell lines. Metab Clin Exp 2006;55: 159–167. 10 Wehmeier KR, Mooradian AD: Autooxidative and antioxidative potential of simple carbohydrates. Free Radic Biol Med 1994;17:83–86. 11 Mooradian AD: Micronutrients in diabetes mellitus; in Ioannides C, Flatt PR (eds): Drugs, Diet and Disease. Mechanistic Approaches to Diabetes. Englewood Cliffs, Prentice Hall, 1995, vol 2, pp 183–200. 12 Mooradian AD: Effect of ascorbate and dehydroascorbate on tissue uptake of glucose. Diabetes 1987;36:1001–1004. 13 Mooradian AD: The antioxidative potential of cerebral microvessels in experimental diabetes mellitus. Brain Res 1995;671:164–169. 14 Kushi LH, Folsom AR, Prineas RJ, et al: Dietary antioxidant vitamins and death from coronary heart disease in postmenopausal women. N Engl J Med 1996;334:1156–1162. 15 Stampfer MJ, Hennekens CH, Manson JE, et al: Vitamin E consumption and the risk of coronary disease in women. N Engl J Med 1993;328:1444–1449. 16 Rimm EB, Stampfer MJ, Ascherio A, et al: Vitamin E consumption and the risk of coronary heart disease in men. N Engl J Med 1993;328:1450–1456.

120

Antioxidants and Diabetes 17 Enstrom JE, Kanim LE, Klein MA: Vitamin C intake and mortality among a sample of the United States population. Epidemiology 1992;3:194–202. 18 Klipstein-Grobusch K, Geleijnse JM, den Breeijen JH, et al: Dietary antioxidants and risk of myocardial infarction in the elderly. The Rotterdam Study. Am J Clin Nutr 1999;69:261–266. 19 D’Odorico A, Martines D, Kiechl S, et al: High plasma levels of alpha- and beta-carotene are associated with a lower risk of atherosclerosis: results from the Bruneck study. Atherosclerosis 2000;153:231–239. 20 Dwyer JH, Paul-Labrador MJ, Fan J, et al: Progression of carotid intima-media thickness and plasma antioxidants: the Los Angeles Atherosclerosis Study. Arterioscler Thromb Vasc Biol 2004;24:313–319. 21 Will JC, Ford ES, Bowman BA: Serum vitamin C concentrations and diabetes: findings from the Third National Health and Nutrition Examination Survey, 1988–1994. Am J Clin Nutr 1999;70:49–52. 22 Mayer-Davis EJ, Bell RA, Reboussin BA, et al: Antioxidant nutrient intake and diabetic retinopathy: the San Luis Valley Diabetes Study. Ophthalmology 1998;105:2264–2270. 23 Ford ES, Will JC, Bowman BA, Narayan KMV: Diabetes mellitus and serum carotenoids: findings from the Third National Health and Nutrition Examination Survey. Am J Epidemiol 1999;149:168–176. 24 Losonczy KG, Harris TB, Havlik RS: Vitamin E and vitamin C supplement use and risk of all cause and coronary heart disease mortality in older persons: the Established Population for Epidemiologic Studies of the Elderly. Am J Clin Nutr 1996;64:190–196. 25 Gaziano JM, Manson JE, Branch LG, et al: A prospective study of consumption of carotenoids in fruits and vegetables and decreased cardiovascular mortality in the elderly. Ann Epidemiol 1995;5:255–260. 26 Salonen JT, Nyyssonen K, Tuomainen TP, et al: Increased risk of non-insulin dependent diabetes mellitus at low plasma vitamin E concentrations: a four year follow up study in men. BMJ 1995;311:1124–1127. 27 Knekt P, Reunanen A, Jarvinen R, et al: Antioxidant vitamin intake and coronary mortality in a longitudinal population study. Am J Epidemiol 1994;139:1180–1189. 28 Shoff SM, Mares-Perlman JA, Cruickshanks KJ, et al: Glycosylated hemoglobin concentrations and vitamin E, vitamin C, and beta carotene intake in diabetic and non-diabetic older adults. Am J Clin Nutr 1993;58:412–416. 29 Gey KF, Puska P: Plasma vitamins E and A inversely correlated to mortality from ischemic heart disease in cross-cultural epidemiology. Ann NY Acad Sci 1989;570:268–282. 30 Coudray C, Roussel AM, Mainard F, et al: Lipid peroxidation level and antioxidant micronutrient status in a pre-aging population; correlation with chronic disease prevalence in a French Epidemiological Study (Nantes, France). J Am Coll Nutr 1997;16:584–597. 31 Mayer-Davis EJ, Monaco JH, Marshall JA, et al: Vitamin C intake and cardiovascular disease risk factors in persons with non-insulin-dependent diabetes mellitus. From the Insulin Resistance Atherosclerosis Study and the San Luis Valley Diabetes Study. Prev Med 1997;26:277–283. 32 Sanchez-Lugo L, Mayer-Davis EJ, Howard G, et al: Insulin sensitivity and intake of vitamins E and C in African, Hispanic, and non-Hispanic white men and women: the Insulin Resistance and Atherosclerosis Study (IRAS). Am J Clin Nutr 1997;66:1224–1231. 33 Jacques PF, Taylor A, Hankinson SE, et al: Long-term vitamin C supplement use and prevalence of early age related lens opacities. Am J Clin Nutr 1997;66:911–916. 34 Stephens NG, Parsons A, Schofield PM, et al: Randomized, controlled trial of vitamin E in patients with coronary disease: Cambridge Heart Antioxidant Study. Lancet 1996;347:781–786. 35 Salonen JT, Nyyssonen K, Salonen R, et al: Antioxidant Supplementation in Atherosclerosis Prevention (ASAP) study: a randomized trial of the effect of vitamins E and C on 3-year progression of carotid atherosclerosis. J Intern Med 2000;248:377–386. 36 Blot WJ, Li JY, Taylor PR, et al: Nutrition intervention trials in Linxian, China: supplementation with specific vitamin/mineral combinations, cancer incidence, and disease specific mortality in the general population. J Natl Cancer Inst 1993;85:1483–1492. 37 Yusuf S, Dagenais G, Pogue J, et al: Vitamin E supplementation and cardiovascular events in high-risk patients. The Heart Outcomes Prevention Evaluation Study Investigators. N Engl J Med 2000;342:154–160. 38 GISSI-Prevenzione Investigators: Dietary supplements with n-3 poly unsaturated fatty acids and vitamin E after myocardial infarction: results of the GISSI-Prevention Trial. Lancet 1999;354:447–455.

121

Mooradian 39 The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group: The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. N Engl J Med 1994;330:1029–1035. 40 Heart Protection Study Collaborative Group: MRC/BHF Heart Protection Study of antioxidant vitamin supplementation in 20,536 high-risk individuals: a randomized placebo-controlled trial. Lancet 2002;360:23–33. 41 Omenn GS, Goodman GE, Thornquist MD, et al: Risk factors for lung cancer and for intervention effects in CARET: the Beta Carotene and Retinol Efficacy Trial. J Natl Cancer Inst 1996;88:1550–1559. 42 Brown BG, Zhao XQ, Chait A, et al: Simvastatin and niacin, antioxidant vitamins, or the combination for the prevention of coronary disease. N Engl J Med 2001;345:1583–1592. 43 de Gaetano G, Collaborative Group of the Primary Prevention Project: Low dose aspirin and vitamin E in people at cardiovascular risk: a randomized trial in general practice. Lancet 2001;357:89–95. 44 Lee IM, Cook NR, Manson JE, et al: Beta-carotene supplementation and incidence of cancer and cardiovascular disease: the Women’s Health Study. J Natl Cancer Inst 1999;91:2102–2106. 45 Hennekens CH, Buring JE, Manson JE, et al: Lack of effect of long-term supplementation with beta carotene on the incidence of malignant neoplasms and cardiovascular disease. N Engl J Med 1996;334:1145–1149. 46 Lee IM, Cook NR, Gaziano JM, et al: Vitamin E in the primary prevention of cardiovascular disease and cancer: the Women’s Health Study: a randomized controlled trial. JAMA 2005;294: 56–65. 47 Vivekananthan DP, Penn MS, Sapp SK, et al: Use of antioxidant vitamins for the prevention of cardiovascular disease: meta-analysis of randomized trials. Lancet 2003;361:2017–2023.

Discussion Dr. Schiffrin: You have mentioned different antioxidants such as vitamin C and vitamin E. What about the status of other intracellular antioxidants such as glutathione? Is it possible that giving glutathione precursors, such as cysteine or proteins rich in cysteine, can improve the glutathione cellular status? This is my first question. What about the administration of arginine? In fact arginine as a substrate of endothelial nitric oxide synthase and a precursor of nitric oxide may well modify the endothelial status. My last question is whether the redox status in diabetes in addition to altering endothelial function could also have an influence on the immune system such as the level of macrophage activation that in turn may increase the inflammatory stress in this type of patient. Is the glutathione status of these patients known? Dr. Mooradian: Glutathione is a very important antioxidant and indeed there are some studies showing that glutathione stores might be reduced in diabetes, especially in uncontrolled diabetes, but these are small studies. There are no studies showing that interventions or replacement with glutathione or its substrates, precursors, would have an impact on any long-term outcome. But the clinicians in the room would know that those intermediates have been used to protect the kidney from other damage during the administration of contrast agents and drug-related hepatic disease, which is another disease entity that involves oxidative stress. So at the moment the short answer is that glutathione precursors are not involved or have not been used in interventional trials to show any outcome measures. Your second question about arginine and endothelial nitric oxide synthase is interesting because it does have some beneficial effects on endothelial cell function, but also remember that one of the sources of oxidative stress is nitric oxide and proteins are modified with nitric oxide. So there is a fine balance whether you are going to achieve a favorable effect with supplementation or not. By the way arginine and a lot of biochemical molecules, including simple molecules such as creatinine or uric acid, all have very important

122

Antioxidants and Diabetes antioxidant activity. In that sense it may have some additional effect as an antioxidant. Again, there are actually no interventional trials on the immune system and macrophage activation, but there are experimental data on the role of macrophages in promoting atherosclerosis, and obviously that is highlighted by the story on myeloperoxidase as a key enzyme in promoting atherosclerosis. I am not aware of any interventional trial targeting myeloperoxidase to prevent premature atherosclerosis. But probably that will come eventually because it is a very hot area of investigation. Dr. Halimi: Could you say some words regarding the zinc status, the tissue contents of zinc in human diabetic patients? With my group I did some studies on this topic in rats with insulin resistance induced by a high fructose diet. Zinc plasma and tissue concentrations were found to be very low, furthermore zinc supplementation was able to normalize insulin sensitivity in these animals. What is your opinion about the loss of zinc in diabetic patients, perhaps in urine partly due to glycosuria, diuretics? Dr. Mooradian: There is a lot of literature on the issue of zinc and as Dr. Halimi mentioned, diabetic individuals, especially those with uncontrolled hyperglycemia and those on diuretic therapy, tend to have excessive excretion of zinc in the urine. But when the zinc stores are measured, be it in plasma or intracellularly as in white or red cells, it is highly variable; sometimes it is normal, sometimes it is high, sometimes it is low, so there is really no consistent effect. One of the reasons is that zinc metabolism is a good example of what I mentioned in terms of the organism protecting itself against micronutrient deficiency. Once there are losses of zinc in the urine, the gut absorption of zinc is multiplied, it increases several fold. Zinc absorption in diabetic individuals is actually quite high for various reasons. One is that the transporters are upregulated, and also the intestinal mucosal area in diabetes, especially in uncontrolled diabetes, is vastly expanded. They tend to absorb zinc much more efficiently than an average individual; so even though they are losing zinc they are also absorbing zinc quite a bit, and therefore they may not have any zinc deficiency. The problem with zinc deficiency or the determination of zinc deficiency is that we really don’t have reliable methods of measuring zinc. Certainly measuring the elemental zinc in plasma or intracellular zinc may not be the best indicator. There are some new methods now that enable estimation of zinc status by measuring certain gene products that are modulated by zinc, but again they have not really caught on in clinical practice. Zinc is an important element in terms of diabetes management, but it doesn’t seem that easy to produce zinc deficiency. Dr. Eshki: Do you believe that controlling for upper levels or recommended allowances on both macro- and micronutrients is essential before looking at specific oxidants and their effect on diabetes? Dr. Mooradian: Are you asking about daily recommended intakes (DRIs) and the variety of nomenclatures in that regard, whether it is recommended dietary allowances or dietary guidelines and dietary recommended ranges? Your point was to first look at the dietary intakes and then see if supplementation on top is useful? Dr. Eshki: The reason I am asking is because normally when dealing with trauma patients under stress, some physicians ask me why I don’t use zinc because they heard that zinc is good at this point. What I tell them is that for me the optimal is to reach the DRI for the patient, and then look at other factors. But when we say, alright, start this way, it is going to become more confusing because we are going to end up with a bag of supplements. Dr. Mooradian: Yes, I understand what you are saying and I fully subscribe to that. In terms of limiting supplementation or intake to the DRIs rather than trying large amounts of micronutrients and supplements. I think at the moment that would be the most rationale thing to recommend. Dr. Tong: I noticed that a lot of dietary recommendations do suggest that patients with cardiovascular disease increase their intake of vegetables and fruits. One of the

123

Mooradian reasons is because it has been claimed that they can get antioxidant products from vegetables and fruits. Are you aware of any study that has looked at the antioxidative capacity in patients taking different proportions of vegetables or different amounts of fruits? Dr. Mooradian: Again the optimal way of getting antioxidants is to consume good looking and good tasting food, that would be vegetables and fruits, and I agree with that recommendation. But in terms of evidence in interventional trials, we have the same problems we were discussing this morning and yesterday that it is very hard to implement an interventional trial with a dietary change and sustain that dietary change over a prolonged period of time in a large number of individuals to be able to show that a particular dietary intervention is effective. At the moment I suspect that it should remain a commonsense approach, a consensus approach. I think people have to emphasize the importance of acquiring antioxidants through dietary means and healthy food choices. But it doesn’t mean that we should not be continuing to look for effective antioxidants in pill form, and if it has any health protective effect that will obviously be a welcome addition to the diet. Dr. Tong: Vitamin E, e.g. some antioxidants from vegetables or natural products, are they going to be different in terms of the isoforms compared to pharmaceutical preparations of vitamin E? Dr. Mooradian: I suspect that it is probably true that the antioxidants in food are different from the more purified elements that are available in pill form from the pharmaceutical industry. The antioxidants in food are very complex since a whole series of compounds has variable degrees of antioxidant activities but is a very rich mixture of compounds. When you select one compound as the candidate element to introduce as a supplement, you are at risk of either choosing the one that may not have as much efficacy as the others when they are all in conjunction, or the one that may end up having some toxicity as well. I don’t want to discourage people from pursuing and looking into specific components of the diet that have important antioxidant activities that may be useful pharmaceutically. Dr. Hill: The large epidemiological studies provide one way to look at this issue, but is anyone doing the clinical trials with fewer numbers of subjects and with more control, where the oxidative state is measured with interventions such as with increased fruit and vegetable consumption or supplements? Dr. Mooradian: It should be done, but as yet it has not been done and that is the major problem in the field. We have been doing study after study with very primitive tools and the results are not always reliable, so we need to have better clinical trials designed. Dr. Chiasson: All these interventional trials are definitely very disappointing and as a matter of fact there is only the CHAOS study, which is questionable as you mentioned. You did mention that one of the problems is that the antioxidant capacity of the subjects was not measured. But do you really think that that is the problem? Why do you think that all these interventional studies are totally negative? Dr. Mooradian: I suspect that there are various potential reasons: one is that the right antioxidants are not being used, but conventional antioxidants are. As I mentioned these are interceptors or scavengers rather than real agents that reduce oxidative stress. It is possible that perhaps the dosing is inadequate, as we have no idea what the appropriate dose will be for these conventional agents to have an effect on oxidative stress. The third potential limitation is that this is a process that takes years and years to develop, and these studies, usually in the range of 4, 5, 6 years, may not be long enough to show the benefit of antioxidant supplementation. What really is needed are studies that go on for 15, 20 years to see an effect. So that is a very difficult point to prove or disprove.

124

Antioxidants and Diabetes Dr. Jianqin Sun: I have three questions. What is the role of the phytochemical substances in antioxidative stress? There are many supplements in either a natural or an artificial form. Which do you think is better; is one better than the other? And the third question, it is quite common that many people take supplements. Should supplements be taken for a long time or for a short time? Dr. Mooradian: The first question is about phenols and naturally occurring antioxidants which are extremely important, there is no question about that, and there are a lot of them. The question is are they effective in the long run: I really don’t know. People have been promoting the use of phenol-containing products, and obviously one of my favorite sources of that would be red wine. Again, there is no clear evidence though that it really makes a big difference. There is a lot of evidence in experimental models in cell cultures preventing oxidation. The other issue of whether the antioxidant should come from natural sources or from pharmaceutical sources: I really don’t know the answer. My naïve thinking is that it should not make a difference. If you have an effective compound, whether it is packaged in the food or packaged in a pill, it should not make a difference. The problem is we don’t know what that effective compound is, so that is where the problem arises. The third question about the duration of treatment is even more difficult. One of the criticism of these interventional trials is that they have not been long enough. So if I see people who are taking supplements, what I do is to ask them to try to limit the supplements to within what I consider to be a reasonably safe margin. In other words if they are taking 1,000 mg vitamin E, I ask them to cut it back to less than 400 mg/day and negotiate a happy medium. The same with carotene and certainly with the retinol. Retinol certainly should not exceed 10,000 or even 8,000 units/day. This is especially an issue for women of reproductive age because retinol in high concentrations is associated with significant teratogenicity. Those women who are taking supplements and planning a pregnancy have to be careful to avoid that. They should be on folate and calcium supplements, possibly a little bit of iron supplementation, but not high concentrations of retinol. Dr. Bantle: In North America the most commonly taken antioxidant supplement is vitamin E, As a participant in a physician’s health study, there is one chance in two I personally am taking it. So I would like to ask what your thoughts are about the recent meta-analysis suggesting that vitamin E may actually increase mortality rates. Dr. Mooradian: You have raised a very important issue that mortality appears to be increased with vitamin E use in this recent meta-analysis. Having said that, I have a very negative bias against meta-analyses in general. I have always said that metaanalysis is to analysis as metaphysics is to physics, and it is really very difficult to compile the data. I understand there are ways of improving your analysis if you are careful and take all the factors into consideration. But nevertheless meta-analyses are like observational trials, they are very good in terms of generating hypotheses. Again it raises the spectra that perhaps indiscriminate use of antioxidants is not the wisest thing to do, and it brings to the forefront the potential that there might be some toxicity involved with the use of very large doses of antioxidants. But I would not drop out of the study because of that meta-analysis; just continue taking your vitamin E, I guess, if you are taking it already.

125

Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 127–137, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

Dietary and Body Weight Control: Therapeutic Education, Motivational Interviewing and Cognitive-Behavioral Approaches for Long-Term Weight Loss Maintenance Alain Golay Service of Therapeutic Education for Chronic Diseases, Diabetes, Obesity, Department of Community Medicine, University Hospital of Geneva, Switzerland

Abstract A diet always induces weight loss in the short term. The loss does not depend on the dietary composition but rather on the caloric deficit. However, a drastic diet often induces binge eating disorders and can lead to a weight gain in the long term. A cognitivebehavioral-nutritional approach allows lasting weight loss and best results with low fat diets in the long term. Therapeutic education is a patient-centered humanistic approach which allows patients to be actors in their own treatment and own diet to improve their success in losing weight and their quality of life. Motivational interviewing and cognitive-behavioral approaches are perfect complements to therapeutic education for long-term weight loss maintenance. Finally, the best diet is the one that the patient can follow in the long term. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

Which Diet to Recommend? Low Carbohydrate or Low Fat? Is the choice of diet really important? A number of controlled studies [1–3] have all proved the same effectiveness, when the caloric intakes are similar, with a balanced diet, a diet poor in carbohydrates and a dissociated diet, over a 6-week or 6-month period (fig. 1). Over 24 weeks, Samaha et al. [4] compared low carbohydrate and low fat diets in 132 obese patients with a body mass index of 43 kg/m2 and a high prevalence 127

Golay

Low carbohydrates

Balanced

Dissociated

Low in carbohydrates 30% Rich in fat 45% Sufficient in protein 25%

Rich in carbohydrates 45% Low in fat 30% Sufficient in protein 25%

Rich in carbohydrates 45% Low in fat 30% Sufficient in protein 25%

Weight loss

Weight loss

Weight loss

7.8 ⫾ 0.5 kg

7.2 ⫾ 0.6 kg

6.7 ⫾ 0.5 kg

Fig. 1. Diets of different compositions and weight loss. Adapted from Alford et al. [1] and Golay et al. [2].

Weight loss (kg)

0 ⫺2

Low fat

⫺4

Low carbohydrate

⫺6 ⫺8 ⫺10 ⫺12 0

6

12

Time (months)

Fig. 2. Low fat and low carbohydrate diets and weight loss. Adapted from Golay et al. [3] and Samaha et al. [4].

of diabetes (39%) or metabolic syndrome (43%). The patients in the low carbohydrate group (maximum 30 g carbohydrates/day) lost significantly more weight (p ⬍ 0.002) than those in the low fat group (500 kcal/day reduction of total energy intake, with a maximum of 30% fat/day). However, Samaha’s study was then supplemented by a follow-up of the same patients over 48 weeks [5]: no significant weight loss difference was found between the 2 groups (fig. 2), nor any lipid modification. How to Maintain Weight Loss? In the National Weight Control Registry in the United States [6], the maintenance of a 5.5-kg weight loss over 5.5 years was achieved through the 128

Diet and Obesity combination of a balanced diet and sustained physical activity (40 min/day). Another study [7] confirms that combining a diet and exercise is more effective than dieting alone. Moreover, physical activity can equally help maintain the weight lost [8, 9]. In addition, it is recognized today that diets poor in fats are more effective in maintaining weight loss over a long period of time [10]. On the contrary, a diet rich in fat induces overconsumption and weight gain through a feeling of non-satiety and the highly caloric content of fats [11]. Overall, patients who have undertaken repeated diets quickly regain the kilos they have lost. When facing the yo-yo phenomenon, eating disorders should be investigated. This is associated with a feeling of restriction, nibbling or compulsions, and has the effect of making patients feel guilty and reducing their self-esteem. Behavioral work, combined with a hypocaloric diet and a program of physical activity, allows weight loss to be maintained [12]. It involves only a small loss of weight but helps alleviate eating disorders, depression and anxiety. A prospective study with a 5-year follow-up [13] confirms that this multidisciplinary approach is entirely beneficial in the long term. Fifty percent of patients maintained their weight loss, and even continued to lose weight. Lately, Wadden et al. [14] confirmed that the combination of medication and group lifestyle modification resulted in more weight loss than either medication or lifestyle modification alone. In conclusion, there is no miracle diet, whatever the current fashion would have us believe. The best diet is one that the patient can follow in the long term. Therefore, it is necessary to draw up dietary advice which is adapted to the individual, combined where possible with teaching about physical activity and, perhaps, cognitive-behavioral therapy. It is important to rule out diets which are too strict and restrictive if we are to avoid eating disorders and weight regain. However, the 3F formula – Fat-Free and Fit – is known to be the most effective approach in the long term for all our patients.

Therapeutic Patient Education to Improve Diet Compliance Therapeutic education is a ‘patient-centered’ humanistic approach which not only allows patients to be actors in their own treatment and diet to improve their quality of life and success in losing weight, but also to reduce potential complications. Therapeutic education of obese and diabetic patients has been carried out successfully for a number of years [13–16]. Its efficiency has been demonstrated on a number of occasions and it has been adapted to other chronic illnesses such as chronic bronchopneumopathy [17], sleep apnea syndrome [18, 19], cardiovascular illnesses [20], etc. This branch of medicine assumed a 129

Golay Therapeutic patient education Work on relapses Transmit Use errors

Knowledge Motivate

Reinforce success

Patient centered

Share the decision

Negotiate

Skills

Behavior

Fig. 3. Therapeutic patient education.

definite identity with the publication of the first official definition in a WHO expert report [21].

Can Therapeutic Patient Education be Therapeutic? The first effect of therapeutic education (fig. 3) is an improvement in patients’ quality of life. Healthcare professionals too often tend to forget this; their overriding aim is to improve therapeutic compliance and reduce complications. With therapeutic education, an 80% reduction in amputations linked to diabetes has been observed [22]. Increasing patients’ knowledge is not the only function of therapeutic education; in fact, its main aim is to make them aware of their problem (whether this involves diagnosis or risk factors) so that they take these notions on board and act more closely on the doctor’s recommendations. That is when the doctor’s teaching becomes ‘therapeutic’; not only does he/she increase patients’ knowledge and improve his/her quality of life, but he/she guides them through the long-term treatment of their problem. Thus, the patients become co-therapists. Therapeutic education not only aims to increase patients’ skills and knowledge, but also to modify their behavior in the long term. Furthermore, its pedagogical approaches are complemented by psychosocial models and by cognitive-behavioral type psychological approaches [12]. For more than 10 years 130

Diet and Obesity

Trigger Frustration at work

Automatic negative thought I’m not good incapable

Consequences

Emotion

Weight gain decompensate the diabetes Guilt

Anger

Behavior Food binge

Fig. 4. Food binges – the vicious circle.

now, these psycho-pedagogical approaches have contributed to a clear improvement in the effectiveness of therapeutic education [23].

Therapeutic Patient Education of Obese Patients Much of our experience has come through the treatment of type-2 diabetic patients suffering from obesity [24]. More than 80% of our type-2 diabetic patients are overweight, very often due to dietary behavior of a binge-eating disorder. The usual nutritional approach is insufficient, since even ideal teaching of dietary knowledge, and the fat contents of foods does not bring about a lasting loss of weight. Even multiplying the number of practical teaching exercises on food management, shopping habits, slimming recipes, etc., only brings a very slight improvement in long-term results. The cognitive-behavioral approach has allowed us to make a great step forward in our interdisciplinary approach and to maintain behavioral changes in the long term [12, 24, 25]. The cognitive-behavioral approach is above all interested in the mechanisms behind the triggers of binges and emotions, as well as automatic negative thoughts (I’m no good, ugly, incapable). The final step of what is called a functional analysis (fig. 4) of a binge episode consists of helping patients to find their own strategies and evaluate the positive consequences arising from them. The functional analysis (fig. 4) of a food binge allows patients to break the vicious circle. For example: frustration at work (trigger) associated with anger (emotion) and negative thought, such as ‘I’m silly to let myself be 131

Golay

Trigger Frustration at work reduced by stress management, negotiation and self-affirmation

Thought Less negative

Consequences

Emotion

Loss of weight stabilized diabetes less guilt

Less anger

Behavior Replacement pleasure

Fig. 5. The virtuous circle following treatment.

insulted’, will induce a binge disorder as usual strategy: ‘I get stuck into food as soon as I get home’. Finally, the consequences of this vicious circle are weight gain, bad glycemic control, a lot of guilt, and even more frustration. By analyzing the flow from trigger to consequence, patients will gain a better understanding of the underlying mechanisms involved. They will move progressively from a vicious circle into a virtuous circle (fig. 5). For example, frustration at work will be reduced by patients learning to be more assertive, to say no politely, to step back, to express their emotions, etc. Expressed emotions will grow less painful and thoughts will become less negative. Meanwhile, replacement strategies for binges will be found which will become both rewarding and pleasurable. As a result, food binges will decrease in number. Finally, the consequences for quality of life and biological results will become more tangible and increasingly felt by the patient.

How to Motivate Patients Too often, poor adherence to treatment is due to the lack of motivation. This is very easy to say, but it is very difficult for healthcare professionals to put themselves in question and realize that they are also partly responsible for patients’ poor compliance! A motivational interview requires certain psycho-pedagogical skills which Miller and Rollnick [26] have proposed and described in full. In addition, clearly defined principles as well as an appropriate attitude and techniques exist for these motivational interviews. 132

Diet and Obesity Motivational Interviewing Motivational interviewing to change a behavior can be used in all therapeutic situations where motivation is central to the change process and where ambivalence exists. Firstly, Miller and Rollnick [26] insist that this interview is a ‘way of being’ with patients and not a varied collection of psychological techniques used to get people to do what they do not want to. The interview aims at exploring patients’ ambivalence in a semi-structured and nonjudgmental manner. During the motivational interviewing, healthcare professionals select information through reflective listening, use the dissonance in the conversation and facilitate change. As such, the principles of the motivational interviewing are above all a Rogerian-type [27] empathic communication which aims to help elaborate patients’ discrepancies and reinforce their feelings of personal efficacy. Another interesting principle involved here is that of rolling with patients’ resistance when they are ambivalent. A typical ambivalence in our obese patients is that they do not follow their dietary plan when they know perfectly well what they have to do to lose weight. Hence, a discrepancy occurs between a value and behavior. The objective of healthcare professionals is to bring patients to consider another point of view and therefore become actors in their own decision-making: ‘What is the least they can change at the lowest psychological cost and greatest benefit?’ A major principle involved in motivational interviewing is reinforcing patients’ personal efficacy and increasing their faith in change. In conclusion, therapeutic education is a patient-centered humanistic approach which allows patients to be actors in their own treatment, to improve their quality of life, their success of weight loss and reduce the risk of potential complications. The motivational interviewing and cognitive behavioral approaches are perfect complements to therapeutic education for longterm weight loss maintenance.

