The effect of allopurinol, a xanthine oxidase inhibitor, on canine experimental ischemic pancreatitis was studied. The animals were divided into nine groups: 1. Group 1. Control with pancreatic ischemia; 2. Group 2. Received allopurinol once, previous to ischemia; 3. Group 3. Received allopurinol once, immediately after ischemia; 4. Group 4. Received allopurinol immediately after ischemia and then daily; and 5. Groups 5, 6, and 7 were controls for the operation, allopurinol, and its vehicle, respectively; 6. Group 8 (pancreatic ischemia) and Group 9 (that received allopurinol after ischemia and daily) were also studied histologically. Serum amylase was determined in all animals. In Groups 1 and 5, following the ischemic period, hyperamylasemia developed and a peak was reached 24 h after ischemia. In Group 2, a significant decrease of amylase levels was found, compared to matched controls immediately after ischemia and then rose, reaching on the fifth day a peak that was less than the controls at 24 h. In Group 3, the serum amylase level increased immediately to values similar to controls; later, there was a drop to levels lower than those found in controls, followed by a peak on the fifth day. In Group 4, there was no significant elevation in the amylase values. Groups 6 and 7 showed no changes of amylasemia. In this experimental model, allopurinol blocked or ameliorated significantly cellular injury, as shown by a decrease of amylase levels in blood, and of histopathological changes, depending on dose and time of administration. These results offer the possibility of a prophylactic therapy for chronic relapsing and idiopathic pancreatitis.