Obesity is one of the most common disorders in clinical practice. Defined (by the NIH) as a body weight 20% or more above "desirable" weight, over one third of adult Americans are overweight. Perched at the center of chronic disease risk and psychosocial disability for millions of Americans, successful management of obesity offers unique patient care and public health opportunities. If all Americans were to achieve a normal body weight, it has been estimated that there would be a 3 year increase in life expectancy, 25% less coronary heart disease, and 35% less congestive heart failure and stroke.

Unfortunately, obesity is also one of the most difficult and frustrating disorders to manage successfully. Considerable effort is expended by primary care providers and patients with little benefit. Using standard treatments in university settings, only 20% of patients lose 20 pounds at two year follow-up while only 5% of patients lose 40 pounds. This lack of clinical success has created a never-ending demand for new weight loss treatments. Approximately 50% of women and 25% of men are dieting at any one time. Americans spent over $30 billion last year on diet books, diet meals, weight-loss classes, diet drugs, exercise tapes, "fat farms," and other weight loss aids!

The challenge for health care providers is to identify those patients with obesity who are most likely to medically benefit from treatment and most likely to maintain weight loss, and to provide them with sound advice, skills for long-term lifestyle change and support. The purpose of this presentation is to review new developments in our understanding of obesity and its treatment to assist in this difficult but important task.

Definitions

Obesity is an excess of body fat; overweight is an excess of body weight, including all components of body composition (muscle, bone, water and fat). In clinical practice, the two are used interchangeably to refer to excess body fat. The two most commonly used terms to quantify obesity are relative weight (RW) and body mass index (BMI). The RW is the actual weight divided by the "desirable weight" (derived from "acceptable weight" tables). The BMI, or Quetelet index, is the actual body weight divided by the height squared (kg per m2). This index more closely corresponds to measurements of body fat and better differentiates "overweight" due to an increase in muscle mass from true obesity.

A recent National Institute of Health Consensus Conference defined obesity (somewhat arbitrarily) as a RW of greater than 120% (BMI > 27 kg/m2). "Morbid" obesity is commonly defined as a RW greater than 200% (BMI >40 kg/m2).

Health Consequences of Obesity

The relationship between body weight and mortality is curvilinear, similar to other cardiovascular risk factors. Most studies have demonstrated a J-shaped or U-shaped relationship, suggesting that the thinnest portion of the population also have an excess mortality. This is thought to be primarily due to the higher rate of cigarette smoking in the thinnest group.

The relationship of body weight to mortality is also effected by age. The body weights associated with the lowest mortality increase with age, and newer weight tables take this into account. In addition, as age increases to over 65, the relationship of body weight and mortality takes on a more striking U-shape. This suggests that although obesity remains an important risk factor in the elderly, undernutrition is also extremely important.

The increase in total mortality related to obesity results predominantly from coronary heart disease (CHD). Evidence is mixed whether obesity is an independent risk factor for coronary heart disease. For example the 1993 cholesterol treatment guidelines omit obesity an a risk factor for CHD, while the previous edition in 1988 included obesity. Nonetheless, obesity is clearly an important risk factor for the development of many other CHD risk factors. Obese individuals age 20-44, for example, have a 3.8 times greater risk of type II diabetes, 5.6 times greater risk of hypertension, and 2.1 times greater risk of hypercholesterolemia. As a result, type II diabetes and stroke also contribute to the increase in obesity-related mortality. The obese also have an increase risk of certain cancers including colon, rectum, and prostate in men and uterus, biliary tract, breast, and ovary in women.

As a result of these conditions, relative weights of 130% are associated with an excess mortality of 35%. Relative weights of 150% have a greater than two-fold excess death rate. Patients with "morbid" obesity (relative weights greater than 200%), have a greater than 10-fold increase in death rates.

Obesity is also associated with a variety of other medical disorders including degenerative joint disease of both weight bearing and non-weight bearing joints, diseases of the digestive tract (gallstones, reflux esophagitis), thromboembolic disorders, heart failure (both systolic and diastolic), respiratory impairment, and skin disorders . Obese patients also have a greater incidence of surgical and obstetric complications and are more prone to accidents. Although obesity is not associated with an increased risk of major psychiatric disorders, obese patients are at increased risk of psychological disorders and social discrimination.

