Cardiac resynchronization therapy (CRT) is an established therapy for patients with heart failure and left ventricular (LV) conduction delay, characterized by a left bundle branch block. Unfortunately, a considerable amount of patients who are eligible according to recent guidelines, do not benefit substantially from CRT. There a multiple reasons for CRT non-response, which may be categorized in patient selection, device implantation and device optimization. This thesis focusses on al three categories.
Firstly, the selection of patients with parameters of mechanical dyssynchrony and discoordination, derived by different strain imaging techniques was evaluated. Several strain imaging techniques, based on speckle tracking echocardiography or cardiac magnetic resonance (CMR) imaging (tagging and feature tracking) are capable of detecting myocardial dyssynchrony. Overall, basic strain parameters were comparable between techniques, although there were significant differences between more complex parameters of mechanical dyssynchrony and discoordination. Two parameters, based on deformation of the interventricular septum, were rather consistent between techniques and are promising for further evaluation. These parameters were categorization of patterns of septal formation and the strain value of the septum at the end of systole.
Secondly, the influence of right ventricular (RV) function on mechanical dyssynchrony and discoordination, as well as on CRT response was evaluated using computer models combined with patient data. Computer models of the cardiovascular system improve our understanding of complex interactions in cardiac (patho)physiology. Although RV dysfunction had relative small influence on mechanical dyssynchrony and discoordination, the influence on CRT response was significant. Despite a considerable amount of LV dyssynchrony, patients with RV dysfunction may not benefit from CRT. RV function should therefore be evaluated before prediction of CRT with markers of mechanical dyssynchrony and discoordination.
Thirdly, the use of quadripolar LV leads for optimization of CRT device implantation and optimization was reviewed. These novel leads increase the number of pacing sites compared to standard bipolar leads, thereby avoiding LV pacing in regions of scar tissue or phrenic nerve stimulation and increasing the chance of pacing near the optimal target. The effect of the quadripolar LV leads electrodes was compared using invasively measured pressure-volume loops. Significant differences on LV pump function between pacing sites within an individual patient were seen. However, these differences could not be predicted by the evaluated electrophysiological parameters. In general, the LV lead should be placed in an area with significant electrical delay, preferably in an anterolateral or lateral position. The effect of multi-point pacing (MPP), pacing the LV with two of the quadripolar lead electrodes, was also tested. Overall, MPP did not improve LV pump function compared to CRT with LV pacing using the optimal electrode of the quadripolar lead. However, some patients do benefit from MPP. Lastly, patient-specific intervals were used to optimize the atrioventricular delay of the pacing configurations of CRT with a quadripolar LV lead. An algorithm based on intrinsic conduction delays proved to be beneficial for LV pump function and provided comparable atrioventricular delays for the different pacing configurations of a quadripolar LV lead.