Dietary Therapy for Obesity: An Emperor With No Clothes
2008; Lippincott Williams & Wilkins; Volume: 51; Issue: 6 Linguagem: Inglês
10.1161/hypertensionaha.106.085944
ISSN1524-4563
Autores Tópico(s)Obesity, Physical Activity, Diet
ResumoHomeHypertensionVol. 51, No. 6Dietary Therapy for Obesity: An Emperor With No Clothes Free AccessArticle CommentaryPDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessArticle CommentaryPDF/EPUBDietary Therapy for Obesity: An Emperor With No Clothes Allyn L. Mark Allyn L. MarkAllyn L. Mark From the Carver College of Medicine, Center on Functional Genomics of Hypertension, and Department of Internal Medicine, University of Iowa, Iowa City. Originally published12 May 2008https://doi.org/10.1161/HYPERTENSIONAHA.106.085944Hypertension. 2008;51:1426–1434Other version(s) of this articleYou are viewing the most recent version of this article. Previous versions: May 12, 2008: Previous Version 1 The prevalence of obesity has increased substantially in the past 3 decades and is projected to increase further in the years ahead. It increases the risk of diabetes mellitus, dyslipidemia, hypertension, cardiovascular disease, sleep apnea, nonalcoholic hepatic steatosis, gallbladder disease, osteoarthritis, and cancer. The prevention and treatment of obesity is, therefore, a leading challenge facing public health and medicine in the 21st century.Two stereotypes have dominated thinking in public health, medicine, and the media about obesity. The first stereotype is that the recent surge in prevalence of obesity reflects almost entirely environmental and psychological factors and excludes an important contribution of genetic biological factors. The second stereotype is that obesity should and can be treated primarily by diet and behavioral modification. In this review, I challenge these tenets.I summarize evidence for a strong genetic neurobiological contribution to adiposity and body weight and assert that common human obesity is, like essential hypertension, a complex multifactorial disease where genetic factors promote sensitivity or resistance to obesity in a toxic environment. This concept of a genetic resistance versus sensitivity to obesity helps explain why many people remain thin in a toxic environment whereas others develop profound obesity.I then discuss evidence that dietary therapy for obesity generally fails to achieve weight loss maintenance. There is mounting indication that the high rate of relapse from weight loss during dietary therapy occurs because of compensatory biological adaptations that promote lack of compliance and effectiveness. Relapse from weight loss during dietary therapy is not caused simply by lack of discipline and will power.Finally, I briefly discuss the alternatives to dietary and behavioral therapy, namely bariatric surgery and pharmacotherapy.As a prelude to my critique of dietary therapy, I begin with a discussion of the role of genetic neurobiological factors in obesity.Contributors to the Increase in Obesity: The Role of Genetic Neurobiological FactorsThe surge in the prevalence of obesity in recent decades, during a time when the gene pool has not changed, has led to the view that environmental changes are the overwhelming contributors to the so-called obesity epidemic. The "Big Two" environmental factors, are said to be: (1) Unending overnutrition related to ubiquitous abundance of low cost/high calorie foods and (2) increasingly sedentary occupations and immobilizing technologies including computers, automobiles, television, and elevators that decrease caloric expenditure.1The emphasis on a "toxic environment" in the obesity epidemic has overshadowed evidence for a strong genetic neurobiological contribution to adiposity and body mass in humans. Evidence is mounting that these two factors—a toxic environment and genetic influences—are not mutually exclusive contributors to obesity. Instead, genetic factors can promote either sensitivity or resistance to obesity in a toxic environment.The Discovery of Leptin and the Revolution in the Genetic Neurobiological Regulation of Appetite and MetabolismThe pioneering discovery of leptin in 1994 by Jeffrey Friedman and colleagues revolutionized understanding of neurobiological regulation of appetite and metabolism.2 In 1973, Coleman conducted studies using parabiosis of obese (ob/ob) mice with diabetes (db/db) or wild-type (+/+) mice.3 He concluded that the ob/ob mice are unable to produce a satiety factor to regulate their food consumption whereas the db/db mice produce a satiety factor, but cannot respond to it. This study remained mysterious for 21 years until Friedman and colleagues discovered the ob or leptin gene and demonstrated that ob/ob mice which display increased food intake and severe early obesity have leptin deficiency from mutations in the leptin gene.2 Shortly thereafter, it was discovered that db/db mice which also display increased food intake and severe obesity have leptin resistance from mutations in