References 1 Alford BB, Blankenship AC, Hagen RD: The effects of variations in carbohydrate, protein and fat content of the diet upon weight loss, blood values and nutrient intake of adult obese women. Am J Diet Assoc 1990;90:534–540. 2 Golay A, Allaz AF, Morel Y, et al: Similar weight loss with low or high-carbohydrate diet. Am J Clin Nutr 1996;63:174–178. 3 Golay A, Allaz AF, Ybarra J, et al: Similar weight loss with low-energy food combining or balancing diets. Int J Obes 2000;24:492–496. 4 Samaha FF, Iqbal N, Seshadri P, et al: A low-carbohydrate as compared with a low-fat diet in severe obesity. N Engl J Med 2003;348:2074–2081. 5 Stern L, Iqbal N, Seshadri P, et al: The effects of low-carbohydrate versus conventional weight loss diets in elderly obese adults. Ann Intern Med 2004;140:778–785. 6 Klem ML, Wing RR, McGuire MT, et al: A descriptive study of individuals successful at longterm maintenance of substantial weight loss. Am J Clin Nutr 1997;66:239–246.

133

Golay 7 Clinical Guidelines on the Identification, Evaluation and Treatment of Overweight and Obesity in Adults – the Evidence Report. Obes Res 1998;6(suppl 2):51S–209S. 8 Meyers AW, Graves TJ, Whelan JP, et al: An evaluation of television-delivered behavioral weight loss program: are the ratings acceptable. J Consult Clin Psychol 1996;64:172–178. 9 Pavlou KN, Krey S, Steffee W: Exercise as an adjunct to weight loss and maintenance in moderately obese subjects. Am J Clin Nutr 1989;49:1115–1123. 10 Klem ML, Wing RR, McGuire MT, et al: A descriptive study of individuals successful at longterm maintenance of substantial weight loss. Am J Clin Nutr 1997;66:239–246. 11 Golay A, Bobbioni E: The role of dietary fat in obesity. Int J Obes Relat Metab Disord 1997;21(suppl 3):S2–S11. 12 Fossati M, Amati F, Painot D, et al: Cognitive-behavioral therapy with simultaneous nutritional and physical activity education in obese patients with binge eating disorder. Eating Weight Disord 2004;9:134–138. 13 Golay A, Buclin S, Ybarra J, et al: New interdisciplinary cognitive-behavioral-nutritional approach to obesity treatment: a 5-year follow-up study. Eating Weight Disord 2004;9: 29–34. 14 Wadden TA, Berkowitz RI, Womble FG, et al: Randomized trial of lifestyle modification and pharmacotherapy for obesity. N Engl J Med 2005;353:2111–2120. 15 Golay A, Bloise D, Maldonato A: The education of people with diabetes; in Pickup J, Williams G (eds): Textbook of Diabetes, ed 2. Oxford, Blackwell, 2002, chapt. 38, pp 1–13. 16 Maldonato A, Segal P, Golay A: The diabetes education study group and its activities to improve the education of people with diabetes in Europe. Patient Educ Couns 2001;44: 87–97. 17 Bourbeau J, Julien M, Maltais F, et al: Reduction of hospital utilization in patients with chronic obstructive pulmonary disease: a disease-specific self-management intervention. Arch Intern Med 2003;163:585–591. 18 Engleman HM, Wild MR: Improving CPAP use by patients with the sleep apnea/hypopnea syndrome (SAHS). Sleep Med Rev 2003;7:81–99. 19 Golay A, Girard A, Grandin S, et al: A new educational program for patients suffering from sleep apnea syndrome. Patient Educ Couns 2005, in press. 20 Eriksson S, Kaati G, Bygren LO: Personal resources, motives and patient education leading to changes in cardiovascular risk factors. Patient Educ Couns 1998;34:159–168. 21 WHO Regional Office for Europe: Therapeutic Patient Education, Continuing Education Programs for Healthcare Providers in the Field of Prevention of Chronic Diseases. Report of a WHO Working Group. Copenhagen, WHO, 1998. 22 Assal JP, Albeanu A, Peter-Riesch B, et al: The cost of training a diabetic patient: effects on prevention of amputation. Diab Metab 1993;19:491–495. 23 Giordan A, Golay A, Jacquemet S, et al: Communication thérapeutique. L’impact d’un message dans le processus d’apprendre. Psychothérapies 1996;16:189–193. 24 Golay A, Volery M, Rieker A, et al: Approche cognitivo-comportementale; in Basdevant A, Guy-Grand B (eds): Médecine de l’obésité. Paris, Médecine-Sciences Flammarion, 2004, pp 246–252. 25 Fossati M, Rieker A, Golay A: Thérapie cognitive en groupe de l’estime de soi chez des patients obèses, un nouvel outil: la fleur de l’estime. J Thér Comportement Cogn 2004;14:29–34. 26 Miller W, Rollnick S: Motivational Interviewing: Preparing People for Change, ed 2. New York, Guilford Press, 2002. 27 Rogers C: Counseling and Psychotherapy: Newer Concept in Practice. Boston, Houghton Miffling, 1957.

Discussion Dr. Chiasson: I would like to come back to the sympathetic nervous system response. It is very interesting because it is one of the reasons why when an extremely low diet is used it doesn’t work. What is the mechanism; how is that regulated? How do you measure that, and what do you think is the problem with the autonomous nervous system?

134

Diet and Obesity Dr. Golay: We have two different techniques. The first one is laser Doppler measurement of the microvascular in the finger and the second one is heart rate monitoring. We have 24-hour heart rate monitoring with a spectral analysis. For the second question, the reason for this unbalance is the many questions behind it. It is known that the sympathetic nervous system is higher in obese patients. Thus, obese patients cannot further increase their system, and then they have no possibility of increasing their fat oxidation. So we have many hypotheses behind this; we are also measuring different hormones. A new one is apelin which can connect the periphery and the heart and perhaps the brain. But this is a new area and is quite interesting. Dr. Katsilambros: I would like to make a comment on the question of Dr. Chiasson. We also had the possibility of measuring heart rate variability in obese and non-obese people under different circumstances. In the last years we have published some papers showing that after a meal the activation of the heart sympathetic nervous system is different depending on whether a person is obese or non-obese. If a person is obese the activation of the heart sympathetic nervous system is low as compared to a lean control person. In addition, if obese and lean people are given fat meals, you observe that they do not react at all, but there is a very slow and nonsignificant increase in the heart sympathetic activity. But if you give an isoenergetic carbohydrate meal to these persons then there is a clear-cut activation of the heart sympathetic activity in the lean people which, however, is not present in the obese people. Dr. Golay: I know your studies and we are doing the same thing. Carbohydrate and proteins are very important for increasing the sympathetic nervous system, but this is not the case for fat. It is even worse with saturated fat. And as we discussed before, n-3 and n-6 are very much better for increasing the sympathetic nervous system. So the key answer is in the fat content of the diet. Dr. T. Wilkin: Can I clarify in relation to spectral analysis? Is it a feature of obesity or is it a feature of insulin resistance that variability is lost? In other words, are there individuals who have a distribution of body fat which is not cause of insulin resistance? Dr. Golay: It is an interesting question. I am a diabetologist and the first measurement I made was in diabetic patients. However, in insulin-resistant patients, we also found a defect. Now we are looking at children from obese patients and I cannot give the data today, but my feeling is that it is probably before insulin resistance or diabetes. Today, it is only a hypothesis. Dr. Zhao: We know that in general vegetarian animals are less aggressive than meat-eating species. So I am just curious if there could be a relationship between the food consumed and the emotional and behavioral responses. So my question is, are there any papers available to show this association? Dr. Golay: I totally agree with you but I don’t have any data showing that hypothesis. We are smiling about this but it is probably true, especially with high sugar. Patients are probably more active and more excited if they are more sugar free. But we need a lot of research in this field. I am sure nutrition is related to emotion, perhaps to depression, etc.; when you have a lack of tryptophan you have a lack of serotonin and then depression. In some studies we have tried to improve the tryptophan intake, and to decrease other proteins. The humor and depression are improved and especially also sugar craving. So a lot of research is needed in this field. Nutrition is a completely open field of research. Dr. Hill: I have two questions about your energy economy concept. First it was done in bariatric patients and there is a lot of speculation that with the surgery you get some changes in the gut hormones which in fact can affect metabolism. Have you found the same results in weight loss produced without bariatric surgery? And the second question is that it looks as though after a year the expected and the actual energy expenditure are the same, suggesting that the energy economy may be more important for weight loss, not so much for weight loss maintenance.

135

Golay Dr. Golay: You are totally right. As to the second question, the energy economy is much smaller at the beginning, and after a weight loss program it is around 200 cal. When you lose weight you also decrease your total energy expenditure so the energy economy at that time for weight maintenance is even more crucial and important. As to your first question, of course we were looking for a normal body weight loss program and we had the same type of results. However, it is more difficult because the variation is bigger. With the bypass weight loss program, we are quite confident with our results from this type of surgical operation. We have at least the same size of stomach and almost the same size of meals, and thus the variation is smaller. One question still remains open: ghrelin suppression after this type of bypass. Dr. T. Wilkin: One of the observations of that energy economy is the reduced conversion peripherally of thyroxin to triiodothyronine, from T4 to T3, which can be quite a striking change for a period of time. Is there any therapeutic potentially possible and advantageous intervention to prevent that occurring? Dr. Golay: Concerning the first comment, it is probably true that the conversion between T4 and T3 is one of the reasons. The second reason is also thermogenesis pathways. What the treatment should be is difficult to say at this time, but definitely not thyroid hormones. We need different studies with leptin replacement. The leptin concentration really drops during a weight loss program in this kind of economy phase, but we were not successful. We have other possibilities, but we have to understand that weight loss during the first months is almost the same for everybody, but the difference is between 3 and 12 months for the second phase, the slow phase. And physical exercise is crucial in this phase. Dr. Ott: I have a question regarding Singapore. As you know Singapore performs nutritional surveys at regular intervals to assess the health status of its population. As a result of this, using television, radio and so on, they recommend consuming four servings of fruits and vegetables per day and physical exercise. They also implement school programs. Do we have any evidence that this kind of activity is effective? I believe the prevalence of diabetes is low in Singapore. Do you have any data? Dr. Golay: In fact, we are doing this kind of program in Geneva today. There are many dieticians in schools, so every teenager can receive this kind of advice, which is part of the program now in schools. We are also trying to have a program for physical exercise paid for by the government. The results are not yet known but I am quite confident that it is the right way to go because we have to start as early as possible. Just before we were talking about economy. We published two papers recently on economy in Switzerland [1, 2]. 98% of our costs for obese patients are due to complications; we paid only 1.5–2% for an obesity program, which is really nothing. We published another paper where we were looking for the effect of weight loss in diabetic patients [3]. We pooled some data with Swedish friends and were able to prove that EUR 14,000/patient/year is saved when a diabetic patient loses 9 kg. It costs a huge amount of money to treat diabetes, myocardial infarctions and all these kind of complications. Dr. Haschke: I am not familiar with the situation in Asia but in Germany the same recommendation has been on the table for several years; i.e. to eat 4–5 servings of vegetables and fruits per day. Two years ago there was the first big population survey on public awareness of the recommendation, how people feel about it, and how they adhere to it. The awareness was very high; 80% of the population knew of this recommendation, but only 5% adhered to it. So I cannot comment on the Asian populations but in Germany it is rather difficult to move forward. Perhaps the recommendations should be phrased in a different way. Dr. Golay: This is exactly what we think. The information is not enough, we should do more than that. We need really to implement behavior courses. Everybody knows that it is very difficult to put into practice. It is like smoking, if you have a lot of people

136

Diet and Obesity smoking around you, you are more prone to smoke and it is difficult to stop. So I think we need time, we need more publicity, we are fighting against big companies like McDonalds and Coca Cola. Dr. Wuersch: Can you comment on your experience with the low glycemic index foods in your treatments? Dr. Golay: There was a big discussion yesterday. For me, the low glycemic index is a good tool, at least for diabetic patients. However, today we are more convinced that it is also helping satiety, and by giving snacks to our obese patients suffering from binge eating disorders (allowed snacks, not nibbling), it is working very well for satiety. When a snack is given they eat less at the next meal. This is important to have a structure for your meals. Dr. L. Wilkin: I want to talk about children for a moment. I am wondering whether it is easier or harder for children to maintain a healthy weight than adults. Perhaps there are a fewer children with so-called eating disorders, binge eating and so on, but maybe children are more susceptible to our obesogenic environment. Would you like to comment? Dr. Golay: It depends on age. For teenagers I think it is easier because you can put a teenager in a certain position to be against the parents, to eat differently from their parents, so it is much easier. But for children, they eat in the family so we need to change the eating habits and behavior of a child with the whole family. That is why we need to go for nutrition courses in school, and to bring the parents into this kind of course. But I know it is a very difficult task. Dr. Eshki: Weight loss programs and weight loss studies have been focusing on the efficacy of the diet alone and not on the safety of the diet. The safety parameters I am referring to are the daily recommended intakes. Don’t you think this missing margin may have big effects on the results of your study and other similar studies in the long-term? Dr. Golay: The safety of a diet for me is the amount of protein, and first you should have enough protein. You can find very strange diets without any proteins and the patients lose lean body mass. In terms of safety, restrictive diets should not be given. Dr. Eshki: When you are talking about protein, and it is used a lot, and you want to reduce or increase it, you can’t ignore the calcium intake for instance because it could cause osteoporosis in the future. So what I am trying to say is that we should not take a narrow view but try to look at the micronutrients also. That is what I mean by safety. Dr. Golay: For the last 5 years we have been doing some studies on micronutrients. I think it is a really interesting field. In obese patients I always found a deficit for oligo elements, for vitamins, for different types of fats. And one of our suggestions today is to supplement n-3. I think in the near future we will have to be more careful with the composition of the diet, especially in diabetes.

References 1 Schmid A, Schneider H, Golay A, Keller U: Economic burden of obesity and its comorbidities in Switzerland. Soz Praventivmed 2005;50:87–94. 2 Ruof J, Golay A, Berne C, et al: Orlistat in responding obese type 2 diabetic patients: metaanalysis findings and cost-effectiveness as rationales for reimbursement in Sweden and Switzerland. Int J Obes (Lond) 2005;29:517–523. 3 Golay A, Ybarra J: Link between obesity and type 2 diabetes. Best Pract Res Clin Endocrinol Metab 2005;19:649–663.

137

Diabetes in the Life Cycle Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 139–153, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

The Accelerator Hypothesis: A Unifying Explanation for Type-1 and Type-2 Diabetes Terence J. Wilkin Department of Medicine, Postgraduate Medical School, Derriford Hospital, Plymouth, UK

Abstract Despite 30 years of research, the cause of type-1 diabetes remains unknown. Meanwhile, its incidence has risen three-fold, its clinical features have become increasingly difficult to distinguish from type-2 diabetes and the contribution of genes to its pathogenesis has changed. The accelerator hypothesis argues that type-1 and type-2 diabetes are the same disorder of insulin resistance set against different genetic backgrounds. It identifies three processes which variably accelerate ␤ cell loss: constitution, insulin resistance and the immune response to it. None of the accelerators leads to diabetes in the absence of weight gain, a trend which the hypothesis deems central to the rising incidence of all diabetes in the industrially developed and developing world. Weight gain causes an increase in insulin resistance, which results in the weakening of glucose control. The rising blood glucose accelerates ␤ cell apoptosis (glucotoxicity) and, by increasing ␤ cell immunogenicity, further accelerates apoptosis in a subset genetically predisposed to an intense immune response. Rather than overlap between the two types of diabetes, the accelerator hypothesis envisages overlay – one a subset of the other. Body mass is central to the development and rising incidence of all diabetes. Only tempo distinguishes type 1 from type 2. The control of weight gain, and with it insulin resistance, could be the means of preventing both by slowing their progression. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

Introduction In the mid 1960s, the Belgian histopathologist Gepts [1] first described lymphocytes infiltrating the islets of children who died within a few days of 139

Wilkin developing diabetes. In 1974, Nerup et al. [2] found an association between childhood diabetes and the immune response or HLA genes on the short arm of chromosome 6. That same year, Bottazzo et al. [3] reported the presence of antibodies to the islet cells in blood samples from people with insulin-dependent diabetes. Together, these observations prompted a radical revision in the understanding of diabetes. The prevailing model of diabetes as a single disorder that could present either in childhood (juvenile onset) or later in life (adult onset) was replaced by a classification that clearly distinguished type-2 diabetes (then almost exclusively a disorder of later adulthood) from type-1 diabetes. Type 1 became known as autoimmune diabetes, caused by a dysregulated immune system that attacks, and ultimately destroys, the ␤ cells. The past three decades have seen an exponential rise in the incidence of diabetes – most noticeably of type 2, owing to the greater numbers involved, but also of type-1 diabetes. While type-2 diabetes is widely understood to be linked to obesity and the insulin resistance it causes, the rise in type 1 has not been satisfactorily explained.

The Issue Diabetes is of two types, or so the textbooks say. Type-1 diabetes is an autoimmune disorder of childhood, characterized by acute onset, ketoacidosis and insulin dependency. Type 2 is a metabolic disorder of middle life, slow in onset and non-insulin-dependent. Once again, the definitions of diabetes need urgent revision. Nowadays, more than half of patients with type-1 diabetes present in adulthood [4], when their onset is slow and many neither develop acidosis nor require insulin for many years. Type-2 diabetes occurs in teenagers [5], sometimes with ketoacidosis [6], and insulin dependency frequently ensues given time. Clinically, there is little other than tempo to distinguish two types of diabetes.

The Hypothesis The accelerator hypothesis argues that type-1 and type-2 diabetes are the same disorder of insulin resistance, set against different genetic backgrounds [7]. The first accelerator – a constitutionally (intrinsically) high rate of ␤ cell apoptosis during life [8] – is necessary for diabetes to develop, but seldom in itself sufficient to cause it. It reflects natural variation within the population. Insulin resistance, the second accelerator, is acquired largely from weight gain and physical inactivity. It further increases the rate of ␤ cell apoptosis through glucotoxicity and lipotoxicity [9, 10]. However, those who develop type-1 diabetes are members of the same population as those who develop 140

The Accelerator Hypothesis type 2, exposed to the same obesogenic environment. A subset with both intrinsic susceptibility and insulin resistance – those with genes that program a particularly intense immune response to the upregulated islets – develops ␤ cell autoimmunity, the third accelerator. The hypothesis views autoimmunity as a response to insulin resistance, rather than the cause of the diabetes – an inflammatory response within the ␤ cells, whose intensity is associated with significant collateral damage. The glucagon-secreting ␣ cells are spared because they are metabolically downregulated in hyperglycemia, rather than upregulated. The author has argued, like others before him [11], that the immune system evolved originally as a ‘housekeeper’, programmed to phagocytose the detritus of natural cell death [12]. It retains that primordial function. From this perspective, autoimmunity will be antigen-driven and specific, its intensity responsive to the rate of apoptosis (antigenic load) and modulated by genetic influences. The issues of self-tolerance and its abrogation, which have always made it difficult conceptually to reconcile autoimmunity with a normal immune system, are not at issue where clones expand appropriately to remove apoptotic bodies. Antibodies in this context are classic immunological adaptors. They link the specific subparticulate molecules to be cleared to nonspecific Fc receptors of phagocytic neutrophils, which in turn engulf the complex and dispose of it through the reticuloendothelial system [13]. Insulin dependency is the end stage towards which all diabetes moves, the rate dependent on the accelerators present, and the notions of type 1 and type 2, insulin- and non-insulin-dependent, are consequently artificial. The development of diabetes is just a matter of time, and tempo is the only feature that distinguishes one type from another. Of the three accelerators, the first is intrinsic and others acquired. Insulin resistance, the second accelerator, is associated with visceral fat mass and is widely believed to explain the epidemic rise of type-2 diabetes in the industrially developed world [14]. The accelerator hypothesis argues that visceral weight gain is also central to type-1 diabetes, as much responsible for its rising incidence as for that of type 2, and the environmental factor in type-1 diabetes that has eluded epidemiology for so long. The concept of an etiological link between the two types of diabetes is not new, and has been suggested before [15], but the evidence is now stronger. Rather than overlap between the two types of diabetes, the accelerator hypothesis envisages overlay. Type-1 diabetes is the same as type 2 except for one essential add-on: intense immune response.

Pathophysiology of Diabetes Type-1 diabetes is associated with autoantibodies and activated lymphocytes which are reactive with ␤ cell antigens [16]. Its course is characterized 141

Wilkin by a symptomless prediabetic phase whose presence can be inferred from immune markers. Pre-type-1 diabetes is a period of accelerated ␤ cell loss, whose tempo varies from acute in those who present young to subacute or chronic in those who present later in life [17]. The differences in tempo are assumed to be under genetic control, since those who develop type-1 diabetes in childhood tend to carry more intensely responsive HLA genes than those who develop it later in life [18, 19]. ␤ cell autoimmunity appears to start early in life, insofar as immune markers predictive of future diabetes can be present as early as at 9 months of age [20]. Type-2 diabetes is characterised by a combination of insulin resistance and defective insulin response that results from it [21]. Blood insulin concentrations are raised, at least initially, but are never sufficient to meet the resistance that entrains them. Like type-1, type-2 diabetes also presents after a variable period of prediabetes, whose presence might be revealed by high fasting insulin/glucose ratios and later by glycosuria or hyperglycemia in circumstances which temporarily increase insulin resistance – typically pregnancy, thyrotoxicosis or a course of anti-inflammatory steroids. Between 17 and 63% of women whose glycosuria during pregnancy is attributable to glucose intolerance will subsequently become diabetic, depending on the series quoted [22]. Of these, a proportion (around 20% according to one study [23]) will develop type-1 diabetes, underscoring the principle to be established here that the prediabetes of type 1 and type 2 differs only in tempo, not in outcome. Both represent a period of accelerated ␤ cell loss. It has long been recognized that islet cells are both metabolically and immunogenically upregulated when functionally stressed by a rising blood sugar [24]. At whatever age it emerges, insulin resistance could be expected to increase ␤ cell stress, and to intensify an immune response in those who are genetically predisposed. The phenomenon of insulin resistance, which as the response to progressively rising body weight has been largely responsible for reducing the age at presentation of type-2 diabetes over recent time, might be doing just the same for type-1 diabetes by promoting the immunological accelerants of ␤ cell death in a progressively younger age group. If there is little clinically to distinguish the two types of diabetes nowadays, there is little fundamentally either.

Insulin Resistance Many theories have sought to account for insulin resistance. The thrifty genotype hypothesis argues for gene selection against muscle proteolysis during an evolutionary history of recurring famine [25, 26]. In contrast, the thrifty phenotype hypothesis [27], which first described an association between low birth weight and insulin resistance, explains the link as a gestational programming of the fetus in response to poor maternal nutrition. More 142

Incidence per 100,000

The Accelerator Hypothesis

27 26 24 22 20 18 16 14 12 10 8 6 4 2 0

Age-standardized annual incidence of type 1 diabetes from 1968 to 1993

Male Female 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 Year

Fig. 1. Scottish Study Group for the Care of Young Diabetics. One of many studies showing the progressive rise in incidence of type-1 diabetes over the past generation.

recently, the fetal insulin hypothesis has cited observations in families with maturity onset diabetes of the young to illustrate the dependence of fetal growth on the genetics of fetal and maternal insulin secretion [28]. It predicts that a gene or combination of genes responsible for insulin resistance will be found which leads both to low weight at birth through insulin resistance and to glucose intolerance later in life (fig. 1). There is a common theme to all three hypotheses: insulin resistance, which might arguably have favored survival in times of famine, leaves many in today’s ‘coke and burger’ culture unable to control their blood sugar. None of the theories, however, estimates how much of today’s diabetes (the attributable proportion) can be accounted for by congenital insulin resistance already present at birth, nor explains the rising incidence of type-2 diabetes which, according to the logic of all three hypotheses, should by now be stable or falling as nutrition in pregnancy improves and gene selection operates to select out the less fit. Insulin resistance acquired through lifestyle change is more likely than genes or gestational experience to underlie the recent rise in diabetes and its ever-younger presentation. Importantly for the hypothesis, glucose clamp studies 20 years ago showed that noninsulinized adults with type-1 diabetes were as insulin resistant as metabolic diabetics of comparable glucose tolerance [29] and there has been further evidence recently. Furthermore, the rise in proinsulin/insulin ratio that has long been the hallmark of insulin resistance in pre-type-2 diabetes has been shown to characterize pre-type-1 diabetes as well [30]. Most seropositive type-2 diabetic adults become type 1 more rapidly than those who are 143

Wilkin seronegative [31] and these observations together provide robust support for the ‘overlay’ and ‘accelerator’ concepts. The slower tempo of progression in adults has thus provided the means of demonstrating that all diabetes is associated with insulin resistance (the second accelerator), that a subgroup advances more rapidly to insulin dependency as a result of autoimmunity (the third accelerator), and that (by implication) this subgroup, had it been free of autoimmunity, would have progressed to diabetes in any case, albeit at a later date.

Susceptibility and Risk Type-2 diabetes is prevalent in industrially developed societies. It affects up to 30% of some populations [32], suggesting that susceptibility to diabetes is common, though not universal, in as much as some of those apparently at greatest risk – the pathologically obese – never develop the disease. The probability of developing a multifactorial disorder such as diabetes is made up of genetic susceptibility and environmental risk. Both contribute a proportion to the probability and, if one rises, the other must inevitably fall. If the rising incidence of type-1 diabetes has been the result of rising environmental risk (obesity and insulin resistance), the genetic contribution must have fallen. Concordance among monozygotic twin pairs is widely believed to reflect the genetic contribution to probability, and lies at around 75% in type-2 diabetes (rising to 95% if those with glucose intolerance alone are included) [33] but only 20% in type 1 [34]. At least, that was historically the case. Recent data from the Danish twin registry suggests that concordance for type-1 diabetes in monozygotic twins is now around 50–60%, and no different from dizygotic twins [35]. Evidence for the rising environmental risk for type-1 diabetes comes from a study in the South of England which shows a progressive rise in body mass index (BMI) at diagnosis over the past 20 years [36]. The hypothesis predicts two outcomes of the corresponding fall in contribution from HLA genes – convergence of the clinical phenotype and loss of the difference in concordance rates between identical and nonidentical twins noted above.

The Epidemiology of Diabetes Over the past 20 years, the incidence of type-2 diabetes in the western world has increased dramatically, a pattern that parallels closely the rising incidence of obesity and (by implication) that of insulin resistance [37]. Furthermore, age at presentation has been falling, such that the incidence of type-2 diabetes in American adolescents has increased 10-fold and in Japanese school children 36-fold within a generation [38]. Rather strikingly, 144

The Accelerator Hypothesis 5

BMI SD score at diagnosis

4

Boys Girls

3 2 1 0 ⫺1 ⫺2 ⫺3 ⫺4 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 Year

Fig. 2. Evidence for the progressive rise in insulin resistance (body mass) deemed by the accelerator hypothesis to be the environmental trigger for type-1 diabetes [from 36].

the same pattern of increasing incidence and younger age at presentation has occurred for type-1 diabetes over the same time period. Several studies in Europe show a doubling in incidence of type-1 diabetes over the last generation, with a clear shift of presentation to younger age groups [39, 40] although the data are largely restricted to children and adolescents. The incidence of type-1 diabetes has been highest around puberty in all populations studied [41, 42]. The earlier peak in girls is consistent with their earlier maturation [43]. The association between type-1 diabetes and puberty has never been satisfactorily explained, but may once again be an expression of insulin resistance. The hormonal changes of puberty (particularly the rise in growth hormone) place demands on insulin production that already damaged islets may be unable to meet. Again, BMI rises rapidly during puberty, and with it insulin resistance [44] (fig. 2). If this latter were the correct explanation, the accelerator hypothesis would predict a correspondingly earlier presentation of diabetes as the BMI previously associated with puberty is reached at a progressively younger age. Tuomilehto et al. [45] have reported how, over recent years, the age-at-onset curve for diabetes has risen to include most of early childhood, all but losing its peripubertal peak. Others have shown independently that weight gain early in childhood is associated with a higher risk of early type-1 diabetes [46, 47] and the same appears to be true for type-2 [2]. Most recently, the Childhood Diabetes in Finland Study Group has reported that a relative 145

BMI SD score at diagnosis⫹ 6 months

Wilkin

5

r⫽0.3, p⬍0.001

Boys Girls

4 3 2 1 0 ⫺1 ⫺2 ⫺3 0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

Age at diagnosis (years)

Fig. 3. The relationship between body mass and age at onset of diabetes in an unselected group of children with type-1 diabetes; the fatter child develops type-1 diabetes younger – true acceleration [from 36].

weight in childhood of ⬎120% is associated with a more than 2-fold greater risk of developing type-1 diabetes [48]. These observations provide important support for the hypothesis. They point to a central role for body mass, and by implication for insulin resistance in the development of type-1 as well as type-2 diabetes. More importantly still, there are now two full length reports to indicate that, within a population with type-1 diabetes, the heaviest children develop type-1 diabetes the youngest – true acceleration [36, 39] (fig. 3).

