Intermediary metabolism in diabetic dogs treated with pancreatic autotransplants and insulin pumps.

Diurnal metabolite profiles were studied in pancreatectomized dogs who had received grafts of their own pancreas. The results were compared to similarly diabetic animals who received exogenous insulin pumped intravenously either peripherally or portally. All animals were studied at least 52 weeks after pancreatectomy. Nondiabetic animals were similarly studied as normal controls. Interestingly, all of the diabetic animals were in excellent metabolic control but significant differences from normal in one or more of the metabolic variables existed in each group. Thus, none of the methods of insulin replacement entirely restored to normal the metabolic state of the pancreatectomized dogs. For example, in the autotransplanted animals, the profiles of glucose, lactate, pyruvate, and alanine manifested either abnormal levels and/or abnormal postprandial excursions. The grafts delivered insulin into the peripheral rather than portal circulations. In the peripherally infused animals, the profiles of pyruvate, alanine, and free fatty acids were also somewhat abnormal. In the portally infused animals the profiles only of pyruvate and alanine showed minor abnormalities without clear statistical significance in some respects as well. The other differences observed were statistically significant (P less than 0.05-0.01). These results suggest that the portal route of insulin delivery may be the route of choice if intermediary metabolism is to approximate normal most closely whether exogenous intravenous insulin is replaced by implanted pumps or endogenous insulin is replaced by pancreatic transplants and that none of these futuristic methods of insulin replacement entirely restore to normal the metabolic state of pancreatectomized dogs.