Abstract
Little is known about electrocardiographic (ECG) characteristics of menopausal women with or at increased risk of coronary heart disease (CHD). Data from 10,101 participants in the Raloxifene Use for The Heart (RUTH) trial were used to correlate baseline ECG abnormalities with clinical characteristics. Baseline characteristics that were statistically significantly associated (p ≤ 0.05) with ECG findings in univariate analyses were used to derive multivariate model selection. Of 59% normal electrocardiograms, 50% were from women with CHD and 69% from women at increased risk of CHD. In the women with CHD, 59% reported a previous myocardial infarction (MI); 43% had a normal electrocardiogram, and 49% had a definite ECG Q-wave MI. Women in the increased-risk group had not reported a previous MI, yet 11% had a definite ECG Q-wave MI. Of women reporting hypertension, 35% had ECG evidence of left ventricular hypertrophy, but 58% did not have an abnormal electrocardiogram. Significantly more women with diabetes in the increased-risk and documented CHD cohorts had abnormal electrocardiograms (p < 0.01 for the 2 cohorts). Percent abnormal electrocardiograms increased with increasing age (55 to 64, 65 to 74, and ≥ 75 years, p < 0.01) in all cohorts. Angina and coronary artery bypass graft surgery, but not percutaneous coronary intervention, predicted an abnormal electrocardiogram. In conclusion, there were high percentages of normal electrocardiograms in the increased-risk and documented CHD groups of RUTH participants, with substantial discrepancy between MI history and ECG MI documentation, and increasing age was the predominant correlate with an abnormal electrocardiogram in all 3 cohorts.