Jonathan Q. Purnell MD1
1Codirector, Metabolic Clinic, and Assistant Professor, Department of Medicine, Oregon Health and Science University
The author has served as a consultant for Amylin Pharmaceuticals, Inc.
The drug bupropion, which is discussed in this chapter, has not been approved by the FDA for use in obesity.
Obesity and its associated disorders are leading causes of morbidity and premature mortality around the world. Obese persons are also vulnerable to low self-esteem and depression because of the psychological and social stigmata that can be associated with obesity. Despite societal prejudicial perceptions that obesity develops because of deficient self-control, research has provided insight into the physiology behind unwanted weight gain. Indeed, during the past decade, the field of body-weight regulation (the study of the homeostatic mechanisms controlling body weight and fat content and the pathophysiology leading to unwanted weight gain or weight loss) has undergone an explosion in research, particularly in the area of neuroendocrine control of appetite and energy expenditure. As with other leading diseases in developed countries, such as hypertension and diabetes, obesity is recognized as a chronic condition resulting from an interaction between environmental influences and an individual's genetic predisposition to weight gain.
The initial evaluation of overweight and obese patients begins with the exclusion of secondary causes of weight gain and the identification of comorbid disorders such as hypertension, diabetes, heart disease, and sleep apnea. Once screening is completed, the approach to the treatment of overweight and obesity is similar to that of other chronic diseases: begin with lifestyle improvements, and then consider medical and surgical options. Although the weight loss that accompanies current therapeutic options is modest on average, the future promises better diagnostic and treatment options for obesity that are based on research into the mechanisms of weight regulation and their role in unwanted weight gain and maintenance of the obese state.
Definition of Obesity
Obesity is an abnormal accumulation of body fat in proportion to body size. Overweight persons have a body-fat proportion that is intermediate between normal and obese. Ideally, an obesity classification system would be based on a practical measurement of body fat that could be performed in the office, would accurately predict disease risk, and would apply to patients from diverse ethnic backgrounds. The most direct measures of body fat, such as underwater weighing or dual-energy x-ray absorptiometry (DXA) scanning, are impractical for use in a clinical setting. Indirect estimates of body fat are clinically more practical.
Classification of Obesity
BODY MASS INDEX
Body mass index (BMI), which is calculated by dividing the body weight in kilograms by height in meters squared, is a classification system that attempts to allow comparison of weights independent of stature across populations. Except in persons who have increased lean weight as a result of intense exercise (e.g., bodybuilders), BMI does correlate with percentage of body fat, but this relationship is independently influenced by sex, age, and race.1 In the United States, data from the second National Health and Nutrition Examination Survey (NHANES II) were used to define obesity in adults as a BMI of 27.3 kg/m2 or more for women and a BMI of 27.8 kg/m2 or more for men.2 These definitions were based on the gender-specific 85th-percentile values of BMI for persons 20 to 29 years of age. In 1998, however, the National Institutes of Health (NIH) Expert Panel on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults adopted the World Health Organization (WHO) classification for overweight and obesity.3 The WHO classification, which predominantly applies to people of European ancestry, assigns an increasing risk for comorbid conditions—including hypertension, type 2 diabetes mellitus, and cardiovascular disease—to persons with higher BMIs [see Table 1] relative to persons of normal weight (i.e., those with a BMI between 18.5 kg/m2 and 25 kg/m2). Asian populations, however, are known to be at increased risk for diabetes and hypertension at lower BMI ranges than those for non-Asian groups.4 Consequently, the WHO has suggested lower cutoff points for consideration of therapeutic intervention in Asians: a BMI of 18.5 to 23 kg/m2 represents acceptable risk, 23 to 27.5 kg/m2 represents increased risk, and 27.5 kg/m2 or higher represents high risk.5
Table 1 Classification of Weight and Risk for Comorbid Conditions2
In addition to an increase in total body fat, a proportionally greater amount of fat in the abdomen or trunk compared with fat in the lower extremities or hips has been associated with increased risk for diabetes, hypertension, and heart disease in both men and women.6Abdominal obesity is commonly reported as a waist-to-hip ratio, but it is most easily quantified by a single circumferential measurement obtained at the level of the superior iliac crest.3 Current guidelines categorize men at increased relative risk for coronary artery disease, diabetes, and hypertension if they have a waist circumference greater than 40 inches (102 cm); women are at increased risk if their waist circumference exceeds 35 inches (88 cm) [see Table 1]. Thus, an overweight person with abnormal fat patterning may be at high risk for these diseases even if that person is not obese by BMI criteria. In those of Asian descent, abdominal (central) obesity is recognized to be a better predictor of comorbidity than BMI.7 Therefore, the WHO has recommended lower waist circumference cutoffs to assign increased risk for comorbidities in this population: 36 inches (90 cm) or more in men and 32 inches (80 cm) or more in women.4
In the United States, the prevalence of overweight has been increasing over the past several decades [see Figure 1]. In the most recently published United States data (1999 to 2002), 65% of adults are overweight (BMI 25 to 30 kg/m2), 30% of the total population are obese (BMI 30 to 40 kg/m2), and 5% have a BMI of 40 kg/m2 or higher.8,9 The prevalence of obesity has also risen in some minority populations, with the highest rates found in some Native American groups, Hispanics, and African Americans; the lowest rates have been found in populations of Asian ancestry [see Figure 2].9,10,11,12 Prevalence rates for obesity in the United States are also highest in populations with less education and lower income levels.12 Internationally, obesity rates are generally lower than those in the United States.13 However, even in societies that traditionally had the lowest prevalence of overweight and obesity, the rates of weight gain are beginning to meet or exceed those of Western societies.14
Figure 1. Time trends of age-adjusted prevalence of overweight (BMI ≥ 25 kg/m2) and obesity (BMI ≥ 30 kg/m2) in United States men and women who are 20 years of age and older.8
Figure 2. Age-adjusted percentage of United States adults who were overweight (BMI ≥ 25 kg/m2), by sex and race/ethnicity, 1999–2002.8 Data for Asians include both men and women in 2003.11
The age at which obesity is most prevalent has also increased. Until NHANES III (1988 through 1994), obesity in the United States peaked between the ages of 40 and 59 years, then declined in the older-age groups.9 According to the most recent NHANES data (1999 to 2002), the prevalence of obesity now remains high past the age of 60 years, reaching 30.5% in men and 34.7% in women.9 In studies that have measured body composition in unselected populations, fat mass also peaks just past middle age in men and women, but percent body fat continues to increase past this age, particularly in men, because of a proportionally greater loss in lean mass.15,16,17 The menopausal period has also been associated with an increase in percent body fat and propensity for central fat distribution, even though total body weight changes very little during this time.18 A propensity for greater abdominal adiposity has also been demonstrated in men,19 in older individuals,20 and in persons with impaired glucose tolerance or type 2 diabetes.21
Etiology and Genetics
Studies of populations, families, adoptions, and twins have established a strong genetic role in determining body weight. Estimates of the genetic contribution to the variance of relative body weight and adiposity range from a low of approximately 20% to a high of 90%.22 The largest study to date to address the contribution of nature versus nurture to body weight, which used a dataset that included over 25,000 twin pairs and 50,000 biologic and adoptive family members, found that genetic factors accounted for 67% of the variance in adiposity in men and women.22 Rarely, childhood-onset obesity will manifest itself as a result of a single-gene obesity syndrome, such as Prader-Willi syndrome or Bartlett-Biedel syndrome, or from a mutation in one of the genes encoding proteins involved with body-weight regulation, such as the pro-opiomelanocortin (POMC) gene, which makes α-melanocortin-stimulating hormone (α-MSH); the melanocortin receptor (MC4); leptin; the leptin receptor; and prohormone convertase enzymes.23
Although the genetics explaining the tendency toward overweight and obesity in the majority of the population remains to be elucidated, over 600 genetic markers have been described in association with obesity-related variables in humans (e.g., BMI, skin-fold thickness, waist-to-hip ratio, fat mass, and percent fat mass).23 With time, discoveries of specific gene products, the role that these proteins play in the pathophysiology of weight regulation, and their interaction with the environment in the expression of unwanted weight gain should lead to more specific pharmacologic treatments for overweight and obese patients who fail to respond adequately to lifestyle measures alone.
