The purpose of the present study is to clarify pathogenesis of cholangitis after pancreatectomy. The clinical study consisted of follow-up survey and hepatobiliary scintigraphy for the patients with type-I reconstruction (e.g. Imanaga procedure) and type-II reconstruction (e.g. Child procedure) after pancreatoduodenectomy. The experimental study was performed in dogs by creating three types of biliary reconstruction, namely, cholecystoduodenostomy (C-D), cholecystojejunostomy (C-J) and Roux-Y cholecystojejunostomy (R-Y). In the follow-up survey, cholangitis occurred in 19.0% of 21 type-I patients and in 33.3% of 18 type-II patients. In the hepatobiliary scintigraphy, type-I demonstrated smooth transit of bile along the reestablishing intestine. Type-II, on the other hand, showed marked stagnation of bile in the excluded loop, which could even trigger cholangitis. The experimental study showed that the results obtained from R-Y were no better than those obtained from C-D and C-J with respect to prevention of cholangitis based on histological, biochemical and bacteriological point of view. These studies suggest that type-I reconstruction carries little risk of causing cholangitis, whereas type-II reconstruction is not effective in preventing cholangitis.