Can inclusion of serum creatinine values improve the Child–Turcotte–Pugh score and challenge the prognostic yield of the model for end-stage liver disease score in the short-term prognostic assessment of cirrhotic patients?


  • *Preliminary data contained in this paper were published as a Letter to the Editor in Gastroenterology 2003; 125: 993–4.

Prof. Roberto Testa, Gastroenterology Unit, Department of Internal Medicine (DIMI), Viale Benedetto XV no. 6 16132 Genova, Italy.
Fax: +39 010 3538638


Abstract: Background: The model for end-stage liver disease (MELD) score is a useful tool to assess prognosis in critically ill cirrhotic patients. However, its short-term prognostic superiority over the traditional Child–Turcotte–Pugh (CTP) score has not been definitely confirmed. The creatinine serum level is an important predictor of survival in patients with liver cirrhosis.

Aims: To evaluate and compare the short-term prognostic accuracy of the CTP, the creatinine-modified CTP, and the MELD scores in patients with liver cirrhosis.

Methods: CTP, creatinine-modified CTP, and MELD scores were calculated in a cohort of 145 cirrhotic patients. The creatinine-modified CTP was calculated as follows: we assessed the mean creatinine serum level and standard deviation (SD) of the 145 study patients, then assigned a score of 1 to patients with creatinine serum levels ≤ to the mean, a score of 2 to patients with creatinine levels between the mean and the mean+1 SD, and a score of 3 to patients with creatinine levels above the mean+1 SD. The creatinine-modified CTP was then calculated by simply adding each patients' creatinine score to their traditional CTP scores. We calculated and compared the accuracy (c-index) of the three parameters in predicting 3-month survival.

Results: The creatinine-modified CTP score showed better prognostic accuracy as compared with the traditional CTP (P=0.049). However, the MELD score proved to be better at defining patients' prognosis in the short-term as compared with both the traditional CTP score (P=0.012) and the creatinine-modified CTP (P=0.047). The excellent short-term prognostic accuracy of the MELD score was confirmed even when patients with abnormal creatinine serum levels were excluded from the analysis (c-index=0.935).

Conclusions: Adding creatinine values to the CTP slightly improves the prognostic usefulness of the traditional CTP score alone. The MELD score has a short-term prognostic yield that is better than what is provided by both the CTP and CTP creatinine-modified scores, even in cirrhotic patients who are not critically ill. The positive results obtained by using the MELD score were confirmed even after excluding patients with impaired renal function.