P155 Effects of upper airway muscle training on upper airway physiology in people with obstructive sleep apnea

Introduction Previous studies demonstrate that oropharyngeal exercises can reduce obstructive sleep apnoea (OSA) severity. However, the physiological mechanisms underlying this improvement are unknown. Thus, this study aimed to evaluate the effects of a speech-pathology led, targeted upper airway muscle training protocol on upper airway physiology. Methods People with mild-moderate OSA (n=12 studied to date, 5 females, 7 males) completed 12 weeks of daily upper airway muscle training targeting the muscles of the tongue and soft palate. Pre- and post-training outcome measures included anterior and posterior tongue muscle strength and endurance and upper airway collapsibility via the upper airway collapsibility index. Results Preliminary findings indicate that 12 weeks of targeted exercise training improved anterior and posterior tongue muscle strength, respectively (54.3±12.7 vs. 61.5±7.7kPa, p<0.01, 50.1±8.5 vs. 58.0±8.5kPa, p<0.01), and anterior and posterior tongue muscle endurance, respectively (15.7±10.9 vs. 24.1±8s, p<0.01, 9.5±4.2 vs. 23.3±17.7s, p<0.01). The upper airway collapsibility index improved post-training (25.5±18.9 vs. 12.26±12.11 %, p=0.03). Conclusions Twelve weeks of upper airway muscle training improved tongue muscle strength, endurance and upper airway collapsibility in people with mild-moderate OSA. These physiological changes provide insight into the potential mechanisms mediating reductions in OSA severity with oropharyngeal exercises. This research was supported by the 2019 Phillips/ASA Sleep Research grant.

P070 Oronasal masks are associated with increased airway collapsibility and increased therapeutic CPAP requirements

Introduction Continuous positive airway pressure (CPAP) delivered via oronasal masks are associated with lower adherence, higher residual AHI and CPAP requirement in comparison to nasal masks. Mechanisms contributing to increased CPAP requirement are not well understood. This physiological study aimed to assess the effect of mask type on upper airway anatomy and collapsibility. Methods 13 OSA patients, underwent a sleep study during which they wore both nasal and oronasal mask for half the night each (order randomized). CPAP was manually titrated to determine therapeutic pressure. Passive upper airway collapsibility was assessed using the Pcrit technique. Participants then underwent an MRI wearing both the nasal and oronasal mask. Cine MRI was used to dynamically assess cross-sectional area of the retroglossal airway across the respiratory cycle with each mask interface. Scans were repeated at 4cmH2O, as well as at the nasal and oronasal therapeutic pressures. Results The oronasal mask was associated with both higher therapeutic pressure requirements (∆M±SEM; +2.6±0.5, p<0.001) and higher Pcrit (+2.4±0.5cmH2O, p=0.001) compared to the nasal mask. The change in therapeutic pressure between masks was strongly correlated with the change in Pcrit (r2= 0.73, p=0.003). Preliminary MRI analyses indicate robust increases in cross-sectional area associated with increasing pressure. After controlling for pressure and breath-phase, the retroglossal area was larger when using a nasal compared to an oronasal mask (+12.42±5.87mm2, p=0.03). Conclusions These preliminary findings suggest that oronasal masks worsen the collapsibility of the airway which likely contributes to the need for an elevated therapeutic pressure relative to nasal masks.

O023 Impact of weight loss of OSA pathophysiology

Introduction Obesity is a major risk factor for developing obstructive sleep apnoea (OSA), however, the underlying mechanisms are not fully understood. We aimed to assess the impact of weight loss on all OSA endotypes (i.e. upper airway collapsibility, muscle compensation, respiratory arousal threshold, and loop gain). Methods We analysed data from 40 OSA patients (collated from 3 centres) who underwent bariatric surgery. Demographics and clinical polysomnograms (PSG) were performed at baseline and at between 6–18 months post-surgery. OSA endotypes were measured during sleep using non-invasive endotyping methods (derived from clinical PSG). Results Participants lost 28±14 kg and had a post-surgery reduction in the AHI of 19.6 (Interquartile range[IQR] -9.8 to -35.4) events/hr [from baseline 39.9 (24.3 to 65.6) events/hr to 17.0 (9.9 to 33.3) events/hr]. Following surgical weight loss, there was significant improvement in collapsibility (∆6.2 [IQR -1.4 to 13]%Veupnoea, p<0.0001), as well as significant reduction in loop gain and arousal threshold (∆-0.06 [-0.17 to 0.009], p<0.001 and ∆-13.7 [-24.8 to -1.8]%Veupnoea, p<0.001 respectively). There was no significant change in muscle compensation. Conclusion Our findings suggest that weight loss improves upper airway collapsibility and reduces loop gain and the arousal threshold, providing novel insights about the mechanisms by which obesity causes OSA. Further analysis is underway to determine whether knowledge of the baseline OSA endotypes (in isolation and/or in combination) can predict which individuals will have a response to weight loss alone.

