Critical Care (Aug 2021)
Machine learning identifies ICU outcome predictors in a multicenter COVID-19 cohort
- Harry Magunia,
- Simone Lederer,
- Raphael Verbuecheln,
- Bryant Joseph Gilot,
- Michael Koeppen,
- Helene A. Haeberle,
- Valbona Mirakaj,
- Pascal Hofmann,
- Gernot Marx,
- Johannes Bickenbach,
- Boris Nohe,
- Michael Lay,
- Claudia Spies,
- Andreas Edel,
- Fridtjof Schiefenhövel,
- Tim Rahmel,
- Christian Putensen,
- Timur Sellmann,
- Thea Koch,
- Timo Brandenburger,
- Detlef Kindgen-Milles,
- Thorsten Brenner,
- Marc Berger,
- Kai Zacharowski,
- Elisabeth Adam,
- Matthias Posch,
- Onnen Moerer,
- Christian S. Scheer,
- Daniel Sedding,
- Markus A. Weigand,
- Falk Fichtner,
- Carla Nau,
- Florian Prätsch,
- Thomas Wiesmann,
- Christian Koch,
- Gerhard Schneider,
- Tobias Lahmer,
- Andreas Straub,
- Andreas Meiser,
- Manfred Weiss,
- Bettina Jungwirth,
- Frank Wappler,
- Patrick Meybohm,
- Johannes Herrmann,
- Nisar Malek,
- Oliver Kohlbacher,
- Stephanie Biergans,
- Peter Rosenberger
Affiliations
- Harry Magunia
- Department of Anesthesiology and Intensive Care Medicine, University Hospital Tübingen, Eberhard-Karls-University Tübingen
- Simone Lederer
- Institute for Translational Bioinformatics and Medical Data Integration Center, University Hospital Tübingen, Eberhard-Karls-University Tübingen
- Raphael Verbuecheln
- Institute for Translational Bioinformatics and Medical Data Integration Center, University Hospital Tübingen, Eberhard-Karls-University Tübingen
- Bryant Joseph Gilot
- Institute for Translational Bioinformatics and Medical Data Integration Center, University Hospital Tübingen, Eberhard-Karls-University Tübingen
- Michael Koeppen
- Department of Anesthesiology and Intensive Care Medicine, University Hospital Tübingen, Eberhard-Karls-University Tübingen
- Helene A. Haeberle
- Department of Anesthesiology and Intensive Care Medicine, University Hospital Tübingen, Eberhard-Karls-University Tübingen
- Valbona Mirakaj
- Department of Anesthesiology and Intensive Care Medicine, University Hospital Tübingen, Eberhard-Karls-University Tübingen
- Pascal Hofmann
- Department of Anesthesiology and Intensive Care Medicine, University Hospital Tübingen, Eberhard-Karls-University Tübingen
- Gernot Marx
- Department of Intensive Care Medicine, University Hospital RWTH Aachen
- Johannes Bickenbach
- Department of Intensive Care Medicine, University Hospital RWTH Aachen
- Boris Nohe
- Center for Anaesthesia, Intensive Care and Emergency Medicine, Zollernalb Klinikum
- Michael Lay
- Center for Anaesthesia, Intensive Care and Emergency Medicine, Zollernalb Klinikum
- Claudia Spies
- Department of Anesthesiology and Operative Intensive Care Medicine (CCM, CVK), Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Berlin Institute of Health, Humboldt-Universität zu Berlin
- Andreas Edel
- Department of Anesthesiology and Operative Intensive Care Medicine (CCM, CVK), Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Berlin Institute of Health, Humboldt-Universität zu Berlin
- Fridtjof Schiefenhövel
- Department of Anesthesiology and Operative Intensive Care Medicine (CCM, CVK), Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Berlin Institute of Health, Humboldt-Universität zu Berlin
- Tim Rahmel
- Department of Anesthesiology, Intensive Care Medicine/Pain Therapy
- Christian Putensen
- Department of Anaesthesiology and Intensive Care Medicine, University Hospital Bonn
- Timur Sellmann
- Department of Anesthesiology and Intensive Care Medicine, Evangelisches Krankenhaus Bethesda
- Thea Koch
- Department of Anesthesiology and Intensive Care Medicine, University Hospital Carl Gustav Carus, Technische Universität Dresden
- Timo Brandenburger
- Department of Anaesthesiology, University Hospital Düsseldorf
- Detlef Kindgen-Milles
- Department of Anaesthesiology, University Hospital Düsseldorf
- Thorsten Brenner
- Department of Anesthesiology and Intensive Care Medicine, University Hospital Essen, University Duisburg-Essen
- Marc Berger
- Department of Anesthesiology and Intensive Care Medicine, University Hospital Essen, University Duisburg-Essen
- Kai Zacharowski
- Department of Anaesthesiology, Intensive Care Medicine and Pain Therapy, University Hospital Frankfurt, Goethe University
- Elisabeth Adam
- Department of Anaesthesiology, Intensive Care Medicine and Pain Therapy, University Hospital Frankfurt, Goethe University
- Matthias Posch
- Department of Anesthesiology and Critical Care, Medical Center - University of Freiburg
- Onnen Moerer
- Center for Anesthesiology, Emergency and Intensive Care Medicine, University of Göttingen
- Christian S. Scheer