Epidemiologic studies have identified several environmental factors that contribute to the continued weight gain documented over the past several decades in westernized countries. The foremost among these factors are an increasingly sedentary lifestyle (e.g., increased car use, community and work environments that discourage activity, and more time spent watching television) and the availability of energy-dense (high-fat, concentrated-sugar), low-fiber foods.24,25,26,27,28 In children, the increased consumption of sugar-added beverages and reduction of dairy intake have also been associated with greater weight gain in prospective studies.29,30 Similar environmental predictors of weight gain have been described in societies adopting Western lifestyles in the transition to First World economies.14,31 Additional societal trends that are thought to have contributed to the increasing weight gain in the United States include smoking cessation (cigarette smoking is known to reduce body weight)25,32 and eating a greater proportion of food away from home, particularly at fast-food restaurants, where food is typically very calorically dense.28,33
Pathophysiology and Pathogenesis
Arguably the most significant recent advances in the science of obesity have been in the area of neuroendocrine control of energy homeostasis, including the understanding of the mechanisms that lead to unwanted weight gain and the counterregulatory systems that restore weight lost after caloric restriction. At its most basic level, body weight is the end result of a balance between energy taken in and energy expended. Weight gain ensues when more energy is consumed than expended. Weight loss occurs through restriction of energy intake, increased energy output, or both. However, this simple model fails to incorporate what are now known to be complex homeostatic systems that counteract voluntary energy perturbations, whether they be forced overfeeding or caloric restriction.
A homeostatic model of weight regulation is conceptually identical to other tightly regulated systems in the body. For example, blood glucose levels reflect input from meals and hepatic stores balanced against clearance through uptake by peripheral tissues and excretion by the kidneys. Glucose levels are kept within a normal range by complex, integrated responses from insulin, glucagon, and other so-called counterregulatory hormones such as catecholamines, cortisol, and growth hormone, which regulate production and clearance. Elevated blood glucose levels and diabetes result when the secretion of primary regulators (i.e., insulin and glucagon) is impaired, when resistance to insulin signaling develops, or both.
Like glucose, body weight is regulated at multiple levels to maintain a normal range or set point through an interaction between systems that control meal-to-meal intake (satiety) and those that control relative fat mass (adiposity) [see Figure 3].34 Although short-term (meal-to-meal) signals such as cholecystokinin have been studied for decades, a long-term afferent signal from the fat tissue, leptin, was not discovered until 1994.35 Leptin is a hormone that is secreted by fat cells in direct proportion to total fat mass, is transported across the blood-brain barrier, and has receptors in hypothalamic nuclei that control appetite and energy expenditure.34 When leptin levels decline with weight loss from caloric restriction or when they increase with overfeeding, altered signaling in central hypothalamic centers become integrated with other input signals (e.g., insulin and ghrelin) to set in motion systems that restore body weight to baseline. Therefore, most obese patients fail to sustain long-term weight loss with calorie restriction alone because of activation of these counterregulatory systems and their promotion of a positive energy balance. In this feedback-loop model of weight regulation, primary obesity results when leptin signaling to central centers is reduced (leptin resistance),34 resulting in uncompensated weight gain that eventually reestablishes energy homeostasis at a higher body-weight set point and blood level of leptin, analogous to insulin levels rising in compensation for acquired insulin resistance.
Figure 3. (a) A feedback model for body-weight regulation in humans based on data from animal models.34 Hypothalamic centers that control long-term energy homeostasis sense fat stores through circulating levels of leptin and insulin. Satiety signals (short-term or meal-to-meal regulators) from the gut are relayed through the brain stem to the hypothalamus, where they are integrated with signals reflecting fat stores. These integrated signals then affect appetite and energy expenditure so as to maintain body weight within a set point range. (b) Leptin controls appetite and energy expenditure in the hypothalamus by alternatively stimulating production of pro-opiomelanocortin (POMC) and α-melanocortin (α-MSH) and inhibiting production of neuropeptide-Y (NPY) and agouti-related protein (AGRP). α-MSH binds to the melanocortin-4 receptor (MC4), which inhibits appetite and increases energy expenditure. NPY and AGRP stimulate appetite while decreasing energy expenditure. Reduced leptin secretion, such as that which occurs after voluntary caloric restriction,59,146leads to enhanced NPY/AGRP signaling, diminished MC4 signaling, and positive energy balance once caloric restriction ceases.147,148 On the other hand, overfeeding leads to increased fat mass and leptin secretion,149 reduced NPY/AGRP signaling, increased MC4 signaling, and negative energy balance148,150 until body weight is restored to baseline. (CCK—cholecystokinin; GLP-1—glucagonlike peptide-1; PYY—peptide YY)
Although this model oversimplifies the complex nature of body-weight regulation, it nonetheless provides a starting point for clinicians in their education of patients about the pathophysiology of obesity and in the rationale for medical management. To achieve sustained weight reduction in overweight and obese patients, interventions must prevent activation of counterregulatory systems that act to restore lost weight by increasing appetite or reducing energy expenditure. Future medical therapies will be based on an understanding of the body's weight regulatory system and will have greater promise for success in maintaining weight loss.
With aging, dysregulation of a number of hypothalamic-pituitary systems may contribute to increased fat mass and sarcopenia. For example, growth hormone secretion diminishes with age.36 Prospective trials in older adults that involved replacing growth hormone and targeting levels of insulinlike growth factor-1 (IGF-1) to the midnormal to upper-normal range have demonstrated improved body composition (less fat, more lean tissue) and, in some studies, reduced central fat.36,37,38 In addition, the decline in testosterone levels in men, the drop in estrogen levels in women at menopause, and increased levels of cortisol in both sexes may also contribute to reduced muscle mass, central fat distribution, or both.18,39,40,41
The history, physical examination, and laboratory evaluation of overweight and obese patients are directed toward three goals: first, to identify secondary causes of obesity [see Differential Diagnosis, below]; second, to identify comorbid conditions [see Figure 4]; and third, to establish the patient's dietary and activity habits.