O015 Upper airway collapsibility measurement under anaesthesia in children: feasibility and utility in predicting perioperative respiratory adverse events

Background Perioperative respiratory adverse events (PRAE) pose significant risk in paediatric anaesthesia, and identifying risk is vital. Perioperative measurement of pharyngeal closing pressure (PCLOSE) is a quick, objective method of assessing upper airway collapsibility that may identify PRAE risk. Aim To investigate if PCLOSE measurement is feasible and predictive of PRAE in children. Method Fifty-six children (1-8years, 34 male, without significant co-morbidities) underwent PCLOSE measurements immediately preceding (pre-PCLOSE) and following (post-PCLOSE) adenotonsillectomy. Measurement was performed under anaesthesia while breathing spontaneously in supine posture with head/jaw neutral. After application of a face mask, inspiratory flow was occluded with an associated decrease in mask/nasal pressure seen with each inspiratory effort. With airway collapse, a plateau developed in minimum pressure observed (= PCLOSE): less collapsible airways occluded at more negative pressures. PCLOSE was averaged over 5–6 sequential efforts, at least 3 times on each occasion. Results Both pre-and post-PCLOSE were successfully measured in 94.6% children without affecting procedure. Pre-PCLOSE and change in PCLOSE from pre- to post- were associated with an increased incidence of PRAE (Poisson regression coefficient 0.083(0.03) (mean, SE); p=0.0054 and 0.03(0.01); p=0.018, respectively). There was no significant association between post-PCLOSE and PRAE. The odd of PRAE occurrence during recovery was 1.5 times higher than in other phases. Conclusion This study demonstrated the feasibility of obtaining PCLOSE. A more collapsible airway before surgery and an increase in collapsibility with surgery were both associated with increased PRAE. PCLOSE measurement could be a valuable risk assessment tool for PRAE in children undergoing surgery.

437 Efficacy of a novel oral appliance and the influence of OSA pathophysiological traits on treatment response

Introduction Oral appliance therapy is a recommended alternative to CPAP for OSA. Approximately 50% of patients have a major reduction in OSA severity but successful treatment outcome remains challenging to predict. Previous prediction methods have focused on clinical variables which have poor predictive value. OSA is recognised as a heterogenous disorder caused by 4 pathophysiological traits. The influence of OSA pathophysiological traits on oral appliance treatment outcome has been explored in recent physiological studies using simplified phenotyping methods. In this preliminary report, we prospectively compared differences in the 4 OSA phenotypes between responders and incomplete responders to a novel oral appliance with built in oral airway using gold standard phenotyping methodology. Methods Data from 22 people with OSA (AHI>10events/h) have been analysed to date. A diagnostic in-laboratory PSG was initially conducted to confirm OSA. A detailed physiology PSG was carried out prior to commencement of oral appliance therapy. For this study night participants were instrumented with standard PSG equipment, nasal mask, pneumotachograph, epiglottic pressure catheter and intramuscular electrodes inserted perorally into the genioglossus to quantify baseline OSA phenotypic traits. Pcrit was quantified via CPAP dial downs and the non-anatomical traits were quantified from naturally occurring apneas and hypopneas. Participants were then fitted with a next generation novel oral appliance with a built-in oral airway (Oventus O2Vent Optima™) and titrated to at least 75% of maximum mandibular advancement. After acclimatization to therapy, participants were invited to undergo a treatment efficacy PSG. Results Oral appliance therapy reduced the AHI by 52% (21[15,31] vs. 11[7,16] events/h, p<0.001). 46% of participants responded to oral appliance therapy based on the definition of AHI < 10events/h. Preliminary analyses indicated that estimates of baseline upper airway collapsibility tended to be different in responders versus non-responders (responders have less collapsible airways). Conclusion The novel oral appliance reduced OSA severity by 50% with resolution of OSA in half of participants. Baseline pharyngeal collapsibility may be an important physiological predictor of treatment outcome. Support (if any):

Detailed evaluation of the upper airway in the Dp(16)1Yey mouse model of Down syndrome

A high prevalence of obstructive sleep apnea (OSA) has been reported in Down syndrome (DS) owing to the coexistence of multiple predisposing factors related to its genetic abnormality, posing a challenge for the management of OSA. We hypothesized that DS mice recapitulate craniofacial abnormalities and upper airway obstruction of human DS and can serve as an experimental platform for OSA research. This study, thus, aimed to quantitatively characterize the upper airway as well as craniofacial abnormalities in Dp(16)1Yey (Dp16) mice. Dp16 mice demonstrated craniofacial hypoplasia, especially in the ventral part of the skull and the mandible, and rostrally positioned hyoid. These changes were accompanied with a shorter length and smaller cross-sectional area of the upper airway, resulting in a significantly reduced upper airway volume in Dp16 mice. Our non-invasive approach, a combination of computational fluid dynamics and high-resolution micro-CT imaging, revealed a higher negative pressure inside the airway of Dp16 mice compared to wild-type littermates, showing the potential risk of upper airway collapse. Our study indicated that Dp16 mice can be a useful model to examine the pathophysiology of increased upper airway collapsibility of DS and to evaluate the efficacy of therapeutic interventions for breathing and sleep anomalies.