Glucose volume of distribution affects insulin sensitivity measured by intravenous glucose tolerance test

Aim: To determine whether glucose volume of distribution (Vd(GLUCOSE)) affects the diagnosis of impaired insulin sensitivity (IS) when using an intravenous glucose tolerance test (IVGTT). Methods: Individuals with distinct levels of IS underwent IVGTT after an overnight fast. The prediabetic group (Prediab; n = 33) differed from the healthy group (Healthy; n = 14) in their larger glycosylated hemoglobin (HbA1c of 5.9 +/- 0.3 vs. 5.4 +/- 0.1%; 41 +/- 4 vs. 36 +/- 1 mmol/mol; p < 0.001), percent body fat (37 +/- 6 vs. 24 +/- 3%; p < 0.001) and cardiovascular fitness level (VO2MAX 22 +/- 5 vs. 44 +/- 5 mL of O-2kg(-1)min(-1); p < 0.001). Ten minutes after intravenous infusion of the glucose bolus (i.e., 35 g in a 30% solution), Vd(GLUCOSE) was assessed from the increases in plasma glucose concentration. IS was calculated during the next 50 min using the slope of glucose disappearance and the insulin time-response curve. Results: Vd(GLUCOSE) was higher in Healthy than in Prediab (230 +/- 49 vs. 185 +/- 21 mLkg(-1); p < 0.001). Vd(GLUCOSE) was a strong predictor of IS (beta standardized coefficient 0.362; p = 0.004). VO2MAX was associated with Vd(GLUCOSE) and IS (Pearson r = 0.582 and 0.704, respectively; p < 0.001). However, body fat was inversely associated with Vd(GLUCOSE) and IS (r = -0.548 and -0.555, respectively; p < 0.001). Conclusions: Since fat mass is inversely related to Vd(GLUCOSE) and in turn, Vd(GLUCOSE) affects the calculations of IS, the IV glucose bolus dose should be calculated based on fat-free mass rather than body weight for a more accurate diagnosis of impaired IS.