Predictive value of ALBI score in assessing prognosis of NSTEMI: a retrospective cohort study

Objective This study aims to evaluate the predictive value of the albumin-bilirubin (ALBI) score for mortality risk in patients with non-ST-segment elevation myocardial infarction (NSTEMI). Methods A retrospective analysis was conducted on 982 patients with NSTEMI. A multivariate Cox proportional hazards model was established to analyze the predictive value of ALBI for mortality after hospital discharge in NSTEMI patients. Additionally, restricted cubic spline (RCS) plots were used to analyze the relationship between ALBI and post-discharge mortality. Operating characteristic (ROC) curves were generated to assess the predictive accuracy of ALBI, and nomograms were constructed to facilitate clinical application in predicting mortality after hospital discharge in patients with NSTEMI. Results Among the 982 participants, 62 (6.3%) developed death during the follow-up. In unadjusted Cox regression models, the ALBI index had a hazard ratio (HR) of 5.07 (95% confidence interval CI: 3.13–8.20, P < 0.001). In the adjusted models, the relationship still persisted. Furthmore, a non-linear and dose–response relationship between the ALBI index and the primary endpoint was observed (non-linear P = 0.050, P overall < 0.001). ROC curve analysis revealed good predictive performance for ALBI. The nomogram model correctly separates patients with and without death risk, with an AUC of 0.835. Our model showed improved prediction of death compared to GRACE, or NT-proBNP (all P < 0.05). Conclusion ALBI was significantly associated with the risk of out-of-hospital death in NSTEMI patients. The novel nomogram based on ALBI, NT-proBNP, and GRACE scores showed an improvement in predicting mortality.