Abstract
BACKGROUND: The synergistic effect of fibrinogen and albumin on long-term mortality remains unclear in senile patients with coronary artery disease (CAD). This study aimed to evaluate the association between fibrinogen-to-albumin ratio (FAR) and 10-year mortality, and to develop a model to predict survival probability in senile patients with CAD. METHODS: In total, 819 senile patients with CAD were enrolled on the basis of the China Geriatric Cardiovascular Comorbidity Study. RESULTS: Compared with patients in the lowest FAR (FAR-Q1) group, the median overall survival (OS) was 2631 days, and patients in the highest FAR (FAR-Q4) group had the shortest OS, with a median of 311 days (p < 0.0001). Multivariate Cox regression suggested FAR as a crucial factor affecting long-term mortality of patients with CAD (HR = 37.75, 95% CI = 4.10-347.98; p < 0.05). Five features associated with long-term mortality were selected using Least Absolute Shrinkage and Selection Operator (LASSO) regression: age, hemoglobin, albumin, FAR, and lnNT-proBNP. The area under the receiver operating characteristic curve (AUC) was 0.838 for multivariate Cox regression and 0.829 for LASSO regression. The restricted cubic spline curve showed a significant J-shaped relationship between FAR and mortality, with a cut-off point of 0.09 (p for nonlinear < 0.001). A time-dependent nomogram was constructed based on five features selected using LASSO regression. The time-dependent AUC remained in the range of 0.69-0.73, indicating the relatively stable power of this model. CONCLUSION: FAR was independently associated with long-term mortality, and a prognostic model based on FAR may aid risk stratification in senile patients with CAD.