Predictive value of intravenous glucose tolerance test insulin secretion less than or greater than the first percentile in islet cell antibody positive relatives of type 1 (insulin-dependent) diabetic patients

Abstract

We have followed-up 35 islet cell antibody-positive first degree relatives of patients with Type 1 (insulin-dependent) diabetes mellitus for an average of 1,300 days with sequential intravenous glucose tolerance tests. At the time of analysis and manuscript submission approximately half (18 of 35) had developed diabetes during follow-up. At initial intravenous glucose tolerance test, 11 had a 1 + 3 min insulin secretion below the first percentile of insulin secretion compared to 225 similarly studied normal control subjects. Six islet cell antibody positive relatives on follow-up developed an intravenous glucose tolerance test less than the first percentile. Fifteen out of 17 (88%) of these islet cell antibody positive relatives with secretion ever found to be below the first percentile are now overtly diabetic (positive predictive value = 88%) and insulin-treated, while only 3 of 18 (17%) without an intravenous glucose tolerance test demonstrating loss of first phase insulin secretion have progressed to diabetes (with approximately 1,300 days of follow-up for both groups relative risk or odds ratio with intravenous glucose tolerance test ever below vs never below the first percentile = 38, p less than 0.001). Intravenous glucose tolerance test response below the first percentile preceded diabetes by an average of 656 days. Even when first phase insulin secretion is below the first percentile, the absolute value of 1 + 3 min insulin above basal insulin correlates with the time to development of diabetes (r = 0.586, p less than 0.001). With our current duration of follow-up, the negative predictive value (intravenous glucose tolerance test never below the first percentile) is 83%, and overall accuracy 86%. Incidence rates of diabetes development amongst our islet cell antibody positive relatives with follow-up while intravenous glucose tolerance test is below the first percentile is 0.48 per year (15 conversions to diabetes amongst 17 relatives in 30.8 patient years of follow-up) vs 0.05 per year (three diabetic patients in 55.5 patient years) with intravenous glucose tolerance test greater than the first percentile.