Daily Ards Research Analysis
Analyzed 14 papers and selected 3 impactful papers.
Summary
Three studies advance precision in ARDS care: a multicenter pediatric study introduces a driving pressure–based oxygenation distension index (ODI) that outperforms the conventional oxygenation index; a stacked-ensemble model predicts ARDS in intra-abdominal sepsis with external validation and interpretability; and a large retrospective analysis links mechanical power and its components to mortality with phenotype-specific effects.
Research Themes
- Ventilator-induced lung injury metrics and physiology-aligned severity indices
- Machine-learning risk stratification for ARDS in sepsis
- Phenotype-informed ventilation strategies using mechanical power
Selected Articles
1. A novel oxygenation distension index (ODI): a driving pressure-based metric compared with the oxygenation index in pediatric PARDS.
In a six-center observational study of 56 children with PARDS, a driving pressure–based oxygenation distension index (ODI) correlated more strongly than the traditional oxygenation index with lung injury markers and better discriminated 30-day mortality. ODI aligns severity assessment with distending forces relevant to ventilator-induced lung injury.
Impact: By directly incorporating driving pressure, ODI offers a physiology-grounded severity metric that outperforms the current standard (OI) in mortality discrimination, potentially reshaping PARDS classification and monitoring.
Clinical Implications: Consider integrating ODI into PARDS severity assessment and risk stratification, especially where advanced respiratory mechanics are measured; prospective validation and workflow integration are needed before widespread adoption.
Key Findings
- ODI correlated more strongly than OI with LUS, PL, DPL, MP/PBW, and MPL/PBW.
- ODI improved 30-day mortality discrimination versus OI (AUC 0.740 vs 0.685; DeLong p=0.040).
- In 56 PARDS patients across six centers, 27% had severe disease; ODI tracked severity better than OI.
Methodological Strengths
- Multicenter physiologic assessment including transpulmonary pressures and lung ultrasound
- Predefined analytic comparisons of ODI versus OI with appropriate discrimination statistics
Limitations
- Small sample size (n=56) limits precision and generalizability
- Observational design; potential measurement heterogeneity across centers
Future Directions: Prospective, larger multicenter validation and interventional trials testing ODI-guided ventilation targets and clinical outcomes.
UNLABELLED: The oxygenation index (OI) is widely used to assess disease severity in pediatric ARDS; however, it relies on mean airway pressure (Pmean), which does not reflect lung-distending forces relevant to ventilator-induced lung injury. Driving pressure (DP) is a determinant of lung strain. We developed and evaluated the oxygenation distension index (ODI) by substituting DP for Pmean in the OI formula. This observational study included children aged 1 month to 18 years with PARDS requiring ≥ 24 h of invasive mechanical ventilation in six centers. Lung ultrasound scores (LUS) and respiratory mechanics, including transpulmonary pressure (PL), transpulmonary driving pressure (DPL), mechanical power (MP), and transpulmonary mechanical power (MPL), were assessed 4-24 h after diagnosis according to PALICC-2 recommendations. Associations of OI and ODI with lung injury markers and clinical outcomes were analyzed. Among 56 patients, 41 (73%) had mild-to-moderate and 15 (27%) had severe PARDS. ODI demonstrated stronger correlations than OI with lung injury markers, including LUS (r = 0.818 vs. 0.501), PL (r = 0.779 vs. 0.697), DPL (r = 0.745 vs. 0.671), MP normalized to predicted body weight (r = 0.518 vs. 0.438), and MPL normalized to predicted body weight (r = 0.634 vs. 0.628). For 30-day mortality prediction, ODI showed higher discriminative performance than OI (AUC 0.740 [95% CI, 0.598-0.866] vs. 0.685 [95% CI, 0.535-0.835]; DeLong p = 0.040). CONCLUSIONS: By incorporating driving pressure, ODI more closely reflects lung injury severity than the conventional oxygenation index and improves discrimination for 30-day mortality in PARDS. WHAT IS KNOWN: • The oxygenation index is the standard metric used for severity classification in pediatric acute respiratory distress syndrome. • Driving pressure has been increasingly recognized as a key physiological determinant of lung injury and clinical outcomes in ARDS. WHAT IS NEW: • A driving pressure-based oxygenation metric, the oxygenation distension index (ODI), shows stronger associations with established markers of lung injury than the conventional oxygenation index in PARDS. • ODI provides improved discrimination for 30-day mortality compared with the oxygenation index in mechanically ventilated children with PARDS.
2. Development and validation of a prognostic model for acute respiratory distress syndrome in critically Ill patients with intra-abdominal sepsis: a multicenter cohort study.
Using MIMIC-IV and eICU-CRD data with external validation, a stacked-ensemble model predicted ARDS during ICU stay in intra-abdominal sepsis with AUCs of 0.811, 0.794, and 0.756. SHAP interpretability highlighted mechanical ventilation as most influential and suggested early vasoactive use associated with lower ARDS risk.
Impact: This externally validated, interpretable model enables early ARDS risk stratification in a high-risk septic population and provides a practical web calculator for clinical decision support.
Clinical Implications: Clinicians can use the web-based calculator to identify intra-abdominal sepsis patients at high risk for ARDS and tailor monitoring, ventilation readiness, and preventive strategies accordingly.
Key Findings
- Stacked-ensemble model achieved AUCs of 0.811 (development), 0.794 (internal validation), and 0.756 (external validation).
- Fourteen predictors retained; mechanical ventilation was the most influential per SHAP.
- Early use of vasoactive agents was associated with reduced ARDS risk.
