OBJECTIVE After weaning from cardiopulmonary bypass (CPB), a decrease in nasopharyngeal temperature (NPT) occurs (afterdrop). The pathophysiology of the afterdrop remains unclear: It might be caused by either inadequate total body rewarming on CPB or to heterogenous distribution of heat during CPB, with subsequent redistribution of heat from the warmer core to the cooler shell tissues. The study objectives were (1) to determine whether post-CPB afterdrop is the result of a negative CPB thermal balance, and (2) to investigate which sites (if any) could best predict the afterdrop. METHODS Prospective evaluation using within-patient comparisons during CPB cooling, CPB rewarming, and 45 minutes post-CPB. METHODS Adult patients gave informed consent before a cardiac surgical procedure in a university hospital. METHODS Eight patients undergoing CABG or valvular replacement with hypothermic CPB (NPT near 29 degrees C) and standardized general anesthesia. METHODS Each patient was studied with temperature monitors (Mon-a-therm 7000; Mallinckrodt-Medexel, Gemenos, France) attached to disposable thermocouple probes placed as follows: urinary bladder, rectum, deltoid, esophagus, nasopharynx, tympanic membrane, and four skin sites. In addition, the temperatures from the thermistors of the pulmonary artery catheter, and the arterial and venous lines of the CPB circuit were considered. Thirteen sites for monitoring temperature were studied. RESULTS Temperatures were recorded every 5 minutes, from the beginning of CPB to the 45th minute after CPB, and thermal exchanges were calculated: change in body heat (QBH), thermal exchanges between the patient and the pump (QCPB), metabolic heat production (Qm) (equal to calculated VO2 at the pump level), and heat loss to the environment (QS) (equal to QBH-QCPB-Qm). Thermal exchanges were obtained in six patients during the plateaus of cooling and rewarming, during the whole CPB phase, and after CPB. It was found that despite a change in QBH during rewarming (1,017 +/- 88 kJ) that was slightly greater than during cooling (-1,008 +/- 104 kJ) (mean +/- SEM), a significant decrease in post-CPB "core" temperature occurred (afterdrop: -1.4 degrees C). Magnitude of the afterdrop was directly related to the magnitude of tympanic membrane cooling and was negatively correlated to the temperature difference between the warmest site (tympanic membrane) and the coolest site (cutaneous thigh temperature) observed at the end of rewarming (r = -0.667; p < 0.05). CONCLUSIONS It is suggested that besides post-CPB heat loss, redistribution of heat may be involved in the mechanism of the afterdrop and that measurements of tympanic membrane and cutaneous thigh temperatures are the best monitors of adequacy of rewarming during CPB.