To characterize the learning curve for laparoscopic cholecystectomy, we compared the first 47 cases (group A), which were performed by two senior attending surgeons who assisted each other when the procedure was introduced into clinical practice (1990-1991), with the first 46 cases (group R) performed by two surgical chief residents who were assisted by members of the teaching faculty in 1992-1993. The patient groups were comparable in terms of age, sex, and anesthetic class, but pathologically proven acute cholecystitis was more common in group R (33% vs. 9%; p < 0.005). To analyze operative procedures and outcomes, we compared operative time, frequency of successful operative cholangiography (attempted in all cases), frequency of conversion to open cholecystectomy, major complication rate, and days of postoperative stay for all patients and for those without complications. Of these parameters, only operative time for nonacute cases differed significantly between the groups (144 min for group A vs. 114 min for group R; p < 0.05). Complications in group A included one ductal injury and one case of postoperative pancreatitis; group R had one ductal injury and two cases of postoperative bleeding. We conclude that (a) the learning curve has similar structure for senior surgeons and resident trainees; and (b) the resident learning curve is not hazardous when teaching assistants are trained in the procedure, which has implications for safe instruction and proctoring of residents and staff.