Left atrial stiffness index predicts atrial fibrillation risk in heart failure with preserved ejection fraction: a nomogram model.

OBJECTIVE: Develop a LASI-based nomogram for predicting atrial fibrillation (AF) risk in heart failure with preserved ejection fraction (HFpEF). METHODS: This retrospective study analyzed 275 HFpEF patients (160 AF, 115 controls). Echocardiography measured LASI (E/e' divided by peak atrial longitudinal strain), left atrial volumes, strains (AP2%, AP4%), and electromechanical delays (SD2: inter-atrial; SD4: intra-left atrial). Multivariate logistic regression identified AF predictors. Nomogram performance was validated by ROC analysis and DCA. RESULTS: Compared to controls, the AF group had significantly larger LVDD/LVSD (P<0.001 both), lower 3D-LAEF% (P<0.001), lower E/e' (P<0.001), higher LASI (P<0.001), larger BSA (P<0.001), higher AP2% (P<0.001), lower AP4% (P<0.001), and longer SD4 (P<0.001). Multivariate analysis identified positive associations with AF risk for: BSA (OR=9.167, P<0.001), AP4% (OR=1.033, P=0.008), SD2 (OR=1.003, P=0.001), and LASI (OR=1.043, P<0.001). Negative associations were found for E/e' (OR=0.889, P=0.002) and SD4 (OR=0.997, P<0.001). ROC AUCs were: LASI=0.666, E/e'=0.707, BSA=0.682, SD2=0.615, AP4=0.666, SD4=0.705. The combined model AUC was 0.801. DCA identified LASI as the optimal single predictor (net benefit 0.3184). CONCLUSION: LASI independently predicted AF risk in HFpEF. The validated nomogram, integrating LASI, BSA, and electromechanical markers (SD2, SD4, AP4%, E/e'), enables precise AF risk stratification, aiding early identification of high-risk patients for targeted intervention.