In one of our earlier studies, an impaired biliary function in diabetes was suggested. We studied bile formation in rats with streptozotocin-induced diabetes (60 mg per kg body weight). Diabetic rats showed hyperglycemia and hypoinsulinemia, but no significant changes in hematocrit, plasma protein concentration or plasma osmolality. Bile flow was significantly (p < 0.05) reduced (-23%) as compared with control animals, despite a higher (p < 0.05) bile acid secretion rate (+56%).
The biliary responses to three choleretic compounds (taurocholate, ursodeoxycholate and insulin), acting in a very different way upon bile formation, were not impaired in diabetes. The study of the relationship between bile acid output and bile flow, after infusion of taurocholate at different doses (0.25 to 1.5 pmoles per min per 100 gm body weight) showed that diabetesinduced cholestasis in the rat is mainly related to a decreased bile acid-independent fraction of the bile flow.
We tested the possible role of hyperglycemia and hypoinsulinemia as cholestatic factors in diabetes. Glucose infusion [300 mM, 150 pl per rnin (Group G)] induced a significant (p < 0.05) reduction in bile flow (-0.33 p1 per min per gm liver) as compared to the basal period. After acute pancreatectomy (P) or mannoheptulose treatment [0.14 mmole per 100 gm body weight (Group M)], similar cholestatic effects were observed (-0.29 and -0.27 pl per rnin per gm liver, respectively). However, plasma glucose and insulin concentrations were higher (p < 0.01) in Group G than in P or M.
Glucose infusion to acutely pancreatectomized (PG) or mannoheptulose-treated (MG) rats induced a similar and marked cholestasis (-0.46 and -0.51 pl per rnin per gm liver, respectively; both p < 0.001). The plasma insulin concentration was not different among the P, M, PG and MG. Nevertheless, a lower (p c 0.05) bile flow was observed in PG and MG, as compared to P and M, respectively. Plasma glucose concentrations were higher (p < 0.001) in PG and MG as compared to P and M.
Bile acid output and plasma osmolality were not significantly different in controls, P, M, PG and MG.