the leptin receptor.4 Leptin decreases food intake and adiposity and increases energy expenditure in the ob/ob but not the db/db mice.5 These seminal discoveries demonstrated dramatically that appetite and metabolism are tightly controlled by genetic neuroendocrine factors and that adipose tissue is "alive" and acts as an endocrine organ regulating appetite, metabolism, and adiposity. Since then, there has been mushrooming evidence for new genetic neurobiological pathways regulating appetite, metabolism, and adiposity. This evidence has derived largely from studies of mutant or genetically engineered mouse models, but there have also been impressive discoveries in humans.Monogenic Human ObesityThere are rare or infrequent Mendelian human counterparts of both the ob/ob mice with mutations in the leptin gene and leptin deficiency6 and the db/db mice with mutations in the leptin receptor and profound leptin resistance.7 Humans with these mutations have voracious appetite, profound early childhood obesity, and other phenotypic abnormalities.6–9 In the children with leptin deficiency, treatment with leptin rapidly and dramatically reduces food intake and leads to normal body weight.9A number of other genes have been linked to severe human obesity.10–12 Pathogenetic mutations in the melanocortin-4 receptor gene have been identified in 5% of children with severe early onset obesity,10 and 3% of children with severe early onset obesity have mutations in the leptin receptor.8 Thus, monogenic forms of human obesity may represent more than 8% of children with early severe obesity, but it is virtually certain that most human obesity is not monogenic. Instead, as with human essential hypertension, most human obesity is a complex multifactorial polygenic disease involving the interaction of susceptibility genes/alleles and environmental stress.Adoption and Twin StudiesStudies of adopted children and twins support a contribution of genetic factors to regulation of adiposity and body mass under physiological conditions. Stunkard et al13 demonstrated that the body mass index (BMI) of adopted children correlates with that of their biological parents (with whom they do not share a common environment) and not with that of their adopted parents (with whom they do share a common environment). In addition, studies of monozygotic twins14,15 reveal a high intrapair correlation in BMI with a lower intrapair correlation in dizygotic twins (Figure 1). There are other remarkable features of adiposity and body mass in monozygotic twins. The intrapair correlation in BMI pertains whether the twins are reared apart or together.16,17 The intrapair concordance persists over many years.15 With controlled overfeeding, there is intrapair concordance not only in weight gain but in the pattern of regional fat distribution.18 These studies demonstrate a strong genetic contribution to the regulation of adiposity and body mass in humans. Indeed, the estimates of hereditability of body mass are as high or higher than those for height and arterial pressure where a large genetic contribution is widely recognized. Body mass is a highly hereditable human trait.12Download figureDownload PowerPointFigure 1. Photographs from Borjeson14 that portray the strong intrapair correlation in body mass in monozygotic or identical twins and a much lower intrapair correlation in dizygotic twins. The intrapair correlation in body mass in monozygotic twins pertains whether the twins are raised apart or together,16,17 and it persists over years15 and in response to controlled overfeeding.18 These twin studies indicate a strong genetic influence on body mass and adiposity in humans.The Concept of Genetic Sensitivity Versus Resistance to Obesity in a Toxic EnvironmentDespite this evidence from adopted children and twins, many would still ask "how can genetic neurobiological factors contribute to the surge in common human obesity in recent decades since the gene pool has not changed during this time?" This question prompts a discussion of (1) susceptibility genes/alleles; (2) the concept of obesity sensitivity and resistance in animals and humans; and (3) the interaction of genetic and environmental factors in complex multifactorial diseases.In the absence of environmental stress, susceptibility alleles (as opposed to disease-causing mutations) in genes will not produce a disease phenotype (Figure 2). It is only in the presence of a "toxic" environment that genetic susceptibility will be expressed as the disease phenotype, eg, hypertension or obesity (Figure 2). Thus, the finding that a disease such as obesity increases in frequency and severity in the presence of a "toxic" environment does not exclude an important contribution of genetic factors to the disease. The contribution of susceptibility genes/alleles to obesity has been shown in humans and experimental animals. Download figureDownload PowerPointFigure 2. A schematic diagram illustrating the interaction of environmental and genetic