Appetite Suppressant Agents: Current Problems and Controversies

 Harold E. Lebovitz, MD

Obesity has become one of the major public health problems in the United States. In the National Health and Nutrition Examination Surveys (NHANES), overweight is defined as a body mass index (BMI) > 27.8 in men and 27.3 in women, with BMI defined as weight in kilograms divided by height in meters squared. Using these criteria, the surveys in 1960–62, 1971–74, and 1976–80 showed that ~24% of adult men and 26% of adult women aged 20–74 years were overweight. In the 1988–91 survey, this figure had risen to 31.7% in men and 34.9% in women.¹

The major medical concerns regarding obesity are the co-morbid conditions to which it contributes.² The more important ones are listed in Table 1. It is because of the morbidity and mortality of these co-morbid conditions that the treatment of obesity has become a major medical target.²

Central Versus General Obesity
Studies such as Framingham,³ The Nurses’ Health,⁴ and the Health Professionals Follow-Up⁵ have indicated that BMI is a significant predictor of the development of type 2 diabetes, hypertension, coronary artery disease, insulin resistance, and even polycystic ovarian disease. BMI is highly correlated with total body fat (correlation coefficient >0.90) but is significantly less correlated with visceral (central) obesity.^6,7

In the mid-1980s, several publications from a large epidemiological study in Gothenburg, Sweden, presented data to show that the risk factor for type 2 diabetes,⁸ ischemic heart disease, and stroke⁹ was visceral obesity rather than total body obesity. It has been shown recently that insulin resistance is related to visceral adipose tissue volume and not to subcutaneous or total adipose tissue volume.^6,7

Visceral adiposity is best identified clinically by the waist circumference and quantitatively for research purposes by computerized tomography or magnetic resonance imaging. The visceral adipose tissue volume is regulated both by the amount of total adiposity and by specific metabolic factors that determine the percentage of accumulated adiposity deposited in the abdomen, as contrasted to the subcutaneous adipose tissue pool.^6,7 In individuals who have mild or moderate increases in adipose tissue mass (BMI < 30 kg/m²), the metabolically determined distribution will be the key determinant of an increased visceral adipose tissue pool. In more severely obese individuals (BMI >30 kg/m²), the visceral adipose tissue pool will be significantly increased as a result of the magnitude of the increase in total adipose tissue pool size, with metabolically determined distribution being less important.

Visceral adipose tissue has a number of important unique characteristics. Its turnover is much more rapid than subcutaneous adipose tissue. Its products drain directly into the portal vein. Its composition may be more influenced by the composition of dietary fat than that of subcutaneous adipose tissue.

Regulation of Adipose Tissue Mass
Adipose tissue mass is regulated by the balance between energy intake and energy expenditure. Genetic and environmental factors regulate both of these processes. Reasonable estimates derived from family and twin studies are that genetic factors contribute ~40% to obesity and environment contributes the other 60%.¹⁰

Major control of adipose tissue mass resides in the hypothalamus, where signals from the periphery that define the state of energy requirements and storage are interpreted and integrated and give rise to alterations in feeding behavior.¹¹ The hypothalamus, through a complex series of redundant mechanisms, coordinates food intake with energy expenditure. These systems are likely to involve leptin (a hormone made by adipose tissue); several peptides synthesized in the hypothalamus, such as neuropeptide Y, insulin, and various gastrointestinal hormones; and input from sympathetic, parasympathetic, and cortical neurons.¹¹ At the present time, intensive investigations of these regulators are being pursued, but their unique relationships to human obesity are unknown. It is known, however, that serotonergic and catecholaminergic pathways in the hypothalamus increase satiety and decrease hunger.

Treatment of Obesity
The treatment of obesity should be based on several key observations.^2,12,13 Obesity is a chronic disorder in which short-term effects are rarely sustained. Obesity in many individuals is a metabolic disorder. Weight reduction programs consisting of dietary intervention, increased physical activity, and behavioral modification have a high failure rate.

