Introduction: Cumulative retrospective evidence suggests that preservation by hypothermic machine perfusion (HMP) reduces the incidence of delayed graft function (DGF) in kidney transplants from donors after cardiac death (DCD). However, two contemporary randomised clinical trials designed to evaluate the effect of MP in DCD kidney transplantation have shown contradictory results. Methods: To address whether HMP has a beneficial effect on the outcome of kidneys from controlled DCD, a comparative analysis between 34 DCD kidneys preserved with conventional static cold storage (SCS) and 83 preserved with HMP was performed. In a subsequent analysis, a clinical score, the deceased donor score system, was used to group the DCD kidneys in four groups (Grades A to D) according to their risk of post-transplant dysfunction. Thus, optimal (Grade A, n=14), borderline (Grade B, n=15) and marginal (Grade C, n=5) kidneys preserved by SCS were compared with optimal (Grade A, n=25), borderline (Grade B, n=39) and marginal (Grade C, n=19) kidneys preserved with HMP. In addition, the ability of perfusion parameters (pressure, flow and resistance) to predict post-transplant outcome was investigated in a group of 50 consecutive DCD kidneys using univariate and multivariate analyses. Results: In the HMP group the incidence of DGF was significantly lower 45% (15/83) vs 79% (27/34) p<0.000) and the length of hospitalization shorter (10 vs 14 p< 0.001). Similarly, 1-year (152±7 vs 193±20, p=0.001) and 5-year graft function was statistically better in the HMP than in the SCS (156±31 vs 186±17 mmol/L, p<0.002). When comparing between grade A kidneys, HMP reduced the incidence of DGF from 71% (10/14) to 24% (6/25). In this group HMP did not have an impact on 1y graft function (128±31 vs 125±36, p= 0.450) but there was improved graft function at 5y (111±18 vs 170±41, p= 0.001). In grade B kidneys HMP reduced the incidence of DGF from 100% (15/15) to 43% (17/39) and there was improved 1-year (163±13 vs 207± 73, p= 0.001) and 5-year graft function (183±27 vs 235±33, p=0.01). Interestingly, in grade C kidneys HMP reduced the incidence of DGF from 100% (5/5) to 68% 13/19) but did not improve 1-year (197±31 25 vs 188±17, p= 0.197) or 5y (231±107 vs 211±23, p= 0.725) graft function. No significant correlation between perfusion pressure, renal flow and intra renal resistance and clinical end-points was found in the regression analyses. Conclusions: This analysis shows that hypothermic machine perfusion significantly reduced the incidence of DGF in DCD kidneys from all DDS grades and improved graft function of optimal (grade A) and borderline (grade B) DCD kidneys. Although HMP did not lead to improved 1-year and 5-yeary graft function in marginal (grade C) DCD kidneys, did enable the evaluation and subsequent use of high-risk DCD kidneys that probably would be discarded without evaluation. Our results suggest that HMP can protect optimal DCD kidneys from ischaemiareperfusion injury but additional strategies to improve short and longterm graft funct...