Numerical simulation of airflow fields in two typical nasal structures of empty nose syndrome: a computational fluid dynamics study.

The pathogenesis of empty nose syndrome (ENS) has not been elucidated so far. Though postulated, there remains a lack of experimental evidence about the roles of nasal aerodynamics on the development of ENS. To investigate the nasal aerodynamic features of ENS andto explore the role of aerodynamic changes on the pathogenesis of ENS. Seven sinonasal models were numerically constructed, based on the high resolution computed tomography images of seven healthy male adults. Bilateral radical inferior/middle turbinectomy were numerically performed to mimic the typical nasal structures of ENS-inferior turbinate (ENS-IT) and ENS-middle turbinate (ENS-MT). A steady laminar model was applied in calculation. Velocity, pressure, streamlines, air flux and wall shear stress were numerically investigated. Each parameter of normal structures was compared with those of the corresponding pathological models of ENS-IT and ENS-MT, respectively. ENS-MT: Streamlines, air flux distribution, and wall shear stress distribution were generally similar to those of the normal structures; nasal resistances decreased. Velocities decreased locally, while increased around the sphenopalatine ganglion by 0.20 ± 0.17 m/s and 0.22 ± 0.10 m/s during inspiration and expiration, respectively. ENS-IT: Streamlines were less organized with new vortexes shown near the bottom wall. The airflow rates passing through the nasal olfactory area decreased by 26.27% ± 8.68% and 13.18% ± 7.59% during inspiration and expiration, respectively. Wall shear stresses, nasal resistances and local velocities all decreased. Our CFD simulation study suggests that the changes in nasal aerodynamics may play an essential role in the pathogenesis of ENS. An increased velocity around the sphenopalatine ganglion in the ENS-MT models could be responsible for headache in patients with ENS-MT. However, these results need to be validated in further studies with a larger sample size and more complicated calculating models.