Monday, October 23, 2023
Every year in the United States, approximately 50 children are placed on transplant waiting lists due to acute liver failure. This condition, which can affect previously healthy children, often escalates rapidly, making timely decisions crucial. While viral hepatitis and drug-induced liver injury have been associated with pediatric acute liver failure, the underlying cause remains undetermined in at least half of the cases. When a child with acute liver failure is admitted to the emergency room, medical teams are faced with a tight timeframe to determine whether a liver transplant is necessary. It's important to note that a liver transplant is a major procedure, involving lifelong immunosuppressant drug use and various complications.
Existing decision support tools such as the Kings College Hospital criteria (KCHC), the pediatric end-stage liver disease (PELD) score, and the liver injury unit (aLIU) score are not well-suited for children with acute liver failure. These tools were originally developed for different purposes: KCHC for adults, PELD for measuring chronic liver disease in children, and aLIU for single-time lab result assessment, failing to adequately account for the instability often seen in children with acute liver failure.
To address this issue, a team of researchers developed what is known as the CHALF score. This score was created using a machine learning-based model trained on data from common tests administered to 147 pediatric acute liver failure patients at CHLA, both in the emergency room and upon hospital admission. By analyzing various factors including patient demographics, diagnoses, and laboratory results throughout their hospitalization, the researchers employed statistical methods to identify the clinical tests and values that could most accurately predict whether a child could survive with their own liver or if a transplant was necessary to prevent liver failure-related mortality.
Source: businesswire.com
