Influence of intercostal muscles contraction on sonographic evaluation of lung sliding: a physiological study on healthy subjects

objectives: to investigate the following: (a) effects of intercostal muscle contraction on sonographic assessment of lung sliding and (b) inter-rater and intra-observer agreement on sonographic detection of lung sliding and lung pulse. methods: we used Valsalva and muller maneuvers as experimental models in which closed glottis and clipped nose prevent air from entering the lungs, despite sustained intercostal muscles contraction. twenty-one healthy volunteers underwent bilateral lung ultrasound during tidal breathing, apnea, hyperventilation, and muller and valsalva maneuvers. the same expert recorded 420 B-mode clips and 420 M-mode images, independently evaluated for the presence or absence of lung sliding and lung pulse by three raters unaware of the respiratory activity corresponding to each imaging. results: during muller and valsalva maneuvers, lung sliding was certainly recognized in up to 73.0% and up to 68.7% of imaging, respectively, with a slight to fair inter-rater agreement for muller maneuver and slight to moderate for valsalva. Lung sliding was unrecognized in up to 42.0% of tidal breathing imaging, and up to 12.5% of hyperventilation imaging, with a slight to fair inter-rater agreement for both. during apnea, interpretation errors for sliding were irrelevant and inter-rater agreement moderate to perfect. even if intra-observer agreement varied among raters and throughout respiratory patterns, we found it to be higher than inter-rater reliability. conclusions: Intercostal muscles contraction produces sonographic artifacts that may simulate lung sliding. clinical studies are needed to confirm this hypothesis. we found slight to moderate inter-rater agreement and globally moderate to almost perfect intra-observer agreement for lung sliding and lung pulse. trial registration: clinical trials.gov registration number. NCT02386696.