Dispersion of repolarization increases after cardiac resynchronization therapy in patients who do not undergo left ventricular reverse remodelling

Abstract

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Wellcome/EPSRC Centre for Medical Engineering CardioInsight Inc. Background The effect of CRT on dispersion of repolarization and arrhythmic risk is unclear. LV epicardial pacing has been associated with increased dispersion of repolarization, which may be due to altered activation and repolarization sequence. However, while CRT-induced ventricular arrhythmias have been reported, evidence from large clinical trials suggest CRT has a favourable effect on arrhythmic risk, with a lower incidence of arrhythmia in patients who undergo LV reverse remodelling. Purpose To investigate the effect of CRT and LV reverse remodelling on dispersion of repolarization using electrocardiographic imaging (ECGi). Methods 11 patients with heart failure and electrical dyssynchrony underwent ECGi after CRT implant and again at 6 months. Reconstructed epicardial electrograms were used to create maps of activation recovery intervals (ARI), an accepted surrogate for action potential duration, which were corrected for heart rate. LV ARI dispersion was calculated as the standard deviation of ARI across the LV epicardium. The methodology is summarized in figure 1. Results Mean age at implant was 74±10 years and 82% of patients were male. 64% had ischaemic aetiology of heart failure, and mean LV ejection fraction was 29±10%. 64% of patients had underlying LBBB, 28% had an RV-paced rhythm and 9% had RBBB. 8 patients had a ≥15% reduction in LV end-systolic volume (LVESV) with CRT at 6 months (volumetric responders). Example ARI maps for 1 patient are shown in figure 2A. There was a significant increase in LV ARI dispersion at 6 months compared to baseline (36.4±7.2ms vs 28.2±7.7ms; P=0.03) [Fig 2B]. In a multiple linear regression analysis, volumetric response was an independent predictor of relative change in LV ARI dispersion from baseline to 6 months (P=0.04). In a sub-analysis, for volumetric responders there was no significant difference in LV ARI dispersion between baseline and CRT at 6 months (36.4 ±6.1 vs 30.1±7.8 ms; P=0.1). In comparison, in volumetric non-responders there was a significant increase in LV ARI dispersion (38.3±1.2 vs 22.6±2.6 ms; P=0.01). The relative change in LV ARI dispersion from baseline to CRT 6-months was greater for volumetric non-responders compared to volumetric responders (70.7 ±21.3% vs 27.0 ±35.4%; P=0.04) [Fig 2C]. There was a moderate negative correlation between relative change in LV ARI dispersion and relative reduction in LVESV (R=-0.5), however this did not meet statistical significance (P=0.12) [Fig 2D]. Conclusion CRT increases dispersion of repolarization at 6 months. However, this potentially arrhythmogenic effect of epicardial pacing was only observed in CRT non-responders, which is in keeping with previous evidence that LV reverse remodelling reduces risk of ventricular arrhythmia.