Abstract
Introduction: Obstructive sleep apnea (OSA) is an extensive public health problem that imposes considerable morbidity. Mandibular advancement splint (MAS) therapy is a well tolerated treatment, but success rates are difficult to predict. Our objective was to investigate the relationship of oropharyngeal airway dimensions, sleep characteristics, patient biometrics, and treatment response within an OSA patient sample. Methods: Records of 33 adults were assessed retrospectively with the use of Dolphin 3D and Image J to measure the airway on pretreatment supine cone-beam computed tomography images and derived lateral cephalograms. The patients used Somnodent (Somnomed; Crows Nest, Australia) MAS appliances, which were titrated over 6-8 weeks. Appliance titration measurements and pre- and posttreatment polysomnograms were assessed. Respiratory disturbance index (RDI), absolute and percentage changes in RDI, non-rapid eye movement (NREM) RDI, rapid eye movement (REM) RDI, supine and nonsupine NREM and REM RDI, and minimal blood-oxygen saturation variables were evaluated. The associations of measurements from 2D and 3D minimal anterior-posterior linear distance and 3D airway variables with MAS treatment response were estimated. Results and Conclusions: Combined effects of baseline total airway volume, body mass index, neck circumference, location of minimal cross sectional area, and OSA severity were associated with treatment response. Patients with higher initial OSA and more superiorly located airway constriction showed enhanced treatment response to MAS therapy. Airway constriction due to maxillofacial disproportions rather than soft tissue obstruction also showed better treatment response. No significant relationships were found in lateral cephalogram measurements.