Rick Schreurs

107 Left and right ventricular response to CRT INTRODUCTION Cardiac resynchronization therapy (CRT) is an established therapy for heart failure patients with a reduced left ventricular (LV) ejection fraction and left bundle branch block (LBBB) [1]. Through biventricular pacing, CRT aims to establish a more synchronous electrical activation of the ventricles and thereby improves cardiac pump function [2] and clinical outcome [1]. However, approximately one-third of the patients that receive CRT do not benefit from this therapy [3–6]. Programming of both atrioventricular (AV) and ventriculoventricular (VV) pacing delays strongly influences the contractile response to CRT, as determined by both ultrasound and maximum rate of LV pressure rise (LV dP/dt max ) [7–9]. However, meta-analyses of multiple optimization methods showed that pacing delay optimization fails to provide long-term improvement in clinical outcome [7]. Suggested reasons for the absence of long-term benefits of such optimization are that the default “out-of-the-box” delays are already fairly good and that the optimization methods employed are not accurate or robust enough. An alternative explanation could be that most pacing delay optimization strategies that have been developed solely take LV function into account. Right ventricular (RV) function is often overlooked, although several studies show that RV failure is an independent predictor of mortality in patients with LV failure with and without CRT [10,11]. Two clinical studies indicated that there was a poor correlation between the pacing delay settings providing the highest LV dP/dt max and RV dP/dt max values [12,13]. In recent years, computational models of cardiac electrophysiology, mechanics of the heart and cardiovascular system have increased our understanding of dyssynchronous heart failure and its treatment with CRT [14]. Right ventricular function and its effect on CRT has however not been studied extensively using a computer modeling approach. Our group has been using the CircAdapt model of the heart and closed-loop circulatory system. While using a simplified cardiac anatomy, the advantages of this model are the inclusion of the entire (systemic and pulmonary) circulation and its high calculation speed (almost real time). In combination with experimental and clinical measurements, CircAdapt has shown to be able to identify and mechanistically understand the electromechanical substrates of the heart that are most responsive to CRT [15–17]. A vast majority of the studies on the heart, also our aforementioned studies, are related largely to the LV. In the present study, we aim to study the changes in both LV and RV contractile response to variations of pacing delay settings in CRT, and to evaluate whether the CircAdapt computer model reliably simulates LV and RV pump function during these interventions. Subsequently we aim to use the computer modeling results to investigate how cardiac output is affected by differences between LV and RV contractile changes. 6