Standard

Risk factors for adverse outcomes during mechanical ventilation of 1152 COVID-19 patients : a multicenter machine learning study with highly granular data from the Dutch Data Warehouse. / on behalf of Dutch ICU Data Sharing Against COVID-19 Collaborators; Fleuren, Lucas M.; Tonutti, Michele et al.

In: Intensive Care Medicine Experimental, Vol. 9, No. 1, 32, 12.2021, p. 32.

Research output: Contribution to journal › Article › Academic › peer-review

Harvard

on behalf of Dutch ICU Data Sharing Against COVID-19 Collaborators, Fleuren, LM, Tonutti, M, de Bruin, DP, Lalisang, RCA, Dam, TA, Gommers, D, Cremer, OL, Bosman, RJ, Vonk, SJJ, Fornasa, M, Machado, T, van der Meer, NJM, Rigter, S, Wils, EJ, Frenzel, T, Dongelmans, DA, de Jong, R, Peters, M, Kamps, MJA, Ramnarain, D, Nowitzky, R, Nooteboom, FGCA, de Ruijter, W, Urlings-Strop, LC, Smit, EGM, Mehagnoul-Schipper, DJ, Dormans, T, de Jager, CPC, Hendriks, SHA, Oostdijk, E, Reidinga, AC, Festen-Spanjer, B, Brunnekreef, G, Cornet, AD, van den Tempel, W, Boelens, AD, Koetsier, P, Lens, J, Achterberg, S, Faber, HJ, Karakus, A, Beukema, M, Entjes, R, de Jong, P, Houwert, T, Hovenkamp, H, Noorduijn Londono, R, Quintarelli, D, Scholtemeijer, MG, de Beer, AA, Cinà, G, Beudel, M, de Keizer, NF, Hoogendoorn, M, Girbes, ARJ, Herter, WE, Elbers, PWG, Thoral, PJ, Rettig, TCD, Reuland, MC, van Manen, L, Montenij, L, van Bommel, J, van den Berg, R, van Geest, E, Hana, A, Boersma, WG, van den Bogaard, B, Pickkers, P, van der Heiden, P, van Gemeren, CCW, Meinders, AJ, de Bruin, M, Rademaker, E, van Osch, FHM, de Kruif, M, Schroten, N, Arnold, KS, Fijen, JW, van Koesveld, JJM, Simons, KS, Labout, J, van de Gaauw, B, Kuiper, M, Beishuizen, A, Geutjes, D, Lutisan, J, Grady, BPX, van den Akker, R, Simons, B, Rijkeboer, AA, Arbous, S, Aries, M, van den Oever, NCG, van Tellingen, M, Dijkstra, A, van Raalte, R, Roggeveen, L, van Diggelen, F, Hassouni, AE, Guzman, DR, Bhulai, S, Ouweneel, D, Driessen, R, Peppink, J, de Grooth, HJ, Zijlstra, GJ, van Tienhoven, AJ, van der Heiden, E, Spijkstra, JJ, van der Spoel, H, de Man, A, Klausch, T, de Vries, H, de Neree tot Babberich, M, Thijssens, O, Wagemakers, L, van der Pol, HGA, Hendriks, T, Berend, J, Silva, VC, Kullberg, B, Heunks, L, Juffermans, N & Slooter, A 2021, 'Risk factors for adverse outcomes during mechanical ventilation of 1152 COVID-19 patients: a multicenter machine learning study with highly granular data from the Dutch Data Warehouse', Intensive Care Medicine Experimental, vol. 9, no. 1, 32, pp. 32. https://doi.org/10.1186/s40635-021-00397-5

APA

on behalf of Dutch ICU Data Sharing Against COVID-19 Collaborators, Fleuren, L. M., Tonutti, M., de Bruin, D. P., Lalisang, R. C. A., Dam, T. A., Gommers, D., Cremer, O. L., Bosman, R. J., Vonk, S. J. J., Fornasa, M., Machado, T., van der Meer, N. J. M., Rigter, S., Wils, E. J., Frenzel, T., Dongelmans, D. A., de Jong, R., Peters, M., ... Slooter, A. (2021). Risk factors for adverse outcomes during mechanical ventilation of 1152 COVID-19 patients: a multicenter machine learning study with highly granular data from the Dutch Data Warehouse. Intensive Care Medicine Experimental, 9(1), 32. [32]. https://doi.org/10.1186/s40635-021-00397-5

