Outcomes of Cardiac Resynchronization Therapy by New York Heart Association Class: A Patient-Level Meta-Analysis

BACKGROUND: Data on the benefits of cardiac resynchronization therapy (CRT) in patients with severe heart failure symptoms are limited. We investigated the relative effects of CRT in patients with ambulatory New York Heart Association (NYHA) IV versus III functional class at the time of device implantation. METHODS AND RESULTS: In this meta-analysis, we pooled patient-level data from the MIRACLE (Multicenter InSync Randomized Clinical Evaluation), MIRACLE-ICD (Multicenter InSync Implantable Cardioversion Defibrillation Randomized Clinical Evaluation), and COMPANION (Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure) trials. Outcomes evaluated were time to the composite end point of the first heart failure hospitalization or all-cause mortality, and time to all-cause mortality alone. The association between CRT and outcomes was evaluated using a Bayesian hierarchical Weibull survival regression model. We assessed if this association differed between NYHA III and IV groups by adding an interaction term between CRT and NYHA class as a random effect. A sensitivity analysis was performed by including data from RAFT (Resynchronization-Defibrillation for Ambulatory Heart Failure). Our pooled analysis included 2309 patients. Overall, CRT was associated with a longer time to heart failure hospitalization or all-cause mortality (adjusted hazard ratio [aHR], 0.79 [95% credible interval [CI], 0.64–0.99]; posterior probability or P=0.044), with a similar association with time to all-cause mortality (aHR, 0.78 [95% CI, 0.59–1.03]; P=0.083). Associations of CRT with outcomes were not significantly different for those in NYHA III and IV classes (ratio of aHR, 0.72 [95% CI, 0.30–1.27]; P=0.23 for heart failure hospitalization/mortality; ratio of aHR, 0.70 [95% CI, 0.35–1.34]; P=0.27 for all-cause mortality alone). The sensitivity analysis, including RAFT data, did not show a significant relative CRT benefit between NYHA III and IV classes. CONCLUSIONS: Overall, there was no significant difference in the association of CRT with either outcome for patients in NYHA functional class III compared with functional class IV.