Figure 4. Evaluation, laboratory testing, and treatment of overweight and obese patients.
A number of the symptoms associated with diseases that can cause or contribute to unwanted weight gain, such as hypothyroidism or Cushing disease, occur frequently in overweight patients. These include fatigue, aches, cold intolerance, constipation, poor exercise tolerance, central obesity, loss of libido, and depression. Deciding when to screen a patient for secondary causes of obesity, therefore, can be a challenge for the practitioner. Establishing a pattern of weight gain may be helpful. A patient with a lifelong history of being heavy and a stable adult weight is unlikely to have a secondary cause of obesity. A sudden or rapid weight gain over a few months or years, however, especially when accompanied by onset of comorbid conditions, may correspond to the prescription of medications that contribute to excess weight gain (especially steroids and newer antipsychotics) or indicate onset of an illness that requires further evaluation.
The history should include questions about diseases for which overweight and obese patients are at higher risk, including hypertension, impaired glucose tolerance or diabetes, hyperlipidemia, heart disease, pulmonary disease, and sleep apnea. These conditions may cause minimal or no symptoms and therefore may be present for months or years before a diagnosis is made. Sleep apnea in particular is a common cause of fatigue and poor concentration or work performance in obese patients; these symptoms are often mistakenly ascribed to an abnormally functioning thyroid gland (despite normal results on thyroid function tests) or a so-called altered metabolism. This diagnosis may be missed unless the clinician specifically asks about characteristic symptoms: restless sleep at night, snoring or observed apnea, fatigue or headache upon awakening and during the daytime, and spontaneous daytime sleep when inactive or while driving. In severely obese patients, increasing peripheral edema, orthopnea, and worsening exercise tolerance may be symptoms of congestive heart failure or pulmonary hypertension and right-sided heart failure from severe sleep apnea. New-onset headaches may indicate normal-pressure hydrocephalus. Gastroesophageal reflux disease usually results in heartburn or an acid taste in the throat. During a period of weight gain, women may develop irregular periods or symptoms of androgen excess. Although commonly diagnosed as polycystic ovary syndrome (PCOS), these findings differ from classic PCOS in that they occur after menarche and are not usually associated with polycystic ovaries.
Finally, inquiring about past and present dietary and activity habits is important for subsequent discussions of medical and surgical management. Most overweight and obese patients will have made numerous attempts to lose weight, through diets, exercise regimens, or commercial weight-loss programs. Because of unrealistic expectations and the inevitable weight regain that occurs, patients are often discouraged or leery of new advice. These failures can also compound feelings of guilt or inadequacy, fueling cycles of worsening self-image and depression. A broad survey of the types of foods people eat can often be accomplished in an office visit; in particular, asking about intake of calorically dense foods, including sodas, and frequency of meals outside the home may identify habits that can be improved. More detailed dietary analysis requires a visit with a nutritionist. Physical impediments such as arthritis, back pain, and asthma should be identified and treated so as to optimize daily activity and adherence to exercise recommendations.
PHYSICAL EXAMINATION AND LABORATORY TESTS
Height and weight measurements in the office are used to classify patients as overweight or obese according to BMI criteria [see Table 1]; however, these criteria may not apply to patients who have gained weight as the result of increased muscle mass from intensive exercise. Evaluation of abdominal obesity requires the use of a tape measure. A waist circumference (obtained at the level of the superior iliac crest) greater than 40 inches (102 cm) in a man or greater than 35 inches (88 cm) in a woman is considered abnormal.
Specific physical findings that might indicate secondary causes of obesity include pretibial edema and delayed tendon reflexes (hypothyroidism), purple striae, supraclavicular fat pad enlargement, and muscle weakness (Cushing syndrome). Other aspects of the clinical evaluation focus on comorbid conditions. Documentation of hypertension requires properly obtained blood pressure measurements (i.e., using the correctly sized cuff for larger persons). Insulin resistance and type 2 diabetes may manifest themselves as acanthosis nigricans—patches of feathery-pigmented skin (hyperkeratotic and hyperpigmented) on the extensor surfaces of the hands and elbows, in the axilla, or on the neck [see 2:I Cutaneous Manifestations of Systemic Diseases]. Hepatomegaly can indicate hepatosteatosis, especially in centrally obese subjects.
With or without acanthosis nigricans, impaired glucose tolerance may be diagnosed by a fasting plasma glucose level between 100 and 125 mg/dl or a 2-hour glucose level between 140 and 200 mg/dl during an oral glucose tolerance test. Type 2 diabetes is diagnosed by two fasting blood glucose measurements of 126 mg/dl or greater, a 2-hour glucose level of 200 mg/dl or more during an oral glucose tolerance test, or a random glucose level of 200 mg/dl or greater and symptoms of diabetes.
Screening for macrovascular risk involves obtaining an electrocardiogram when appropriate and carefully examining the patient for xanthomata, which can indicate the presence of elevated blood levels of chylomicrons (eruptive xanthoma), type III hyperlipidemia (palmar xanthoma or tuberoeruptive xanthoma), or familial hypercholesterolemia (tendon xanthoma). Each of these physical manifestations of hyperlipidemia, although rare in a primary care practice, indicates a severe or potentially life-threatening condition that requires urgent diagnosis and treatment. A fasting lipid profile should be obtained to complete the cardiovascular risk assessment, and if necessary, treatment should be instituted according to guidelines from the National Cholesterol Education Program Expert Panel.42 This panel also incorporated several nonlipid risk factors for cardiovascular disease into its recommendations for clinical care by defining criteria for a condition that has become known as the metabolic syndrome (also called syndrome X, the deadly quartet, and the insulin-resistance syndrome). The metabolic syndrome includes the most common abnormalities of lipid and glucose metabolism that accompany abdominal obesity [see Table 2]. Identifying these abnormalities in a patient allows the practitioner to better assign that patient's risk for diabetes and coronary artery disease.43 Similar criteria are now recognized by the Centers for Disease Control and Prevention (CDC) as the dysmetabolic syndrome X and have been assigned a diagnosis code (277.7) in the International Classification of Diseases, 9th Revision (ICD-9). Screening laboratory tests for hepatosteotosis include a liver panel. In addition, all overweight patients should have documentation of normal thyroid function with a thyroid-stimulating hormone level.
Table 2 Criteria for Metabolic Syndrome42,152
Although obesity is associated with abnormal levels of a number of hormones and cytokines, including leptin, ghrelin, interleukins, and tumor necrosis factor, measurement of these variables should be limited to research protocols and are not currently recommended for general clinical practice.
It is important for clinicians to be alert for secondary medical causes of obesity but also to be aware that, in most cases, treatment of these coexisting diseases rarely leads to complete reversal of the obese state. As an example, hypothyroidism is relatively common in the general population and may be present in an obese patient, but the weight loss that might be expected with thyroid hormone replacement is limited and variable.