Regional fat distribution: Recent investigations suggest that the location of the excess body fat (regional fat distribution) is a major determinant of the degree of excess morbidity and mortality due to obesity. At least three components of body fat are associated with obesity-related adverse health outcomes. These are the total amount of body fat (expressed as a percentage of body weight), the amount of subcutaneous truncal or abdominal fat (upper body fat), and the amount of visceral fat located in the abdominal cavity. These three are partly correlated with each other but exhibit a fairly high degree of independence. Each of these components of body fat is associated with varying degrees of metabolic abnormalities and independently predict adverse health outcomes. In each of 6 prospective epidemiologic studies, increased abdominal obesity was associated with increased cardiovascular and total mortality.

Body fat distribution can be assessed by a number of measurement techniques. Measurements of skinfolds (subscapular and triceps) reflect subcutaneous fat. Measurement of circumferences (waist and hip) reflect both abdominal and visceral fat. CT and MR scans measure subcutaneous and visceral fat. Clinically, measurement of the waist and hip circumference is most useful. The waist is measured at the umbilicus and the hips at the greater trochanter. A waist to hip ratio of 1.0 and 0.8 are considered normal in men and women, respectively. Ratios above these values reflect abdominal and/or visceral obesity and a greater risk of obesity-related disorders.

Etiology of Obesity

Numerous line of evidence suggest strong genetic influences on the development of obesity. Most convincing are genetic studies of adoptees and twins. In a study of 800 Danish adoptees, there was no relationship between the body weight of adoptees and their adopting parents but a close correlation with the body weights of their biological parents. In a study of approximately 4000 twins, a much closer correlation between body weights was found in monozygotic than dizygotic twins. In this study, genetic factors accounted for approximately two thirds of the variation in weights. More recent studies of twins reared apart and the response of twins to overfeeding showed similar results. Studies of regional fat distribution in twins has also shown a significant (but not complete) genetic influence.

The exact mechanisms by which such genetic factors result in obesity are considerably less clear. Differences in both energy intake and energy expenditure have been investigated. Genetic influences on control of appetite and eating behavior have long been considered. Animal studies have demonstrated the influence of dozens of factors on eating behavior, and it is likely that similar factors are at work in humans. Observational studies of eating behavior have suggested that the obese both eat more food and do so more rapidly than the non-obese.

Differences in energy expenditure are also likely to be at least partially determined by genetic influences. Differences in the resting metabolic expenditure (RME), for example, could easily result in considerable differences in body weight since RME accounts for approximately 60-75% of total energy expenditure. The RME can vary by as much as 20% between individuals of the same age, sex, and body build; such differences could account for approximately 400 kcals of energy expenditure per day! Recent evidence suggests that the metabolic rate is similar in family members, and as expected, individuals with lower metabolic rates are more likely to gain weight. Differences in the thermic effect of food, the amount of energy expended following a meal, may also contribute to obesity. Although some investigators have shown a decreased thermic effect of food in the obese, others have not.

Environmental factors are also clearly important in the development of obesity. Decreased physical activity and food choices that result in increased energy intake also clearly contribute to the development of obesity. Medical illness and some medications can also result in obesity, but such instances account for less than 1% of cases. Hypothyroidism and Cushing's syndrome are the most common. Diseases of the hypothalamus can also result in obesity but these are quite rare. Major depression, which usually results in weight loss, can occasionally present with weight gain. Consideration of these causes is particularly important when evaluating unexplained, recent weight gain.

Thus, the etiology of obesity is multifactorial, and almost certainly under both genetic and environmental influences.

Indications for Weight Loss (Patient Selection)

Weight loss is indicated to treat the above conditions in any patient who is obese (RW>120%, BMI>27 kg/m2). In fact weight loss is the first-line treatment for most of these disorders. In addition, many patients with relative weights of 100-120% (BMI 25-27 kg/m2) who have one of these conditions (particularly hypertension, diabetes, lipid disorders or significant psychosocial disability) will often dramatically benefit from weight reduction.

