Abstract Background Noninvasive carbon dioxide monitoring is an essential component of chronic obstructive pulmonary disease (COPD) management. However, the comparative accuracy of end-tidal (PetCO2) and transcutaneous CO2 (PtcCO2) monitoring in COPD remains uncertain. This study aimed to systematically evaluate the agreement between PetCO2 obtained using prolonged expiration with an integrated correction algorithm (PetCO2-PA) and PtcCO2 with partial pressure of arterial CO2 (PaCO2) in patients with COPD. Methods In this single-center study, 83 patients with COPD (48 in a stable phase and 35 during acute exacerbation) underwent simultaneous measurement of PetCO2-PA, PtcCO2, and arterial blood gas (ABG) analysis. Agreement between noninvasive and arterial CO2 measurements was assessed using Bland-Altman analysis (bias and limits of agreement), intraclass correlation coefficients (ICC), and receiver operating characteristic (ROC) curve analysis for the detection of hypercapnia, defined as PaCO2 ≥45 mmHg and ≥50 mmHg. Results Both PtcCO2 and PetCO2-PA demonstrated strong correlations with PaCO2 (r = 0.91 and r = 0.88, respectively; p 0.0001). Bland-Altman analysis revealed a mean bias of − 1.7 mmHg (limits of agreement: −8.6 to 5.1) for PtcCO2 and −2.4 mmHg (−9.9 to 5.1) for PetCO2-PA. This close agreement persisted across disease states (stable vs. exacerbation) and the presence or not of hypercapnia. Both noninvasive methods exhibited excellent diagnostic accuracy for hypercapnia (PaCO2 ≥45 or 50 mmHg), each achieving an area under the curve (AUC) greater than 0.94, with no statistically significant inter-method differences. The proportions of measurements exceeding the clinical acceptability thresholds of ± 4 mmHg and ±7 mmHg did not differ significantly between techniques. Conclusion PetCO2-PA demonstrated clinically acceptable agreement with PaCO2, non-inferior to PtcCO2, in patients with COPD. Owing to its cost-effectiveness, rapid operation, and portability, PetCO2-PA represents a practical and reliable alternative for initial screening of respiratory failure in COPD management. This abstract is funded by: None
Zha et al. (Fri,) studied this question.