Publication: Predictive model for the optimal glomerular filtration rate in living kidney transplant recipients
Issued Date
2014-01-01
Resource Type
ISSN
18732623
00411345
00411345
Other identifier(s)
2-s2.0-84896474882
Rights
Mahidol University
Rights Holder(s)
SCOPUS
Bibliographic Citation
Transplantation Proceedings. Vol.46, No.2 (2014), 469-473
Suggested Citation
T. Srithongkul, N. Premasathian, A. Vongwiwatana, W. Uwatanasombat, K. Vareesangthip Predictive model for the optimal glomerular filtration rate in living kidney transplant recipients. Transplantation Proceedings. Vol.46, No.2 (2014), 469-473. doi:10.1016/j.transproceed.2013.11.096 Retrieved from: https://repository.li.mahidol.ac.th/handle/20.500.14594/34445
Research Projects
Organizational Units
Authors
Journal Issue
Thesis
Title
Predictive model for the optimal glomerular filtration rate in living kidney transplant recipients
Other Contributor(s)
Abstract
Background Recipient glomerular filtration rate (GFR) after living kidney transplantation (KT) is influenced by many factors. Defining the appropriate level of recipient GFR post-KT is helpful. The aim of this study was to establish a predictive model to estimate the optimal recipient GFR at 1 week post-KT. Methods We retrospectively analyzed 211 living KTs without delayed or slow graft function. Estimated GFR was calculated using the Cockcroft-Gault (CG) formula. Donor kidney volume was obtained from routine computed tomographic angiography (CTA) by work station GE (AW 4.20) program. Multivariate analysis was carried out with automated backward selection to establish the predictive model. The bias, precision, and accuracy of our model were also determined by application of the model to another 37 living KTs. Results In multivariate analysis, the significant parameters to predict recipient GFR were donor age (P =.025) and kidney volume (P <.0001) and both were incorporated in the predictive model; predicted CG recipient GFR = 28.325 + (donor kidney volume x 0.282) - (0.297 x donor age). The correlation coefficient (R) is 0.5. Application to another group revealed that our model had high precision (14.45 mL/min), small positive bias (0.24 mL/min), and high percentage (81%) of predicted value, which was within 30% of the observed recipient GFR post-KT. Conclusion Our predictive model included donor age and donor kidney volume and could be used to estimate the optimal recipient GFR post-KT. This could be helpful to identify early graft dysfunction and to make a decision if further invasive investigation such as allograft biopsy is necessary. © 2014 by Elsevier Inc. All rights reserved.