Impact of Liver Fibrosis Scores on Short-Term Mortality in ICU Patients with Acute Myocardial Infarction: Evidence from the MIMIC-IV Database and a Chinese Independent Cohort

Bingfu WangYulong SongYujian FanZhiqiang Zhao^*

• First Teaching Hospital of Tianjin University of Traditional Chinese Medicine, Tianjin, China

Background Acute myocardial infarction (AMI) has a sudden onset and a high short-term mortality rate. While liver fibrosis scores have shown prognostic value in cardiovascular diseases, their role in predicting short-term outcomes in AMI remains unclear. This study aims to systematically evaluate the predictive role of liver fibrosis scoring systems (FIB-4, APRI, NFS, BARD) in short-term clinical outcomes for patients with AMI. Methods This retrospective cohort study analyzed AMI patients from the Medical Information Mart for Intensive Care IV (MIMIC-IV) and the cardiac intensive care unit (CCU) of First Teaching Hospital of Tianjin University of Traditional Chinese Medicine. Primary endpoints were intensive care unit (ICU) and 30-day all-cause mortality. The study used logistic and Cox regression, Kaplan–Meier curves with log-rank tests, restricted cubic spline (RCS), subgroup analyses, and receiver operating characteristic (ROC) curves to assess the associations between liver fibrosis scores and short-term mortality. Results In ICU mortality prediction, area under the curves (AUCs) for FIB-4 were 0.672 (MIMIC) and 0.726 (independent cohort); for APRI, 0.674 and 0.747, respectively; and for NFS, 0.593 and 0.624. Regarding the risk stratification of 30-day mortality, Kaplan–Meier survival curves confirmed the ability of the scores to stratify patients by risk. In both cohorts, the survival curves for groups stratified by FIB-4 were significantly different (log-rank test, P < 0.0001). NFS and APRI also demonstrated a significant ability to differentiate survival outcomes across cohorts (log-rank test, P < 0.05). Multivariate-adjusted logistic regression and Cox regression analyses in both cohorts showed that patients in the highest tertile (T3) of FIB-4, NFS, and APRI scores exhibited a significantly elevated risk of ICU and 30-day all-cause mortality compared with those in the lowest tertile (T1) (all P < 0.05). RCS analyses revealed significant nonlinear associations between FIB-4/APRI and 30-day mortality. Conclusion Elevated liver fibrosis scores (FIB-4, APRI, NFS) are independently associated with increased short-term mortality in patients with AMI. As simple, non-invasive markers, they show promise for the early risk stratification of these high-risk patients and can serve as a useful adjunct for identifying individuals at higher risk.