BACKGROUND
Pediatric acute liver failure (PALF) is a rare yet severe condition characterized by rapidly progressing hepatic dysfunction and multisystem organ failure (MOF) syndrome in previously healthy children. It poses several clinical challenges to physicians, one of them being the identification of the children who won’t recover without liver transplantation (LT).
OBJECTIVES
This study was aimed at reviewing the experience of our pediatric intensive care unit and identifying mortality risk factors. Our results might also enrich our reflection about living donor liver transplantation (LDLT) and its indication in the difficult ethical context of PALF.
PATIENTS AND METHODS
We collected the retrospective data of all PALF children admitted from March 1989 to December 2016. Demographic, clinical, laboratory, and short-term outcome data were recorded. Statistical analysis was performed in order to identify mortality risk factors.
RESULTS
In total, 100 children (54 girls) were included. Their mean age was 59.9 months, with 40% of them being less than 2 years old. Hepatic encephalopathy, shock, and respiratory failure were diagnosed in 73, 39, and 32 patients, respectively. The main causes of PALF varied depending on age, and etiology remained undetermined in 35 % of cases. In our study population, 14 children had contraindications to LT and died, 24 did not meet our criteria for LT, and 22 survived. Furthermore, 62 children were considered as candidates for LT, among whom 36 received a liver from a deceased donor and 3 underwent LDLT. 11 patients who were on the LT waiting list died and 12 were withdrawn from this list because of spontaneous recovery. Overall survival was 61%; survival after LT was 69%. Statistical analysis identified respiratory failure (OR= 2.9), acute renal failure (OR=2.9), and blood lactate level (OR =1.5) as significant mortality risk factors. A risk staging system was developed based on these three factors to predict mortality in children (AUROC =0.83 [0.74 – 0.91] p<0.001). The mean waiting time before LT was 2 days. When comparing two periods, 1989-2002 (n=33 children) and 2003-2016 (n=6), we observed an increase in the mean waiting time, from 1.5 to 4.8 days. Children died on average 6 days after being registered on the LT waiting list.
CONCLUSION
Our results confirm the poor outcome of PALF and suggest that respiratory and acute renal failures should be added to the LT criteria. Given that most children candidates for LT died from MOF and that the waiting time seems to increase, LDLT should be proposed 3 or 4 days after the decision regarding LT was made.