Association between left atrial stiffness index and atrial fibrillation recurrence following radiofrequency ablation

Abstract Background Growing evidence indicates that the left atrial stiffness index (LASI) serves as a straightforward and reliable surrogate for left atrial (LA) function. Although LASI has been linked to atrial fibrillation (AF) recurrence, supporting evidence remains limited. Obesity is also implicated in the pathophysiology of AF. Currently, no existing research has investigated the correlation between the Z-score-standardized product of LASI and body mass index (z-LASI-BMI) and AF recurrence. To the best of our knowledge, this is the first study to investigate the association between z-LASI-BMI and AF recurrence after radiofrequency catheter ablation (RFCA). We further evaluated the predictive value of LASI and z-LASI-BMI for post-ablation AF recurrence. Methods In this retrospective study conducted at the Second Affiliated Hospital of Xi’an Jiaotong University, 240 patients who underwent initial RFCA for AF were consecutively enrolled between September 2022 and October 2024. The analysis evaluated the association between left atrial stiffness parameters (LASI and z-LASI-BMI) and post-ablation AF recurrence. Results Over a 12-month observation period post-RFCA, 25.0% of participants exhibited AF recurrence. Multivariable regression analysis confirmed that both LASI and z-LASI-BMI independently predicted AF recurrence. Restricted cubic spline models revealed a positive linear relationship between LASI and z-LASI-BMI and AF recurrence rates. Additionally, adding the z-LASI-BMI index into the fully adjusted baseline risk model significantly improved the prediction of AF recurrence, as reflected by increases in the C-statistic, continuous net reclassification improvement, and integrated discrimination improvement, but requires external validation. Conclusions Both LASI and z-LASI-BMI were independently associated with AF recurrence after ablation. z-LASI-BMI improved discrimination beyond LASI alone but remains exploratory and requires external validation.