Tissue kallikrein in severe acute pancreatitis in patients treated with high-dose intraperitoneal aprotinin.

Abstract

The activation of the kallikrein-kinin system is thought to be one of the pathophysiologic mechanisms in acute pancreatitis. A radioimmunoassay for human urinary tissue kallikrein was developed and used to measure tissue kallikrein peritoneal exudate and plasma from 48 patients with severe acute pancreatitis. All patients were treated with intraperitoneal lavage. One group (n = 22) received high doses of the protease inhibitor aprotinin (aprotinin group), and the other group, saline (control group). Levels of kallistatin in peritoneal exudates and plasma were measured with an enzyme immunoassay. A large increase in tissue kallikrein was observed in the peritoneal exudate, which declined in both groups after multiple lavages. Complexing of liberated tissue kallikrein with kallistatin was evidenced by gel filtration in both peritoneal exudates and plasma in both groups. The decrease in kallistatin observed in both peritoneal exudate and plasma is therefore regarded as being due not only to repeated lavage, but also to true consumption of the binding protein. Some of the liberated tissue kallikrein in the peritoneal fluid and plasma was complexed to aprotinin. In the control group, six patients were operated on because of pancreatic necrosis, compared with none in the aprotinin group. The levels of tissue kallikrein in the lavage fluid were lower in the control group, but this was the result of the very low tissue kallikrein values in the six patients operated on for pancreatic necrosis. Levels of kallistatin in plasma and peritoneal exudate in these six patients were lower on the day of admission compared with the other patients, and the plasma levels continued to be lower during the first week. Large amounts of tissue kallikrein were found to be released into the peritoneal exudates in acute pancreatitis. Lavages effectively cleared the released tissue kallikrein. The tissue kallikrein was complexed to kallistatin, whereas in the aprotinin group, also to aprotinin both in plasma and in peritoneal fluid. The partitioning of kallikrein between the two inhibitors was the result of the interaction between enzyme and inhibitors and the turnover of the complexes formed. The low admission levels of kallistatin in the six patients operated on because of pancreatic necrosis suggest that kallistatin may act as an early marker of severity in acute pancreatitis.