Bundle-branch block morphology and other predictors of outcome after cardiac resynchronization therapy in Medicare patients.

BACKGROUND: Clinical trials of cardiac resynchronization therapy (CRT) have enrolled a select group of patients, with few patients in subgroups such as right bundle-branch block (RBBB). Analysis of population-based outcomes provides a method to identify real-world predictors of CRT outcomes. METHODS AND
RESULTS: Medicare Implantable Cardioverter-Defibrillator Registry (2005 to 2006) data were merged with patient outcomes data. Cox proportional-hazards models assessed death and death/heart failure hospitalization outcomes in patients with CRT and an implantable cardioverter-defibrillator (CRT-D). The 14 946 registry patients with CRT-D (median follow-up, 40 months) had 1-year, 3-year, and overall mortality rates of 12%, 32%, and 37%, respectively. New York Heart Association class IV heart failure status (1-year hazard ratio [HR], 2.23; 3-year HR, 1.98; P<0.001) and age ≥ 80 years (1-year HR, 1.74; 3-year HR, 1.75; P<0.001) were associated with increased mortality both early and late after CRT-D. RBBB (1-year HR, 1.44; 3-year HR, 1.37; P<0.001) and ischemic cardiomyopathy (1-year HR, 1.39; 3-year HR, 1.44; P<0.001) were the next strongest adjusted predictors of both early and late mortality. RBBB and ischemic cardiomyopathy together had twice the adjusted hazard for death (HR, 1.99; P<0.001) as left BBB and nonischemic cardiomyopathy. QRS duration of at least 150 ms predicted more favorable outcomes in left BBB but had no impact in RBBB. A secondary analysis showed lower hazards for CRT-D compared with standard implantable cardioverter-defibrillators in left BBB compared with RBBB.
CONCLUSIONS: In Medicare patients, RBBB, ischemic cardiomyopathy, New York Heart Association class IV status, and advanced age were powerful adjusted predictors of poor outcome after CRT-D. Real-world mortality rates 3 to 4 years after CRT-D appear higher than previously recognized.