Body Mass: The Elusive ‘Trigger’ of Type-1 Diabetes If the rising incidence and earlier presentation of diabetes were to be explained by an ever heavier population at all ages, weight gain would have as much a role in the changes of demography of type-1 diabetes as it does in that of type 2. Ascribing the rising incidence of type-1 diabetes to metabolic, rather than to immunological, factors has novel and important implications. Clinical research over the past 30 years has focused almost exclusively on exogenous factors (viruses, toxins, allergens) deemed able to initiate, facilitate or intensify autoimmune damage to the ␤ cell. Although many have been proposed [49], none has been confirmed. The prevalence of obesity meanwhile has trebled. Insulin resistance, resulting from a combination of obesity and physical inactivity, is a serious candidate for the ‘elusive’ environmental factor responsible for the rising incidence of type-1 diabetes and, as such, a true accelerator. 146

The Accelerator Hypothesis The rise in incidence of diabetes in ‘westernized’ countries over recent time has occurred over a period too brief for changes in the gene pool to exert an influence. Similarly, there is no evidence of falling birth weight over the same period to suggest a deterioration in fetal quality and attendant risk of insulin resistance, as the fetal origins hypothesis proposes – indeed, birth weights have risen. More likely, it is the progressive increase in body weight after birth, with the rising prevalence of obesity at all ages over the past 30 or more years, which is responsible. Most importantly, epidemiological observation suggests that the progressive rise in type-1 diabetes only began in the middle of the 20th century, the time after the end of the Second World War when modern wealth and lifestyles started to involve populations rather than merely individuals [50]. The accelerator hypothesis at its simplest can be reduced to the unfavorable interplay of two phenomena: insulin resistance and accelerated ␤ cell apoptosis. Apoptosis occurs throughout life at a variable rate, but is intrinsically higher in those who are susceptible to diabetes. Many people of normal body mass, despite intrinsically more rapid ␤ cell apoptosis, never develop glucose intolerance because insulin secretory reserve remains sufficient in such circumstances to maintain control of blood glucose concentrations. However, insulin resistance – whether present at birth or acquired through the accumulation of visceral fat during life – makes demands on insulin secretion that in some cannot be met. Immune damage of the ␤ cells is an additional accelerator, restricted by genotype to small and independent minorities of the intrinsically diabetessusceptible and nonsusceptible populations alike. At its most aggressive, the so-called autoimmunity might be sufficient to cause diabetes of itself, though an accelerator different from insulin resistance would be needed to account for the increase in apoptosis which provokes it. The hypothesis predicts that in older patients, where ␤ cell autoreactivity is less aggressive, those with islet cell autoimmunity most likely to develop diabetes will already have the diabetic phenotype – a low ␤ cell mass and high insulin resistance. Those who do not might be expected to remain seropositive but healthy, to succumb only as and when the intrinsic ␤ cell mass wanes and/or insulin resistance rises with the corpulence of advancing age. Tempo, and tempo alone, it is argued, distinguishes what in the past have been viewed as two separate types of diabetes. The three phases in the progression to overt diabetes – prediabetes, chemical diabetes and clinical diabetes can all be identified in both types, differing only (though sometimes substantially) in their relative duration. Not to take account of these phases and their differences in tempo will make it conceptually difficult to regard as equivalent two diabetic states where insulin is needed from the outset in one, but only (if ever) after a long period of clinical diabetes in the other. The requirement for insulin in both cases is nevertheless reached at exactly corresponding points in the progression of prediabetes [19]. All people with diabetes 147

Wilkin move towards this point of insulin dependence – some very quickly, others perhaps not in a lifetime. A hypothesis postulating body mass as a primary risk factor in the etiology of type-1, as well as type-2, diabetes is novel, but eminently testable. Ultimately, it will be necessary to establish whether strategies to reduce the second accelerator (insulin resistance) in those at risk from type-1 diabetes, through weight loss, metformin, or one of the new thiazolidinediones [51], is paralleled by a deceleration in the third – autoimmune damage to the ␤ cells. The notion that type 1 could represent merely the accelerated development of type-2 diabetes is important if it implies that strategies currently on trial to suppress the immunological accelerator of type-1 diabetes (e.g. anti-CD3 [52, 53]) leave unchanged the insulin resistance which provoked it. The control of weight gain, and with it insulin resistance, could be the fundamental means of averting both.

References 1 Gepts W: Pathologic anatomy of the pancreas in juvenile diabetes mellitus. Diabetes 1965;14: 619–633. 2 Nerup J, Platz P, Anderssen OO: HLA antigens and diabetes mellitus. Lancet 1974;ii:864–866. 3 Bottazzo GF, Florin-Christensen A, Doniach D: Islet-cell antibodies in diabetes mellitus with autoimmune polyendocrine deficiencies. Lancet 1974;ii:1279–1283. 4 Molbak AG, Christau B, Marner B, et al: Incidence of insulin-dependent diabetes in age groups over 30 years in Denmark. Diabet Med 1994;11:650–655. 5 Rosenbloom AL, Joe JR, Young RS, Winter WE: Emerging epidemic of type 2 diabetes in youth. Diabetes Care 1999;22:345–354. 6 Aizawa T, Funase Y, Katakura M, et al: Ketosis-onset diabetes in young adults with subsequent non-insulin-dependency, a link between IDDM and NIDDM? Diabet Med 1997;14: 989–991. 7 Wilkin TJ: The accelerator hypothesis: weight gain as the missing link between type I and type II diabetes. Diabetologia 2001;44:914–922. 8 Mauricio D, Mandrup-Poulsen T: Apoptosis and the pathogenesis of IDDM: a question of life and death. Diabetes 1998;47:537–543. 9 Maedler K, Donath MY: Beta-cells in type 2 diabetes: a loss of function and mass. Horm Res 2004;62(suppl 3):67–73. 10 Robertson RP, Harmon J, Tran PO, Poitout V: Beta-cell glucose toxicity, lipotoxicity, and chronic oxidative stress in type 2 diabetes. Diabetes 2004;53(suppl 1):S119–S124. 11 Grabar P: Autoantibodies and the physiological role of immunoglobulins. Immunol Today 1983;4:337–340. 12 Wilkin TJ: Autoimmunity: attack or defence? Autoimmunity 1989;3:57–73. 13 Roitt I, Brostoff J, Male D: Immunology. London, Churchill Livingstone, 1995, pp 1.5–1.6. 14 Zimmet P: Globalization, coca-colonization and the chronic disease epidemic: can the Doomsday scenario be averted? J Intern Med 2000;247:301–310. 15 Wilkin TJ: Early nutrition and diabetes mellitus (editorial). Br Med J 1992;306:283–284. 16 Atkinson MA, Maclaren NK: The pathogenesis of insulin-dependent diabetes mellitus. N Engl J Med 1994;331:1428–1436. 17 Eisenbarth GS, Gianani R, Yu L, et al: Dual-parameter model for prediction of type I diabetes mellitus. Proc Assoc Am Phys 1998;110:126–135. 18 Demaine AG, Hibberd ML, Mangles D, Millward BA: A new marker in the HLA class I region is associated with the age at onset of IDDM. Diabetologia 1995;38:623–628. 19 Caillat-Zucman S, Garchon HJ, Timsit J, et al: Age-dependent HLA genetic heterogeneity of type 1 insulin-dependent diabetes mellitus. J Clin Invest 1992;90:2242–2250.

148

The Accelerator Hypothesis 20 Ziegler AG, Hummel M, Schenker M, Bonifacio E: Autoantibody appearance and risk for development of childhood diabetes in offspring of parents with type 1 diabetes: the 2-year analysis of the German BABYDIAB Study. Diabetes 1999;48:460–468. 21 Turner RC, Holman RR, Matthews DR, Peto J: Relative contributions of insulin deficiency and insulin resistance in maturity-onset diabetes. Lancet 1982;i:596–598. 22 Kjos SL, Buchanan TA: Gestational diabetes mellitus. N Engl J Med 1999;341:1749–1756. 23 Damm P, Kuhl C, Buschard K, et al: Prevalence and predictive value of islet cell antibodies and insulin autoantibodies in women with gestational diabetes. Diabet Med 1994;11: 558–563. 24 Bjork E, Kampe O, Karlsson FA, et al: Glucose regulation of the autoantigen GAD65 in human pancreatic islets. J Clin Endocrinol Metab 1992;75:574–576. 25 Neel JV: Diabetes mellitus: a ‘thrifty’ genoype rendered detrimental by ‘progress’? Am J Hum Genet 1962;14:353–362. 26 Reaven GM: Hypothesis: muscle insulin resistance is the (‘not so’) thrifty genotype. Diabetologia 1998;41:482–484. 27 Hales CN, Barker DJP: Type 2 (non insulin-dependent) diabetes: the thrifty phenotype hypothesis. Diabetologia 1992;35:595–601. 28 Hattersley AT, Tooke JE: The fetal insulin hypothesis: an alternative explanation of the association of low birth weight with diabetes and vascular disease. Lancet 1999;353:1789–1792. 29 Gray RS, Borsey DQ, Irvine WJ, et al: Non-insulin-treated ICA positive diabetics are equally insulin-resistant. Diabetes Metab 1983;9:292–296. 30 Rodriguez-Villar C, Conget I, Casamitjana R, et al: High proinsulin levels in late pre-IDDM stage. Diabetes Res Clin Pract 1997;37:145–148. 31 Turner R, Stratton I, Horton V, et al: UKPDS 25: autoantibodies to islet-cell cytoplasm and glutamic acid carboxylase for prediction of insulin requirement in type 2 diabetes. UK Prospective Diabetes Study Group. Lancet 1997;350:1288–1293. 32 Zimmet PZ, McCarty DJ, de Courten MP: The global epidemiology of non-insulin-dependent diabetes mellitus and the metabolic syndrome. J Diabetes Complications 1997;11:60–68. 33 Medici F, Hawa M, Ianari A, et al: Concordance rate for type 2 diabetes mellitus in monozygotic twins: an actuarial analysis. Diabetologia 1999;42:146–150. 34 Kaprio J, Tuomilehto J, Koskenvuo M, et al: Concordance for type 1 (insulin-dependent) and type 2 (non-insulin-dependent) diabetes mellitus in a population-based cohort of twins in Finland. Diabetologia 1992;35:1060–1067. 35 Kyvik KO: Concordance for diabetes in 2–20 year old twins. Diabetologia 2004;47(suppl 1):A112. 36 Betts P, Mulligan J, Ward P, et al: Increasing body weight predicts the earlier onset of insulindependent diabetes in childhood: testing the ‘accelerator hypothesis’ (2). Diabet Med 2005;22:144–151. 37 National Task Force on the Prevention and Treatment of Obesity: overweight, obesity, and health risk. Arch Intern Med 2000;160:898–904. 38 Kitagawa T, Owada M, Urakami T, Yamauchi K: Increased incidence of non-insulin dependent diabetes mellitus among Japanese schoolchildren correlates with an increased intake of animal protein and fat. Clin Pediatr (Phila) 1998;37:111–115. 39 Onkamo P, Vaananen S, Karvonen M, Tuomilehto J: Worldwide increase of type 1 diabetes – analysis of the data on published incidence trends. Diabetologia 1999;42:1395–1403. 40 Zhao HX, Stenhouse E, Soper C, et al: Incidence of childhood-onset type 1 diabetes mellitus in Devon and Cornwall, England, 1975–1996. Diabet Med 1999;16:1030–1035. 41 Akerblom HK, Reunanen A: The epidemiology of insulin-dependent diabetes mellitus (IDDM) in Finland and in northern Europe. Diabetes Care 1985;8(suppl 1):10–16. 42 Staines A, Bodansky HJ, Lilley HE, et al: The epidemiology of diabetes mellitus in the United Kingdom: the Yorkshire Regional Childhood Diabetes Register. Diabetologia 1993;36: 1282–1287. 43 Staines A, Bodansky HJ, Lilley HE, et al: The epidemiology of diabetes mellitus in the United Kingdom: the Yorkshire Regional Childhood Diabetes Register. Diabetologia 1993;36: 1282–1287. 44 Moran A, Jacobs DR Jr, Steinberger J, et al: Insulin resistance during puberty: results from clamp studies in 357 children. Diabetes 1999;48:2039–2044. 45 Tuomilehto J, Virtala E, Karvonen M, et al: Increase in incidence of insulin-dependent diabetes mellitus among children in Finland. J Epidemiol 1995;24:984–992.

149

Wilkin 46 Johansson C, Samuelsson U, Ludvigsson J: A high weight gain in early life is associated with an increased risk of type 1 (insulin-dependent) diabetes mellitus. Diabetologia 1994;37:91–94. 47 Hypponen E, Kenward MG, Virtanen SM, et al: Infant feeding, early weight gain and risk of type 1 diabetes. Childhood Diabetes in Finland (DiME) Study Group. Diabetes Care 1999;22: 1961–1965. 48 Hypponen E, Virtanen SM, Kenward MG, et al: Obesity, increased linear growth, and risk of type 1 diabetes in children. Childhood Diabetes in Finland Study Group. Diabetes Care 2000;23:1755–1760. 49 Dahlquist G: The aetiology of type 1 diabetes: an epidemiological perspective. Acta Paediatr Suppl 1998;425:5–10. 50 Gale EA: The rise of childhood type 1 diabetes in the 20th century. Diabetes 2002;51: 3353–3361. 51 Day C: Thiazolidinediones. Diabet Med 1999;16:179–192. 52 Harlan DM, von Herrath M: Immune intervention with anti-CD3 in diabetes. Nat Med 2005;11:716–718. 53 Herold KC, Gitelman SE, Masharani U, et al: A single course of anti-CD3 monoclonal antibody hOKT3gamma1(Ala-Ala) results in improvement in C-peptide responses and clinical parameters for at least 2 years after onset of type 1 diabetes. Diabetes 2005;54:1763–1769.

Discussion Dr. Eshki: Thank you for this very informative presentation. I have witnessed the condition in the neonatal intensive care unit in which infants develop diabetes because they have been given diets inappropriate for their age. There are several studies looking at infant diet and type-1 diabetes [1, 2]. Does your theory cover such factors? Dr. T. Wilkin: Insulin resistance is a post receptor defect, and the child who grows up to become a type-1 child lives in exactly the same obesogenic environment as a child who grows up to become a type-2 individual. We tended to think of two separate populations, type-2 insulin resistance, type-1 autoimmune, but they live in the same obesogenic world. They have exactly the same experiences which lead to increased insulin resistance. There is no reason to believe that there would be a different insulin resistance in the type-1 than there would be in the type-2 individual. Dr. Eshki: It is just because insulin is formed in a different area than the insulin receptors, that is what I was looking at. So maybe the formation of insulin receptors could also be a factor where perhaps in some cases it is not well enough formed for them to provide the insulin available. Dr. T. Wilkin: I think current understanding would lead us to the argument that insulin resistance is indeed a post-receptor issue. Dr. Chiasson: It is a very provocative and interesting hypothesis. I agree that it is interesting that perhaps with a different paradigm we will approach the prevention of type-1 differently, and we may be able to delay the development of type-1 diabetes, be it type-1 or type-2. Now in type-2 diabetes insulin resistance is mostly an environmentally induced factor. It is very likely, yet it still has to be proven, that the major genetic defect in type-2 is the ␤-cell, maybe the ␤-cell mass or the ␤-cell susceptibility to whatever environmental factor, obesity, decreased physical activity, and insulin resistance. Now the fact that insulin resistance or obesity or both, because usually one correlates with the other, would tend to increase or accelerate the development of type-1 in subjects susceptible for type-1 is not unexpected because the ␤-cells are under strain. If they have whatever the problem at the ␤-cell level then obesity would accelerate, would increase the stress on the ␤-cell. So to me, the insulin resistance per se will accelerate both type-1 and type-2, but that doesn’t necessarily mean that the

150

The Accelerator Hypothesis problem at the ␤-cell level is the same. As you said, correlations are always interesting but do not prove a cause and effect relationship. So I challenge the hypothesis. Dr. T. Wilkin: I think what we will always find that the individual who is more immune-susceptible, that is to say whose HLA genes instruct a more reactive response, will always present first. The accelerator hypothesis I see as revolving around a gradient of ␤-cell loss, which we are all subject to. Most of us live our lives and ␤-cell loss is never sufficient to cause us diabetes. From that point on there are accelerators. One of those accelerators, as I think we agree, is insulin resistance which has mechanisms which are increasingly being understood. The second of these, which is my accelerator number 3, is the immune response which is graded according to immune response genes. Now I believe all that is doing is to accelerate the rate, so that of a group of individuals who is going to get diabetes, those in whom it is most accelerated, will inevitably be children, and those in whom it is least accelerated may well die before it ever happens. I don’t look outside that paradigm for an answer at the moment because I think that insulin resistance is driving both the non-immune and the immune component of ␤-cell damage. Dr. Chiasson: I am not an immunologist, but I understand that the HLA genes are not diabetic genes. There are a number of genes that have been identified, but whether it is in the same area, I don’t know. The genes themselves as well as the antibodies that we are looking at do not necessarily cover everything. That must be left open. Also in type-2, there are going to be a number of genes related to the ␤-cell function as well as perhaps insulin action. At least in the prevention studies that we have done so far, the only thing that we have been able to modify is the insulin action, insulin sensitivity, not ␤-cell function. Dr. T. Wilkin: Are you talking of type-2 prevention studies, or type-1? Dr. Chiasson: Type-1 would most likely be the same. At the present time I don’t think that in type-1 we have any tools that can help us to modify the susceptibility of the ␤-cell, assuming that it is different than the susceptibility in type-2. We don’t have any data one way or the other, but it is clear to me that they are susceptible, phenotypically they are different. As a matter of fact Dr. Slama told me that a lean type-2 is a type-1 who is ignoring himself. So more and more we find antibodies in those lean type-2 individuals, and they evolve differently over time, and the insulin resistance is different. The insulin resistance in type-1 exists; it does not compare to type-2. Dr. T. Wilkin: Again I would answer by saying that you have two fundamental contributors to the probability of disease, one of them is insulin resistance and one of them is HLA genetic. In order to develop diabetes you would need to have a higher insulin resistance if you don’t have the genetic contribution. In type-1 diabetes you would need less insulin resistance to have the same effect because you have the help of the third accelerator which is the immune response. Dr. Slama: I found your hypothesis very clever. I accept your hypothesis that there are some accelerators in the phenomenon. But I have two arguments. First of all you have also to show not only that the patients are a bit overweight as compared to normal but they do have insulin resistance. When insulin resistance has been studied by glucose clamp in type-1 diabetics it was found that it is really insulin resistance at the beginning where the blood glucose is very high but declining to normal very quickly when blood glucose is normalized, making this phenomenon an accelerator but not a causal factor. My second argument: if you try to find out if there is vitamin B12 deficiency in a large population, you will find a prevalence of X. Then in the same community there will be iron deficiency, much more anemia in this population, and the more iron deficiency the less you will find a role for autoimmune disease in this community. Does that mean that the disease is the same as iron deficiency, of course not. When you say that genetics is decreasing, it is decreasing relatively but in

151

Wilkin absolute value it does not change. So I accept your hypothesis of the accelerator, but I do not accept that because the accelerator might be verified that the original disease has disappeared. Dr. T. Wilkin: Let me take the first question. I would give the same response as I did to Dr. Chiasson, that the amount of the insulin resistance you would require to develop a diabetes is going to be less if the contribution that it makes is helped by an immune accelerator being present as well. The second question is an interesting one because your implication is that even though childhood diabetes is increasing, and it undoubtedly is, you will always have conventional type-1 diabetes because in absolute terms it has not changed. I actually don’t agree with that, I think you will lose the phenotype of a type-1 diabetic because I think he will become overweight, acanthotic, seropositive, T-cell reactive and quite impossible to distinguish from what we know now as a type-2. I think there will be a convergence because the slim type-1 diabetic, as we have known him in the past, is now obese. Dr. Slama: How can you say so when you say you will find T-cell activation and you will not be able to separate them? There is T-cell stimulation, so they are different. Perhaps our tools are not fine enough to distinguish this very well, but it doesn’t mean that it doesn’t exist. Dr. T. Wilkin: There is T-cell stimulation for both groups now which there wasn’t previously. We never knew type-2 diabetes in childhood 30 years ago. The real evidence comes from the slide that I showed twice in which there is a secular rise in the weight at diagnosis of children who are labeled type-1, and that has been linear since the 1980s. The variance at each time point is much the same, the mean has progressively risen, we are progressively losing sight of the cachectic, scrawny, low body weight child that is often portrayed as the characteristic presentation of type-1 diabetes. I think we will lose it completely if things go on the way they are. Dr. Schiffrin: If I remember correctly, the genetic susceptibility for diabetes type-1 and celiac disease shows a strong overlap. There are Swedish studies showing that the two diseases coexist in some patients. Now if I got your paradigm right, if the insulin resistance or the environment is becoming so important, the coexistence of type-2 diabetes should be less associated with celiac disease if the genetic background is less important. Is that right? Dr. T. Wilkin: You are absolutely correct about the association between celiac disease and type-1 diabetes, but the connection there is genetic. If the phenotype of type-1 changes as Dr. Slama was saying, there will always be the same genetic distribution within the population. I have no doubt that there will always be an association of celiac disease with a proportion of individuals who have diabetes. But that is a genetic association, it is not an association with the phenotype of a disease which I think is becoming indistinguishable for the reasons that I have indicated. Dr. Golay: I come from the type-2 research field, and we are more and more able to prove that we have a ␤-cell defect in type-2. I propose a third type (half type-1) between a clear type-1 (complete insulin defect) and type-2 with insulin resistance and a slow partial defect. Dr. T. Wilkin: I understand what you are saying, but I am not sure I would give it that explanation. Dr. Golay: We have a third group of diabetic patients instead of having the same disease for type-1 and type-2. Dr. T. Wilkin: The rates at which conversion occurs from type-2 to type-1 is a function of the third accelerator. I think that conversion occurs very rapidly in childhood and the earlier in childhood the more rapid, and that is because the tempo is so telescoped in childhood by the intensity of the immune response; that would be my argument. When you have got through childhood you have, as it were, used up the

152

The Accelerator Hypothesis population with that particular genotype of extremely intense reactivity. Teenagers who develop type-1 have a different genotype distribution than those who are 2 or 3 years old when they develop type-1. When you move from teenagers into adulthood, the genotype changes again. When you come to adults, where the third accelerator is contributing very little, then the tempo of conversion is very slow, but I see no fundamental conceptual difference in the two. Dr. Metzger: There are two points I wanted to make. One, I don’t believe that your analogy of the cachectic type-1 patient is very strong evidence because that represented the long-term symptomatic uncontrolled state of diabetes, and we tend to see a diagnosis made much earlier now. In analogy we see very few sustained long-term symptomatic patients with Graves disease before we make a diagnosis, whereas 30 years ago it was very different. It seems to me that the acceleration in the onset of type-1 diabetes would not necessarily increase the true incidence of type-1 diabetes. We may simply be seeing a shift to the left in onset, and whether the lifetime risk of type-1 diabetes has changed in the last 30 years I think would be an interesting part of evaluating the status of type-1 diabetes. Dr. T. Wilkin: There is some evidence that would help on that as well. There are at least two, possibly three reports in the literature which would suggest that there is indeed a shift. I think the accelerator hypothesis would predict just that, that there is a group in the population with higher susceptibility genes who will accelerate faster if given the stimulus through insulin resistance. But they would borrow from the older group as a result, so that you would see a shift to an earlier age at onset, and earlier age at onset is very much the observation that we make in type-1 diabetes. Dr. Bantle: I think there is evidence that if you treat people with early type-1 diabetes with insulin and control glucose, you cause weight gain. Presumably you increase insulin resistance but, at the same time, you preserve insulin production for some time. Would the accelerator hypothesis be able to explain how that would happen? Dr. T. Wilkin: I think it would. I think it might predict that if you give an external source of insulin you would rest the endogenous ␤-cell so it would be less antigenic because it would be less metabolically upregulated, and as a less antigenic ␤-cell it would invite a less intense immune response and therefore might survive longer. That is speculation but I think it was the premise on which the DPT-1 study was based on using low dose insulin to try and preserve the ␤-cell.

References 1 Strotmeyer ES, Yang Z, LaPorte RE, et al: Infant diet and type 1 diabetes in China. Diabetes Res Clin Pract 2004;65:283–292. 2 Paronen J, Knip M, Savilahti E, et al: Effect of cow’s milk exposure and maternal type 1 diabetes on cellular and humoral immunization to dietary insulin in infants at genetic risk for type 1 diabetes. Finnish Trial to Reduce IDDM in the Genetically at Risk Study Group. Diabetes 2000;49:1657–1665.

153

Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 155–169, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

Diet and Medical Therapy in the Optimal Management of Gestational Diabetes Mellitus Boyd E. Metzger Northwestern University Feinberg School of Medicine, Chicago, IL, USA

Abstract Gestational diabetes mellitus (GDM), a common medical complication of pregnancy is increasing in prevalence among all populations in parallel with the global increase in obesity and type-2 diabetes mellitus (DM). Although controversy regarding the perinatal consequences of GDM continues, efforts to identify the severity of maternal glucose intolerance associated with clinically important adverse outcomes are ongoing. Medical therapies beyond the traditional ‘standard’ medical nutrition therapy (diet) or insulin are being explored (oral glyburide and metformin); however, less costly alternatives such as more intensive lifestyle modification need to be evaluated. Such approaches are also applicable after GDM and are known to delay or prevent progression to DM in these high-risk subjects. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

Introduction Gestational diabetes mellitus (GDM) is defined as ‘glucose intolerance with onset or first recognition during pregnancy’ [1] and it is ordinarily detected in the last half of gestation in parallel with increasing severity of insulin resistance. In contrast to those whose glucose metabolism remains normal, the ␤ cells of women who develop GDM fail to fully compensate with adequate insulin secretion. In the midst of the global increases in obesity, metabolic conditions and type-2 diabetes, it is expected that GDM would also be seen more frequently. However, universal screening for GDM is not routinely practiced in many parts of the world, and diagnostic tests and criteria are not

155

Metzger Table 1. Increasing incidence of gestational diabetes in northern California (1991–2000) Year

All ethnic groups1 White (non-Hispanic)2 African American2 Hispanic2 Asian2

1991 %

1993 %

1995 %

1997 %

2000 %

5.1 3.9 4.1 7.2 7.2

5.2 4.1 5.1 6.9 8.0

6.2 5.1 4.7 8.1 8.3

7.5 5.7 5.8 9.8 11.0

7.2 5.7 6.4 8.3 9.7

Adapted from Ferrara et al. [2]. 1Adjusted for age and race-ethnicity. 2Adjusted for age.

standardized. Evidence for an increase in GDM was reported in Australia more than a decade ago. The data summarized in table 1 were reported from the Kaiser Permanente Health Plan of Northern California in early 2004 [2]. In this program, screening for GDM is standardized and applied universally. There is more than 85% compliance with the protocol. The data in table 1 were derived from more than 265,000 pregnancies screened for glucose intolerance. The key finding is that, during the 10-year period in which these data were collected, the overall incidence (using the criteria of Carpenter and Coustan [3] throughout) showed an increase of 41%, adjusting for the influences of maternal age and ethnic mix. In 2005, similar findings were reported in a multi-ethnic population enrolled in the Kaiser Permanente of Colorado [4] in the years 1994–2002. An Increase in GDM has also been reported in India [5]; however, detailed data on the population-wide incidence from prior and current years are lacking. Developing an optimal strategy for dealing with the increasing prevalence of GDM presents major challenges. In addition to the failure to use the same criteria for diagnosis of GDM, there has been much controversy about the specificity of the relationships to hyperglycemia and ‘the level of hyperglycemia, short of overt diabetes, that conveys increased risk’ [6] and, until recently, whether treatment of GDM can improve outcomes [7]. Historically, the treatment of GDM has focused on correcting maternal hyperglycemia as a means of slowing fetal growth in an effort to avoid macrosomia and its associated morbidities, in particular risk of birth injury, cesarean delivery and neonatal morbidities. However, as will be illustrated, the pathophysiology and pathogenesis of GDM are complex and an optimal approach to treatment may involve more than simply correcting hyperglycemia. Furthermore, children who have been exposed to the intrauterine environment of diabetes or 156

Optimal Management of GDM GDM are at increased risk of obesity and altered glucose metabolism in later life [8].

Maternal Nutrition and Fetal Growth in GDM It is well known that the state of maternal nutrition (both macro- and micronutrient) before and during pregnancy is a critical factor influencing fetal growth and pregnancy outcome. Guidelines for optimal weight gain and recommendations for daily intake of micronutrients during pregnancy have for many years been offered by bodies such as the FAO/WHO/UNU [9] and the Institute of Medicine of the National Academy of Science [10]. However, the benefit of specific supplementations during pregnancy is, in most cases, uncertain (iodine and iron excepted). In April 2004, an entire Nestlé Nutrition Workshop was devoted to the topic [11]. Studies addressing these issues in pregnancy complicated by diabetes are very sparse. Nevertheless, these same guidelines for maternal nutrient intake and weight gain are generally used in the presence of diabetes (type 1, type 2 or GDM). Influences of Obesity Insulin Resistance Maternal weight prior to pregnancy, weight gain during pregnancy, maternal height, number of previous pregnancies, and gender of the fetus all influence infant size at birth. Women who are obese prior to pregnancy tend to gain less weight and give birth to heavy babies (large for gestational age) more frequently than those whose pre-gravid weight is normal [12]. Catalano et al. [13] found that insulin sensitivity prior to pregnancy is an important predictor of birth weight and may account for the association between maternal pre-gravid weight and birth weight. In women with GDM, there is little if any correlation between maternal weight gain and birth weight; however, studies that examine associations with the weight of babies born to mothers with GDM are all potentially confounded by treatment of hyperglycemia. As indicated above, obesity is associated with an increased risk of large for gestational age (macrosomic) babies even when glucose levels remain normal [12]. Nevertheless, a collaborative study in Chicago, Ill., and Seoul, Korea, found that GDM was associated with a similar increment in the number of large babies above that of the general population in the more obese subjects in Chicago and in the mostly non-obese Korean population [14]. Nutrient Delivery Circulating concentrations of micro- and macronutrients (lipids, glucose and free amino acids) are altered during pregnancy and they are further modified by the presence of diabetes. More than 25 years ago, GDM was first characterized as a ‘panfuel’ metabolic disturbance with alterations in lipids (free 157

Metzger

FFA

␮mol/l

700

500

300

Triglyceride

mg/dl

400

300

200

140

Serine

␮mol/l

120

100

80

Isoleucine

␮mol/l

80

60

40

8:00

13:00

18:00

0:00

8:00

Meals at time (h)

Fig. 1. Diurnal patterns of plasma free fatty acids (FFA), triglycerides and individual amino acids (serine and isoleucine) in the third trimester of normal pregnancy (䊉) and ‘mild’ gestational diabetes (⌬; fasting plasma glucose ⬍105 mg/dl [5.8 mmol/l]). Adapted from Metzger et al. [15].

