par Balasse, Edmond
Référence Diabète & métabolisme, 2, 2, page (87-94)
Publication Publié, 1976
Article révisé par les pairs
Résumé : Diabetic hyperketonemia results primarily from an excessive production of ketone bodies by the liver. Two factors ought to be present in order to induce hyperketogenesis: increased amounts of free fatty acids should be delivered to the liver from adipose tissue; and the liver should be in a ketogenic state. These two conditions are present in diabetic ketoacidosis owing to the simultaneous existence of insulinopenia and hyperglucagonemia. Under the influence of a low insulin/glucagon ratio, adipose tissue lipolysis is enhanced and intrahepatic fatty acid metabolism is switched towards β oxidation and ketogenesis, away from triglyceride synthesis. Thus, the bi hormonal insulin glucagon system is implicated in the regulation of ketone metabolism as it is in glucose homeostasis. Another recent step in our knowledge on ketone metabolism is the in vitro and in vivo demonstration of a reduced capacity of certain extrahepatic tissues to utilize ketones in diabetic ketosis. This defect probably results from insulin lack. Thus, both increased production and relative underutilization of ketones contribute to diabetic hyperketonemia. Finally, it has recently been suggested that several homeostatic feed back mechanisms involving ketone bodies could be operative in diabetic ketosis as they are in starvation ketosis. For instance, ketone bodies could inhibit mobilization of gluconeogenic aminoacids and thus contribute to protein sparing; according to their antilipolytic properties, ketones could also tend to restrain the development of diabetic ketosis. The classical treatment of diabetic coma with high doses of insulin has been recently questioned by several authors who showed that small doses (5-10 U/hr) can be equally effective. Another controversy has emerged regarding the mode of administration of insulin. Arguments are developed in this review in favor of the following therapeutic regimen: insulin should not be given by the i.m. route; insulin should be given i.v. either as a constant infusion or, more simply, as repeated i.v. boluses; and the dose of insulin should not be lower than 15-20 U/hr owing to the possible resistance to smaller doses, especially in infected patients. The possible mechanisms of cerebral edema occasionally seen during treatment of diabetic coma are discussed, and the factors which can be held responsible for an osmotic dysequilibrium between brain and plasma are analyzed. Finally, regarding the bicarbonate controversy, we emphasize the fact that possible drawbacks of alcali administration (cerebrospinal fluid acidosis, cerebral anoxia, hypokaliemic action) should not preclude its use in severe acidosis, which can markedly depress myocardial and respiratory functions.