Abstract
Survivors of pediatric liver transplantation are at risk for developing complications related to posttransplant immunosuppressive medications. Withdrawal is possible in selected patients but carries the risk of graft rejection and loss. We modeled the effect of withdrawing immunosuppressive medications on survival, cost, and quality-adjusted life-years (QALYs) in a hypothetical cohort of pediatric patients who received transplantation for biliary atresia with stable liver enzymes and no recent episodes of rejection, and who were free from immunosuppression-related adverse effects. A decision analysis tree was developed, and Monte Carlo simulations were used to track patients through the model during a 10-year time course with 1-year cycles. Data from the literature were used to assign probabilities to major clinical events and preference-based utility scores to the values of health outcomes. One-way and probabilistic sensitivity analyses were used to evaluate the impact of uncertainty. Patients following the withdrawal strategy had a 10-year survival rate of 95.8% and experienced 8.61 QALYs versus 88.6% survival and 8.01 QALYs for those taking immunosuppressive medications. Each additional QALY is attained at a cost of -$18,992.41 and was therefore cost saving. Patients in our model who had their immunosuppression withdrawn had improved survival and QALYs with lower costs. Although every effort was made to validate the model, it is limited by the accuracy of the underlying assumptions. Therefore, clinical trials are needed to determine predictors of successful immunosuppression withdrawal to allow for personalization of medication regimens. Survivors of pediatric liver transplantation are at risk for developing complications related to posttransplant immunosuppressive medications. Withdrawal is possible in selected patients but carries the risk of graft rejection and loss. We modeled the effect of withdrawing immunosuppressive medications on survival, cost, and quality-adjusted life-years (QALYs) in a hypothetical cohort of pediatric patients who received transplantation for biliary atresia with stable liver enzymes and no recent episodes of rejection, and who were free from immunosuppression-related adverse effects. A decision analysis tree was developed, and Monte Carlo simulations were used to track patients through the model during a 10-year time course with 1-year cycles. Data from the literature were used to assign probabilities to major clinical events and preference-based utility scores to the values of health outcomes. One-way and probabilistic sensitivity analyses were used to evaluate the impact of uncertainty. Patients following the withdrawal strategy had a 10-year survival rate of 95.8% and experienced 8.61 QALYs versus 88.6% survival and 8.01 QALYs for those taking immunosuppressive medications. Each additional QALY is attained at a cost of -$18,992.41 and was therefore cost saving. Patients in our model who had their immunosuppression withdrawn had improved survival and QALYs with lower costs. Although every effort was made to validate the model, it is limited by the accuracy of the underlying assumptions. Therefore, clinical trials are needed to determine predictors of successful immunosuppression withdrawal to allow for personalization of medication regimens.
