Background The critical need for precise risk stratification in severe liver cirrhosis is underscored by its substantial 30-day mortality rates, demanding reliable tools to guide clinical interventions.

Objective To establish a machine learning-driven prognostic model for short-term mortality prediction in decompensated cirrhosis through comprehensive analysis of critical care data.

Methods This retrospective cohort study analyzed 1,044 carefully curated cases from the MIMIC-IV database, randomly divided into training (n=740) and validation (n=304) sets. We developed a machine learning model incorporating multidimensional clinical parameters, with rigorous evaluation and internal validation. Short-term survival was analyzed via bootstrap-validated Cox proportional hazards regression. Prognostic heterogeneity across INR-based strata was examined.

Results The final prediction model incorporated eight significant predictors: age (OR 1.040, 95% CI 1.022-1.058), international normalized ratio (OR 1.496, 95%CI 1.297-1726), creatinine (OR 1.210, 95%CI 1.109-1.319), platelets (OR 0.995, 95%CI 0.993-0.997), leukocytes (OR 1.120, 95%CI 1.083-1.159), total bilirubin (OR 1.031, 95%CI 1.006-1.057), peptic ulcer (OR 0.297, 95%CI 0.120-0.734), and metastatic solid tumors (OR 3.001, 95%CI 1.211-7.436). The model demonstrated excellent discrimination with an AUC of 0.851 (95%CI: 0.800–0.901) in the validation cohort. Cox regression analysis confirmed these findings and identified additional associations with aspartate aminotransferase and red blood cell levels. Subgroup analysis revealed significant mortality variations across different INR ranges (P<0.001).

Conclusions Our prediction model identifies high-risk cirrhotic patients and highlights critical prognostic factors, offering clinicians a valuable tool for risk stratification and timely intervention. The strong correlation between laboratory markers, complications, and outcomes underscores the importance of close monitoring in this population.

The authors have declared no competing interest.

Yes

I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.

Yes

The details of the IRB/oversight body that provided approval or exemption for the research described are given below:

We have obtained access to the MIMIC database without the need for ethical review.

I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.

Yes

I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).

Yes

I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.

Yes