Sequential analysis of variable markers for predicting outcomes in pediatric patients with acute liver failure.

AIM: Our aim was to analyze serial changes in the predictive variables and a scoring system retrospectively adapted to evaluate outcomes in pediatric patients with acute liver failure (ALF).

METHODS: We retrospectively collected data on 65 patients with ALF. The 65 patients were divided into two groups according to the need for liver transplantation (LT) as follows: LT group (n = 54) and non-LT group (n = 11). The early determination scoring system of the indications for LT proposed by the Intractable Hepato-Biliary Diseases Study Group of Japan (JIHBDSG) was used in our study. The area under the receiver operating characteristic curve (AUROC) was calculated for the JIHBDSG score between the LT group and non-LT group at the time of diagnosis (day 0) and day 3, and day 5 after the diagnosis.

RESULTS: A JIHBDSG score of >3 at day 5 was found to identify the patients requiring LT with 83.7% sensitivity, 81.8% specificity, and 83.3% diagnostic accuracy. Based on a comparison of AUROC values, the JIHBDSG score on day 5 (AUROC 0.91) was higher than that on day 0 (AUROC 0.75) and day 3 (AUROC 0.84).

CONCLUSION: We showed that a serial analysis of the JIHBDSG score might be useful for predicting outcomes of ALF in pediatric patients who fulfilled the criteria of LT indication in our center. However, further studies are needed to validate our results.