ASN's Mission

ASN leads the fight to prevent, treat, and cure kidney diseases throughout the world by educating health professionals and scientists, advancing research and innovation, communicating new knowledge, and advocating for the highest quality care for patients.

learn more

Contact ASN

1401 H St, NW, Ste 900, Washington, DC 20005

email@asn-online.org

202-640-4660

The Latest on Twitter

Kidney Week

Abstract: TH-PO1122

Construction of a Predictive Model of Delayed Graft Function Using Machine Learning Techniques

Session Information

Category: Transplantation

  • 1902 Transplantation: Clinical

Authors

  • Costa, Silvana Daher, Federal university of ceara, Fortaleza, Brazil
  • Oliveira, Claudia M., Centro de Pesquisas Doenças Hepatorenais, Fortaleza, Brazil
  • Fernandes, Paula frassinetti Cbc, Universidade Federal do Ceara and Universidade Estadual do Ceara, Fortaleza, Brazil
  • Daher, Elizabeth De Francesco, Federal University of Ceara, Fortaleza, Brazil
  • De andrade, Luis gustavo Modelli, UNESP, Univ Estadual Paulista, Botucatu, Brazil
  • Esmeraldo, Ronaldo M., Hospital Geral de Fortaleza, Fortaleza, Brazil
  • Sandes-Freitas, Tainá Veras, Universidade Federal do Ceará, Fortaleza, Brazil
Background

Brazilian studies have been reported incidences of delayed graft function (DGF), 2-3 fold the incidences described by American and European cohorts. The available predictive models of DGF, based on distinct clinical realities, are not validated in our population.
The purpose of the study was evaluate the accuracy of the available predictive models and, since none of these models present good accuracy, construct a local prediction model

Methods

Retrospective cohort study including DD KT performed between Jan 2014 and Dec 2017 in two transplant centers (n=443). The predictive DGF models tested were those described by Irish et al., Jeldres et al., Chapal et al., and Zaza et al. For the construction of the new predictive model, machine learning was used

Results

Patients were predominantly men (56.7%), young adults (44.2 ± 14.7 years), mixed race (84.4%), who remained 46.8 ± 45.2 months on dialysis. Donors had a mean age of 31 ± 12.7 years, most of them died from trauma (70.9%), 5.4% were hypertensive (HA), 0.7% were diabetic, and the final creatinine was 1.1 ± 0.6 mg/dL. Only 4.3% were expanded criteria donors. 83.1% of the grafts were perfused with HTK and the mean cold ischemia time (CIT) was 20.9 ± 4 hours. The incidence of DGF in this sample was 53%. The predictive models of DGF available presented regular or poor discriminant power: Irish (AUC 0.686), Chapal (AUC 0.638), Jeldres (AUC 0.613), Zaza (AUC 0.591). The three models with the best performance were decision tree, neural networks and support vector machine. In the final model, the variables considered were: from recipients : age, diabetes and time on dialysis; from donors: age, body mass index, HA, serum sodium , creatinine phosphokinase, final creatinine, cause of death, high dose of vasoactive drugs and diuresis; CIT. The final model showed excellent discriminant power (AUC 0.942)

Conclusion

The incidence of DGF in the sample was high, despite the predominance of standard criteria donors. In addition to variables classically associated with DGF, variables related to donor maintenance were pointed out in non-linear statistical methodologies. The available predictive models had poor accuracy in predicting DGF in our population. The developed model presented excellent performance.

Funding

  • Government Support - Non-U.S.