The Early Insulin Response During an Oral Glucose Tolerance Test Exhibits Marked Day-To-Day Variability in Free-Living Adults.

The oral glucose tolerance test (OGTT) is often performed in large epidemiology or intervention studies with measures of the early insulin response from this test used to estimate β-cell function. To determine the day-to-day variability of the insulin response and other OGTT measures, we studied 25 healthy adults [10 normal glucose tolerance (NGT), 6 impaired glucose tolerance (IGT) and 9 type 2 diabetes (DM)] with a standard 75 g OGTT performed on two separate days [median 7 days apart (range 5-14 days)]. Subjects were told to follow their same daily routine and to fast overnight. Glucose and insulin were measured at -10, -5, 0, 30, 60, 90. and 120 minutes. Insulin sensitivity was determined by 1/HOMA (22.5/fasting insulin × fasting glucose). The insulin response was determined as the ratio of the change in insulin for the change in glucose over the first 30 minutes (ΔI/ΔG 30 min) and the ratio of the incremental area under the curve from 0-120 minutes for insulin and glucose (incAUCins/incAUCglu). The within subject % coefficient of variation (CV) was computed. Results None of the OGTT variables or weight differed significantly between the study days. Each variable on study day 1 was highly correlated with the same variable on study day 2 (r = .89-.99, p < .001). CVs were not significantly different between glucose tolerance categories. Conclusions The early insulin response (ΔI/ΔG 30 min) during the OGTT shows marked within-subject variability from day to day. Studies that use this as a measure of β-cell function need to take this variability into account when calculating sample size.