The first phase insulin response to intravenous glucose is highly reproducible

To determine the reproducibility of the first phase insulin response to intravenous glucose, ten normal subjects underwent two intravenous glucose tolerance tests separated by at least two weeks. Intravenous dextrose (0.3 g/kg) was administered over 2 min by continuous infusion and arterialised-venous samples were taken from a retrogradely cannulated hand vein in the opposite arm. Within subjects, median coefficient of variation for the 3 min insulin was 4.0% (range 1.2-24.3%) and median coefficient of variation for the 0-10 min area was 6.7% (range 1.7-18.8%). These coefficients of variation are close to those of the assay itself (less than 10%). Despite this, between subject responses varied by greater than sixfold. In conclusion, contrary to previous reports the intravenous glucose tolerance test is highly reproducible. This makes it a very valuable tool for further studies of the pathogenesis of diabetes.