Prediction model for probability of deceased donor kidney transplantation among transplant candidates in waiting list
Issued Date
2025-12-01
Resource Type
eISSN
14712369
Scopus ID
2-s2.0-105009723658
Journal Title
BMC Nephrology
Volume
26
Issue
1
Rights Holder(s)
SCOPUS
Bibliographic Citation
BMC Nephrology Vol.26 No.1 (2025)
Suggested Citation
Sutharattanapong N., Ingsathit A., Thotsiri S., Wiwattanathum P., Thammanichanond D., Thakoorabutr P., Thakkinstian A., Kantachuvesiri S. Prediction model for probability of deceased donor kidney transplantation among transplant candidates in waiting list. BMC Nephrology Vol.26 No.1 (2025). doi:10.1186/s12882-025-04221-0 Retrieved from: https://repository.li.mahidol.ac.th/handle/20.500.14594/111174
Title
Prediction model for probability of deceased donor kidney transplantation among transplant candidates in waiting list
Author's Affiliation
Corresponding Author(s)
Other Contributor(s)
Abstract
Background: The rapid increase in end-stage kidney disease (ESKD) patients compared to available organ donors remains a significant challenge. This study aimed to externally validate and update the Scientific Registry of Transplant Recipients (SRTR) prediction model for receiving deceased donor kidney transplantation (KT) among Thai patients on the waiting list. Methods: This retrospective study included 3,923 patients registered for deceased donor KT. The original SRTR score was calculated. The primary outcome was the time to deceased donor KT. Cox regression analysis was used to evaluate associations between the SRTR score, additional predictors, and time to KT. The SRTR model was revised by re-estimating individual coefficients and updated by incorporating significant predictors. Discriminative performance was assessed using Harrell’s C-index. Results: The original SRTR model performed poorly in our cohort, with a C-index of 0.521. After revision, the C-index slightly improved to 0.662. The updated model, incorporating nine predictors (sex, blood group, diabetes, peripheral vascular disease, ESKD etiology, dialysis vintage, history of prior kidney transplantation, panel reactive antibody, and anti-HBs antibody status), achieved a C-index of 0.744, indicating good discriminative performance. Conclusion: The updated prediction model demonstrated good predictive performance and may be applied to optimize the management of transplant candidates on the waiting list in the Thai population. Clinical trial number: Not applicable.