Predicting atrial fibrillation ablation outcome: The CAAP-AF score

Abstract

Background Patients with a variety of clinical presentations undergo atrial fibrillation (AF) ablation. Long-term ablation success rates can vary considerably. Objective The purpose of this study was to develop a clinical scoring system to predict long-term freedom from AF after ablation. Methods We retrospectively derived the scoring system on a development cohort (DC) of 1125 patients undergoing AF ablation and tested it prospectively in a test cohort (TC) of 937 patients undergoing AF ablation. Results The demographics of the DC patients were as follows: age 62.3 ± 10.3 years, male sex 801 (71.2%), left atrial size 4.30 ± 0.69 cm, paroxysmal AF 348 (30.9%), number of drugs failed 1.3 ± 1.1, hypertension 525 (46.7%), diabetes 100 (8.9%), prior stroke/transient ischemic attack 78 (6.9%), prior cardioversion 528 (46.9%), and CHADS2 score 0.87 ± 0.97. Multivariate analysis showed 6 independent variables predicting freedom from AF after final ablation: coronary artery disease (P =.021), atrial diameter (P =.0003), age (P =.004), persistent or long-standing AF (P