Scheidt-Nave C, Barrett-Connor E, Wingard D L, Cohn B A, Edelstein S L
Department of Community and Family Medicine, University of California, San Diego, School of Medicine, La Jolla 92093-0607.
Am J Epidemiol. 1991 Mar 15;133(6):565-76. doi: 10.1093/oxfordjournals.aje.a115928.
The relation between fasting plasma glucose and mortality from ischemic heart disease was examined in a population of 3,458 nondiabetics (aged 40-79 years) in Rancho Bernardo, California, who were free of heart disease at baseline and were followed for an average of 14 years, from 1972 to 1987. A linear increase of age-adjusted ischemic heart disease mortality rates with fasting glucose was observed in men, and a threshold relation at 110 mg/100 ml plasma glucose was observed in women. On the basis of Cox proportional hazards analysis, the observed sex differential in the association between plasma glucose and ischemic heart disease mortality proved to be statistically significant and independent of the effects of age, systolic blood pressure, body mass index, plasma cholesterol, plasma triglyceride, and, in women, estrogen use. In sex-specific multivariate models, interaction terms representing a threshold effect at glucose levels greater than or equal to 110 mg/100 ml were statistically significant in women (p = 0.007), but not in men, and interaction between sex and the glucose threshold term was observed in multivariate analysis of men and women combined (p = 0.07). The authors conclude that sex differences in the effect of fasting glycemia on ischemic heart disease mortality among nondiabetics exist and are unexplained. Elucidation may hold a key to the sex difference in heart disease and the relatively greater importance of diabetes and impaired glucose tolerance as risk factors for ischemic heart disease in women compared with men.
