Can laryngeal ultrasound detect laryngeal movements during mechanically assisted cough?

BACKGROUND: Mechanical insufflation-exsufflation (MI-E) is used to augment ineffective cough, but treatment efficacy depends on appropriate upper airway response. Transnasal flexible laryngoscopy (TFL) is the reference standard for assessing laryngeal responses, but is invasive. This study evaluated whether laryngeal ultrasonography (US) can assess laryngeal responses during MI-E. METHODS: In this cross-sectional study, healthy volunteers underwent MI-E with both passive and active cough efforts, while TFL and US were performed simultaneously. Laryngeal US was applied using anterior and lateral approaches. Recordings were analysed retrospectively frame-by-frame by two blinded raters. Assessability was defined as appropriate visualisation of the vocal folds or aryepiglottic folds. Visualisation success rates and concordance between laryngeal US and TFL findings were calculated across pre-specified MI-E cycles. Inter-rater reliability for US interpretation was assessed in a subset. RESULTS: 30 adult participants and 875 MI-E cycles were included. At least one laryngeal US approach yielded assessable data in all participants, with lower assessability in males. Across MI-E cycles, visualisation success rates were higher for TFL than for laryngeal US (93.5% versus 68%). Concordance between US and TFL was 78% and 79% during passive and active insufflation, 80% and 88% during passive and active exsufflation, and 92% during active cough compression. Inter-rater reliability was substantial (κ=0.62). CONCLUSION: This exploratory study suggests that laryngeal US may offer a feasible noninvasive approach to assess laryngeal responses during MI-E, but TFL remains the reference standard. Further validation in patient populations is required.