CASTLE-AF generally aligned with the rest of the literature in showing better outcomes among patients with atrial fibrillation (Afib) and heart failure who received catheter ablation, according to a meta-analysis.
Compared to drug therapy alone, Afib ablation was linked to lower odds of all-cause mortality (9.0% vs 17.6%, RR 0.52, 95% CI 0.33-0.81) and heart failure hospitalizations (16.4% vs 27.6%, RR 0.60, 95% CI 0.39-0.93) among six trials with at least 6 months of follow-up.
Improvements in other metrics also tended to favor Afib ablation over standard therapy, reported Vivek Reddy, MD, of Icahn School of Medicine at Mount Sinai in New York, and colleagues, in their report published online in the Annals of Internal Medicine:
- Left ventricular ejection fraction (LVEF): mean difference 6.95% (95% CI 3.0%-10.9%)
- 6-minute walk test distance: mean difference 20.93 meters (95% CI 5.91-35.95)
- Peak oxygen consumption (VO2max): mean difference 3.17 mL/kg per minute (95% CI 1.26-5.07)
- Quality of life per Minnesota Living with Heart Failure Questionnaire: mean difference -9.02 points (95% CI -19.75 to 1.71)
Moreover, the results did not change substantively when the controversial CASTLE-AF trial was excluded. That study came under significant criticism and was unique in key ways.
“The favorable outcomes reported with catheter ablation probably were driven mostly by a reduction in Afib burden and an improvement in LVEF,” the authors wrote. “To appreciate how substantial this improvement is, note that treatment with angiotensin-converting enzyme inhibitors, which also has a mortality benefit, results in a 5% LVEF improvement in patients with HFrEF [heart failure with reduced ejection fraction].”
That most patients had robust rate control over follow-up in these trials bolsters the theory that Afib itself is an important cause of LV systolic dysfunction, independently of tachycardia, Reddy and colleagues said.
CASTLE-AF was both the biggest trial and the one with longest follow-up (60 months) among the studies included in the meta-analysis. It was also unique in that it had half of patients treated for paroxysmal Afib, whereas the rest of the trials involved patients with persistent Afib. Critics have pointed to several limitations of CASTLE-AF, including the highly selected population, failure to meet enrollment targets, and significant loss to follow-up.
It was a weakness of the study that CASTLE-AF dominated the results, Reddy’s group acknowledged, but a sensitivity analysis that left out the CASTLE-AF data showed similar results, albeit with significant heterogeneity (I2=71%). The open-label trial designs and potential for patient selection bias across the board were also limitations.
The meta-analysis now shows a numerical trend toward more serious adverse events with Afib ablation (7.2% vs 3.8%, RR 1.68, 95% CI 0.58-4.85), though the investigators said this was driven by a small trial from 2011.
They noted four trials of Afib ablation in HFrEF with results to come: RAFT-AF (ongoing), CATCH-AF (ongoing), AFARC-LVF (presumably completed, status unknown), and AMICA (presumably completed, pending results).
Reddy reported grants and personal fees from Abbott, Biosense Webster, Boston Scientific, CardioNXT/AFTx, and Medtronic; personal fees from Biotronik, CardioFocus, Cardionomic, EBR Systems, and Impulse Dynamics; and stock ownership in Acutus Medical, Affera, Apama Medical, Autonomix, Axon, Backbeat, BioSig, Circa Scientific, Corvia Medical, East End Medical, EPD Solutions, EPIX Therapeutics, EpiEP, Eximo, Farapulse, Javelin, Keystone Heart, LuxCath, Manual Surgical Sciences, MedLumics, Middle Peak Medical, Newpace, Surecor, ValCare, and VytronUS.