Cardiac Resynchronization Therapy Improves Outcomes in Patients With Intraventricular Conduction Delay But Not Right Bundle Branch Block: A Patient-Level Meta-Analysis of Randomized Controlled Trials

Abstract

Background: Benefit from cardiac resynchronization therapy (CRT) varies by QRS characteristics; individual randomized trials are underpowered to assess benefit for relatively small subgroups.

Methods: The authors analyzed patient-level data from pivotal CRT trials (MIRACLE [Multicenter InSync Randomized Clinical Evaluation], MIRACLE-ICD [Multicenter InSync ICD Randomized Clinical Evaluation], MIRACLE-ICD II [Multicenter InSync ICD Randomized Clinical Evaluation II], REVERSE [Resynchronization Reverses Remodeling in Systolic Left Ventricular Dysfunction], RAFT [Resynchronization-Defibrillation for Ambulatory Heart Failure], BLOCK-HF [Biventricular Versus Right Ventricular Pacing in Heart Failure Patients with Atrioventricular Block], COMPANION [Comparison of Medical Therapy, Pacing and Defibrillation in Heart Failure], and MADIT-CRT [Multicenter Automatic Defibrillator Implantation Trial - Cardiac Resynchronization Therapy]) using Bayesian Hierarchical Weibull survival regression models to assess CRT benefit by QRS morphology (left bundle branch block [LBBB], n=4549; right bundle branch block [RBBB], n=691; and intraventricular conduction delay [IVCD], n=1024) and duration (with 150-ms partition). The continuous relationship between QRS duration and CRT benefit was also examined within subgroups defined by QRS morphology. The primary end point was time to heart failure hospitalization (HFH) or death; a secondary end point was time to all-cause death.

Results: Of 6264 patients included, 25% were women, the median age was 66 [interquartile range, 58 to 73] years, and 61% received CRT (with or without an implantable cardioverter defibrillator). CRT was associated with an overall lower risk of HFH or death (hazard ratio [HR], 0.73 [credible interval (CrI), 0.65 to 0.84]), and in subgroups of patients with QRS ≥150 ms and either LBBB (HR, 0.56 [CrI, 0.48 to 0.66]) or IVCD (HR, 0.59 [CrI, 0.39 to 0.89]), but not RBBB (HR 0.97 [CrI, 0.68 to 1.34]; P_interaction <0.001). No significant association for CRT with HFH or death was observed when QRS was <150 ms (regardless of QRS morphology) or in the presence of RBBB. Similar relationships were observed for all-cause death.

Conclusions: CRT is associated with reduced HFH or death in patients with QRS ≥150 ms and LBBB or IVCD, but not for those with RBBB. Aggregating RBBB and IVCD into a single "non-LBBB" category when selecting patients for CRT should be reconsidered.

Registration: URL: https://www.

Clinicaltrials: gov; Unique identifiers: NCT00271154, NCT00251251, NCT00267098, and NCT00180271.

Keywords: bundle-branch block; cardiac resynchronization therapy; defibrillators, implantable; meta-analysis.

Figure 1.

Consort diagram depicting the application of exclusion criteria allowing for creation of the final study cohort.

Figure 2.

Forest plot depicting the adjusted model assessing association between CRT and outcomes (a, HFH or death and b, all-cause death) overall and by trial.

Figure 3.

Forest plots depicting an adjusted model assessing the association between CRT and outcomes (a, HFH or death and b, all-cause death) among subgroups defined by QRS morphology (LBBB, RBBB, IVCD) and duration (≥150 ms or <150 ms).

Figure 4.

Relationship between QRS duration, CRT, and outcomes (a, HFH or death and b, all-cause death) within each subgroup defined by QRS morphology. The black lines depict point estimates and the red lines depict the 95% posterior credible intervals. Y axes depict HR for heart failure hospitalization or death. X axes depict QRS duration. The vertical dotted lines indicate QRS durations at which the 95% CI crosses a HR of 1.0 (neutrality), indicating strong evidence of benefit.

Figure 5.

Relationship between QRS duration and HFH or death within each subgroup defined by sex and QRS morphology. The black lines depict point estimates and the red lines depict the 95% posterior credible intervals. Y axes depict HR for heart failure hospitalization or death. X axes depict QRS duration. The vertical dotted lines indicate QRS durations at which the 95% CI crosses a HR of 1.0 (neutrality), indicating strong evidence of benefit.