Clinical Focus ›› 2025, Vol. 40 ›› Issue (11): 978-987.doi: 10.3969/j.issn.1004-583X.2025.11.003

Previous Articles     Next Articles

Association between the De Ritis ratio and 28-day mortality in patients with cerebrovascular accident: A MIMIC-Ⅳ database analysis

Ren Dezhi1,2, Xu Siyao1,2, Wang Shuai3, Duan Jun2()   

  1. 1. China-Japan Friendship Hospital (Institute of Clinical Medical Sciences), Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100029,China
    2. Department of Critical Care Medicine, China-Japan Friendship Hospital, Beijing 100029,China
    3. Department of Emergency, Tongren Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai 200336, China
  • Received:2025-09-23 Online:2025-11-20 Published:2025-12-02
  • Contact: Duan Jun E-mail:13691362130@163.com

Abstract:

Objective To evaluate the association between the aspartate transaminase (AST)/alanine transaminase (ALT) (De Ritis) ratio and 28-day mortality in patients with cerebrovascular accident (stroke) and to assess its predictive value across clinical subgroups. Methods We performed a retrospective cohort study using the MIMICIV database (2008-2019). From 6, 165 stroke admissions identified by ICD codes I60, I63, and I69, 1, 674 patients were retained after excluding cases with excessive missing data. Missing values were imputed by mean substitution; variables with >20% missingness were excluded. Candidate predictors were selected using LASSO regression. A multivariable Cox proportional hazards model-including restricted cubic splines to evaluate nonlinearity-was used to estimate the association between the De Ritis ratio and 28-day mortality, adjusting for potential confounders such as Sequential Organ Failure Assessment (SOFA)/Acute Physiology and Chronic Health Evaluation Ⅱ (APACHE Ⅱ) scores, organ support measures, and relevant biomarkers. Model discrimination and calibration were assessed using ROC-AUC and calibration plots. Preplanned subgroup analyses explored effect modification by comorbidities. Results The De Ritis ratio was an independent predictor of 28-day mortality. KaplanMeier analysis showed significantly lower survival in the highest quartile (Q4) compared with lower quartiles (P<0.01). In univariate Cox models, mortality risk increased with higher De Ritis quartiles (Q3: HR=2.073, 95%CI: 1.477-2.91; Q4: HR=2.066, 95%CI: 1.452-2.939; both P<0.01). The multivariable model retained the De Ritis ratio as an independent predictor and achieved an AUC of 0.764. Restricted cubic spline analysis identified an inflection point at a De Ritis ratio of approximately 2: Risk increased with rising ratio up to 2, while the relationship beyond 2 was less consistent. Decision curve analysis demonstrated clinical net benefit across a wide range of threshold probabilities. Subgroup analyses indicated stronger predictive value in older patients and in those without acute kidney injury, cirrhosis, or coronary artery disease. Conclusion The De Ritis (AST/ALT) ratio is a promising prognostic biomarker for stroke. It may help clinicians better estimate shortterm mortality risk, inform treatment decisions, and ultimately improve patient outcomes.

Key words: stroke, De Ritis ratio, MIMIC-Ⅳ database

CLC Number: