No abstract

The purpose of this study was to investigate the effect of breathing route on the collapsibility of the pharyngeal airway in patients with obstructive sleep apnea by using computational fluid dynamics technology. MethodsThis study examined Japanese men with obstructive sleep apnea. Computed tomography scans of the nose and pharynx were taken during nasal breathing with closed mouth, nasal breathing with open mouth, and oral breathing while they were awake. Three-dimensional reconstructed stereolithography models and digital unstructured grid models were created and airflow simulations were performed using computational fluid dynamics software. ResultsAirflow velocity was significantly higher during oral breathing than during nasal breathing with open or closed mouth. No significant difference in maximum velocity was noted between nasal breathing with closed and open mouth. However, airflow during nasal breathing with open mouth was slow but rapidly sped up at the lower level of the velopharynx, and then spread and became a disturbed, unsteady stream. In contrast, airflow during nasal breathing with closed mouth gradually sped up at the oropharyngeal level without spreading or disturbance. Negative static pressure during oral breathing was significantly decreased; however, there were no significant differences between nasal breathing with closed or open mouth. ConclusionsComputational fluid dynamics results during nasal and oral breathing revealed that oral breathing is the primary condition leading to pharyngeal airway collapse based on the

ObjectiveNasal obstruction is a common problem in continuous positive airway pressure (CPAP) therapy for obstructive sleep apnea and limits treatment compliance. The purpose of this study is to model the effects of nasal obstruction on airflow parameters under CPAP using computational fluid dynamics (CFD), and to clarify quantitatively the relation between airflow velocity and pressure loss coefficient in subjects with and without nasal obstruction.MethodsWe conducted an observational cross-sectional study of 16 Japanese adult subjects, of whom 9 had nasal obstruction and 7 did not (control group). Three-dimensional reconstructed models of the nasal cavity and nasopharynx with a CPAP mask fitted to the nostrils were created from each subject’s CT scans. The digital models were meshed with tetrahedral cells and stereolithography formats were created. CPAP airflow simulations were conducted using CFD software. Airflow streamlines and velocity contours in the nasal cavities and nasopharynx were compared between groups. Simulation models were confirmed to agree with actual measurements of nasal flow rate and with pressure and flow rate in the CPAP machine.ResultsUnder 10 cmH2O CPAP, average maximum airflow velocity during inspiration was 17.6 ± 5.6 m/s in the nasal obstruction group but only 11.8 ± 1.4 m/s in the control group. The average pressure drop in the nasopharynx relative to inlet static pressure was 2.44 ± 1.41 cmH2O in the nasal obstruction group but only 1.17 ± 0.29 cmH2O in the control group. The nasal obstruction and control groups were clearly separated by a velocity threshold of 13.5 m/s, and pressure loss coefficient threshold of approximately 10.0. In contrast, there was no significant difference in expiratory pressure in the nasopharynx between the groups.ConclusionThis is the first CFD analysis of the effect of nasal obstruction on CPAP treatment. A strong correlation between the inspiratory pressure loss coefficient and maximum airflow velocity was found.

Maxillomandibular advancement surgery is useful for treatment of sleep apnea. However, preoperative analysis and evaluation to facilitate decision-making regarding the direction and distance of maxillomandibular movement has primarily consisted of morphological analysis; physiological function is not evaluated. To improve preoperative prediction, this study used fluid simulation to investigate the characteristics and effects of airway changes associated with maxillomandibular movement. A one-dimensional model with general applicability was thus developed. Actual measurements of flow in patients were used in this fluid simulation, thus achieving an analysis closer to clinical conditions. The simulation results were qualitatively consistent with the actual measurements, which confirmed the usefulness of the simulation. In addition, the results of the one-dimensional model were within the error ranges of the actual measurements. The present results establish a foundation for using accumulating preoperative measurement data for more-precise prediction of postoperative outcomes.

This paper, as Part II of the study on wake-disturbed boundary layer, is aimed at investigation of the effects of free-stream turbulence on wake-induced transition of the boundary layer under a favorable pressure gradient. Hot-wire probe measurements are also made on the wake-disturbed boundary layer to obtain ensemble-averaged shape factor contours on the distance-time diagrams. These data are then used to examine how the favorable pressure gradient and the free-stream turbulence affects time-resolved behaviors of the boundary layer subjected to periodic wakes. In addition, likewise in Part I, the heat transfer data are compared with the transition model proposed by Funazaki (1996) in order to check the capability of the model under the favorable pressure gradient as well as the free-stream turbulence.