Key takeaways
-
This study aimed to develop the first clinically relevant tools to reliably predict outcomes in neonatal acute liver failure (ALF).
-
The use of existing predictive models for pediatric ALF is limited in neonatal ALF due to significant differences between the two populations.
-
Researchers identified the high prognostic value of alpha-fetoprotein at admission and developed admission and peak models for neonatal ALF that accurately predict outcomes.
-
Findings from this study may help clinicians with risk stratification to better counsel families and inform management decisions, such as liver transplantation candidacy.
Research study background
Neonatal acute liver failure (ALF), a rare condition with high mortality rates in infants under 30 days old, lacks reliable clinical or laboratory tools for prognosis. Elevated serum alpha-fetoprotein (AFP) values, proteins produced by the liver and yolk sac of a fetus, have been proposed as a factor associated with better outcomes in neonatal ALF. However, the prognostic accuracy has not been well defined. Prediction models established for pediatric ALF aren’t fully applicable for neonates due to a number of factors, including significant differences in etiology, management, outcomes and the lack of neonates included in ALF research.
Researchers in the Pediatric Liver Center at Children’s Hospital Colorado recognized the critical need for prognostic biomarkers to guide family counseling and clinical decision-making for liver transplantation. For this study, published in the Journal of Pediatrics, they analyzed neonatal ALF cases admitted between 2005 and 2022. Investigators aimed to develop neonatal-specific predictive models for survival with native liver (SNL) and to compare their accuracy to existing models for pediatric ALF. Neonatal ALF was defined as acute liver injury with prothrombin time (PT) ≥20 seconds or international normalized ratio (INR) ≥2. Among the 51 neonates studied, the most common causes of ALF were ischemia/hypoxia, infection and gestational alloimmune liver disease (GALD). Of these patients, 22 were SNL, 27 died without transplant and two received a liver transplant, one of which died.
First, researchers looked to AFP levels. Compared to neonates who received transplants or died, those with SNL had significantly higher AFP levels at admission. This difference was even more pronounced when looking at patients without a diagnosis of GALD, a disease where patients are known to have elevated levels of AFP. In non-GALD cases, lower AFP levels predicted death or liver transplant with 75% sensitivity and 100% specificity. The team then developed two neonate-specific models, an admission model with INR and ammonia and a peak model with PT and ammonia. The higher the neonatal ALF score, the lower the SNL rates. Both models predicted outcomes with good accuracy.
Clinical applications
This study highlights the high prognostic value of AFP at admission, especially in neonates without GALD, and introduces the first neonate-specific ALF models with clinically relevant admission scores. These findings enhance understanding of neonatal ALF and offer tools to care for these patients as a distinct clinical subset of pediatric ALF. While these models show promise in improving liver transplant decisions and reducing mortality, further studies are needed to validate their accuracy and impact.
Featured researchers
Priya Rolfes, MD
Pediatric Transplant Hepatology Fellow
Digestive Health Institute
Children's Hospital Colorado
![](/globalassets/doctors/sundaram-100.jpg?v=49e832)
Shikha Sundaram, MD
Medical Director, Liver Program
Gastroenterology and Transplant Hepatology
Children's Hospital Colorado
![](/globalassets/news/2022/cso-ron-sokol-3.jpg?v=490548)
Ronald Sokol, MD
Chief Scientific Officer, Child Health
Digestive Health Institute
Children's Hospital Colorado
Director, Colorado Clinical and Translational Sciences Institute
University of Colorado School of Medicine
![](/globalassets/doctors/sarah-taylor.jpg?v=48d78d)
Sarah Taylor, MD
Pediatric gastroenterologist
Digestive Health Institute
Children's Hospital Colorado
Associate professor
Pediatrics-Gastroenterology, Hepatology and Nutrition
University of Colorado School of Medicine