Abstract: PO2347
Predicting Allograft Survival in Young Pediatric Kidney Transplant Recipients
Session Information
- Pediatric Nephrology: Glomerular Disease and Transplantation
October 22, 2020 | Location: On-Demand
Abstract Time: 10:00 AM - 12:00 PM
Category: Pediatric Nephrology
- 1700 Pediatric Nephrology
Authors
- Manca Barayre, Florian, Emory University School of Medicine, Atlanta, Georgia, United States
- Harambat, Jerome, Bordeaux University Hospital, Bordeaux, France
- Le Page, Amelia, Monash University Faculty of Medicine Nursing and Health Sciences, Clayton, Victoria, Australia
- Marks, Stephen D., Great Ormond Street Hospital For Children NHS Foundation Trust, London, London, United Kingdom
- Prestidge, Chanel, Starship Children's Health, Newmarket, Auckland, New Zealand
- Sypek, Matthew P., Australia and New Zealand Dialysis and Transplant Registry, Adelaide, South Australia, Australia
- McDonald, Stephen P., Australia and New Zealand Dialysis and Transplant Registry, Adelaide, South Australia, Australia
- Patzer, Rachel E., Emory University School of Medicine, Atlanta, Georgia, United States
- Hogan, Julien, Emory University School of Medicine, Atlanta, Georgia, United States
Background
Kidney transplantation (kTx) presents specific challenges in younger recipients. There are no predictive model of renal allograft loss in young pediatric recipients to inform donor selection. We aimed to develop and validate a predictive model of graft loss in an international cohort of young kTx recipients.
Methods
We included first-time kTX recipients under 5 years of age in the USA, Australia, New Zealand, UK and France between 2005 and 2018. A multivariate Cox regression was used to develop a predictive model of graft loss or death on the US cohort. Model discrimination (C-statistics) and calibration were assessed internally and externally on the non-US cohort.
Results
2543 kTx in children <5 years old were included. 10-year overall graft survival rate was 80.0% [95% CI = 77.7% – 82.2%]. Given the interaction between some predictors and recipient’s age, we developed two models stratified on recipient age (cut-off: 36 months) including donor/recipient body surface area ratio, ischemia time, donor weight and immunological matching. Immunological matching was a stronger predictor among older recipients, while morphological variables were stronger predictors in younger recipients. C-statistics on the training cohort were 0.63 (95% CI = 0.57 –0.68) and 0.65 (95% CI = 0.59 – 0.71) and the models were well calibrated. Figure 1 presents the discrimination of the models at different time horizons and the calibration at 10 years on the validation cohort.
Conclusion
We confirm the overall good renal allograft survival in children transplanted under the age of 5. We developed and validated predictive models of graft loss or death based on pre-transplant factors in this population that may be used to inform donor selection.
AUC and 95% CI at different horizons in the validation cohort and 10-year calibration plots in validation cohorts (A, C: ≤ 36 months old / B, D: > 36 months old)