In cirrhosis, several prognostic scores have been developed to predict overall survival. The MELD (Model for End-Stage Liver Disease), MELD-Na and MELD3.0 scores have been proposed as a criteria for allocating organs.

The aim of our study was to explore MELD 3.0 score for predicting survival in decompensated Tunisian cirrhotic patients. 

We conducted an observational, retrospective study involving all patients with decompensated cirrhosis managed in Gastroenterology department of Taher Maamouri hospital during the period from January 2010 to December 2021. Mortality was assessed using the Kaplan Meier model. The performance of the scores was assessed using the AUROC (Area under the Receiver Operating Characteristic curve).

One hundred and fifty-two patients were included with an average age of 62 years. The main causes of chronic liver disease were hepatitis C virus (HCV) infection (30.9%) and hepatitis B virus (HBV) infection (32.2%).  The mean value of the MELD, MELD-Na and MELD3.0 scores was 17.78, 17.74 and 19.4, respectively. The mean follow-up time was 17.1 months. The overall mortality at 3-month, 6-month and 1-year was 18.4%, 24.3% and 35.5% respectively.

In the prediction of 3-month mortality, the AUROCs of the MELD, MELD-Na and MELD3.0 scores were 0.786; 0.841 and 0.850 respectively. The cut-off value of the MELD3.0 score was 24 (sensitivity 68%, specificity 88%). A statistically significant correlation was noted between the MELD3.0 score ≥24 and 3-month mortality (p=0.04).

Overall, MELD3.0 shows promising value with the highest AUROC in all assessed scoring systems. MELD 3.0 has the best performance for predicting mortality at various time points. Multicenter and long-term studies with larger samples would be helpful for the scoring systems to predict the mortality more precisely in decompensated liver cirrhosis patients.