“Cardiac resynchronisation therapy”: Does the haemodynamic improvement of biventricular pacing truly arise from cardiac resynchronisation?

Abstract

In this thesis I have explored some of the fundamental concepts which underpin biventricular pacing (commonly called cardiac resynchronisation therapy, CRT). As a therapy, its impact on survival, and symptoms is impressive. By adopting the name cardiac resynchronisation therapy, a common assumption is that these benefits come from ensuring resynchronisation of the ventricles in the failing heart. In this thesis I explore how biventricular pacing delivers its benefit, and whether there are other dimensions beyond resynchronisation which deserve more attention. I first performed a meta-analysis to quantify what the actual symptomatic benefit from biventricular pacing is in the randomised controlled trials. A non-response rate of one-third is often quoted to biventricular pacing, but my analysis demonstrated that once the effect seen in the control arms is deducted the incremental symptomatic response rate is closer to 15%. I explored more acute markers of response, and how they are used for optimisation of biventricular pacing. I composed a review of different technologies available for optimisation, and a developed a step wise approach to develop the ideal optimisation scheme. Left ventricular outflow tract (LVOT) Doppler is one commonly used measure for optimisation, and my analysis concluded that a much larger number of beats is required for precise optimisation., I evaluated a novel method to acquire and trace around large numbers of LVOT Doppler velocities, and assessed whether breath holding is required. I discovered that breath holding did not have a significant impact on the magnitude or variability of measurements, and quiet breathing may be the easier way to acquire a larger number of beats for precise measurements. An algorithm using multiple alternations of systolic blood pressure between reference and tested pacing setting has been developed by my supervisors for reproducible AV optimisation, I used this technology to explore current techniques, and explore concepts in biventricular pacing: I evaluate the different methods for manufacturer specific electrogram-based AV optimisation. I found that agreement between the different methods is poor, and none agree with the haemodynamic optimum. I explored the apparent discrepancy studies have reported on the effect of VV optimisation. By performing VV optimisation by using four different methods for holding the AV delay constant (A-LV constant, A-RV constant, time to first ventricular lead constant, and time to second ventricular lead constant), I discovered that the acute haemodynamic effect was predominantly determined by the time to the first paced ventricle. To explore the influence of pure AV optimisation in heart failure I examined a group of patients with PR prolongation and demonstrated a significant improvement in acute haemodynamic response with AV optimised pacing of the His bundle. Temporary pacing of the His allows us to maintain the same, narrow QRS morphology and thus examine the pure effect of AV optimisation, an mean increment of 4 mmHg in systolic pressure is seen, approximately 60% of that seen in heart failure with LBBB. This also demonstrates the pure effect of AV shortening without an associated adverse haemodynamic effect of right ventricular pacing. I explored the role of lead position and whether the AV optimum varies between different LV lead positions. The AV optimum, did not significantly differ, suggesting high precision measurements at one AV delay could be used to determine the best lead position which in my study was occasionally in a position which would usually be considered non-conventional (anterior basal wall). I finally explored the role of biventricular pacing in non-LBBB morphologies looking at outcome studies. My analysis showed that biventricular pacing has a harmful effect in narrow QRS and the effect increases with the duration of time, indicating that this is due to a physiologically adverse effect of pacing. As this is the case, one can make a case for switching off biventricular pacing in such patients.