The outcomes of intra-arterial urokinase versus surgery for acute peripheral arterial occlusion (PAO) were compared. Patients at a university hospital who had received intraarterial urokinase for PAO were identified by computer and pair-matched on the basis of comorbidities, age, sex, and site of occlusion to computer-selected patients who had undergone surgery. Only patients with category I or II ischemia were considered. The study period for the urokinase group was February 1995 through January 1996, and the period for the surgery group was June 1993 through January 1996. Twenty-eight patients in each group met the selection criteria. Patients who had received urokinase had a significantly shorter median length of stay (8.5 days) than patients in the surgery group (13 days) and significantly fewer infectious complications (2 versus 10). No differences in amputation rates, total hospital costs, or mortality rates were detected. Patients who received intra-arterial urokinase for PAO had a shorter length of stay in the hospital and fewer infectious complications than those who underwent surgery.