Nephrology Dialysis Transplantation

Real-time prediction of intradialytic hypotension using machine learning and cloud computing infrastructure.

Hanjie Zhang, Lin-Chun Wang, Sheetal Chaudhuri, Aaron Pickering, Len Usvyat, John W Larkin, Peter Waguespack, Zuwen Kuang, Jeroen P Kooman, Franklin W Maddux, Peter Kotanko


Background: In maintenance hemodialysis patients, intradialytic hypotension (IDH) is a frequent complication that has been associated with poor clinical outcomes. Prediction of IDH may facilitate timely interventions and eventually reduce IDH rates.

Methods: We developed a machine learning model to predict IDH in in-center hemodialysis patients 15-75 min in advance. IDH was defined as systolic blood pressure (SBP) <90 mmHg. Demographic, clinical, treatment-related and laboratory data were retrieved from electronic health records and merged with intradialytic machine data that were sent in real-time to the cloud. For model development, dialysis sessions were randomly split into training (80%) and testing (20%) sets. The area under the receiver operating characteristic curve (AUROC) was used as a measure of the model's predictive performance.

Results: We utilized data from 693 patients who contributed 42 656 hemodialysis sessions and 355 693 intradialytic SBP measurements. IDH occurred in 16.2% of hemodialysis treatments. Our model predicted IDH 15-75 min in advance with an AUROC of 0.89. Top IDH predictors were the most recent intradialytic SBP and IDH rate, as well as mean nadir SBP of the previous 10 dialysis sessions.

Conclusions: Real-time prediction of IDH during an ongoing hemodialysis session is feasible and has a clinically actionable predictive performance. If and to what degree this predictive information facilitates the timely deployment of preventive interventions and translates into lower IDH rates and improved patient outcomes warrants prospective studies.

Keywords: end-stage kidney disease; intradialytic hypotension; machine learning; real-time prediction.

© The Author(s) 2023. Published by Oxford University Press on behalf of the ERA.

About the Author

Hanjie Zhang, MSc, PhD

Supervisor of Biostatistics and Applied Artificial Intelligence /Machine Learning

Hanjie joined RRI in 2014. She received a master’s degree in statistics from Columbia University, New York, and a PhD in medical science from the University of Maastricht, The Netherlands. Hanjie has been involved in the design of several large cluster-randomized clinical trials and complex statistical analyses in collaboration with the Medical Office, FMCNA...

Dr. Peter Kotanko, MD

RRI Research Director

SVP, Corporate Research & Development

Peter Kotanko, MD is Research Director at the Renal Research Institute (RRI), New York. Prior to joining RRI, from 1997 to 2007 he served as Vice Chair of a department of internal medicine at an academic teaching hospital in Graz, Austria. Prior to moving to Graz in 1989, he worked 1982-89 at the Department of Physiology and the University Clinic of Internal Medicine, Innsbruck, Austria...