Integrated biomarkers and cardiac phenotypes associated with atrial fibrillation: evidence from real-world hospital data

Background: This study aimed to investigate the association between the B-type natriuretic peptide (BNP)/ left ventricular ejection fraction (LVEF) and atrial fibrillation (AF) and to evaluate its discriminative performance in a real-world hospitalized population. Methods: We conducted a hospital-based observational study that included 270 consecutive patients hospitalized at the Second Affiliated Hospital of Wannan Medical University between August 2023 and December 2025. BNP/LVEF was calculated by dividing plasma BNP levels by LVEF measured by transthoracic echocardiography. AF was diagnosed based on cardiac rhythm recordings during hospitalization or physician diagnosis. We used a multivariate logistic regression model to analyze the association between BNP/LVEF and atrial fibrillation. Restricted cubic spline analysis was used to explore potential non-linear relationships. Receiver operating characteristic (ROC) curves were used to assess discriminatory performance, and pre-specified subgroup analyses were performed to evaluate effect modification. Results: Among 270 patients, 123 (45.6%) had atrial fibrillation. In all adjusted models, higher BNP/LVEF values were independently associated with an increased risk of atrial fibrillation in fully adjusted model (OR, 1.02; 95% CI, 1.01-1.04). Restricted cubic spline analysis revealed a significant non-linear association (P-nonlinear 0.001), with a sharp increase in the risk of atrial fibrillation at lower BNP/LVEF levels, followed by a plateau. BNP/LVEF demonstrated good discriminatory ability for atrial fibrillation (AUC = 0.84) in the full unadjusted model. Subgroup analysis showed significant heterogeneity across sex, smoking, alcohol consumption, and coronary heart disease status. Conclusion: BNP/LVEF is independently and nonlinearly associated with atrial fibrillation and demonstrates good discriminative performance. This integrative marker may provide clinically useful information for AF risk stratification in hospitalized patients.