Effect of Intrapulmonary Gas Composition on Lung Ultrasound A-Line Artifacts and Grayscale Intensity: A Prospective Exploratory Study with Heliox in Healthy Volunteers

Background/Objectives: Lung ultrasound (LUS) of the aerated lung relies on artifacts generated at the pleural interface, but the contribution of intrapulmonary gas to these artifacts has not been studied in humans. Heliox, a helium–oxygen mixture with low density and acoustic impedance, allows the gas phase of the lung to be modified without altering its structure. We examined whether heliox breathing alters A-line artifacts and grayscale intensity in LUS images. Methods: In this prospective exploratory study, 52 healthy volunteers underwent LUS at four thoracic locations during ambient-air breathing and subsequently during heliox (80% He/20% O2) breathing. A-lines were counted manually, and grayscale intensity and horizontal reverberations were quantified in ImageJ. Paired differences were analyzed with a clustered Wilcoxon signed-rank test and cluster-bootstrap 95% confidence intervals (CIs). Reliability of A-line counting was assessed in a stratified random subsample (13 participants, 104 images) by blinded test–retest and inter-observer re-counting. Results: Heliox breathing increased global grayscale intensity (mean difference 6.7 grayscale units, 95% CI 4.1–9.2; p < 0.001), most markedly in the deepest image compartment (10.1 units, 95% CI 6.5–13.9), and the number of visible A-lines (0.45, 95% CI 0.30–0.62; p < 0.001). Horizontal reverberations did not differ. In the blinded subsample, the heliox–air difference in A-lines decreased from 0.44 in the original unblinded count to 0.21 in the blinded test–retest and 0.04 in the second reader’s count, and the 95% limits of agreement for manual counting (approximately ±1.2 A-lines per image) exceeded the prespecified minimum relevant difference of one A-line. Conclusions: Heliox breathing was associated with a consistent, depth-dependent increase in the grayscale intensity of lung ultrasound images, suggesting that intrapulmonary gas composition contributes to image brightness. The concomitant change in A-line counts was directionally consistent but smaller than the resolution of manual counting at the imaging depth used and therefore remains inconclusive.