Stratification for Identification of Prognostic Categories In the Acute RESpiratory Distress Syndrome (SPIRES) Score

Abstract

OBJECTIVES: To develop a scoring model for stratifying patients with acute respiratory distress syndrome into risk categories (Stratification for identification of Prognostic categories In the acute RESpiratory distress syndrome score) for early prediction of death in the ICU, independent of the underlying disease and cause of death. DESIGN: A development and validation study using clinical data from four prospective, multicenter, observational cohorts. SETTING: A network of multidisciplinary ICUs. PATIENTS: One-thousand three-hundred one patients with moderate-to-severe acute respiratory distress syndrome managed with lung-protective ventilation. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The study followed Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis guidelines for prediction models. We performed logistic regression analysis, bootstrapping, and internal-external validation of prediction models with variables collected within 24 hours of acute respiratory distress syndrome diagnosis in 1,000 patients for model development. Primary outcome was ICU death. The Stratification for identification of Prognostic categories In the acute RESpiratory distress syndrome score was based on patient’s age, number of extrapulmonary organ failures, values of end-inspiratory plateau pressure, and ratio of Pao 2 to Fio 2 assessed at 24 hours of acute respiratory distress syndrome diagnosis. The pooled area under the receiver operating characteristic curve across internal-external validations was 0.860 (95% CI, 0.831–0.890). External validation in a new cohort of 301 acute respiratory distress syndrome patients confirmed the accuracy and robustness of the scoring model (area under the receiver operating characteristic curve = 0.870; 95% CI, 0.829–0.911). The Stratification for identification of Prognostic categories In the acute RESpiratory distress syndrome score stratified patients in three distinct prognostic classes and achieved better prediction of ICU death than ratio of Pao 2 to Fio 2 at acute respiratory distress syndrome onset or at 24 hours, Acute Physiology and Chronic Health Evaluation II score, or Sequential Organ Failure Assessment scale. CONCLUSIONS: The Stratification for identification of Prognostic categories In the acute RESpiratory distress syndrome score represents a novel strategy for early stratification of acute respiratory distress syndrome patients into prognostic categories and for selecting patients for therapeutic trials.

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

“…We also examined whether the ML models provided an improvement in the prediction of ICU mortality when compared with the SPIRES score that we previously reported (4). We calculated measures to assess the validation of the prediction models, related to calibration and discrimination, by studying the external validity of the models developed in 1,000 patients and tested in 303 patients (36, 37) (detailed in Supplemental File, http://links.lww.com/CCM/H413).…”

“…Once most relevant variables were selected by a genetic algorithm (GA) in the dataset of 1,000 patients, this dataset was divided into five folders to perform five-fold randomized cross-validation repeated 100 times using machine learning. AIC = Akaike information criterion, BIC = Bayesian information criterion, RF = random forest, SPIRES = a four-variable score as an acronym for ”Stratification for Prognostic categories In the Acute RESpiratory distress syndrome” (see Villar et al [4]), XGBoost = extreme gradient boosting.…”

“…Using a large number of patients with moderate-to-severe ARDS admitted to a network of ICUs from several geographical areas of Spain, we aimed to assess the value of ML techniques to predict ICU mortality using variables collected within the first 24 hours of diagnosis of moderate-to-severe ARDS. We compared the prognostic abilities of ML and logistic regression models with each other, and with our previously described Stratification for identification of Prognostic categories In acute RESpiratory distress syndrome (SPIRES) score (4).…”

Association of platelet count with all‐cause mortality from acute respiratory distress syndrome: A cohort study

“…

Acute respiratory distress syndrome (ARDS) is characterized by noncardiogenic pulmonary edema and refractory hypoxemia, which is a serious complication associated with high mortality. [1][2][3] Despite lung protection ventilation and intravenous steroids, many patients are still at risk of respiratory failure and death. 4,5 Previous studies have evaluated the risk factors for death in patients with ARDS.

…”

Clinical relevance of timing of assessment of ICU mortality in patients with moderate-to-severe Acute Respiratory Distress Syndrome

“…We performed an ancillary analysis of data derived from 1580 adult patients with moderate-to-severe ARDS 16 managed with lung-protective MV in a network of ICUs under the Spanish Initiative for Epidemiology, Stratification, and Therapies of ARDS (SIESTA) (A full list of members and their affiliations appears in the Supplemental File ), as previously described 1 , 9 – 13 . The current study was conducted in two steps.…”