Weight loss in obese individuals should be linked to benefits on co-morbid conditions and not to cosmetic results. Relatively modest weight loss (5–10% of body weight) may have significant benefit on co-morbid conditions, particularly if this weight loss preferentially reduces visceral adiposity.^14,15

These observations have led to a renewed interest in drug therapy for obesity. A number of long-term weight reduction studies with pharmacological agents have established that a 5–10% decrease in body weight can be achieved in significant numbers of obese individuals.² These results require continued administration of the agents. As has been suggested by recent reports, some of these drug regimens are associated with serious and unacceptable side effects.^16,17

The drugs being reviewed in this publication are listed in Table 2. These drugs act either by stimulating beta-adrenergic systems in the hypothalamus, which results in a decrease in appetite, or by activating serotonin-responsive neurons, which inhibits appetite. Drugs can alter monoamine activity by three possible mechanisms: 1) by acting as a monoamine agonist, 2) by stimulating the release of the monoamine from the axon terminal, or 3) by inhibiting the re-uptake of the monoamine from the synapse into the nerve terminal (where it is converted to inactivate metabolites).

An agent that is a pure re-uptake inhibitor increases synaptic monoamine levels approximately two- to threefold.¹⁸ An agent that releases monoamines from nerve terminals will increase monoamine levels locally more than tenfold. A monoamine agonist will similarly increase activity many fold.

Specific Appetite Suppressant Drugs: Effects on Weight

Noradrenergic agents
Diethylpropion, phendimetrazine, and phentermine are derivatives of beta-phenethylamine and stimulate the release of norepinephrine from nerve endings.^12,13,19 Mazindol is a tricyclic compound that blocks the re-uptake of norepinephrine.^13,19 In 1975, Scoville²⁰ analyzed for the Food and Drug Administration (FDA) the results of more than 200 studies of the effects of noradrenergic agents on weight loss in obese subjects. The studies included had a duration of 4–20 weeks. Scoville concluded that drug-treated patients lost 0.23 kg/week more than placebo-treated patients. He noted no difference in effectiveness among the various noradrenergic drugs.

Serotonergic agents
d,l-fenfluramine was introduced in 1975 as an appetite suppressant.²¹ Though the drug resembles amphetamines chemically, its mode of action is quite different. d,l-fenfluramine partially inhibits the re-uptake of serotonin but also stimulates the release of serotonin from nerve endings.^19,21 The increase in serotonin in the synapse is thought to stimulate neurons that cause satiety and decrease appetite. Dexfenfluramine is the biologically active isomer in d,l-fenfluramine.

Several 6-month to 1-year open label studies comparing d,l-fenfluramine treatment to placebo showed that d,l-fenfluramine causes more weight loss than placebo (greater by ~5% of body weight), is more beneficial in a structured program, and achieves better results than behavioral modification.^12,13,19,21

Dexfenfluramine plus diet was shown in a large, 1-year multicenter European trial to cause significantly greater weight loss than placebo plus diet.²² Both placebo and dexfenfluramine patients achieved their maximum weight loss at 6 months. Mean weight loss with dexfenfluramine was 10.26% of initial body weight and with placebo was 7.18% of initial body weight. Nearly twice as many patients on dexfenfluramine lost 10% or more body weight than those on placebo (34.9 vs. 17%). Other studies have shown similar results, though an occasional small study has shown that dexfenfluramine has no greater effect on weight loss than does placebo.²¹

Combination of noradrenergic and serotonergic agents
The study that established the credibility of long-term appetite suppressant drug therapy as a feasible approach to obesity management was carried out by Weintraub and colleagues and reported in a series of publications in 1992.^23-25 They treated 121 obese patients with a combination of fenfluramine (60 mg) and phentermine (15 mg) as an adjunct to behavior, nutrition, and exercise counseling in a 34-week, double-blind, placebo-controlled trial (Figure 1). During the first 6-week, no-drug, lead-in period, patients lost 4.2 kg. During the 28-week, double-blind portion of the study, drug-treated patients lost 14.3 kg, while placebo-treated patients lost only 4.6 kg. From week 35 to week 60, all patients were treated with continuous, intermittent, or augmented combination drug therapy.

As shown in Figure 1, weight loss was maintained in continuously or intermittently treated patients. Active drug treatment of previously placebo-treated individuals induced weight loss that equaled that of the initially drug-treated patients. Those few people who were resistant to drug effect initially remained so on increased doses (augmented treatment).

The patients were continued on various drug therapy programs for years 2 and 3. The number of patients remaining in the study at 2 years was 83 and at 3 years was 59. In spite of continued drug therapy, there was some weight gain (weight loss was 10.3 kg at 2 years and 9.4 kg at 3 years).

Following discontinuation of drug therapy, the patients regained weight even though lifestyle interventions were maintained. Weight change at 4 years was 1.4 kg less than at the start of the study.

