Letter by Lehmann Regarding Articles “Electrocardiographic Abnormalities That Predict Coronary Heart Disease Events and Mortality in Postmenopausal Women: The Women’s Health Initiative” and “Electrocardiographic Predictors of Incident Congestive Heart Failure and All-Cause Mortality in Postmenopausal Women: the Women’s Health Initiative”
To the Editor:
I read with interest the extensive analyses recently published by Rautaharju et al1 identifying traditional and nontraditional ECG predictors of coronary heart disease (CHD) events and mortality and incident congestive heart failure and all-cause mortality2 in a population of more than 38 000 postmenopausal women (Women’s Health Initiative participants). QT interval was the studied ECG index of repolarization duration, with rate adjustment achieved by using a linear function of the RR interval (QTrr). The authors reported that in their multiadjusted multiple ECG variable models, a prolonged QTrr (≥437 ms) was predictive of combined fatal and nonfatal CHD events among study subjects with preexisting cardiovascular disease (CVD) but not in those free of CVD and was predictive of CHD-related mortality in the CVD-free group. Furthermore, prolonged QTrr was predictive of incident congestive heart failure (regardless of preexisting CVD status) and all-cause mortality (CVD group only). The lower 95% confidence limits for the corresponding statistically significant hazard ratios ranged from 1.02 to 1.25.
Although patients with major ventricular conduction defects (QRS duration ≥120 ms) were excluded from the study, the authors did not include QRS duration among the covariates in their statistical models. This is a potentially important methodologic omission, considering that QRS duration constitutes the initial portion of the QT interval. Indeed, Dr Rautaharju previously reported3 that QRS duration had a modest but highly significant (P<0.001) influence on the QT interval in a regres-sion model that included a rate correction factor. Moreover, from a pathophysiological standpoint, QRS duration (even if <120 ms) is likely to be relatively more prolonged in individuals with varying degrees of myocardial fibrosis and/or left ventricular hypertrophy, a well known epidemiological risk factor for adverse cardiac prognosis. However, as recently reviewed,4 none of the widely cited prior studies investigating rate-corrected QT interval as an ECG predictor of cardiovascular mortality in broad populations has addressed the potential confounding effect of QRS duration. The Women’s Health Initiative database affords Rautaharju et al an excellent opportunity to fill in this persistent methodological gap in the literature, at least as it pertains to women.
Rautahajru PM, Kooperberg C, Larson JC, LaCroix A. Electrocardiographic abnormalities that predict coronary heart disease events and mortality in postmenopausal women: the Women’s Health Initiative. Circulation. 2006; 113: 473–480.
Rautaharju PM, Kooperberg C, Larson JC, LaCroix A. Electrocardiographic predictors of incident congestive heart failure and all-cause mortality in postmenopausal women: the Women’s Health Initiative. Circulation. 2006; 113: 481–489.