Construction of a Waiting Time Predictive Model for Kidney Transplant with Deceased Donor in the State of São Paulo

J. Silva¹, M. Contti¹, M. Valiatti¹, H. Nga¹, G. Santos¹, M. Perosa², G. Ferreira³, L. Modelli de Andrade¹

¹UNESP, Botucatu, Brazil, ²Hospital Leforte, São Paulo, Brazil, ³Santa Casa de Juiz de Fora, Juiz de Fora, Brazil

Meeting: 2021 American Transplant Congress

Abstract number: 34

Keywords: Allocation, Kidney transplantation, Prediction models, Waiting lists

Topic: Clinical Science » Kidney » Kidney Deceased Donor Allocation

Session Information

Session Name: Kidney Deceased Donor Allocation

Session Type: Rapid Fire Oral Abstract

Date: Saturday, June 5, 2021

Session Time: 4:30pm-5:30pm

Presentation Time: 5:05pm-5:10pm

Location: Virtual

*Purpose: Chronic kidney disease is an important public health problem and kidney transplant is the therapy of choice when possible. The transplant system in the State of São Paulo, Brazil is a valuable sample. Few investigations study the waiting time for kidney transplant with a deceased donor, therefore, developing a predictive model can contribute to better allocating the patients. Objectives: determine the predictors for waiting time on the list for kidney transplant, verify the applicability of the allocation criteria, and create a predictive model of waiting time.

*Methods: Retrospective cohort study. All patients listed for transplantation between Jan/2000 to Dec/2017 in the database of the São Paulo State Department of Health were included. Variables studied: age, sex, race, baseline disease, regional reference centres, dialysis duration, ABO blood group, panel class I, HLA-A, HLA-B, HLA-DR, blood transfusions, pregnancies, and previous transplants. The data were randomly separated: 75% for training and 25% for model validation testing. Cox regression was performed having the transplant as the outcome. Sensitivity analyses were carried out in regional reference centres and regression analyses were carried out with competitive outcome.

*Results: We analysed 54,055 records. In the period, 28.5% of the patients were transplanted (n = 13,694), WITH a higher probability in the first 50 months. The main factors that reduced the chance of transplantation were: Panel> 80%, belonging to the regional School of Medicine of the University of São Paulo (FMUSP) and blood type O. Factors associated with higher chance of transplantation: age <18 years, presence of anti-HBc and blood type AB. A predictive model was obtained capable of predicting the waiting time on the transplant list with excellent agreement in internal validation (c-index = 0.70).

*Conclusions: Allocation system that is effective in prioritizing recipients under 18 and patients with greater compatibility in the HLA system. Patients with reduced chances of transplantation were those who were sensitized, those from blood group O, and those with HLA homozygosity. Regional differences were found which favoured centres with a lower number of patients placed on the transplant list. A predictive model that can help in the predictability of the transplant was created.

To cite this abstract in AMA style:

Silva J, Contti M, Valiatti M, Nga H, Santos G, Perosa M, Ferreira G, Andrade LModellide. Construction of a Waiting Time Predictive Model for Kidney Transplant with Deceased Donor in the State of São Paulo [abstract]. Am J Transplant. 2021; 21 (suppl 3). https://atcmeetingabstracts.com/abstract/construction-of-a-waiting-time-predictive-model-for-kidney-transplant-with-deceased-donor-in-the-state-of-sao-paulo/. Accessed May 23, 2025.

« Back to 2021 American Transplant Congress