Key pointsr Impaired pharyngeal anatomy and increased airway collapsibility is a major cause of obstructive sleep apnoea (OSA) and a mediator of its severity.r Upper airway reflexes to changes in airway pressure provide important protection against airway closure. r This study shows increased pharyngeal collapsibility and attenuated genioglossus reflex responses during expiration in people with OSA.Abstract Upper airway collapse contributes to obstructive sleep apnoea (OSA) pathogenesis. Pharyngeal dilator muscle activity varies throughout the respiratory cycle and may contribute to dynamic changes in pharyngeal collapsibility. However, whether upper airway collapsibility and reflex responses to changes in airway pressure vary throughout the respiratory cycle in OSA is unclear. Thus, this study quantified differences in upper airway collapsibility and genioglossus electromyographic (EMG) activity and reflex responses during different phases of the respiratory cycle. Twelve middle-aged people with OSA (2 female) were fitted with standard polysomnography equipment: a nasal mask, pneumotachograph, two fine-wire intramuscular electrodes into the genioglossus, and a pressure catheter positioned at the epiglottis and a second at the choanae (the collapsible portion of the upper airway). At least 20 brief (ß250 ms) pressure pulses (ß−11 cmH 2 O at the mask) were delivered every 2-10 breaths during four conditions: (1) early inspiration, (2) mid-inspiration, (3) early expiration, and (4) mid-expiration. Mean baseline genioglossus EMG activity 100 ms prior to pulse delivery Amal Osman is a Research Associate at the Adelaide Institute for Sleep Health, Flinders University. She completed her PhD at the University of New South Wales and Neuroscience Research Australia. She is investigating development of simple clinical tools to measure upper airway collapsibility in obstructive sleep apnoea, a critical component in the pathogenesis of this common sleep-related breathing disorder. Current interests include understanding the mechanisms that contribute to airway closure in sleep apnoea. The hope is that this work brings us closer to developing new clinical tools to predict and optimize responses to targeted and novel treatments. J Physiol 598.3 and genioglossus reflex responses were quantified for each condition. The upper airway collapsibility index (UACI), quantified as 100 × (nadir choanal − epiglottic pressure)/nadir choanal pressure during negative pressure pulses, varied throughout the respiratory cycle (early inspiration = 43 ± 25%, mid-inspiration = 29 ± 19%, early expiration = 83 ± 19% and mid-expiration = 95 ± 11% (mean ± SD) P < 0.01). Genioglossus EMG activity was lower during expiration (e.g. mid-expiration vs. mid-inspiration = 76 ± 23 vs. 127 ± 41% of early-inspiration, P < 0.001). Similarly, genioglossus reflex excitation was delayed (39 ± 11 vs. 23 ± 7 ms, P < 0.001) and reflex excitation amplitude attenuated during mid-expiration versus early inspiration (209 ± 36 vs. 286 ± 80%, P = 0.009). These findings ...