Predicting 1-year successful clinical use of an arteriovenous access for hemodialysis using machine learning
Arteriovenous (AV) access is important to support long-term hemodialysis; however, a significant proportion fail due to inadequate maturation or other complications. Tools that can predict long-term AV access success may guide clinical decision-making but remain limited. We developed machine learning (ML) algorithms that predict 1-year successful clinical use of an AV access using pre-operative data. The Vascular Quality Initiative (VQI) was used to identify patients who underwent surgical AV fistula/graft creation between 2011-2024. We identified 111 pre-operative demographic, clinical, and anatomic features. Six ML models were trained with 10-fold cross-validation. Overall, 59,674 patients underwent AV access creation and 28,304 (47.4%) had 1-year successful clinical use of their index AV access. The best prediction model was XGBoost, achieving an AUROC of 0.90. In comparison, logistic regression had an AUROC of 0.70. The XGBoost model accurately predicted 1-year successful clinical use of an AV access for hemodialysis, performing better than logistic regression.
Authors
- Mohamad A. Hussain (ORCID: https://orcid.org/0000-0003-2471-8167)
- Ori D. Rotstein (ORCID: https://orcid.org/0009-0004-4388-5352)
- Graham Roche‐Nagle (ORCID: https://orcid.org/0000-0002-3560-6788)
- Charles de Mestral (ORCID: https://orcid.org/0000-0001-7177-7480)
- Naomi Eisenberg (ORCID: https://orcid.org/0000-0002-1237-9244)
- Duminda N. Wijeysundera (ORCID: https://orcid.org/0000-0002-5897-8605)
- Derek Beaton (ORCID: https://orcid.org/0000-0001-6118-4366)
- Elisa Greco (ORCID: https://orcid.org/0000-0002-7580-3444)
- Douglas S. Lee (ORCID: https://orcid.org/0000-0001-7078-745X)
- Muhammad Mamdani (ORCID: https://orcid.org/0000-0001-5199-6344)
- Leen Al‐Omran
- Ben Li (ORCID: https://orcid.org/0000-0002-7191-1034)
- Mohammed Al-Omran
Institutions
- Brigham and Women's Hospital (US)
- St. Michael's Hospital (CA)
- University Health Network (CA)
- Alfaisal University (SA)
- University of Toronto (CA)
- King Faisal Specialist Hospital & Research Centre (SA)
- St Michael’s Hospital (IE)
- Artificial Intelligence in Medicine (Canada) (CA)
- Unity Health Toronto
- Ontario Medical Association (CA)
Publication Details
- Journal
- npj Digital Medicine
- Published
- 2025-12-04
- DOI
- https://doi.org/10.1038/s41746-025-02187-9
- Primary Topic
- Central Venous Catheters and Hemodialysis
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Brigham and Women's Hospital
- Physicians' Services Incorporated Foundation
- University of Toronto
- Canadian Institutes of Health Research