Intra-ventricular resynchronization for optimal left ventricular function during pacing in experimental left bundle branch block

Abstract

ObjectivesWe sought to investigate to what extent intra-ventricular asynchrony (intraVA) and inter-ventricular asynchrony (interVA) determine left ventricular (LV) function in canine hearts with left bundle branch block (LBBB) during ventricular pacing. BackgroundPacing therapy improves LV pump function in patients with heart failure and abnormal ventricular conduction supposedly due to resynchronization. However, the relationship between LV pump function and measures of asynchrony is not well established. MethodsIn 15 experiments, LV (various sites) and biventricular (BiV) pacing was performed at atrioventricular (AV) delays of 20 to 140 ms. Measured were the maximum rate of increase (dP/dtmax) of LV pressure and LV stroke work (SW) (conductance catheter), interVA (time delay between the upslope of LV and RV pressures), and intraVA (from endocardial electrical activation maps). ResultsInduction of LBBB increased interVA (−6.4 ± 8.6 to −28.4 ± 8.5 ms [RV earlier]) and intraVA (4.9 ± 2.4 to 18.0 ± 3.3 ms), whereas LV dP/dtmaxand SW decreased (−13 ± 18% and −39 ± 24%, respectively). During LBBB, LV and BiV pacing increased LV dP/dtmaxand SW (mean increases 14% to 21% and 11% to 15%, respectively) without changing diastolic function or preload. Optimal improvement in LV function was obtained consistently when intraVA returned to pre-LBBB values, while interVA remained elevated. Normalization of intraVA required AV delays shorter than the baseline PQ time during LV apex and BiV pacing, thus excluding endogenous LV activation, but AV delays virtually equal to the baseline PQ time (difference 4 ± 9 ms, p = NS) during pacing at (mid)lateral LV sites to obtain fusion between pacing-induced and endogenous activation. ConclusionsIn LBBB hearts, optimal restoration of LV systolic function by pacing requires intra-ventricular resynchronization. The optimal AV delay to achieve this depends on both the site of pacing and baseline PQ time.