Weight loss to prevent complications of obesity in patients without active medical complications or metabolic consequences of obesity is more controversial. In young and middle aged individuals, particularly those with a family history of obesity-related disorders, treatment should be based on the degree of obesity (RW >120%) and body fat distribution. Such individuals with upper body obesity (waist-hip ratios >1.0 in men and >0.8 in women) should be considered for treatment; those individuals with lower body obesity, and no medical complication or metabolic consequences of obesity, should be reassured and followed for development of additional upper body obesity and its metabolic consequences.

A medical or psychosocial indication for weight loss is necessary but not sufficient, however, to begin treatment. Only motivated patients should be begun on a treatment regimen. Considerable effort should be made to assess the patient's motivation for significant diet and exercise changes. Questions should focus on how the current attempt compares to previous attempts, a realistic assessment of the patient's goals for amount and rate of weight loss, the extent to which outside stresses, emotional disorders, or substance abuse might impair the attempt, and the degree to which others can provide support. Patient motivation can be further assessed by requiring the patient to complete specific pre-treatment assignments. For example, patients can be asked to complete a 3-5 day diet record and to submit an exercise plan which includes both the type of aerobic exercise that the patient plans to begin and how the patient plans to fit it into their schedule. Diet Therapy

Standard dietary treatment of obesity should utilize the same nutritional principles as diet recommendations for healthy people. Total fat intake should be limited to 30% or less of total calories, protein 15%, and carbohydrate (primarily complex carbohydrates) 55% or greater. A total energy intake should be recommended to result in a daily energy deficit of 500-1000 calories. Since one pound of fat equals approximately 3500 calories, these deficits will result in a 1-2 pound weight loss per week.

The recommended calorie intake can be derived by multiplying the patient's "desirable" weight (the midpoint of the "acceptable weight" table) by 30 calories per kilogram, and then subtracting the desired energy deficit.

Example:

45 yo man, 5'10", weighs 200 pounds "desirable" weight = 160 pounds (73 kg) estimated energy requirement = 2190 kcals per day recommended energy intake to lose 2 lbs per week = 1190 kcals

No dietary manipulation of macronutrients or other nutritional components can change these basic thermodynamic concepts; yet, virtually all popular diets are based on attempts to circumvent thermodynamics.

It is impossible for the clinician to keep up with each new popular diet. Categorizing diets, however, can allow generalizations which allow for a basic understanding of the most common diets. Major categories of diets include low carbohydrate diets, vegetarian diets, single-food diets, high carbohydrate diets, very-low-calorie diets, and prepackaged diet programs.

Low carbohydrate diets

Low carbohydrate diets are the most resilient popular diet concept. Examples include the Atkins diet, the Stillman diet, and the Scarsdale diet. These diets are based on the correct observation that at equal calorie intakes, low carbohydrate diets result in more rapid weight loss than high carbohydrate diets. Unfortunately, the greater weight loss observed during low carbohydrate feeding is entirely due to changes in water balance. During carbohydrate restriction, ketonuria increases and results in greater sodium excretion and water loss. The resumption of carbohydrate feeding reverses this process and results in sodium and water retention. No other differences in body composition or weight loss are observed. Low carbohydrate diets are by definition high in fat and/or protein and are thus unsuitable for long term weight loss. These diets are also commonly deficient in calcium and dietary fiber.

Vegetarian diets

Vegetarian diets are typically low fat, high carbohydrate, high fiber diets and thus consistent with dietary goals. Vegetarian diets, however, can be either nutritionally adequate or inadequate depending on food selection. Diets which restrict all animal products (vegan) require particular attention to protein intake and vitamin B 12. Diets which include dairy products (lacto-vegetarian) or dairy and eggs (lacto-ovo vegetarian) are easier to plan. Most large weight loss programs include a vegetarian option

Single-food diets

Single-food diets are based on the concept that its not only what you eat but when you eat it that is important. The Beverly Hills diet, for example, suggests that by ingesting foods one at a time, digestion is made more efficient, resulting in fewer calories getting "stuck" and less weight gain. The diet relies primarily on fresh fruit and is inadequate in protein, niacin, calcium and iron. As expected, it commonly results in diarrhea.

High carbohydrate diets

High carbohydrate diets are typically balanced, hypocaloric diets. Most are also high fiber diets. Although the intake of large quantities of dietary fiber offers no particular weight loss benefit, these diets are consistent with current dietary recommendations and can be encouraged. Most large commercial diet programs such as Weight Watchers, TOPS, etc. use high carbohydrate diets and can be recommended to patients.

