Progression of hyperglycemia is mediated by loss of β-cell function, but we lack a marker which would be sufficiently accurate and convenient to use in large prospective trials. To determine if the 1 hour OGTT plasma glucose (1hrOGTT) could serve such a purpose, we compared it to standard measures of the disposition index (DI, insulin secretion*insulin action) in 584 European adults without known diabetes in the Relationship between Insulin Sensitivity and Cardiovascular Disease Risk (RISC) study dataset, to predict progression over 3 years. Measures included DImodeling (OGTT modeling β-cell glucose sensitivity*insulin sensitivity), DI-ISI ([0-30min ΔC-peptide/Δglucose]*[1/fasting insulin]), DIclamp (acute C-peptide response to glucose*euglycemic clamp glucose disposal [M/I]), and DIhybrid (OGTT modeling β-cell glucose sensitivity*M/I). Using ADA OGTT criteria, for predicting progression from normal glucose to any dysglycemia (diabetes [DM, FPG >126 or 2hrPG >200 mg/dl] or prediabetes [IFG, FPG 100-125, or IGT, 2hrPG 140-199 mg/dl], n=115), the area under receiver operating characteristic curves (ROC) for the 1hrOGTT was 0.637 vs. 0.607, 0.602, 0.565, and 0.582 for DImodeling, DI-ISI, DIclamp, and DIhybrid, respectively. For non-high risk progressing to high risk dysglycemia ([DM] or [IFG + IGT], n=40), the 1hrOGTT ROC was 0.790 vs. 0.738, 0.737, 0.647, and 0.729 for the DI indices, respectively, and for non-DM progressing to DM (n=6), the 1hrOGTT ROC was 0.910, vs. 0.899, 0.822, 0.741, and 0.837 for the DI indices, respectively.
Conclusions: The 1hrOGTT plasma glucose level appears to be comparable or superior to the DI based on more labor-intensive and costly direct assessments of insulin secretion and action in predicting progression of hyperglycemia over 3 years. Consideration should be given to use of the 1hrOGTT glucose as both an accurate and convenient indirect measure of β-cell function in clinical trials, and possibly also of the risk of development and progression of hyperglycemia in clinical practice.
B.T. Legvold: None. A. Zhang: None. M.K. Rhee: Research Support; Self; Boehringer Ingelheim Pharmaceuticals, Inc. L.R. Staimez: None. K. Utzschneider: Other Relationship; Self; Medtronic. L.P. Savoye: None. J.R. Petrie: Advisory Panel; Self; Novo Nordisk A/S. Consultant; Self; Advanced Clinical Intelligence, IQVIA. Research Support; Self; AstraZeneca, Janssen Pharmaceuticals, Inc. Speaker’s Bureau; Self; Merck KGaA, Novo Nordisk A/S. B. Balkau: None. L.S. Phillips: Research Support; Self; AbbVie Inc., Diasome Pharmaceuticals, Inc., Eli Lilly and Company, GlaxoSmithKline plc., Novartis Pharmaceuticals Corporation, Novo Nordisk Inc., Pfizer Inc., Sanofi-Aventis. Other Relationship; Self; Diasyst, Janssen Pharmaceuticals, Inc.
U.S. Department of Veterans Affairs (CX001025, BX003340, CX001737); National Institutes of Health (DK099716, DK091958, DK098246, DK111024, AI133172)