Hypercortisolemia of Cushing syndrome is a rare cause of unwanted weight gain, but clinicians should have a low threshold for screening for this disease when patients experience large amounts of weight gain in a short period, especially when the weight gain is accompanied by hypertension, diabetes, or muscle weakness. Deficiencies of growth hormone or gonadal steroids are also associated with modest increases in body adiposity. Growth hormone deficiency can lead to reduced muscle mass and increased fat mass, which is improved with hormone replacement therapy.44 Similar changes in body composition have been described in hypogonadal men45 and in postmenopausal women.18Unfortunately, obesity is often accompanied by low levels of IGF-1 and, in men, low testosterone levels because of low sex-hormone-binding globulin levels. To distinguish obesity-associated low testosterone from a true deficiency state, free testosterone levels can be measured. In addition, weight loss will increase both IGF-1 and total testosterone levels in obese patients but not in patients with true deficiencies.
A number of medications can lead to unwanted weight gain and obesity; if possible, such patients should be switched to alternative agents [see Table 3]. Drug-related weight gain occurs most commonly during long-term glucocorticoid treatment of inflammatory conditions (e.g., asthma and inflammatory arthritis), with immunosuppression after transplantation, and with cancer chemotherapy. When possible, reducing or discontinuing a glucocorticoid in favor of an alternative medication can reverse this weight gain. Patients with type 1 or type 2 diabetes often gain weight after starting therapy; this weight gain is proportional to the degree of improved glycemic control and results from a reduction in glucosuria and improvement in metabolic efficiency.46 Long-term studies have shown that intensive insulin treatment of type 1 diabetes can result in excessive weight gain and obesity in up to 25% of patients; in type 2 diabetes, intensive glycemic control with insulin, a sulfonylurea, or one of the thiazolidinediones may also result in greater weight gain than that predicted by improved glycemic control alone.47 Therapy with metformin plus nighttime long-acting insulin may reduce or prevent this extra weight gain,48 and newer diabetes medications, such as pramlintide and exenatide, can improve glycemic control in both type 1 and type 2 diabetes with a modest weight loss.47 Neuropsychotropic drugs, particularly newer antipsychotic and antiseizure medications, have been associated with weight gain (sometimes massive), obesity, and diabetes.49,50
Table 3 Medications Commonly Associated with Weight Gain and Obesity, with Possible Alternative Agents47,49,152
As a first step in the management of obesity, appropriate follow-up testing and treatment should be provided for any secondary causes of obesity and comorbid conditions identified during screening. Then, the approach to the treatment of obesity is similar to that of other chronic conditions, such as hypertension, hypercholesterolemia, and diabetes. Intervention starts with lifestyle measures for 3 to 6 months. For obesity, these lifestyle interventions include improved diet and increased activity. For patients whose weight does not change with lifestyle intervention alone or whose weight loss is insufficient to lower their long-term health risk, consideration is then given to pharmacologic or surgical management. An NIH expert panel has suggested that patients whose BMI is 30 or more or who have a BMI of 27 or more plus obesity-related risk factors (i.e., diabetes, hypertension, or hyperlipidemia) could be considered for pharmacologic therapy.3Patients with a BMI of 40 or more or a BMI of 35 or more plus obesity-related risk factors could be considered for surgical therapy.
The weight-loss goal for the treatment of obesity is sustained weight loss of 5% or more of initial body weight. Although this goal does not result in attainment of a normal body weight (BMI of 19 to 25) in the majority of patients, it still represents a weight loss that can be achieved with available intervention modalities and that has been associated with lower morbidity, including reductions in risk for diabetes and heart disease.51,52
For some patients who are experiencing a period of weight gain, weight stability may be their primary goal. This is especially common in patients who have just completed a low-calorie weight-loss program and are struggling to remain below their initial weight.
NONMEDICAL (LIFESTYLE) THERAPY
Hypocaloric diets have been a mainstay recommendation by the medical community for obese patients. These diets range from a moderate reduction in daily intake (200 to 500 fewer calories a day) to more stringent, very low calorie diets (600 to 800 total calories a day), which require careful follow-up by a nutritionist and a physician to avoid life-threatening electrolyte disorders and symptomatic cholelithiasis. Although it is possible to achieve short-term weight loss with these strategies, long-term weight loss is poor even when behavior-modification weight-maintenance programs are continued. Analyses of published data on long-term weight-loss maintenance showed that approximately 50% of the initial lost body weight is regained within the first 1 to 2 years, and 95% or more is regained by 5 years after the completion of the calorie-restriction phase.53,54 This restoration of lost body weight after a period of calorie-restriction-induced weight loss can be explained by the reduction of fat-dependent feedback signals to the brain, such as leptin, which then activate counterregulatory systems to restore body weight to baseline [see Pathophysiology and Pathogenesis, above].
The long-term failure to maintain weight loss after caloric restriction also indicates that the central set point for body weight is not reset at a lower body weight with the passage of time. Another important implication of these data is that obesity treatments lacking mechanisms that interfere with this counterregulatory system will likely fail to allow long-term weight-loss maintenance. Some existing therapies do result in limited weight loss without activation of appetite (see below) and can be combined with caloric restriction for improved long-term weight-loss maintenance; these include a low-fat diet,55,56 exercise,56 and pharmacologic treatments.57,58
An increase in dietary-fat intake leads to obesity in animal studies and has been associated with a higher prevalence of overweight and obesity in many human-population studies.25 Prospectively randomizing overweight and obese persons to ad libitum feeding (eating until one feels full, then stopping) of a fat-restricted diet results in a spontaneous reduction in caloric intake and subsequent modest weight loss, compared with results in persons on a diet that contains a higher amount of fat typical of Western societies.59 This calorie reduction occurs despite a concomitant fall in leptin levels, in contrast to the increase in appetite that follows a fall in leptin levels with weight loss from caloric restriction.59,60,61 By implication, an increase in dietary fat results in a state of central leptin resistance, requiring a higher level of body fat and leptin levels to attain a new body-weight equilibrium,62 whereas dietary-fat restriction leads to partial improvement in leptin signaling, resulting in spontaneous reduction in appetite and body weight.
The average amount of weight loss attributable to a low-fat diet in these studies, however, is only on the order of 3 to 4 kg (6.6 to 8.8 lb).59 In addition, the weight-loss responses of persons to a low-fat diet can vary tremendously, with some individuals losing 13 kg (28.7 lb) or more and others losing no weight or even gaining weight.63 This variable response to a lifestyle intervention, such as a change in a specific diet component, is common in chronic diseases whose expression results from an interaction between a genetic predisposition and environmental influence. In patients with hypertension, for example, blood pressure reductions in response to restriction of dietary salt are also heterogeneous.
