Superior mesenteric artery blood flow increases significantly after hypoglycaemia in healthy humans. Glucagon has vasoactive properties but its role in hypoglycaemic hyperaemia is unclear. To assess this role, we studied the superior mesenteric artery blood flow response to hypoglycaemia of patients with uncomplicated Type 1 (insulin-dependent) diabetes mellitus of at least 10 years duration; a group known to have defective glucagon response to hypoglycaemia. Hypoglycaemia was induced using an intravenous infusion of soluble human insulin (2.5 m-unitsċmin -1 ċkg -1 ) discontinued at a plasma glucose of 2.5 mmol/l. Superior mesenteric artery blood flow was measured using transcutaneous duplex Doppler ultrasound. Plasma samples were assayed for glucose, insulin, glucagon, catecholamines, growth hormone and cortisol. Plasma glucose concentration fell to a nadir of 1.8 (0.3) mmol/l in patients and 1.4 (0.1) mmol/l in controls. Plasma glucagon concentration was unchanged in patients from a baseline level of 111.7 (13.1) ng/l but rose in controls from 105 (8.5) to a peak of 239 (3.1) ng/l ( P < 0.001). Superior mesenteric artery blood flow increased in both groups: from 385 (29) to 921 (100) ml/min (140% increase; P < 0.05) in patients and from 517 (50) to 790 (67) (53% increase; P < 0.001) in controls. This study shows that patients with Type 1 diabetes have a normal splanchnic vascular hyperaemic response to hypoglycaemia despite defective glucagon counter-regulation. These results support our previous work suggesting that glucagon is not a major mediator of this response; it seems likely that circulating adrenaline is the major regulatory mechanism.

1. Previous studies have suggested that glucagon in supraphysiological doses may mediate postprandial and hypoglycaemia-induced splanchnic vasodilatation in man and experimental animals. There are no reported studies investigating the role of glucagon in doses producing circulating concentrations within the physiological range. 2. Two separate studies were performed. In study 1, superior mesenteric artery blood flow was measured by Doppler ultrasound in six normal subjects during either saline or glucagon infusion at 1, 3 and 6 ng min −1 kg −1 , which resulted in circulating glucagon levels within the physiological range. Mean superior mesenteric artery blood flow fell during the 3 and 6 ng min −1 kg −1 glucagon infusions (3 ng min −1 kg −1 : −31.8%, range −20 to −56% of baseline; 6 ng min −1 kg −1 : −20.7%, range −8 to −53% of baseline; P < 0.05). 3. In study 2, superior mesenteric artery blood flow was measured during hypoglycaemia induced by an insulin infusion in 12 normal subjects. In six of these subjects the effect of suppression of glucagon release during hypoglycaemia was assessed by preteatment with the somatostatin analogue octreotide (0.8 μg/kg subcutaneously) given 30 min before the insulin infusion. 4. The nadir in blood glucose concentration at the hypoglycaemic reaction was similar in both groups and glucose recovery was complete by 60 min after the hypoglycaemic reaction. Plasma catecholamine concentrations rose in both groups after the hypoglycaemic reaction. 5. Superior mesenteric artery blood flow rose at the hypoglycaemic reaction in both groups despite suppression of glucagon release with octreotide. 6. These data suggest that at physiological concentrations glucagon does not mediate splanchnic vasodilatation in man.

