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

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, Pao 2/Fio 2, 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.

Comparing Traditional Regression and Machine Learning Models in Predicting Acute Respiratory Distress Syndrome Mortality

“…The article (1) illustrates that the distinctions between these models are minimal. Upon examination of the supplementary data, one can observe similar metrics across these models.…”

“…To the Editor: I n the study by Villar et al (1), the authors delve into the predictive modeling of mortality in moderate to severe acute respiratory distress syndrome (ARDS). We appreciate the pioneering integration of data science techniques into this medical prediction framework.…”

“…Jesús Villar, MD, PhD, FCCM [1][2][3] Jesús M. González-Martin, PhD 2,3 Tamas Szakmany, MD, PhD, FCCM 4,5 The authors reply:…”

“…W e thank Valiente Fernández al (1) for their interest in our recent Predicting Outcome and STratification of severity in ARDS (POSTCARDS) study (2) on predicting ICU mortality in patients with acute respiratory distress syndrome (ARDS) using machine learning (ML) published in Critical Care Medicine.…”

POSTCARDS from a SIESTA: Crossing the Translational and Generalizability Gap for Predictive Models of Acute Respiratory Distress Syndrome-Related Mortality*

“…In this issue of Critical Care Medicine , the study by Villar et al (1) tests a new mortality prediction score in Spanish Initiative for Epidemiology, Stratification and Therapies for Acute Respiratory Distress Syndrome (SIESTA) (ALIEN, STANDARDS, STANDARDS-2) and externally validates in prevalence and outcome of acute hypoxemic respiratory failure (PANDORA) data. This novel dataset in SIESTA contains three trials with 1,000 patients with moderate-to-severe acute respiratory distress syndrome (ARDS).…”