Anticoagulation strategy of choice for atrial fibrillation ablation - a network meta-analysis of complication rates

Direct oral anticoagulants (DOACs) and vitamin K antagonists (VKA) interrupted (i) or uninterrupted (u) are used for atrial fibrillation (AF) ablation. Small studies have suggested some differences in efficacy for interrupted and uninterrupted regimens, as well as individual DOACs (dabigatran (D), apixaban (A), rivaroxaban (R), edoxaban (E). A network meta-analysis would provide further clarification of different regimens safety.

To evaluate total complications (bleeding and thromboembolic) of interrupted and uninterrupted strategy of individual DOACs and VKA.

We analysed 47 published studies that directly compared the complication rates of individual DOACs (D, A, R, E) with VKA, either interrupted or uninterrupted, in patients undergoing AF ablation. The primary outcomes were total complications – combined bleeding and thromboembolic events. A network meta-analysis was used to pool effect sizes comparing each individual regimen, which were expressed as odds ratios (OR) with 95% confidence intervals (CI), and a pairwise meta-analysis using a random effects model.

A total of 22.188 patients were included (DOACs: 9.972; uVKA: 9.781; iVKA: 3.435). iDOACs were used in 6.451 patients (iD: 3.964; iR: 1.964; iA: 462; iE: 61) and uDOACs in 3.521 (uA: 1.177; uD: 1.987; uE: 357). There were no differences in complication rates among individual DOACs, either interrupted or uninterrupted (Figure 1B, Figure2.A,B). uVKA was superior to iVKA but similar to individual DOACs, except iD, which had a better safety profile. iVKA had the worst safety profile (Figure 1A)

This meta-analysis shows that safety is similar among D, A, R, E, either interrupted or uninterrupted strategy. iD has a significantly lower complication rate than uVKA offering a safer profile. Our data supports the use of DOACs or uVKA as anticoagulation regimens peri AF ablation.

Forest Plots of Pairwise Estimates