Vancouver

on behalf of Dutch ICU Data Sharing Against COVID-19 Collaborators, Fleuren LM, Tonutti M, de Bruin DP, Lalisang RCA, Dam TA et al. Risk factors for adverse outcomes during mechanical ventilation of 1152 COVID-19 patients: a multicenter machine learning study with highly granular data from the Dutch Data Warehouse. Intensive Care Medicine Experimental. 2021 Dec;9(1):32. 32. doi: 10.1186/s40635-021-00397-5

Author

on behalf of Dutch ICU Data Sharing Against COVID-19 Collaborators ; Fleuren, Lucas M. ; Tonutti, Michele et al. / Risk factors for adverse outcomes during mechanical ventilation of 1152 COVID-19 patients : a multicenter machine learning study with highly granular data from the Dutch Data Warehouse. In: Intensive Care Medicine Experimental. 2021 ; Vol. 9, No. 1. pp. 32.

BibTeX

@article{2ee91d5c3bd141b3911eb8b3fba3b42c,
title = "Risk factors for adverse outcomes during mechanical ventilation of 1152 COVID-19 patients: a multicenter machine learning study with highly granular data from the Dutch Data Warehouse",
abstract = "Background: The identification of risk factors for adverse outcomes and prolonged intensive care unit (ICU) stay in COVID-19 patients is essential for prognostication, determining treatment intensity, and resource allocation. Previous studies have determined risk factors on admission only, and included a limited number of predictors. Therefore, using data from the highly granular and multicenter Dutch Data Warehouse, we developed machine learning models to identify risk factors for ICU mortality, ventilator-free days and ICU-free days during the course of invasive mechanical ventilation (IMV) in COVID-19 patients. Methods: The DDW is a growing electronic health record database of critically ill COVID-19 patients in the Netherlands. All adult ICU patients on IMV were eligible for inclusion. Transfers, patients admitted for less than 24 h, and patients still admitted at time of data extraction were excluded. Predictors were selected based on the literature, and included medication dosage and fluid balance. Multiple algorithms were trained and validated on up to three sets of observations per patient on day 1, 7, and 14 using fivefold nested cross-validation, keeping observations from an individual patient in the same split. Results: A total of 1152 patients were included in the model. XGBoost models performed best for all outcomes and were used to calculate predictor importance. Using Shapley additive explanations (SHAP), age was the most important demographic risk factor for the outcomes upon start of IMV and throughout its course. The relative probability of death across age values is visualized in Partial Dependence Plots (PDPs), with an increase starting at 54 years. Besides age, acidaemia, low P/F-ratios and high driving pressures demonstrated a higher probability of death. The PDP for driving pressure showed a relative probability increase starting at 12 cmH2O. Conclusion: Age is the most important demographic risk factor of ICU mortality, ICU-free days and ventilator-free days throughout the course of invasive mechanical ventilation in critically ill COVID-19 patients. pH, P/F ratio, and driving pressure should be monitored closely over the course of mechanical ventilation as risk factors predictive of these outcomes.",
keywords = "COVID-19, Machine learning, Mortality prediction, Risk factors",
