The Relationship between Upper Airway Anatomy and Obesity in Patients with Obstructive Sleep Apnea

Rationale: Obesity is the most important risk factor for obstructive sleep apnea (OSA). However, the complex relationship between obesity and upper airway anatomy (craniofacial structure, soft tissues, and airway caliber) has not been robustly examined in patients with OSA. Objectives: To evaluate the relationship between obesity, on the basis of body mass index (BMI), and upper airway anatomic structures in adult patients with moderate or severe OSA. Methods: In this cross-sectional study, 583 patients with apnea-hypopnea index ≥15 events/h (mean age, 53.7 ± 10.4 yr; 81.0% men) were included from the Iceland Sleep Apnea Cohort. Airway sizes, soft tissue volumes, and craniofacial dimensions were quantified using three-dimensional magnetic resonance imaging. We examined how upper airway anatomy was associated with BMI using linear regression (continuous BMI) and analysis of covariance (BMI categories), adjusting for age, sex, and apnea-hypopnea index. Results: Most upper airway anatomy was significantly associated with BMI among patients with OSA. Higher BMI was associated with a different airway shape, including larger minimum anteroposterior distance at both the retropalatal and retroglossal regions and smaller minimum lateral distance in the retropalatal region. All pharyngeal soft tissues were larger with greater BMI, including the volumes of the tongue (and tongue fat), soft palate (and soft palate fat), lateral walls, fat pads, epiglottis, and pterygoids. Patients with lower BMIs had smaller craniofacial measures (e.g., distances between hyoid, retropogonion, and third cervical vertebrae, intramandibular volume, and nasooropharyngeal areas) and more retrognathia. BMI was only weakly associated with the proportion of mandibular space occupied by soft tissues (with no difference among BMI groups), suggesting comparable intraoral "crowdedness" among patients with OSA at different degrees of obesity, albeit for different reasons. Conclusions: The present results support associations between obesity and airway shape, soft tissue volumes, and craniofacial measures among patients with moderate to severe OSA. These relationships provide insights into anatomic traits leading to OSA in lean and obese patients and can inform more personalized treatment options.