An apparent paradox has been reported in that the average percent fat content of the American diet is dropping, yet the weight of the American population keeps increasing. Although it is true that the average percentage of total calories from fat in the American diet has declined over the past several decades (from 36% to 34%, according to the most recent NHANES data),64 this did not occur because Americans have been eating less fat (daily dietary fat intake was 81.9 g in 1972 and increased to 85.5 g in 1990) but, rather, because total calories increased, leading to a lower fat percentage.65 In contrast, studies that documented weight loss with a lower fat intake did so by lowering the absolute amount of fat in the diet. It is worth noting that the levels of fat restriction leading to weight loss in these studies were not severe. Severe fat restriction (< 20% of total calories) may not be sustainable for many patients because of limited food options and palatability.
Increasing dietary-carbohydrate intake while lowering total fat intake results in modest spontaneous reduction in caloric intake and weight loss in overweight and obese persons. In the studies that documented this effect, the additional carbohydrates were derived from fruits, vegetables, and grain products, and the resulting increase in dietary fiber also may have played a role in greater satiety and weight loss. In society (especially in young people), however, dietary carbohydrates have increasingly been consumed in the form of processed foods sweetened with sucrose or fructose. These simple carbohydrates (especially fructose) may potentially have deleterious effects on insulin resistance, lipid levels, and body weight when consumed in large amounts.29,66
Paradoxically, severe carbohydrate restriction (< 30 g/day) may also lead to modest spontaneous weight loss without initial activation of appetite. Such severe carbohydrate restriction initially mobilizes glycogen stores in the liver and induces ketogenesis, and the resulting diuresis accounts for some of this weight loss.67 At one time, the ability to draw meaningful conclusions about the longer-term safety and efficacy of low-carbohydrate diets was impeded by the paucity of controlled studies and the variability of carbohydrate restriction from study to study (from < 20 to ≥ 200 g/day).68 Although methodological issues remain, randomized, controlled studies have now shown that during the first 6 months of diet treatment, persons placed on a low-carbohydrate diet lose weight more rapidly than those placed on a low-calorie, low-fat diet.69,70,71 Subsequently, however, individuals on the low-carbohydrate diet either stop losing weight or regain weight, and by 1 year, weight loss is the same with the two diets.70,71,72 The average 1-year weight loss ranged from 2.5 to 5.1 kg in these studies, and both diets had a high dropout rate, of approximately 30% to 40%.70,71,72 Contrary to popular beliefs about the potential adverse effects of consuming diets high in fat or carbohydrates, lipid levels and glucose metabolism improved with both diets in proportion to weight loss.69,70,71,72,73
Increasing dietary-protein intake has also been associated with weight loss. In one of the few prospective, randomized studies of an ad libitum high-protein diet (fat restricted), obese patients experienced significantly greater weight loss than obese control subjects who followed a regular diet or a low-fat, high-carbohydrate diet over a period of 6 months.74 After 12 months, however, total weight loss on the high-protein diet was attenuated and no longer differed from that seen with the high-carbohydrate diet.74
Most nutrition societies recommend limiting protein intake to approximately 10% to 15 % of daily calories because of concerns regarding long-term health consequences of high intake of protein (especially animal protein). These concerns include the possible association of increased protein intake with intestinal cancers, bone disease, and renal disease. To date, prospective studies have shown that increasing dietary protein increases the glomerular filtration rate,75 which may be harmful to patients with existing renal disease or diabetes, but the long-term effect of increased glomerular filtration rate in otherwise healthy persons is not known.
Increased dietary fiber has been shown to improve body weight and cardiovascular risk factors.76,77 Typical high-fiber foods include fruits, vegetables, oat and wheat bran, and legumes, which are also low in fat. Even after controlling for low-fat content, however, diets higher in fiber result in reduced intake and a weight loss of approximately 2 kg.76 Although it is possible to increase fiber through the use of supplements such as psyllium or methylcellulose, the current intake of fiber in the United States of about 15 g/day could be increased to 25 to 30 g/day by avoiding calorically dense, refined-sugar foods and increasing consumption of fruits, vegetables, and whole-grain products.
Summary of dietary recommendations
Overall, an initial recommendation to lower dietary-fat intake and increase dietary-fiber intake for weight loss is reasonable and supported by the scientific literature. Long-term studies of greater than 1 year have not been conducted to show that this weight loss is sustained. Nevertheless, animal models of obesity, population studies, and prospective studies of 1 year or less of low-fat, high-fiber diets versus high-fat diets have documented that a high-fat diet is detrimental to body weight and that restriction of dietary fat to 25% to 30% of calories and an increase of 10 to 15 g of fiber a day result in a significant, albeit limited, weight loss for the average patient.
It is important that clinicians discourage unrealistic expectations about weight loss from a low-fat, high-fiber diet so that patients do not become disillusioned with this therapy. Also, patients should be informed that low-fat, high-fiber diets have been shown to reduce numerous health risks, especially when instituted as part of an overall lifestyle change that includes exercise.50,78 For this dietary advice to be effective, however, it is often necessary to refer patients to a nutritionist for evaluation and follow-up.
A low-fat diet can be achieved by substituting either carbohydrate or protein for fat; this allows tailoring of the diet to the individual patient. Some patients may respond better to a high-carbohydrate diet in terms of food preferences and weight loss, whereas others might have better responses to a high-protein diet, although all these diet variations have shown only moderate weight loss.72 A low-carbohydrate diet cannot currently be recommended for clinical practice, because it has not been shown to be superior to other diets70,72; the long-term health outcomes of sustained ketosis are uncertain; and increased intake of saturated fat and trans-fatty acids may negate the benefit of weight loss by increasing serum cholesterol and triglyceride levels in some patients.72,73 Many questions concerning the effects of a higher protein intake (up to 30% of total calories) on patients' renal function remain unanswered. For this reason, high-protein diets should be avoided in patients with existing renal disease and diabetes.
