The Association of a classical left bundle Branch Block Contraction Pattern by vendor-independent strain echocardiography and outcome after cardiac resynchronization therapy

Abstract

Background: The association of a Classical left bundle branch block (LBBB) contraction pattern and better outcome after cardiac resynchronization therapy (CRT) has only been studied using vendor-specific software for echocardiographic speckle-tracked longitudinal strain analysis. The purpose of this study was to assess whether a Classical LBBB contraction pattern on longitudinal strain analysis using vendor-independent software is associated with clinical outcome in CRT recipients with LBBB. Methods: This was a retrospective cohort study including CRT recipients with LBBB, heart failure, and left ventricular (LV) ejection fraction ≤35%. Speckle-tracked echocardiographic longitudinal strain analysis was performed retrospectively on echocardiograms using vendor-independent software. The presence of a Classical LBBB contraction pattern was determined by consensus of two readers. The primary end point was a composite of time to death, heart transplantation or LV assist device implantation. Secondary outcome was ≥15% reduction in LV end-systolic volume. Intra- and inter-reader agreement of the longitudinal strain contraction pattern was assessed by calculating Cohen's κ. Results: Of 283 included patients, 113 (40%) were women, mean age was 66 ± 11 years, and 136 (48%) had ischemic heart disease. A Classical LBBB contraction pattern was present in 196 (69%). The unadjusted hazard ratio for reaching the primary end point was 1.93 (95% confidence interval, 1.36-2.76, p < 0.001) when comparing patients without to patients with a Classical LBBB contraction pattern. Adjusted for ischemic heart disease and QRS duration < 150 milliseconds the hazard ratio was 1.65 (95% confidence interval, 1.12-2.43, p = 0.01). Of the 123 (43%) patients with a follow-up echocardiogram, 64 of 85 (75%) of patients with a Classical LBBB contraction pattern compared to 13 of 38 (34%) without, had ≥15% reduction in LV end-systolic volume (p < 0.001). Cohen's κ were 0.86 (95% confidence interval, 0.71-1.00) and 0.42 (95% confidence interval, 0.30-0.54) for intra- and inter-reader agreement, respectively. Conclusion: Using vendor-independent strain software, a Classical LBBB contraction pattern is associated with better outcome in CRT recipients with LBBB, but inter-reader agreement for the classification of contraction pattern is only moderate.