Evaluation of the OPTN Six-Status Heart Allocation System and a Combined Prognostic Model for Waitlist Risk Evaluation

KEY POINTSO_ST_ABSQuestionC_ST_ABSDoes the current national six-status heart allocation system rank waitlisted transplant patients based on medical urgency? Can this heart allocation systems prognostic performance be improved? FindingsThe six-status system is poorly calibrated, lacks sufficient statistical discrimination, and underestimates risk in the highest-risk patients, or those with an observed six-month mortality probability greater than 2%. By combining the current status system with six additional patient characteristics (previous transplant, ventilation, mean pulmonary capillary wedge pressure, willingness to accept a donor after cardiac death, diabetes status, and most recent creatinine), we correctly predicted greater than 80% of six-month waitlist mortalities in 7,706 study participants. MeaningThis study challenges the safety and efficacy of the current national heart allocation system. ImportanceIn December 2016, the Organ Procurement and Transplantation Network (OPTN) approved a bylaw that restructured the national heart allocation policy from a three-status to a six-status system. This new allocation system, which aimed to assign the highest priority to the patients with the highest mortality risk, went into effect on October 18, 2018. Since then, studies have identified limitations with the current system. However, no changes have been made to improve the national heart allocation system. ObjectiveGiven the clear importance and impact of ranking patients based on medical urgency, we carefully evaluated the six-status heart allocation system to determine its correlation with observed mortality, or calibration, and its ability to predict six-month patient mortality risk and waitlist survival. We identified six additional patient characteristics associated with waitlist mortality and combined them with the six-status score to significantly improve the current allocation systems ability to predict six-month waitlist mortality. DesignA retrospective, secondary analysis of the Scientific Registry of Transplant Recipients (SRTR) database of heart transplant candidates and recipients waitlisted from October 18, 2018, to December 31, 2024. SettingThe United States ParticipantsSingle-organ heart transplant candidates, 18 years of age and older who were placed on the waitlist (N = 19,275). Patients listed multi-organ transplantation were excluded. ExposuresAll-cause waitlist mortality Main Outcomes and MeasuresThe primary outcome of this study was the validation of the calibration and prognostic performance of the current heart allocation system. The secondary outcome is a simple model that greatly improves upon the current systems ability to accurately (>80%) predict waitlisted patient mortality. ResultsWith a mean calibration slope of 0.94 (0.66, 1.21) and an area under the receiver operating curve of 0.71 (0.47, 0.87), the current allocation system is poorly calibrated, has only moderate statistical discrimination, and underestimates patient risk in the most critically ill patients. Hazard and time-series regression analysis confirmed that the six-status system does not adequately rank patients based on medical urgency. Our combined model demonstrates that the national allocation system can be improved. Conclusions and RelevanceWhile the current heart distribution system accounts for some patient risk factors, a more objective and accurate model is needed to achieve the OPTNs strategic objective to more reliably model and predict patient risk and survival likelihood. Our model more accurately predicts patient waitlist mortality and will better inform waitlist management and improve waitlist survival by prioritizing medical urgency.