Objective To compare the effects of snfentanil and fentanyl combined anesthesia on stress responses in pediatric patients undergoing surgical repair of congenital heart defect with cardiopulmonary bypass (CPB) .Methods Twenty-four pediatric patients aged 2-6 yr undergoing surgical repair of congenital heart defect were randomized into 2 groups (n=12 each): sufentanil combined anesthesia group (group S) and fetanyl combined anesthesia group (group F). The patients were premedicated with intramuscular morphine 0.2 mg/kg, scopolamine 0.01 mg/kg and ketamine 5 mg/kg. Anesthesia was induced with midazolam 0.1 mg/kg and snfentanil 0.7 μg/kg (in group S) or fentanyl 5 μg/kg (in group F). Tracheal intubation was facilitated with vecuronium 0.15 mg/kg. The patients were mechanically ventilated. PCO_2 was maintained at 35-45 mm Hg. Anesthesia was maintained with propofol infusion at 6-9 mg·kg~(-1)·h~(-1) and intermittent in boluses of midazolam,vecuronium and 3 doses of sufentanil (0.7, 1.5, 1.5 μg/kg) or fentanyl (5, 10, 10 μg/kg). Arterial blood samples were taken after entering the operation room, at 1 min after tracheal intubation, 1 min after splitting of sternum, immediately after rewarming, 10 min after termination of CPB and 24 h after operation for determination of plasma ACTH, cortisol, glucagon, lactic acid and blood glucose. Results The MAP and HR were maintained in the normal range during operation in both groups. The plasma concentrations of ACTH, cortisol, glucaon and lactic acid were significantly lower in group S than in group F. Plasma ACTH cortisol, glucagon, lactic acid and blood glucose increased significantly during operation as compared with the baseline before anesthesia in both groups. Conclusion Sufentanil combined anesthesia can inhibit stress response in pediatric patients more effectively during surgical repair of congenital cardiac defect performed under CPB.