OBJECTIVE Pediatric acute liver failure is a rare disorder which results in death or the need for liver transplantation in 25-50% of cases. The adults scores are unable to predict survival without liver transplantation of pediatric patients. The present study assessed the use the of indocyanine green plasma disappearance rate as a tool to predict the evolution of pediatric patients with acute liver failure. METHODS All patients met the criteria of acute liver failure according to the Pediatric Acute Liver Failure Study Group. King's College, Clichy's criteria and ICG-PDR were obtained on admission or when acute liver failure was diagnosed and repeated every 12-24 hours, respectively. RESULTS Thirteen out of 48 patients suffered an irreversible liver damage. Seven of them underwent a liver transplantation and 6 died on the waiting. A total of 154 ICG-PDR measurements were taken during the study (Median 12.4 %/min, r:6.2 - 26.3). The ICG-PDR was significantly lower in patients who suffered irreversible liver damage compared with those who survived without liver transplantation (median ICG-PDR 4.1 %/min; r:4.0 - 5.7 vs median ICG-PDR 20.3 %/min; r: 9.1 - 30.1; respectively. P < 0.001). Using a ROC curve the cutoff of ICG-PDR for assessing the need for liver transplantation was set at 5.9 %/min (sensitivity 92.3%, specificity 97.1%). Sensitivity, specificity, PPV, NPV and DA for ICG-PDR were higher than the King's College and Clichy's criteria. CONCLUSIONS ICG-PDR is a powerful tool that would improve the categorization of patients with pediatric acute liver failure.