Abstract P116: The Oral Glucose Tolerance Test is Highly Reproducible for the Diagnosis of Diabetes in Africans: The Africans in America Study

Abstract

Introduction: As the diabetes epidemic spreads across Africa, the oral glucose tolerance test (OGTT) has the potential to provide important information on glucose tolerance category for the individual and across the population. However, for both patient care and the collection of prevalence statistics, there needs to be a high degree of certainty that the diagnosis made by a single OGTT is reproducible. There are no rigorous studies on the reproducibility of the OGTT in Africans. Objective: By performing two OGTT in African-born blacks living in the Washington DC area, our goal was to evaluate the reproducibility of: a) the glucose tolerance category b) time to glucose peak, c) shape of glucose curve, d) glucose concentration at 60 minutes. Methods: The participants were 90 African-born blacks (66% male (59/90), age 40±11y (mean±SD), BMI 27.6±4.6 kg/m 2 ), who self-identified as healthy and were enrolled in the Africans in America cohort. Two OGTT were performed 10±8 days apart. At each OGTT plasma samples were taken at -15, 0, 30, 60, 90 and 120 minutes. Diabetes (DM), prediabetes (pre-DM) and normal glucose tolerance (NGT)was defined by ADA guidelines for glucose. The shape of the glucose curve was characterized as monophasic, biphasic or indeterminate. During the OGTT, glucose at 60 minutes ≥ 155 mg/dL was defined as elevated. The κ-statistic was used to evaluate reproducibility (slight 0.00 to 0.20, fair 0.21 to 0.40; substantial 0.61 to 0.80, excellent 0.81 to 1.0). Results: At OGTT-1, DM, pre-DM and NGT occurred in 26% (23/90), 27% (24/90) and 48% (43/90), resp. At OGTT-2, DM, pre-DM and NGT occurred in 21% (19/90), 28% (25/90) and 51% (46/90) resp. Reproducibility for the diagnosis of glucose tolerance status (NGT, pre-DM, or DM) was just substantial (κ=0.66). However, reproducibility for the diagnosis of diabetes (DM or no DM) was excellent (κ=0.88). Furthermore, no one without DM at OGTT-1 was identified as having DM at OGTT-2. The reproducibility of time to glucose peak was only fair (κ=0.32), but 100% of participants with DM had peak glucose after 30 min. Similarly, the shape of the glucose curve had fair reproducibility (κ=0.21), but no one with DM had a biphasic curve. For elevated 60 min glucose, reproducibility was substantial (κ=0.62) and 96% (22/23) of the participants with DM had elevated glucose at 60 min. Conclusion: When duplicate OGTT were performed in Africans, the diagnosis of DM and the morphology of the glucose curves were highly reproducible.