Abstract 15233: Cardiac Resynchronization Therapy is Associated With Increased Mortality in Right Bundle Branch Block Patients
Introduction: The proportion of right bundle branch block (RBBB) patients in cardiac resynchronization therapy (CRT) trials is limited, and the effect of CRT in RBBB has not been fully established. In RBBB (as opposed to LBBB) the His-Purkinje system rapidly activates left ventricular (LV) endocardium. Overdriving normal LV activation with epicardial LV pacing may be inefficient/harmful in RBBB.
Hypothesis: CRT-defibrillators (CRT-D) are associated with increased mortality in RBBB patients, but reduced mortality in LBBB patients compared to implantable defibrillators (ICD) alone.
Methods: We analyzed Medicare records from CRT-D and ICD recipients from 2002-2009 that were followed up for up to 48 months. To reduce treatment selection bias, ICD patients were propensity score matched to CRT-D patients based on 17 variables including age, sex, diabetes, hypertension, ischemic cardiomyopathy, atrial fibrillation, Charlson score and prior heart failure (HF) hospitalization. In RBBB, 1:1 matching resulted in 7,298 CRT-D and 7,298 ICD patients, and for LBBB 1:1 matching with replacement resulted in 54,218 CRT-D and 20,763 ICD patients. Adjusted hazard ratios (HRs) were calculated to compare CRT-D to ICD separately for RBBB and LBBB.
Results: In RBBB, CRT-D (compared to ICD) had a 13% higher death risk (HR: 1.13 [95% confidence interval: 1.07-1.18]) while in LBBB, CRT-D had a 5% lower death risk (HR: 0.95 [0.92-0.97]) (Figure). For HF hospitalization or death, RBBB CRT-D patients had a 15% higher risk than ICD (HR: 1.15 [1.10-1.20]) while LBBB CRT-D patients had a 12% lower risk (HR: 0.88 [0.86-0.90]).
Conclusions: In a large Medicare population, CRT was associated with increased mortality in RBBB, but decreased mortality in LBBB. These findings should be interpreted with caution as propensity scores can be influenced by unmeasured confounders (e.g. ejection fraction, NYHA HF class, QRS duration). Additional research on CRT in RBBB should include these covariates.
- © 2013 by American Heart Association, Inc.