ASN's Mission

To create a world without kidney diseases, the ASN Alliance for Kidney Health elevates care by educating and informing, driving breakthroughs and innovation, and advocating for policies that create transformative changes in kidney medicine throughout the world.

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: SA-PO824

Distinct Phenotypes of Kidney Retransplantation by Machine Learning Consensus Clustering in the United States

Session Information

Category: Transplantation

  • 2002 Transplantation: Clinical

Authors

  • Mao, Michael A., Mayo Clinic, Jacksonville, Florida, United States
  • Mao, Shennen, Mayo Clinic, Jacksonville, Florida, United States
  • Thongprayoon, Charat, Mayo Clinic Minnesota, Rochester, Minnesota, United States
  • Vaitla, Pradeep, University of Mississippi Medical Center, Jackson, Mississippi, United States
  • Jadlowiec, Caroline, Mayo Clinic Arizona, Scottsdale, Arizona, United States
  • Acharya, Prakrati C., Texas Tech Health Sciences Center, El Paso, Texas, United States
  • Leeaphorn, Napat, St. Luke's Health System, Kansas City, Missouri, United States
  • Kaewput, Wisit, Phramongkutklao College of Medicine, Bangkok, Thailand
  • Pattharanitima, Pattharawin, Thammasat University, Pathum Thani, Thailand
  • Cheungpasitporn, Wisit, Mayo Clinic Minnesota, Rochester, Minnesota, United States
Background

The application of machine learning may provide a novel understanding of unique phenotypes of kidney retransplant recipients that will allow identification of new strategies to improve outcomes. Our study aimed to characterize kidney retransplant recipients using an unsupervised machine learning approach.

Methods

We performed consensus cluster analysis using recipient-, donor-, and transplant-related characteristics in 17,443 kidney retransplant recipients from the 2010-2019 OPTN/UNOS database . We identified each cluster’s key characteristics using the standardized mean difference of >0.3. Posttransplant outcomes including acute allograft rejection, death-censored graft failure, and mortality were compared among the assigned clusters.

Results

Consensus cluster analysis identified three distinct clusters of kidney retransplant recipients. The key characteristics of cluster 1 were white patients who received pre-emptive kidney retransplant or had dialysis duration less than 1 year before receiving kidney retransplant from living, female, and older donors. Cluster 1 patients had lower PRA, cold ischemia time, use of kidney machine perfusion, and occurrence of delayed graft function, but had more private insurance. In contrast, cluster 2 patients had higher PRA and received kidney retransplant from non-ECD deceased donors with a lower number of HLA mismatches. Cluster 3 patients received kidney retransplant from non-ECD deceased donors with a higher number of HLA mismatches. Cluster 1 had the most favorable patient and graft survival while cluster 3 had the worst patient (HR 2.17) and graft survival (HR 2.64).

Conclusion

Unsupervised machine learning approach characterized kidney retransplant recipients based on their pattern of clinical characteristics into three clinically distinct clusters with differing posttransplant outcomes.

Standardized differences across the three clusters for each baseline parameter.