The objective of this investigation was to test the capacity of recombinant human pancreatic secretory trypsin inhibitor (rhPSTI) to provide prophylaxis against experimental pancreatitis. Acute hemorrhagic pancreatitis was induced by intraductal injection of sodium taurocholate in rats and by intraductal injection of bile in dogs. In one treatment group of rats the injection of taurocholate was preceded by injection of rhPSTI. In a second group of rats the rhPSTI was given intraperitoneally starting 15 min after the induction of acute pancreatitis. The survival rate in a control group of rats was 13%. In contrast, the survival rate in groups receiving rhPSTI intraductally or intraperitoneally was 80% and 63%, respectively. The survival rate in a control group of dogs was 40% at 24 h and 0% at 48 h. In contrast, all the dogs receiving a single intraductal dose of rhPSTI, either immediately before the bile injection or mixed with the bile, survived for up to 6 weeks. Detailed biochemical and immunohistologic studies in the dog indicate that, whereas rhPSTI cannot prevent the initial bile-induced injury, it does prevent the subsequent development of that injury to the point where there is massive damage to the pancreas and the surrounding tissues, and changes in blood chemistry. The development of the initial injury is, therefore, presumed to involve activation of trypsinogen. Since rhPSTI prevents the serious consequences of experimental pancreatic injury by blocking the action of trypsin, and since the pathobiochemistry of human acute pancreatitis also implies an important role for trypsin, it is possible that rhPSTI could protect humans from the pancreatitis that complicates endoscopic retrograde cholangiopancreatography and endoscopic papillotomy.