Introduction
Visual assessment of the quality of perfusion remains a fundamental, yet poorly evaluated, aspect of donor kidney appraisal. We aimed to evaluate the consistency of grading of perfusion and its association with organ utilisation and graft outcomes.  
Methods
A retrospective analysis of all deceased-donor kidney transplants from 2000-2016 was performed using the UK Transplant Registry database. Perfusion quality was graded at retrieval and implanting as: 1 = good, 2 = fair, 3 = poor, 4 = patchy. Multivariable analyses were performed to determine whether grade of perfusion is an independent predictor of discard rates, delayed graft function/primary non-function (DGF/PNF) rates and graft survival. 
Results
Analysis included 31,167 organs from 15,750 donors, of which 2,556 (8.2%) were discarded, most commonly due to ‘poor perfusion’ (15%). In transplanted organs, consistency of grading at retrieval vs. implanting centre was poor (Kappa=0.179), with only 17.2% of grade 4 organs retaining the same grade. Grade of perfusion at retrieval was independently associated with discard rates (p<0.001), which increased from 6.5% at grade 1, to 41.8% at grade 3, before falling to 27.1% at grade 4.
Perfusion grade at retrieval was associated with DGF/PNF rates (6.6% vs. 2.8% for grade 3 vs. 1), but not long-term graft survival (p=0.454). However, perfusion grade at implanting was a significant independent predictor of both outcomes. DGF/PNF rates increased progressively with the perfusion grade at implanting (p<0.001), but graft survival only declined significantly between grades 1 and 3 (p=0.002), with grade 1 and 4 organs having similar outcomes (HR: 1.03, p=0.764). 
Conclusions
Perfusion grading at retrieval has poor consistency with implanting and is associated with organ utilisation, despite not being associated with long-term outcomes. Outcomes become worse based on implanting grade 1-3 but not grade 4, suggesting the need for re-evaluation of the grading system.

BTSNHSBT-19