author = "{on behalf of Dutch ICU Data Sharing Against COVID-19 Collaborators} and Fleuren, {Lucas M.} and Michele Tonutti and {de Bruin}, {Daan P.} and Lalisang, {Robbert C.A.} and Dam, {Tariq A.} and Diederik Gommers and Cremer, {Olaf L.} and Bosman, {Rob J.} and Vonk, {Sebastiaan J.J.} and Mattia Fornasa and Tomas Machado and {van der Meer}, {Nardo J.M.} and Sander Rigter and Wils, {Evert Jan} and Tim Frenzel and Dongelmans, {Dave A.} and {de Jong}, Remko and Marco Peters and Kamps, {Marlijn J.A.} and Dharmanand Ramnarain and Ralph Nowitzky and Nooteboom, {Fleur G.C.A.} and {de Ruijter}, Wouter and Urlings-Strop, {Louise C.} and Smit, {Ellen G.M.} and Mehagnoul-Schipper, {D. Jannet} and Tom Dormans and {de Jager}, {Cornelis P.C.} and Hendriks, {Stefaan H.A.} and Evelien Oostdijk and Reidinga, {Auke C.} and Barbara Festen-Spanjer and Gert Brunnekreef and Cornet, {Alexander D.} and {van den Tempel}, Walter and Boelens, {Age D.} and Peter Koetsier and Judith Lens and Sefanja Achterberg and Faber, {Harald J.} and A. Karakus and Menno Beukema and Robert Entjes and {de Jong}, Paul and Taco Houwert and Hidde Hovenkamp and {Noorduijn Londono}, Roberto and Davide Quintarelli and Scholtemeijer, {Martijn G.} and {de Beer}, {Aletta A.} and Giovanni Cin{\`a} and Martijn Beudel and {de Keizer}, {Nicolet F.} and Mark Hoogendoorn and Girbes, {Armand R.J.} and Herter, {Willem E.} and Elbers, {Paul W.G.} and Thoral, {Patrick J.} and Rettig, {Thijs C.D.} and Reuland, {M. C.} and {van Manen}, Laura and Leon Montenij and {van Bommel}, Jasper and {van den Berg}, Roy and {van Geest}, Ellen and Anisa Hana and Boersma, {W. G.} and {van den Bogaard}, B. and Peter Pickkers and {van der Heiden}, Pim and {van Gemeren}, {Claudia C.W.} and Meinders, {Arend Jan} and {de Bruin}, Martha and Emma Rademaker and {van Osch}, {Frits H.M.} and {de Kruif}, Martijn and Nicolas Schroten and Arnold, {Klaas Sierk} and Fijen, {J. W.} and {van Koesveld}, {Jacomar J.M.} and Simons, {Koen S.} and Joost Labout and {van de Gaauw}, Bart and Michael Kuiper and Albertus Beishuizen and Dennis Geutjes and Johan Lutisan and Grady, {Bart P.X.} and {van den Akker}, Remko and Bram Simons and Rijkeboer, {A. A.} and Sesmu Arbous and Marcel Aries and {van den Oever}, {Niels C.Gritters} and {van Tellingen}, Martijn and Annemieke Dijkstra and {van Raalte}, Rutger and Luca Roggeveen and {van Diggelen}, Fuda and Hassouni, {Ali el} and Guzman, {David Romero} and Sandjai Bhulai and Dagmar Ouweneel and Ronald Driessen and Jan Peppink and {de Grooth}, {H. J.} and Zijlstra, {G. J.} and {van Tienhoven}, {A. J.} and {van der Heiden}, Evelien and Spijkstra, {Jan Jaap} and {van der Spoel}, Hans and {de Man}, Angelique and Thomas Klausch and {de Vries}, Heder and {de Neree tot Babberich}, Michael and Olivier Thijssens and Lot Wagemakers and {van der Pol}, {Hilde G.A.} and Tom Hendriks and Julie Berend and Silva, {Virginia Ceni} and Bob Kullberg and Leo Heunks and Nicole Juffermans and Arjan Slooter",
note = "Funding Information: Partially funded by grants from ZonMw (project 10430012010003, file 50-55700-98-908), Zorgverzekeraars Nederland and the Corona Research Fund. The sponsors had no role in any part of the study. Publisher Copyright: {\textcopyright} 2021, The Author(s).",
year = "2021",
month = dec,
doi = "10.1186/s40635-021-00397-5",
language = "English",
volume = "9",
pages = "32",
journal = "Intensive Care Medicine Experimental",
issn = "2197-425X",
publisher = "Springer Science + Business Media",
number = "1",

}

RIS

TY - JOUR

T1 - Risk factors for adverse outcomes during mechanical ventilation of 1152 COVID-19 patients

T2 - a multicenter machine learning study with highly granular data from the Dutch Data Warehouse

AU - on behalf of Dutch ICU Data Sharing Against COVID-19 Collaborators

AU - Fleuren, Lucas M.

AU - Tonutti, Michele

AU - de Bruin, Daan P.

AU - Lalisang, Robbert C.A.

AU - Dam, Tariq A.

AU - Gommers, Diederik

AU - Cremer, Olaf L.

AU - Bosman, Rob J.

AU - Vonk, Sebastiaan J.J.

AU - Fornasa, Mattia

AU - Machado, Tomas

AU - van der Meer, Nardo J.M.