1. Superior mesenteric artery blood flow was examined by Doppler ultrasound in six male subjects aged 19–23 years during the infusion of saline (control), 10 and 40 ng of adrenaline min −1 kg −1 for 30 min, or propranolol and 10 ng of adrenaline min −1 kg −1 for 30 min, on four separate occasions. 2. Adrenaline infusion resulted in significant peak mean (SEM) rises in circulating adrenaline concentrations during the infusion period only [control, 0.20 (0.05) nmol/l; 10 ng of adrenaline min −1 kg −1 , 1.37 (0.29) nmol/l; 40 ng of adrenaline min −1 kg −1 , 3.73 (0.40) nmol/l; 10 ng of adrenaline min −1 kg −1 and propranolol, 1.48 (0.16) nmol/l, P < 0.001 versus control]. These values are within the physiological range. 3. Superior mesenteric artery blood flow rose in a dose-dependent manner during the adrenaline infusions alone, but not during the infusion of adrenaline and propranolol [mean (95% confidence interval) area under the curve: control, −4.2 (−11 to +2.7)%; 10 ng of adrenaline min −1 kg −1 , +4 (−1 to 11.9)%; 40 ng of adrenaline min −1 kg −1 , +34 (+ 6.5 to + 61.5)%; 10 ng of adrenaline min −1 kg −1 and propranolol, −8.4 (−23 to +6)%]. 4. Superior mesenteric artery resistance fell during the adrenaline infusions alone and rose during the combined adrenaline and propranolol infusion [mean (SEM) area under the curve: control, 6.4 (2.7)%; 10 ng of adrenaline min −1 kg −1 , −2.9 (2.5)%; 40 ng of adrenaline min −1 kg −1 , −15 (1.4)%; 10 ng of adrenaline min −1 kg −1 and propranolol, 16.9 (10)%]. 5. These data suggest that splanchnic vasodilatation is mediated via a β-adrenergic mechanism.

1. Splanchnic haemodynamic changes were studied in seven healthy subjects during hypoglycaemia induced by the intravenous infusion of insulin. Superior mesenteric artery blood flow and cardiac output were examined non-invasively by a Doppler ultrasound technique. 2. Blood glucose concentration fell from 4.5 (0.14) mmol/l basally to 1.5 (0.09) mmol/l [mean ( sem ), P < 0.003] at the hypoglycaemic reaction (‘R’) and recovered to baseline by ‘R’ + 60 min. There was an associated rise in plasma glucagon, adrenaline and noradrenaline levels. 3. Superior mesenteric artery blood flow rose at ‘R’ from a basal value of 532 (38) ml/min to a peak of 803 (73) ml/min at ‘R’+10 min [mean ( sem ), P < 0.005] and remained significantly elevated until ‘R’ + 40 min. Resistance in this vessel fell by 33% at ‘R’+ 10 min ( P < 0.005) and remained significantly low until ‘R’ + 40 min. 4. Cardiac output rose by 33% at ‘R’ ( P < 0.004) and returned to normal by ‘R’ + 20 min. This was associated with a 24% rise in pulse rate ( P 0.03), but no change in stroke volume or mean arterial pressure. Total peripheral resistance fell by 21% at ‘R’ ( P 0.005) and had returned to normal by ‘R’ + 20 min. 5. The sustained rise in splanchnic blood flow during hypoglycaemic recovery may be of homoeostatic importance by providing metabolic fuel to the liver for gluconeogenesis.

1. The effects of the subcutaneous administration of a long-acting somatostatin analogue (octreotide) or of placebo on the splanchnic blood flow response to a mixed solid meal has been examined in eight normal subjects by using a transcutaneous Doppler ultrasound technique. Each subject was studied on two occasions more than 1 week apart. 2. On the control day, feeding had a pronounced effect on both superior mesenteric artery and portal venous blood flows, causing a peak rise of 82% in superior mesenteric artery blood flow at 15 min and of 75% in portal venous blood flow at 30 min post-prandially ( P < 0.001). Blood flows remained elevated 2 h after the meal. Pulse and blood pressure showed no significant changes from baseline. 3. Octreotide reduced fasting superior mesenteric artery blood flow by 59% ( P < 0.05) and portal venous blood flow by 49% ( P < 0.01) and blunted the normal post-prandial rise. Pulse and blood pressure did not change in response to either the injection or the ingestion of the meal. 4. Octreotide suppressed the release of insulin, glucagon and pancreatic polypeptide in response to feeding and resulted in post-prandial hyperglycaemia. 5. The mechanism of action of octreotide on splanchnic blood flow is uncertain. It may be mediated via a direct vascular effect or it may act via suppression of vasoactive intestinal hormones.