The conclusions Weintraub and associates drew from this study were as follows: 1) long-term appetite suppressant therapy is effective; 2) appetite suppressant therapy is additive to lifestyle changes; and 3) body weight set point is not corrected by appetite suppressants, since discontinuation of drug treatment results in restoration of weight to pre-therapy values.

Serotonin re-uptake inhibitors
The classic serotonin re-uptake inhibitors are fluoxetine and sertroline. During early studies of the effects of fluoxetine on the treatment of depression, it was noted that weight loss was a side effect. Subsequent studies showed a dose-response effect of fluoxetine on weight loss in obese patients.²⁶ Fluoxetine treatment (60 mg/day) of obese individuals causes a weight loss of ~3–4 kg, as compared to placebo.^12,27 Weight loss, when it does occur, is seen in the first few months of treatment and frequently is not sustained with chronic treatment (>6 months). Large, multicenter trials of fluoxetine treatment for obesity have not been published, and the drug has not been approved for treatment of obesity by the FDA.

Sibutramine: combined serotonin and norepinephrine re-uptake inhibitor
Sibutramine is a novel aryclyclobutyl-alkylamine that is converted to active metabolites that are both serotonin and norepinephrine re-uptake inhibitors.¹⁸ This drug differs from dexfenfluramine and phentermine in that it does not stimulate the release of monoamines from the nerve terminals.

It has been shown in several pivotal studies to cause significant weight loss in obese individuals.18 Doses of 5, 10, and 20 mg once daily resulted in weight losses of 2.1, 3.5, and 5.1 kg in excess of that in placebo controls.¹⁸ Weight loss maximizes at about 6 months and remains stable with continued treatment (longest study 1 year). At 6 months, ~36% of patients treated with 20 mg/day had a weight loss of 10% or more, and 63% had a weight loss of 5% or more.¹⁸ Similar data are observed with 12 months of treatment at 15 mg/day.

Benefits of Weight Loss on Co-Morbid Conditions
Weight loss itself will cause the benefits described in Table 3. In studies of the effect of appetite suppressants on co-morbid conditions, it is necessary to determine whether the observed effects are due just to the weight loss itself or whether some of the effects may be related to direct effects of the drugs. For example, dexfenfluramine, when given to obese type 2 diabetic patients, has been reported to improve fasting plasma glucose and oral glucose tolerance after 1 week of therapy, when no significant weight loss had occurred.²⁸

Dexfenfluramine treatment has been reported to improve insulin sensitivity as measured by the glucose clamp.^15,28 Marks and associates found that weight loss following 3 months of dexfenfluramine treatment selectively decreased visceral adipose tissue.¹⁵ This was associated with an increase in insulin sensitivity, a reduction in total cholesterol and serum triglycerides, and a decrease in fasting plasma glucose and glycated hemoglobin. Without comparable placebo-treated weight loss in those studies, it is difficult to determine which, if any, effects might be due to dexfenfluramine rather than to the accompanying weight loss.²⁹

A 1-kg decrease in body weight has been shown to decrease systolic blood pressure by 1.6 and diastolic blood pressure by 1.3 mm/Hg, respectively.³⁰

Weight loss associated with sibutramine therapy decreases serum triglycerides and low-density lipoprotein (LDL) cholesterol and raises serum high-density lipoprotein (HDL) cholesterol in direct relationship with the magnitude of weight loss.¹⁸ For example, a 5–10% weight loss decreases plasma triglycerides a mean of 12.9% and plasma LDL cholesterol 3.8% and raises plasma HDL cholesterol 4.5%. Weight loss of 10–15% of body weight decreases plasma triglycerides 23.6% and plasma LDL cholesterol 8.7%, while raising plasma HDL cholesterol 5.4%.

Complications of Appetite Suppressant Drug Treatment
Appetite-suppressant drug treatment is associated with a number of both major and minor side effects. The side effects are unique for each class of appetite suppressant agents.

Noradrenergic agents frequently cause insomnia, dry mouth, restlessness, and some increase in blood pressure and heart rate. They are contraindicated with the use of monoamine oxidase inhibitors.^12,19

The major side effects of sibutramine are a rise in mean systolic and diastolic blood pressure of 2–3 mm/Hg and a mean heart rate increase of 3–5 beats per minute. Dry mouth and insomnia may occur.¹⁸

The serotonergic drugs d,l-fenfluramine and dexfenfluramine have been implicated recently in the development of two major catastrophic illnesses: primary pulmonary hypertension^16,31,32 and valvular heart disease.^17,33-35 Additionally, there are significant concerns that these drugs may also lead to brain neurotoxicity.³⁶ The less serious side effects of the serotonergic drugs include headache, dry mouth, diarrhea, drowsiness, depression, and thinking abnormalities.

