Pediatric acute liver failure (PALF) is a complex disease that has multiple causes and is associated with a high risk of death. For a child with PALF, liver transplantation can be a lifesaving procedure. However, there are two primary reasons that the decision to pursue transplantation is difficult for the child's clinicians and family members. First, there is a very real chance that the child's liver will recover on its own if the child is able to survive for enough time without a transplant. Second, in nearly 50% of cases of PALF, the cause of the disease is unknown, and this makes it very difficult to predict the outcomes for these patients in the absence of transplantation. Clinicians are faced with the dilemma of whether to place a child on the list to receive a liver from a deceased donor or whether to wait for the child to recover. A child's family member must decide whether to become a living donor, as occurs in over 14% of cases in which patients receive transplants for acute liver failure. There is some concern that transplantation may be pursued """"""""too early,"""""""" when waiting may allow the damaged liver to recover without surgery. To address this problem, the proposed study will use data from the PALF registry, which is the largest and most clinically detailed longitudinal database of PALF patients in the world. The study will be a collaborative effort of two research groups that are uniquely suited to evaluate factors influencing when to pursue transplantation. The Principal Investigator's modeling group has extensive experience in mathematically modeling the progression of liver disease, applying simulation models to organ allocation policies, and representing complex medical decisions. The overall goals of the study are to improve clinical decision making in cases of PALF by (1) developing quantitative models of prognosis and disease progression and (2) using these models to calibrate optimization models of the transplant decision. These goals will be accomplished through two specific aims.
Aim 1 is to characterize and quantitatively describe the short-term natural history of children with PALF. This will require estimating statistical models of pretransplant and post transplant survival, as well as developing quantitative models of disease progression.
Aim 2 is to construct a proof-of-concept model evaluating the optimal timing of liver transplantation for children with PALF, including the possibility of transplanting an organ from a living donor. We will modify our existing simulation model of the US organ allocation system to include PALF, will validate the model to ensure that it represents the time course of PALF, and will use the model to examine transplant timing practices in children with PALF.
Acute liver failure in children is a devastating, often fatal condition for which one of the potential treatments is liver transplantation. However, the decision to choose to treat a child with a transplant is difficult, because even if a child is very ill, there is a chance that the child's liver would recover on its own, without surgery, and that the transplant would therefore be unnecessary. The best strategy for deciding when to wait a little longer or when to undertake transplantation is not known with certainty. The research we describe here will use statistics and mathematical models to better predict how fast liver disease will progress in children. The goal of the research is to improve decisions regarding the use of liver transplantation for young patients with acute liver disease.
