Evaluation of Upper Airway Characteristics Using Computational Fluid Dynamics

Introduction: Obstructive sleep apnea (OSA) involves a disruption of nocturnal respiration due to cessation of airflow. The research objectives were to: 1) determine if upper airway characteristics (minimal cross-sectional area and total volume) change significantly between low risk, healthy non-OSA subjects and OSA subjects during inhalation and exhalation using cone-beam computed tomography (CBCT) imaging, and 2) if pressure (minimum, maximum, and pressure range) changes significantly in steady-state k-ω model computational fluid dynamics (CFD) fluid flow simulations. Materials and Methods: CBCT imaging was collected at both end-inhalation and end-exhalation for 16 low-risk healthy adult non-OSA subjects recruited at the UIC College of Dentistry, Department of Orthodontics and compared to existing CBCT data from 7 OSA subjects. The CBCT images were imported into Dolphin Imaging System (Chatsworth, CA), segmented into stereolithography (STL) files of the naso- and oro- pharyngeal spaces for area and volumetric measurements. Subsequently, 5 non-OSA subject and 5 OSA subject STL models met pre-processing criteria for CFD simulations using ANSYS Fluent Meshing (Canonsburg, PA). An unstructured mesh consisting of hexahedral, tetrahedral, and octagonal cells was generated and the Commercial CFD package Fluent (version R2021) was used to solve the standard dual equation turbulence model (k-ω). Both objective and supplemental descriptive measures were obtained from a combination of imaging, simulations, and questionnaires and subsequent statistical analyses were performed using IBM SPSS with both parametric and nonparametric tests to evaluate statistical significance at 0.05 level. Results/Conclusion: Regarding Dolphin area and volumetric assessments, there were statistically significant mean differences in Total Volume, Minimum CSA between non-OSA and OSA groups at inhalation and exhalation, p-values equal (p=0.002, 0.003, 0.004, and 0.007), respectively. There were statistically significant mean differences in volume and min CSA between the inhalation and exhalation for the non-OSA group, (p<0.001 and 0.002), respectively. Lastly, there was a statistically significant mean difference in total volume pre- and post-OA therapy in OSA subjects on inhalation only (p=0.006). With respect to the CFD simulations, limitations of the CFD simulation method used and the outstanding data available that could not be implemented rendered testing and subsequent results to be inconclusive, requiring further investigation. An observed trend, however, was that the OSA subject group for the CFD simulation featured airway models whose point of lowest pressure coincided with the area of maximum constriction. Further studies of increasing power and increasing accuracy of simulation method in regard to the airway are needed to further characterize the upper airway in order to learn more about its function in ideal conditions and in obstructive sleep apnea.