Model for End-Stage Liver Disease (MELD) score system to evaluate patients with viral hepatitis on the waiting list: better than the Child-Turcotte-Pugh (CTP) system?

BACKGROUND: The Model for End-Stage Liver Disease (MELD), based on creatinine, bilirubin, and International normalized ratio (INR), has been shown to be superior to the Child-Turcotte-Pugh (CTP) score in predicting 3-month mortality among patients on the transplant waiting list due to end-stage liver disease (ESLD). An additional advantage of MELD is the possibility to add "adjustment points" for exceptional patients at risk for death because of liver disease not identified by changes in the used parameters, as occurs in the case of hepatocellular carcinoma (HCC). Although it is useful, MELD has some important limitations: There are no differences for patients with or without ascites, and for the absence of other laboratory parameters involved in the etiology of disease. In this study, we evaluated dropouts of patients on the waiting list for orthotopic liver transplantation (OLT) based upon the characteristics of these subjects before and after introduction of the MELD score.

METHODS: All patients on the OLT waiting list from June 1, 2006 to June 30, 2007 were enrolled in the MELD group (A) and evaluated with CHILD and MELD score, while those listed from January 1, 2004 to May 31, 2005 were enrolled in pre-MELD group (B) to be evaluated with CHILD. In these subjects we assessed the drop out frequency and waiting time and we compared the results to assess possible differences (U Mann-Whitney Test; P<.05).

RESULTS: The total number of patients included in this study was 176: 116 patients in Group A and 60 in Group B. We had a drop-out frequency of 21% with a median of 9+/-6 S.E. months in Group A, while 9% with a median of 15+/-8 months S.E. in Group B. The dropout frequencies were as follows: Group A--16 deaths (1 HCC--15 disease complications) while in Group B we had 13 drop outs, 10 exitus (4 HCC and 6 disease complications) and three exclusions for nonmedical reasons. In Group A we had a higher number of deaths due to disease complications than in group B (P<.05). Further, we had 32 OLTx in Group A and 45 in Group B. Survival rate did not show any differences between the two groups while number needed to harm was 11.

CONCLUSIONS: The use of MELD score in this group of patients produced an advantage for HCC, but seemed to cutoff patients with viral hepatitis complications during the waiting time. Particularly, about one in every 11 patients may receive an harm using this score system. Other parameters should be introduced as adjustment points to make the MELD score suitable also for patients with infectious liver diseases.