Predicting and ameliorating graft function in deceased donor kidney transplantation

Student thesis: Doctoral ThesisDoctor of Philosophy


Thesis summary


Deceased donor kidney transplant programmes are increasingly successful, with improving long-term outcomes (1, 2). However, there remains a shortfall in the number of kidney transplants performed relative to the number of patients who would benefit from one. Expansion of the donor pool to include older donors, those with more comorbidities or with a high terminal creatinine has resulted in falling waiting lists (2-4) but greater variability in graft outcomes. The ability to predict transplant outcomes has therefore become increasingly important as the use of ‘higher risk’ donors results in poorer graft outcomes (5). In particular, expansion of the donor pool has resulted in inferior early graft function, with up to half of kidney transplants from donation after circulatory death donors requiring haemodialysis in the first week of transplantation (6). Ten to 20% of kidneys retrieved from deceased donors are discarded, primarily due to uncertainties about organ quality and long-term graft survival (3, 7-9). The ability to predict graft function prior to transplantation would greatly inform the decision-making process of organ allocation and recipient selection, whilst also identifying grafts that would be amenable to interventions aimed at ameliorating transplant function. Predicting graft outcomes may also lead to less organ wastage and increased transplantation rates.

This thesis focuses on three main approaches that attempt to predict graft function: histopathological analysis of the kidney prior to transplantation, analysis of known donor demographic data from registry analyses, and organ performance during ex vivo normothermic machine perfusion. The predictive ability of these tools has yet to be fully determined. Furthermore, these approaches may indirectly improve graft function, by informing decision making, or (in the case of normothermic machine perfusion) directly provide an opportunity to improve organ quality.


• To determine more reliable methods of predicting poor kidney transplant function (early and late)
• To determine means of ameliorating graft dysfunction prior to kidney transplantation


1. Examine the predictive ability of donor kidney histology on transplant outcomes, including delayed graft function
2. Examine the national registry to determine whether the duration of delayed graft function impacts longer term graft function and survival
3. Examine the safety and feasibility of ex vivo normothermic perfusion and its efficacy ameliorating early graft function
4. Examine the use of ex vivo normothermic perfusion to reduce post-reperfusion syndrome
5. Examine the use of white cell filtration during ex vivo normothermic perfusion as a means of ameliorating early graft parameters.


1. Chronic donor histological changes at implantation were not associated with long-term graft and patient survival in this study. The systematic use of preimplantation kidney biopsies increased organ discard rate when an organ utilisation algorhythm was applied.
2. In donation after circulatory death donor kidney transplantation, delayed graft function lasting longer than 14 days was predictive of inferior patient and graft survival. In contrast, delayed graft function of any duration is associated with inferior outcomes in donation after brain death donor kidney transplantation.
3. Ex vivo normothermic perfusion did not reduce delayed graft function or reperfusion syndrome following deceased donor kidney transplantation. However, these studies may have been underpowered. These studies primarily demonstrated the safety and feasibility of short duration EVNP prior to transplantation.
4. White cell filtration during EVNP with may reduce the number of donor-derived inflammatory cells in the perfusate. However this did not appear to improve early graft functional parameters.


Early graft dysfunction can have a long-term impact on deceased donor kidney transplant function and patient survival. Methods of predicting graft dysfunction are imperfect and can mislead clinicians into unnecessary organ discard. Normothermic machine perfusion technology offers the potential for pre-transplant organ conditioning that may provide improved outcomes for patients.
Date of Award1 Jul 2023
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorAnthony Dorling (Supervisor) & Chris J Callaghan (Supervisor)

Cite this