Search

The lung allocation score prioritizes candidates for a lung transplant in the United States. As the country adopts the continuous distribution framework for organ allocation, we must reevaluate lung allocation score assumptions to maximize transplant benefit.We used Scientific Registry of Transplant Recipients data to study the impact of these changes: (1) updating cohorts; (2) transitioning from 1- to 5-year posttransplant survival; (3) using time-varying effects for non-proportional hazards; and (4) weighting waitlist and posttransplant area under the curve differently. Models were compared using Spearman correlations and C-statistics. The thoracic simulation allocation model characterized transplant rates and proportions of recipient subgroups under the current and new systems.Posttransplant areas under the curve models were estimated with recipients aged ≥12 from January 1, 2014, to December 31, 2018. All models had similar C-statistics and Spearman correlations, indicating similar predictive performance and posttransplant area under the curve rankings. Five-year posttransplant area under the curve across age and diagnosis groups varied more than 1-year groups. Using the thoracic simulation allocation model, 1- and 5-year posttransplant model under the curve models showed similar transplant rates and recipient characteristics under the current system, but under continuous distribution, 5-year posttransplant area under the curve resulted in increased transplant rates with more recipients younger and in diagnosis groups B and C.Incorporating equally weighted waitlist and posttransplant models using 5-year posttransplant survival detected the largest variability in survival under the continuous distribution system, which could improve long-term survival in the United States.

BACKGROUND The thoracic simulated allocation model (TSAM) is used by the Scientific Registry of Transplant Recipients to predict the relative effect of organ allocation policy changes. A new lung allocation policy changing the first unit of allocation from donation service area to 250 nautical miles took effect on November 24, 2017. We studied TSAM's ability to correctly predict trends caused by changes in allocation policy. METHODS We compared the population characteristics from the TSAM cohort, 6,386 lung transplant candidates from 2009 to 2011, with the observed cohort of 7,601 candidates from the year before the policy change on November 24, 2017, and the year after. Simulations were run 10 times. Waitlist mortality and transplant rates were calculated and compared with observed mortality and transplant rates in the years before and after the policy change. RESULTS TSAM correctly predicted no change in overall waitlist mortality or transplant rates with the policy change. Observed waitlist mortality values were higher, as were transplant rates, because of increased organ donation and population change. TSAM predicted increased transplant rates for diagnosis group D (idiopathic pulmonary fibrosis), decreased rates for group A (chronic obstructive pulmonary disease), and increased rates for candidates with lung allocation score ≥50, but these changes did not occur in the waitlist and transplant populations after the policy change. CONCLUSIONS TSAM correctly predicted the relative trends caused by a change in allocation policy but smaller sub-group predictions were not seen.