Very-low-calorie diets

A major development in the dietary treatment of obesity is the use of safe and effective very-low-calorie diets (VLCD's). Also known as protein-sparing modified fasts and protein-formula-liquid diets, these diets restrict calorie intake to 400-800 kcals per day. Patients ingest only preformulated, usually liquid, food that provides adequate protein, vitamins and minerals. Additional intake is limited to 2-3 quarts of calorie-free beverages per day. The major advantage of these diets is the "complete removal of patients from the food environment" to facilitate compliance. In addition, the significant energy deficit results in rapid weight loss, usually 3-4 pounds per week, encouraging the patient to continue. Major ongoing concerns about these diets include their cost, side effects and complications and long term results.

A recent concern, for example, has been the observation that significant numbers of patients using VLCD's develop gallstones. Studies using serial ultrasound examinations of the gallbladder before, during and after VLCD's have shown that approximately 25% of patients develop gallstones or "sludge." Approximately 25% of these become symptomatic and require surgery, approximately 50% resolve spontaneously, and approximately 25% remain asymptomatic. VLCD-related gallstones can be prevented with ursodeoxycholic acid and possibly by aspirin. Some authorities recommend the use of ursodeoxycholic acid (or aspirin) in individuals at particularly high risk of gallstone disease (eg those with a prior history).

Patients on VLCD's have marked improvements in obesity-related metabolic parameters. Within weeks blood sugars, blood lipids, and blood pressure are significantly reduced. Most diabetic, hypertensive and hypercholesterolemic patients can have medications markedly reduced or discontinued.

As with standard diet therapy of obesity, VLCD's require compliance during the diet and long term nutritional and behavioral changes to maintain weight loss. Well planned programs which combine VLCD's with behavior modification, exercise, and social support report improved long term results. For example, Nunn et. al. reported an average weight loss of 55 pounds with 75% and 52% of the loss maintained at one and 2 1/2 year follow-up respectively. Hartman et al reported maintenance of an average of 24 pounds after 2-3 year follow-up. The largest study, reviews the 18 month follow-up for over 4000 patients treated with Optifast. 25% of patients dropped out of the program early. Of those remaining in the program, 68% lost considerable weight but did not reach their goal. Of this group only 5-10% maintained weight loss after 18 months. Of the 32% of patients who attained their goal weight, 30% of women and 50% of men maintained their weight loss at 18 months. Thus, of the entire initial cohort, approximately 15% of patients met their weight loss goal (typically >40 pounds of weight loss) and maintained it at 1 1/2 years. Although these results are significantly better than results from most standard programs (in which approximately only 5% of patients lose and maintain 40 pounds at one year follow-up), the expense and risks of such programs continue to raise considerable controversy.

A recent trend in the use of liquid supplements, has been the use of similarly formulated 800 kcal/day diets. Most authorities feel that at 800 kcals per day or more there is little risk of the medical complications seen at lower levels of energy, including gallstones. Thus, although such patients continue to require extensive counseling and support during and after such programs, considerably less medically monitoring (and expense) is necessary. A recent comparison of 400 kcal and 800 kcal liquid diets demonstrated that although the rate of weight loss was greater in the lower calorie group (as expected), the long-term results were equivalent.

Liquid diets

A variety of other liquid diets are available commercially over the counter. Those that provide <800 kcals per day are potentially quite dangerous due to the lack of medical supervision and their use in individuals with mild obesity (who develop greater degrees of protein loss than those with more significant obesity). UltraSlimfast is a variant on this concept available in most supermarkets. Patients are instructed to use two shakes per day (approximately 200 kcals each) and then eat a sensible dinner (400-600 kcals). Since this results in >=800 kcals per day, and contains adequate quantities of protein, vitamins and minerals, it is generally considered to be safe. Although no studies have been published on its efficacy, it is unlikely to work for most individuals since most problem eaters take in most of their calories at the evening meal. If such individuals were able to eat a "sensible dinner' (and no additional calories just before and afterwards) in the first place, they would lose weight successfully without the product. It too also suffers from the difficulty of weight maintenance after the patient makes the transition to regular food.

