INTRODUCTION

Living donor kidney transplantation accounts for around half of all paediatric renal transplantation and results in improved allograft survival, although there is limited data comparing the effect of HLA matching on outcomes. The UK 2006 Kidney Allocation Scheme (NKAS) prioritises children with good HLA matching. Good HLA matching at first transplantation can reduce the risk of HLA sensitisation.

The aim of this study was to determine the relative importance of donor type and HLA matching on allograft outcomes in paediatric renal transplant recipients, to determine whether there is ever justification to wait for a well HLA matched donation after brain death (DBD) kidney if there is a poorly HLA matched living donor (LD) kidney available.

MATERIALS AND METHODS

Data were obtained from the UK Transplant Registry on all paediatric (<18 years) renal transplant recipients who received a DBD or LD kidney-only transplant between 2000 and 2011. HLA A, B and DR mismatch were categorised according to the UK NKAS and split into two groups (good and poor HLA match).

Kaplan-Meier analysis was used to estimate five-year renal allograft survival by donor type and by HLA match group, differences were tested using the univariate log-rank test. A Cox proportional hazards risk-adjusted model was included to account for explanatory variables including transplant year and recipient age. Data were fully anonymised and ethical principles adhered to.

RESULTS

1,378 paediatric renal transplant recipients were analysed; 804 (58%) received a DBD donor kidney, 574 (42%) received an LD kidney. Figure 1 shows the superior five-year renal allograft survival for children receiving a poorly HLA matched LD kidney transplant (88%, 95% CI 84-91%) compared to children receiving a well HLA matched DBD kidney transplant (83%, 95% CI 80-86%, log rank test p = 0.03).

The Cox proportional hazards risk-adjusted model showed a lower risk of renal allograft failure in children who received a poorly HLA matched LD kidney compared to children who received a well HLA matched DBD kidney, although this did not reach statistical significance (Hazard Ratio 0.77, 95% CI 0.53-1.11).

For both donor types there was no significant difference in five-year renal allograft survival in children who received a good HLA match compared with those that received a poor HLA match (DBD donor p = 0.08, living donor p = 0.47).

CONCLUSIONS

In children, well HLA matched DBD renal transplant outcomes are not superior when compared to poorly HLA matched LD renal transplants. This supports previous evidence showing that with current immunosuppression, HLA matching has a diminished effect on short-term renal allograft outcomes. Although this study does not assess long-term outcomes and the risk of HLA sensitisation, it is difficult to justify preferentially waiting for an improved HLA matched DBD kidney when a poorer HLA matched LD kidney transplant is available.