Machine Perfusion: Not Just for Marginal Kidney Donors

Abstract

Interest in machine perfusion (MP) for donated kidneys has markedly increased in the past decade as a means to improve graft function, although the donor populations in which it should be applied have not yet been resolved. All adults undergoing de-novo isolated kidney transplantation from standard-criteria donors in the UNOS database 2005 to 2011 were reviewed with the primary endpoint of delayed graft function (DGF), defined as dialysis within seven days of transplantation, in those who received kidneys that underwent MP versus cold storage (CS) alone. Three methods were used to control for differences between groups. Multivariable logistic regression was performed, adjusting for donor and recipient characteristics significantly associated with DGF. Rates were also compared in a cohort of propensity-matched MP vs CS recipients. Finally, a paired-kidney study was performed, where one kidney underwent MP and the contralateral underwent CS. There were 36,323 patients, with unadjusted DGF rates of 18.6 per cent (n = 1830/9882) and 22.4 per cent (n = 5931/26,441; P < 0.001) in the MP vs CS groups, respectively. After multivariable analysis, the odds ratio for DGF in the MP group was 0.59 (P < 0.001) versus CS. In the propensity-matched cohort, there were 8929 patients each in the MP and CS groups. DGF occurred in 16.8 per cent of the MP group vs 25.3 per cent with CS (P < 0.001, OR 0.59). In the paired-kidney study, rates of DGF were 16.7 per cent vs 24.3 per cent (P < 0.001) in the 1665 recipients each in the MP versus CS groups (OR 0.6). In conclusion, machine perfusion is beneficial in reducing DGF even when standard donors are utilized, and thus should not be limited to marginal kidneys.