158

Optimal Management of GDM fatty acids (FFAs), triglycerides, lipoproteins), amino acids (especially branched chain amino acids) and other glucose-regulating hormones as well as glucose [15] (fig. 1). Obesity is also associated with changes in the circulating concentration of multiple nutrients (glucose, FFAs, triglycerides, cholesterol, branched chain amino acids), and it is likely that insulin resistance also plays a central role in these changes. For example, it is well known that normal pregnancy is associated with a decline in the fasting plasma glucose (FPG) concentration of approximately 0.5–0.6 mmol/l, and most of the change has occurred by the end of the first trimester. However, Mills et al. [16] found no decline in FPG in obese pregnant women (pre-pregnancy body mass index ⱖ30) with normal glucose tolerance. Yogev et al. [17] performed continuous glucose monitoring of subcutaneous interstitial fluid via a glucose sensor for 72 h in normal weight and obese non-diabetic pregnant women in the third trimester [17]. Compared to the normal weight subjects, the obese group had higher postprandial and mean values during the day, lower values during the night and similar fasting and pre-meal glucose concentrations.

Pathophysiology and Pathogenesis of GDM Normal pregnancy and GDM are states of profound insulin resistance and the mediating factors are similar, if not identical. However, women who develop GDM tend also to be insulin-resistant when not pregnant and become even more insulin-resistant during pregnancy [18]. GDM develops when ␤ cells are unable to compensate the increasing insulin resistance with sufficiently greater insulin secretion to maintain normal fasting and/or postprandial glucose concentrations [18, 19]. Placental Factors For many years, clinical and research evidence has indicated a role for the placenta in the development of insulin resistance and the list of putative mediators of this effect is long (table 2). The placenta is also an endocrine organ and numerous studies have demonstrated potential mechanisms for hormones of placental origin to contribute to insulin resistance. Furthermore, insulin sensitivity is restored quickly after the placenta is expelled. In the last several years, it has been demonstrated that the placenta synthesizes and potentially secretes an array of mediators and cytokines (e.g. leptin, C-reactive protein, tumor necrosis factor (TNF)-␣, interleukins 6 and 8) that may influence maternal metabolism and insulin sensitivity directly or indirectly [20]. Radaelli et al. [21] compared patterns of gene expression in placentas from pregnancies with normal glucose metabolism or GDM. They found major differences in the expression profiles with increases in markers and mediators of inflammation (interleukins, leptin and TNF-␣ receptors and downstream adaptors) in GDM. Whether these alterations of gene expression 159

Metzger Table 2. Factors of placental origin that putatively contribute to insulin resistance Large placenta size Human placental lactogen

Progesterone Human growth hormone (placental variant) Corticotrophin-releasing factor/cortisol Leptin Resistin

Human chorionic gonadotropin Prolactin Estradiol

Interleukin 6 Interleukin 8 Tumor necrosis factor-␣ C-reactive protein

⌬ Insulin sensittivity (10⫺2 mg·kg⫺1FFM·min⫺1/␮U·ml⫺1

2.5

0.0

⫺2.5

⫺5.0

⫺7.5

⫺10.0 ⫺1

0

1

2

3

⌬ TNF-␣ (pg/ml)

Fig. 2. The correlation between a change in TNF-␣ and a change in insulin sensitivity from before pregnancy to late pregnancy in subjects with normal carbohydrate metabolism (䊉) or gestational diabetes (䉬) is plotted. Adapted from Kirwan et al. [23, fig. 4] with permission.

represent primary events or occur as a secondary response to the increased delivery of nutrients, hormones and mediators from maternal tissues and their accumulation/storage in the placenta, e.g. triglycerides and glycogen [22], remains to be determined. Kirwan et al. [23] measured insulin sensitivity serially during pregnancy in a group of women with normal glucose tolerance or GDM and correlated these indices with measures of body composition, plasma TNF-␣, leptin cortisol and reproductive hormones. Levels of TNF-␣, leptin cortisol, fat mass, and all reproductive hormones measured were increased in late gestation compared to early in pregnancy; however, only a change in TNF-␣ from early to late pregnancy predicted the magnitude of change in insulin sensitivity over the same time (fig. 2). 160

Optimal Management of GDM Adipocyte-Derived Mediators Leptin is synthesized in both adipose tissue and placenta. Plasma leptin concentrations are increased during pregnancy. Its potential contribution to insulin resistance is uncertain [24]. Adiponectin is another adipocyte-derived protein that may also play a role in insulin resistance and predisposition to altered glucose metabolism. Adiponectin levels tend to be low in insulinresistant states including GDM and have been found to correlate with ␤-cell function in late pregnancy [25].

Treatment of GDM Rationale As pointed out above, there is considerable controversy about ‘the level of hyperglycemia, short of overt diabetes that conveys increased perinatal risk’ [6]. However, there is a general consensus that lowering maternal capillary blood glucose concentrations to ⱕ95 mg/dl (5.3 mmol/l) in the fasting state, ⱕ140 mg/dl (7.8 mmol/l) at 1 h or ⱕ120 mg/dl (6.7 mmol/l) 2 h after meals may reduce the risk of excessive fetal growth to approximate the risk in the general population [1], and a major objective of treating GDM is to reduce adverse perinatal events, primarily those associated with excess weight or adiposity of the newborn (cesarean delivery, birth trauma, neonatal morbidities). Goals The primary goal that is most often articulated for the treatment GDM is the restoration of fasting and post-meal glucose values to within normal ranges. Some investigators have provided evidence that it is more ‘cost-effective’ to assess fetal abdominal circumference (AC) by ultrasound in all women with GDM to identify those at low risk of having a large baby (AC ⬍75th percentile) and concentrate therapeutic efforts on those at a much higher risk of delivering a large baby (AC ⱖ75th percentile) [26]. Metabolic Management Lifestyle Modification Nutritional Therapy. Medical nutrition therapy (MNT) is referred to as the ‘cornerstone’ of medical management of GDM; however, relatively little information is available to allow evidence-based recommendations for specific dietary approaches to management of GDM [1]. Furthermore, MNT of GDM is not presently a major focus of research. The clinical guidelines for micro- and macronutrient intake and desired maternal weight gain closely parallel those that are recommended for ‘normal, healthy pregnancies’ [9, 10]. The significant differences are in recommendations concerning sources of dietary carbohydrate, with emphasis placed on increasing complex carbohydrates and reducing or eliminating monosaccharide, sucrose and other 161

Metzger oligosaccharides. Calorie restriction aimed at achieving weight loss (the standard initial dietary recommendation when not pregnant) continues to be viewed as experimental for GDM. Exercise Therapy. A number of investigators have demonstrated that exercise can effectively improve glycemia in GDM. The optimal frequency and intensity of exercise have not been determined and the overall impact on perinatal and long-term outcomes has not been fully assessed. With 3 or more sessions of exercise (⬎15 min duration), the maximum impact on maternal glucose levels may not be seen for 2–4 weeks [1]. Intensified Metabolic Management Failure to achieve or to maintain maternal glycemia within the targets mentioned above after initiating MNT, or signs of excessive fetal growth indicate the need for more intensive metabolic therapy. Historically, treatment with insulin has been used in such instances [1]. There is no formally accepted standard protocol for therapy of GDM with insulin. Beneficial effects of therapy with insulin have been reported with a variety of regimens, including once daily administration of a fixed dose of ‘NPH’ insulin. It is likely that twice daily use of combinations of biosynthetic ‘Regular’ and ‘NHP’ human insulin (in premixed fixed or flexible proportions) represents the most widely used algorithm. In the last decade, additional treatment options have been advocated, including use of ‘rapid-acting’ insulin analogs (lispro and aspart, glulisine insulins), a long-acting insulin analog (insulin glargine), or oral anti-diabetic medication (sulfonylurea [glyburide], metformin). The advent of new and/or alternate therapeutic options raises issues of cost/benefit, appropriate endpoints, safety, and efficacy. As implied in the Introduction, the time may be at hand to develop indicators or measures of optimal metabolic control that go beyond indices of hyperglycemia alone. For example, do subjects with similar levels of FPG or postprandial glucose values that are achieved by different interventions also have similar levels of FFA, other lipids, individual amino acids, etc., and are the outcomes comparable? Few studies have been designed or analyzed to address questions such as these. However, Langer et al. [27] have recently reported better perinatal outcomes (i.e, lower rate of large for gestational age babies, less macrosomia (birth weight ⱖ4,000 g), lower rate of cesarean delivery, and less frequent neonatal morbidities assessed by a ‘composite outcome’ score), in pregnancies of obese GDM subjects (pre-pregnancy body mass index ⱖ30) treated with insulin and ‘well-controlled’ than in ‘well-controlled’ diet-treated obese subjects, even though the initial and final glycemic measurements did not differ the 2 groups of subjects. Rapid-Acting Insulin Analogs The rapid-acting insulin analogs (lispro, aspart and glulisine insulin) play an established role outside pregnancy in the management of type-1 DM and 162

Optimal Management of GDM they are commonly used in patients with type-2 DM who require multiple daily injections. When used prior to pregnancy, rapid-acting analogs are commonly continued during gestation although their safety and efficacy during pregnancy have not been established by controlled clinical trials. The use of these analogs during pregnancy varies in different regions and among individual physicians. The potential use of these analogs in the management of GDM has been explored to a limited extent. Jovanovic et al. [28] found no difference in immunological response to the administration of regular human insulin or lispro insulin (Humalog™) in women with GDM, and lispro insulin was not detectable in cord blood. The area under the glucose curve following a meal test was less and episodes of hypoglycemia fewer with lispro insulin. Human NPH insulin was used to provide ‘basal’ needs. No fetal or neonatal abnormalities were noted in either treatment group [28]. Data are not currently available from pregnancies in which intermediate (insulin detemir) and/or long-acting analogs (insulin glargine) have been used in the treatment of GDM. The cost of treatment with an insulin analog is considerably greater than the cost of using regular and NPH human insulin. Therefore, the safety, efficacy and clinical benefit should be demonstrated before the wide use of insulin analog therapy of GDM is endorsed. Medical Therapy with Glyburide In the past, the American Diabetes Association [29] and the American College of Obstetricians and Gynecologists [30] have recommended that oral medications not be administered for the treatment of GDM because of concerns about increased risks of congenital malformations and neonatal hypoglycemia that might be associated with their use. After finding very limited transplacental passage of glyburide in perfusion studies with term human placentas [31], Langer et al. [32] performed a randomized clinical trial comparing the results of treating GDM (meeting their criteria for use of ‘intensified medical therapy’) with insulin and glyburide. The primary endpoint, glycemic control, was comparable in the 2 groups, perinatal outcomes did not differ and glyburide was not detectable in cord blood taken at delivery. Only 8/201 (4%) subjects in the glyburide treatment group failed to have an adequate glycemic response requiring transfer to therapy with insulin. The conclusion was reached that ‘in women with gestational diabetes, glyburide is a clinically effective alternative to insulin therapy’ [32]. These results have generated cautious optimism about the potential use of this oral agent in the treatment of GDM. Although additional studies of glyburide treatment by others have followed, to date none has involved a sample approaching the size used in the initial clinical trial. Medical Therapy with Metformin For nearly 50 years, metformin has been used in the treatment of type-2 DM. Reports of its limited use during pregnancy have intermittently been 163

Metzger published. Although substantial amounts of metformin cross the placenta, in recent years interest in its use in childbearing women has increased substantially. A major factor in this trend is the evidence that administration of metformin to women with the polycystic ovary syndrome enhances ovulation and improves fertility. Some reports also suggest that administration of metformin during pregnancy as well as during induction of ovulation can reduce fetal loss and lessen the risk of developing GDM. A prospective, double-blind, randomized, placebo-controlled, clinical trial has not been completed to address the issues regarding the benefit and safety of exposure of the mother and fetus to metformin for the duration of pregnancy [33]. The potential effects in the offspring must be evaluated through childhood and puberty into adulthood. There is also renewed interest in the use of metformin for the treatment of some patients with type-2 DM or GDM. In Australia and New Zealand, a randomized clinical trial of metformin versus insulin treatment of GDM has been initiated and guidelines have been published in Australia regarding the potential use of metformin in GDM subjects poorly controlled with diet but refusing therapy with insulin, and as an adjunct to insulin in the presence of severe obesity and very severe insulin resistance [34]. As mentioned above, the potential effects in the offspring must be evaluated through childhood and puberty into adulthood.

References 1 Metzger BE, Coustan DR: Summary and recommendations of the Fourth International Workshop-Conference on Gestational Diabetes Mellitus. The Organizing Committee. Diabetes Care 1998;21(suppl 2):B161–B167. 2 Ferrera A, Kahn HS, Quesenberry CP, et al: An increase in the incidence of gestational diabetes mellitus: Northern California 1991–2000. Obstet Gynecol 2004;103:526–533. 3 Carpenter MW, Coustan DR: Criteria for screening tests for gestational diabetes. Am J Obstet Gynecol 1982;144:763–773. 4 Dabelea D, Snell-Bergeon JK, Hartsfield CL, et al: Increasing prevalence of gestational diabetes mellitus (GDM) over time and by birth cohort. Diabetes Care 2005;28:579–584. 5 Seshiah V, Balaji V, Balaji MS, et al: Gestational diabetes mellitus in India. J Assoc Physicians India 2004;52:707–711. 6 HAPO Study Cooperative Research Group: The Hyperglycemia and Adverse Pregnancy Outcome (HAPO) Study. Int J Gynaecol Obstet 2002;78:69–77. 7 Crowther CA, Hiller FE, Moss JR, et al, for the Australian Carbohydrate Intolerance Study in Pregnant Women (ACHOIS) Trial Group: Effect of treatment of gestational diabetes mellitus on pregnancy outcomes. N Engl J Med 2005;352:2477–2486. 8 Silverman BL, Purdy LP, Metzger BE: The intrauterine environment: implications for the offspring of diabetic mothers. Diab Rev 1996;4:21–35. 9 Energy and Protein Requirements. Report of a joint FAO/WHO/UNU Expert Consultation. World Health Organ Tech Rep Ser 1985;724:1–206. 10 Institute of Medicine and Food and Nutrition Board: Nutrition during Pregnancy. Washington, National Academy Press, 1990. 11 Hornstra G, Uauy R, Yang X (eds): The Impact of Maternal Nutrition on the Offspring. Nestlé Nutrition Workshop Series Pediatric Program. Vevey/Basel, Nestec/Karger, 2005, vol 55.

164

Optimal Management of GDM 12 Kliegman RM, Gross T: Perinatal problems of the obese mother and her infant. Obstet Gynecol 1985;66:299–306. 13 Catalano PM, Drago NM, Amini SB: Maternal carbohydrate metabolism and its relationship to fetal growth and body composition. Am J Obstet Gynecol 1995;172:1464–1470. 14 Jang HJ, Cho NH, Min Y-K, et al: Increased macrosomia and perinatal morbidity independent of maternal obesity and advanced maternal age in Korean women with gestational diabetes mellitus. Diabetes Care 1997;20:1582–1593. 15 Metzger BE, Phelps RL, Freinkel N, Navickas IA: Effects of gestational diabetes on diurnal profiles of plasma glucose, lipids, and individual amino acids. Diabetes Care 1980;3:402–409. 16 Mills JL, Jovanovic L, Knopp R, et al: Physiological reduction in fasting plasma glucose concentration in the first trimester of normal pregnancy: the Diabetes in Early Pregnancy Study. Metabolism 1998;47:1140–1144. 17 Yogev Y, Ben-Haroush A, Chen R, et al: Diurnal glycemic profile in obese and normal weight nondiabetic pregnant women. Am J Obstet Gynecol 2004;191:949–953. 18 Catalano PM, Tyzbir ED, Roman NM, et al: Longitudinal changes in insulin release and insulin resistance in nonobese pregnant women. Am J Obstet Gynecol 1991;165:1667–1672. 19 Buchanan TA, Xiang AH: Gestational diabetes. J Clin Invest 2005;115:485–491. 20 Coughlan MT, Permezel M, Georgiou HM, Rice GE: Repression of oxidant-induced nuclear factor-kappaB activity mediates placental cytokine responses in gestational diabetes mellitus. J Clin Endocrinol Metab 2004;89:3584–3594. 21 Radaelli T, Varastehpour A, Catalano P, Haugel-De Mouzon S: Gestational diabetes induces placental genes for chronic stress and inflammatory pathways. Diabetes 2003;52:2951–2958. 22 Diamant YZ, Metzger BE, Freinkel N, Shafrir E: Placental lipid and glycogen content in human and experimental diabetes mellitus. Am J Obstet Gynecol 1982;144:5–11. 23 Kirwan JP, Haugel-De Mouzon S, Lepercq J, et al: TNF-alpha is a predictor of insulin resistance in human pregnancy. Diabetes 2002;51:2207–2213. 24 Ceddia RB, Koistinen HA, Zierath JR, Sweeney G: Analysis of paradoxical observations on the association between leptin and insulin resistance. FASEB J 2002;16:1163–1176. 25 Retnakaran R, Hanley AJG, Raif N, et al: Adiponectin and beta cell dysfunction in gestational diabetes: pathophysiological implications. Diabetologia 2005;48:993–1001. 26 Buchanan TA, Kjos SL, Montoro MN, et al: Use of fetal ultrasound to select metabolic therapy for pregnancies complicated by mild gestational diabetes. Diabetes Care 1994;17:275–283. 27 Langer O, Yogev Y, Xenakis EM-J, Brustman L: Overweight and obese in gestational diabetes: the impact on pregnancy outcome. Am J Obstet Gynecol 2005;192:1768–1776. 28 Jovanovic L, Ilic S, Pettitt DJ, et al: Metabolic and immunologic effects of insulin lispro in gestational diabetes. Diabetes Care 1999;22:1422–1427. 29 American Diabetes Association: Gestational diabetes mellitus. Diabetes Care 2003;26(suppl 1): S103–S105. 30 Diabetes and Pregnancy. ACOG Technical Bulletin No. 200 (replaces No. 92). Washington, American College of Obstetricians and Gynecologists, 1994, pp 359–366. 31 Elliott BD, Langer O, Schenker S, Johnson RF: Insignificant transfer of glyburide occurs across the human placenta. Am J Obstet Gynecol 1991;165:807–812. 32 Langer O, Conway DL, Berkus MD, et al: A comparison of glyburide and insulin in women with gestational diabetes mellitus. N Engl J Med 2000;343:1134–1138. 33 Ehrmann DA: Polycystic ovary syndrome. N Engl J Med. 2005;352:1223–1236. 34 Simmons D, Walters BNJ, Rowan JA, McIntyre HD: Metformin therapy and diabetes in pregnancy. Med J Aust 2004;180:462–464.

Discussion Dr. Katsilambros: It was very interesting to note that in your initial slides it was clearly shown that Chinese women presented much higher gestational diabetes as compared to Caucasians. This is very curious since the diabetes rate per se in the general Chinese population is lower than that in Caucasians. Could it be that this reflects the fact that Chinese women are thinner? You have shown during pregnancy thin

165

Metzger women tend to gain a lot of weight as compared to women who are heavier before pregnancy. Could it also be that Chinese women produce relatively more estrogens during pregnancy or even that their tissues are relatively more sensitive to estrogens than the tissues of Caucasian women? Please also take into account the fact that in China only one child per woman is the common policy. I say that because we know that the greater the number of pregnancies the greater the diabetic tendency. Dr. Metzger: I failed to make it clear that these Chinese and other Asian women were people who were born in China or Vietnam, and then migrated and lived in Australia. So the Westernization effect of the first generation is what we are seeing. We have seen the same thing in Korean women moving to Chicago. So the high expression of diabetes is seen in the new environment. However, in other countries I do believe that gestational diabetes is also increased. There have been reports of an increase in India, although I have to say that these population-base studies are not adjusted for confounding factors as those from California. In studies that have looked at this, greater weight is associated with gestational diabetes even within populations with relatively little obesity. So it looks as if the same factors worsen their insulin resistance as occur in the areas where there is more obesity. Ms. Franz: Traditionally the focus of nutrition therapy for gestational diabetes has been on blood glucose control and glucose outcomes. Are you suggesting that weight control and/or caloric intake should receive more attention? Dr. Metzger: I would like to see them raised. Yes, in terms of other outcomes I believe in judging if glyburide therapy is equivalent to insulin, or is one form of insulin equivalent to another, we probably should be looking at some of the other metabolic changes that are part of diabetes and gestational diabetes as well as perinatal outcomes to help us determine what really does normalize the environment most effectively. Dr. Slama: Have you any experience with increased physical activity as promoted by Jovanovic in the past? More and more of our patients are treated with glargine and ultra-rapid analogs. What do you do when such perfectly controlled diabetic girls become pregnant while using glargine and analogs? Dr. Metzger: There have been two approaches to physical activity in published studies. One is the upper arm approach that Jovanovic has used, and other groups simply used walking and more traditional forms of increasing physical activity with similar effects on improving glucose levels in patients with gestational diabetes. So I believe the full component of lifestyle intervention should be encouraged more than we have done in the past. Your question with regard to analogs deals more with people with existing diabetes, and my approach has been as follows. The short-acting insulin analogs have now been used for a decade although some initial concerns were raised. There have been no systematic clinical trials, but there also has been no evidence that they have any adverse effects. So women who are using short-acting analogs can continue that with very little concern and very little additional discussion. With the long-acting analogs such glargine insulin, I would encourage my colleagues to discuss the pros and cons of their use during pregnancy with the women before they become pregnant. There have been no clinical trials, although there are some animal data that have raised concern that high doses of glargine might increase the levels or might interact with IgF receptors in addition to insulin receptors that could potentially be harmful. There are no published data in large numbers. so the safety issues are not fully resolved. I encourage people to have a discussion with their female patients before glargine insulin is started as to whether they would want to use it during pregnancy. Women that have been brought in good control and are comfortable with a given regimen, find it very disconcerting to be asked to make a change. So if it can possibly be done before pregnancy, it is much easier. Having said

166

Optimal Management of GDM that, I have never had anybody who wanted to discontinue glargine insulin once the pregnancy has started. Dr. Chiasson: I was surprised by your conclusion suggesting that the glucagonlike peptide-1 inhibitor of dipeptidyl peptidase-4 (DPP4) could be one of the tools we could use in the future because the enzyme DPP4 is involved in a number of metabolic pathways, and the safety has yet to be proven. In the future we may find that it is safe. In fact gestational diabetes, at least at the beginning, is a postprandial disease. I was wondering why you would not suggest an approach that specifically addresses the postprandial plasma glucose rise, such as ␣-glucosidase inhibitors, which could be tested? These seem, at least certainly ␣-glucosidase inhibitors, to be safe, non-toxic and most likely without any detrimental effects, and could be a safe drug for gestational diabetes. Dr. Metzger: I am not really optimistic that we will be using DPP4 inhibitors in the near future unless they are very specifically known not to cross the placenta. However, glucagon-like peptide-1 agonists might be approachable in the near future and would help restore insulin secretion. Other agents have the burden of transplacental delivery, and ␣-glucosidase inhibitors are absorbed to a limited extent. They carry warnings for pregnancy because the 2 or 3% that is absorbable is hepatically active and potentially hepatotoxic in large doses. So they also face the burden of concern for introduction in pregnancy. Dr. Velarde: You mentioned that some of the patients who were diagnosed with gestational diabetes already had substantial weight gain at the time of diagnosis. This implies that perhaps they had some insulin resistance going on for weeks, maybe at the time of conception. While the interventions that you mentioned may prevent the respiratory complications, macrosomia, large for gestational age infants, do you think we should be more aggressive with the early diagnosis and treatment of these patients to prevent the congenital malformations? Dr. Metzger: Congenital malformations that occur in women who we label as having gestational diabetes are probably related to one of two things. One, they had diabetes before pregnancy and had overt hyperglycemia at the time they became pregnant which wasn’t detected. Second, there is increasing evidence that obesity, independent of hyperglycemia, is associated with a higher risk of malformations. The true mediators of those malformations are not known. The background against which we need to compare malformations in gestational diabetes are women of the same weight and other characteristics who don’t have abnormal glucose levels. I do think that many women in whom we diagnose or who even develop gestational diabetes over the course of pregnancy have abnormalities earlier than we can detect. Decisions concerning the time, or multiple times during pregnancy to screen for gestational diabetes are made pragmatically. We have many places in the world where it is not feasible to test once for gestational diabetes. So it is hard to make a set of recommendations to do earlier screening when in many instances it is not feasible to even screen once. The focus on screening at around 28 weeks of gestation is based on a pragmatic yield versus cost basis with still enough time to intervene for potential benefit. Dr. T. Wilkin: At the NIH earlier this year in Bethesda there was a presentation suggesting that there may be gene methylation of the germline of the offspring that may have long-term transgenerational effects. I just wonder if you could comment on whether you look beyond the next generation, to the one after that, and after that, because the impact of hyperglycemia and insulin resistance in the gestationally diabetic female doesn’t just stop with the early effects on the offspring. Dr. Metzger: This is a topic about gestational diabetes that I am very interested in. There is certainly strong evidence that the exposure to the intrauterine environment of diabetes has lasting consequences. The animal model data have supported

167

Metzger this for a couple of decades; there have been very good epidemiological studies from the Pima population that Dabelea et al. [1] conducted, and we have done a long-term study of children from diabetic mothers, a cohort we followed for 20 years [2]. The findings are complementary. So there is strong evidence that exposure in the intrauterine environment to diabetes accelerates or promotes both obesity and glucose intolerance in the next generation. I believe that preventing the long-term consequences of diabetes is as important as the perinatal issues. We are pretty successful with the perinatal issues, we don’t know that we can successfully intervene to prevent the long-term effects. Dr. Hill: Are high levels of physical activity or high physical fitness protective against gestational diabetes? Dr. Metzger: I think they are. Of course the data come from long-term observational studies, not from prospective trials of any kind. I believe the Nurses Health Study has shown a reduced prevalence in gestational diabetes associated with higher levels of physical activity [3]. Many years ago I was in communication with an investigator in an African country where physical activity remained extremely high and women remained in the field until they delivered and came back to the field a few hours later. They performed a series of studies during pregnancy in which there was very little weight gain, no increase in body fat, and glucose tolerance tests which showed no change in glucose tolerance at all during pregnancy. So I suspect that we can probably intervene under certain circumstances to reduce the amount of insulin resistance that occurs during normal pregnancy, and that would in a similar way probably reduce the incidence in gestational diabetes. Dr. Jianqin Sun: Do you have any data on the prevalence of the macrosomia among diabetic mothers and normally pregnant women? Dr. Metzger: The incidence of macrosomia in normal pregnancy varies a lot in different populations around the world, and one of the major factors is the amount of obesity and insulin resistance in the normal population. In patients with gestational diabetes, the observed frequency of macrosomia is a very complex thing to interpret because once we make a diagnosis we intervene with treatment that is intended to prevent the macrosomia that we measure as an outcome. So there are very few data that would say how much macrosomia would occur in untreated gestational diabetes. Where untreated gestational diabetes with more severe hyperglycemia exists, the incidence is very high. In the hyperglycemia and adverse outcome study that we are currently doing, we will have data on how much macrosomia occurs with lesser levels of hyperglycemia because that is a purely observational study. Dr. Knowler can you add something with regard to outcomes in untreated Pima? Dr. Knowler: We have been studying pregnancy and pregnancy outcome in the Pima Indians for a period of about 40 years during which time there have been tremendous improvements in the treatment of diabetic pregnancy. Unfortunately some of the adverse outcomes such as the risk of diabetes in the offspring, do not show any sign of declining overtime. So I agree with the conclusion you made, treatment of diabetes in pregnancy certainly can be effective in reducing a lot of the perinatal outcomes but it is not clear whether it would be effective in the long-term adverse outcomes in the children.

References 1 Dabelea D, Knowler WC, Pettitt DJ: Effect of diabetes in pregnancy on offspring: follow-up research in the Pima Indians. J Matern Fetal Med 2000;9:83–88.

168

Optimal Management of GDM 2 Silverman BL, Rizzo TA, Cho NH, Metzger BE: Long-term effects of the intrauterine environment. The Northwestern University Diabetes in Pregnancy Center. Diabetes Care 1998;21(suppl 2):B142–B149. 3 Solomon CG, Willett WC, Carey VJ, et al: A prospective study of pregravid determinants of gestational diabetes mellitus. JAMA 1997;278:1078–1083.

169

Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 171–181, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

Do Meal Replacement Drinks Have a Role in Diabetes Management? Herwig H. Ditschuneit Medizinische Klinik, Universitätsklinikum Ulm, Ulm, Germany

Abstract The poor effectiveness of conventional dietary treatment for weight loss and weight maintenance in patients with type-2 diabetes may be improved by a meal replacement strategy that provides a strong structured meal plan with reasonable opportunity for dietary variety. Typical meal replacement programs fix the intake of one or two meals per day with a calorie-controlled, nutritionally balanced commercial formulation, and allow prudent additional meals and snacks. In obese subjects, diets with meal replacements have proven to be more efficient than conventional diets. Patients on the meal replacement regimen lost 7.3 and 8.4% of initial body weight after 12 weeks and 4 years, respectively, whereas the patients on the conventional diet had lost 1.4% and 3.2% of initial body weight after 12 weeks and 4 years, respectively. The meal replacement plan has also proven to be effective in patients with type-2 diabetes. After 6 and 12 months, patients in the meal replacement group achieved on average a weight loss of 5.24 and 4.35% of their initial body weight, respectively. In contrast, after 6 and 12 months, patients on the individualized diet plan achieved on average a weight loss of 2.85 and 2.36% of their initial body weight, respectively. Meal replacements offer a promising strategy for treating obese patients with type-2 diabetes. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

Obesity and Type-2 Diabetes Epidemiological studies show that increasing body weight is associated with an increasing risk for type-2 diabetes (T2D) [1]. More than 90% of individuals with T2D are obese. Conversely the prevalence of T2D is 46% among individuals with a BMI of 30 (kg/m2) and higher. Obesity is a contributing factor in the development of T2D in an estimated 60–90% of patients with this condition [2]. The data indicate that the current epidemic obesity may be the major causative factor in the worldwide increase of the prevalence of diabetes. 171

Ditschuneit Excess body fat, in particular surplus of visceral fat, is recognized to result in metabolic disturbances. The amount of intra-abdominal fat and the waist circumference have proven as independent predictors of the incidence of T2D [3].

