AB-453070-3 VERY-EARLY SYMPTOMATIC RECURRENCE IS ASSOCIATED WITH INCREASED RISK OF LATE RECURRENCE AMONG ATRIAL FIBRILLATION PATIENTS AFTER RADIOFREQUENCY ABLATION

Abstract

The effect of very-early (within 48 h) symptomatic recurrence (VESR) of atrial fibrillation (AF) after radiofrequency catheter ablation (RFCA) on late recurrence (LR) (after 3 months of RFCA) has been seldomly reported. To explore the relationship between VESR and LR among post-RFCA patients. We enrolled AF patients who registered in the Chinese Atrial Fibrillation Registry study and received the first RFCA procedure from June 2018 to December 2021 at Beijing Anzhen Hospital. Patients were divided into four groups based on VESR and early recurrence (ER) (from 48 h to 3 months after RFCA): group A (no VESR, no ER); group B (VESR but no ER); group C (ER but no VESR); and group D (both VESR and ER). Their clinical characteristics and risk of LR events were compared among groups. The study included 6887 patients; VESR was found in 330 (4.79% ) patients. The proportion of women in patients with VESR was higher (group B vs. group A: 40.65% vs. 32.03%, P=0.043; group D vs. group A: 39.13% vs. 32.03%, P=0.032), and the proportion of hypertensive patients was lower (group B vs. group A: 41.46% vs. 53.64%, P=0.007; group D vs. group A: 44.44% vs. 53.64%, P=0.009). With an average follow-up of 17.7 months, the Kaplan-Meier curve showed that LR risk in VESR patients was significantly higher than in other patients in both paroxysmal (log-rank, P<0.001) and persistent (log-rank, P<0.001) AF patients (P for interaction=0.118). In VESR patients, LR risk between group B and group D patients was significantly different (log-rank, P<0.001). Multivariate Cox regression showed that group B and group D were associated with increased risk of LR (HRB:A=2.161, P<0.001; HRD:A=7.401, P<0.001). In VESR patients, compared with the combination of variables in the APPLE score (Age, Persistent AF, imPaired eGFR, Left atrial diameter, left ventricular Ejection fraction), the combination of group and the APPLE score variables significantly improves AUC for predicting LR at 6 (73.2% vs. 58.2%, P<0.001), 12 (72.0% vs.60.5%, P<0.001), and 18 (69.2% vs. 50.3%, P=0.002) months. VESR is associated with increased risk of LR in both paroxysmal and persistent AF patients. Further, classification of VESR patients based on recurrence between 48 h and 3 months improves the prediction potential of LR risk in these patients.