Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Observational Study
. 2023 Dec 1;51(12):1638-1649.
doi: 10.1097/CCM.0000000000006030. Epub 2023 Aug 31.

Predicting ICU Mortality in Acute Respiratory Distress Syndrome Patients Using Machine Learning: The Predicting Outcome and STratifiCation of severity in ARDS (POSTCARDS) Study

Jesús Villar  1   2   3 , Jesús M González-Martín  1   2 , Jerónimo Hernández-González  4 , Miguel A Armengol  5 , Cristina Fernández  2 , Carmen Martín-Rodríguez  6 , Fernando Mosteiro  7 , Domingo Martínez  8 , Jesús Sánchez-Ballesteros  9 , Carlos Ferrando  10 , Ana M Domínguez-Berrot  11 , José M Añón  12 , Laura Parra  13 , Raquel Montiel  14 , Rosario Solano  15 , Denis Robaglia  16 , Pedro Rodríguez-Suárez  1   17 , Estrella Gómez-Bentolila  2 , Rosa L Fernández  1   2 , Tamas Szakmany  18   19 , Ewout W Steyerberg  20 , Arthur S Slutsky  3   21 ; Predicting Outcome and STratifiCation of severity in ARDS (POSTCARDS) Network
Collaborators, Affiliations
Observational Study

Predicting ICU Mortality in Acute Respiratory Distress Syndrome Patients Using Machine Learning: The Predicting Outcome and STratifiCation of severity in ARDS (POSTCARDS) Study

Jesús Villar et al. Crit Care Med. .

Abstract

Objectives: To assess the value of machine learning approaches in the development of a multivariable model for early prediction of ICU death in patients with acute respiratory distress syndrome (ARDS).

Design: A development, testing, and external validation study using clinical data from four prospective, multicenter, observational cohorts.

Setting: A network of multidisciplinary ICUs.

Patients: A total of 1,303 patients with moderate-to-severe ARDS managed with lung-protective ventilation.

Interventions: None.

Measurements and main results: We developed and tested prediction models in 1,000 ARDS patients. We performed logistic regression analysis following variable selection by a genetic algorithm, random forest and extreme gradient boosting machine learning techniques. Potential predictors included demographics, comorbidities, ventilatory and oxygenation descriptors, and extrapulmonary organ failures. Risk modeling identified some major prognostic factors for ICU mortality, including age, cancer, immunosuppression, Pa o2 /F io2 , inspiratory plateau pressure, and number of extrapulmonary organ failures. Together, these characteristics contained most of the prognostic information in the first 24 hours to predict ICU mortality. Performance with machine learning methods was similar to logistic regression (area under the receiver operating characteristic curve [AUC], 0.87; 95% CI, 0.82-0.91). External validation in an independent cohort of 303 ARDS patients confirmed that the performance of the model was similar to a logistic regression model (AUC, 0.91; 95% CI, 0.87-0.94).

Conclusions: Both machine learning and traditional methods lead to promising models to predict ICU death in moderate/severe ARDS patients. More research is needed to identify markers for severity beyond clinical determinants, such as demographics, comorbidities, lung mechanics, oxygenation, and extrapulmonary organ failure to guide patient management.

PubMed Disclaimer

Conflict of interest statement

Drs. Martín-Rodríguez and Rodríguez-Suárez received support for article research from the National Institutes of Health. Dr. Rodríguez-Suárez disclosed work for government. Dr. Szakmany received funding from PAION UK and Thermo Fisher UK; he disclosed that they are a trustee of Intensive Care National Audit & Research Centre and an Associate Editor for Social Media for Critical Care Explorations. Dr. Slutsky received funding from Signal-1. Dr. Villar, Dr. Añón, Dr. Ferrando, Ms. Fernández, and Dr. González-Martin received grant support from the Instituto de Salud Carlos III, Madrid, Spain (CB06/06/1088). Dr. Hernández-González is a Serra Húnter fellow. Dr. Slutsky was funded by the Canadian Institutes of Health Research (grants numbers 137772 and FDN143285). The remaining authors have disclosed that they do not have any potential conflicts of interest.

Comment in

References

    1. Villar J, Slutsky AS: Golden anniversary of the acute respiratory distress syndrome: Still much work to do!. Curr Opin Crit Care. 2017; 23:4–9
    1. Ferring M, Vincent JL: Is outcome from ARDS related to the severity of respiratory failure? Eur Respir J. 1997; 10:1297–1300
    1. Villar J, Blanco J, del Campo R, et al.; Spanish Initiative for Epidemiology, Stratification & Therapies for ARDS (SIESTA) Network: Assessment of PaO 2 /FiO 2 for stratification of patients with moderate and severe acute respiratory distress syndrome. BMJ Open. 2015; 5:e006812
    1. Villar J, González-Martín JM, Ambrós A, et al.; Spanish Initiative for Epidemiology, Stratification and Therapies of ARDS (SIESTA) Network: Stratification for identification of prognostic categories in the acute respiratory distress syndrome (SPIRES) score. Crit Care Med. 2021; 49:e920–e930
    1. Brower RG, Matthay MA, Morris A, et al.; Acute Respiratory Distress Syndrome Network: Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000; 342:1301–1308

Publication types

LinkOut - more resources