Prepackaged meals

Another important development in the treatment of obesity is the tremendous expansion of commercial weight loss programs which provide the client with prepackaged meals in combination with varying degrees of nutrition education, exercise, and behavior modification. As with VLCD's, these programs offer the advantage of "removing the patient from the food environment," by limiting the patient's choices in regards to shopping, meal planning, food preparation and at meal time. As with VLCD's, these programs may be helpful to some patients who need considerable structure to initiate weight loss. Unfortunately, these programs also have difficulty at maintaining weight loss when the patient again makes the transition to regular foods. Because these diets contain >800 kcals per day intensive medical monitoring is not necessary during dieting.

Health consequences of dieting and weight loss

Effects of weight loss on morbidity and mortality: Surprisingly few studies have examined the effects of weight loss on morbidity and mortality. Studies examining the effect of weight loss on cardiovascular risk factors generally show beneficial changes with weight loss as predicted. Descriptive studies on mortality, however, show inconsistent results. Some studies show reduction in mortality in the obese (Wannamethee, 1990; Lean,1990) and some show increased mortality (Garrision, 1985; Pamuk, 1993; Wilcosky, 1990). Such descriptive studies are unable to clarify whether changes in mortality is caused by the weight change, whether disease or other factors which contribute to disease such as cigarettes causes weight loss, or whether both are related to a third factor. No randomized trials of long-term effects of voluntary weight loss on mortality have been published.

Effects of weight fluctuations.

Since so many Americans are dieting at any one time, and having so little long-term success, considerable interest has been focused on the potential adverse effects of weight cycling ("yo-yo dieting"). At least 9 potential adverse effects of weight cycling have been hypothesized, primarily from animal studies. These include making further weight loss more difficult, increasing total body fat and central obesity, increasing subsequent calorie intake, increasing food efficiency, decreasing energy expenditure, increasing levels of adipose-tissue lipolytic enzymes and liver lipogenic enzymes, increasing insulin resistance, increasing blood pressure, and increasing blood cholesterol and triglyceride levels. Most experts currently feel that these phenomen occur inconsistently, if at all.

More recently, seven descriptive studies have attempted to address this question by looking at the impact of weight fluctuations on CHD incidence, CHD mortality and total mortality. Five demonstrated an effect on at least one of these endpoints, while two others did not thus suggesting a link between weight variability and disease outcomes. It should be noted however, that these studies were not designed to assess the contribution of dieting (and voluntary weight loss) to weight fluctuation. Other factors which influence weight were also not considered.

Thus, at the present time, it remains uncertain dieting-induced voluntary weight fluctuations have a negative health impact in humans. This remains an important question, however, and reinforces the idea that casual attempts at quick weight loss should be avoided. At the present time, however, committed attempts at long-term weight loss should not be discouraged because of the potential of regaining weight.

Metabolic rate

As every dieter has observed, the rate of weight loss slows during the course of dieting. Because this can be quite discouraging to the unwary patient (or uninformed physician), it is important to inform the patient prior to initiating a weight loss diet that this is likely to occur. Weight loss is most rapid during the initial days of hypocaloric feeding due to changes in sodium and water balance. This is due to early loss of glycogen and protein (both contain water) and, depending on the degree of calorie deficit and type of diet, to sodium losses associated with ketonuria. Following this initial phase, weight loss will depend on the extent of energy deficit. With time, however, the rate of weight loss slows again as the body's metabolic rate decreases and the energy deficit becomes smaller. This change in metabolic rate can be 2-3 times greater than that predicted from changes in body weight. The lower the energy content of the diet, the lower the metabolic rate will be. Although it was initially suggested that exercise occurring during a period of hypocaloric feeding could prevent this decrease in metabolic rate, recent studies have suggested that it has no beneficial effect during hypocaloric feeding (but will increase the post-diet metabolic rate by preserving lean body mass).

