Background: Cirrhotic patients have many critical situations. The recognition of risk factors that can stratify a population of cirrhotic patients into subgroups with different survival is of great prognostic value for the clinician. Numerous attempts have been made to develop a reliable prognostic survival model for cirrhosis. (Chatzicostas et al 2003) Some current ICU prognostic models [APACHE, OSF and SOFA] were used to stratify cirrhotics into risk categories, but few cirrhotics were included in the original model development. (Cholongitas et al 2006) Liver-specific scores [CTP and MELD] might be useful in these circumstances.Objectives: evaluate whether ICU prognostic models perform better compared with liver-disease specific ones in cirrhotics admitted to ICU and defining the predictors of poor mortality in a group of cirrhotic patients admitted to Hepatology intensive care unit.Methods: The survival/mortality of 301 cirrhotic patients admitted from July 2007 to March 2012 was respectively studied. The CTP, MELD, and APACHE II scores were computed for each patient within the first 24 hours of their admission. Patient disposition was classified as either survival or non survivors. Group comparisons between survivors and non-survivors were carried out using multiple logistic regressions to identify risk factors directly correlated with mortality. Finally, sensitivity, specificity, positive Predictive and negative predictive values were computed for the three prognostic scoring systems.Results: 172 patients non-survivors (57.1%) and 129 patients were survivors (42.9%). CTP, APACHE Π and MELD mean scores of survivors (10.2, 22.7, 21.8Respectively) were lower than non-survivors (11.3, 26.6 and 31.4, respectively) (p<0.001). Sensitivity of CTP, APACHEΠ and MELD were (75.6%, 72% and 86.6% respectively), the overall predictive accuracy of MELD was 81% greater than the CTP (67%) and APACHE Π (71%) scores. Usage of mechanical ventilation and vasopressors were associated with high mortality.Conclusion: The present study showed that MELD score has the highest overall predictive accuracy among the three scoring systems. However, it is prudent to say, that these scoring systems should always go hand in hand with critical clinical analysis and good decision making.