Rationale: Oral appliances (OAs) are commonly used as an alternative treatment to continuous positive airway pressure for patients with obstructive sleep apnea (OSA). However, OAs have variable success at reducing the apnea–hypopnea index (AHI), and predicting responders is challenging. Understanding this variability may lie with the recognition that OSA is a multifactorial disorder and that OAs may affect more than just upper-airway anatomy/collapsibility. Objectives: The objectives of this study were to determine how OA alters AHI and four phenotypic traits (upper-airway anatomy/collapsibility and muscle function, loop gain, and arousal threshold), and baseline predictors of which patients gain the greatest benefit from therapy. Methods: In a randomized crossover study, 14 patients with OSA attended two sleep studies with and without their OA. Under each condition, AHI and the phenotypic traits were assessed. Multiple linear regression was used to determine independent predictors of the reduction in AHI. Measurements and Main Results: OA therapy reduced the AHI (30 ± 5 vs. 11 ± 2 events/h; P < 0.05), which was driven by improvements in upper-airway anatomy/collapsibility under passive (1.9 ± 0.7 vs. 4.7 ± 0.6 L/min; P < 0.005) and active conditions (2.4 ± 0.9 vs. 6.2 ± 0.4 L/min; P < 0.001). No changes were seen in muscle function, loop gain, or the arousal threshold. Using multivariate analysis, baseline passive upper-airway collapsibility and loop gain were independent predictors of the reduction in AHI (r2 = 0.70; P = 0.001). Conclusions: Our findings suggest that OA therapy improves the upper-airway collapsibility under passive and active conditions. Importantly, a greater response to therapy occurred in those patients with a mild anatomic compromise and a lower loop gain.

Study Objectives:. Obstructive sleep apnea (OSA) results from the interaction of several physiological traits; specifically a compromised upper airway

Upper airway surgery is often recommended to treat patients with obstructive sleep apnea (OSA) who cannot tolerate continuous positive airways pressure. However, the response to surgery is variable, potentially because it does not improve the nonanatomical factors (ie, loop gain [LG] and arousal threshold) causing OSA. Measuring these traits clinically might predict responses to surgery. Our primary objective was to test the value of LG and arousal threshold to predict surgical success defined as 50% reduction in apnea-hypopnea index (AHI) and AHI <10 events/hour post surgery. We retrospectively analyzed data from patients who underwent upper airway surgery for OSA (n = 46). Clinical estimates of LG and arousal threshold were calculated from routine polysomnographic recordings presurgery and postsurgery (median of 124 [91–170] days follow-up). Surgery reduced both the AHI (39.1 ± 4.2 vs. 26.5 ± 3.6 events/hour; p < .005) and estimated arousal threshold (−14.8 [−22.9 to −10.2] vs. −9.4 [−14.5 to −6.0] cmH₂O) but did not alter LG (0.45 ± 0.08 vs. 0.45 ± 0.12; p = .278). Responders to surgery had a lower baseline LG (0.38 ± 0.02 vs. 0.48 ± 0.01, p < .05) and were younger (31.0 [27.3–42.5] vs. 43.0 [33.0–55.3] years, p < .05) than nonresponders. Lower LG remained a significant predictor of surgical success after controlling for covariates (logistic regression p = .018; receiver operating characteristic area under curve = 0.80). Our study provides proof-of-principle that upper airway surgery most effectively resolves OSA in patients with lower LG. Predicting the failure of surgical treatment, consequent to less stable ventilatory control (elevated LG), can be achieved in the clinic and may facilitate avoidance of surgical failures.

Upper airway collapsibility is a key determinant of obstructive sleep apnea (OSA) which can influence the efficacy of certain non-continuous positive airway pressure (CPAP) treatments for OSA. However, there is no simple way to measure this variable clinically. The present study aimed to develop a clinically implementable tool to evaluate the collapsibility of a patient’s upper airway. Collapsibility, as characterized by the passive pharyngeal critical closing pressure (P_crit), was measured in 46 patients with OSA. Associations were investigated between P_crit and data extracted from patient history and routine polysomnography, including CPAP titration. Therapeutic CPAP level, demonstrated the strongest relationship to P_crit (r²=0.51, p < .001) of all the variables investigated including apnea-hypopnea index, body mass index, sex, and age. Patients with a mildly collapsible upper airway (P_crit ≤ −2 cmH₂O) had a lower therapeutic CPAP level (6.2 ± 0.6 vs. 10.3 ± 0.4 cmH₂O, p < .001) compared to patients with more severe collapsibility (P_crit > −2 cmH₂O). A therapeutic CPAP level ≤8.0 cmH₂O was sensitive (89%) and specific (84%) for detecting a mildly collapsible upper airway. When applied to the independent validation data set (n = 74), this threshold maintained high specificity (91%) but reduced sensitivity (75%). Our data demonstrate that a patient’s therapeutic CPAP requirement shares a strong predictive relationship with their P_crit and may be used to accurately differentiate OSA patients with mild airway collapsibility from those with moderate-to-severe collapsibility. Although this relationship needs to be confirmed prospectively, our findings may provide clinicians with better understanding of an individual patient’s OSA phenotype, which ultimately could assist in determining which patients are most likely to respond to non-CPAP therapies.

Oronasal masks are frequently used for continuous positive airway pressure (CPAP) treatment in patients with obstructive sleep apnea (OSA). The aim of this study was to (1) determine if CPAP requirements are higher for oronasal masks compared to nasal mask interfaces and (2) assess whether polysomnography and patient characteristics differed among mask preference groups. Retrospective analysis of all CPAP implementation polysomnograms between July 2013 and June 2014. Prescribed CPAP level, polysomnography results and patient data were compared according to mask type (n = 358). Oronasal masks were used in 46%, nasal masks in 35% and nasal pillow masks in 19%. There was no difference according to mask type for baseline apnea-hypopnea index (AHI), body mass index (BMI), waist or neck circumference. CPAP level was higher for oronasal masks, 12 (10–15.5) cm H2O compared to nasal pillow masks, 11 (8–12.5) cm H2O and nasal masks, 10 (8–12) cm H2O, p < 0.0001 (Median [interquartile range]). Oronasal mask type, AHI, age, and BMI were independent predictors of a higher CPAP pressure (p < 0.0005, adjusted R2 = 0.26.). For patients with CPAP ≥ 15 cm H2O, there was an odds ratio of 4.5 (95% CI 2.5–8.0) for having an oronasal compared to a nasal or nasal pillow mask. Residual median AHI was higher for oronasal masks (11.3 events/h) than for nasal masks (6.4 events/h) and nasal pillows (6.7 events/h), p < 0.001. Compared to nasal mask types, oronasal masks are associated with higher CPAP pressures (particularly pressures ≥ 15 cm H2O) and a higher residual AHI. Further evaluation with a randomized control trial is required to definitively establish the effect of mask type on pressure requirements. A commentary on this article appears in this issue on page 1209. Deshpande S, Joosten S, Turton A, Edwards BA, Landry S, Mansfield DR, Hamilton GS. Oronasal masks require a higher pressure than nasal and nasal pillow masks for the treatment of obstructive sleep apnea. J Clin Sleep Med 2016;12(9):1263–1268.