V CALISA1, J CRAIG1,2, S CHADBAN3, W LIM4, K HOWARD2, J CHAPMAN5, S MCDONALD6,7, P CLAYTON6, G WONG1,2,5
1Centre for Kidney Research, The Children’s Hospital at Westmead, Sydney, New South Wales, 2School of Public Health, University of Sydney, Sydney, New South Wales, 3Renal Transplant Unit, Royal Prince Alfred Hospital, Sydney, New South Wales, 4Department of Renal Medicine, Sir Charles Gairdner Hospital, Perth, Western Australia, 5Centre for Transplant and Renal Research, Westmead Hospital, Sydney, New South Wales, 6ANZDATA Registry, 7Central and Northern Adelaide Renal and Transplantation Service, Royal Adelaide Hospital, Adelaide, South Australia.
Aim: To determine the health benefits of risk-based, deceased-donor kidney allocation, compared to current practice.
Background: Risk-based allocation, using Kidney Donor Risk Index (KDRI) and Estimated Post-Transplant Survival (EPTS) score, has been proposed to improve utility and equity in allocation.
Methods: Risk-based matching algorithms, utilising KDRI-EPTS matching of the lowest percentile (10-30%) of scores, was applied to recipients transplanted in Australia, 1994-2013. Recipients >60 years were preferentially allocated grafts in the highest percentile (5-15%) of KDRI scores. Probabilistic Markov models, terminating in first graft failure or death, compared risk-based allocation against current practice. Incremental KDRI score, waiting time, life years, quality-adjusted life years and graft years under risk-based matching algorithms, compared to current practice, were determined.
Results: KDRI-EPTS matching reduced median KDRI scores by 0.19 (p<0.001) and waiting time for transplant by 0.64 years (p<0.001), for recipients ≤30 years but increased median KDRI score by 0.27 (p<0.001) and waiting time by 0.94 years (p<0.001) for recipients >60 years. Prioritising the top 15% of KDRI scores for recipients >60 years, decreased their median waiting time by 9.6% (p<0.001) but increased median KDRI by 11.5% (p<0.001). KDRI-EPTS matching threshold of 30% incurred maximum health benefit of 0.319 life years (p<0.001), across the population, with an increase of 0.4 life years (p<0.001) and a decrease of 0.31 life years (p=0.030) for recipients aged ≤30 and >60 years, respectively. Prioritisation of high KDRI donors for recipients >60 years did not change their health outcomes (p=0.739).
Conclusions: Risk-based matching incurs health benefits across the population, benefitting recipients ≤45 years and disadvantaging those >60 years. Preferentially allocating high KDRI grafts to recipients >60 years does not provide them significant health benefits.