Prediction of gastrointestinal bleeding hospitalization risk in hemodialysis using machine learning

Abstract Background Gastrointestinal bleeding (GIB) is a clinical challenge in kidney failure. INSPIRE group assessed if machine learning could determine a hemodialysis (HD) patient’s 180-day GIB hospitalization risk. Methods An eXtreme Gradient Boosting (XGBoost) and logistic regression model were...

Full description

Saved in:
Bibliographic Details
Main Authors: John W. Larkin, Suman Lama, Sheetal Chaudhuri, Joanna Willetts, Anke C. Winter, Yue Jiao, Manuela Stauss-Grabo, Len A. Usvyat, Jeffrey L. Hymes, Franklin W. Maddux, David C. Wheeler, Peter Stenvinkel, Jürgen Floege, on behalf of the INSPIRE Core Group
Format: Article
Language:English
Published: BMC 2024-10-01
Series:BMC Nephrology
Subjects:
Online Access:https://doi.org/10.1186/s12882-024-03809-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850182484487045120
author John W. Larkin
Suman Lama
Sheetal Chaudhuri
Joanna Willetts
Anke C. Winter
Yue Jiao
Manuela Stauss-Grabo
Len A. Usvyat
Jeffrey L. Hymes
Franklin W. Maddux
David C. Wheeler
Peter Stenvinkel
Jürgen Floege
on behalf of the INSPIRE Core Group
author_facet John W. Larkin
Suman Lama
Sheetal Chaudhuri
Joanna Willetts
Anke C. Winter
Yue Jiao
Manuela Stauss-Grabo
Len A. Usvyat
Jeffrey L. Hymes
Franklin W. Maddux
David C. Wheeler
Peter Stenvinkel
Jürgen Floege
on behalf of the INSPIRE Core Group
author_sort John W. Larkin
collection DOAJ
description Abstract Background Gastrointestinal bleeding (GIB) is a clinical challenge in kidney failure. INSPIRE group assessed if machine learning could determine a hemodialysis (HD) patient’s 180-day GIB hospitalization risk. Methods An eXtreme Gradient Boosting (XGBoost) and logistic regression model were developed using an HD dataset in United States (2017–2020). Patient data was randomly split (50% training, 30% validation, and 20% testing). HD treatments ≤ 180 days before GIB hospitalization were classified as positive observations; others were negative. Models considered 1,303 exposures/covariates. Performance was measured using unseen testing data. Results Incidence of 180-day GIB hospitalization was 1.18% in HD population (n = 451,579), and 1.12% in testing dataset (n = 38,853). XGBoost showed area under the receiver operating curve (AUROC) = 0.74 (95% confidence interval (CI) 0.72, 0.76) versus logistic regression showed AUROC = 0.68 (95% CI 0.66, 0.71). Sensitivity and specificity were 65.3% (60.9, 69.7) and 68.0% (67.6, 68.5) for XGBoost versus 68.9% (64.7, 73.0) and 57.0% (56.5, 57.5) for logistic regression, respectively. Associations in exposures were consistent for many factors. Both models showed GIB hospitalization risk was associated with older age, disturbances in anemia/iron indices, recent all-cause hospitalizations, and bone mineral metabolism markers. XGBoost showed high importance on outcome prediction for serum 25 hydroxy (25OH) vitamin D levels, while logistic regression showed high importance for parathyroid hormone (PTH) levels. Conclusions Machine learning can be considered for early detection of GIB event risk in HD. XGBoost outperforms logistic regression, yet both appear suitable. External and prospective validation of these models is needed. Association between bone mineral metabolism markers and GIB events was unexpected and warrants investigation. Trial registration This retrospective analysis of real-world data was not a prospective clinical trial and registration is not applicable. Graphical Abstract
format Article
id doaj-art-f4272cea88a24d7fa5d86c3b4f57a0ee
institution OA Journals
issn 1471-2369
language English
publishDate 2024-10-01
publisher BMC
record_format Article
series BMC Nephrology
spelling doaj-art-f4272cea88a24d7fa5d86c3b4f57a0ee2025-08-20T02:17:37ZengBMCBMC Nephrology1471-23692024-10-0125111610.1186/s12882-024-03809-2Prediction of gastrointestinal bleeding hospitalization risk in hemodialysis using machine learningJohn W. Larkin0Suman Lama1Sheetal Chaudhuri2Joanna Willetts3Anke C. Winter4Yue Jiao5Manuela Stauss-Grabo6Len A. Usvyat7Jeffrey L. Hymes8Franklin W. Maddux9David C. Wheeler10Peter Stenvinkel11Jürgen Floege12on behalf of the INSPIRE Core GroupFresenius Medical Care, Global Medical OfficeFresenius Medical Care, Global Medical OfficeFresenius Medical Care, Global Medical OfficeFresenius Medical Care, Global Medical OfficeFresenius Medical Care, Global Medical OfficeFresenius Medical Care, Global Medical OfficeFresenius Medical Care, Global Medical OfficeFresenius Medical Care, Global Medical OfficeFresenius Medical Care, Global Medical OfficeFresenius Medical Care AG, Global Medical OfficeUniversity College LondonDept of Renal Medicine, Karolinska University HospitalDivisions of Nephrology and Cardiology, University Hospital RWTH AachenAbstract Background Gastrointestinal bleeding (GIB) is a