Abstract: FR-PO025
Exploring the Diversity of Clinical Profiles and Outcomes in Kidney Transplant Recipients with Limited Education Using Clustering Analysis
Session Information
- AI, Digital Health, Data Science - II
November 03, 2023 | Location: Exhibit Hall, Pennsylvania Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Augmented Intelligence, Digital Health, and Data Science
- 300 Augmented Intelligence, Digital Health, and Data Science
Authors
- Tangpanithandee, Supawit, Mayo Clinic Minnesota, Rochester, United States
- Thongprayoon, Charat, Mayo Clinic Minnesota, Rochester, Minnesota, United States
- Miao, Jing, Mayo Clinic Minnesota, Rochester, Minnesota, United States
- Jadlowiec, Caroline, Mayo Clinic, Phoenix, Arizona, United States
- Mao, Shennen, Mayo Clinic, Jacksonville, Florida, United States
- Mao, Michael A., Mayo Clinic, Jacksonville, Florida, United States
- Leeaphorn, Napat, Mayo Clinic, Jacksonville, Florida, United States
- Krisanapan, Pajaree, Mayo Clinic Minnesota, Rochester, Minnesota, United States
- Cheungpasitporn, Wisit, Mayo Clinic Minnesota, Rochester, Minnesota, United States
Background
Education level has been identified as a potential predictor of post-transplant outcomes for kidney transplant recipients. However, there is a lack of research focusing on recipients with lower education levels and their unique clinical characteristics. Therefore, the objective of our study was to utilize unsupervised machine learning techniques to cluster kidney transplant recipients with lower education levels.
Methods
We performed consensus clustering analysis on 20,474 kidney transplant recipients with education levels below college/university using recipient, donor, and transplant data from the OPTN/UNOS database (2017-2019). We identified significant characteristics for each cluster and compared posttransplant outcomes.
Results
Most recipients had completed high school (86%) and were non-white (64%). We identified four clusters: Cluster 1 comprised young, non-diabetic patients receiving kidneys from young, non-hypertensive, non-ECD deceased donors with lower KDPI. Cluster 2 included preemptive or early dialysis initiators, predominantly white, receiving kidneys from living donors. They showed better outcomes. Cluster 3 consisted of young kidney re-transplant recipients with higher PRA and fewer HLA mismatches. Cluster 4 involved older, diabetic patients receiving kidneys from lower-quality donors. Cluster 2 exhibited the best outcomes, while clusters 1, 3, and 4 had higher risks of graft failure and patient mortality.
Conclusion
Unsupervised machine learning successfully clustered kidney transplant recipients with lower education levels into four distinct groups, each with unique clinical profiles and varying posttransplant outcomes. Cluster 2 demonstrated the best outcomes, while clusters 1, 3, and 4 had higher risks of graft failure and patient mortality. These findings have implications for personalized care and risk stratification in kidney transplant recipients with lower education levels.