The hyperglycemic property of certain sulfonamides was first discovered by one of us (LoubatiBres, 1944(LoubatiBres, , 1946) during a study of relationships between the chemical structures and hypoglycemic action of thiodiazol derivatives.The author found that the sulfonamides either increased, decreased, or maintained the blood sugar constant (Loubatikres, 1946, p. 63-64).Diazoxide* (7-Chloro-3-methyl-2H-1,2,4-benzothiadiazine-1, 1 Dioxide) (FIG-URE 1) is a sulfonamide whose sulfonamide radical is included in a ring. It is a hypotensive substance that does not have any diuretic action. Initial clinical tests on hypertensive patients revealed a hyperglycemic action, especially in patients predisposed to diabetes.Let us consider the animal experiments in which the hyperglycemic action of diazoxide was shown to be due mainly to an excess of catecholamines in the blood co-workers, 1964, 1965). In these studies, they had to know if the excess of catecholamines was due to corrective reflex mechanisms that follow hypotension produced by the drug (essentially the secretion of epinephrine or norepinephrine by the adrenal medulla) ;.or if it was due to a direct stimulating action of the drug on the secretion of catecholamines by the adrenal medulla. We can also ask ourselves whether diazoxide is, in any way, responsible for an eventual depletion of tissue catecholamines.The experiments in which adrenergic blocking drugs (Tabachnick and associates, 1964, 1965; Kvam and Stanton, 1964) were found capable of reducing the hyperglycemic action of diazoxide favor the hypothesis of the intervention of catecholamines in the action of diazoxide, but cannot be considered as irrefutable proof.We also consider unconvincing the experiments supporting the hypothesis that the hyperglycemic action of diazoxide may be due to the reduction of the secretory rhythm of insulin.To demonstrate this phenomenon in vitro, the authors worked on pancreas fragments of the rat, rabbit, and miniature pig, incubated in an oxygenated medium (Frerichs and Creutzfeldt, 1965; Howell and Taylor, 1966). However, there is no doubt that such an experimental method, while valuable, only permits a study of the drug's effect on the most superficial islets of Langerhans of the pancreatic tissue. Therefore, it appeared to be more convenient and physiological to operate on the intact pancreas of the rat, perfused by its own arteriovenous and capillary system, measuring the variations of insulin in the postpancreatic portal vein blood.With the idea of showing in vivo an eventual slowing down of insulin secretion by diazoxide, the authors looked, above all, for the amount of insulin in the peripheral blood.We considered it more convincing to operate on the anesthetized dog and to take blood directly from the pancreaticoduodenal vein in order to detect varia-The diazoxide used in our experiments was kindly supplied to our laboratory by Schering Corporation.
226