AU - Rigter, Sander

AU - Wils, Evert Jan

AU - Frenzel, Tim

AU - Dongelmans, Dave A.

AU - de Jong, Remko

AU - Peters, Marco

AU - Kamps, Marlijn J.A.

AU - Ramnarain, Dharmanand

AU - Nowitzky, Ralph

AU - Nooteboom, Fleur G.C.A.

AU - de Ruijter, Wouter

AU - Urlings-Strop, Louise C.

AU - Smit, Ellen G.M.

AU - Mehagnoul-Schipper, D. Jannet

AU - Dormans, Tom

AU - de Jager, Cornelis P.C.

AU - Hendriks, Stefaan H.A.

AU - Oostdijk, Evelien

AU - Reidinga, Auke C.

AU - Festen-Spanjer, Barbara

AU - Brunnekreef, Gert

AU - Cornet, Alexander D.

AU - van den Tempel, Walter

AU - Boelens, Age D.

AU - Koetsier, Peter

AU - Lens, Judith

AU - Achterberg, Sefanja

AU - Faber, Harald J.

AU - Karakus, A.

AU - Beukema, Menno

AU - Entjes, Robert

AU - de Jong, Paul

AU - Houwert, Taco

AU - Hovenkamp, Hidde

AU - Noorduijn Londono, Roberto

AU - Quintarelli, Davide

AU - Scholtemeijer, Martijn G.

AU - de Beer, Aletta A.

AU - Cinà, Giovanni

AU - Beudel, Martijn

AU - de Keizer, Nicolet F.

AU - Hoogendoorn, Mark

AU - Girbes, Armand R.J.

AU - Herter, Willem E.

AU - Elbers, Paul W.G.

AU - Thoral, Patrick J.

AU - Rettig, Thijs C.D.

AU - Reuland, M. C.

AU - van Manen, Laura

AU - Montenij, Leon

AU - van Bommel, Jasper

AU - van den Berg, Roy

AU - van Geest, Ellen

AU - Hana, Anisa

AU - Boersma, W. G.

AU - van den Bogaard, B.

AU - Pickkers, Peter

AU - van der Heiden, Pim

AU - van Gemeren, Claudia C.W.

AU - Meinders, Arend Jan

AU - de Bruin, Martha

AU - Rademaker, Emma

AU - van Osch, Frits H.M.

AU - de Kruif, Martijn

AU - Schroten, Nicolas

AU - Arnold, Klaas Sierk

AU - Fijen, J. W.

AU - van Koesveld, Jacomar J.M.

AU - Simons, Koen S.

AU - Labout, Joost

AU - van de Gaauw, Bart

AU - Kuiper, Michael

AU - Beishuizen, Albertus

AU - Geutjes, Dennis

AU - Lutisan, Johan

AU - Grady, Bart P.X.

AU - van den Akker, Remko

AU - Simons, Bram

AU - Rijkeboer, A. A.

AU - Arbous, Sesmu

AU - Aries, Marcel

AU - van den Oever, Niels C.Gritters

AU - van Tellingen, Martijn

AU - Dijkstra, Annemieke

AU - van Raalte, Rutger

AU - Roggeveen, Luca

AU - van Diggelen, Fuda

AU - Hassouni, Ali el

AU - Guzman, David Romero

AU - Bhulai, Sandjai

AU - Ouweneel, Dagmar

AU - Driessen, Ronald

AU - Peppink, Jan

AU - de Grooth, H. J.

AU - Zijlstra, G. J.

AU - van Tienhoven, A. J.

AU - van der Heiden, Evelien

AU - Spijkstra, Jan Jaap

AU - van der Spoel, Hans

AU - de Man, Angelique

AU - Klausch, Thomas

AU - de Vries, Heder

AU - de Neree tot Babberich, Michael

AU - Thijssens, Olivier

AU - Wagemakers, Lot

AU - van der Pol, Hilde G.A.

AU - Hendriks, Tom

AU - Berend, Julie

AU - Silva, Virginia Ceni

AU - Kullberg, Bob

AU - Heunks, Leo

AU - Juffermans, Nicole

AU - Slooter, Arjan

N1 - Funding Information: Partially funded by grants from ZonMw (project 10430012010003, file 50-55700-98-908), Zorgverzekeraars Nederland and the Corona Research Fund. The sponsors had no role in any part of the study. Publisher Copyright: © 2021, The Author(s).

