Abstract
[Doctor of Medicine thesis]. Cardiac resynchronization therapy (CRT) has become an important therapeutic strategy for heart failure (HF) patients with impaired left ventricular (LV) systolic function and prolonged QRS duration. The benefit of CRT as an adjunct to pharmacological therapy is now well established, with sustained improvements in quality of life, hospitalization rates and mortality. However, even in carefully selected patients, the response to CRT is often unpredictable with a considerable number of nonresponders (30-50%). Although the reasons for this nonresponse are not entirely clear, studies have suggested that non-optimal left ventricular (LV) lead positioning, lack of electrical dyssynchrony, suboptimal device programming and myocardial scar burden play an important role.More recently, a wealth of evidence has pointed to the limitations of 12-lead electrocardiography, suggesting it may not accurately reflect the presence or complexities of electrical dyssynchrony in the failing heart. As the efficacy of CRT is primarily achieved through LV resynchronization, there has been renewed interest in the development of techniques that enable better characterization of cardiac electric activation patterns and identification of electrical dyssynchrony. This approach would appear logical, given that CRT is primarily an ‘electrical therapy’, designed to treat an underlying electrical conduction abnormality. Vectorcardiography (VCG), which was first described in 1920, offers an alternative interpretation of the 12-lead ECG. Its resurgence in the field of CRT has emerged from the recognition that VCG parameters can provide information on dyssynchrony beyond that currently provided by the 12-lead ECG. Prominent amongst these is vectorcardiographic QRS area (QRSarea), which has been shown to be superior to QRSd and QRS morphology in predicting response to CRT.
The work presented herein is structured into two major sections. First, we investigate the role of QRSarea as a novel predictor of response to CRT. Using a combination of different study designs, our results demonstrate that QRSarea is a better predictor of CRT response than QRSd and QRS morphology. We also show that CRT-induced ΔQRSarea can be used to help quantify LV resynchronization and to predict long-term clinical outcomes following CRT. Importantly, we are the first to show that a concomitant reduction in both QRSarea and QRSd is associated with the best clinical outcomes after CRT, indicating that ECG and VCG can be used in conjunction to help improve patient selection for CRT.
In the second part of this thesis, we focus on the development and validation of a novel, vector-based 3D electroanatomical modelling system. Using a novel computational method, ECGSync combines the surface ECG-derived vectorcardiogram with cardiac magnetic resonance imaging to estimate, by inverse solution, the 3-dimensional sequence of LV activation. Accordingly, we show that ECGSync can noninvasively map ventricular electrical activity and accurately locate the site of latest electrical activation prior to CRT implantation. Furthermore, we demonstrate that novel ECGSync-derived markers of dyssynchrony can help predict CRT response. Our findings suggest that VCG may have great potential to improve the clinical application of CRT.
Date of Award | Aug 2022 |
---|---|
Original language | English |
Supervisor | Francisco Leyva-Leon (Supervisor) & Keqing Wang (Supervisor) |