Relationship of postload plasma glucose to mortality with 19-yr follow-up. Comparison of one versus two plasma glucose measurements in the Chicago Peoples Gas Company Study.

Abstract

To explore the relationship of one versus two postload plasma glucose measurements to 19-yr mortality in men in the Chicago Peoples Gas Company Study. One-hour postload plasma glucose was measured twice (1.4 yr apart, 1965-1967) for 873 nondiabetic men 34-65 yr of age. Serum cholesterol, blood pressure, height, and weight were measured. Data on smoking habits were obtained. Mortality follow-up was complete through July 1986, 19 yr after second glucose load. In prospective analyses, with each of three glucose values--first, second, and mean of first and second--coronary, cardiovascular, and all-cause age-adjusted mortality rates were higher in quintiles 3, 4, and 5 compared with quintile 1, with a significant linear trend. Multivariate analyses with glycemia as a continuous variable confirmed a positive independent association between plasma glucose and mortality endpoints with first measurement and mean of two values but not with second. With dichotomization of plasma glucose (greater than or equal to 11.2 mM [greater than or equal to 200 mg/dl]) for 30 men hyperglycemic both times, CHD and CVD death rates were significantly higher (odds ratios 2.3-2.7) compared with those for 758 men consistently nonhyperglycemic. In contrast, for those hyperglycemic once only, death rates were not significantly higher. Results of multivariate analyses were consistent with these findings. These data indicate a significant relationship of asymptomatic hyperglycemia on repeat examinations to coronary and cardiovascular mortality independent of other factors measured in the study.