SUMMARY Dogs were prepared with Pavlov pouches which could be perfused with saline so as to detect rapidly changes occurring in acid secretion.Dose-response studies were performed by the intravenous injection of insulin (0-06, 0-12, 0.25, 0-5, or 1-0 unit/kg). Both the magnitude of peak response and the duration of response were dose-related and both were related to the magnitude of fall in plasma glucose.These results are not consistent with an 'all-or-none' hypothesis for insulin-stimulated gastric acid secretion, but favour the alternative hypothesis that insulin provides a graded glycopenic stimulus producing a graded secretory response.It has long been held that the gastric secretory response to insulin hypoglycaemia is an 'all-or-none' phenomenon, initiated when the blood sugar falls below a certain threshold value (40-50 mg/100 ml) and giving a peak response whose magnitude is unrelated to the degree of hypoglycaemia. These assumptions have recently been questioned by Baron (1970), who suggested that insulin hypoglycaemia provided a graded glycopenic stimulus producing a graded vagal response. On repeated studies in one human subject Baron found that peak acid output after insulin was significantly correlated with the lowest blood glucose achieved after insulin, or with the fall-or rate of fall of blood glucose. In these studies peak acid output was defined as that in the two consecutive 15-minute samples giving the highest acid output. It is possible, however, that peak acid output in a much shorter period may be the same with all doses of insulin, but that the duration of the secretory response may be greater when hypoglycaemia is prolonged. The 'all-or-none' assumption would in this case be valid, but would appear refuted when long collection periods were used to define acid output. It was to test this hypothesis that the present experiment was carried out. Dogs were prepared with innervated (Pavlov) pouches which were perfused with saline so that changes in acid secretion could be detected rapidly. Dose response studies were then performed to determine the secretory response to insulin.
MethodsTwo mongrel dogs (20.4 and 26-3 kg) were each prepared with a gastric fistula and an innervated fundic pouch (Pavlov). Two Gregory cannulae drained the pouch, at its upper and lower ends (Fig. 1). After a three-week recovery period secretory studies were begun in each dog. These were performed after an overnight fast, during which water but not food was available. The animals stood on tables supported by cloth slings. During each test 154 mM NaCl was infused into a leg vein at a rate just sufficient to maintain patency in a polyethylene cannula which was used to obtain blood samples. In an initial test histamine dihydro-chloride was infused intravenously at a dose of 0-16 mg/kg/hr in order to provide an index of 'maximal' pouch secretion. Pouch secretions were collected as 15-minute samples for two hours. Gastric fistula secretion in this and subsequent studies was collected and discarded.In subsequent studies ...