Discussion
Two accurate nasal cavity and paranasal sinus models were created, and numerical simulation of nasal ventilation was performed. In our research, in the simulation that was performed using the nasal cavity model without the sinuses, the magnitude of the relationship was correct, but the simulation value tended to be lower than the actual measurement, although a similar tendency was observed.15 By including the paranasal sinuses, the simulation was expected to approach the measured value because it would be closer to the actual shape; however, this was not the case. This discrepancy can be attributed to the fact that the airflow in the model shown in Figure 4 hardly entered the paranasal sinuses.
There are two possible explanations for the observation that Model 1 had a higher nasal resistance value than did model 2. 1) In Model 1, the cross-sectional area became smaller than that of Model 2 as it approached the choana; thus, the flow velocity increased from the continuity equation. As the flow velocity increased, the pressure decreased based on Bernoulli’s principle.
2) The difference in the cross-sectional areas of the nasopharynx compared with the choana was larger in Model 1 than in Model 2. Thus, an energy loss occurred because of the reverse pressure gradient according to the rapid expansion of the cross-sectional area; it became a pressure loss, and the pressure drop became even larger (Fig. 6).
Ventilation into the sinuses remains uncertain, but it is unlikely that there is no ventilation. In our study, Model 1 exhibited ventilation in the maxillary and ethmoidal sinuses exclusively. Kumar et al. reported the air flow in the periods pre- and post-endoscopic sinus surgery using numerical simulation.16 However, the 3D model reported by them in pre-surgery did not connect the nasal cavity with the frontal sinus and maxillary sinus. Therefore, it was too difficult to create an accurate 3D model. The natural ostium connecting the sinuses and the nasal cavity may not be accurately created. To date, it has been reported that the nasal sinus model is used to examine the correlation between the nasal air permeability measurement and the numerical simulation, and, although they are similar, they do not completely match.14 Radulesco et al. reported the measured and simulated nasal resistance before septoplasty using 22 nasal cavity models. The perceptions of the patients and the measured and simulation nasal resistance exhibited strong correlations. However, the measured and simulation nasal resistance were poorly correlated.17 Tretiakow et al. investigated the workflow for creating a 3D model for accurate CFD.18We think that it is necessary to create an accurate nasal sinus model for more accurate simulation.
Our results suggest that the length of the inferior turbinate and the cross-sectional area of the nasopharynx and the choana affect the nasal resistance value. Hariri et al. reported that inferior turbinate weight loss reduced nasal resistance in 3 of 5 models, whereas it remained unchanged in the remaining two models. In the two cases without change, the nasal resistance was affected by factors other than the inferior turbinate.19 The relationship between the length of the inferior turbinate and the cross-sectional area encompassing the nasopharynx, as in this example, may also affect the nasal resistance value. However, the sample size used in our investigation was 2 volunteers, which was too small to conclude on the effect of nasal resistance in this context. A larger sample size is needed for further investigation.