Abstract
Background and Objectives: Acute respiratory failure (ARF) has a heterogeneous course in the emergency department (ED), and early prediction of noninvasive mechanical ventilation (NIMV) failure is difficult. The PaCO(2)-ETCO(2) gap reflects ventilation-perfusion mismatch and increased physiologic dead space; however, the prognostic value of its short-term change during NIMV is unclear. This study evaluated baseline, post-treatment, and delta (post-pre) PaCO(2)-ETCO(2) gap values for predicting intubation, intensive care unit (ICU) admission, and mortality in ED patients with ARF receiving NIMV. Materials and Methods: This prospective observational study enrolled adults (≥18 years) treated with NIMV in a tertiary ED. Exclusion criteria included GCS < 15, intoxication, pneumothorax, trauma, pregnancy, gastrointestinal bleeding, need for immediate intubation/CPR, or incomplete data. ETCO(2) was recorded within the first 3 min of NIMV and at 30 min; concurrent arterial blood gases provided PaCO(2). The PaCO(2)-ETCO(2) gap was calculated at both time points and as delta. Outcomes were intubation, ICU admission, and mortality. ROC analyses determined discriminatory performance and cutoffs using the Youden index. Results: Thirty-four patients were included (50% female; mean age 73.26 ± 10.07 years). Intubation occurred in 9 (26.5%), ICU admission in 20 (58.8%), and mortality in 10 (29.4%). The post-treatment gap and delta gap were significantly higher in intubated patients (p = 0.007 and p = 0.001). For predicting intubation, post-treatment gap > 10.90 mmHg yielded AUC 0.807 (p = 0.007; sensitivity 77.8%, specificity 76.0), while delta gap > 2.90 mmHg yielded AUC 0.982 (p = 0.001; sensitivity 88.9%, specificity 92.0). Delta gap also predicted ICU admission (cutoff > 0.65 mmHg; AUC 0.746, p = 0.016) and mortality (cutoff > 2.90 mmHg; AUC 0.865, p = 0.001). Conclusions: In ED ARF patients receiving NIMV, an increasing PaCO(2)-ETCO(2) gap-especially the delta gap-was associated with higher risks of intubation, ICU admission, and mortality, supporting serial CO(2) gap monitoring as a practical early warning marker of deterioration.