Background: The serum C-reactive protein-to-albumin ratio (CAR) has been identified as an adverse prognostic indicator in a variety of diseases. Nevertheless, there have been not been any studies reporting a relationship between CAR and the prognosis of chronic heart failure (CHF). This study was designed to evaluate the association between CAR and all-cause mortality in CHF patients with different ejection fractions. Methods: A total of 1221 heart failure (HF) patients were enrolled at the First Affiliated Hospital of Kunming Medical University due to acute exacerbation of chronic HF from January 2017 to October 2021. The main outcome was all-cause mortality. After collecting baseline characteristics and laboratory results from all patients, we classified all participants into four groups based on CAR quartile (G1-G4). Kaplan-Meier survival curves and multivariate Cox proportional hazard models were employed to investigate the association between CAR and all-cause mortality in the patients. Furthermore, receiver operating characteristic (ROC) curves were constructed for CARs, and the area under the curve (AUC) was calculated. Results: After excluding ineligible patients, we ultimately included 1196 patients with CHF. The mean age was 66.38 +/- 12.521 years, and 62% were male. According to the Kaplan-Meier analysis, with different ejection fractions, the risk of all-cause mortality was always highest for G4 (CAR >63.27) and lowest for G1 (CAR <= 7.67). Cox multivariate regression analyses indicated that the CAR was an independent predictor of all-cause mortality in all HF patients and in patients with different HF subtypes. According to the ROC curves, the AUC for the CAR was 0.732 (p < 0.001), with a sensitivity of 66.2% and the specificity of 72.7%. CAR had a greater predictive value for all-cause mortality than did C-reactive protein (CRP). Conclusions: An elevated serum CAR was independently associated with an increased risk of all-cause death, regardless of heart failure subtype.