Following the period of hypocaloric feeding (resumption of normal energy intakes), the metabolic rate will increase, but to a level below that observed before beginning the diet. In this instance, the extent of decrease is appropriate for the amount of weight lost. Nonetheless, an individual who has successfully lost weight, will have a significantly reduced total energy expenditure as compared to before weight loss. This is not only due to reductions in metabolic rate, but also reduction from the thermic effect of food (the individual eats less), and in differences in physical activity (it takes less energy to perform the same amount of activity for a smaller person). Thus, in order to maintain weight loss, individuals need to consume less energy than before dieting (and increase energy expenditure by increasing the amount of physical activity). Exercise

Aerobic exercise offers a number of significant advantages to patients attempting to achieve long term weight loss. First and foremost, exercise increases energy expenditure, helping to create the energy deficit necessary for weight loss. Unfortunately, the amount of energy expended during most aerobic exercises (walking, jogging, swimming, etc.) for the typical periods of time performed (20-30 minutes 4-5 times per week) is modest, approximately 500-1000 kcals per week. Thus, exercise can be predicted to have little effect on short term weight loss. Clinical trials reflect this modest effect: some studies demonstrate weight loss with exercise alone or extra weight loss when exercise plus diet is compared to diet alone, but some studies do not.

The importance of exercise for successful maintenance of weight loss, is more clearly established. In addition to the cumulative effect of increased energy expenditure (500-1000 kcals/week x 52 weeks=7-15 pounds per year), exercise effects the composition of the body substance lost during weight loss. When exercise is directly compared to diet, or when exercise plus diet is compared to diet alone, exercise results in greater preservation of lean body mass. That is, for each pound of weight lost, less fat and more muscle is lost during weight loss programs without exercise. This is particularly important since the body's resting metabolic expenditure (a major portion of the total daily energy expenditure) is closely correlated with lean body mass.

The observation that much of the long-term impact of exercise is through preservation of lean body mass has resulted in an increased interest in the potential role of resistance training (weight lifting, circuit training, etc.). Preliminary results suggest that resistance training during dieting does result in maintenance of lean body mass compared to dieting alone. Thus, highly motivated patients can be instructed to add resistance training to their aerobic exercise program.

Regular aerobic exercise results in an number of other benefits to the obese patient including improved cardiovascular training effect (increased exercise tolerance), decreased appetite (per calorie expended), a general sense of well-being, decreased blood pressure (in hypertensives), improved glucose metabolism and insulin action (in diabetics), improved blood lipids (in lipid disorders) and, in the long-term, decreased cardiovascular and all-cause mortality.

Young patients with mild-moderate obesity can be begun directly on a regular aerobic exercise program. Patients are commonly instructed to select two exercises and to perform either one of them 4-5 time per week for 20-25 minutes per day. Patients are taught to take their pulse and to generate a sustained tachycardia at 70-80% of their maximum predicted heart rate. Older patients and patients with severe obesity are instructed to begin walking programs without initial concern about meeting target heart rates. As weight loss proceeds, and patients become used to exercising regularly, patients can be advanced to formal aerobic programs.

Although many authorities recommend formal exercise stress testing prior to prescribing an exercise program for all patients, our practice has been to reserve testing for sedentary men over 40 and women over 50; for patients with known or suspected coronary artery disease; and for patients with two additional cardiovascular disease risk factors (in addition to obesity and sedentary lifestyle). Patients on VLCD's should be prescribed exercise programs in the same manner as patients on standard weight loss programs.

Behavior Modification and Social Support

Sustained weight loss requires long term changes in eating behavior. Patients must learn specific skills to facilitate decreased calorie intake and increased energy expenditure. Although formal behavior modification programs are available most patients can be taught basic behavioral strategies in the office.

The single most useful behavioral skill is planning and record keeping. Patients can be taught to plan both menus and exercise programs in advance. Patients are then instructed to record actual dietary and exercise behaviors. While the act of record keeping itself will aid in behavioral change, the availability of records will also help the health care provider

assess progress and make specific suggestions for additional problem solving. Specific reward systems are also useful for many patients. Refundable financial contracts have been shown to be effective in a number of small studies.

Social support is an additional essential component for any successful weight loss program. Most successful programs use peer group support. Involvement of family members is also particularly important. A comprehensive review of published results of weight loss programs strongly suggests that close provider-patient contact is a better predictor of success than the particular weight loss intervention.

