The effects of drugs used to treat obesity on the autonomic nervous system

Objective: Body fatness is partly under hypothalamic control with effector limbs, which include the endocrine system and the autonomic nervous system (ANS). In previous stud ies we have shown, in both obese and never-obese subjects, that weight increase leads to increased sympathetic and decreased parasympathetic activity, whereas weight decrease leads to decreased sympathetic and increased parasympathetic activity. We now report on the involvement of such ANS mechanisms in the action of anti-obesity drugs, independent of change in weight. Research Methods and Procedures: Normal weight males (ages 22 to 38 years) were fed a solid food diet, carefully measured to maintain body weight, for at least 2 weeks, as inpatients at the Rockefeller University General Clinical Research Center. In a single-blind, placebo/drug/placebo design, eight subjects received dexfenfluramine, seven phentermine (PHE), and seven sibutramine (SIB). ANS measures of parasympathetic and sympathetic activity included: determination of amount of parasympathetic control (PC) and sympathetic control (SC) of heart period (interbeat interval), using sequential pharmacological blockade by intravenous administration of atropine and esmolol. These autonomic controls of heart period are used to estimate the overall level of parasympathetic and sympathetic activities. Norepinephrine, dopamine, and epinephrine levels in 24-hour urine collections were measured and also resting metabolic rate (RMR). Results: Sufficient food intake maintained constant body weight in all groups. PHE and SIE produced significant increases in SC but no change in PC or in RMR. In contrast, dexfenfluramine produced marked decreases in SC, PC, and RMR. For all three drugs, the effects on urine catecholamines directly paralleled changes in cardiac measures of SC. Discussion: ANS responses to PHE and SIE were anticipated. The large, and unanticipated, response to dexfenfluramine suggests further study to determine whether there could be any relation of these ANS changes to the adverse cardiovascular effects of treatment with dexfenfluramine.