We evaluated the effect of the novel protease inhibitor nafamostat on rat necrotizing pancreatitis through different routes of administration.

Three hours after the induction of severe pancreatitis, the rats received intravenous gabexate or intravenous or local mesenteric intra-arterial nafamostat. At 9 hours, ascites and bronchoalveolar lavage fluid were collected for the evaluation of capillary leakage (Evans blue extravasation). Pancreas and lung were excised for histologic features, myeloperoxidase, and trypsinogen activation peptide. Twenty-four hour survival was evaluated.

Only the intravenous infusion of nafamostat significantly reduced myeloperoxidase (11.7 ± 2.3 vs 18.3 ± 1.8 mU/mg; P < .05) and capillary leakage in lungs (Evans blue dye, 1.6 ± 0.3 vs 2.6 ± 0.3; P < .05). Only intra-arterial infusion of nafamostat significantly diminished capillary peritoneal leakage (Evans blue dye, 3.6 ± 0.9 vs 9.4 ± 0.4; P <.01). Typsinogen activation peptide levels were significantly reduced in all groups, but only intra-arterial infusion did so to baseline. Histologic inflammation in the pancreas was most significantly reduced after intra-arterial infusion (0.92 ± 0.08 vs 2.91 ± 0.06; P <.05). No form of protease inhibition reduced mortality rates.

The effects of protease inhibition depend on the route of administration. Nafamostat has maximal effects on the pancreas and peritoneal capillary leakage when delivered by way of local intra-arterial infusion, and shows a greater reduction of lung leukocyte infiltration and capillary leakage by the intravenous route. Nafamostat is more effective than gabexate. (Surgery 2001;130:175-81.)