Predicting 90-day survival of patients with COVID-19: Survival of Severely Ill COVID (SOSIC) scores
Matthieu Schmidt, Bertrand Guidet, Alexandre Demoule, Maharajah Ponnaiah, Muriel Fartoukh, Louis Puybasset, Alain Combes, David Hajage, for the COVID-ICU Investigators, Alain Mercat, Pierre Asfar, François Beloncle, Julien Demiselle, Tài Pham, Arthur Pavot, Xavier Monnet, Christian Richard, Alexandre Demoule, Martin Dres, Julien Mayaux, Alexandra Beurton, Cédric Daubin, Richard Descamps, Aurélie Joret, Damien du Cheyron, Frédéric Pène, Jean‐Daniel Chiche, Mathieu Jozwiak, Paul Jaubert, Guillaume Voiriot, Muriel Fartoukh, Marion Teulier, Clarisse Blayau, Erwen L’Her, Cécile Aubron, Laetitia Bodénes, Nicolas Ferrière, Johann Auchabie, Anthony Le Meur, Sylvain Pignal, Thierry Mazzoni, Jean‐Pierre Quenot, Pascal Andreu, Jean-Baptiste Roudau, Marie Labruyère, Saad Nseir, Sébastien Préau, Julien Poissy, Daniel Mathieu, Sarah Benhamida, Rémi Paulet, Nicolas Roucaud, Martial Thyrault, Florence Daviet, Sami Hraiech, Gabriel Parzy, Aude Sylvestre, Sébastien Jochmans, Anne-Laure Bouilland, Mehran Monchi, Marc Danguy des Déserts, Quentin Mathais, Gwendoline Rager, Pierre Pasquier, Jean Reignier, Amélie Seguin, Charlotte Garret, Emmanuel Canet, Jean Dellamonica, Clément Saccheri, Romain Lombardi, Yanis Kouchit, Sophie Jacquier, Armelle Mathonnet, Mai-Ahn Nay, Isabelle Runge, Frédéric Martino, Laure Flurin, Amélie Rollé, Michel Carles, Rémi Coudroy, Arnaud W. Thille, Jean‐Pierre Frat, Maeva Rodriguez, Pascal Beuret, Audrey Tientcheu, Arthur Vincent, Florian Michelin, Fabienne Tamion, Dorothée Carpentier, Déborah Boyer, Christophe Girault, Valérie Gissot, Stéphan Ehrmann, Charlotte Salmon Gandonnière, Djlali Elaroussi, Agathe Delbove, Yannick Fédun, Julien Huntzinger, Eddy Lebas
Abstract
BACKGROUND: Predicting outcomes of critically ill intensive care unit (ICU) patients with coronavirus-19 disease (COVID-19) is a major challenge to avoid futile, and prolonged ICU stays. METHODS: The objective was to develop predictive survival models for patients with COVID-19 after 1-to-2 weeks in ICU. Based on the COVID-ICU cohort, which prospectively collected characteristics, management, and outcomes of critically ill patients with COVID-19. Machine learning was used to develop dynamic, clinically useful models able to predict 90-day mortality using ICU data collected on day (D) 1, D7 or D14. RESULTS: Survival of Severely Ill COVID (SOSIC)-1, SOSIC-7, and SOSIC-14 scores were constructed with 4244, 2877, and 1349 patients, respectively, randomly assigned to development or test datasets. The three models selected 15 ICU-entry variables recorded on D1, D7, or D14. Cardiovascular, renal, and pulmonary functions on prediction D7 or D14 were among the most heavily weighted inputs for both models. For the test dataset, SOSIC-7's area under the ROC curve was slightly higher (0.80 [0.74-0.86]) than those for SOSIC-1 (0.76 [0.71-0.81]) and SOSIC-14 (0.76 [0.68-0.83]). Similarly, SOSIC-1 and SOSIC-7 had excellent calibration curves, with similar Brier scores for the three models. CONCLUSION: The SOSIC scores showed that entering 15 to 27 baseline and dynamic clinical parameters into an automatable XGBoost algorithm can potentially accurately predict the likely 90-day mortality post-ICU admission (sosic.shinyapps.io/shiny). Although external SOSIC-score validation is still needed, it is an additional tool to strengthen decisions about life-sustaining treatments and informing family members of likely prognosis.