Overview
Atrioventricular node ablation with permanent pacing (the "pace-and-ablate" strategy) controls ventricular rate in atrial fibrillation that has not responded to drug therapy or catheter ablation. Chronic right ventricular pacing, however, can cause pacing-induced cardiomyopathy (PICM), and it is not well established which patients are at risk and should therefore receive conduction system pacing or biventricular pacing instead.
REPACE is a multicentre retrospective cohort of consecutive patients who underwent atrioventricular node ablation from January 2011 onwards at five centres in Switzerland, the United Kingdom and Poland: Liverpool Heart and Chest Hospital, Inselspital Bern, University Hospital Basel, Geneva University Hospitals and the Silesian Centre for Heart Diseases, Zabrze. More than 2,000 patients ablated up to March 2024 have been included to date; patients ablated after that date at the participating centres are being added.
The study has three aims: to validate, in an independent multicentre population, a previously published risk model for pacing-induced cardiomyopathy; to describe the procedural technique and outcomes of atrioventricular node ablation across different health systems; and to estimate the health-economic impact of the pace-and-ablate strategy.
Pacing-induced cardiomyopathy is defined as an absolute reduction in left ventricular ejection fraction of at least 10 percentage points to a value below 50%. All potential endpoints are adjudicated by a clinical events committee blinded to the classification made at the site.
Description
Data collection. Baseline demographics, comorbidity, device and procedural data, electrocardiographic and echocardiographic measurements, follow-up events and mortality were abstracted from hospital records at each participating centre into a central electronic case report form.
Endpoint adjudication. Every potential endpoint is reviewed by two independent adjudicators blinded to the classification made at the site, with a third reviewer consulted where they disagree. Pacing-induced cardiomyopathy is assessed on every adjudicated event, independently of how that event is otherwise classified.
Analysis populations. The validation of the risk model is performed in the patients who had neither cardiac resynchronisation therapy nor conduction system pacing at baseline, because a decline in left ventricular function in an already resynchronised patient represents non-response to resynchronisation rather than pacing-induced cardiomyopathy. Patients with resynchronisation at baseline are described and contribute to the procedural and health-economic analyses. No lower limit of baseline ejection fraction is applied: since the definition requires a fall of at least 10 percentage points to a value below 50%, patients already below 50% at baseline can meet it.
Health-economic analysis. Direct costs are derived from Hospital Episode Statistics in the United Kingdom, individual procedure costs from the SwissEPnet registry for Switzerland, and diagnosis-related group and hospital costs for Switzerland and Poland, covering hospital charges, device costs and subsequent care for complications and follow-up. Cost-utility and cost-effectiveness analyses use Markov models to compare pacing strategies.
Eligibility
Inclusion Criteria:
- Age 18 years or older at the time of atrioventricular node ablation
- Atrial fibrillation, undergoing atrioventricular node ablation with a permanent cardiac implantable electronic device already in situ or implanted as part of the pace-and-ablate strategy
- Availability of baseline echocardiographic data, including left ventricular ejection fraction
Exclusion Criteria:
\* Documented refusal of the further use of health-related data
Note: cardiac resynchronisation therapy or conduction system pacing at baseline is not an eligibility exclusion. Such patients are included in the cohort and in the procedural and health-economic analyses, but are excluded from the validation of the pacing-induced cardiomyopathy risk model, in which a decline in left ventricular function would represent non-response to resynchronisation. No lower limit of baseline ejection fraction is applied.
