Model for end-stage liver disease, model for end-stage liver disease-sodium and Child-Turcotte-Pugh scores over time for the prediction of complications of liver cirrhosis

Background/Aims: There has been no report concerning the predictive capability of each scoring system in determining the development of complications of liver cirrhosis such as variceal bleeding and/or hepatic encephalopathy. Methods: We retrospectively studied 128 patients with liver cirrhosis [92 males; mean (standard deviation) 54.2 (11.2) years] admitted to our institution from March 2004 to April 2006. Seventy-three patients (57.0%, group 1) were admitted because of complications of cirrhosis and 55 patients (43.0%, group 2) were admitted for causes unrelated to complications of cirrhosis. We calculated values for model for end-stage liver disease (MELD), MELD-sodium (MELD-Na) and Child-Turcotte-Pugh (CTP) scores on admission and at 3 and 6 months before admission. Each delta score was defined as the difference in the scores of 3 and 6 months before admission. Results: The relative risk for complications in the patients with ΔMELD/3 months ≥ 1.35, ΔMELD-Na/3 months ≥ 1.35 and ΔCTP/3 months ≥ 1 was 2.05 [95% confidence intervals (CI) 1.47-2.85, P < 0.01], 2.04 (95% CI 1.45-2.88, P < 0.01) and 1.98 (95% CI 1.39-2.81, P < 0.01) respectively. The area under the receiver-operating characteristic curves of ΔMELD/3 months, ΔMELD-Na/3 months and ΔCTP/3 months for the occurrence of cirrhotic complications were 0.691, 0.694 and 0.722 respectively. A higher ΔMELD/3 months (≥1.35), ΔMELD-Na/3 months (≥ 1.35) and ΔCTP/3 months (≥ 1) was associated with decreased survival. Conclusions: Delta model for end-stage liver disease/3 months, ΔMELD-Na/3 months and ΔCTP/3 months were clinically useful parameters for predicting the occurrence of cirrhotic complications.