- A web-based risk calculator was deployed for clinical use.
Methodological Strengths
- External validation in an independent cohort and use of SHAP for interpretability
- Robust feature selection (Boruta, LASSO, logistic regression) and stacked ensemble of 10 base learners
Limitations
- Retrospective data sources may harbor unmeasured confounding and coding biases
- External validation from a single center may limit geographic generalizability
Future Directions: Prospective impact studies to assess whether model-guided interventions reduce ARDS incidence and improve outcomes; broader multi-regional external validations.
BACKGROUND: To develop and externally validate a machine learning-based model for predicting the risk of acute respiratory distress syndrome (ARDS) in patients with intra-abdominal sepsis. METHODS: Data were obtained from the MIMIC-IV and the eICU-CRD database, including patients diagnosed with intra-abdominal sepsis. ARDS occurrence during intensive care unit (ICU) stay was defined as the primary outcome. Feature selection was performed using a combination of the Boruta algorithm, LASSO regression, and logistic regression. Ten base machine learning algorithms were trained and integrated into a stacked ensemble model. Model performance was systematically evaluated, and interpretability was assessed using SHapley Additive exPlanations (SHAP). External validation was conducted in an independent cohort of patients with intra-abdominal sepsis admitted to the First Affiliated Hospital of Xinjiang Medical University between 2016 and 2024. A web-based risk prediction calculator was subsequently developed to facilitate clinical decision support. RESULTS: Among 1,120 patients included from the MIMIC-IV and eICU-CRD databases, 554 (49.46%) developed ARDS during their ICU stay. Fourteen predictors were retained, including mechanical ventilation, use of vasoactive agents, history of chronic pulmonary disease, Sequential Organ Failure Assessment (SOFA) score, Glasgow Coma Scale (GCS) score, key vital signs, and routine laboratory indicators. The stacking model achieved areas under the receiver operating characteristic curve (AUC) of 0.811 in the development cohort, 0.794 in the internal validation cohort, and 0.756 in the external validation cohort. SHAP analysis identified mechanical ventilation as the most influential predictor, while early vasoactive agents use was associated with a reduced ARDS risk. CONCLUSION: A stacked ensemble model for predicting ARDS risk in patients with intra-abdominal sepsis demonstrated robust performance, stability, and interpretability. This model provides a practical tool for early risk stratification and informed clinical decision-making.
3. Exploring the association of mechanical power with mortality and phenotypes among patients with acute respiratory distress syndrome: a retrospective analysis.
In 1,333 ARDS patients from MIMIC-IV, higher mechanical power was independently associated with increased 28-day mortality, with an optimal threshold of 18.7 J/min. The elastic-dynamic component drove risk most strongly, and three ARDS phenotypes exhibited distinct mortality profiles and responses to mechanical power.
Impact: Defines a clinically relevant mechanical power threshold and reveals phenotype-specific risk relationships, informing personalized ventilation strategies beyond traditional oxygenation metrics.
Clinical Implications: Targeting lower mechanical power (e.g., <18.7 J/min) and adjusting respiratory rate may reduce risk; integrating phenotype-informed strategies could refine ventilator management and trial design.
Key Findings
- Mechanical power <18.7 J/min associated with lower 28-day mortality.
- Elastic-dynamic component had strongest mortality association; resistive component was not significant.
- Respiratory rate emerged as the strongest predictor of mortality.
- Three ARDS phenotypes showed distinct mortality and differential association with high mechanical power.
Methodological Strengths
- Large sample size with multiple statistical approaches (logistic, Cox, KM) and unsupervised clustering
- Component-level analysis of mechanical power to elucidate physiologic drivers
Limitations
- Single-database retrospective design limits causal inference and may include residual confounding
- Lack of prospective validation or intervention to test MP-targeted strategies
Future Directions: Prospective trials testing mechanical power–targeted ventilation and phenotype-adapted strategies; external validation across diverse ICUs.
INTRODUCTION: Mechanical power (MP) quantifies the energy delivered from a ventilator to the respiratory system and is a key contributor to ventilator-induced lung injury. This study evaluated the association between MP and mortality in patients with acute respiratory distress syndrome (ARDS), and examined whether this relationship differs across data-driven ARDS phenotypes. METHODS: Patients with ARDS requiring invasive ventilation were identified from the MIMIC-IV database. The association between MP and mortality was assessed using logistic regression, Kaplan-Meier survival analysis, and Cox proportional hazards models. The optimal MP threshold was determined using maximally selected rank statistics. Unsupervised clustering was used to identify ARDS phenotypes and evaluate phenotype-specific responses to MP. RESULTS: A total of 1,333 patients were included. An MP < 18.7 J/min was associated with significantly lower 28-day mortality. Among MP components, the elastic-dynamic component showed the strongest association with mortality; the elastic-static component had a weaker association, and the resistive component was not significant. Respiratory rate was the strongest predictor of mortality. Three phenotypes were identified. Phenotype I (mechanical stress-dominant): poor respiratory mechanics and the highest mortality. Phenotype II (oxygenation-preserved with mild inflammation): better oxygenation, less organ dysfunction, and the lowest mortality. Phenotype III (systemic hyperinflammation with metabolic dysregulation): significant laboratory abnormalities, strong association with high MP, and increased mortality. DISCUSSION: High mechanical power (MP) was independently associated with increased mortality in patients with ARDS. An MP threshold of 18.7 J/min demonstrated prognostic relevance for mortality risk stratification, and the association between MP and outcomes varied across ARDS phenotypes, highlighting the potential value of phenotype-informed ventilation strategies.