Biliary atresia (BA) is the leading indication to perform a pediatric liver transplantation (LT). Timely hepatoportoenterostomy (HPE) attempts to interrupt the natural history and allow for enteric bile flow; however, most patients who are treated with HPE require LT by the age of 10 years. We determined the cost-effectiveness of foregoing HPE to perform primary LT (pLT) in children with BA compared with standard-of-care HPE management. A Markov model was developed to simulate BA treatment over 10 years. Costs were measured in 2018 US dollars and effectiveness in life-years (LYs). The primary outcome was incremental cost-effectiveness ratio (ICER) between treatments. Model parameters were derived from the literature. In the base model, we assumed similar LT outcomes after HPE and pLT. Sensitivity analyses on all model parameters were performed, including a scenario in which pLT led to 100% patient and graft survival after LT. Children undergoing HPE accumulated $316,692 in costs and 8.17 LYs per patient. Children undergoing pLT accumulated $458,059 in costs and 8.24 LYs per patient, costing $1,869,164 per LY gained compared with HPE. With parameter variation over plausible ranges, only post-HPE and post-LT costs reduced the ICER below a typical threshold of $100,000 per LY gained. On probabilistic sensitivity analysis, 93% of iterations favored HPE at that threshold. With 100% patient and graft survival after pLT, pLT cost $283,478 per LY gained. HPE is more economically favorable than pLT for BA. pLT is unfavorable even with no graft or patient loss. The ability to predict those patients who may experience high costs after HPE or low costs after LT may help identify those patients for whom pLT could be considered.
ASJC Scopus subject areas