Sipahi et al reported in their meta-analysis of 5 cardiac resynchronization therapy (CRT) trials that QRS duration was an important predictor of response to CRT.1 They concluded that patients with a QRS of 150 milliseconds (ms) or greater had a reduction in heart failure events, whereas those with a QRS less than 150 ms did not. Unfortunately, there was no analysis of QRS morphology, eg, type of bundle branch block. Although they proposed a future meta-analysis to further refine QRS duration cutoffs, they failed to mention that QRS morphology may be equally or more important than QRS duration.