Prognostic value of CT-derived biatrial strain in predicting recurrence after atrial fibrillation ablation

Background Cardiac computed tomography (CT) is routinely performed before atrial fibrillation (AF) ablation for anatomical assessment. Recent advances in CT-based feature-tracking enable quantification of atrial strain, allowing functional evaluation of atrial remodeling; however, the prognostic relevance of CT-derived biatrial strain remains uncertain. Purpose To evaluate the association between CT-derived left atrial (LA) and right atrial (RA) strain and AF recurrence and to determine whether CT-defined biatrial functional phenotypes provide incremental prognostic value beyond structural indices. Methods We retrospectively analyzed 422 consecutive patients who underwent first radiofrequency AF ablation with preprocedural cardiac CT between 2016 and 2024. LA and RA strain were quantified using CT-derived feature-tracking analysis from routine imaging datasets. Patients were categorized according to median strain values to define biatrial functional phenotypes. The primary endpoint was 3-year AF/atrial tachycardia recurrence. Kaplan–Meier analysis and multivariable Cox regression models were performed with sequential adjustment for age, sex, AF type, and CT-derived LA volume index. Results The study population had a mean age of 64.7 ± 10.2 years, 32.2% were male, and 51.5% had persistent AF. Median LA strain was 14.6% and median RA strain was 20.7%, with strong inter-atrial correlation (r = 0.71, p < 0.001). During a median follow-up of 1.9 years (IQR 0.9–3.0), 139 patients (32.9%) experienced recurrence. Three-year recurrence rates were 23.8% in the LA-high/RA-high group, 37.3% in LA-high/RA-low, 47.6% in LA-low/RA-high, and 45.8% in LA-low/RA-low (log-rank p = 0.00018). After adjustment for age and sex, both the LA-only dysfunction group (HR 2.88, 95% CI 1.60–5.18) and the biatrial dysfunction group (HR 2.26, 95% CI 1.51–3.36) were significantly associated with recurrence, whereas isolated RA dysfunction showed borderline significance. With additional adjustment for AF type and LA volume index, reduced LA strain remained independently associated with recurrence. Trend analysis demonstrated a graded increase in risk across CT-defined phenotypes (HR per step 1.30, 95% CI 1.14–1.47; p < 0.001). Conclusions CT-derived atrial strain enables non-invasive functional assessment of biatrial remodeling. Although LA and RA strain are strongly correlated, recurrence after AF ablation is predominantly associated with impaired LA function.