Electromechanical investigations in cardiac resynchronisation therapy

Abstract

Cardiac Resynchronisation Therapy (CRT) is an established treatment for systolic heart failure with certain electrical and mechanical parameters. However not all patients implanted with CRT with these parameters will improve symptomatically or in cardiac function. In order to enhance the response rate to CRT methods have been developed to select patients more appropriately and therapy delivery technology has progressed. In this thesis I assess implications for CRT response using cardiac magnetic resonance (CMR) imaging techniques, 12 lead electrocardiography (ECG) markers and non-invasive body surface electrical mapping technology. Initially we studied CMR derived 3-dimensional meshes of cardiac morphology in order to test whether left ventricular shape impacts response to CRT. We show that patients were more likely to respond to CRT if they exhibited an asymmetric thickening pattern, with greater hypertrophy of the antero-lateral wall with respect to the postero-septal. Following on from this I consider the interaction between 12 lead ECG markers (QRS duration & bundle branch block morphology) and CMR imaging (scar and contraction pattern data) with the potential for CRT response. In the first of these studies I review left bundle branch block patients and mechanical contraction patterns derived from CMR; patients with strict left bundle branch block and a type 2 or ‘U-shaped’ contraction pattern are highly likely to volumetrically remodel following CRT, with the majority of these patients becoming ‘super-responders’. The following study examines the interaction between QRS duration, MRI derived mechanical dyssynchrony and scar in order to consider whether mechanical dyssynchrony in narrow QRS patients is amenable to treatment with CRT. We demonstrate that 25% of patients with systolic heart failure and a narrow QRS have mechanical dyssynchrony, but this seems to be secondary to scar, which may give insight into non-response and potential harm from CRT in this group. Multisite pacing is a newer technique that may offer improved response rates to CRT. In the next section I will present the acute haemodynamic data from multi site and multi vein pacing protocols in left bundle branch block patients; the results indicate that myocardial scar seems integral for improvement with multi site left ventricular pacing over traditional biventricular pacing in these patients. Finally, I performed a prospective clinical study using body surface ECG mapping to look at mapping differences between right ventricular pacing sites in left bundle branch block patients as well as studying high fidelity electrical predictors of response to CRT. We establish that right ventricular septal pacing is closer in activation pattern to intrinsic left bundle branch block than apical pacing both in an in vivo and in silico model. Lastly, in our cohort of patients studied, baseline activation parameters derived from mapping do not enhance prediction of response to CRT, but may offer assistance in targeting left ventricular lead position at implant in order to maximise response rate.

Date of Award	1 Apr 2021
Original language	English
Awarding Institution	King's College London
Supervisor	Reza Razavi (Supervisor)