Background
A high proportion of patients undergoing catheter ablation (CA) for atrial fibrillation (AF) experience recurrence of arrhythmia. This meta-analysis aims to identify pre-ablation serum biomarker(s) associated with arrhythmia recurrence to improve patient selection before CA.

Methods
A systematic approach following PRISMA reporting guidelines was utilised in libraries (Pubmed/Medline, Embase, Web of Science, Scopus) and supplemented by scanning through bibliographies of articles. Biomarker levels were compared using a random-effects model and presented as odds ratio (OR). Heterogeneity was examined by meta-regression and subgroup analysis.

Results
In total, 73 studies were identified after inclusion and exclusion criteria were applied. Nine out of 22 biomarkers showed association with recurrence of AF after CA. High levels of N-Terminal-pro-B-type-Natriuretic Peptide [OR (95% CI), 3.11 (1.80–5.36)], B-type Natriuretic Peptide [BNP, 2.91 (1.74–4.88)], high-sensitivity C-Reactive Protein [2.04 (1.28–3.23)], Carboxy-terminal telopeptide of collagen type I [1.89 (1.16–3.08)] and Interleukin-6 [1.83 (1.18–2.84)] were strongly associated with identifying patients with AF recurrence. Meta-regression highlighted that AF type had a significant impact on BNP levels (heterogeneity R2 = 55%). Subgroup analysis showed that high BNP levels were more strongly associated with AF recurrence in paroxysmal AF (PAF) cohorts compared to the addition of non-PAF patients. Egger’s test ruled out the presence of publication bias from small-study effects.

Conclusion
Ranking biomarkers based on the strength of association with outcome provides each biomarker relative capacity to predict AF recurrence. This will provide randomised controlled trials, a guide to choosing a priori tool for identifying patients likely to revert to AF, which are required to substantiate these findings.