Postimplant QRS narrowing may predict clinical response after cardiac resynchronization therapy (CRT), but identification of nonresponders remains difficult. We studied the predictive value of electrocardiographic characteristics for mortality or cardiac transplantation in patients after CRT. Patients who had electrocardiograms available for review from before and after CRT device implantation were identified from a clinical database. Bivariate and multivariate Cox regression analyses were performed for the end point of death or transplantation. Of 337 patients (age 65+/-12 years, 76% men, left ventricular ejection fraction 22+/-12%, pre-QRS 175+/-30 ms), 84 died and 7 underwent transplantation during a follow-up of 27+/-15 months. Variables predictive of death or transplantation included QRS increase after CRT (45% vs 32%, p=0.03), older age, higher New York Heart Association class, lower left ventricular ejection fraction, and higher tertile of postimplant QRS (p=0.04), but not preimplant rhythm, QRS duration, or QRS morphology. After adjusting for confounding variables, independent predictors of mortality were older age (hazard ratio [HR] 1.03, 95% confidence interval [CI] 1.00 to 1.05, p=0.04), lack of angiotensin-converting enzyme inhibitor or angiotensin receptor blocker (HR 2.17, 95% CI 1.16 to 4.08, p<0.02), and longer postimplant QRS by tertile (HR 1.50, 95% CI 1.09 to 2.05, p=0.01). In conclusion, wider QRS after CRT device implantation is an independent predictor of mortality or transplantation. In patients with increased QRS durations despite CRT, closer follow-up or reassessment for alternative management strategies may be warranted.