variance in complex multifactorial polygenic diseases involving the interaction of susceptibility genes/alleles and toxic environmental factors. Animals and humans with susceptibility genes that render them prone to a multifactorial disease will not manifest the disease phenotype, such as obesity, in a healthy environment. It is only in the presence of a toxic environment that genetic susceptibility will be expressed as the disease phenotype. In a toxic environment, the disease phenotype will emerge in the genetically prone animals or humans, but not in the genetically resistant population. An increase in the frequency or prevalence of a disease phenotype in a toxic environment does not exclude a strong genetic influence on the emergence of the phenotype.There are both rat and mouse models that display striking genetic sensitivity or resistance to obesity when fed high-fat diet. Two rat models merit emphasis—obesity prone and resistant Sprague-Dawley rats developed and characterized extensively by Levin et al19–21 and obesity prone and resistant Wistar rats developed by Hill et al22,23 and recently studied extensively by MacLean et al.24,25In humans, both epidemiological and experimental studies support the concept that there are individuals with striking differences in sensitivity versus resistance to obesity. Morbid obesity is increasing at a much faster rate than moderate obesity in the United States.26 There has been a disproportionate increase in the number of severely obese people in recent years. In contrast, the BMI of adults in the lowest percentiles has not changed nearly as much as the BMI of those in the highest percentiles.27 These observations should be obvious to anyone who has walked the streets of the USA in the past 25 years. These epidemiological data strongly suggest that there are humans who are susceptible to obesity in a toxic environment and others who are relatively resistant.There is also experimental evidence for the concept of obesity sensitivity and resistance in humans. In 16 nonobese young adults overfed by 1000 kcal per day for 8 weeks, Levine et al28 observed a 10-fold difference in weight gain ranging from 0.36 kg to 4.23 kg (Figure 3). In earlier studies of 12 pairs of identical twins, Bouchard et al18 observed that the variability in weight gain during controlled overfeeding was strongly influenced by genetic factors. Weight gain among the 24 individuals overfed by 1000 kcal per day for 12 weeks ranged from 4.3 to 13.3 kg. The strongest predictor of weight gain, particularly in visceral fat, with controlled overfeeding was the amount gained by the subject's identical twin. These studies indicate that the concept of sensitivity versus resistance to obesity in a toxic environment has a substantial genetic underpinning. Download figureDownload PowerPointFigure 3. A summary of experimental evidence from a study by Levine et al28 in which 16 nonobese volunteers who had controlled overfeeding for 8 weeks displayed a 10 variability in weight gain. This finding supports the concept of obesity sensitivity and resistance during overfeeding in humans. A study of controlled overfeeding in identical twins by Bouchard et al18 indicated that the strongest predictor of weight gain with overfeeding was the amount gained by the subject's identical twin. This study suggests that genetic factors are major determinants of obesity sensitivity or resistance.18In summary, adoption and twin studies provide evidence for a substantial genetic contribution to body mass and adiposity in humans. Mendelian monogenic forms of early childhood obesity provide incontrovertible evidence for powerful genetic neurobiological influences on appetite and adiposity in humans. Finally, there are both epidemiological and experimental data supporting the concept of genetic susceptibility or resistance to diet-induced weight gain in humans. This concept of obesity sensitivity and resistance may explain why some individuals develop profound obesity in a toxic environment whereas others do not. As Bray has said: "genes load the gun and a permissive or toxic environment pulls the trigger."29The Concept of Gestational and Perinatal Metabolic Imprinting of Offspring and Epigenetic Mechanisms in ObesityIn addition to susceptibility genes/alleles influencing sensitivity or resistance to obesity, there is evidence for a maternal gestational and perinatal "metabolic imprinting" of the offspring.30,31 In humans, this concept has emerged primarily from epidemiological studies,31 but it has received strong support from experimental studies in rodents.30,32 Maternal obesity and high-fat diet during gestation and lactation promote obesity and insulin resistance in offspring of rats.30A Notable Feature of This Metabolic Imprinting in Rats Is Its Relationship to the Genotype of the MotherIn rats selectively bred to be either genetically prone or resistant to the development of diet-induced obesity, maternal obesity during gestation leads to increased