RESUMO
Weaning patients from mechanical ventilation (MV) is a critical and resource intensive process in the Intensive Care Unit (ICU) that impacts patient outcomes and healthcare expenses. Weaning methods vary widely among providers. Prolonged MV is associated with adverse events and higher healthcare expenses. Predicting weaning readiness is a non-trivial process in which the positive end-expiratory pressure (PEEP), a crucial component of MV, has potential to be indicative but has not yet been used as the target. We aimed to predict successful weaning from mechanical ventilation by targeting changes in the PEEP-level using a supervised machine learning model. This retrospective study included 12,153 mechanically ventilated patients from Medical Information Mart for Intensive Care (MIMIC-IV) and eICU collaborative research database (eICU-CRD). Two machine learning models (Extreme Gradient Boosting and Logistic Regression) were developed using a continuous PEEP reduction as target. The data is splitted into 80% as training set and 20% as test set. The model's predictive performance was reported using 95% confidence interval (CI), based on evaluation metrics such as area under the receiver operating characteristic (AUROC), area under the precision-recall curve (AUPRC), F1-Score, Recall, positive predictive value (PPV), and negative predictive value (NPV). The model's descriptive performance was reported as the variable ranking using SHAP (SHapley Additive exPlanations) algorithm. The best model achieved an AUROC of 0.84 (95% CI 0.83-0.85) and an AUPRC of 0.69 (95% CI 0.67-0.70) in predicting successful weaning based on the PEEP reduction. The model demonstrated a Recall of 0.85 (95% CI 0.84-0.86), F1-score of 0.86 (95% CI 0.85-0.87), PPV of 0.87 (95% CI 0.86-0.88), and NPV of 0.64 (95% CI 0.63-0.66). Most of the variables that SHAP algorithm ranked to be important correspond with clinical intuition, such as duration of MV, oxygen saturation (SaO2), PEEP, and Glasgow Coma Score (GCS) components. This study demonstrates the potential application of machine learning in predicting successful weaning from MV based on continuous PEEP reduction. The model's high PPV and moderate NPV suggest that it could be a useful tool to assist clinicians in making decisions regarding ventilator management.
RESUMO
RATIONALE: Steroid profiles in combination with a corticotropin stimulation test provide information about steroidogenesis and its functional reserves in critically ill patients. OBJECTIVES: We investigated whether steroid profiles before and after corticotropin stimulation can predict the risk of in-hospital death in sepsis. METHODS: An exploratory data analysis of a double blind, randomized trial in sepsis (HYPRESS [HYdrocortisone for PRevention of Septic Shock]) was performed. The trial included adult patients with sepsis who were not in shock and were randomly assigned to placebo or hydrocortisone treatment. Corticotropin tests were performed in patients prior to randomization and in healthy subjects. Cortisol and precursors of glucocorticoids (17-OH-progesterone, 11-desoxycortisol) and mineralocorticoids (11-desoxycorticosterone, corticosterone) were analyzed using the multi-analyte stable isotope dilution method (LC-MS/MS). Measurement results from healthy subjects were used to determine reference ranges, and those from placebo patients to predict in-hospital mortality. MEASUREMENTS AND MAIN RESULTS: Corticotropin tests from 180 patients and 20 volunteers were included. Compared to healthy subjects, patients with sepsis had elevated levels of 11-desoxycorticosterone and 11-desoxycortisol, consistent with activation of both glucocorticoid and mineralocorticoid pathways. After stimulation with corticotropin, the cortisol response was subnormal in 12% and the corticosterone response in 50% of sepsis patients. In placebo patients (n = 90), a corticotropin-stimulated cortisol-to-corticosterone ratio > 32.2 predicted in-hospital mortality (AUC 0.8 CI 0.70-0.88; sensitivity 83%; and specificity 78%). This ratio also predicted risk of shock development and 90-day mortality. CONCLUSIONS: In this exploratory analysis, we found that in sepsis mineralocorticoid steroidogenesis was more frequently impaired than glucocorticoid steroidogenesis. The corticotropin-stimulated cortisol-to-corticosterone ratio predicts the risk of in-hospital death. Trial registration Clinical trial registered with www. CLINICALTRIALS: gov Identifier: NCT00670254. Registered 1 May 2008, https://clinicaltrials.gov/ct2/show/NCT00670254 .
