Abstract

Objective: To verify what information from oral glucose tolerance testing (OGTT) independently predicts mortality.

Research Design and Methods: 1401 initially non-diabetic participants from the Baltimore Longitudinal Study of Aging, aged 17-95 years, with one or more OGTT (median=2, range 1-8) with insulin and glucose measurements measured every 20 minutes over 2 hours. Proportional hazard using the longitudinally collected data and Bayesian model averaging were used to examine the association of OGTT measurements individually and grouped with mortality adjusting for covariates.

Results: Participants were followed for a median 20.3 years (range 0.5 – 40 years). The first hour OGTT glucose and insulin levels increased only modestly with age; whereas levels during the second hour increased more than 4% per decade. Individually, the 100- and 120-minute glucose measures and the fasting and 100-minute insulin levels were all independent predictors of mortality. When all measures were considered together, only higher 120-minute glucose was a significant independent risk factor for mortality.

Conclusion: The steeper rise with age of the OGTT 2-hour glucose values and the prognostic primacy of the 120 minute glucose value for mortality is consistent with previous reports and suggests the value of using the OGTT in clinical practice.