Comparison of four LBBB definitions for predicting mortality in patients receiving cardiac resynchronization therapy

Background: Left bundle branch block (LBBB) is considered an important prognostic parameter in cardiac resynchronization therapy (CRT). We aimed to evaluate, in a sizeable cohort of patients with CRT, long-term mortality, and morbidity according to four different electrocardiographic definitions of LBBB.

Methods: This longitudinal cohort study included consecutive patients who underwent CRT device implantation in our institution in years 2006-2014. Two endpoints were assessed: (a) death from any cause or urgent heart transplantation, and (b) death from any cause or heart failure admission. All preimplantation ECGs were analyzed by three physicians blinded to outcome and categorized as LBBB or non-LBBB according to four definitions.

Results: A total of 552 CRT patients entered survival analysis. According to the conventional definition, 350 (63.4%) patients had LBBB, and the Marriott, WHO/AHA, and Strauss definitions identified LBBB in 254 (46.0%), 218 (39.5%) and 226 (40.9%) patients, respectively. During the 9 years of observation, 232 patients died, the combined endpoint was met by 292 patients. The Strauss LBBB definition was significantly better to the other definitions in predicting survival (Kaplan-Meier analysis with comparison of C-statistics). Multivariate Cox regression model showed that LBBB was the major determinant of all-cause mortality with the Strauss definition having the lowest hazard ratio (0.51) of the four studied definitions.

Conclusions: Criteria included in various definitions of LBBB result in a diagnosis of LBBB in divergent groups of patients. Differences in LBBB definitions have clinical consequences, as patients without 'complete/true' LBBB probably get no mortality benefit from CRT.