Left Ventricular Enlargement, Cardiac Resynchronization Therapy Efficacy, and Impact of MultiPoint Pacing

Background: Left ventricular (LV) epicardial pacing results in slowly propagating paced wavefronts. We postulated that this effect might limit cardiac resynchronization therapy efficacy in patients with LV enlargement using conventional biventricular pacing with single-site LV pacing, but be mitigated by LV stimulation from 2 widely spaced sites using MultiPoint pacing with wide anatomic separation (MPP-AS: >= 30 mm). We tested this hypothesis in the multicenter randomized MPP investigational device exemption trial. Methods: Following implant, quadripolar biventricular single-site pacing was activated in all patients (n=506). From 3 to 9 months postimplant, among patients with available baseline LV end-diastolic volume (LVEDV) measures, 188 received biventricular single-site pacing and 43 received MPP-AS. Patients were dichotomized by median baseline LVEDV indexed to height (LVEDVIMedian). Outcomes were measured by the clinical composite score (primary efficacy end point), quality of life, LV structural remodeling (up arrow EF >5% and down arrow ESV 10%) and heart failure event/cardiovascular death. Results: LVEDVIMedian was 1.1 mL/cm. Baseline characteristics differed in patients with LVEDVI>Median versus LVEDVI <= Median. Among patients with LVEDVI>Median, biventricular single-site pacing was less efficacious compared to patients with LVEDVI <= Median (clinical composite score, 65% versus 79%). In contrast, MPP-AS programming generated greater clinical composite score response (92% versus 65%, P=0.023) and improved quality of life (-31.0 +/- 29.7 versus -15.7 +/- 22.1, P=0.038) versus biventricular single-site pacing in patients with LVEDVI>Median. Reverse remodeling trended better with MPP-AS programming. In patients with LVEDVI>Median, heart failure event rate increased following the 3-month randomization point with biventricular single-site pacing (0.0150 +/- 0.1725 in LVEDVI>Median versus -0.0190 +/- 0.0808 in LVEDVI (<= Median), P=0.012), but no heart failure event occurred in patients with MPP-AS programming between 3 and 9 months in LVEDVI>Median. All measured outcomes did not differ in patients receiving MPP-AS and biventricular single-site pacing with LVEDVI <= Median. Conclusions: Conventional biventricular single-site pacing, even with a quadripolar lead, has reduced efficacy in patients with LV enlargement. However, the greatest response rate in patients with larger hearts was observed when programmed to MPP-AS pacing.