Catheter ablation in end-stage heart failure with atrial fibrillation: an hierarchical endpoint analysis of the CASTLE-HTx trial

Abstract

Background The CASTLE-HTx trial (NCT04649801) showed that the combination of catheter ablation and guideline-directed medical therapy (GDMT) was associated with a lower likelihood of a composite of death from any cause, implantation of a left-ventricular assist device (LVAD), or heart transplantation (HTx) in patients with end-stage heart failure (HF) and atrial fibrillation (AF). Aims This is an ancillary analysis with the generalized pairwise comparison methodology of the main outcomes of CASTLE-HTx. Methods In CASTLE-HTX, 194 patients were randomized to catheter ablation and GDMT (n = 97) or medical therapy alone (n = 97). The first hierarchical outcome was a composite of (1) death from any cause, (2) urgent HTx, (3) implantation of an LVAD, and (4) frequency of hospitalizations for worsening HF. Secondary analysis also included AF burden and left-ventricular ejection fraction improvement. Treatment effects are reported as net treatment benefit (NTB) and win odds, with corresponding confidence intervals (CIs). Results Patients randomized to ablation had more wins with respect to death from any cause (28.4%/12.4%), HTx (4.3%/2.1%), LVAD implantation (4.2%/0.6%), and hospitalizations for worsening HF (24.8%/11.3%). 61.6% of pairs favoured ablation, 26.4% medical therapy, and 12.0% were tied. This resulted in a restricted NTB at 3 years of 35.3% (95% CI, 19.8-49.0, P < .001) favouring ablation (win odds: 2.09; 95% CI, 1.49-2.92). In the secondary analysis, 67.0% of pairs favoured ablation against 28.6% of pairs favouring medical therapy. The restricted NTB was 38.4% (95% CI, 22.8-52.0, P < .001) in favour of ablation (win odds: 2.25; 95% CI, 1.59-3.17). Conclusions This pairwise analysis of CASTLE-HTx confirms the clinical benefits of early catheter ablation in addition to GDMT among AF patients in end-stage HF. The treatment benefit was primarily driven by mortality and HF hospitalizations.[GRAPHICS].