obesity (metabolic imprinting) only in the offspring of the mothers with a genetic predisposition to obesity.30 Thus, the interaction of genetic sensitivity and a toxic environment in the mother during gestation can stamp the offspring with a lifelong predisposition to obesity.Epigenetic mechanisms, ie, hereditary changes in gene function/expression that occur without changes in DNA sequence, may be involved in metabolic imprinting.32,33 These epigenetic changes in gene expression may be triggered by environmental influences, such as maternal nutrition during gestation. The practical implication is that the resulting changes in gene function and susceptibility to obesity may persist throughout life in the offspring.In summary, maternal obesity and high-fat diet during gestation and lactation can create a long-term sensitivity to obesity in the offspring. This metabolic imprint is apparently linked to an interaction between the mother's genetic predisposition to obesity and her metabolic environment during gestation and lactation. The gestational influences on long term susceptibility to obesity may be transmitted in part by epigenetic mechanisms. This maternal metabolic imprinting of the offspring introduces another source of biological predisposition to obesity in a toxic environment. Thus, both susceptibility genes/alleles and metabolic imprinting (epigenetic effects) can create a predisposition or sensitivity to obesity that is enduring and beyond purely cognitive and psychological control.Dietary Therapy for Obesity: An Emperor With No ClothesA second stereotype is that obesity should and can be treated primarily by dietary and behavioral therapy. What has led to this stereotype? Until recently, we have lacked understanding of the biology of obesity. This led to the view that the surge in common human obesity is solely a behavioral and environmental problem and should, therefore, be treated by behavioral and dietary therapy.The antidote to this stereotype is relapse from weight loss during dietary therapy and its biological basis. Over 5 decades, it has been demonstrated repeatedly that dietary therapy fails to achieve weight loss maintenance.34–36 Stunkard commented on this in 1958.34 Kramer et al confirmed it in 1989.35 Rosenbaum et al stated in 1997 that "approximately two thirds of persons who lose weight will regain it within one year, and almost all persons who lose weight will regain it within five years."36 One might argue that weight loss programs have improved and that these reports do not reflect the success of the best current programs. Unfortunately, this is not true. In a meta-analysis reported last year, Dansinger et al37 found that dietary counseling produced only modest weight loss and that even this diminished over time. Wilfey et al recently performed a randomized controlled trial to determine whether "maintenance interventions" augmented the effects of initial weight loss programs in obese children.38 One of two maintenance programs improved short-term efficacy of weight loss treatment in the children, but these effects waned over the next 1 to 2 years. In an editorial commentary, Rhodes and Ludwig39 commented: "… the data provided a sobering message—despite a statistically significant effect" in maintaining weight loss "… the effects of all interventions diminished over time and even at their peak were small." In addition, Tsai and Wadden40 recently evaluated the effectiveness of commercial behavioral weight loss programs. These programs include trained staff, low-calorie diets, physical activity programs, support groups, and counseling on behavioral modification. Despite these inducements to weight loss, these programs produced meager maintenance of weight loss. Many would suggest that the answer lies in more research to improve compliance and effectiveness of dietary and behavioral therapy. After decades of trying, and given the biological mechanisms that promote loss of compliance with dietary and behavioral therapy, I doubt that this will succeed.Let me acknowledge—a small percentage of people who lose weight by diet and behavioral modification maintain their weight loss. In 1993, Wing, Hill and colleagues41,42 started the National Weight Control Registry (NWCR) to recruit and study a group of people who had succeeded at long-term weight loss maintenance, defined as weight loss of at least 30 pounds maintained for at least 1 year. In 2006, Hill41 reported that the 6000 individuals in the NWCR had maintained average weight loss of over 70 pounds for an average of almost 6 years. Among these subjects, 4 behaviors and strategies stood out: (1) eating a moderately low fat, high carbohydrate diet; (2) consistent self-monitoring of body weight, food intake, and physical activity; (3) eating breakfast every day; and (4) very high levels of physical activity equivalent to 60 minutes/d of moderate intensity physical activity.What can we conclude from the study of people who have successfully maintained weight