Primary Pulmonary Hypertension
In the 1960s, an epidemic of primary pulmonary hypertension in association with the anoretic agent aminorex fumarate occurred in Switzerland, Germany, and Austria.¹⁶ In the early 1990s, French investigators reported a cluster of cases of primary pulmonary hypertension in individuals who had used derivatives of fenfluramine.^16,21

The International Primary Pulmonary Hypertension Study Group did a case-control study of 95 patients with primary pulmonary hypertension and 355 matched controls.¹⁶ The use of anoretic drugs (mainly fenfluramine derivatives) was associated with a 6.3 odds ratio for primary pulmonary hypertension. When anoretic agents were used for more than 3 months, the odds ratio was 23.1. The absolute risk was estimated to be 3/50,000 treated patients.

Some cases of primary pulmonary hypertension improve after the anoretic agent is discontinued, but many progress further. One recently reported patient took fenfluramine and phentermine for only 23 days and died 8 months later of plexogenic pulmonary hypertension.³²

Valvular Heart Disease
A recent article in the New England Journal of Medicine reported that 24 women treated with fenfluramine-phentermine with no previous history of heart disease presented with cardiovascular symptoms or a heart murmur.¹⁷ They were found to have abnormal valvular morphology and regurgitation demonstrated by echocardiography.¹⁷ Both left- and right-side heart valves were involved. Eight of the women also had newly documented pulmonary hypertension. Cardiac surgical intervention had been required in 5 patients. Duration of anoretic drug treatment was 1–28 months, with a mean of 11 ± 6.9 months.

In the same issue of the journal, Graham and Green34 of the FDA reported that 28 additional cases were reported to them with similar findings. The patients’ median age was 45 years, and all were women. The median daily dose of fenfluramine was 60 mg and of phentermine was 30 mg. Six patients had undergone valve replacement. Additional cases treated with fenfluramine alone (2 patients) and dexfenfluramine alone (4 patients) were reported to them.

Brain Serotonin Neurotoxicity
The potential of fenfluramine and dexfenfluramine to cause neurotoxic side effects has been of considerable concern. Fenfluramines cause dose-related chronic reductions in serotonin axonal markers in all animal species in which it has been tested and with all routes of drug administration. The relative doses that cause these neurotoxic effects have been estimated to be of the same order of magnitude as those used to treat human obesity. No human studies have been carried out, and the pathological and clinical potential for neurotoxicity is not known.

Lessons Learned and Future Perspective
The emergence of severe complications associated with serotonergic appetite-suppressant drug treatment of obesity should not be allowed to lessen the enthusiasm generated in the past decade for developing effective drug treatment of obesity. The fundamental problems and principles defined in the report of the National Task Force on the Prevention and Treatment of Obesity still prevail.

Several key principles provide the cornerstone for any pharmacological intervention in obese individuals. Pharmacological agents must be adjuncts to lifestyle intervention, i.e., diet and in-creased physical activity. Pharmacological agents should be used to reduce co-morbid conditions and not for cosmetic purposes. Pharmacological agents must be safe, effective, and nonaddictive for long-term use, since chronic weight loss itself does not reset the body weight control setting of the hypothalamus. A weight loss of 5–20% of initial body weight is sufficient to significantly improve co-morbid conditions.

The association of fenfluramine and dexfenfluramine with the development of both valvular heart disease and primary pulmonary hypertension would seem to preclude these and related drugs as useful agents in the treatment of obesity. It is of considerable importance to understand the mechanism causing these complications, since future drug development of appetite suppressant agents must eliminate these side effects. Dexfenfluramine and fenfluramine are serotonin releasers, as well as serotonin re-uptake inhibitors. Such drugs cause much higher local concentrations of serotonin than do classic serotonin re-uptake inhibitors, such as fluoxetine.

The failure to identify valvular abnormalities and pulmonary hypertension with fluoxetine, which has been used extensively for many years, suggests the possibility that pure re-uptake inhibitors, which cause much smaller rises in local serotonin concentrations and restrict the rises to the synaptic clefts, are the appetite suppressant drugs of choice until more specific neurohormonal regulators are developed.

