Symptoms, hormones, and glucose fluxes during a gradual hypoglycaemia induced by intraperitoneal vs venous insulin infusion in Type I diabetes.

Abstract

Intraperitoneal (IP) insulin infusion with programmable implantable pumps is associated with a reduction in hypoglycaemic events when compared to intensive diabetes management with subcutaneous insulin in patients with Type 1 diabetes mellitus. The mechanism may involve more physiological insulin kinetics, lower peripheral insulin levels or a specific hepatic action of portal insulin on hypoglycaemic counter regulation. To investigate the latter two hypotheses, we performed two hypoglycaemic clamps (controlled blood glucose decrement to 2.2 mmol l-1) in random order in 12 Type 1 diabetic patients. Insulin was infused either IP or IV for 150 min, at rates chosen to generate similar peripheral insulin levels (1 mU/kg-1 min-1 IV or 2 mU/kg-1 min-1 IP, n = 6) to evaluate direct hepatic action, or at similar rates (1 mU/kg-1 min-1 IV and IP, n = 6) to evaluate IP indirect effects via lower peripheral insulinaemia. Hepatic glucose production and glucose utilization were measured by [6.6 2H] glucose dilution technique. Glucose production was lower (1.7 +/- 0.4 vs 0.5 +/- 0.4 mg kg-1 min-1, p < 0.05), and utilization was similar at the end of the matched-insulinaemia IV and IP clamps, respectively. By contrast, glucose production was higher (1.7 +/- 0.5 IV vs 2.7 +/- 0.3 IP mg kg-1 min-1, p < 0.01) and glucose utilization lower (4.4 +/- 1.0 IV vs 3.3 +/- 0.2 IP mg kg-1 min-1, p < 0.05) with IP delivery at the end of the matched-dose clamps. Counterregulatory hormones and hypoglycaemic symptoms increased similarly in all clamps. In summary, IP insulin, when compared to IV insulin at similar delivery rates, but not at similar insulinaemia, is associated with a less negative glucose balance (glucose production-glucose utilization) during hypoglycaemia. Such a mechanism may play a role in the reduced hypoglycaemic risk seen with IP implantable pumps.