The Authors’ Reply
Indeed, the global aim of the new French allocation system is to minimize death on the waiting list and offer the most equitable allocation system. The system will extend nationwide graft sharing to all candidates.
The implementation of a benefit-based approach to graft allocation requires accurate estimates of posttransplant mortality. Yet existing models for predicting posttransplant mortality have only moderate discriminatory performance.2,3 The absence of donor factors and consideration for donor-recipient interactions may explain the limited accuracy of these models. Within this context, we are developing a French data-driven Transplanted Patient Risk Score including recipient and donor factors in addition to donor-recipient interactions. If this model has a good predictive accuracy, we would consider introducing posttransplant survival in the allocation system.
In addition, Thomas Egan raises question about use of candidate characteristics only at listing. In the future allocation system, the change over time of the components of the risk score will be taken account with mandatory updates of the variables included in the score every 3 months in stable candidates and every 3 days in patients requiring inotropic infusion and/or short-term mechanical circulatory support.
Finally, we expect that an urgency-based cardiac allocation system using candidate’s characteristics instead of therapies and medical management can be exported to other countries. Because of differences in graft availability, candidate characteristics and management and healthcare systems from one country to another, a validation of the CRS is required before use in other countries.