This study evaluated whether breath-by-breath phase III segments of oxygrams and capnograms recorded at the airway opening are qualitatively and quantitatively similar to those obtained directly from peripheral alveolar gas in mechanically ventilated patients. The investigation used video-assisted lobectomy procedures as an opportunity to obtain direct lung parenchyma samples while patients were anesthetized and mechanically ventilated.
The authors hypothesized that the phase III portion of expiratory oxygrams and capnograms collected at the airway opening should be qualitatively similar to phase III obtained directly at the lung acini. The aim was to test this hypothesis by comparing continuous gas tracings recorded at the airway opening with direct alveolar gas samples taken from a peripheral lung lobe in patients undergoing thoracic surgery.
Mechanically ventilated adult patients scheduled for video-assisted lobectomy were used as the clinical model. The operative setting allowed safe access to the peripheral lung parenchyma of the lobe that would be excised, enabling direct sampling of alveolar gas during ongoing mechanical ventilation and anesthesia.
In ten anesthetized patients immediately prior to lobectomy, the investigators punctured the lung lobe scheduled for removal with a 20 G angiocath inserted to approximately 5 mm depth. The angiocath was connected via a sampling line to both sidestream and mainstream capnography systems to record alveolar CO2 and O2 signals.
Simultaneous recordings were obtained at the standard airway opening (the endotracheal tube proximal opening) and from the direct alveolar sampling line. The configuration permitted breath-synchronized, oscillatory gas signals to be observed and compared between alveolar and airway sites on a breath-by-breath basis.
Alveolar CO2 and O2 samples displayed an oscillating pattern in synchrony with the mechanical breath cycles. The alveolar capnograms and oxygrams were characterized by an absence of a discernible phase I, a very short phase II, and a present phase III. Expired gas tracings recorded at the airway opening showed a phase III that appeared qualitatively similar to the alveolar phase III.
The investigators quantified the slope of phase III for expired curves at the airway opening and compared these to slopes from alveolar mainstream capnograms. The median slope of phase III at the airway opening was reported as 2.20 mmHg s-1 with an inter-quartile range (IQR) of 1.28 mmHg s-1. The median slope for alveolar mainstream capnograms was reported as 2.18 mmHg s-1 with an IQR of 1.24 mmHg s-1. The comparison yielded a p value of 0.76, indicating no statistically significant difference in phase III slope between the two sampling locations as reported in the abstract.
Based on the similarity in phase III morphology and in the measured slopes between airway opening and direct alveolar samples, the authors conclude that sampling of alveolar gas at the airway opening can reliably represent mean alveolar gas characteristics in mechanically ventilated patients. The findings emphasize the role of continuous gas diffusion through the alveolar-capillary membrane and support that both convection- and diffusion-dependent inhomogeneities contribute to the origin of phase III in expiratory gas tracings.
These results have implications for the interpretation of capnography and expirography during mechanical ventilation, suggesting that airway measurements—when phase III is adequately sampled—reflect alveolar gas behavior in the studied setting.
The abstract provides the principal methods and key comparative outcomes but does not present full patient demographic details, ventilatory settings (other than that patients were mechanically ventilated), or additional quantitative analyses beyond the reported medians, IQRs, and p value. Full methodological and numerical details, including potential confounders and broader statistical reporting, are accessible in the full text linked through the publisher. The sample size reported in the abstract was ten patients, reflecting a small surgical cohort used to obtain direct alveolar samples.
The demonstrated similarity between airway and direct alveolar phase III slopes supports the utility of noninvasive airway sampling for assessing mean alveolar gas composition in ventilated patients, at least in the intraoperative thoracic surgery model used here. This may reinforce confidence in bedside capnography and volumetric expired gas analysis as proxies for alveolar gas behavior. Further detail and broader validation across different disease states, ventilatory modes, and lung regions would require additional studies; such extensions were not reported in the abstract.
This summary is based on the abstract of the article titled "Direct measurements of alveolar gas in mechanically ventilated humans" (Physiol Meas. 2026;47(8); doi: 10.1088/1361-6579/ae9036). The PubMed record (PMID 42497902) and the journal full-text link provide access to the complete report for readers seeking comprehensive methods and data.