Prevention of T2D There are numerous studies giving evidence that weight loss in obese subjects and particularly in obese subjects with impaired glucose tolerance (IGT) may prevent development of T2D. In a small well-performed study in Malmö Sweden [4] overweight men with IGT were enrolled in a 5-year intervention program emphasizing a low-calorie diet and physical training. At the 5-year follow-up participants showed a mean weight loss of 2.3 kg whereas control subjects showed a mean weight gain of 0.5 kg. In the intervention group 75.8% showed improved glucose tolerance whereas 10.6% progressed to T2D. In contrast, in the control group 67.1% showed deterioration in glucose tolerance and 28.6% progressed to T2D. In agreement with this study in large diabetes prevention trials with lifestyle changes, including calorie restriction and encouragement to increase physical activity, the incidence of diabetes was reduced in the lifestyle intervention groups. In the Da Qing IGT and Diabetes Study [5] the cumulative incidence of diabetes at 6 years in subjects with IGT at baseline was 67.7% in the control group. In the diet, exercise, and diet-plus-exercise intervention groups the risk of developing diabetes compared to the control group was reduced by 31, 46 and 42%, respectively. In the Finland Study [6], the effect of the intervention on the incidence of diabetes was most pronounced among subjects who made comprehensive changes in lifestyle. Compared to subjects undergoing typical care (control group) in those with weight loss the incidence of diabetes was reduced by more than 50%. In the Diabetes Prevention Program [7] the incidence of diabetes in persons at risk after a mean follow-up of 2.8 years was 11.0 cases per 100 personyears in the placebo group but only 4.8 in the lifestyle group.

Effects of Weight Loss in T2D In patients with T2D weight loss has been demonstrated to improve biomarkers of good health, metabolic disturbances, HbA1c and vascular complications [8]. The most striking results have been reported after massive weight loss induced by surgery. In one study [9], before surgery 49% of the patients had IGT or diabetes and after 14 years with a mean weight loss of 32.8% more than 90% maintained normal levels of plasma glucose. 172

Do Meal Replacement Drinks Have a Role in Diabetes Management? In the Swedish SOS Study [10], after 10 years in the surgery group, 36% of the diabetic patients had recovered from diabetes; in contrast in the control group only 13% of patients had recovered. In the surgery group after 10 years 7% of patients and in the control group 24% of patients had developed diabetes. These data emphasize the importance of obesity contributing to the pathogenesis of T2D and provide the rationale for developing appropriate management strategies for patients with T2D. It may be difficult to sort out benefits that can be attributed to medical nutrition therapy, diabetes education, and exercise. There are three metaanalysis studies looking at diabetes education and a variety of weight loss methods showing that nutrition interventions have the largest effect on weight loss. The weight loss again improves the metabolic control. This has been demonstrated impressively in the Diabetes Treatment Study from Northern Ireland [11] in which a successful long-term weight low was achieved with diet. The average weight loss of 9 kg after 6 months of treatment was sustained over the 6-year study. All the participating patients had recently diagnosed T2D, and diabetes was managed by diet alone in 87% of the patients at 1 year and 71% at 6 years.

Medical Nutrition Therapy in Diabetes On the basis of the medical literature and sound clinical practice, medical nutrition therapy with effective weight reduction improves glycemic control, metabolic disturbances and vascular complications in T2D. But current nutritional recommendations for patients with T2D in general do not attach great importance to weight loss. The goals of medical nutrition therapy aimed at by the ADA focus on attaining and maintaining optimal metabolic outcomes, on preventing and treating the chronic complications of diabetes, on improving health through healthy food choices and physical activity, and on addressing individual nutritional needs [12]. There is evidence that all these goals are easily reached or somewhat better reached with a reduction of body weight. However, it seems to be a common belief among physicians that nutritional therapy in T2D is not efficient. To optimize glycemic control physicians often start with drug or insulin therapy before the effects of diet treatment are assessable [13]. In view of recent consistent and strong evidence that weight reduction improves insulin sensitivity and glycemic control in T2D, the Joslin Diabetes Center and Joslin Clinic recommended clinical guidelines for overweight and obese adults with T2D, prediabetes or high risk of developing T2D that consider weight reduction as one of the prime objectives of any nutrition recommendations for patients with diabetes. Among others the guidelines advise that weight reduction should be individualized and continued until BMI reaches ⱕ25 kg/m2 or an until a BMI goal which has been agreed 173

Ditschuneit upon is reached, that individuals should learn and practice portion control as an effective way of weight control, and that meal replacements may be used [14].

Management of Weight Loss in Diabetics It has been reported that achievement of weight loss and weight maintenance in patients with T2D seems to be more difficult than in people without diabetes. Poor adherence to the dietary recommendations and physiological adaptations that occur with dieting [15] may explain the poor outcomes. In addition, intense blood glucose control in diabetics may counteract with weight loss efforts. The results of the UK Prospective Diabetes Study show a significant increase in weight in the intensive-treated group compared with the conventional diet group by 3.1 kg for the cohort at 10 years [16]. There is a wide range of various dietary treatments which are offered by health professionals for the treatment of obese patients with T2D. Most dietary treatments aim at a reduction of energy intake below energy needs. However, despite giving detailed recommendations patients find compliance extremely difficult and the effectiveness of standard dietary treatment in general appears to be poor, particularly in the long term. Thus, for weight loss in diabetic patients, programs are needed to increase the amount of weight loss and to facilitate and improve long-term weight maintenance.

Very Low Calorie Diets One approach for effective weight reduction are very low calorie diets (VLCDs). These diets provide 600–800 kcal/day given as calorie-controlled, vitamin and mineral-fortified liquid meals taken as the sole nutrient source. VLCDs have been proven effective in obese patients and also in obese subjects with T2D. In diabetic patients VLCDs induce weight loss and improve glycemic control. A meta-analysis showed that treatment with VLCDs over 12 weeks decreased body weight by 9.6% of initial body weight and fasting plasma glucose values decreased to about 50% of initial values after 2 weeks and remained low for the 12 weeks of treatment [8]. Concomitant with the decrease of plasma glucose concentrations glycosylated hemoglobin levels and insulin levels decreased. Furthermore, excellent acceptance, compliance and safety are documented [17]. While VLCDs in obese subjects with T2D induce remarkable results of short-term weight loss, some observations show that VLCDs despite substantial initial weight loss do not maintain weight loss in the long term [18]. It can 174

Do Meal Replacement Drinks Have a Role in Diabetes Management? be assumed that the liquid diets will not be maintained long-term because subjects on VLCDs are excluded from their normal daily activities.

Meal Replacements The poor effectiveness of conventional dietary treatment may be improved by use of a meal replacement program. Incorporating meal replacements into traditional lifestyle interventions has been proven as a successful strategy for weight control in obese subjects. Meal replacements are characterized by liquid formulas, powder formulations reconstituted with water or milk, or nutrition bars that are typically fortified with vitamins and minerals and with or without fiber. Meal replacements as a tool for weight management is very popular among people trying to lose weight. In a dietician-led program more than 60% of subjects chose meal replacements at least once daily as their preferred weight loss strategy [19]. In a typical meal replacement plan one or two full-energy meals are replaced by a low-energy, nutritionally balanced product, while one meal is prepared with prudent self-selected conventional food. Snacks are replaced by nutrition bars or a low-fat dairy product or a piece of fruit. A weight loss program with meal replacements in place of one or two daily meals has been shown to improve compliance with an energy-restricted diet in comparison with simply providing food plans [20]. A meal replacement plan has proven significantly more effective than conventional diet plans in obese subjects [21]. The use of a meal replacement plan was also effective in overweight and obese subjects working in highstress jobs (police, hospital health professionals, flight crew members and firefighters). After 12 months of treatment, the study participants had lost 8.4% of their initial body weight and 1 year later 66% of the firefighters and aviation personnel had retained ⬎80% of their weight loss at week 12 and half of these weighed less than at week 12 [22]. In a meta- and pooling analysis of 6 studies weight loss induced by a partial meal replacement plan at 3 months and at 1 year exceeded the weight loss induced by a conventional diet plan. The weight loss of those completing the study ranged between 2.61 and 4.35 kg (⬃3.7% of initial body weight) in the conventional energy-restricted diet group, and between 6.97 and 7.31 kg (⬃7.8%) in the meal replacement group. At the 1-year evaluation 64% of the subjects in the conventional diet group had dropped out compared to 47% of subjects in the meal replacement group [20]. The magnitude of weight loss in the partial meal replacement group at 1 year was in the range of that often observed in effective pharmacologic weight control studies. Recently, in a review evaluating the role of meal replacements in obesity treatment, the conclusion was drawn that meal replacements are a valid alternative dietary strategy in the treatment of obesity and may aid long-term maintenance of weight [23]. 175

Ditschuneit A meal replacement plan has also proven effective in subjects with T2D. Weight loss with a liquid meal replacement plan was compared with an isocaloric energy-restricted diet. Maximum weight loss was achieved in both programs after 3 months and glycemic control improved in both groups of patients. However, at 1 year between the groups there were no significant differences in weight [24]. In another study with a meal replacement plan for 12 weeks subjects were randomized into three groups using either a meal replacement containing lactose, fructose, and sucrose, a meal replacement in which fructose and sucrose were replaced with oligosaccharides, or an exchange diet plan recommended by the ADA [25]. Weight loss was significantly greater in the meal replacement groups than in the group with an exchange diet plan. During weight loss, there were reductions in glucose, insulin and HbA1c concentrations. In addition, the oral antidiabetic medications were reduced in some patients. Thus, it has been shown in this clinical study that meal replacements can be used safely as a part of a comprehensive treatment program for T2D patients. In a meta- and pooling analysis [20], about 20% of the study population at baseline was diabetic. There was no difference in weight loss between diabetic and nondiabetic subjects at 3 months. However, patients with diabetes did not maintain their weight loss at 1 year to the same extent as nondiabetic subjects. These findings are consistent with earlier weight loss studies of diabetic patients that have also shown a reduced long-term weight loss compared to nondiabetic patients [26]. The long-term efficiency of a meal replacement plan based on soy-based meal replacements was evaluated in a randomized prospective study in type-2 diabetics. The study compared the effects of a meal replacement plan with the effects of an individualized diet plan, as recommended by the ADA [27]. In this study, weight loss was achieved in both diet groups over the 12 months’ study period. In the meal replacement group weight was reduced by 5.2 and 4.3 kg at 6 and 12 months, respectively; in the individualized diet plan group weight was reduced by 2.8 and 2.3 kg at 6 and 12 months, respectively. In the meal replacement group, glucose concentrations were lower than in the individualized diet group at 3, 6 and 12 months, respectively. The levels of HbA1c in the individualized diet group were reduced but were not significantly lower than at baseline; in contrast in the meal replacement group the mean HbA1c level was reduced and significantly lower than in the individualized diet group for the entire study period.

Why Diets with Incorporated Meal Replacements Improve Weight Loss and Weight Maintenance Meal replacements are very popular in US adults trying to lose weight. According to a telephone survey, 15% of women and 13% of men were using 176

Do Meal Replacement Drinks Have a Role in Diabetes Management? meal replacements for weight loss [28]. There are various reasons for this popularity. Meal replacements help people lose weight by providing a controlled amount of calories and fat in a prefixed portion. They are convenient, easy to purchase, easy to store, and require little preparation. They are available nearly everywhere and reasonably prized. Meal replacements reduce the number of decisions people have to make daily about what to eat and reduce their exposure to tempting foods that might result in overeating. Two meals are replaced, but the one remaining meal offers the opportunity to maintain a normal life, reduces barriers to dietary adherence, and helps to instill regular eating patterns. In addition, the remaining meal may increase the accuracy of calorie estimation and estimation of portion size, provides a simple guide for healthy nutrition, and positively affects self-monitoring. Meal replacements incorporated in the diet provide the nutrient composition necessary for longterm weight maintenance and healthy nutrition. Within the weight management programs the meal replacement plan is allowing varying degrees of flexibility and variety in eating. The intake of one or two meals per day with a nutritionally balanced formula means a strong structure and the self-selection of additional meals and snacks means sufficient dietary variation. The meal replacement plan is a logical and easy-to-use plan for patients to follow, as supported by strong evidence.

Outlook In contrast to the standard dietary approach, diet plans which include meal replacements have been proved as an appropriate tool for weight reduction and for long-term weight loss maintenance in obese subjects and also in obese patients with T2D. As individuals with diabetes appear to be less successful in weight loss and weight loss maintenance than nondiabetic individuals, the meal replacement strategy is particularly challenging in individuals with diabetes. The weight loss results of studies with meal replacements have made feasible a large long-term study: Action for Health in Diabetes. This study with approximately 5,000 overweight volunteers with T2D sponsored by the NIDDK and the NIH (www.niddk.nih.gov/patient/SHOW/lookahead.htm) will evaluate the long-term health effects of an intensive lifestyle intervention designed to achieve and maintain weight loss. This program is compared to a control condition involving a program of diabetes support and education. To help participants achieve and maintain weight loss, diet strategies in the form of meal replacements (e.g. prepared meals and liquid formula), exercise strategies and optional weight loss medications are utilized. The primary outcome is the occurrence of cardiovascular events including myocardial infarctions and strokes and cardiovascular deaths. The study will be conducted over a 13-year period, and the results are not available before the year 2012. 177

Ditschuneit References 1 Tulloch-Reid MK, Williams DE, Looker HC, et al: Do measures of body fat distribution provide information on the risk of type 2 diabetes in addition to measures of general obesity? Diabetes Care 2003;26:2556–2561. 2 Colditz GA, Willett WC, Stampfer MJ, et al: Weight as a risk factor for clinical diabetes in women. Am J Epidemiol 1990;132:501–513. 3 Wang Y, Rimm EB, Stampfer MJ, et al: Comparison of abdominal adiposity and overall obesity in predicting risk of type 2 diabetes among men. Am J Clin Nutr 2005;81:555–563. 4 Eriksson K-F, Lindgarde F: Prevention of type 2 (non-insulin-dependent) diabetes mellitus by diet and physical exercise. The 6-Year Malmo Feasibility Study. Diabetologia 1991;34:891–898. 5 Pan X-R, Li G-W, Hu Y, et al: Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance. Diabetes Care 1997;20:537–544. 6 Tuomiletho J, Lindstrom J, Eriksson JL, et al: Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344:1343–1350. 7 Knowler WC, Barrett-Connor E, Fowler SE, et al: Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393–403. 8 Anderson JW, Kendall CWC, Jenkins DJA: Importance of weight management in type 2 diabetes: review with meta-analysis of clinical studies. J Am Coll Nutr 2003;5:331–339. 9 Pories WJ, Swanson MS, MacDonald KG, et al: Who would have thought it? An operation proves to be the most effective therapy for adult-onset diabetes mellitus. Ann Surg 1995;222: 339–350. 10 Sjöström L, Lindroos A-K, Peltonen M, et al: Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric surgery. N Engl J Med 2004;351:2683–2693. 11 Hadden DR, Blair ALT, Wilson EZ, et al: Natural history of diabetes presenting age 40–69 years: A prospective study of the influence of intensive dietary therapy. Q J Med 1986;59: 579–598. 12 American Diabetes Association: Nutrition principles and recommendations in diabetes (Position Statement). Diabetes Care 2004;27(suppl 1):S36–S46. 13 Keller U: Why does nutrition therapy so often fail in non-insulin-dependent diabetes? What measures bring success? Ther Umsch 1995;52:501–508. 14 Joslin Diabetes Center: Clinical Nutrition Guideline for Overweight and Obese Adults with Type 2 Diabetes, Prediabetes or at High Risk for Developing Type 2 Diabetes. Boston, Joslin Diabetes Center, Publications Department, 2005, Publ. No. 617-226-5815. 15 Pi-Sunyer FX: Weight loss in type 2 diabetic patients. Diabetes Care 2005;28:1526–1527. 16 UK Prospective Diabetes Study: Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 1998;352:837–853. 17 Henry RR, Wallace P, Olefsky JM: Effects of weight loss on mechanisms of hyperglycemia in obese non-insulin-dependent diabetes mellitus. Diabetes 1986;35:990–999. 18 Wing RR: Use of very-low-calorie diets in the treatment of obese persons with non-insulindependent diabetes mellitus. J Am Diet Assoc 1995;95:569–572. 19 Bowermann S, Bellman M, Saltsman P, et al: Implementation of a primary care physician network obesity management program. Obes Res 221;9(suppl 4):321S–325S. 20 Heymsfield SB, van Mierlo CAJ, van der Knaap HCM, et al: Weight management using a meal replacement strategy: meta and pooling analysis from six studies. Int J Obes 2003;27:537–549. 21 Flechtner-Mors M, Ditschuneit HH, Johnson TD, et al: Metabolic and weight loss effects of long-term dietary intervention in obese patients: four-year results. Obes Res 2000;8:399–402. 22 Winick C, Rothacker DC, Norman RL: Four worksite weight loss programs with high-stress occupations using a meal replacement product. Occup Med 2002;52:25–30. 23 Keogh JB, Clifton PM: The role of meal replacements in obesity treatment. Obes Rev 2005;6: 229–234. 24 Hensrud DD: Dietary treatment and long-term weight loss and maintenance in type 2 diabetes. Obes Res 2001;9(suppl 4):348S–353S. 25 Yip I, Go VL, DeShields S, et al: Liquid meal replacements and glycemic control in obese type 2 diabetes patients. Obes Res 2001;9(suppl 1):341S–347S. 26 Wing RR, Marcus MD, Epstein LH, Salata R: Type II diabetic subjects lose less weight than their overweight nondiabetic spouses. Diabetes Care 1987;10:563–566.

178

Do Meal Replacement Drinks Have a Role in Diabetes Management? 27 Li Z, Hong K, Saltsman P, et al: Long-term efficacy of soy-based meal replacements vs an individualized diet plan in obese type 2 diabetes patients: relative effects on weight loss, metabolic parameters, and C-reactive protein. Eur J Clin Nutr 2005;59:411–418. 28 Levy AS, Heaton AW: Weight control practices of U.S. adults trying to lose weight. Ann Intern Med 1993;119:661–666.

Discussion Dr. Hill: I really want to congratulate you for doing the long-term studies, because in the area of obesity treatment short-term studies tell us very little. Do you think of meal replacements as a drug so that their use will continue forever? If people don’t continue to use them, won’t they regain weight? So even though you have done a longterm study, the question is what happens over the even longer term? Will people have to return to real foods eventually and what will happen to their weight when they do? Dr. Ditschuneit: That is a very important issue in the treatment of obesity with meal replacements. Meal replacements are considered to be drugs, at least by some patients. The question when to return to conventional food is nearly always present. Many patients are doing well with the meal replacement plan and follow it to achieve the body weight that they strive for. After that, more and more they return to conventional meals. In the case of weight regain, they can again use meal replacements. Corresponding to the weight changes there will be a balance between meal replacements and real food. Dr. Katsilambros: I would like also to congratulate you for the long-term study. How much protein and how much carbohydrate are there in the meal replacements? Dr. Ditschuneit: In the 4-year study, each formula in the meal replacement plan contained 17 g protein and 34 g carbohydrates. Dr. Katsilambros: The reason I am asking that is that if you combine carbohydrate and protein you have a very high insulin secretion. If you give preventively more proteins then the insulin secretion is less and perhaps the hunger is less. Dr. Ditschuneit: In the weight loss phase with two meal replacements and one conventional meal per day and also in the weight maintenance phase with one meal replacement and two conventional meals, 45–50% of energy was derived from carbohydrates, 20–25% from protein and 30% from fat. Dr. Katsilambros: What kind of carbohydrates? Dr. Ditschuneit: In the study most of the carbohydrate in the meal replacements was sugar. The remaining conventional meals contained mainly complex carbohydrates. We preferred putting glucose into the meal replacement because we had the impression that the patients felt better. We think that somehow the sympathetic nervous system was activated through stimulation of insulin secretion after uptake of glucose and amino acids. Dr. Katsilambros: A stimulation but not high, if they are not combined with carbohydrates. Dr. Ditschuneit: Intensive and detailed interviews with the patients were performed, and the carbohydrate and protein intake was modified if needed. Dr. Katsilambros: How good was the compliance in the conventional group? What was the reason why they did not loose much weight? Dr. Ditschuneit: Compliance was very good in both groups of patients. Because all the patients had been transferred by a general physician to the university hospital, motivation to follow the study protocol was high. In addition the meal replacements were free for the patients. Compliance in the conventional group was similar to that in the meal replacement group. One reason was that the patients in the conventional

179

Ditschuneit group were also promised the meal replacements for free after 3 months. We assume that the weight loss in the patients of the conventional group was lower because they were more tempted by dietary mistakes and due to the lower adherence to the dietary prescriptions. For the patients it has been shown that it is easier to take a portion of controlled meal replacement than to prepare a meal from self-selected conventional foods. Dr. Katsilambros: But how was the compliance in the conventional group? Dr. Ditschuneit: Compliance was the same as in the meal replacement group. For the reasons mentioned earlier, compliance within the first 3 months was complete with regard to visits to the hospital. Dr. Katsilambros: It is very impressive then why they did not lose weight. I remember long ago at the University of Ulm you used the zero diet. What happened to these people? Did you have a follow-up? I am very curious to know. Dr. Ditschuneit: Between 1970 and 1980 we had experience with starvation in obese patients. This situation was always transient and described as the zero diet. The zero diet was an opportunity for morbidly obese patients to lose considerable amounts of weight within a short time. We have no long-term follow-up of these patients. Dr. Foreyt: I agree with Dr. Hill, these data just didn’t exist before your studies on obesity, so congratulations. I have some questions regarding the maintenance phase. How much lifestyle intervention was necessary? How much counseling did they get when they came in every month or every 2 months? Did you just pass out the meal replacements or did you actually do group therapy or individual treatment for an hour, what happened? Did you measure how much meal replacement they actually took? Was it the counseling or was it the meal replacements or both? Dr. Ditschuneit: In the maintenance phase patients came in every month, and later at least every 2 months. Group sessions were done regularly. In addition individual consultations and training were done every month by a nutritionist. The time needed for a visit was variable and depended on individual problems. At each visit the number of meal replacements that the patient had used were counted and registered. The long-term outcome of the study was the result of regular and individual counseling as well as the simplicity of the meal replacement program. Dr. T. Wilkin: Did you suggest that the analysis was on the basis of cases available or intention to treat? Dr. Ditschuneit: The data were analyzed on the basis of all available cases, intentionto-treat. and of the last values traced back to baseline. Dr. T. Wilkin: What percentage of those on the meal replacements did you lose at 4 years? Dr. Ditschuneit: At 4 years, of the 100 patients we lost a total of 25: 14 in the meal replacement group and 11 in the conventional group. Dr. Gerasimidi-Vazeou: In your study what was the volume of a meal that you gave to replace first the breakfast and second the lunch? Dr. Ditschuneit: The volume of the meal replacement was 250 ml for breakfast as well as lunch. Dr. Golay: My concern is for clinicians. Most of the time the patients cannot take liquid diets forever and they eat them on top of the regular diet so they gain weight. In terms of patient education, it is also not a really good idea to propose liquid diets forever. It is better to teach them to eat a regular diet. It is a big problem for me to propose a liquid diet to our patients. Dr. Ditschuneit: I agree that patients cannot take liquid diets forever and we do not propose this. With a partial meal replacement plan, patients have a structured meal plan and can be taught a healthy diet.

180

Do Meal Replacement Drinks Have a Role in Diabetes Management? Dr. Golay: I am very pleased to hear that. For a mother with children it is difficult to eat a liquid diet all the time. My best patients for a liquid diet are surgeons; they like this kind of meal because it is fast. Dr. Ditschuneit: We do not recommend partial meal replacement for children. Dr. Bantle: I have two comments. First for Dr. Golay, I have no concern about using these products. We use them both in research and in clinical practice and they actually provide better nutrition than meals for many people because they are fortified with vitamins and minerals. You can demonstrate that calcium intake is increased by using meal replacements. The second comment pertains to weight loss in people with type-2 diabetes. I agree that it is more difficult to accomplish than it is in non-diabetic populations. But I don’t think that is because they comply with treatment less well. I think the more likely explanation is that the treatment reduces urine glucose so, as they lose weight and plasma glucose declines, the calories previously lost in the urine are retained. Said in a different way, they are actually too thin for their caloric intake because of glucosuria.

181

The Role of Drugs and Diet Therapy – Alone and Together Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 183–196, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

Physical Activity in Prevention and Management of Obesity and Type-2 Diabetes James O. Hill, Jennifer Stuht, Holly R. Wyatt, Judith G. Regensteiner Center for Human Nutrition, University of Colorado School of Medicine, Denver, CO, USA

Abstract Obesity and type-2 diabetes can be considered diseases of physical inactivity. Physically activity protects against type-2 diabetes through its positive effects on weight management and on the metabolic pathways involved in glycemic control that are not weight-dependent. Increasing physical activity is one of the most effective strategies both for preventing type-2 diabetes and for managing it once it is present. However, we still face an enormous challenge in getting people to achieve sustainable increases in physical activity. A promising strategy is to get people walking more, starting small and increasing gradually over time. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

Introduction Physical inactivity is a critical factor in the etiology and treatment of type-2 diabetes. There is a clear inverse relationship between physical activity and insulin resistance [1], and the risk of developing type-2 diabetes is decreased with physical activity [2–5]. The relationship between physical activity and the risk of type-2 diabetes holds across different methods of assessing physical activity from self-reports to measured cardiorespiratory fitness [2–5]. For example, figure 1 shows how the incidence of type-2 diabetes per 1,000 personyears varies with cardiorespiratory fitness in men [4]. Increased cardiorespiratory fitness (which can be increased by physical activity) reduced the risk 183

Rate per 1,000 person-years

Hill/Stuht/Wyatt/Regensteiner

8 7 6 5 4 3 2 1 0

Rate per 1,000 person-years

P⬍0.001

Ageⱖ45 years

a

c

P ⱕ0.01

9 8 7 6 5 4 3 2 1 0

Age⬍45 years

9 8 7 6 5 4 3 2 1 0

b

P⫽0.006

Low fitness Moderate fitness High fitness

P ⬍0.001

BMI ⱖ 27 kg/m2

BMI ⬍ 27 kg/m2

25 20 P⫽0.004

15 P ⬍0.001

P⫽ 0.001

10 P⬍0.002 5 0

Parental diabetes

No parental diabetes

d

Impaired fasting glucose

Normal fasting glucose

Fig. 1. Incidence of type-2 diabetes per 1,000 person-years by cardiorespiratory a), body mass index (BMI; b), history of fitness levels according to age group (a d). parental diabetes (cc), and impaired fasting glucose (d

of developing type-2 diabetes in both overweight and non-overweight individuals, in young and older individuals, in those with and without impaired fasting glucose, and in those with and without parents with diabetes. A similar negative relationship was found between self-reported physical activity and the risk of type-2 diabetes in a group of 87,253 middle-aged women [2]. Physical activity impacts type-2 diabetes in at least two ways. First, increased physical activity is associated with less overweight and obesity (both through prevention and treatment of obesity). Since overweight and obesity increase the risk of type-2 diabetes [6], less overweight and obesity is associated with less type-2 diabetes. Second, physical activity helps in the prevention and treatment of type-2 diabetes through weight-independent effects on metabolic pathways involved in diabetic control [1, 7, 8].

Obesity and Type-2 Diabetes A strong association between indices of overweight and obesity and the development of type-2 diabetes has been reported in a great many studies. As 184

Physical Activity and Type-2 Diabetes 100 Age-adjusted relative risk

90 80 70 60 50 40 30 20 10 35 ⭓

9 –3

4.

9 2. –3

33

9 0. –3

31

9 29

–2

8.

9 27

6.

9 4.

–2 25

9 3.

–2 24

9 2.

–2 23

–2 22

⬎

22

0

BMI (kg/m2)

Fig. 2. Body mass index (BMI) and the age-adjusted relative risk (95% confidence interval) for non-insulin-dependent diabetes in US women.

one example, Colditz et al. [9] followed women, 30–55 years of age, over 14 years. They demonstrated that the risk of developing type-2 diabetes increased rapidly with increases in body mass index (BMI; fig. 2). Individuals with BMIs in the healthy range (18.5–24.9 kg/m2) were at very low risk of developing type-2 diabetes. Also, weight gain in adults, regardless of BMI, appears to be a risk factor for development of type-2 diabetes [9, 10]. Since overweight and obesity increase the risk of development of type-2 diabetes, the next question is whether weight loss in these groups reduces the risk of development of type-2 diabetes. Strong evidence in support of weight loss as a means of preventing type-2 diabetes in individuals who are overweight or obese comes from the Diabetes Prevention Program (DPP) [11] and from the Finnish study on the prevention of type-2 diabetes [12]. In these studies, overweight and obese individuals, who were at high risk of developing type-2 diabetes (i.e. impaired glucose tolerance) and who were assigned to a lifestyle-intervention program that involved modest weight loss, reduced their risk of developing type-2 diabetes by about 60%. In the DPP, the lifestyle group achieved a 7% reduction in body weight and maintained a 5% weight reduction over 4 years. The lifestyle intervention in the DPP included modification of diet and increases of 150 min/week of physical activity. In summary, the available data suggest that increases in body weight that lead to overweight and obesity increase the risk of type-2 diabetes. Modest weight loss in overweight and obese individuals substantially reduces the risk of type-2 diabetes. 185

Hill/Stuht/Wyatt/Regensteiner Physical Activity and Obesity Physical Activity in Prevention of Obesity Results of epidemiological studies are consistent in that those who are physically active are less likely to gain weight over time and become obese than those who are not. Several studies [13–16] have found that individuals who remain physically active or increase physical activity over time are less likely to gain weight than those who remain sedentary or who reduce physical activity over time. There are very few prospective, randomized trials examining the ability of physical activity to prevent weight gain. Such trials are solely needed in order to establish a causal relationship between physical activity and prevention of weight gain and in order to identify how much physical activity is required to prevent weight gain. This amount may vary between populations and between life stages of individuals.