Sibutramine, which is tentatively scheduled to be marketed in early 1998 as an appetite suppressant for weight loss, is a specific serotonin and noradrenergic neuronal re-uptake inhibitor. Studies available indicate that it is not a serotonin releaser and that its effect on hypothalamic serotonin levels is an increase of approximately two- to threefold, consistent with its re-uptake inhibitor effect. If current studies on its side effects show it not to be associated with heart valve abnormalities or pulmonary hypertension, it will clearly meet the guidelines suggested by the National Task Force on the Prevention and Treatment of Obesity.

An area of potentially useful clinical investigation is combination treatment of fluoxetine with noradrenergic agents. Fluoxetine alone does not appear to sustain significant weight loss beyond 6 months. Perhaps combination with noradrenergic agents would improve the chronic efficacy of fluoxetine.

Continued efforts to find appropriate pharmacological adjuncts to facilitate weight loss in obese individuals with co-morbid conditions should remain a high priority for clinical investigators and clinicians alike.

References

¹Kuczmarski RJ, Flegal KM, Campbell SM, Johnson CL: Increasing prevalence of overweight among U.S. Adults: the National Health and Nutrition Examination Surveys, 1960 to 1991. JAMA 272:205-11, 1994.

²National Task Force on the Prevention and Treatment of Obesity: Long-term pharmacotherapy in the management of obesity. JAMA 276:1907-15, 1996.

³Hubert HB, Feinlieb M, McNamara PM, Castelli WP: Obesity as an independent risk factor for cardiovascular disease. Circulation 67:968-77, 1977.

⁴Manson JE, Willet WC, Sampler MJ, Colditz GA, Hunter DJ, Hankinson SE, Hennekens CH, Speirer FE: Body weight and mortality among women. N Engl J Med 333:677-85, 1995.

⁵Chan JM, Rimm EB, Colditz GA, Stamper MJ, Willet WC: Obesity, fat distribution, and weight gain as risk factors for clinical disease in men. Diabetes Care 17:961-69, 1994.

⁶Banerji MA, Chaiken RL, Gordon D, Kral JG, Lebovitz HE: Does intra-abdominal adipose tissue in black men determine whether NIDDM is insulin-sensitive or insulin-resistant. Diabetes 44:141-46, 1995.

⁷Banerji MA, Lebowitz J, Chaiken RL, Gordon D, Kral JG, Lebovitz HE: Relationship of visceral adipose tissue and glucose disposal is independent of sex in black NIDDM subjects. Am J Physiol 273:E425-32, 1997.

⁸Ohlson LO, Larsson B, Svardsudd K, Welin L, Eriksson H, Wilhelmsen L, Bjorntorp P, Tibblin G: The influence of body fat distribution on the incidence of diabetes mellitus. Diabetes 34:1055-58, 1985.

⁹Lapidus L, Bengtsson C, Larsson B, Pennert K, Rybo E, Sjostrom L: Distribution of adipose tissue and risk of cardiovascular disease and death: a 12-year follow-up of participants in the population study of women in Gothenburg, Sweden. Br Med J 289:1257-61, 1984.

¹⁰Stunkard AJ: The Salmon Lecture. Some perspectives on human obesity: its causes. Bull NY Acad Med 64:902-23, 1988.

¹¹Rosenbaum M, Leibel RL, Hirsch J: Obesity. N Engl J Med 337:396-407, 1997.

¹²Bray GA: Use and abuse of appetite-suppressant drugs in the treatment of obesity. Ann Intern Med 119:707-13, 1993.

¹³Ryan DH: Medicating the obese patient. Endocrinol Metab Clin North Am 25:989-1004, 1996.

¹⁴Despres JP: Lipoprotein metabolism in visceral obesity. Int J Obesity 15:45-52, 1991.

¹⁵Marks SJ, Moore NR, Clark ML, Strauss BJ, Hockaday TD: Reduction of visceral adipose tissue and improvement of metabolic indices: effects of dexfenfluramine in NIDDM. Obes Res 4:1-7, 1996.

¹⁶Abenhaim L, Moride Y, Brenot F, Rich S, Benichou J, Kurz X, Higenbottam T, Oakley C, Wouters E, Aubier M, Simonneau G, Begaud B: Appetite suppressant drugs and the risk of primary pulmonary hypertension. N Engl J Med 335:609-16, 1996.

¹⁷Connolly HM, Crary JL, McGoon MD, Hensrud DD, Edwards BS, Edwards WD, Schaff HV: Valvular heart disease associated with fenfluramine-phentermine. N Engl J Med 337:581-88, 1997.

