Abstract: FR-PO1039

KDPI and Allograft eGFR in Deceased Donor Kidney Transplant Recipients over Follow Up: 1995-2013

Session Information

Category: Transplantation

  • 1702 Transplantation: Clinical and Translational

Authors

  • Sexton, Donal J., The Irish Longitudinal Study on Ageing (TILDA), Trinity College Dublin., Dublin, Ireland
  • O'kelly, Patrick, Beaumont Hospital, Dublin 9, Co Dublin, Ireland
  • Kennedy, Claire, Beaumont Hospital, Dublin 9, Co Dublin, Ireland
  • de Freitas, Declan G., Beaumont Hospital, Dublin 9, Co Dublin, Ireland
  • O'Seaghdha, Conall M., Beaumont Hospital, Dublin 9, Co Dublin, Ireland
  • Conlon, Peter J., Beaumont Hospital, Dublin 9, Co Dublin, Ireland
Background

The KDPI has been validated for prediction of deceased donor graft outcomes and popularized in organ allocation in the United States. Whether KDPI predicts eGFR over long-term follow up has not been extensively characterised.

Methods

We performed a retrospective analysis of eGFR (CKD EPI equation) and graft outcomes over follow up in the National Kidney Transplant Service of Ireland database for the years 2006–2013. Associations of the composite KDPI score with eGFR at various time points over follow up were modeled using linear regression, linear mixed effects models and time to event strategies respectively.

Results

N=877 patients had complete data regarding KDPI calculation, N=148 allografts failed. Median (IQR) KDPI score was 52 (49). At year 1 eGFR ml/min/1.73m2 were: 69.2 (57.6-82.5) (N=172) quartile 1 of KDPI, 59.4 (50.2-74.5) (N=171) in quartile 2 KDPI, 51.9 (42.2-62.2) (N=176) in quartile 3 KDPI, 48.3 (38.5-58.5) (N=162) in quartile 4 KDPI, P<0.001. At year 5 eGFR in these quartiles were 74.5 (60.5-88.2) (N=110), 64.6 (51.7-83.2) (N=104), 48.8 (34.4-61.6) (N=85), and 43.6 (37.5-58.1) (N=69), P<0.001. On repeated measures analysis with linear mixed effects models with a random participant specific intercept and a random time effect, KDPI (fixed effect covariate) associated with eGFR over follow up (see Fig 1): estimate (se) -0.25 (0.02) P<0.001 for KDPI. The variability in eGFR over 5 years alligned with KDPI score was 21% after adjusting for recipient age (time-varying covariate).

Conclusion

KDPI score predicted eGFR at multiple time points over long term follow up in this study. However, taking the KDPI as a composite of non-modifiable donor factors, a considerable portion of eGFR variability is likely attributable non-donor factors.

A plot of mean eGFR by quartile of KDPI score in deceased donor transplants in Ireland over 5 years follow up.