In order to investigate the effect of long-term treatment with dexfenfluramine (dF) on 24-hour energy expenditure (EE), 10 obese females were studied in a double-blind design. Shortly before and 4 weeks after cessation of a 13 months treatment period with either dF (30 mg/day) or placebo (PL) the 24-hour EE was measured. The measurements were performed using a 24 m3 direct heat sink calorimeter with continuous real time measurements of evaporative and sensible heat losses. The patients performed a standardized program of exercise, rest and meals. The measurements were performed at 24 degrees C and at a humidity between 3 and 11 g/m3. Discontinuation of dF treatment did not change energy expenditure significantly from placebo, neither when expressed in kJ/kg lean body mass nor in kJ/kg body weight. After cessation of treatment total 24-hour EE decreased likewise nonsignificantly by 2.9 percent in the dF group and by 4.0 percent in the PL group. EE measured over 24 hours was subdivided into day and night periods and into resting energy expenditure as well as a measurement of the heat losses over a period of 3 hours after a meal. This subdivision of the EE showed similar nonsignificant differences. The conclusion is therefore that dF possesses no significant thermogenic effect during long-term administration in human obese subjects.