Biventricular versus conduction system pacing after atrioventricular node ablation in heart failure patients with atrial fibrillation

Abstract

Funding Acknowledgements Type of funding sources: None. Background Atrioventricular node ablation (AVNA) with biventricular (BiV) pacing is an established treatment option for heart failure (HF) patients with drug refractory atrial fibrillation (AF) (1). However, compared to conduction system pacing (CSP) modalities, including His bundle pacing (HBP) and left bundle branch pacing (LBBP), BiV pacing delivers non-physiological ventricular activation (2). Purpose To compare clinical outcomes of BiV pacing and both CSP modalities in HF patients with symptomatic AF who underwent AVNA. Methods Consecutive AF patients with LV ejection fraction (LVEF <50%) who received either BiV pacing or CSP in conjunction with AVNA between May 2015 and July 2021 were retrospectively analysed. Procedural characteristics, electrocardiographic, echocardiographic, and clinical parameters were assessed at baseline and 6 months after the procedure. Results Fifty-five patients (male 43.6%, age 71 years (IQR 10), LVEF 39% (IQR 14)) were included. Thirteen patients (23.6%) received BiV pacing, 30 patients (54.5%) HBP and 12 patients (21.8%) LBBP. All groups had similar baseline characteristics, acute success rate and adverse events. Post-procedural QRS duration was significantly shorter (p<0.01) in CSP (118 ms (IQR 28)) than in BiV pacing (172 ms (IQR 18)). While NYHA class improved in both HBP (p<0.01) and LBBP (p=0.01), it did not improve in BiV group (p=0.1) At follow-up, end systolic volume (ESVi) decreased in both HBP (48±20 to 32±12 mL/m2, p<0.01) and LBBP (62±22 to 52±22 mL/m2, p=0.02), but did not differ in BiV pacing group (51±12 to 53±14 mL/m2, p=0.6). Similarly, LVEF increased in HBP (form 39% (IQR 16) to 53% (IQR 14), p<0.01) and LBBP (from 41% (IQR 23) to 40% (IQR 25), p=0.04), but did not change in BiV group (from 38% (IQR 5) to 37% (IQR 6), p=0.9). Significantly lower (p<0.01) pacing thresholds were achieved in LBBP (0.75 V at 0.5 ms (IQR 0.3)) than in HBP group (1.0 V at 0.5 ms (IQR 1)). Two patients in HBP group were switched to right ventricular pacing due to rise in HBP threshold. In the remaining patients threshold remained stable during follow-up. Conclusion Conduction system pacing modalities showed superior symptomatic and echocardiographic improvement compared to BiV pacing after AVNA. While LBBP offered lower and more stable pacing parameters, there were no differences in clinical outcomes and echocardiographic remodelling when compared to HBP.