Physical Activity in Obesity Treatment While it is possible to lose weight with physical activity alone, the amount of physical activity required for substantial weight loss is well beyond what is feasible for most Americans. Wing [17] reviewed several studies in which physical activity alone was used for weight loss. While the amount of weight loss with physical activity was significantly greater than 0, it was in the order of only a few pounds. Similarly, Wing [17] reviewed several studies in which weight loss with diet alone was compared with weight loss with diet plus exercise. It was concluded that most studies found no significant differences in total weight loss but that in just about every study the absolute amount of weight lost was a little higher when diet and exercise were used. This is not surprising given that weight loss is a function of the degree of energy imbalance and that a much greater energy imbalance can be created with food restriction than with increased physical activity. Even though adding physical activity to food restriction adds little when it comes to initial weight loss, there may be other advantages of engaging in physical activity during weight loss that increase the chances that the weight loss will be maintained. Physical activity could result in a higher proportion of weight loss coming from fat and less from fat-free mass loss [18]. This could reduce the drop in metabolic rate that accompanies weight loss. Another advantage of engaging in physical activity during weight loss is that it may better prepare the person to be able to engage in sufficient amounts of physical activity to keep weight off.

186

Physical Activity and Type-2 Diabetes Role of Physical Activity in Maintenance of Weight Loss In many studies, high levels of physical activity have been found to predict success in long-term weight loss maintenance [19–22]. For example, subjects in the National Weight Control Registry who are maintaining an average weight loss of about 30 kg for about 5.5 years, report expending about 2,800 kcal/week (⬃60–90 min/day) in physical activity [19]. Less than 10% of the successful weight loss maintainers report that they are maintaining their weight loss with diet alone. Decreases in physical activity in this group predict weight regain over time. Others [20–22] have found that amounts of physical activity equivalent to about 60–90 min/day are associated with successful weight loss maintenance. Role of Physical Activity in Management of Obesity and Type-2 Diabetes In addition to lowering the risk of type-2 diabetes, physical activity and high cardiorespiratory fitness can lower the risk of macrovascular disease, hypertension, and some cancers [23]. Increased physical activity is also associated with reduced all-cause mortality, even in individuals who are already overweight or obese and in those who have type-2 diabetes [23–25]. Non-Weight-Dependent Effects of Physical Activity on Prevention and Management of Type-2 Diabetes Physical activity has impacts on metabolism that seem to improve management of type-2 diabetes independent of weight loss [1, 7, 8]. Increased physical activity reduces insulin resistance, improves insulin sensitivity, and increases glucose disposal rates even independent of changes in body weight or body fatness [1, 7, 8]. In addition to insulin resistance and glycemic control, other factors which likely play a role in causing exercise impairments and which might be amenable to treatment include endothelial dysfunction, cardiac abnormalities and mitochondrial abnormalities [7, 8]. How Much Physical Activity Can Prevent Obesity and Type-2 Diabetes Increasing physical activity is widely recommended as a strategy for prevention and management of obesity and type-2 diabetes. Increasing physical activity levels in the population could have a major effect on reducing the incidence of both diseases. However, there are no definitive data to allow specific recommendations about the amount of physical activity that may be

187

Hill/Stuht/Wyatt/Regensteiner effective in the prevention and management of these diseases. The exception is that maintaining a significant weight loss seems to require about 60–90 min/ day of physical activity [19–22]. It has been suggested that far less physical activity is required to prevent weight gain and that increases in walking as small as 2,000 extra steps each day (about 15–20 min or about 1.6 km) can prevent the gradual weight gain seen in most of the population. The American Diabetes Association recommends 30 min/day of physical activity for individuals with type-2 diabetes, which is based on recommendations from other credible groups [26] and on improvements in overall health. It is likely that the minimum effective dose of physical activity to prevent or manage obesity and type-2 diabetes will vary somewhat from person to person. However, when it comes to the prevention/treatment of obesity and type-2 diabetes, any increase in physical activity seems beneficial. The epidemiological work by Wei et al. [4] and Blair and Brodney [23] has consistently suggested that the greatest benefit of physical activity on the prevention of chronic diseases and mortality comes from getting the most sedentary individuals to make small increases in physical activity. Strength training on a regular basis may also have a beneficial effect on glucose tolerance [8]. However, such training may not improve maximal oxygen consumption. Thus, strength training should probably be considered an adjunct rather than the primary form of training for persons with type-2 diabetes, since aerobics produces both types of benefit.

How to Get People Active While there is great agreement on the benefits of physical activity, there is still a big challenge in getting people to increase physical activity. The environment we have constructed is not one that encourages physical activity, and in fact many technological advances of the last decades have likely caused substantial declines in physical activity [27]. Increases in physical activity can be promoted in different ways. In general this can be though promoting planned physical activity or exercise, and through promoting increases in lifestyle physical activity. The latter involves showing people how to increase physical activity (i.e. walking) throughout their usual day without the need to set aside planned exercise time. For individuals maintaining a significant weight loss, both strategies will likely be required in order to achieve sufficient physical activity. For prevention of the 0.45–0.90 kg of weight gained each year by the average American, either strategy alone could be sufficient. Walking may be one of the best ways to get people to increase physical activity. A study by Hu et al. [28] suggests that vigorous exercise is not required to reduce the risk of type-2 diabetes. They found that walking was negatively associated with the risk of developing diabetes. In a prospective study, 188

Physical Activity and Type-2 Diabetes DiLoreto et al. [29] evaluated the impact of walking on patients with type-2 diabetes. They proscribed walking for their patients and, in a post hoc analysis, found significant health and financial benefits to increased walking. These included improvements in blood pressure, lipids, and glucose. Further, the yearly cost of medications decreased with increased walking. While any amount of walking was beneficial, the greatest benefits occurred with increases in physical activity of ⬎10 metabolic equivalents (METs)/h/week. This is consistent with the idea of helping people make small rather than large behavior changes. Step counters (pedometers) have been used effectively to produce gradual increases in physical activity. While physical activity levels of ⬎30 min/day may be desirably, any increases in physical activity can benefit health and help in the maintenance of a healthy body weight. Special Considerations for Physical Activity and Type-2 Diabetes Exercise testing is recommended for many persons with type-2 diabetes before beginning an exercise conditioning program because of the high prevalence of occult cardiovascular disease, symptoms of which may only be manifested during exercise [30, 31]. Specifically, many patients with diabetes may have ischemia or infarction without angina. Most experts recommend that persons with type-2 diabetes be tested if the individual has previously been inactive, has had diabetes for more than 10 years, or is over the age of 35 [30, 31]. However, because of the potential of false-positive exercise tests, further more invasive testing may be required to substantiate a positive stress test. Exercise testing can also be used to provide an exercise prescription. Although measurement of oxygen consumption is the most reliable way to assess exercise capacity and create an exercise prescription, it is often not practical to measure it in the clinical setting. Therefore, maximal heart rate measured during the treadmill test can be used as a crude but reasonable substitute. One exception is that patients with autonomic neuropathy as a complication of diabetes may not be able to achieve an age-predicted maximal heart rate which may reduce the sensitivity of the test. In these patients, if an accurate exercise prescription is of importance, maximal oxygen consumption should be measured. The presence of type-2 diabetes alone does not require a supervised exercise conditioning program. However, there are several specific groups of patients with type-2 diabetes for whom a more formal exercise conditioning program is desirable including patients with some types of heart disease. Safety Considerations As with any population, increasing physical activity has some potential risk for the patient with type-2 diabetes. Individuals who have underlying 189

Hill/Stuht/Wyatt/Regensteiner coronary heart disease, as in non-diabetic patients with heart disease, exercise may theoretically precipitate angina, myocardial infarction, arrhythmias or even sudden death. As with other patients with coronary heart disease, physical activity is contraindicated in the presence of unstable angina. High intensity aerobic exercise and isometric exercise are contraindicated in the patient with proliferative retinopathy because of an increased risk of developing retinal or vitreous hemorrhages and retinal detachment. However, moderate intensity aerobic exercise, such as walking, is an acceptable modality of treatment. Patients with peripheral neuropathy should not engage in exercise which may traumatize the insensitive foot (such as jogging). In addition, properly fitted footwear and checking of the feet for injury after exercise are recommended precautions. Data evaluating the potential problem of exercise-induced hypoglycemia in type-2 diabetes patients taking oral agents or insulin are lacking. However, non-diabetic individuals taking oral hypoglycemic drugs developed hypoglycemia during prolonged exercise. A reasonable precaution is to monitor blood glucose frequently upon initiating a physical activity program. The positive effects of physical activity in persons with type-2 diabetes may require adjusting medications and physicians should be alert for this need when their patients begin increasing physical activity.

References 1 Gautier JF, Scheen A, Lefebevre PJ: Exercise in the management of non-insulin-dependent (type 2) diabetes mellitus. Int J Obes 1995;19(suppl 4):S58–S61. 2 Manson JE, Rimm EB, Stampfer MJ, et al: Physical activity and incidence of non-insulin dependent diabetes mellitus in women. Lancet 1991;338:774–778. 3 Helmrich SP, Ragland DR, Leung RW, et al: Physical activity and reduced occurrence of noninsulin dependent diabetes mellitus. N Engl J Med 1991;325:147–152. 4 Wei M, Gibbons LW, Mitchell TL, et al: The association between cardiorespiratory fitness and impaired fasting glucose and type 2 diabetes mellitus in men. Ann Intern Med 1999;130: 89–96. 5 Lynch J, Helmrich SP, Lakka TA, et al: Moderately intense physical activities and high levels of cardiorespiratory fitness reduce the risk of non-insulin dependent diabetes mellitus in middle-aged men. Arch Intern Med 1996;156:1307–1314. 6 Bray GA: Contemporary Diagnosis and Management of Obesity. Newtown, Handbooks in Health Care, 1998. 7 Colberg SR, Swain DP: Exercise and diabetes control: a winning combination. Phys Sportsmed 2000;28:online version. 8 White RD, Sherman C: Exercise in diabetes management. Phys Sportsmed 1999;4:online version. 9 Colditz GA, Willett WC, Rotnitzky A, et al: Weight gain as a risk factor for clinical diabetes in women. Ann Intern Med 1995;122:481–486. 10 Chan JM, Rimm EB, Colditz GA, et al: Obesity, fat distribution, and weight gain as risk factors for clinical diabetes in men. Diab Care 1994;17:961–969. 11 Diabetes Prevention Program Research Group: Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346;393–402. 12 Tuomiletho J, Lindstrom J, Eriksson JG, et al: Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344: 1343–1350. 13 Haapanen N, Miilunpalo S, Pasanen M, et al: Association between leisure time physical activity and 10-yr body mass change among working-aged men and women. Int J Obes Relat Metab Disord 1997;21:288–296.

190

Physical Activity and Type-2 Diabetes 14 Dipietro L, Dziura J, Blair SN: Estimated change in physical activity levels (PAL) and prediction of 5-year weight change in middle-aged men: the aerobics center longitudinal study. Med Sci Sports Exerc 2005, in press. 15 Schmitz KH, Jacobs DRJ, Leon AS, et al: Physical activity and body weight: associations over ten years in the CARDIA Study. Coronary Artery Risk Development In Young Adults. Int J Obes Relat Metab Disord 2000;24:1475–1487. 16 Williamson DF, Madans J, Anda RF, et al: Recreational physical activity and ten-year weight change in a US national cohort. Int J Obes Relat Metab Disord 1993;17:279–286. 17 Wing RR: Physical activity in the treatment of the adulthood overweight and obesity: current evidence and research issues. Med Sci Sports Exerc 1999;31(suppl):S547–S552. 18 Ross R, Janssen I: is abdominal fat preferentially reduced in response to exercise induced weight loss? Med Sci Sports Exerc 1999;31(suppl):S568–S572. 19 Klem ML, Wing RR, McGuire MT, et al: A descriptive study of individuals successful at long term maintenance of substantial weight loss. Am J Clin Nutr 1997;66:239–246. 20 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. 21 Weinsier WL, Hunter GR, Desmond RA, et al: Free-living activity energy expenditure in women successful and unsuccessful at maintaining a normal body weight. Am J Clin Nutr 2002;75:499–504. 22 Jakicic JM, Winters C, Lang W, Wing RR: Effects of intermittent exercise and use of home exercise equipment on adherence, weight loss and fitness in overweight women. JAMA 1999;282: 1554–1560. 23 Blair SN, Brodney S: Effects of physical inactivity and obesity on morbidity and mortality: current evidence and research issues. Med Sci Sport Exerc 1999;31(suppl 1):S646–S662. 24 Wei M, Gibbons LS, Kampert JB, et al: Low cardiorespiratory fitness and physical inactivity as predictors of mortality in men with type 2 diabetes. Ann Intern Med 2000;132:605–611. 25 Church TS, Cheng YJ, Earnest CP: Exercise capacity and body composition as predictors of mortality among men with diabetes. Diabetes Care 2004;27:83–88. 26 US Department of Health and Human Services: Physical Activity and Health: Report of the Surgeon General. Atlanta, US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, 1996. 27 Hill JO, Wyatt HR, Reed GW, Peters JC: Obesity and the environment: where do we go from here? Science 2003;299:853–855. 28 Hu FB, Sigal RJ, Rich-Edwards JW, et al: Walking compared with vigorous physical activity and risk of type 2 diabetes in women. JAMA 1999;282:1433–1439. 29 DiLoreto CD, Fanelli C, Lucidi P: Make your diabetic patients walk. Diabetes Care 2005;28: 1295–1302. 30 Sigal RJ, Kenny GP, Wasserman DH, Castaneda-Sceppa C: Physical activity/exercise and type 2 diabetes. Diabetes Care 2004;27:2518–2539. 31 Consensus Development Conference on Diet and Exercise in Non-Insulin-Dependent Diabetes Mellitus. National Institutes of Health. Diabetes Care 1987;10:639–644.

Discussion Dr. Mooradian: The issue of being fit and its correlation with mortality or outcome measures: the data you showed, specifically the data in type-2 diabetes and any other data in the literature, are mostly correlations and they are very hard to pin down to fitness. A lot of the time people who tend to be fit are also healthier individuals, so overall other co-morbidities will be less so. It is very hard to sort out fitness as an independent predictor of outcome. Another issue is that, as far as I know, there are absolutely no data to show that, as you recommend, increasing food intake and exercise actually improves longevity or health, other than the non-tangible benefits of exercise that we are all aware of, and there are many benefits of exercise. In contrast, a lot of data show that if you restrict food you improve life expectancy invariably in many species. So if

191

Hill/Stuht/Wyatt/Regensteiner you want to make a recommendation based on outcome, it seems to me that it is better to restrict food rather than liberalize it, and then do a lot of exercise. Dr. Hill: Let me deal with both those issues because these are really good questions. I am glad you asked about the fit and fat issue. Even though people call it a controversy, it isn’t. I believe the data suggest that whatever your fatness level it is better to be fitter, and whatever your fitness level it is better to be leaner. We can produce significant weight loss in patients with a BMI of ⬎40, but most will still be overweight or obese. Our treatment for these individuals should be aimed both at producing weight loss and increasing physical fitness. As we are more successful at doing this, we will create more fit and fat people. The research on caloric restriction and longevity is very interesting. There were some data presented at the obesity meetings in Vancouver suggesting that increased exercise may also positively impact longevity. The question is whether it is calorie restriction or avoiding positive energy balance that increases longevity. Dr. Metzger: Do you see any allowance for an inherent influence on the level of physical activity? The reason I bring this up is that there have been some reports that already from intrauterine time onward babies predisposed to have more body fat and individuals predisposed to be heavy already have a reduced level of physical activity. Dr. Hill: This is an interesting concept to consider. We are learning that the intrauterine environment can have a major impact on metabolism and weight later in life. I don’t think we have any data about this, but it is certainly possible that the intrauterine environment could affect later physical activity. Dr. Bantle: We commonly think of exercise as a means of expending energy and that is the mechanism whereby it influences weight. But I wonder about another possibility. Distance runners are uniformly lean and I ask myself why. Is it because they can’t eat enough to keep up with their exercise? In fact, what they eat is absolutely astounding in terms of amounts and calories. So my question is, could there be another mechanism involved here? Is it possible that the hypothalamus responds to the level of habitual physical activity and adjusts body configuration to conform to the need? In effect, the hypothalamus might say, ‘We have to run 50 km/week and we need to stay lean to do that’. Dr. Hill: We know the brain is very important in body weight regulation but it is not totally clear what is being regulated. The brain could be regulating an amount of energy intake, an amount of physical activity (or energy expenditure) or an amount of body weight or body fat. Distance runners are a unique group and it is possible that they are regulating around a high level of physical activity. Alternatively they may be regulating a high level of energy intake and the high levels of physical activity are necessary to keep them lean. I think that people can modify their physical activity level to some extent and that the level of physical activity affects the sensitivity of many metabolic processes in the body. I see the sedentary state as the abnormal state so rather than have a note from your doctor to exercise, you should have a note from your doctor to be sedentary. There is some evidence from the Pima Indians that leptin regulation may be different in the Mexican Pimas who are much more physically active than the Pimas in Arizona. So I think we may find that the level of physical activity plays a role in the precision of regulation of some metabolic processes. The brain and specifically the hypothalamus are almost certainly involved. Dr. Chiasson: I was wondering about the obese population that you followed in the registry. You said that in those who maintain moderate exercise, their body weight is gradually reduced through exercise. Is that a failure to moderate exercise or a failure to follow diet over time? Dr. Hill: The short answer is we don’t know. What we know is that the people who are successful at weight loss maintenance report high levels of physical activity and

192

Physical Activity and Type-2 Diabetes people whose exercise decreases over time gain weight. There are several studies that suggest that people who are only moderately active are unable to maintain their weight loss. It seems to take a lot of exercise to maintain large weight losses. I think the amount of exercise you have to do is related to the amount of weight you lose. The more weight you are keeping off, the more you have to exercise. Dr. Chow: It is possible to induce someone to engage in some exercise; perhaps you can effect a change of mind so they will be more self-disciplined which means that they probably will also be more compliant in diet and follow the guidelines from the medical profession. Do you agree, and do you have any evidence for this? Dr. Hill: That is a good question and that is one of the alternative hypotheses for why exercise is always such a great marker of success in weight loss maintenance. It could simply be a marker of compliance, so if a person is exercising regularly they are also more likely to be eating healthily. One reason I think exercise itself is important is that when we estimate how much metabolic rate should have changed with weight loss for people in the National Weight Loss Registry, the amount of exercise they are doing is very close to this value. Thus, I think the advantage of high levels of exercise is that it allows these people to maintain a lower weight with having to engage in constant food restriction. Dr. Slama: For the sake of the discussion I would like to recall the words of Winston Churchill. When he was asked the recipe for his long life he said: no sport, no sport at all, Scotch and cigar every day. I would like to challenge the idea put forward by Elliott Joslin who said that you cannot treat diabetes without exercise, insulin and diet, and more precisely that you cannot control diabetes without exercise. We did a study in our department showing that there is a population of diabetic people who is very well controlled with a very low level of exercise. We also have type-1 and type-2 diabetic people who are very badly controlled with a very high level of exercise. In any direction you can find something, but you may also observe a large population of patients with a very low level of exercise and perfect blood glucose control. But of course exercise has a much more important effect on life: cardiovascular disease prevention. Dr. Hill: You make a very good point. In the National Weight Control Registry for example there are about 9% of people who are maintaining weight loss without exercise. So I do think there is going to be a population for whom exercise may not be effective and diet is critically important. I think there is probably a population for whom exercise is tremendously important and exercise alone may be sufficient to maintain a healthy weight. Most people are going to be in the middle where exercise is going to help but diet changes are also needed. The more I look at this issue, the more I believe that exercise is more effective the earlier in the cascade of chronic disease you are. I think it is the most effective in preventing obesity and diabetes in lean people. I think the more you go down the path toward diabetes and cardiovascular disease, the more exercise you have to do to positively impact health. Ms. June Chan: You mentioned that in the National Weight Control Registry most people use walking and some use resistance training. Do you have any data showing that resistance training plus aerobic exercises is going to be better than aerobics alone in terms of weight maintenance or obesity prevention? Dr. Hill: We looked at the proportion of people in the registry who reported engaging in resistance training or weight lifting versus the proportion in a general survey of the American population. A higher proportion of women in the National Weight Control Registry (15–20%) reported engaging in resistance training as compared to women in the general public (5–6%). We do not know if this is a factor in their success or not. There are some data in the literature showing that resistance training has similar effects to aerobic activity in obesity treatment. I don’t think resistance training can

193

Hill/Stuht/Wyatt/Regensteiner totally take the place of aerobic activity, but it can be a positive companion to aerobic activity. Dr. T. Wilkin: I was interested in Dr. Bantle’s comments on the hypothalamus and the possibility that it censures the physical activity that is accomplished. We study physical activity in children, and about 18 months ago reported in the British Medical Journal [1] a study in 3 different schools with very different opportunities for physical activity. In one of the schools it was 9 h and in another school under 2 h, so it was a very big range. As you might expect the activity of the children in the school that gave the most opportunity did much more during the day time. We used accelerometers to measure this, so these would be reasonably objective measurements we were making. However, when these children got home in the evening they just flopped. The children who got under 2 h perked up in the evening. If you added the out-of-school to the inschool activities, you got the same over the whole of the range. This lead us to do a number of other studies, all of which put together suggest very strongly that there is an activity stat, at least in children, which regulates the amount of activity they do. I suspect that as children become teenagers and beyond there may be social cues that may override this, but it is probably most strong in children. Dr. Hill: I think it is a possibility. I know there are some data in elderly people showing that when they were given a supervised exercise program that they were more sedentary in the rest of the day. But I am very skeptical that the amount of physical activity is fixed. Again, people in the National Weight Loss Registry were able to maintain large, permanent increases in physical activity. Dr. Barclay: You mentioned the importance of breakfast, and we often hear that it is better to get your calories more in the morning than in the evening. What is the evidence for that in terms of weight maintenance and weight loss? Dr. Hill: There is a fair amount of evidence that eating breakfast has a positive effect on body weight. First of all there are several epidemiology studies showing that people who eat breakfast are leaner than those who do not. There are studies reporting that people who eat breakfast end up eating fewer calories during the day than those who do not. Finally there are studies showing that satiety is highest for food eaten early during the day compared to food eaten later in the day. So we have a lot of circumstantial evidence suggesting the importance of breakfast for weight management. Dr. Golay: You propose 60 min of exercise for obese patients but in fact they should have even less than that because they are obese. The energy expenditure in obese patients is much higher, so I would propose even less than 15 min for obese compared to lean patients. Dr. Hill: I don’t think 15 min is enough. You are correct that obese people have a higher cost of exercise, but I still think it takes about an hour a day of exercise to maintain a large weight loss. I think this may be because subjects are making up for some sort of metabolic price of being obese. Dr. Metzger: I want to come back to the question about the calorie distribution through the day. There are animal models from 45–50 years ago showing that calories remained constant. If a large proportion of calories is administered to experimental animals in the latter half of the day, equivalent to our dinner time, body composition is significantly affected resulting in more hepatic lipid synthesis, increased body fat at the same isocaloric intake. I think there have been some human studies that would be consistent with that, but obviously they can’t be as interventional as the animal studies. Dr. Hill: I think we see that pattern. I know you all see it in your obese subjects who skip breakfast, eat a light lunch and start eating at 3 or 4 in the afternoon and eat all through the night. This may be the worst meal pattern. I think that by eating breakfast you break up that pattern and this may be beneficial. Dr. Ho: I have a question regarding the relationship between the exercise and the timing of meals.

194

Physical Activity and Type-2 Diabetes Dr. Hill: I believe that the most important decision is whether to exercise or not regardless of when you exercise. The literature is mixed regarding the interaction of eating and exercise. My reading of the literature is that if there is an interaction, it is small. The far bigger effect is whether you do any exercise or not. Dr. Halimi: Regarding type-2 diabetic patients, one of the major objectives is the control of glycemia, not only for reducing the high cardiovascular risk but mainly for preventing microangiopathy. In your opinion what is the best duration for reducing glycemia if we consider that the body doesn’t burn the same fuel according to the duration of a physical exercise and what training could change? Dr. Hill: So you asking how the interaction of exercise and meals affects glycemia during the day? I think that is an interesting question and that is where perhaps exercise and meals might play a role. If it is glycemia that is the issue, then the timing of exercise in type-2 diabetes could be important. Dr. Halimi: Not only because of the time of the meal. After 30 or 45 min the fuel utilized changes, and there is some evidence in favor of a longer duration of physical exercise in type-2 diabetics when compared to obese patients. Dr. Hill: There is a tradeoff between the duration and intensity of exercise. The effects depend on intensity and duration. The more intense you exercise the more carbohydrate versus fat is oxidized. Moderate intensity longer duration exercise burns proportionally more fat but less calories. For weight management in non-diabetics, the total amount of energy expended in physical activity is probably more important than the timing and intensity. For diabetics, the timing and intensity may be more important and could affect glycemia during the day. Ms. Franz: A concern I have with the Weight Loss Registry is that it may lead individuals to have unrealistic weight loss goals. While there are participants in weight loss programs who will lose larger amounts of weight and those who will lose none, the majority will likely lose 4.5–7.5 kg at 12 months and, with continued support, maintain a weight loss of 3–4 kg [2, 3]. They will not be as successful at weight loss as the individuals in the Weight Loss Registry but will still experience health benefits as evidenced by participants in the Diabetes Prevention Program. Women participating in a weight loss program expected to lose 34% of their body weight, and despite a weight loss of 16%, they reported being unsatisfied with their weight loss [4]. Baseline expectations are also reported to be an independent predictor of attrition in obese patients entering a weight loss program; the higher the expectations, the higher the attrition at 12 months [5]. Although we can certainly learn from the participants in the Weight Loss Registry, it is important that persons attempting weight loss have realistic weight loss goals. Dr. Hill: I agree totally. The whole point of developing the registry was not to look at the prevalence of successful weight loss maintenance but simply to look for similar behavior in those who are most successful. People in the registry have achieved a level of success that most people do not achieve. However, I think the registry helps us identify the kinds of behaviors that could help more people be successful.

References 1 Mallam KM, Metcalf BS, Kirkby J, et al: Contribution of timetabled physical education to total physical activity in primary school children: cross sectional study. BMJ 2003;327: 592–593. 2 Curioni CC, Lourenco PM: Long-term weight loss after diet and exercise: a systematic review. Int J Obes (Lond) 2005;29:1168–1174. 3 Douketis JD, Macie C, Thabane L, Williamson DF: Systematic review of long-term weight loss studies in obese adults: clinical significance and applicability to clinical practice. Int J Obes (Lond) 2005;29:1153–1167.

195

Hill/Stuht/Wyatt/Regensteiner 4 Foster GD, Wadden TA, Vogt RA, Brewer G: What is a reasonable weight loss? Patients’ expectations and evaluations of obesity treatment outcomes. J Consult Clin Psychol 1997;65: 79–85. 5 Dalle Grave R, Calugi S, Molinari E, et al, QUOVADIS Study Group: Weight loss expectations in obese patients and treatment attrition: an observational multicenter study. Obes Res 2005;13:1961–1969.

196

Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 197–206, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

The Role of Lifestyle Modification in Dysmetabolic Syndrome Management John P. Foreyt Department of Medicine, Baylor College of Medicine, Houston, TX, USA

Abstract Lifestyle modification should be the primary therapeutic intervention in individuals with the dysmetabolic syndrome, given the fact that obesity, unhealthy diet, and physical inactivity are primary underlying risk factors for its development. Most individuals with the dysmetabolic syndrome need to lose weight through dietary changes and increases in physical activity. Modest weight losses may significantly improve all aspects of the syndrome. Because individuals differ in their lifestyles, tailoring interventions to meet the specific needs of each person will maximize the chances of success. Assessment of the individual with the dysmetabolic syndrome involves quantification of obesity, diets and dietary patterns, physical activity, emotional problems, and motivation. To help individuals make lifestyle changes, a number of behavior modification strategies have shown good efficacy. These strategies include a tailored problem-solving intervention, involving goal-setting, self-monitoring, stimulus control, cognitive restructuring, stress management, relapse prevention, social support, and contracting. The frequency of self-monitoring is an especially important strategy for continued success. Research studies have clearly demonstrated the power of lifestyle modification for long-term behavioral change. Lifestyle modification appears effective in delaying or preventing the development of the dysmetabolic syndrome. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

Introduction The dysmetabolic syndrome is among the fastest growing disease entities in the world today. In the United States, e.g. almost 25% of adults 20 years and older have the syndrome [1]. Its prevalence increases with age, with it affecting almost 50% of US adults 60 years and older. Although the syndrome has many names, including ‘Reaven’s syndrome’, ‘syndrome X’, the ‘insulin resistance syndrome’, and the ‘metabolic syndrome’, its diagnosis centers 197

Foreyt around a constellation of metabolic derangements including central obesity, dyslipidemias (i.e. high plasma levels of triglycerides and low levels of highdensity lipoprotein cholesterol), hypertension, insulin resistance, and glucose intolerance, along with increased prothrombotic and inflammatory markers [2]. The National Cholesterol Education Program, the World Health Organization, and other national and international organizations have delineated differing but related criteria for its diagnosis [2]. This constellation of metabolic abnormalities greatly increases the risk of developing cardiovascular disease [3–6]. It appears that in a few years the dysmetabolic syndrome will overtake cigarette smoking as the strongest risk factor for the development of heart disease in the US. Sadly, the syndrome also is afflicting a growing number of children and adolescents. The prevalence of the syndrome in US children aged 12–19 years is about 1 in 10 [7]. In overweight/obese children, 1 in 3 have the syndrome. Among severely obese children, the prevalence reaches 50%. Two thirds of all adolescents have at least 1 metabolic abnormality [8, 9]. Given the rising prevalence in obesity among children and adolescents, these data are not surprising.