loss? They are a small minority of people who have lost weight. They are highly self-motivated. The primary strategy of more than 90% of these people who are successful in maintaining weight loss is "a high level of physical activity", ie, moderate intensity physical activity for about 60 minutes/d.41Can we produce and maintain such robust increases in physical activity in people who are not self-motivated? Tate, Wing, et al43 conducted a randomized clinical trial comparing standard behavioral therapy (SBT) with high levels of physical activity plus SBT. For 18 months, individuals assigned to high physical activity had an intensive support program. At the end of the 18 months, those assigned to high physical activity plus SBT were indeed performing higher physical activity, but, 1 year after the support program ended, physical activity had declined and there was no difference in activity and weight loss between the 2 groups. This study demonstrates that in most people there is relapse from adherence to high physical activity just as there is relapse from dietary restriction.Mechanisms of Relapse From Weight Loss: PsychologicalThere are 2 possible explanations for failure to maintain weight loss during dietary and behavioral therapy. The first is behavioral or psychological, and the second is biological. The psychological explanation holds that obese people lack the will power and discipline to eat less and exercise more. This view is convenient because it shifts the blame to the patient for our lack of effective treatment, but it lacks evidence. To the contrary, in 1961 Albert Stunkard, a psychiatrist who has made many scholarly contributions to understanding human obesity, and his colleagues examined the concept that obese humans have distinctive personality characteristics that predispose them to obesity. In a systematic comparison of obese and lean men reported, Stunkard and his colleagues reported that "no distinctive personality features were found."44 Nevertheless, obese people continue to be stigmatized as lacking the discipline to prevent and treat obesity. As Friedman stated: "In most circles, the conventional wisdom on obesity's cause has not changed appreciably from the time of Galen, who clearly held obese individuals responsible for their state… . Through the ages obese individuals have been held accountable for their condition."45Mechanisms for Relapse From Weight Loss: BiologicalAs a counterpoint to the simplistic notion that relapse from weight loss during dieting has a psychological explanation, there is mounting evidence that "voluntary efforts to reduce weight are resisted by potent compensatory biological responses."45In obesity prone rats, MacLean et al24,25 demonstrated that weight-reduced obese rats manifested a persistent decrease in energy expenditure and an increased drive to eat. As a result, when food became available, the rats displayed an exceptionally high rate of weight regain.24,25 Food restriction and weight loss are accompanied by decreases in plasma leptin (Figure 4). This results in decreases in catabolic hypothalamic proopiomelanocortin (POMC) and increases in anabolic hypothalamic neuropeptide Y (NPY)21 (Figure 4). These neurochemical changes stimulate food intake and minimize energy expenditure and, thereby, promote weight regain (Figures 4 and 5). As Levin has said, with food restriction and weight loss, "the drive to regain lies mainly in the brain."21Download figureDownload PowerPointFigure 4. A schematic depiction of changes in leptin and neuropeptides during food restriction.21 These compensatory adapations promote hunger and decrease metabolism. This undermines adherence to low calorie diets and promotes weight regain.Download figureDownload PowerPointFigure 5. In the midst of famine, the biological adapations to restricted food intake would protect against starvation and promote survival (A), but in contemporary environments with an abundance of food and increasingly sedentary lifestyles, these biological adaptations make it difficult to sustain weight loss during dietary therapy (B)These and other studies in experimental animals provide a contemporary framework for understanding observations made 4 decades ago in obese humans during diet-induced weight loss. Hirsch and colleagues46 studied responses to diet-induced weight loss in obese humans in the 1960s. Hirsch later summarized his studies: "my colleagues and I found that patients in their obese state had no specific constellation of behavioral abnormalities or psychological aberrations. Nothing convincingly implicated any psychiatric or psychological disturbance as the cause of their obesity. After these same patients lost weight, however, they manifested many behavioral and physiological alterations. They developed a marked preoccupation with food and dieting, and their physical and mental activity generally slowed down."47Subsequently, Leibel, Rosenbaum, and Hirsch48 demonstrated that weight loss and weight maintenance