Medications for treatment of obesity

Medications for the treatment of obesity are widely available both over the counter and with prescription. Medications include amphetamines (with high abuse potential-DEA Schedule II); non-amphetamine schedule IV appetite suppressants fenfluramine (Pondimin), phentermine (Ionamin), diethylproprion (Tenuate, Tepanil), mazindol (Sanorex, Mazanor) and pemoline (Cylert); the over-the-counter medication, phenylpropanolamine; and the antidepressants fluoxitene (Prozac) and sertraline (Zoloft).

Considerable controversy exists as to the efficacy of these agents and specific indications for their use. Numerous barriers to the use of medications for obesity have been recently described by advocates for their use. These include public perception of obesity as a disease of lack of willpower, expectation that medications should "cure" obesity, hindrance by state licensing agencies, limited research on long term efficacy, and the abuse potential of some of the medications. Numerous short term studies of appetite-suppressant drugs (ASD's) have been conducted. A recent meta-analysis of 36 studies using mazindol and fenfluramine showed that after a median duration of 12 weeks these drug as resulted in a mean weight loss 3 kg grater than placebo. In those studies with follow-up, discontinuation of the drugs most commonly resulted in weight gain. Advocates of drug therapy point out, however, that this is true for medications for most chronic diseases e.g. hypertension, cholesterol, diabetes, peptic ulcer disease; stopping a successful medication results in a recurrence of the condition being treated. In these cases, as well, medications do not cure the illness. A recent long-term study with phentermine 15 mg and fenfluramine 60 mg showed 9.4 kg weight loss after 3 years (Of the 121 who started the study 51 completed 190 weeks; these mean results do not include those who dropped out). Most common side effects included dry mouth, GI disturbances, somnolence, and nervousness.

Advocates of ASD's interpret these result to suggest that long-term use of such medications are safe and efficacious and merit a reevaluation in clinical practice, especially when used as an adjunct to a comprehensive weight loss program or to prevent weight gain following successful weight loss. At the UCSF Weight Management Program our practice is to limit the use of medications to morbidly obese patients with severe complications of obesity. Our initial experience in such patients, however, has not been encouraging. Anecdotal experience from others suggest that medications work best in those who need them the least, e.g. motivated patients with mild obesity.

Surgery

Although generally considered to be the last resort for the treatment of obesity, over 100, 000 obese patients have had surgical therapy. Few controlled trials exist, and the development of rational indications for surgery has been difficult.

Gastric operations are now the procedures of choice. Most popular are the vertical-banded (Mason) gastroplasty in which a smaller stomach pouch is created and gastric bypass procedures. Although both procedures result in significant weight loss, randomized trials comparing the two procedures tend to favor gastric bypass procedures. This is particularly true in patients who consume large amount of sweets. Perioperative mortality averages <1%, but ranges between 0-4% in different centers. Complications are common and include wound dehiscence, peritonitis, nausea and vomiting, vitamin deficiencies, and hair loss. When all necessary reversals, revisions, and patients lost to follow-up are considered, failure rates approach 50%.

Jejunoileal bypass was the first major surgical procedure popularized for the treatment of obesity. Although weight loss was effectively produced and maintained (average 100 pounds in 5 years), diarrhea and fluid and electrolyte disorders occurred chronically in over half of patients. One-third of patients develop progressive liver disease. This procedure has been abandoned. Two newer bypass operations under investigation are a biliointestinal operation and a biliopancreatic operation. Jaw wiring is another commonly used, but not recommended, surgical treatment. Initial weight loss is similar to gastric procedures on average but data on weight maintenance is quite variable. Suction lipectomy, or liposuction, is a surgical procedure permitting the removal of fat from specific areas of the body. Usually performed by plastic surgeons, 5 pounds of fat can be removed with each procedure in an attempt to reshape thighs and waists resistant to more traditional weight loss and exercise treatments. No advantageous metabolic changes are induced by the procedure.

Summary

No magic bullet exists for the very difficult, but medically important, task of weight loss. Although many of the most problems encountered in medical practice can be treated by weight loss alone, only motivated patients should be begun on weight loss programs. Weight loss treatments vary considerably in terms of risk, cost, and efficacy. For most patients with mild or moderate obesity, a multifactorial approach including diet, exercise, behavior modification and social support can be prescribed. Close patient-provider contact and long-term follow-up with emphasis on exercise are key ingredients for success. Motivated patients with severe obesity should be considered for supervised VLCD's, again emphasizing long-term dietary change, exercise, behavior modification, and social support.