PY - 2021/12

Y1 - 2021/12

N2 - Background: The identification of risk factors for adverse outcomes and prolonged intensive care unit (ICU) stay in COVID-19 patients is essential for prognostication, determining treatment intensity, and resource allocation. Previous studies have determined risk factors on admission only, and included a limited number of predictors. Therefore, using data from the highly granular and multicenter Dutch Data Warehouse, we developed machine learning models to identify risk factors for ICU mortality, ventilator-free days and ICU-free days during the course of invasive mechanical ventilation (IMV) in COVID-19 patients. Methods: The DDW is a growing electronic health record database of critically ill COVID-19 patients in the Netherlands. All adult ICU patients on IMV were eligible for inclusion. Transfers, patients admitted for less than 24 h, and patients still admitted at time of data extraction were excluded. Predictors were selected based on the literature, and included medication dosage and fluid balance. Multiple algorithms were trained and validated on up to three sets of observations per patient on day 1, 7, and 14 using fivefold nested cross-validation, keeping observations from an individual patient in the same split. Results: A total of 1152 patients were included in the model. XGBoost models performed best for all outcomes and were used to calculate predictor importance. Using Shapley additive explanations (SHAP), age was the most important demographic risk factor for the outcomes upon start of IMV and throughout its course. The relative probability of death across age values is visualized in Partial Dependence Plots (PDPs), with an increase starting at 54 years. Besides age, acidaemia, low P/F-ratios and high driving pressures demonstrated a higher probability of death. The PDP for driving pressure showed a relative probability increase starting at 12 cmH2O. Conclusion: Age is the most important demographic risk factor of ICU mortality, ICU-free days and ventilator-free days throughout the course of invasive mechanical ventilation in critically ill COVID-19 patients. pH, P/F ratio, and driving pressure should be monitored closely over the course of mechanical ventilation as risk factors predictive of these outcomes.

AB - Background: The identification of risk factors for adverse outcomes and prolonged intensive care unit (ICU) stay in COVID-19 patients is essential for prognostication, determining treatment intensity, and resource allocation. Previous studies have determined risk factors on admission only, and included a limited number of predictors. Therefore, using data from the highly granular and multicenter Dutch Data Warehouse, we developed machine learning models to identify risk factors for ICU mortality, ventilator-free days and ICU-free days during the course of invasive mechanical ventilation (IMV) in COVID-19 patients. Methods: The DDW is a growing electronic health record database of critically ill COVID-19 patients in the Netherlands. All adult ICU patients on IMV were eligible for inclusion. Transfers, patients admitted for less than 24 h, and patients still admitted at time of data extraction were excluded. Predictors were selected based on the literature, and included medication dosage and fluid balance. Multiple algorithms were trained and validated on up to three sets of observations per patient on day 1, 7, and 14 using fivefold nested cross-validation, keeping observations from an individual patient in the same split. Results: A total of 1152 patients were included in the model. XGBoost models performed best for all outcomes and were used to calculate predictor importance. Using Shapley additive explanations (SHAP), age was the most important demographic risk factor for the outcomes upon start of IMV and throughout its course. The relative probability of death across age values is visualized in Partial Dependence Plots (PDPs), with an increase starting at 54 years. Besides age, acidaemia, low P/F-ratios and high driving pressures demonstrated a higher probability of death. The PDP for driving pressure showed a relative probability increase starting at 12 cmH2O. Conclusion: Age is the most important demographic risk factor of ICU mortality, ICU-free days and ventilator-free days throughout the course of invasive mechanical ventilation in critically ill COVID-19 patients. pH, P/F ratio, and driving pressure should be monitored closely over the course of mechanical ventilation as risk factors predictive of these outcomes.

KW - COVID-19

KW - Machine learning

KW - Mortality prediction

KW - Risk factors

UR - http://www.scopus.com/inward/record.url?scp=85118905299&partnerID=8YFLogxK

U2 - 10.1186/s40635-021-00397-5

DO - 10.1186/s40635-021-00397-5

M3 - Article

C2 - 34180025

VL - 9

SP - 32

JO - Intensive Care Medicine Experimental

JF - Intensive Care Medicine Experimental

SN - 2197-425X

IS - 1

M1 - 32

ER -

ID: 20333471