Treatment of the Dysmetabolic Syndrome The American Heart Association, working with the National Institutes of Health National Heart, Lung and Blood Institute and the American Diabetes Association, has provided a series of recommendations for the treatment of the dysmetabolic syndrome [10]. These recommendations include medications for managing the dyslipidemias, hypertension, and impaired glucose tolerance. However, the primary treatment for the dysmetabolic syndrome is lifestyle modification, focusing on diet and physical activity. In particular, modest weight losses significantly improve all aspects of the syndrome. Effective lifestyle changes also will aid in its prevention.

Lifestyle Modification Works The largest, most comprehensive and impressive study to date to document the beneficial effects of lifestyle modification is the Diabetes Prevention Program (DPP) [11]. The DPP randomly assigned 3,234 non-diabetic individuals with elevated fasting and post-load plasma glucose concentrations to placebo, metformin (850 mg twice a day), or a lifestyle modification invention with the goals of a 7% weight loss and 150 min of physical activity per week. Average follow-up was 2.8 years. Results were dramatic. The lifestyle modification group reduced the incidence of type-2 diabetes by 58%, the metformin group by 31%, compared to placebo. The lifestyle modification intervention was significantly more effective than metformin. Beneficial effects of the 198

Role of Lifestyle Modification lifestyle intervention were even seen on nontraditional cardiovascular risk factors including concentrations of C-reactive protein and fibrinogen [12]. In a similar study done in Finland, 522 middle-aged, overweight men and women with impaired glucose tolerance were randomly assigned to either a lifestyle intervention or control group [13]. Each participant in the intervention group received individualized counseling aimed at weight reduction, improved diet, and increased physical activity. The mean follow-up was 3.2 years. The lifestyle intervention group had a significant reduction of 58% in the relative risk of developing diabetes, similar to the results of the DPP. The reduction in risk was directly associated with beneficial changes in lifestyle. The DPP researchers also reported results on the 53% of participants (n ⫽ 1,711) who had the metabolic syndrome at baseline. The incidence of the metabolic syndrome was reduced by 41% in the lifestyle group and by 17% in the metformin group, compared with placebo [14]. The researchers attributed the dramatic effect of lifestyle modification on both the prevention of incident metabolic syndrome and a reduction of its overall prevalence primarily to reductions in waist circumference and in blood pressure. They concluded that their results demonstrate the value of lifestyle intervention in both the prevention and treatment of the syndrome, above and beyond improvement in glycemia alone and that lifestyle intervention may reduce risk in individuals with impaired glucose tolerance. Currently, the ongoing Look AHEAD (Action for Health in Diabetes) study has as its primary hypothesis that an intensive lifestyle intervention similar to DPP to reduce weight and increase physical activity will reduce cardiovascular morbidity and mortality [15]. In more than 5,000 obese adults with type-2 diabetes the study is comparing the long-term (up to 11.5 years) effects of an intensive lifestyle intervention designed to achieve and maintain a modest weight loss through decreased caloric intake and increased physical activity versus a control condition of diabetes support and education on the combined incidence of serious cardiovascular events, i.e. cardiovascular death, nonfatal myocardial infarction, and non-fatal stroke. Additional research goals include comparisons of cardiovascular disease risk factors, mortality, diabetes-related metabolic factors and complications, intervention safety, indices of health, quality of life, and a number of economic outcomes. A significant number of participants also have the dysmetabolic syndrome. This study is the first longterm randomized controlled trial to assess the effects of lifestyle intervention with reduced mortality as a primary outcome. The results of the DPP and Finnish Diabetes Prevention Study (FDPS) document the importance of lifestyle modification as the primary therapeutic intervention in individuals with the dysmetabolic syndrome, given the fact that obesity, unhealthy diet, and physical inactivity are considered to be the primary underlying risk factors for the development of the syndrome. Results from the Look AHEAD trial may also help document the role of lifestyle change as significant in reducing mortality. Through lifestyle modification aimed at the 199

Foreyt Table 1. Components of a lifestyle modification intervention

Goal setting Self-monitoring Stimulus control Cognitive restructuring Stress management Relapse prevention Social support Contracting

development of a healthy diet, increased physical activity, and behavior modification, weight loss is the primary key to treating the dysmetabolic syndrome.

Components of a Lifestyle Modification Intervention Most individuals with the dysmetabolic syndrome need to lose weight through dietary changes and increases in physical activity. Because individuals differ in their lifestyles, tailoring interventions to meet the specific needs of each person will maximize the chances of success (table 1).

Beginning Lifestyle Change Assessment Assessment of the individual with dysmetabolic syndrome involves quantification of obesity, diets and dietary patterns, physical activity, emotional problems, and motivation [16]. Lifestyle modification strategies are typically most helpful for individuals with a body mass index of ⬍40. For individuals with severe obesity, much more aggressive approaches, such as bariatric surgery, may be required. Registered dietitians are the best professionals for assessing diets and dietary patterns. Physical activity can be assessed by asking the number of minutes per day spent walking briskly or similar exercise. Emotional problems, which oftentimes make weight loss much more difficult, can be identified through the use of brief psychological questionnaires, such as the Beck Depression Inventory [17] or similar instruments. For individuals who do not appear motivated to change their diet or physical activity levels, personalizing each of the dysmetabolic syndrome risk factors and explaining how modest weight losses may improve them can sometimes be helpful. For individuals who do seem willing to make changes, encouragement from the physician can oftentimes help. Basic education about small steps which individuals can make in changing their lifestyles may help motivate them to begin a program. For individuals who are already motivated the lifestyle changes described next have been shown to be extremely useful. 200

Role of Lifestyle Modification Lifestyle Change Strategies Diet and Physical Activity The recognized diet for weight loss involves a balanced eating plan, including a deficit of about 500–1,000 kcal/day, resulting in a safe weight loss of 0.45–0.90 kg/week (1–2 lb) [10, 18]. Physical activity recommendations typically include adding 30–60 min/day of brisk walking or the equivalent on most days of the week [10]. Behavior Modification To help individuals make these dietary and physical activity changes, a number of behavior modification strategies have shown good efficacy. These strategies include a tailored problem-solving intervention, involving goalsetting, self-monitoring, stimulus control, cognitive restructuring, stress management, relapse prevention, social support, and contracting [19]. Goal-Setting Many individuals with the dysmetabolic syndrome have unrealistic goals regarding their ability to lose weight. Obese individuals frequently express a desire to lose more than 30% of their weight [20], although research studies indicate that patients who participate in a lifestyle change intervention lose about 8–10%. It is important to remember that a modest loss of about 8% generally will lead to improvement in the constellation of risk factors of the dysmetabolic syndrome. The DPP and Finnish prevention study, for example, both showed that modest weight losses through lifestyle change significantly reduced the incidence of diabetes in individuals at high risk. Unfortunately, this amount of weight loss is disappointing to some individuals who still are overweight or obese. An emphasis on achieving small, short-term goals, such as walking an extra 20 min/day and then focusing on the increased psychological feelings of enhanced well-being can sometimes help individuals understand and appreciate the gains that they have made. Accomplishing modest dietary or physical activity goals oftentimes improves self-esteem and serves as encouragement to try new ones. Setting unrealistic goals, such as losing weight too quickly through fasting or severely depriving diets, or attempting strenuous exercise, frequently results in disappointment, discouragement, and ultimately failure. Reevaluation of goals should be done on a regular basis, with revisions made as needed. Self-Monitoring Research suggests that self-monitoring is the most important of all behavioral change strategies [21]. In the DPP study, for example, the frequency of self-monitoring was related to success at achieving both the physical activity goal and the weight loss goal [22]. Self-monitoring involves three aspects: self-observation, self-recording, and feedback. The primary purpose of self-monitoring is to raise awareness. If an individual is going to successfully change physical activity and diet, it is important to know one’s activity level and what one is 201

Foreyt eating. The typical way to raise awareness of habits is through a diary in which individuals write down what they eat and the number of minutes that they are physically active. They then look up the number of calories they ate and calculate the number of calories burned through their brisk walking or other activity (calculating about 6 cal burned/min of walking). Pedometers are also helpful for raising awareness, with about 100 cal expended for each 2,000 steps. Individuals typically do not like to keep diaries. They also tend to underestimate their intake by about one third, and overestimate their activity by about one half. It really does not matter. The primary purpose of monitoring is raising awareness, not accuracy of recording. Monitoring helps remind individuals of what they are trying to do, i.e. healthy eating and more exercise. Monitoring weight by weighing on a regular schedule, i.e. daily or once a week, is also important. Food and activity diaries, and monitoring weight on a regular schedule, are essential components of long-term habit change in individuals with the dysmetabolic syndrome. If an individual agrees to make a single lifestyle change strategy, self-monitoring should be the one chosen. Stimulus Control Stimulus control involves identifying and confronting the individual’s idiosyncratic barriers to losing weight [23]. Frequently the food and activity diaries can help identify the problems that individuals are having when they stray from their healthy eating pattern or activity plan. Common problems include eating outside the home, traveling extensively, late night eating, or weather too hot, too rainy, or too cold to exercise. Using a problem-solving approach by encouraging the individual to come up with a realistic plan to the specific problem can oftentimes lead to a solution. A few ideas that individuals can be encouraged to come up with when they are confronted with barriers include: taking meals from home to the office rather than eating in the cafeteria; calling restaurants ahead of time and asking for low-fat, low-calorie meals; carrying calorie-controlled meals when on trips; planning sensible snacks in the home, or agreeing to exercise in a shopping mall during inclement weather. Cognitive Restructuring The belief that losing weight will improve all aspects of an obese individual’s life and that somehow all problems will disappear is not unusual. Unfortunately, it usually is not true. Cognitive restructuring involves strategies to help individuals change the unrealistic beliefs they may think about themselves and replace them with more positive, realistic ones [24]. Individuals are taught to identify self-enhancing, self-affirming thoughts about themselves. Repeating healthy self-affirmations such as ‘I will walk for 20 min before breakfast’, ‘small changes can make a big difference’, ‘I am no longer putting off my life until I reach some magic weight’, and repeating them daily can be excellent motivators for lifestyle change. 202

Role of Lifestyle Modification Stress Management Emotional factors frequently interfere with the development of healthy lifestyles. Stressful life events oftentimes lead to unwanted eating and subsequent weight gain. Recognizing stressors and learning strategies to deal with them can help individuals manage their eating patterns more effectively. Physical activity is an especially good strategy for reducing feelings of stress because it raises an individual’s sense of well being. The regular practice of meditation or learning techniques like progressive muscle relaxation can be useful for managing everyday stresses [25]. Relapse Prevention Lapses from one’s diet or exercise program are common. Unplanned events, emergencies, boredom, and other factors interfere with well-intentioned plans. Life gets in the way. Understanding that lapses occur and are to be expected may help reduce the probability of a relapse (giving up), leading to weight regain. Learning strategies to deal with transgressions, such as those that likely occur during holidays, can help prevent a major collapse in one’s long-term goals. Reviewing and understanding what happened and developing a strategy to deal with a situation if it occurs again can be extremely useful in preventing its recurrence [26]. Social Support Family members who eat the same healthy food and exercise together can be a strong support system for individuals with the dysmetabolic syndrome. Good friends, colleagues at work, the individual’s physician, and others can play a major role in keeping motivation high. Social support systems work because supportive individuals serve as excellent role models, provide assistance in confronting and dealing with obstacles to change, and encourage self-acceptance [20]. Contracting Formalizing agreements to make specific behavioral changes can be useful for some individuals. Contracting involves having individuals agree to one or more specific behaviors that they will do over a short period of time. The agreed upon behaviors should be simple, achievable, and realistic. Agreeing ‘to do better’ or ‘to be a better person’ is not helpful. Agreeing to walk for 20 min on Monday, Wednesday, and Friday at 8:00 in the morning is hopefully simple, achievable, and realistic. Likewise, reducing the number of desserts from every night to three times a week may also be achievable. Writing down the agreements and signing the contract usually helps in achieving the goals. Contracting is particularly useful for motivating short-term behavioral change. Changing contracts frequently helps [20]. 203

Foreyt Conclusions Research studies have clearly demonstrated the power of lifestyle modification for long-term behavioral change. The DPP provided excellent evidence of the potential of lifestyle modification intervention for the reduction of risk in individuals with the dysmetabolic syndrome. Weight loss is the key for most individuals with the syndrome. Lifestyle change strategies, including self-monitoring, stimulus control, cognitive restructuring, stress management, relapse prevention, social support, and contracting, increase the chances for adoption of a healthy diet and a habit of regular physical activity. The frequency of self-monitoring is an especially important strategy for continued success. Lifestyle modification appears effective in delaying or preventing the development of the dysmetabolic syndrome [27]. Acknowledgment This work was supported in part by grant No. DK058299 from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK).

References 1 Ford ES, Giles WH, Dietz WH: Prevalence of the metabolic syndrome among US adults: findings from the Third National Health and Nutrition Examination Survey. JAMA 2002;287: 356–359. 2 Grundy SM, Brewer HB Jr, Cleeman JI, et al: Definition of metabolic syndrome: report of the National Heart, Lung and Blood Institute/American Heart Association conference on scientific issues related to definition. Circulation 2004;109:433–438. 3 Anderson JL, Horne BD, Jones HU, et al: Which features of the metabolic syndrome predict the prevalence and clinical outcomes of angiographic coronary artery disease? Cardiology 2004;101:185–193. 4 Solymoss BC, Boruassa MG, Campeau L, et al: Effect of increasing metabolic syndrome score on atherosclerotic risk profile and coronary artery disease angiographic severity. Am J Cardiol 2004;93:159–164. 5 Lakka HM, Laaksonen DE, Lakka TA, et al: The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. JAMA 2002;288:2709–2716. 6 Isomaa B, Almgren P, Tuomi T, et al: Cardiovascular morbidity and mortality associated with the metabolic syndrome. Diabetes Care 2001;24:683–689. 7 Weiss R, Dziura J, Burgert TS, et al: Obesity and the metabolic syndrome in children and adolescents. N Engl J Med 2004;350:2362–2374. 8 Cook S, Weitzman M, Auinger P, et al: Prevalence of a metabolic syndrome phenotype in adolescents: findings from the Third National Health and Nutrition Examination Survey, 1988–1994. Arch Pediatr Adolesc Med 2003;157:821–827. 9 De Ferranti SD, Gauvreau K, Ludwig DS, et al: Prevalence of the metabolic syndrome in American adolescents: findings from the Third National Health and Nutrition Examination Survey. Circulation 2004;110:2494–2497. 10 Grundy SM, Hansen B, Smith SC Jr, et al: Clinical management of metabolic syndrome: report of the American Heart Association/National Heart, Lung, and Blood Institute/American Diabetes Association conference on scientific issues related to management. Circulation 2004;109:551–556.

204

Role of Lifestyle Modification 11 Knowler WC, Barrett-Connor E, Fowler SE, et al, Diabetes Prevention Program Research Group: Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393–403. 12 Haffner S, Temprosa M, Crandall J, et al, Diabetes Prevention Program Research Group: Intensive lifestyle intervention or metformin on inflammation and coagulation in participants with impaired glucose tolerance. Diabetes 2005;54:1566–1572. 13 Tuomilehto J, Lindstrom J, Eriksson JG, et al: Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344: 1343–1350. 14 Orchard TJ, Temprosa M, Goldberg R, et al: The effect of metformin and intensive lifestyle intervention on the metabolic syndrome: the diabetes prevention program randomized trial. Ann Intern Med 2005;142:611–619. 15 Ryan DH, Espeland MA, Foster GD, et al, Look AHEAD Research Group: Look AHEAD (Action for Health in Diabetes): design and methods for a clinical trial of weight loss for the prevention of cardiovascular disease in type 2 diabetes. Control Clin Trials 2003;24:610–628. 16 Foreyt JP: Need for lifestyle intervention: how to begin. Am J Cardiol 2005;96:11E–14E. 17 Steer RA, Cavalieri TA, Leonard DM, Beck AT: Use of the Beck depression inventory for primary care to screen for major depression disorders. Gen Hosp Psychiatry 1999;21:106–111. 18 Krauss RM, Eckel RH, Howard B, et al: AHA dietary guidelines: revision 2000: a statement for healthcare professionals from the Nutrition Committee of the American Heart Association. Circulation 2000;102:2284–2299. 19 Foreyt JP, Pendleton VR: Management of obesity. Prim Care Rep 2000;6:19–30. 20 Foster GD, Wadden TA, Vogt RA, Brewer G: What is a reasonable weight loss? Patients’ expectations and evaluations of obesity treatment. J Consult Clin Psychol 1997;65:79–85. 21 Baker RC, Kirschenbaum DS: Self-monitoring may be necessary for successful weight control. Behav Ther 1993;24:377–394. 22 Wing RR, Hamman RF, Bray GA, et al, Diabetes Prevention Program Research Group: Achieving weight and activity goals among diabetes prevention program lifestyle participants. Obes Res 2004;12:1426–1434. 23 Foreyt JP, Goodrick GK: Attributes of successful approaches to weight loss and control. Appl Prev Psychol 1994;3:209–215. 24 Foreyt JP, Poston WS 2nd: What is the role of cognitive-behavior therapy in patient management? Obes Res 1998;6(suppl 1):18S–22S. 25 Poston WS 2nd, Foreyt JP: Successful management of the obese patient. Am Fam Physician 2000;61:3615–3622. 26 Wadden TA, Foster GD, Wang J, et al: Clinical correlates of short and long-term weight loss. Am J Clin Nutr 1922;56:274–278. 27 Pritchett AM, Foreyt JP, Mann DL: Treatment of the metabolic syndrome: the impact of lifestyle modification. Curr Atheroscler Rep 2005;7:95–102.

Discussion Dr. Slama: I have two points to make. One I don’t know if it is a joke or a philosophical question, and the second one is much more serious. My first point is that if you say that 20 years from now 100% of the people will be fat, then what will be the definition of normal? Will we be normal, or will they be abnormal? Do we say that elephants are too fat? So 20 years from now we will all be normal. My serious question is that I read in the newspaper that for the first time in the history the increase in life expectancy is leveling off and even now slightly decreasing. Might it be that we are now paying the price for our very recent lifestyle? Dr. Foreyt: I will answer your second question first. That came out of a paper just published by Olshansky et al. [1] in the New England Journal of Medicine. What they did was take all the data and show that this extended life expectancy is starting to level off, and they think now it is going to start decreasing because of obesity. They

205

Foreyt particularly show data, especially in children who are developing adult risk factors earlier and earlier, that this generation of children will die before their parents do. So according to Olshansky et al. this extending lifestyle is ending. Regarding your first point about 100% of Americans being obese or overweight by 2040; as you know obesity itself is excess body fat not weight. So the body mass index doesn’t fit for everybody, but the average man in the United States has about 20% body fat, the average woman has about 25% body fat. With a body mass index of 30 a woman has about 37% body fat, and a man has about 25% body fat. So that equates very well if we look at body fat in men and women and look at that with respect to when this is going to happen, and it looks as though it is going to happen by 2040 in all of us in the United States, but all the other countries are catching up quickly. That curve is not starting to level off yet in any of the countries. The International Obesity Task Force data show that the lines are going up in all countries. I have not yet seen a country where it is leveling off or going down; not pessimistic, realistic. Dr. Eshki: I want to make a recommendation. When looking at diseases in the past century, we see that the major killers, such as heart disease, cancer, stroke, and diabetes, are on the rise. Yet there is a lot of money being spent on research and health care. Looking at the preceding theory which focuses on the predisposing, reinforcing and enabling factors, I see the enabling factors as a major player when fighting this battle. Health professionals and the public seem to be fighting this battle alone. The government and industrial sectors seem to be watching this battle take place and don’t seem to know that their intervention would make a difference. For instance, in Saudi Arabia there is a city called Medina where the governor issued a law banning smoking there. Furthermore, there are billboards in the streets around the city which are intended to educate the public about the dangers of smoking. As a result of this in the past 2 years health professionals have witnessed a dramatic reduction in the prevalence of lung cancer. So do you think the other two sectors should join the battle or should we continue to spend more money and lives, and hope for a happy ending? Dr. Foreyt: That is a great question and it is being debated over and over; at every meeting I go to that is the major focus of the meeting. I think it has to do with personal responsibility. There is a role for communities, a role for local governments, and there is a role for the federal government. What that role should be is being debated now. But ultimately it has got to do with each individual; we all have to play a part.

Reference 1 Olshansky SJ, Passaro DJ, Hershow RC, et al: A potential decline in life expectancy in the United States in the 21st century. N Engl J Med 2005;352:1138–1145.

206

Bantle JP, Slama G (eds): Nutritional Management of Diabetes Mellitus and Dysmetabolic Syndrome. Nestlé Nutr Workshop Ser Clin Perform Program, vol 11, pp 207–218, Nestec Ltd., Vevey/S. Karger AG, Basel, © 2006.

Critical Review of the International Guidelines: What Is Agreed upon – What Is Not? Nicholas Katsilambros, Stavros Liatis, Konstantinos Makrilakis First Department of Propaedeutic Medicine, University of Athens Medical School, Laiko General Hospital, Athens, Greece

Abstract The nutrition recommendations of 6 major scientific organizations (the American Diabetes Association, the Diabetes and Nutrition Study Group of the European Association for the Study of Diabetes, the Canadian Diabetes Association, the Joslin Diabetes Center and Joslin Clinic, the American Association of Clinical Endocrinologists and Diabetes UK) are reviewed. They all agree that weight loss (with reduction in energy intake and increase in physical activity) is an important therapeutic strategy in all overweight/obese individuals who have or are at risk of type-2 diabetes. Very low carbohydrate diets are not considered appropriate. The recommended proportion varies slightly (from 40 to 65%). The concept of the glycemic index is stressed as important in nearly all guidelines. Fiber intake is advised, up to 50 g/day, if tolerated. Protein intake (for normal kidney function) is advised to range from 10 to 20% of total energy. A low fat diet (⬍30–35%) is recommended by all. Saturated fat and transfatty acids should be restricted to ⬍10% and dietary cholesterol to ⬍300 mg/day. Monounsaturated fatty acids are generally considered beneficial and should replace saturated fat or carbohydrates in low-fat diets. Polyunsaturated fatty acids (PUFAs) should comprise about 10%, with the n ⫺ 3 PUFAs being more beneficial, especially for high triglyceride levels. Alcohol intake has cardioprotective effects when used in moderation. Routine supplementation of the diet with antioxidants and vitamins is not necessary. Copyright © 2006 Nestec Ltd., Vevey/S. Karger AG, Basel

Introduction Type-2 diabetes mellitus is one of the most costly and burdensome chronic diseases of our time and a condition that is increasing in epidemic proportions worldwide [1]. Its complications are a significant cause of morbidity and 207

Katsilambros/Liatis/Makrilakis mortality and a tremendous economic burden to society. Some of the risk factors for its development, such as obesity, physical inactivity and high energy diet, can potentially be modified. Compelling evidence now exists, from welldesigned randomized studies [2–4], that the disease can be prevented or delayed in subjects at high risk of its development, i.e. subjects with impaired glucose tolerance or impaired fasting glucose. The interventions studied include lifestyle modifications (with diet and exercise) and drug treatment. Weight loss with lifestyle modification seems to be the most effective way of preventing diabetes mellitus so far, given the fact that it addresses other cardiovascular disease risk factors as well (hypertension and dyslipidemia). In the established diabetic also, physical activity and diet continue to be a fundamental form of therapy. Medical nutrition therapy (MNT) guidelines for persons with diabetes have changed a lot over the previous few decades. Before the discovery of insulin, in 1921, starvation therapies were applied, with nearly total restriction of food. After the introduction of insulin in the treatment regimen, the fear that ‘sugar is bad in diabetes’ led to the adoption of low carbohydrate diets which were consequently high in fat. Later on, it was realized that high fat diets were the ones causing problems, leading to increased cardiovascular risk. Thus, a whole array of studies was initiated in search of the optimal composition of the diet in diabetic persons. Since many issues are still topics of scientific debate, they endorse the principle of individualization in MNT for diabetes and set guidelines accordingly. Today there is no one ‘diabetic’ diet. The recommended diet can only be defined as a prescription based on the assessment of treatment goals and outcomes, taking into consideration the individual needs and preferences of the people. Nutrition recommendations represent a thoughtful synthesis of a multitude of current data. The nutrition recommendations of the major scientific organizations, the American Diabetes Association (ADA) [5], the Diabetes and Nutrition Study Group (DNSG) of the European Association for the Study of Diabetes (EASD) [6], the Canadian Diabetes Association [7], the Joslin Diabetes Center and Joslin Clinic [8], the American Association of Clinical Endocrinologists (AACE) [9] and Diabetes UK [10], will be reviewed here and their similarities and disagreements analyzed. It should be emphasized that the recommendations are based on the best available evidence of these scientific organizations, categorized according to their strength. The ADA grades their recommendations into four categories: those with strong supporting evidence, those with some supporting evidence, those with limited supporting evidence and those based on expert consensus. The DNSG of the EASD grade their recommendations according to their strength of evidence based on 5 evidence classes (Ia, Ib, IIa, IIb, III) and a separate class (IV) reserved for statements from expert committees based on the Scottish Intercollegiate Guidelines Network [11]. The classes of the recommendations are: 208

Critical Review of the International Guidelines Ia Ib IIa IIb III

Meta-analysis of randomized controlled trials At least one randomized controlled trial At least one well-designed controlled study without randomization At least one other type of well-designed quasi-experimental study Well-designed nonexperimental descriptive studies, such as comparative studies, correlation studies and case studies IV Expert committee reports or opinions and/or clinical experiences of respected authorities The grades of the recommendations are as follows: (A) Requires at least one randomized controlled trial as part of a body of literature of overall good quality and consistency addressing the specific recommendation (evidence levels Ia, Ib). (B) Requires the availability of well-conducted clinical studies but no randomized clinical trials on the topic of recommendation (evidence levels IIa, IIb, III). (C) Requires evidence obtained from expert committee reports or opinions and/or clinical experiences of respected authorities. Indicates an absence of directly applicable clinical studies of good quality (evidence level IV). The Joslin Diabetes Center follows the same pattern of evidence as the ADA, whereas the Canadian Diabetes Association and the AACE do not provide a grading system for their recommendations. The Diabetes UK recommendations are based on the ADA and DNSG technical reviews.

Aims and Goals of Nutritional Advice There is a general agreement among the various scientific organizations mentioned above regarding the goals of MNT in diabetes. These goals are generally the following, as addressed in the ADA statement [5]: • Attain and maintain recommended metabolic outcomes, including glucose and HbA1c levels, LDL cholesterol, HDL cholesterol, triglyceride levels, blood pressure and body weight. • Prevent and treat the chronic complications and comorbidities of diabetes. Modify nutrient intake and lifestyle as appropriate for the prevention and treatment of obesity, dyslipidemia, cardiovascular disease, hypertension and nephropathy. • Improve health through healthy food choices and physical activity. • Address individual nutritional needs, taking into consideration personal and cultural preferences and lifestyle, while respecting the individual’s wishes and willingness to change. Goals of MNT that apply to specific situations include the following: • For youth with type-1 diabetes mellitus, MNT should provide adequate energy to ensure normal growth and development. Insulin regimens should be integrated into usual eating and physical activity habits. 209

Katsilambros/Liatis/Makrilakis For youth with type-2 diabetes mellitus, who are usually overweight/ obese, appropriate changes in eating and physical activity habits should be facilitated. • For pregnant and lactating women, adequate energy and nutrients needed for optimal outcomes should be provided. • For older adults, nutritional and psychological needs of an aging individual should be addressed. • For individuals treated with insulin or insulin secretagogues, selfmanagement education for treatment (and prevention) of hypoglycemia, acute illness and exercise-related blood glucose problems should be provided. • For individuals at risk of diabetes, physical activity and food choices that facilitate moderate weight loss, or at least prevent weight gain, should be encouraged. The DNSG of the EASD does not provide specific aims and goals of nutritional advice in its most recent publication [6], although the previous one in 1999 generally – albeit briefly – agrees with the aforementioned statements of ADA [12]. The Canadian Diabetes Association [7], the AACE [9] and Diabetes UK [10] also generally and briefly agree with them. The Joslin Diabetes Center and Joslin Clinic additionally emphasizes the goals of improving postprandial hyperglycemia, postprandial hypertriglyceridemia, body fat distribution with reduction of visceral fat and reduction of cardiovascular risk, as evidenced by improvement of endothelial function and endothelial markers and reduction of inflammatory cytokines [8]. All the scientific organizations (with the exception of the DNSG of the EASD, which does not make any specific recommendation) generally agree that, given the complexity of the dietary issues, a registered dietitian, familiar with the components of diabetes MNT should be part of the team that provides dietary advice to the diabetic individual. •

Body Weight and Energy Balance There is a general consensus among the various scientific organizations that weight loss is an important therapeutic strategy in all overweight or obese individuals who have type-2 diabetes or are at risk of developing type-2 diabetes. The primary approach to achieving weight loss, in the vast majority of cases, is a therapeutic lifestyle change, which includes a reduction in energy intake and an increase in physical activity. The ADA recommends a moderate decrease in caloric balance (by 500–1,000 kcal/day) that will result in a slow but progressive weight loss of about 1–2 lb/week. For most patients, a weight loss diet is advised to provide at least 1,000–1,200 kcal/day for women and 1,200–1,600 kcal/day for men. The DNSG of the EASD does not make specific caloric recommendations, but stresses that a weight reduction of as little as 10% of the initial body 210

Critical Review of the International Guidelines weight will have very beneficial effects on insulin sensitivity, blood pressure and lipid levels in overweight diabetic persons. The goal should be to decrease BMI under 25 kg/m2 and to prevent weight regain thereafter. They also stress that overweight patients with type-1 diabetes may also become insulin resistant and weight loss may lead to a reduction in insulin dose and improved glycemic control. The Joslin Diabetes Center and Joslin Clinic recommends a modest and gradual weight reduction of 1 lb every 1–2 weeks as an optimal therapeutic target, with a reduction of daily caloric intake by 250–500 kcal. Total daily caloric intake should not be less than 1,000–1,200 kcal/day for women and 1,200–1,600 kcal/day for men. The AACE emphasizes the fact that simply a negative caloric balance can decrease insulin resistance in type-2 diabetes and that loss of as few as 10–20 lb (4.5–9 kg) will be helpful. For patients with type-1 diabetes, the patient must understand the action and duration of the insulin being used as well as the effect of the timing of the intake of food on the insulin action. Flexibility of insulin dosing and timing must be taught to the patient. Diabetes UK recommends that a loss of 1–2 kg per month by means of a sustained energy deficit of approximately 500 kcal/day should be regarded as satisfactory. Furthermore, given the fact that body weight tends to increase with aging up to the sixth decade, avoidance of further weight gain may be considered a success in some patients. The Canadian Diabetes Association does not have specific recommendations regarding the total amount of energy that needs to be consumed by diabetic persons in order to lose weight. It should be noted that although weight loss is a major objective in the guidelines of all scientific organizations, there is actually a lack of prospective trials with hard end points (mortality). Also, no specific recommendations are provided on how to lose weight. Consequently, there is obviously a need for long-term, well-designed trials in order to assess the benefit from weight management in patients with diabetes. In this regard, the Look AHEAD: Action for Health in Diabetes Program aims to examine, in overweight volunteers with type-2 diabetes, the long-term effects of an intensive lifestyle intervention program, designed to achieve and maintain weight loss by decreased caloric intake and increased physical activity. The composite end point includes: cardiovascular death (including fatal myocardial infarction and stroke), nonfatal myocardial infarction, and nonfatal stroke.

