Dynamic Parameters within the First 72 H of ICU Admission Predict Extubation Failure and 28-Day Mortality in Severe Pneumonia-Induced ARDS: A Retrospective Cohort Study.

Abstract

Objective: To evaluate dynamic parameter changes within 72 h of intensive care unit (ICU) admission for predicting extubation failure and 28-day mortality in patients with severe pneumonia-induced acute respiratory distress syndrome (ARDS).

Methods: This retrospective cohort study enrolled 424 adults with severe pneumonia-induced ARDS receiving invasive ventilation ≥ 48 h during January 2023-August 2025. We collected clinical data and calculated 72-h changes (Δ) in key parameters: physiological stress [maximum respiratory rate (RR), mean heart rate (HR)] and disease progression [ΔPaO₂/FiO₂, Δ blood lactate (BLA), Δ procalcitonin (PCT), Δ Sequential Organ Failure Assessment (SOFA)]. Outcomes were extubation failure (first spontaneous breathing trial failure or reintubation ≤ 48 h) and 28-day mortality. Multivariable logistic and Cox regression models were built, with discrimination assessed by ROC curves.

Results: Extubation failure and 28-day mortality rates were 23.82% (101/424) and 29.48% (125/424), respectively. For extubation failure, independent risk factors included older age, higher APACHE II score at admission, immunosuppression, higher maximum RR, higher mean HR, and increased ΔBLA, ΔPCT, and ΔSOFA (protective factor: increased ΔPaO₂/FiO₂) (p < 0.05). The prediction model had an AUC of 0.912 (95% CI, 0.883-0.941). For 28-day mortality, independent risk factors were higher APACHE II score at admission, higher maximum RR, increased ΔBLA, and increased ΔSOFA (p < 0.05), with a time-dependent AUC of 0.755 (95% CI, 0.729-0.781). A significant association was observed between extubation failure and 28-day mortality, with a markedly higher mortality rate in patients with extubation failure compared to those with successful extubation (81.19% vs. 13.31%, p < 0.001).

Conclusion: Dynamic parameters within 72 h of ICU admission are predictors of extubation failure and 28-day mortality in severe pneumonia-induced ARDS, offering a tool for early risk stratification. Extubation failure was also strongly associated with increased short-term mortality, underscoring its clinical significance as an adverse outcome during the disease course.