Aims: Accurate selection of patients with left bundle branch block (LBBB) may help increasing response to cardiac resynchronization therapy (CRT). There is no agreement on LBBB definition. The aim of the study was to investigate the prevalence of 'true-LBBB' according to Strauss in patients undergoing CRT.
Methods and results: The study population included 414 consecutive patients (71.9% men; mean age 69.7 ± 9.6 years), who underwent CRT according to 2010 European Society of Cardiology (ESC) guidelines. Patients were classified into three groups: traditional LBBB according to American Heart Association, LBBB according to Strauss and intraventricular conduction delay (IVCD). Subsequently, they were re-classified into classes of recommendations, according to the current 2013 ESC Guidelines. Traditional LBBB was recorded in 229 patients (55%), an LBBB according to Strauss in 153 (37%) and an IVCD in 32 (8%). Patients with an LBBB according to Strauss showed a significantly more prolonged QRS duration (P < 0.001), greater baseline end-systolic and end-diastolic volumes (P = 0.011 and P = 0.013, respectively) compared with those with IVCD. The prevalence of mid-QRS notching in at least two contiguous leads was 100% in LBBB according to Strauss; 24% in traditional LBBB and 21.9% in IVCD (P < 0.001). At multivariate analysis, PR interval less than 200 ms and QRS of at least 150 ms were independent predictors of mid-QRS notching [odds ratio (OR) 1.78; 95% confidence interval (95% CI) 1.10-2.88; P = 0.02 and OR 2.88; 95% CI 1.80-4.62;P < 0.0001]. Applying stricter criteria for LBBB according to Strauss, a significant reduction in Class I recommendation and an increase in Class II was observed (90.1 vs. 37%; P < 0.0001 and 9.9 vs. 63%; P < 0.0001).
Conclusions: Applying stricter criteria, only 37% of patients undergoing CRT showed a true-LBBB according to Strauss. Accurate identification of true-LBBB may have a potential additional value in better selecting patients.