¹⁸Submission to FDA Sept 26, 1996. Sibutramine hydrochloride monohydrate NDA 20-632. Knoll Pharm.

¹⁹Goldstein DJ, Porvin JH: Long-term weight loss: the effect of pharmacologic agents. Am J Clin Nutr 60:647-57, 1994.

²⁰Scoville BA: Review of amphetamine-like drugs by the Food and Drug Administration: clinical data and value judgments. In Obesity in Perspective. Bray GA, Ed. DHEW Publication No. NIH-75-708, Washington DC, US Govt. Printing Office, 1975, p. 441-43.

²¹Davis R, Faulds D: Dexfenfluramine: an updated review of its therapeutic use in the management of obesity. Drugs 52:696-724, 1996.

²²Guy-Grand B, Appelbaum M, Crepaldi G, Gries A, Lefebvre P, Turner P: International trial of long-term dexfenfluramine in obesity. Lancet 2:1142-45, 1989.

²³Weintraub M, Sundaresan PR, Madan M, Schuster B, Balder A, Lasagna L, Cox C: Long-term weight control study I (week 0-34). Clin Pharmacol Ther 51:586-94, 1992.

²⁴Weintraub M, Sundaresan PR, Schuster B, Ginsberg G, Madan M, Blader A, Stein EC, Byrne L: Long term weight control study II (weeks 34-104). Clin Pharmacol Ther 51:595-601, 1992.

²⁵Weintraub M, Sundaresan PR, Schuster B, Averbuch M, Stein EC, Byrne L: Long-term weight control study III, IV, V (weeks 104 -156, 156-190, 190-210). Clin Pharmacol Ther 51:602-607, 608-14, 615-18, 1992.

²⁶Levine LR, Enas GG, Thompson WL, Byyny RL, Daver AD, Kirby RW, Kreindler TG, Levy B, Lucas CP, McIlwain HH: Use of fluoxetine, a selective serotonin-uptake inhibitor in the treatment of obesity: a dose-response study. Int J Obesity 13:635-45, 1989.

²⁷Gray DS, Fujioka K, Devine W, Bray GA: Fluoxetine treatment of the obese diabetic. Int J Obesity 16:193-98, 1992.

²⁸Scheen AJ, Paolisso G, Salvatore T, Lefebvre PJ: Improvement of insulin-induced glucose disposal in obese patients with NIDDM after 1 week treatment with d-fenfluramine. Diabetes Care 14:325-32, 1991.

²⁹Wiley KA, Molyneaux LM, Yue DK: Obese patients with type 2 diabetes poorly controlled by insulin and metformin: effects of adjunctive dexfenfluramine therapy on glycaemic control. Diabetic Med 11:701-704, 1994.

³⁰Seasson J, Fagard R, Lijnen P, Amery A: Body weight, sodium intake and blood pressure. J Hypertension 7 (Suppl 1):519-23, 1989.

³¹Manson JAE, Faich GA: Pharmacotherapy for obesity—do the benefits outweigh the risks? N Engl J Med 335:659-60, 1996.

³²Mark EJ, Patalas ED, Chang HT, Evans RJ, Kessler SC: Fatal pulmonary hypertension associated with short-term use of fenfluramine and phentermine. N Engl J Med 337:602-606, 1997.

³³Cannistra LB, Davis SM, Bauman AG: Valvular heart disease associated with dexfenfluramine. N Engl J Med 337:636, 1997.

³⁴Graham DJ, Green L: Further cases of valvular heart disease associated with fenfluramine-phentermine. N Engl J Med 337:635, 1997.

³⁵Curfman GD: Diet pills: redux. N Engl J Med 337:629-30, 1997.

³⁶McCann UD, Seiden LS, Rubin LJ, Ricaurte GA: Brain serotonin neurotoxicity and primary pulmonary hypertension from fenfluramine and dexfenfluramine. JAMA 278:666-72, 1997.

Harold E. Lebovitz, MD, is a professor of medicine and chief, Division of Endocrinology, at State University of New York Health Science Center, in Brooklyn.

Note of disclosure: Dr. Lebovitz sits on advisory panels for Knoll Pharmaceutical, Smith Kline Beecham, Amylin, and Bayer Corporation. He receives honoraria from Knoll Pharmaceutical, Bristol-Myers Squibb, Bayer, Warner-Lambert, and Amylin. He receives research funding from Knoll Pharmaceutical, Bayer, and Smith Kline Beecham. These companies are involved in manufacturing drugs for the treatment of obesity and/or diabetes mellitus.