Backgrounds/aims: The change of MELD (Model for End-stage Liver Disease) score over time (DeltaMELD) has been proposed as a tool to predict the survival in cirrhotic patients. The aims of the study were to assess ability of the DeltaMELD to predict the survival and compare them with the initial MELD and CP score.
Methods: MELD score was serially determined at least twice with more than two-month interval in 120 cirrhotic patients. We analyzed the clinical factors associated with the variation of MELD score. The predictive power of 6, 12 and 24 months mortality for DeltaMELD, initial MELD and CP score was compared by c-statistics. Patient survival was also compared at 6, 12 and 24 months according to the cut off values of DeltaMELD/month, initial MELD and CP score.
Results: Increased MELD score was associated with biochemical and clinical parameters such as esophageal variceal bleeding and onset of hepatic encephalopathy. The area under receiver operating characteristic (ROC) curve for DeltaMELD/month was 0.928 (P<0.001) compared with 0.575 for MELD score and 0.636 for CP score at 6 month-mortality; the area was 0.727, 0.594 and 0.657 at 12 month-mortality; 0.693, 0.587 and 0.639 at 24 month-mortality, respectively. The patients with DeltaMELD/month more than 1.0 had resulted in the higher mortality at 6, 12 and 24 months. The DeltaMELD/month was associated with mortality and was an independent prognostic predictor with a risk ratio of 1.679 (95% CI: 1.381-2.042, P<0.001).
Conclusions: Determination of DeltaMELD could be better prognostic predictor for patients with liver cirrhosis than initial MELD and CP score.