Data-driven prediction of continuous renal replacement therapy survival

• Published in: Nature communications 2024-06, Vol.15 (1), p.5440-12, Article 5440
• Main Authors: Zamanzadeh, Davina, Feng, Jeffrey, Petousis, Panayiotis, Vepa, Arvind, Sarrafzadeh, Majid, Karumanchi, S. Ananth, Bui, Alex A. T., Kurtz, Ira
• Format: Article
• Language: English
• Subjects: 631/114/1305; 631/114/1314; 631/114/2413; 692/4022/1950/2777; Aged; Algorithms; Continuous Renal Replacement Therapy - methods; Data collection; Dialysis; Electronic Health Records; Electronic medical records; Error analysis; Female; Hemodialysis; Humanities and Social Sciences; Humans; Kidneys; Learning algorithms; Machine Learning; Male; Middle Aged; multidisciplinary; Patients; Predictions; Renal replacement therapy; Renal Replacement Therapy - methods; Renal Replacement Therapy - mortality; ROC Curve; Science; Science (multidisciplinary); Subgroups; Survival; Therapy; Uncertainty
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-024-49763-3

Continuous renal replacement therapy (CRRT) is a form of dialysis prescribed to severely ill patients who cannot tolerate regular hemodialysis. However, as the patients are typically very ill to begin with, there is always uncertainty whether they will survive during or after CRRT treatment. Because of outcome uncertainty, a large percentage of patients treated with CRRT do not survive, utilizing scarce resources and raising false hope in patients and their families. To address these issues, we present a machine learning-based algorithm to predict short-term survival in patients being initiated on CRRT. We use information extracted from electronic health records from patients who were placed on CRRT at multiple institutions to train a model that predicts CRRT survival outcome; on a held-out test set, the model achieves an area under the receiver operating curve of 0.848 (CI = 0.822–0.870). Feature importance, error, and subgroup analyses provide insight into bias and relevant features for model prediction. Overall, we demonstrate the potential for predictive machine learning models to assist clinicians in alleviating the uncertainty of CRRT patient survival outcomes, with opportunities for future improvement through further data collection and advanced modeling. Despite decades of use in clinical care, only half of individuals who receive continuous renal replacement therapy (CRRT) benefit, and no consensus exists around who should be placed on CRRT. Here, the authors use electronic health record data from multiple institutions to improve prediction of CRRT response before initiating treatment