INTRODUCTION

We investigated whether pacing-induced electrical dyssynchrony at the time of cardiac resynchronization therapy (CRT) device implantation was associated with chronic CRT response.

METHODS AND RESULTS

We included a total of 69 consecutive heart failure patients who received a CRT device. Left (LVp-RVs) and right (RVp-LVs) pacing-induced interlead delays were measured intraoperatively and used to determine if there was paced left ventricular (LV) dyssynchrony, defined as present when LVp-RVs is larger than RVp-LVs. CRT response was defined as a reduction in LV end-systolic volume ≥15%, 6 months after implantation. Paced left ventricular dyssynchrony (PLVD) was associated with ischemic cardiomyopathy (ICM) (χ : 8; P = .005) but not with QRS morphology nor with pacing lead positions. In a univariate analysis, PLVD (odds ratio [OR], 6.53; 95% confidence interval [CI], 2.2-18.9; P = .001), atypical left bundle branch block (LBBB) (OR, 3.3; 95% CI, 1.2-9.4; P = .022), and ICM (OR, 5.2; 95% CI, 1.6-17; P = .006) were associated with nonresponse. In a multivariate analysis, both PLVD (OR, 9.74; 95% CI, 2.8-33.9; P < .0001) and atypical LBBB (OR, 5.6; 95% CI, 1.5-20.3; P = .009) were independently associated with nonresponse. Adding PLVD to a model based on QRS morphology provided a significant and meaningful incremental value to predict LV reverse remodeling after CRT (χ to enter: 8; P < .005). Computer simulations corroborate these findings by showing that, while intrinsic electrical dyssynchrony is a prerequisite, the level of pacing-induced dyssynchrony modulates acute CRT response.

CONCLUSION

In addition to the intrinsic electrical substrate, PLVD is strongly associated with less LV reverse remodeling, demonstrating that measuring the electrical substrate during pacing has additional value for prediction of CRT response in an already well-selected patient population.