- Department of Anesthesiology, University Medicine Greifswald
- Daniel Sedding
- Department Cardiology, Angiology and Intensive Care Medicine, University Hospital Halle (Saale)
- Markus A. Weigand
- Department of Anesthesiology, Heidelberg University Hospital
- Falk Fichtner
- Department of Anesthesiology and Intensive Care, Leipzig University Hospital
- Carla Nau
- Department of Anesthesiology and Intensive Care, University Medical Center Schleswig-Holstein, Campus Lübeck, University of Lübeck
- Florian Prätsch
- Department of Anaesthesiology and Intensive Care Therapy, Otto-Von-Guericke-University Magdeburg
- Thomas Wiesmann
- University Hospital Marburg, UKGM, Philipps University Marburg
- Christian Koch
- Department of Anesthesiology, Intensive Care Medicine and Pain Therapy, University Hospital Giessen and Marburg, Justus-Liebig University Giessen
- Gerhard Schneider
- Department of Anesthesiology and Intensive Care, School of Medicine, Klinikum Rechts Der Isar, Technical University of Munich
- Tobias Lahmer
- Klinik Und Poliklinik Für Innere Medizin II, Klinikum Rechts Der Isar der, Technischen Universität München
- Andreas Straub
- Department for Anesthesiology, Intensive Care Medicine, Emergency Medicine and Pain Medicine, St. Elisabethen Klinikum
- Andreas Meiser
- Department of Anesthesiology, Intensive Care Medicine and Pain Medicine, Saarland University Hospital Medical Center
- Manfred Weiss
- Department of Anesthesiology and Intensive Care Medicine, Ulm University
- Bettina Jungwirth
- Department of Anesthesiology and Intensive Care Medicine, Ulm University
- Frank Wappler
- Department of Anaesthesiology and Intensive Care Medicine, Cologne-Merheim Medical Centre, Witten/Herdecke University
- Patrick Meybohm
- Department of Anaesthesiology, Intensive Care, Emergency and Pain Medicine, University Hospital Wuerzburg, University Wuerzburg
- Johannes Herrmann
- Department of Anaesthesiology, Intensive Care, Emergency and Pain Medicine, University Hospital Wuerzburg, University Wuerzburg
- Nisar Malek
- Department of Internal Medicine 1, University Hospital Tübingen
- Oliver Kohlbacher
- Institute for Translational Bioinformatics and Medical Data Integration Center, University Hospital Tübingen, Eberhard-Karls-University Tübingen
- Stephanie Biergans
- Institute for Translational Bioinformatics and Medical Data Integration Center, University Hospital Tübingen, Eberhard-Karls-University Tübingen
- Peter Rosenberger
- Department of Anesthesiology and Intensive Care Medicine, University Hospital Tübingen, Eberhard-Karls-University Tübingen
- DOI
- https://doi.org/10.1186/s13054-021-03720-4
- Journal volume & issue
-
Vol. 25,
no. 1
pp. 1 – 14
Abstract
Abstract Background Intensive Care Resources are heavily utilized during the COVID-19 pandemic. However, risk stratification and prediction of SARS-CoV-2 patient clinical outcomes upon ICU admission remain inadequate. This study aimed to develop a machine learning model, based on retrospective & prospective clinical data, to stratify patient risk and predict ICU survival and outcomes. Methods A Germany-wide electronic registry was established to pseudonymously collect admission, therapeutic and discharge information of SARS-CoV-2 ICU patients retrospectively and prospectively. Machine learning approaches were evaluated for the accuracy and interpretability of predictions. The Explainable Boosting Machine approach was selected as the most suitable method. Individual, non-linear shape functions for predictive parameters and parameter interactions are reported. Results 1039 patients were included in the Explainable Boosting Machine model, 596 patients retrospectively collected, and 443 patients prospectively collected. The model for prediction of general ICU outcome was shown to be more reliable to predict “survival”. Age, inflammatory and thrombotic activity, and severity of ARDS at ICU admission were shown to be predictive of ICU survival. Patients’ age, pulmonary dysfunction and transfer from an external institution were predictors for ECMO therapy. The interaction of patient age with D-dimer levels on admission and creatinine levels with SOFA score without GCS were predictors for renal replacement therapy. Conclusions Using Explainable Boosting Machine analysis, we confirmed and weighed previously reported and identified novel predictors for outcome in critically ill COVID-19 patients. Using this strategy, predictive modeling of COVID-19 ICU patient outcomes can be performed overcoming the limitations of linear regression models. Trial registration “ClinicalTrials” (clinicaltrials.gov) under NCT04455451.
Keywords
