Résumé
BACKGROUND: Patients with decompensated liver cirrhosis usually resulted in admission to the intensive care unit (ICU) during hospitalization. When admitted to the ICU, the mortality was high. The aim of this study is to identify multiple prognostic factors for mortality and to analyze the significance of prognostic survival model with each scoring system in patients with decompensated liver cirrhosis who was admitted to the ICU. METHODS: From January 2008 to December 2008, 60 consecutive patients with decompensated liver cirrhosis were admitted in the ICU and retrospectively reviewed. Prognostic models used were Child-Turcotte-Pugh (CTP), model for end-stage liver disease (MELD), model for end-stage liver disease with incorporation of serum sodium (MELD-Na), acute physiology and chronic health evaluation (APACHE) II, and sequential organ failure assessment (SOFA). The predictive prognosis was analyzed using the area under the receiver's operating characteristics curve (AUC). RESULTS: The median follow up period was 20 months, and ICU mortality was 17% (n = 10). A total of 24 patients (40%) died during the study period. The average survival of five prognostic models was related with the severity of the disease. All of the five systems showed significant differences in the cumulative survival rate, according to the scores on admission, and the MELD-Na had the highest AUC (0.924). Multivariate analysis showed that bilirubin and albumin were significantly related to mortality. CONCLUSIONS: The CPT, MELD, MELD-Na, APACHE II, and SOFA may predict the prognosis of patients with decompensated liver cirrhosis. The MELD-Na could be a better prognostic predictor than other scoring systems.