Macronutrient Composition of the Diet The various scientific organizations have small differences in their recommendations for the macronutrient composition of the diet (including carbohydrates, protein and fat) in diabetic persons (table 1). 211

Katsilambros/Liatis/Makrilakis Table 1. Recommendations of the various scientific organizations regarding macronutrient content of the diet in diabetes Organization

CHO

Protein

Fat

ADA

45–65% (at least 130 g/day; previously 60–70% combined with MUFA)

15–20% (microalbuminuria: 0.8–1.0 g/kg/day; overt nephropathy: 0.8 g/kg/day)

Low fat diet SAFA ⬍10% (SAFA ⬍7% if LDL-C ⬎100 mg/dl) PUFA ⬃10% Trans-FAs minimized Cholesterol ⬍300 mg/day (cholesterol ⬍200 mg/day if LDL-C ⬎100 mg/dl)

10–20%

⬍35% (⬍30% in overweight persons) SAFA ⫹ trans-FAs: ⬍10% [lower intake (⬍8%) if LDL-C is elevated] MUFA 10–20% PUFA ⬍10% Cholesterol ⬍300 mg/day

GI important Fiber up to 50 g/day DNSG of EASD 45–60% GI important Fiber ⬎40 g/day Free sugars ⬍50 g/day

Joslin Clinic

⬃40% (at least 130 g/day) GI important Fiber up to 50 g/day (minimum 20–35 g/day)

AACE

55–60%

Canadian

50–55%

Diabetes Association

GI important Sucrose ⬍10%

212

(T1DM with overt nephropathy: 0.8 g/kg/day; microalbuminuria in T1DM/T2DM or macroalbuminuria in T2DM: no firm recommendation)

20–30% Patients with nephropathy: consult a nephrologist 15–20% In microalbuminuria: 10–15% 15–20%

30–35% SAFA ⬍10% (⬍7% if LDL-C ⬎100 mg/dl) Cholesterol ⬍300 mg/day (⬍200 mg/day if LDL-C ⬎100 mg/dl) ⬍30% (⬍15% if overweight or dyslipidemic) ⬍30% SAFA ⫹ trans-FAs: ⬍10% PUFA ⬍10% (n-3 PUFA preferred) MUFA preferred

Critical Review of the International Guidelines Table 1. (continued) Organization

CHO

Protein

Fat

Diabetes UK

45–60% (combined with MUFA: 60–70%) GI important Fiber: no quantitative recommendation Sucrose ⬍10%

Not ⬎1 g/kg/day

⬍35% SAFA ⫹ trans-FAs: ⬍10% n-6 PUFA ⬍10% n-3 PUFA: twice weekly MUFA 10–20% (MUFA combined with CHO: 60–70%)

CHO ⫽ Carbohydrates; FAs ⫽ fatty acids; LDL-C ⫽ low density lipoprotein cholesterol; SAFA ⫽ saturated fatty acids; T1DM ⫽ type-1 diabetes mellitus.

The ADA have, in their most recent clinical practice recommendations [5], changed their guidelines regarding the carbohydrate content of the diet. In their previous recommendations [13, 14] they stated that carbohydrate and monounsaturated fat combined should provide 60–70% of the total daily energy intake (based on an expert consensus statement), without giving specific individual figures for the carbohydrates or the monounsaturated fat. (They recommend that contribution of carbohydrates and monounsaturated fat in energy intake should be individualized, based on nutrition assessment, metabolic profiles and weight and treatment goals.) In their 2005 recommendations [5], however, the amount of carbohydrate in the diet is specified as 45–65% of total energy intake, with the provision that the absolute quantity of carbohydrate is at least 130 g/day, due to the absolute requirement of the brain and the central nervous system for glucose as an energy source. The recommendations for protein and fat intake are not mentioned in the new guidelines of the ADA and are assumed to be the same as the previous ones, based on the 2002 Technical Review [14]. Protein intake is recommended at 15–20% of total energy (provided that renal function is normal) and total fat should be reduced to facilitate weight loss. The primary dietary fat goal in persons with diabetes is mentioned as a decrease in saturated fat intake to ⬍10% of total energy (⬍7% if low density lipoprotein cholesterol is ⬎100 mg/dl) and dietary cholesterol to ⬍300 mg/day (⬍200 mg/day if low density lipoprotein cholesterol is ⬎100 mg/dl). It is still considered as desirable that monounsaturated fat should replace saturated fat. Trans-unsaturated fat intake should be minimized and polyunsaturated fat reduced to ⬃10% of total energy. A great deal of scientific debate was generated over the previous years regarding the utility and importance of the glycemic index (GI) and its incorporation in the dietary recommendations [15–20]. The ADA, in their previous practice guidelines [13, 14], did not consider that there was sufficient data of 213

Katsilambros/Liatis/Makrilakis long enough duration, to justify the incorporation of the GI into the nutritional guidelines. The total amount of carbohydrates in meals and snacks was considered more important than their source or type. In their most recent guidelines however [5], based on a review of new evidence [21], the ADA concludes that both the amount (grams) of carbohydrate as well as the type of carbohydrate in a food influence the blood glucose level, and that the use of the GI can provide an additional benefit over that observed when total carbohydrate is considered alone. The GI concept is adopted by the DNSG of the EASD, the Joslin Diabetes Center and Joslin Clinic, the Canadian Diabetes Association and Diabetes UK as well and is considered able to discriminate foods in quite a satisfactory way [22]. In conclusion, as regards macronutrient composition of the diet (table 1), all scientific organizations agree that very low carbohydrate diets are not appropriate for people with diabetes. The recommended proportion varies slightly among them (from 40 to 65%), being lowest in the Joslin Diabetes Center recommendation (⬃40%). The concept of the GI is stressed as important in almost all guidelines now. Fiber intake is advised up to an amount of 50 g/day, as long as it can be tolerated. Foods containing carbohydrates from whole grains, fruits, fresh vegetables, legumes and low-fat milk should be preferred, because of their high fiber and low GI contents. If desired, and if blood glucose levels are satisfactory, moderate intakes of free sugars (up to 50 g/day of sucrose) may be incorporated in the diet of individuals with diabetes as well. It should be noted that the recommended range of carbohydrate intake (40–65% total energy for most experts) is based on the limits for total fat and protein intakes. For patients with persistently raised triglyceride levels a trial of carbohydrate intake at the lower end of the recommended intake range may be appropriate. Carbohydrate-rich, low GI foods are suitable as carbohydrate-rich choices provided the other attributes of the foods are appropriate [23]. Protein intake (for people with normal kidney function) is advised to range from 10 to 20% of total energy, with the exception of the Joslin Clinic that recommends 20–30% of total caloric intake to be provided by protein (although not based on strong scientific evidence). According to the Joslin Clinic recommendations, emerging data suggest that these diets aid in the sensation of fullness, whereas low protein meal plans are associated with increased hunger. Thus, lean protein together with healthy fats may serve to reduce appetite and assist patients in achieving and maintaining a lower calorie level [24]. For type-1 diabetic persons with established nephropathy (proteinuria) a lower amount of protein intake (0.8 g/kg/day) is recommended by the ADA and the DNSG of the EASD. In individuals with microalbuminuria, the ADA recommends 0.8–1.0 g/kg/day of protein intake, whereas the DNSG of the EASD state that there is not sufficient evidence to make a firm recommendation for them, as well as for type-2 diabetic persons with macroalbuminuria. The Joslin Clinic recommends that patients with nephropathy 214

Critical Review of the International Guidelines should consult a nephrologist before they increase total or percentage protein in their diet. Regarding fat intake, a low fat diet (⬍30–35%) is recommended by all scientific organizations. There is an unanimous consensus that saturated fat and trans-fatty acids should be restricted to ⬍10% of total energy intake and dietary cholesterol to ⬍300 mg/day. Monounsaturated fatty acids (MUFA) are generally considered beneficial and should replace saturated fat or carbohydrates in low fat diets. Olive oil consumption (the richest MUFA-containing fat and an indispensable component of the Mediterranean diet [25]) is equivalent to polyunsaturated fatty acids (PUFA) when compared as regards blood glucose and blood lipid levels [26]. PUFA should comprise about 10% of total caloric intake with the n-3 PUFA (from oily fish and plant sources) being more beneficial, especially for high triglyceride levels. It should also be noted that no controlled intervention studies in subjects with diabetes mellitus having sufficient power to demonstrate that effects of dietary fat on cardiovascular or other disease endpoints exist. We only have very limited data from observational studies. These recommendations are mainly based on studies in nondiabetic subjects. The percentage of people who actually adhere to these recommendations of nutrient intake is very limited in the various countries [27].

Alcohol Intake The ADA recommend that if individuals choose to drink alcohol, daily intake should be limited to one drink for adult women and two drinks for adult men. It should be avoided by pregnant women and people with other medical problems, like pancreatitis, advanced neuropathy, severe hypertriglyceridemia or alcohol abuse. Since it can have both hypoglycemic or hyperglycemic effects in people with diabetes, alcohol should be consumed during meals. It has been shown that moderate amounts of alcohol ingestion (5–15 g/day) is associated with a decreased risk of coronary heart disease. The DNSG also agrees with the recommendation of moderate alcohol intake in diabetes (10 g/day for women, 20 g/day for men) provided that it is consumed during meals (especially by patients using insulin). The Joslin Clinic does not have any specific recommendations for alcohol, whereas the AACE recommends that patients with diabetes should avoid or limit the use of alcohol, because predicting or anticipating its effect on blood glucose is difficult. The Canadian Diabetes Association recommends limiting intake to 1–2 drinks/day, whereas Diabetes UK suggests sensible drinking for the general population. Moderate alcohol consumption of 1–3 units daily probably has a cardioprotective effect. In conclusion, it looks like alcohol intake does have a cardioprotective effect when used in moderation by diabetic persons [28] and as long as it is 215

Katsilambros/Liatis/Makrilakis consumed sensibly (especially with meals by people using insulin) has no detrimental effect. Antioxidants and Vitamins It is generally agreed by all scientific organizations that routine supplementation of the diet with antioxidants and vitamins is not necessary. The DNSG of the EASD recommends that the consumption of foods naturally rich in dietary antioxidants (tocopherols, carotenoids, vitamin C, flavonoids, polyphenols, phytic acid), trace elements and other vitamins should be encouraged. The daily consumption of a range of vegetables and fruit is encouraged as well as regular consumption of wholegrain breads, cereal and oily fish. Salt intake is advised to be limited to ⬍6 g/day. Prevention of Type-2 Diabetes During the last few years well-conducted, randomized studies have unequivocally shown that type-2 diabetes mellitus can be effectively prevented or delayed by lifestyle modification programs in people at risk of developing it [2–4]. The ADA, the DNSG of the EASD and Diabetes UK have incorporated diabetes prevention guidelines in their nutritional recommendations. The ADA gives emphasis to the fact that structured programs that focus on lifestyle changes, including education, reduced fat and energy intake, regular physical activity and regular participant contact, can reduce the risk of developing diabetes. The DNSG of the EASD state that weight reduction and maintenance of weight loss in overweight individuals is a critical component of the lifestyle modification program, which may be expected to reduce the risk of developing type-2 diabetes. The appropriate macronutrient composition of the diet is a total fat ⬍30% of energy intake, saturated fat ⬍10% and fiber intake ⬎15 g/1,000 kcal. Diabetes UK state that structured programs of lifestyle change which emphasize weight loss by reduced energy and fat intake and increased physical activity can reduce the risk of overweight people with impaired glucose tolerance to develop type-2 diabetes. In both studies of diabetes prevention mentioned above [3, 4] frequent ingestion of wholegrain products, vegetables, fruits, low-fat milk and meat products, soft margarines and vegetable oils rich in MUFAs was the means of facilitating the appropriate macronutrient composition of the diet for achieving a weight loss of 5–7% of the initial body weight. Thus, for people who are overweight/obese, especially if there is a strong family history of diabetes or if they have impaired glucose metabolism, weight loss with diet and exercise should be strongly advised. 216

Critical Review of the International Guidelines References 1 King H, Aubert RE, Herman WH: Global burden of diabetes, 1995–2025. Prevalence, numerical estimates and projections. Diabetes Care 1998;21:1414–1431. 2 Pan XR, Li GW, Hu YH, et al: Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance. The Da Qing IGT and Diabetes Study. Diabetes Care 1997;20: 537–544. 3 Tuomilehto J, Lindstrom J, Ericsson JG, et al, Finnish Diabetes Prevention Study Group: Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344:1343–1350. 4 Knowler WC, Barrett-Connor E, Fowler S, et al, Diabetes Prevention Program Research Group: Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393–403. 5 American Diabetes Association: Position Statement: Standards of medical care in diabetes. Diabetes Care 2005;28(suppl 1):S4–S36. 6 Mann JI, De Leeuw I, Hermansen K, et al, Diabetes and Nutrition Study Group (DNSG) of the European Association for the Study of Diabetes (EASD): Evidence-based nutritional approaches to the treatment and prevention of diabetes mellitus. Nutr Metab Cardiovasc Dis 2004;14:373–394. 7 Wolever T, Gougeon R, Freeze C, et al: Nutrition therapy. Canadian Diabetes Association. Clinical Practice Guidelines Expert Committee, 2003 (accessed online at www.diabetes.ca). 8 Joslin Diabetes Center and Joslin Clinic: Clinical nutrition guidelines for overweight and obese adults with type 2 diabetes, prediabetes or at high risk of developing type 2 diabetes, 2005, pp 1–5 (accessed online at www.joslin.org). 9 The American Association of Clinical Endocrinologists Medical Guidelines for the Management of Diabetes Mellitus: The AACE system of intensive diabetes self-management – 2002 update. Endocr Pract 2002;8(suppl 1):45–48. 10 Connor H, Annan F, Bunn E, et al, Nutrition Subcommittee of the Diabetes Care Advisory Committee of Diabetes UK: The implementation of nutritional advice for people with diabetes. Diabet Med 2003;20:786–807. 11 Scottish Intercollegiate Guidelines Network. SIGN Guidelines: An introduction to SIGN methodology for the development of evidence-based clinical guidelines, 1999 (www.show. scot.nhs.uk/sign/home.htm). 12 The Diabetes and Nutrition Study Group (DNSG) of the European Association for the Study of Diabetes (EASD), 1999: Recommendations for the nutritional management of patients with diabetes mellitus. Eur J Clin Nutr 2000;54:353–355. 13 American Diabetes Association: Clinical Practice Recommendations 2004: Nutrition principles and recommendations in diabetes. Diabetes Care 2004;27(suppl 1):S36–S46. 14 Franz MJ, Bantle JP, Beebe CA, et al: Evidence-based nutrition principles and recommendations for the treatment and prevention of diabetes and related complications. Diabetes Care 2002;25:148–198. 15 Mann J, Hermansen K, Vessby B, Toeller M, Diabetes Nutrition Study Group of the European Association for the Study of Diabetes: Evidence-based nutritional recommendations for the treatment and prevention of diabetes and related complications. A European perspective (letter). Diabetes Care 2002;25:1256–1258. 16 Franz MJ, Bantle JP: Response to the Diabetes Nutrition Study Group of the European Association for the Study of Diabetes. Diabetes Care 2002;25:1258–1259. 17 Irwin T: New dietary guidelines from the American Diabetes Association (letter). Diabetes Care 2002;25:1262. 18 Franz MJ, Bantle JP: Response to Irwin. Diabetes Care 2002;25:1262–1263. 19 Wolever T: American Diabetes Association evidence-based nutrition principles and recommendations are not based on evidence (letter). Diabetes Care 2002;25:1263–1264. 20 Franz MJ, Bantle JP: Response to Wolever. Diabetes Care 2002;25:1264–1265. 21 ADA Statement: Dietary carbohydrate (amount and type) in the prevention and management of diabetes. A statement by the American Diabetes Association. Diabetes Care 2004;27:2266–2271. 22 Bornet FR, Costagliola D, Rizkalla et al: Insulinemic and glycemic indexes of six starch-rich foods taken alone and in a mixed meal by type 2 diabetics. Am J Clin Nutr 1987;45:588–595.

217

Katsilambros/Liatis/Makrilakis 23 Katsilambros N, Philippides P, Davoulos G, et al: Sesame-derived candies and glycaemic response in type II diabetic subjects. Diabetes Nutr Metab 1991;4:325–327. 24 Brinkworth GD, Noakes M, Parker B, et al: Long-term effects of advice to consume a highprotein, low-fat diet, rather than a conventional weight-loss diet, in obese adults with type 2 diabetes: one-year follow-up of a randomized trial. Diabetologia 2004;47:1677–1686. 25 Trichopoulou A, Costacou T, Bamia C, Trichopoulos D: Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med 2003;348:2599–2608. 26 Katsilambros N, Kostalas G, Michalakis N, et al: Metabolic effects of long-term diets enriched in olive oil or sunflower oil in non-insulin-dependent diabetes. Nutr Metab Cardiovasc Dis 1996;6:164–167. 27 Thanopoulou A, Karamanos B, Angelico F, et al, Multi-Centre Study of the Mediterranean Group for the Study of Diabetes (MGSD): Nutritional habits of subjects with type 2 diabetes mellitus in the Mediterranean basin: comparison with the non-diabetic population and the dietary recommendations. Multi-Centre Study of the Mediterranean Group for the Study of Diabetes (MGSD). Diabetologia 2004;47:367–376. 28 Pitsavos C, Makrilakis K, Panagiotakos DB, Chrysohoou C, et al: The J-shape effect of alcohol intake on the risk of developing acute coronary syndromes in diabetic subjects: the CARDIO2000 II Study. Diabet Med 2005;22:243–248.

Discussion Ms. Franz: It is essential that nutrition recommendations be evidence-based, but it is also important to have evidence that recommendations can be implemented in the ‘real world’ and that outcomes from free-living subjects are similar to the findings from subjects in controlled research settings. For example, a number of small, short-term studies have reported benefits from a diet with 30% of energy intake from protein [1]. However, in a study comparing long-term compliance to diets low in fat and high in either protein or carbohydrate, at the 52-week follow-up protein intake was similar in each group suggesting that long-term it is difficult to change protein intake [2]. Dr. Katsilambros: Thank you very much for your comments.

References 1 Gannon MC, Nuttall FQ, Saeed A, et al: An increase in dietary protein improves the blood glucose response in persons with type 2 diabetes. Am J Clin Nutr 2003;78:734–741. 2 Brinkworth GD, Noakes M, Keogh JB, et al: Long-term effects of a high-protein, low-carbohydrate diet on weight control and cardiovascular risk markers in obese hyperinsulinemic subjects. Int J Obes Relat Metab Disord 2004;28:661–670.

218

Subject Index

Acarbose, diabetes prevention trials 32, 33, 40–42 Accelerator hypothesis accelerators beta cell autoimmunity 140, 141, 147, 150, 151 insulin resistance 140, 142–144, 150 obesity 145–148 diabetes risk and susceptibility 144 epidemiology of diabetes 144–146 overview 139–141 pathophysiology of diabetes type-1 141, 142, 150, 151 type-2 142, 150, 151 Adiponectin, gestational diabetes mellitus pathophysiology 161 Alcohol, dietary guidelines 215, 216 Angiotensin-converting enzyme (ACE), diabetes prevention trials with inhibitors and receptor antagonists 35, 36 Antioxidants, see also specific antioxidants diet considerations 124, 125 dietary guidelines 216 intervention trials in diabetes 114–118, 124 observational studies in diabetes 109–114 oxidative stress and hyperglycemia 108, 109

postprandial glycemia and suppression in serum 51 recommendations 119, 120 supplementation pros and cons 118, 119 Bariatric surgery costs 40 diabetes prevention trials 34, 35, 40 effects on hypertension 10 energy economy in patients 135, 136 Beta cell apoptosis, accelerator hypothesis 140, 141, 147, 150, 151 Body mass index, see Obesity Celiac disease, type-1 diabetes association 152 Cortisol, adipose tissue production 8 Diet, see also Glycemic index; Meal replacement drinks; Obesity; Weight loss antioxidants 124, 125 breakfast effect on body weight 194 diabetes prevention trials 99, 100, 103, 104, 216 exercise interactions with meals 194, 195 fiber effects 100 guidelines for diabetics alcohol 215, 216

219

Subject Index Diet (continued) guidelines for diabetics (continued) American Diabetes Association 98, 99, 208, 209, 213–215 American Heart Association 99 Chinese Ministry of Health 99 energy balance and weight loss 210, 211 European Association for the Study of Diabetes 99, 208, 209 fatty acids 104 goals 209, 210 macronutrients 211–215 mortality studies 99 popular weight loss diets 101, 104, 105 salt intake 105 very low calorie diets 174, 175 Dipeptidyl peptidase-4 (DDP4), inhibitors in gestational diabetes mellitus management 167 Dyslipidemia, see Low-density lipoprotein; Triglycerides Dysmetabolic syndrome, see Metabolic syndrome Exercise, see also Lifestyle modification children 194 diabetes type-2 exercise testing and prescription 189 intensity and modes of physical activity 187, 188, 193 management 187 metabolism effects 187 safety 189, 190 duration versus intensity 195 gestational diabetes mellitus management 162, 166–168 meal interactions 194, 195 obesity exercise prescription 194 prevention 186 treatment 186 weight loss goals 195 maintenance 187, 193 mechanisms 192 physical activity increase strategies 188, 189, 193 physical inactivity and diabetes risks 183, 184

220

Fat dietary guidelines 213, 215 limiting in diet 71, 105 Fiber dietary fiber hypothesis 43, 44 effects in diet 100 Free fatty acids (FFAs) low glycemic index food effects 46, 47 metabolic syndrome levels 5–7, 10–12 Fructose corn syrup production 84, 85 diet responses in diabetics 85–87 food sources 84, 85 glycemic index 85 metabolism 84 obesity relationship 88–90, 92, 93, 101 rat studies 94, 95 sex differences in response 92, 93 sweetness 84 Gestational diabetes mellitus (GDM) definition 155 generational effects 167, 168 incidence trends 156, 165, 166 macrosomia prevalence 168 maternal nutrition and fetal growth effects of obesity insulin resistance 157 nutrient delivery 157, 159 pathophysiology adipocyte factors 161 placental factors 159, 160 screening 155, 156 treatment dipeptidyl peptidase-4 inhibitors 167 exercise 162, 166–168 glyburide 163, 166 goals 161 insulin therapy 162, 163 medical nutrition therapy 161, 162, 166 metformin 163, 164 rationale 161 Glucor®, mechanism and prescription 76 ␣-Glucosidase inhibitors, see Glucor® Glutathione, status in diabetes 122, 123 Glyburide, gestational diabetes mellitus management 163, 166 Glycemic control antioxidant supplementation 51

Subject Index importance in complication prevention 50, 51 traditional Chinese medicine trials 20, 21 Glycemic index (GI) classification 54, 58, 59 definition 57, 58 dietary fiber hypothesis 43, 44 dietary recommendation utilization 213, 214 energy expenditure effects 54 examples of foods 59, 85, 86 food company labeling 64, 65, 71 fructose 85 limitations 45, 59–61 low glycemic food effects blood lipid response 55, 56, 61, 74 clinical effects 48, 49, 54 clinical recommendations 77, 78, 80, 81 epidemiological evidence for health effects 49, 62, 100 free fatty acid response 46, 47 glucose absorption 45, 46, 55, 74 HbA1c response 61, 62, 71 weight loss studies 63, 64, 69, 70 measurement of food response 44, 45, 53 HbA1c, low glycemic index food effects 61, 62, 71 Herbal therapy, see Traditional Chinese medicine Hormone replacement therapy, diabetes prevention trials 36, 37 Hypertension bariatric surgery effects 10 metabolic syndrome 12 Insulin resistance accelerator hypothesis 140, 150 body mass index and risks 145–148 diabetes treatment implications 153 fetal insulin hypothesis 143 gestational diabetes mellitus 157 metabolic syndrome 5–8 oxidative stress and hyperglycemia 108, 109 thrifty genotype hypothesis 142 thrifty phenotype hypothesis 142

Insulin therapy gestational diabetes mellitus 162, 163 preprandial use 76 Leptin body weight regulation 192 gestational diabetes mellitus pathophysiology 159–161 Life expectancy, trends 205, 206 Lifestyle modification, see also Diet; Exercise;; Weight loss benefits over pharmacotherapy 39, 40 components assessment 200 behavior modification 201 cognitive restructuring 202 contracting 203 diet and physical activity 201 goal-setting 201 relapse prevention 203 self-monitoring 201, 202, 204 social support 203 stimulus control 202 stress management 203 diabetes prevention studies 49, 50 metabolic syndrome studies 198–200 prospects for study 204 Low-density lipoprotein (LDL) fructose response 87, 88 low glycemic index food effects 55, 56, 61, 74 Macrosomia, see Gestational diabetes mellitus Meal replacement drinks children 181 composition 175, 179 duration of use 179, 180 popularity 176 prospects for study 177 weight loss compliance 179, 180 efficacy 175, 176, 180 mechanisms 176, 177 Medical nutrition therapy (MNT) benefits in type-2 diabetes 173, 174 diabetes trials 66 gestational diabetes mellitus 161, 162, 166 goals 209, 210 guidelines, see Diet individualization 208

221

Subject Index Metabolic syndrome blood pressure 12 China 16 definitions 1–3, 5, 9, 10 diagnosis 198 epidemiology 3, 4, 197 etiology 5–8 free fatty acid levels 5–7, 10–12 insulin resistance 5–8 lifestyle modification intervention trials 198–200 obesity role 7, 8, 11, 12 traditional Chinese medicine, see Traditional Chinese medicine treatment 6, 7 Metformin diabetes prevention trials 32, 33, 40–42 gestational diabetes mellitus management 163, 164 mechanism of action 76 Motivational interviewing, weight loss patients 133 National Weight Control Registry 193, 195 Novonorm®, hypoglycemic activity 76 Obesity, see also Diet; Weight loss bariatric surgery effects on hypertension 10 body mass index and insulin resistance risks 145–148 breakfast effect on body weight 194 diabetes risks 171, 172, 184, 185 etiology 41 exercise prescription 194 prevention trials 186 weight loss goals 195 maintenance 187, 193 mechanisms 192 fructose effects 88–90, 92, 93, 101 low glycemic index food weight loss studies 63, 64, 69, 70 metabolic syndrome role 7, 8, 11, 12 trends 205, 206 Orlistat, diabetes prevention trials 33, 34 Oxidative stress, see Antioxidants Physical activity, see Exercise Pioglitazone, diabetes prevention trials 38

222

Pravastatin, diabetes prevention trials 36 Protein, dietary guidelines 214, 215 Rehmannia six recipe, diabetes management 25–28 Salt intake guidelines 105 Starlix®, hypoglycemic activity 76 Stress management, lifestyle modification 203 Syndrome X, see Metabolic syndrome Thyroid hormone, energy economy regulation 136 Traditional Chinese medicine (TCM), diabetes management adverse effects 22 antidiabetic drug therapy combination 20, 21, 27–29 glycemic control trials 20, 21 herbs 17–19, 25–28 modes 16, 17 overview 15, 16 recommendations 22 symptom relief trials 17 Triglycerides, fructose effects 88–90, 92, 94 Troglitazone, diabetes prevention trials 33 Tumor necrosis factor-␣ (TNF-␣), gestational diabetes mellitus pathophysiology 159, 160 Very low calorie diets (VLCDs), weight loss 174, 175 Vitamin A, intervention trials in diabetes 118 Vitamin C, intervention trials in diabetes 118 Vitamin E intervention trials in diabetes 114, 118 mortality impact in diabetes 125 observational studies in diabetes 109 supplementation in glycemic control 51 Weight loss benefits in type-2 diabetes 172, 173 children 137 diabetes prevention studies 172 diet, see also Diet; Meal replacement drinks

Subject Index guidelines for energy balance 210, 211 low carbohydrate versus low fat 127, 128 low glycemic index food studies 63, 64, 69, 70, 137 popular weight loss diets 101, 104, 105 recommendations 136, 137 safety 137 very low calorie diets 174, 175 difficulty in diabetics 174 exercise in maintenance 187, 192 maintenance 128, 129

motivational interviewing 133 sympathetic nervous system response in obese patients 134, 135 therapeutic patient education diet compliance 129, 130 food binge cycle breaking 131, 132 obese patients 131, 132 quality of life outcomes 130 Xenical®, diabetes prevention trials 34, 40 Zinc, status in diabetes 123

223