Increasing energy expenditure through exercise has been another mainstay of obesity therapy. Without counterregulation (alteration in appetite or non-exercise-based energy expenditure), an increase in activity should lead to continued and sustained weight loss. Prospective intervention studies have shown, however, that the average amount of weight loss attributable to exercise alone (no caloric restriction in addition to the exercise) is small, ranging from 1 to 4 kg (2.2 to 8.8 lb).79,80 Further weight loss presumably is limited by alterations in nonexercise energy expenditure that compensate for the increase in exercise-induced energy expenditure; increased appetite can be discounted as the source of counterregulation, because most studies show little change in energy intake.81
Current recommendations are to participate in 3 to 5 hours of moderate to vigorous activity per week. For many patients, especially those with limited mobility, simply increasing activity may be an initial goal. As with alterations in the macronutrient content of the diet, exercise leads to variable degrees of weight loss, ranging from little or none to a substantial amount. Even with moderate weight loss, which is typical, patients should be encouraged to continue with increased activity because of the numerous health benefits attributable to being fit.82,83
Combined Diet and Exercise
A number of studies have prospectively examined the effect of combined diet and exercise interventions on health outcomes. Sustained weight loss with caloric restriction can be improved with a low-fat diet84,85 or regular exercise.79,86 In a national survey seeking to determine the characteristics of persons who have been able to sustain a weight loss of at least 13.6 kg (30 lb) for 1 year, responders reported that on average, they followed a diet consisting of approximately 24% fat87 and expended an average of 2,827 kcal/wk in exercise (roughly the equivalent of walking 28 miles/wk).88 This level of activity is nearly three times more than the 1,000 kcal/wk recommended by the American College of Sports Medicine for the minimum weekly exercise for the purposes of reducing body weight.88 Prospective studies of dietary-fat restriction and exercise have also demonstrated reduced progression, or even reversal, of heart disease in patients with known cardiovascular disease78 and up to a 58% reduction in the incidence of type 2 diabetes in overweight patients with impaired glucose tolerance.50
Because of the overall health benefits, the initial treatment of overweight and obese patients should include increased activity in addition to dietary-fat restriction and increased dietary fiber. If caloric restriction is recommended, the likelihood of long-term weight-loss maintenance may be improved by the addition of these interventions. Dietary and activity advice should be implemented using individualized, sustainable behavioral and life style changes. Again, it is important for both clinicians and patients not to have unreasonable expectations for weight loss from diet and exercise. For some patients, the best that may be achieved through lifestyle improvements is prevention of further weight gain. Patients who fail to meet unreasonable goals may become frustrated and return to a less healthy lifestyle. Even when little or no weight loss ensues, these lifestyle changes offer many health benefits, including improved lipid levels, increased insulin sensitivity, and reduced risk for progression of cardiovascular disease and for onset of type 2 diabetes.89 After institution of lifestyle measures, if a patient's body weight remains above the guideline cutoff points, medical and surgical treatment options remain.90
Once patients have become overweight or obese, lifestyle interventions play an important role in reducing comorbidities, but they typically have modest effects on body weight (see above). These patients will then require lifelong medical or surgical management to achieve more meaningful weight loss (see below). This failure of the body-weight set point to remain at a lower level in the average patient greatly increases the importance of prevention of unwanted weight gain. Prevention requires targeted improvements in food choices for pregnant mothers and children and, more importantly, a reversal of the societal trend toward reduced activity levels.
The hurdles for implementing these simple recommendations are, however, considerable. Increasing intake of healthy foods includes not only overcoming personal and cultural preferences for higher-fat foods but also consideration of the economic costs of food at home and at schools (calorically dense foods are often cheaper than fruits and vegetables) and the impact of family and work demands on ability to eat at home or purchase prepared foods.91 Increasing activity does not necessarily require going to a gym on a regular basis. Rather, it includes maintaining physical activity in routine daily events, such as physical education in schools, designing work spaces and buildings to promote walking and stair use, planning urban environments to promote more pedestrian activity (e.g., safer streets, more sidewalks, increased density of commercial and residential properties), and overcoming a cultural reliance on automobiles for routine travel.92 Even with immediate institution of lifestyle improvements, though, it is likely that reversing the current decades-long trend in weight gain will take several generations.
The Food and Drug Administration has approved several prescription medications for the treatment of obesity [see Table 4].90 These medications fall into two categories: centrally acting drugs, which suppress appetite, and peripherally acting drugs, which reduce fat absorption. For example, phentermine and sibutramine act centrally, reducing appetite by promoting the release of norepinephrine from presynaptic terminals (phentermine) and inhibiting the uptake of both norepinephrine and serotonin (sibutramine) in central nuclei. Orlistat acts peripherally, inhibiting the action of lipases in the brush border of the intestine and thereby reducing lipid absorption.
Table 4 Pharmacologic Agents Approved by the Food and Drug Administration for the Treatment of Obesity
Clinicians who treat obese patients with medical therapy should keep in mind four important principles. First, most studies have included an initial treatment phase in which patients are placed on a hypocaloric diet (usually, daily caloric intake is reduced by 500 to 1,000 kcal) at the time of randomization or just before drug treatment. Second, the weight loss with obesity agents varies considerably: some patients lose a dramatic amount of weight; others lose only a little weight; and still others lose some weight only to regain it despite continuation of the medication. In any case, average weight loss with currently approved medications does not usually exceed 10% of the baseline weight. Third, weight loss is greatest during the first 3 to 6 months, followed by a plateau at a new lower weight even with continuation of the therapy. Intermittent therapy (i.e., 3 months on therapy followed by 3 months off, 3 months on, etc.) does not increase weight loss93,94 and has been associated with increased side effects (e.g., dry mouth).93 Finally, every drug-treatment study that has included posttreatment follow-up has shown rapid weight regain toward baseline after discontinuance of the medication. The plateauing of weight after initial weight loss and the regaining of weight after medication discontinuance have been interpreted to indicate that the medication became ineffective with time or failed because weight loss was neither sufficient nor sustained after cessation of the treatment. However, pharmacologic therapy for obesity is no different from therapy for other common chronic diseases. For instance, patients with hypertension often have a variable response to a first-line agent. In some cases, blood pressure lowering may be insufficient and may require the addition of a second medication, and hypertension returns when the medication is discontinued. Once a person begins receiving medical therapy for obesity, continued efficacy is evidenced by sustained weight-loss maintenance. Treatment should therefore be continued indefinitely unless the weight is regained or significant side effects develop.
Agents Approved for Short-Term Use
Benzphetamine, phendimetrazine, diethylpropion, mazindol, and phentermine are approved by the FDA for the short-term treatment of obesity (weeks). All but phentermine are rarely, if ever, used in clinical practice today. Benzphetamine and phendimetrazine are both Drug Enforcement Administration (DEA) schedule III drugs, with higher abuse potential than the others. Diethylpropion and mazindol have indications and side effects identical to those of phentermine (all are DEA schedule IV), but they are less well studied and have not had the same acceptance in clinical practice as phentermine; consequently, diethylpropion and mazindol are not extensively discussed in this chapter.
Phentermine inhibits appetite and causes an average weight loss of 8.7 kg (19.2 lb) (net weight loss of 5.1 kg [11.2 lb] when compared with placebo).95 The agent is available as a 30 mg resin and in 15 mg and 37.5 mg tablets. Doses in excess of 37.5 mg are not recommended because of unacceptable side effects.
Little information is available about subsequent improvement in health outcomes with phentermine treatment alone, because studies of this drug have historically been short term, and the drug was frequently used in combination with fenfluramine. A group of postmenopausal women treated with phentermine and a low-calorie diet experienced a 14% weight loss, along with reduction in low-density lipoprotein (LDL) cholesterol and triglyceride levels and an increase in high-density lipoprotein (HDL) cholesterol levels over 9 months.96 In patients with diabetes, despite a net loss of 3.8 kg (8.4 lb) compared with placebo over 6 months, use of phentermine produced no improvement in glycemic control or glycosuria and no significant reduction in hypoglycemic drug use.97
As a result of central nervous system activation by phentermine, patients may experience anxiousness, insomnia, palpitations, and dry mouth. In case reports, phentermine treatment has been associated with vasospasm, psychosis, and ischemic events,98,99,100 although in a larger cohort study, phentermine was not associated with stroke.101
Agents Approved for Long-Term Use
The FDA approved sibutramine for the medical treatment of obesity in 1997. The recommended duration for therapy was 1 year, but treatment may be continued beyond that time if no significant side effects occur and sustained weight loss is documented.
