Reduction of Glycemic Potentiation: Sensitive Indicator of β-Cell Loss in Partially Pancreatectomized Dogs

To determine which test of islet function is the most sensitive indicator of subclinical β-cell loss, we studied six conscious dogs before and 1 and 6 wk after removal of the splenic and uncinate lobes [64 ± 2% pancreatectomy (PX)]. To assess hyperglycemic potentiation, acute insulin secretory responses (AIR) to 5 g i.v. arginine were measured at the fasting plasma glucose (FPG) level after PG was clamped at ∼250 mg/dl and after PG was clamped at a maximally potentiating level of 550–650 mg/dl. FPG levels were unaffected by PX (112 ± 4 mg/dl pre-PX vs. 115 ± 5 mg/dl 6 wk after PX, P NS). Similarly, basal insulin levels remained constant after PX (11 ± 2 μU/ml pre-PX vs. 11 ± 1 μU/ml 6 wk after PX, P NS). The AIR to 300 mg/kg i.v. glucose decreased slightly from 42 ± 9 μU/ml pre-PX to 32 ± 5 μU/ml 6 wk after PX (P NS), and thus the β-cell loss was underestimated. In contrast, insulin responses to arginine declined markedly after PX. The AIR to arginine obtained at FPG levels declined from 23 ± 3 μU/ml pre-PX to 13 ± 2 μU/ml 6 wk after PX (P = .04). The AIR to arginine obtained at PG levels of ∼250 mg/dl declined even more, from a pre-PX value of 56 ± 7 μU/ml to 21 ± 4 μU/ml 6 wk after PX (P = .02). However, the largest decline in AIR to arginine occurred at PG levels of 550–650 mg/dl (113 ± 13 μU/ml pre-PX vs. 28 ± 7 μU/ml 6 wk after PX, P = .001), thus indicating a marked decrease of maximal glycemic potentiation. Basal levels of glucose and insulin and AIRs to glucose and arginine obtained 1 wk after PX were similar to those obtained 6 wk after PX. The second purpose of the study was to explore mechanisms by which compensation of this β-cell loss occurs and by which hyperglycemia is avoided. One possibility is a reduction of glucagon secretion, but immunoreactive glucagon levels and glucagon secretory responses to arginine measured at three PG levels remained unchanged after PX. Similarly, tissue sensitivity to insulin, measured with euglycemic clamps at two elevated insulin levels, and insulin clearance remained unchanged after PX. Thus, the mechanism for maintenance of euglycemia after partial PX in the dog remains unclear. In summary, a reduction in glycemic potentiation of the insulin response to arginine was found to be the best means of detecting the subclinical β-cell loss induced by a two-thirds PX in dogs.