Background/aims: MELD-Na (model for end-stage liver disease with incorporation of serum sodium) was suggested to provide better survival prediction than MELD alone for patients with end stage liver disease. However, there is no data verifying the usefulness of MELD-Na for predicting short term mortality of cirrhotic patients in Korea. This study was aimed to determine whether MELD-Na would be more accurate in predicting short term mortality than other scoring systems such as Child-Turcotte-Pugh (CTP) or MELD.
Methods: Data from 355 patients admitted due to liver cirrhosis were retrospectively reviewed. The cumulative survival rates were obtained. Prediction of mortality rate for three months and one year were analyzed using the area under the receiver's operating characteristics curve (AUC).
Results: One hundred patients (28%) died during the study period. All of the three systems showed significant differences in the cumulative survival rate according to the scores on admission (p0.001). The AUC of CTP, MELD, and MELD-Na in predicting three-months mortality were 0.828, 0.845, and 0.862 (p0.05), and the AUC of each score system for death within one year were 0.792, 0.800, and 0.831, respectively (p0.05). The AUC of MELD-Na in predicting short term death were the highest, although it was not statistically significant. Multivariate analysis showed that only MELD-Na was significantly related to three-month mortality (p=0.012).
Conclusions: MELD-Na is more appropriate in predicting short term mortality, but larger scale studies are needed to confirm the superiority of MELD-Na to MELD and CTP in patients with liver cirrhosis.