secondary to a decrease in caloric intake in humans are accompanied by a compensatory decrease in total energy expenditure. This mirrors the changes in energy expenditure in obese rats undergoing dietary restriction and weight loss. This decrease in energy expenditure would protect against starvation during famine, but in an environment prone to obesity it makes maintenance of weight loss difficult (Figure 5).Dietary adherence declines dramatically within 4 to 12 months and is another major cause of relapse from weight loss during dieting.49 Despite claims to the contrary, this occurs with all diet programs including Atkins, Weight Watchers, and Ornish.49 Not surprisingly, this decrease in adherence has been attributed to a lack of will power and discipline and blamed on the patients. But there is now incontrovertible evidence for compensatory biological adaptations that increase appetite and undermine adherence to weight reducing diets (Figure 4).Food restriction and weight loss are accompanied by decreases in leptin (Figure 4). Leptin exerts a potent influence on appetite. Children with congenital leptin deficiency have aggressive food-seeking behavior.6 Before leptin treatment, a young child with leptin deficiency consumed in excess of 1100 calories at a single meal, which is approximately half the average daily intake of an adult. With only a few injections of leptin, this was reduced by 84% to 180 calories a day, the typical intake of a normal child.27 Ghrelin, a gastrointestinal hormone that stimulates appetite, might also be involved in loss of dietary adherence during dieting.50 Ghrelin normally rises before meals and falls after every meal.50 It is increased during diet-induced weight loss, so that with weight-reducing diets, there is more ghrelin stimulating appetite.50 Thus, there is mounting evidence that appetite is a tightly regulated biological variable. With dieting and weight loss, there are homeostatic adaptations that stimulate appetite and decrease dietary adherence (Figure 4).Schwartz et al51 have reviewed evidence that the homeostatic control system of anabolic (NPY and agouti related peptide) and catabolic (proopiomelanocortin and cocaine-amphetamine-regulated transcript) neuronal pathways regulated by leptin are inherently more vigorous in defending against weight loss than against weight gain.In addition, there are nonhomeostatic (reward, cognitive, emotional) cortico-limbic controls that promote ingestive behavior.52 In a world of plentiful food, this nonhomeostatic control system has assumed increasing importance in regulating ingestive behavior and may override satiety.52 As Berthoud reports, it is simplistic to believe that these nonhomeostatic mechanisms are purely cognitive and can be easily controlled by discipline or will power.52 These nonhomeostatic control pathways are under profound neurobiological control.The Paradox of the Limitations and Dominance of Dietary TherapyMany leaders in this field have commented on the high rate of relapse from adherence to dietary and behavioral treatment of obesity and its biological basis.21,27,29,34,36,47 In an era when we pride ourselves on practicing evidence-based medicine, why then does dietary and behavioral therapy still reign if it is "an emperor with no clothes"? I propose 4 reasons: (1) It is highly profitable for industry and academia; (2) It puts the responsibility for failure on the patient and not the physician; (3) We have not had effective and safe drug treatment for obesity; and (4) In the midst of an "epidemic", public health officials and physicians are loathe to acknowledge that we do not have effective prevention.If Not Dietary and Behavioral Therapy, Then What?If dietary and behavioral therapy is as I suggest "an emperor with no clothes", then what are the alternatives? It is not my purpose to provide an extensive review of bariatric surgery and drug therapy, but some comments seem necessary.Bariatric SurgeryBariatric surgery produces substantial and sustained weight loss in patients with morbid obesity.53–58 Other treatments do not.53–56 The Swedish Obesity Subjects (SOS) study is the largest prospective (albeit nonrandomized) trial of bariatric surgery.55 In 1703 patients followed for 10 years after surgery in the SOS study, there was considerable benefit in lifestyle, diabetes, and dyslipidemia.55Dixon et al56 recently evaluated the effects of bariatric surgery (laparoscopic adjustable gastric banding or LAGB) on glycemic control and need for diabetes medications in 60 obese (BMI 30 to 40), type 2 diabetic patients randomized to either surgery or conventional diabetes, dietary and behavioral therapy. In patients followed for 2 years, the surgical group achieved greater weight loss, glycemic control, and remission of diabetes. One potential caveat of the general applicability of this study is that the magnitude of weight loss with LAGB (20.7% of body weight) was greater than that observed by other investigators.59Surprisi
Referência(s)