BackgroundCardiac resynchronization therapy (CRT) response relies on two factors: when and where to pace. These factors may be enhanced by dynamic atrioventricular delays (AVDs) (e.g., SyncAV CRT, Abbott) and multi-site left ventricular (LV) pacing (e.g., MultiPoint Pacing [MPP], Abbott). Their individual and combined synchronization contributions have not been evaluated across a comprehensive spectrum of pacing configurations. The objective is to distinguish the acute electrical synchrony achieved by (i) static vs. dynamic AVDs, (ii) single- vs. multi-site LV pacing, and (iii) with vs. without RV pacing. MethodsCRT-indicated patients with LBBB and intact AV conduction (PR<250ms) were enrolled and evaluated during implant. Acute changes in 12-lead ECG QRS duration (QRSd) were evaluated during: intrinsic conduction, biventricular pacing (BiV), biventricular MPP, LV-only single-site pacing (LVSS), and LV-only MPP (LVMPP). CRT modes were evaluated with static AVDs and optimized SyncAV AVDs. ResultsCRT implant and QRSd evaluation were completed in 85 patients (71% male, 34% ischemic, 179ms PR). The median intrinsic QRSd of 165ms was reduced by BiV, MPP, LVSS, and LVMPP without SyncAV to 144ms (by 14%), 142ms (16%), 155ms (8%), and 149ms (12%), respectively (P<0.01 vs. intrinsic). BiV+SyncAV, MPP+SyncAV, LVSS+SyncAV, and LVMPP+SyncAV reduced the intrinsic QRSd significantly further to 128ms (by 23%), 124ms (26%), 131ms (21%), and 129ms (24%) (P<0.0001, each corresponding pair). ConclusionsMPP combined with SyncAV achieved the narrowest QRSd, in the overall population and in the most patients, by delivering ventricular pacing from all available sites (RV+LV1+LV2) while timed with dynamic AVDs.