The average weight loss on the highest currently approved dose, 15 mg, is 7.0 kg (15.4 lb) (net weight loss of 5.7 kg [12.6 lb] when compared with placebo).102 In a 2-year study of sibutramine therapy, a weight loss of 10.2 kg (22.5 lb) (net weight loss of 5.5 kg [12.1 lb] when compared with placebo) persisted for up to 18 months, at which time a slight upward trend became evident.57 With placebo, in contrast, patients began to regain weight immediately after completion of the hypocaloric phase, which was at 6 months. After 2 years, significantly greater proportions of patients taking sibutramine, compared with control subjects, had maintained 5% and 10% weight loss. These results may have limited applicability, however, because only persons who completed the initial 6-month weight-loss phase and lost at least 5% of their initial body weight by caloric restriction were subsequently randomized to placebo or continued sibutramine therapy. This represents a 23% dropout rate before randomization to drug therapy. In addition, approximately 30% of the patients treated with sibutramine and 50% of the patients given placebo withdrew from the study. A higher pretreatment body weight103 and weight loss of at least 1.8 kg (4 lb) during the first month of therapy102 have been shown to predict continued weight-loss response with longer treatment.
Sibutramine treatment is also associated with improvements in lipid levels, including lower triglyceride levels and higher HDL levels.57,94,104Sibutramine has been safely used in patients with type 2 diabetes, with improvements in lipid levels being similar to those reported in nondiabetic patients.105 Glycemic control also improves in diabetic patients on sibutramine therapy, with the greatest reductions in hemoglobin A1c occurring in those patients who lose more than 10% of their initial body weight.105 Common side effects of sibutramine treatment include dry mouth, constipation, insomnia, palpitations, and headache. Sibutramine consistently raises average blood pressures slightly and pulse rate more so, even with weight loss.57,102 In populations not selected for hypertension, studies have documented increases in average diastolic pressure of 0 to 3.4 mm Hg, increases in systolic blood pressure of 0 to 2.7 mm Hg, and increases of average pulse rate of 4.1 to 6 beats/min.57,94,102 Similar results with sibutramine have been shown for hypertensive patients receiving a variety of antihypertensive medications.106,107 Therefore, sibutramine treatment can be used if blood pressure is controlled, but blood pressure and pulse rate should be monitored routinely.108 However, because of the potential for added cardiovascular demand, sibutramine should not be used in patients with a diagnosis of cardiovascular disease, heart failure, arrhythmia, or stroke. To date, sibutramine has not been associated with either valvular disease or pulmonary hypertension.
Orlistat inhibits lipases in the gastrointestinal lumen, thereby antagonizing triglyceride hydrolysis and reducing fat absorption by roughly 30%. Because orlistat is not absorbed to any significant extent, its primary mechanism of action is thought to be through providing what is in effect a low-fat diet, thereby promoting lower caloric intake and weight loss [see Dietary-Fat Restriction, above], as well as improved weight-loss maintenance, when combined with a low-calorie diet.109 As with sibutramine, orlistat treatment should be continued as long as the patient maintains weight loss and avoids significant side effects.
To minimize side effects related to fat malabsorption, candidates for orlistat treatment are first placed on a diet containing only 30% of calories from fat. When combined with a calorie-restricted diet, orlistat treatment (120 mg with each meal) results in an average weight loss of 7.2 to 13 kg (16 to 28.7 lb) (net weight loss of 1.3 to 5.6 kg [2.9 to 12.3 lb] when compared with placebo). After 1 year, patients taking orlistat maintain greater weight loss than those taking placebo [see Table 4].58,110,111 During continued follow-up for another year, regaining of weight is seen but remains less than that in patients given placebo.57,110,111
Patients who lose weight with orlistat also experience a significant reduction in levels of total and LDL cholesterol (approximately 4% to 8%, which is significantly lower than reductions in patients given placebo).110,111 Levels of triglycerides and HDL cholesterol are either reduced or left unchanged; these results are not different from those in the placebo group. Blood pressure and insulin levels also decrease with weight loss in patients on orlistat therapy.58,110,111 Similar improvements in lipid levels, along with improvements in glycemic control, have been documented in obese persons with type 2 diabetes who lose weight with orlistat.112
Gastrointestinal side effects may occur in up to 80% of patients when they begin therapy with orlistat (such side effects are also seen in 50% to 60% of patients given placebo), but this incidence diminishes with time. Symptoms include abdominal discomfort, flatus, fecal urgency, oily spotting, and fecal incontinence. When administered to patients who adhere to a low-fat diet, orlistat is generally well tolerated. The fat malabsorption that accompanies orlistat treatment can also lead to reductions in fat-soluble vitamins, but in prospective studies, the average levels for vitamins A, D, E, and β-carotene remained in the normal range.58,110,113 In one study, only 2.4% of orlistat-treated patients had documented below-normal levels of β-carotene; only 3.1%, below-normal levels of vitamin D; and only 1.6%, below-normal levels of vitamin E.110 Nevertheless, orlistat should not be given to patients with existing malabsorptive states, and it is recommended that patients take a daily multivitamin supplement during therapy.
Non-FDA-Approved Medical Therapy for Obesity
Observational data from studies of patients with depression suggested that treatment with selective serotonin reuptake inhibitors (SSRIs) may result in weight loss, but this effect is slight and short-lived. Moreover, in prospective studies, SSRI treatment is sometimes associated with mild weight gain.49 Bupropion is currently approved as an antidepressant and for smoking cessation. The mechanism of action of bupropion is not precisely known, but it includes weak inhibition of norepinephrine and dopamine reuptake. Combining bupropion (300 to 400 mg/day) with moderate daily caloric restriction has been shown to result in greater weight loss than placebo (net weight loss of 3 to 4 kg [6.6 to 8.8 lb], on average) after 6 months in obese patients with no symptoms or mild symptoms of depression.114,115,116 Bupropion is contraindicated in patients with seizures, anorexia nervosa, and bulimia.
Deficiencies of growth hormone and sex steroids have been associated with higher body-fat content and greater central obesity. Replacing growth hormone reduces fat mass, increases lean mass, and reduces central obesity. Similar improvements in body composition have been reported with testosterone replacement in hypogonadal men. Although obese persons have documented abnormalities in the hypothalamic-pituitary-gonadal and somatotropic axes, treatment of overweight or centrally obese persons with growth hormone, sex steroids, or both has not been shown to produce clear improvements in body composition, fat distribution, or comorbid conditions.
