Usefulness of electrocardiographic QRS/T angles with versus without bundle branch blocks to predict heart failure (from the Atherosclerosis Risk in Communities Study).

Repolarization abnormalities in the setting of bundle branch blocks (BBB) are generally ignored. We used Cox regression models to determine hazard ratios (HRs) with 95% confidence intervals (CIs) for incident heart failure (HF) associated with wide spatial and frontal QRS/T angle (upper twenty-fifth percentile of each) in men and women with and without BBB. This analysis included 14,478 participants (54.6% women, 26.4% blacks, 377 [2.6%] with BBB) from the Atherosclerosis Risk in Communities Study who were free of HF at baseline. Using No-BBB with normal spatial QRS/T angle as the reference group, the risk for HF in multivariable adjusted models was increased 51% for No-BBB with wide spatial QRS/T angle (HR 1.51, 95% CI 1.37 to 1.66), 48% for BBB with normal spatial QRS/T angle (HR 1.48, 95% CI 1.17 to 1.88), and the risk for incident HF was increased more than threefold for BBB with wide spatial QRS/T angle (HR 3.37, 95% CI 2.47 to 4.60). The results were consistent across subgroups by gender. Similar results were observed for the frontal plane QRS/T angle. In the pooled BBB group excluding right BBB, a positive T wave in lead aVR and heart rate 70 bpm and higher were also potent predictors of incident HF similar to the QRS/T angles. In conclusion, both BBB and wide QRS/T angles are predictive of HF, and concomitant presence of both carries a much higher risk than for either predictor alone. These findings suggest that repolarization abnormalities in the setting of BBB should not be considered benign or an expected consequence of BBB.