Resting electrocardiographic abnormalities suggestive of asymptomatic ischemic heart disease associated with non-insulin-dependent diabetes mellitus in a defined population.

Abstract

The prevalence of ischemic heart disease (IHD) in older adults by glucose tolerance status was evaluated in 2,223 white men and women, aged 50-89 years, in the Rancho Bernardo cohort who were studied between 1984 and 1987. Impaired glucose tolerance (IGT) and non-insulin-dependent diabetes mellitus (NIDDM) were classified according to World Health Organization criteria. End points of ischemic heart disease were defined by Rose Questionnaire and resting electrocardiogram (ECG) according to the Minnesota Code. IHD by electrocardiographic changes was classified as asymptomatic (without history of chest pain or overt IHD) or symptomatic (with history). IHD by all criteria combined was significantly more common in men and women with NIDDM, and in women with IGT, than in those with normal glucose tolerance. The prevalence of myocardial infarction, defined by major Q wave, Rose Questionnaire chest pain criteria, or personal history, was higher in persons with NIDDM than in persons without; the difference was highly significant in women (odds ratio, 2.08 [1.22, 3.56]; p = 0.009). Angina pectoris was not significantly related to NIDDM or IGT in either sex. Electrocardiographic evidence of asymptomatic IHD was significantly more prevalent in both men and women with NIDDM as compared with those with normal glucose tolerance (odds ratios, 1.75 [1.10, 2.81] for men and 1.80 [1.07, 3.01] for women; p less than 0.05). This significant association persisted after excluding persons on digitlis or diuretic therapy and, in women, was also independent of the effect of major known IHD risk factors. These population-based data are consistent with clinical reports suggesting an association of diabetes with silent myocardial infarction or ischemia. The presence of ischemic resting electrocardiographic abnormalities in the asymptomatic diabetic patient is likely to have prognostic and therapeutic implications.