Predictors of short-term atrial fibrillation recurrence after pulmonary vein isolation: the role of left atrial strain and structural parameters

Abstract Introduction Atrial fibrillation (AF) is the most common arrhythmia, significantly affecting quality of life. Pulmonary vein isolation (PVI) is a key treatment, but recurrence remains a challenge. Left atrial (LA) structural and functional parameters, including LA strain (LAS) assessed via echocardiography, have emerged as potential predictors of post-ablation AF recurrence. Purpose To assess baseline comorbidities and echocardiographic differences between recurrent and non-recurrent patients undergoing PVI for AF. Methods This retrospective, single-center study included AF patients who underwent first-time PVI from 2015 to 2024 with prior formal echocardiographic evaluation. Echocardiographic images were re-evaluated, and LAS and volume measurements were analyzed when image quality allowed. AF recurrence was assessed at 6 and 12 months using 12-lead ECGs or available rhythm Holter monitoring. Results Of 216 patients who underwent PVI in our center, 72 met inclusion criteria. Sixty completed 6-month follow-up, and 51 completed 12-month follow-up. The median time from echocardiography to PVI was 312 days [IQR 120–610]. AF recurrence occurred in 9 patients (15.0%) at 6 months and 15 (29.4%) at 12 months. Recurrent patients at 12 months were older (69.5 vs. 60.2 years, p=0.01), while prior comorbidities and antiarrhythmic drug use were similar between groups. Time from AF diagnosis to ablation was longer in recurrent patients but not statistically significant (6 months: 1435 vs. 960 days, p=0.39; 12 months: 1435 vs. 960 days, p=0.26). At 12 months, AF recurrence was associated with a larger pre-PVI indexed LA diameter (23.3±3.5 vs. 20.5±2.6 mm/m², p=0.01) and lower E′ lateral (9.5±3.0 vs. 12.5±3.0 cm/s, p=0.01). At 6 months, recurrent patients also had significantly lower pre-PVI E′ lateral (9.3±3.5 vs. 12.0±2.9 cm/s, p=0.02). Although LA reservoir strain trended lower in recurrent patients at 12 months (20.2±7.3% vs. 25.7±7.7%, p=0.08), it did not reach statistical significance. Other echocardiographic parameters, including conduit and contraction strain, indexed LA volume, left ventricular global longitudinal strain, and E/e′ ratio, showed no significant differences between groups. Conclusion This study has several limitations. The small sample size and low inclusion rate limit generalizability and may introduce selection bias. Additionally, the long median time from echocardiography to PVI could affect the accuracy of pre-procedural measurements. Despite growing evidence suggesting that LA strain analysis could predict AF recurrence, our study did not find it to be a significant predictor. However, given the study's limitations, this does not rule out its potential prognostic value. Larger, prospective studies with higher inclusion rates and shorter echocardiography-to-PVI intervals are needed to clarify whether strain is a strong predictor of recurrence.