EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Assessing SOFA score trajectories in sepsis using machine learning: A pragmatic approach to improve the accuracy of mortality prediction

Lars Palmowski iD, Hartmuth Nowak, Andrea Witowski, Björn Koos, Alexander Wolf, Maike Weber, Daniel Kleefisch, Matthias Unterberg, Helge Haberl, Alexander von Busch, Christian Ertmer, Alexander Zarbock und 21 weitere

PLOS ONE · 28.03.2024 · 19 (3) · e0300739

DOI: 10.1371/journal.pone.0300739 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS ONE, online erschienen am 28.03.2024. Verfasst von Lars Palmowski und 32 weiteren Autor:innen. Beteiligte Einrichtung: Klinik für Anästhesiologie, Intensivmedizin und Schmerztherapie, Universitätsklinikum Knappschaftskrankenhaus Bochum, Ruhr Universität Bochum, Bochum…. Hauptthemen laut MEDLINE: Organ Dysfunction Scores, Sepsis. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Faculty of Health at Witten/Herdecke University, Germany, EFRE.NRW.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

An increasing amount of longitudinal health data is available on critically ill septic patients in the age of digital medicine, including daily sequential organ failure assessment (SOFA) score measurements. Thus, the assessment in sepsis focuses increasingly on the evaluation of the individual disease’s trajectory. Machine learning (ML) algorithms may provide a promising approach here to improve the evaluation of daily SOFA score dynamics. We tested whether ML algorithms can outperform the conventional ΔSOFA score regarding the accuracy of 30-day mortality prediction.

Methods

We used the multicentric SepsisDataNet.NRW study cohort that prospectively enrolled 252 sepsis patients between 03/2018 and 09/2019 for training ML algorithms, i.e. support vector machine (SVM) with polynomial kernel and artificial neural network (aNN). We used the Amsterdam UMC database covering 1,790 sepsis patients for external and independent validation.

Results

Both SVM (AUC 0.84; 95% CI: 0.71–0.96) and aNN (AUC 0.82; 95% CI: 0.69–0.95) assessing the SOFA scores of the first seven days led to a more accurate prognosis of 30-day mortality compared to the ΔSOFA score between day 1 and 7 (AUC 0.73; 95% CI: 0.65–0.80; p = 0.02 and p = 0.05, respectively). These differences were even more prominent the shorter the time interval considered. Using the SOFA scores of day 1 to 3 SVM (AUC 0.82; 95% CI: 0.68 0.95) and aNN (AUC 0.80; 95% CI: 0.660.93) led to a more accurate prognosis of 30-day mortality compared to the ΔSOFA score (AUC 0.66; 95% CI: 0.58–0.74; p < 0.01 and p < 0.01, respectively). Strikingly, all these findings could be confirmed in the independent external validation cohort.

Conclusions

The ML-based algorithms using daily SOFA scores markedly improved the accuracy of mortality compared to the conventional ΔSOFA score. Therefore, this approach could provide a promising and automated approach to assess the individual disease trajectory in sepsis. These findings reflect the potential of incorporating ML algorithms as robust and generalizable support tools on intensive care units.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Organ Dysfunction ScoresSepsisHumansRetrospective StudiesIntensive Care UnitsMachine LearningPrognosisROC Curve

Autor:innen und Institutionen

  • Klinik für Anästhesiologie, Intensivmedizin und Schmerztherapie, Universitätsklinikum Knappschaftskrankenhaus Bochum, Ruhr Universität Bochum, Bochum, Germany.
  • Zentrum für Künstliche Intelligenz, Medizininformatik und Datenwissenschaften, Universitätsklinikum Knappschaftskrankenhaus Bochum, Bochum, Germany.
  • Medizinische Fakultät, Medizinisches Proteom-Center, Ruhr Universität Bochum, Bochum, Germany.
  • Zentrum für Proteindiagnostik (PRODI), Ruhr Universität Bochum, Bochum, Germany.
  • Klinik für Anästhesiologie, Operative Intensivmedizin und Schmerztherapie, Universitätsklinikum Münster, Münster, Germany.
  • Klinik für Anästhesiologie und Operative Intensivmedizin, Universitätsklinikum Bonn, Bonn, Germany.

ORCID-Profile: Lars Palmowski, Ulrich Limper, Thomas Köhler, Tim Rahmel

Förderung

  • Faculty of Health at Witten/Herdecke University, Germany
  • EFRE.NRW

Zitieren

APA

Palmowski, L., Nowak, H., Witowski, A., Koos, B., Wolf, A., Weber, M., Kleefisch, D., Unterberg, M., Haberl, H., von Busch, A., Ertmer, C., Zarbock, A., Bode, C., Putensen, C., Limper, U., Wappler, F., Köhler, T., Henzler, D., Oswald, D., …, & SepsisDataNet.NRW research group (2024). Assessing SOFA score trajectories in sepsis using machine learning: A pragmatic approach to improve the accuracy of mortality prediction. PLOS ONE, 19(3), e0300739. https://doi.org/10.1371/journal.pone.0300739

Vancouver

Palmowski L, Nowak H, Witowski A, Koos B, Wolf A, Weber M, et al. Assessing SOFA score trajectories in sepsis using machine learning: A pragmatic approach to improve the accuracy of mortality prediction. PLOS ONE. 2024;19(3):e0300739. doi:10.1371/journal.pone.0300739

BibTeX
@article{Palmowski2024,
  author = {Palmowski, Lars and Nowak, Hartmuth and Witowski, Andrea and Koos, Björn and Wolf, Alexander and Weber, Maike and Kleefisch, Daniel and Unterberg, Matthias and Haberl, Helge and von Busch, Alexander and Ertmer, Christian and Zarbock, Alexander and Bode, Christian and Putensen, Christian and Limper, Ulrich and Wappler, Frank and Köhler, Thomas and Henzler, Dietrich and Oswald, Daniel and Ellger, Björn and Ehrentraut, Stefan F and Bergmann, Lars and Rump, Katharina and Ziehe, Dominik and Babel, Nina and Sitek, Barbara and Marcus, Katrin and Frey, Ulrich H and Thoral, Patrick J and Adamzik, Michael and Eisenacher, Martin and Rahmel, Tim and SepsisDataNet.NRW research group, SepsisDataNet.NRW research group},
  title = {Assessing SOFA score trajectories in sepsis using machine learning: A pragmatic approach to improve the accuracy of mortality prediction},
  journal = {PLOS ONE},
  year = {2024},
  volume = {19},
  number = {3},
  pages = {e0300739},
  doi = {10.1371/journal.pone.0300739}
}
Kostenlos starten →