Abstract
ObjectiveCirrhosis with acute kidney injury is associated with a high mortality rate, particularly in patients receiving continuous renal replacement therapy. This study aimed to develop a nomogram to predict in-hospital mortality in patients with cirrhosis and acute kidney injury receiving continuous renal replacement therapy.MethodsA retrospective study was conducted using the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. Patients with cirrhosis with acute kidney injury who underwent continuous renal replacement therapy were identified. Predictors were selected using least absolute shrinkage and selection operator regression, and a multivariable logistic regression model was developed. Model performance was evaluated using area under the receiver operating characteristic curve (AUC), calibration analysis, and decision curve analysis. Internal validation was performed via bootstrapping.ResultsThe following three independent predictors of in-hospital mortality were identified in the 452 eligible patients: international normalized ratio at continuous renal replacement therapy initiation, presence of sepsis, and vasoactive drug use. The nomogram showed strong discrimination (AUC = 0.828, 95% confidence interval: 0.790-0.866) and consistent internal validation (AUC = 0.825, 95% confidence interval: 0.781-0.870). Calibration and decision curve analyses indicated good agreement and clinical usefulness.ConclusionsA nomogram incorporating international normalized ratio, sepsis, and vasoactive drug use may help estimate short-term mortality risk in patients with cirrhosis and acute kidney injury receiving continuous renal replacement therapy. As the analysis was based on single-center data, and no external validation was performed, the findings should be interpreted cautiously and verified in future studies.