Machine perfusion and long-term kidney transplant recipient outcomes across allograft risk strata.

Background: The use of machine perfusion (MP) in kidney transplantation lowers delayed graft function (DGF) and improves 1-year graft survival in some, but not all, grafts. These associations have not been explored in grafts stratified by the Kidney Donor Profile index (KDPI). Methods: We analyzed 78 207 deceased-donor recipients using the Scientific Registry of Transplant Recipients data from 2006 to 2013. The cohort was stratified using the standard criteria donor/expanded criteria donor (ECD)/donation after cardiac death (DCD)/donation after brain death (DBD) classification and the KDPI scores. In each subgroup, MP use was compared with cold storage. Results: The overall DGF rate was 25.4% and MP use was associated with significantly lower DGF in all but the ECD-DCD donor subgroup. Using the donor source classification, the use of MP did not decrease death-censored graft failure (DCGF), except in the ECD-DCD subgroup from 0 to 1 year {adjusted hazard ratio [aHR] 0.56 [95% confidence interval (CI) 0.32-0.98]}. In the ECD-DBD subgroup, higher DCGF from 1 to 5 years was noted [aHR 1.15 (95% CI 1.01-1.31)]. Also, MP did not lower all-cause graft failure except in the ECD-DCD subgroup from 0 to 1 year [aHR = 0.59 (95% CI 0.38-0.91)]. Using the KDPI classification, MP did not lower DCGF or all-cause graft failure, but in the ≤70 subgroup, higher DCGF [aHR 1.16 (95% CI 1.05-1.27)] and higher all-cause graft failure [aHR 1.10 (95% CI 1.02-1.18)] was noted. Lastly, MP was not associated with mortality in any subgroup. Conclusions: Overall, MP did not lower DCGF. Neither classification better risk-stratified kidneys that have superior graft survival with MP. We question their widespread use in all allografts as an ideal approach to organ preservation.