clinical challenge in kidney failure. INSPIRE group assessed if machine learning could determine a hemodialysis (HD) patient’s 180-day GIB hospitalization risk. Methods An eXtreme Gradient Boosting (XGBoost) and logistic regression model were developed using an HD dataset in United States (2017–2020). Patient data was randomly split (50% training, 30% validation, and 20% testing). HD treatments ≤ 180 days before GIB hospitalization were classified as positive observations; others were negative. Models considered 1,303 exposures/covariates. Performance was measured using unseen testing data. Results Incidence of 180-day GIB hospitalization was 1.18% in HD population (n = 451,579), and 1.12% in testing dataset (n = 38,853). XGBoost showed area under the receiver operating curve (AUROC) = 0.74 (95% confidence interval (CI) 0.72, 0.76) versus logistic regression showed AUROC = 0.68 (95% CI 0.66, 0.71). Sensitivity and specificity were 65.3% (60.9, 69.7) and 68.0% (67.6, 68.5) for XGBoost versus 68.9% (64.7, 73.0) and 57.0% (56.5, 57.5) for logistic regression, respectively. Associations in exposures were consistent for many factors. Both models showed GIB hospitalization risk was associated with older age, disturbances in anemia/iron indices, recent all-cause hospitalizations, and bone mineral metabolism markers. XGBoost showed high importance on outcome prediction for serum 25 hydroxy (25OH) vitamin D levels, while logistic regression showed high importance for parathyroid hormone (PTH) levels. Conclusions Machine learning can be considered for early detection of GIB event risk in HD. XGBoost outperforms logistic regression, yet both appear suitable. External and prospective validation of these models is needed. Association between bone mineral metabolism markers and GIB events was unexpected and warrants investigation. Trial registration This retrospective analysis of real-world data was not a prospective clinical trial and registration is not applicable. Graphical Abstracthttps://doi.org/10.1186/s12882-024-03809-2BleedingGastrointestinalHospitalizationKidney FailurePredictive Modeling
spellingShingle John W. Larkin
Suman Lama
Sheetal Chaudhuri
Joanna Willetts
Anke C. Winter
Yue Jiao
Manuela Stauss-Grabo
Len A. Usvyat
Jeffrey L. Hymes
Franklin W. Maddux
David C. Wheeler
Peter Stenvinkel
Jürgen Floege
on behalf of the INSPIRE Core Group
Prediction of gastrointestinal bleeding hospitalization risk in hemodialysis using machine learning
BMC Nephrology
Bleeding
Gastrointestinal
Hospitalization
Kidney Failure
Predictive Modeling
title Prediction of gastrointestinal bleeding hospitalization risk in hemodialysis using machine learning
title_full Prediction of gastrointestinal bleeding hospitalization risk in hemodialysis using machine learning
title_fullStr Prediction of gastrointestinal bleeding hospitalization risk in hemodialysis using machine learning
title_full_unstemmed Prediction of gastrointestinal bleeding hospitalization risk in hemodialysis using machine learning
title_short Prediction of gastrointestinal bleeding hospitalization risk in hemodialysis using machine learning
title_sort prediction of gastrointestinal bleeding hospitalization risk in hemodialysis using machine learning
topic Bleeding
Gastrointestinal
Hospitalization
Kidney Failure
Predictive Modeling
url https://doi.org/10.1186/s12882-024-03809-2
work_keys_str_mv AT johnwlarkin predictionofgastrointestinalbleedinghospitalizationriskinhemodialysisusingmachinelearning
AT sumanlama predictionofgastrointestinalbleedinghospitalizationriskinhemodialysisusingmachinelearning
AT sheetalchaudhuri predictionofgastrointestinalbleedinghospitalizationriskinhemodialysisusingmachinelearning
AT joannawilletts predictionofgastrointestinalbleedinghospitalizationriskinhemodialysisusingmachinelearning
AT ankecwinter predictionofgastrointestinalbleedinghospitalizationriskinhemodialysisusingmachinelearning
AT yuejiao predictionofgastrointestinalbleedinghospitalizationriskinhemodialysisusingmachinelearning
AT manuelastaussgrabo predictionofgastrointestinalbleedinghospitalizationriskinhemodialysisusingmachinelearning
AT lenausvyat predictionofgastrointestinalbleedinghospitalizationriskinhemodialysisusingmachinelearning
AT jeffreylhymes predictionofgastrointestinalbleedinghospitalizationriskinhemodialysisusingmachinelearning
AT franklinwmaddux predictionofgastrointestinalbleedinghospitalizationriskinhemodialysisusingmachinelearning
AT davidcwheeler predictionofgastrointestinalbleedinghospitalizationriskinhemodialysisusingmachinelearning
AT peterstenvinkel predictionofgastrointestinalbleedinghospitalizationriskinhemodialysisusingmachinelearning
AT jurgenfloege predictionofgastrointestinalbleedinghospitalizationriskinhemodialysisusingmachinelearning
AT onbehalfoftheinspirecoregroup predictionofgastrointestinalbleedinghospitalizationriskinhemodialysisusingmachinelearning