Selected References

Andres R. Muller DC, Sorkin JD. Long term effects of change in body weight on all-cause mortality: A review. Ann Int Med 1993:119:737-743

Alford, B.B., Blankenship A.C., and Hagen R.D. "The effects of variations in carbohydrate, protein, and fat content of the diet upon weight loss, blood values, and nutrient intake of adult obese women." J Am Diet Assoc 1990;50:534-40.

Ballor, D. L., Katch V.L., Becque M.D., and Marks C.R.. "Resistance weight training during caloric restriction enhances lean body weight maintenance." Am J Clin Nutr 1988;47:19-25.

Blair SN, Shaten J, Brownell KD, et al. Body weight change, all-cause mortality and cause-specific mortality in the Multiple Risk Factor Intervention Trial. Ann of Int Med 1993;119:749-757

Bouchard, C. Is weight fluctuation a risk factor? N Engl J Med 1991; 324: 1887-9

Bouchard, C., Tremblay, A., Despres J. et al.The response of long-term overfeeding in identical twins. N Engl J Med 1990; 322: 1477-82

Bray, G.A. Pathophysiology of obesity Am J Clin Nutr 1992;55:488S-494S

Bray, GA. Barriers to the Treatment of Obesity. Ann Int Med 1991; 115:152-3

Brownell KD and Rodin J. The dieting maelstrom: Is it possible and advisable to lose weight? Am Psychologist 1994;49:781-791

Consensus Development Conference Panel. Gastrointestinal surgery for severe obesity: Consensus Development Conference Statement. Ann Int Med 1991; 115:956-61

Danford, D. and Fletcher, S.W. Methods for voluntary weight loss and control: National Institues of Health Technology Assessment Conference. Ann Int Med 1993;119: 641-770

Foster GD at al. A controlled comparison of three very-low-calorie diets: Effects on weight, body composition, and symptoms. Am J Clin Nutr 1992; 55:811

Frank, A. Futility and avoidance: Medical professionals in the treatment of obesity. JAMA 1993;269:2132-2133

Jeffery RW, Wing RR, French SA. Weight cycling and cardiovascular risk factors in obese men and women. Am J Clin Nutr 1992; 55: 641-644.

Lee IM, Paffenberger RS. Change in body weight and longevity. JAMA 1992;268:2045

Lee IM, Manson JE, Hennekens CH, Paffenber RS. Body weight and mortality: A 27 year follow-up of middle-aged men. JAMA 1993; 270:2823-2828

Levy AS, Heaton AW. Weight control practices of US adults trying to lose weight. Ann Int Med 1993; 119:661-666.

Lissner L, Odell PM, D'Agostino RB, et al. Variability of body weight and health outcomes in the Framingham population NEJM 1991;324:1839-1844.

Manson JE, Colditz GA, Stampfer MJ, Willett WC, et al. A prospective study of obesity and risk of coronary heart disease in women. NEJM 1990; 322:882-889

Stahl KA, Imperiale TF. An overview of the efficacy and safety of fenfluramine and mazindol in the treatment of obesity. Arch Fam Med 1993; 2:1033-37

Stamler, J. Epidemic Obesity in the United States. Arch Int Med 1993;153:1040-43

Stunkard, A.J., Harris J.R., Pedersen N.L., and McClearn G.E. "The body mass index of twins who have been reared apart" NEJM 1990;322:1483-87.

Wadden, TA, Foster GD, Letizia KA, and Mullen JL. "Long-term effects of dieting on resting metabolic rate in obese outpatients." JAMA 1990;264:707-11.

Wilber JF. Neuropeptides, appetite regulation and human obesity. JAMA 1991;266:257

Wilson GT. Relationship of dieting and voluntary weight loss to psychological functioning and binge eating. Ann Int Med 1993; 119:727-730

Williamson, D.F. Madans J., Anda R.F. et al. Smoking Cessation and Severity of weight gain in a national cohjort. N Engl J Med 1991;324:739-45

Yanowski SZ. Are anorectic agents the "magic bullet" for obesity? Arch Fam Med 1993; 2:1025-26

Special thank to Dr. Robert B. Baron MD, MS UCSF Division of General Internal Medicine