With improved safety from technical advances and demonstrated efficacy, bariatric (obesity) operations clearly have a role in the current management of severely obese patients.90 The mechanisms whereby these operations result in sustained weight loss are poorly understood, but they likely include alterations in the gut-derived hormonal and neural inputs to the central nervous system.117 In general, these procedures can be classified into one of three types: restriction of food passage, malabsorption of nutrients, or a combination of the two [see Table 5]. As an example of a purely restrictive procedure, vertical-banded gastroplasty involves the formation of a small stoma for the passage of food. The rationale behind this procedure is to increase a sense of fullness and reduce food intake. Procedures resulting in malabsorption typically involve bypassing sections of small or large intestine. Early procedures in which large sections, or even the entire length, of the small intestine were bypassed often resulted in severe malabsorption and sometimes hepatic failure and death. Subsequent modifications have led to bypass of shorter sections and lower morbidity. Patients must understand that bypass surgery is anatomically irreversible in most cases and has a potentially high postoperative complication rate.
Table 5 Most Commonly Used Bariatric Procedures118,153,154
Several large-scale studies have demonstrated the efficacy of various bariatric procedures in severely obese patients. One study of gastric bypass reported sustained loss of approximately 50% of baseline body weight (at 5- to 10-year follow-up).118 In the ongoing Swedish Obesity Subjects (SOS) Intervention Study of 1,157 severely obese persons (average BMI of 42 kg/m2), weight loss in the first postoperative year ranged from 21% after gastric banding to 38% after gastric bypass.119 Some weight regain was found on 10-year follow-up, but the gastric banding patients remained 13% below and the gastric bypass patients 25% below their initial body weight.119
A more recent technique known as laparoscopic gastric banding, in which a restrictive band is placed around the upper stomach, has also shown efficacy in sustained weight loss in studies up to 4 years. The weight loss with laparoscopic gastric banding is generally felt to be similar to that with vertical-banded gastroplasty but not as great as that with gastric bypass.120
An important benefit of bariatric surgery is that the large weight losses bring improvements in the comorbid conditions that accompany obesity,121,122 including lowering of lipid levels and blood pressure and reduced rates of diabetes, progression to diabetes, and sleep apnea.119,122 Moreover, cohort and population-based studies have demonstrated that weight loss after bariatric surgery is associated with both reduced mortality and reduced use of health care resources.123
Complication rates for bariatric surgery will vary by surgeon, site, and technique. The most common operative-related complications, occurring in up to 30% of patients, are wound infections, atelectasis or pneumonia, and hernia.118,124,125,126,127 Serious but rare complications include anastomotic stenosis or leakage, thromboembolism, and bowel obstruction. Perioperative death rates have ranged from 0.2% to 1.3%.118,125,126,127 The laparoscopic techniques offer greater safety than open procedures without sacrificing efficacy.128,129Nesidioblastosis has been described in a small series of obese patients after gastric bypass operations.130 These patients experienced serious or life-threatening postprandial hypoglycemic episodes. A number of questions remain, however, regarding the frequency of this finding, the validity of the link between the gastric bypass and beta cell hypertrophy, and how to best do a workup of patients who may have this condition. Nevertheless, clinicians should include nesidioblastosis in the differential diagnosis of gastric bypass patients who complain of symptoms that have heretofore been ascribed to the so-called dumping syndrome (i.e., rapid transit of food into the small intestine associated with rapid increases in blood volume to the gut and insulin secretion): weakness, dizziness, palpitations, and near-syncope.
Because of the potential for rapid weight loss after bariatric surgery, follow-up by a nutritionist and physician is important to ensure that patients preserve lean mass and maintain hydration and to monitor for symptomatic cholelithiasis. Lifelong alteration of dietary habits may be required to ensure the continued success of the surgery, and dietary supplements may be needed to prevent nutritional deficiencies [seeTable 5].
Surgical therapy for obesity results in significant, sustained weight loss in a majority of patients with severe obesity.121,122 With regard to total weight loss and improved disease outcomes, the gastric bypass procedure has the greatest support from published studies,121,122whereas the gastric banding procedure has the lowest published rates of morbidity (11%) and mortality (0.05%).120 With the considerable improvements in comorbidity and quality of life that result from this weight loss, bariatric surgery is not only efficacious but also cost-effective in the management of severe obesity.122,123,131
Complications of Obesity
Persons who are overweight or obese and have central adiposity are at increased risk for hyperlipidemia, hypertension, and cardiovascular disease mortality.6,132,133,134,135 In addition, obesity and central adiposity are both strong risk factors for the development of type 2 diabetes.11,136 Other diseases with a higher incidence in obese persons include gallstones, high uric acid levels and gout, hepatic steatosis, osteoarthritis, obesity hypoventilation, atrial fibrillation, nephrolithiasis, and certain cancers.137,138 Sleep apnea is likely underdiagnosed in overweight and obese patients139 and should be strongly considered in patients with complaints of fatigue, daytime somnolence, snoring, restless sleep, and morning headaches. People who are overweight and obese also carry a significant psychosocial burden. Obesity is accompanied by lower self-esteem in children140 and prospectively predicts depression in adults.141 Young adults who were overweight as adolescents are less likely to marry, will complete less schooling, and have lower incomes than nonoverweight peers.142 The economic costs of the treatment of obesity and its comorbidities are high: estimates from 1995 United States data put the costs at $99.2 billion, and of this, $52 billion, or 5.7% of the United States national health expenditure, went to pay direct medical costs.143
Mortality increases when men and women become overweight or obese, in large part because of comorbid conditions (see above). This impact of obesity on mortality is greatest in younger age groups and diminishes with age144,145 and may be significantly influenced by the patient's fitness status. Prospective observational studies suggest that mortality and risk associated with coronary artery disease are highest in the least fit and lowest in the most fit, independent of body weight.82,83 Evidence of lower mortality after specific lifestyle, medical, or surgical interventions for obesity is currently lacking. Instead, decreased severity of a number of diseases and cardiovascular risk factors associated with increased mortality have been demonstrated after even modest weight loss.89 Insulin resistance, hyperlipidemia, hypertension, sleep apnea, diabetes, and cardiovascular disease have all been shown to improve with weight loss that follows intensive lifestyle interventions (e.g., low-fat, high-fiber diet and exercise) and medical therapy.51,52,58,78,89,104,105 With the greater weight loss that accompanies bariatric surgery, resolution of diseases such as diabetes and sleep apnea119,123 and increased survival124 have been reported. Optimal therapy for obesity, however, remains elusive. Simple caloric restriction alone results in short-term weight loss (months to years), but without additional interventions such as a low-fat diet, exercise, pharmacologic therapy, or a combination of these, 95% or more of the weight initially lost will be regained within 5 years.54 Lifestyle interventions alone, including a low-fat, high fiber diet and increased activity, have only small effects on body weight but result in significant improvements in obesity-related morbidity.51 Current pharmacologic therapy is effective in achieving weight loss of up to 10% of initial body weight for at least up to 2 years.57,110 Bariatric surgery has been shown to increase weight loss to up to 50% of initial weight for at least 10 years,118,119 but it carries the risks of surgery and, with most operations, irreversible anatomic modification. Nonetheless, for patients with severe obesity who are at high risk for morbidity and mortality, bariatric surgery can offer hope for improved survival while more effective medical therapies are being developed.
Figure 3 Seward Hung.
Editors: Dale, David C.; Federman, Daniel D.