We studied three groups of patients without previous renal impairment, undergoing elective coronary artery bypass surgery. Group H (n = 7) underwent open heart surgery using moderate hypothermia (28 degrees C); Groups N and M (n = 8, each) were managed at normothermia. The extracorporeal circuit was primed with Hartmann's solution 2.5 L with the addition of mannitol 0.5 g/kg in Group M. Serum concentrations of sodium and creatinine, and the urinary concentrations of microalbumin and N-acetyl-beta-D-glucosaminidase (NAG) were measured in each patient at six different time intervals: T0, 6 h prior to surgery; T1, between sternotomy and 45 min into cardiopulmonary bypass (CPB); T2, in the interval from 45 min into, to prior to weaning off CPB; T3, from coming off CPB to skin closure; T4, in the first 6 h in the intensive care unit; and T5, at 6 days postoperatively. Creatinine clearance (CCR) and fractional sodium excretion (FENA) were calculated at each time point. Urine output during CPB at Interval T2 was significantly higher in Group H compared to Group N (P = 0.03) but not Group M. We found no significant differences in CCR, FENA, microalbuminuria, and urinary NAG among the three groups at any time. However, there were overall significant changes in measured variables over time compared to baseline. We conclude that CPB is associated with a significant alteration in renal function as shown by increased FENA, microalbuminuria, and urinary NAG. The use of hypothermic or normothermic CPB and the use of prophylactic mannitol did not produce any significant modification of these changes.