Left ventricular global longitudinal strain and mechanical dispersion predict response to multipoint pacing for cardiac resynchronization therapy

Abstract

To evaluate the acute effect of multipoint pacing (MPP) and search for the better baseline predictors of response to MPP for guiding patient selection. We enrolled 46 heart failure patients scheduled for implantation of MPP-enabled cardiac resynchronization therapy (CRT) devices. An acute pacing protocol including conventional CRT and MPP pacing configurations was performed after implantation. Echocardiography was used at baseline and during pacing test, and response was defined as left ventricular (LV) end-systolic volume (ESV) reduction ≥ 15% at 6-month follow-up. MPP response was present in 32 (69%) patients. Responders showed significantly superior LV ejection fraction, global longitudinal peak strain (GLPS), and mechanical dispersion (MD) with MPP than with conventional CRT (all P < 0.05). Baseline GLPS (OR 1.524; 95% CI 1.031-2.251; P = 0.034) and MD (OR 1.048; 95% CI 1.016-1.081; P = 0.003) were independent predictors of MPP response in multivariate regression analysis. Both |GLPS| and MD were significantly correlated with percentage change in LVESV (▵LVESV%) at 6-month follow-up (r = 0.731 and r = 0.696, respectively; all P < 0.001). |GLPS| ≥ 5.0% combined with MD ≥ 120 ms predicted MPP response with the optimal sensitivity of 91% and specificity of 71% (AUC = 0.848, P < 0.001). MPP tends to be superior to conventional CRT in improving acute response. |GLPS| and MD can successfully predict response to MPP, and their combination can further improve the prediction accuracy of response.