Creation of an idealized nasopharynx geometry for accurate computational fluid dynamics simulations of nasal airflow in patient‐specific models lacking the nasopharynx anatomy

Abstract

Virtual surgery planning based on computational fluid dynamics (CFD) simulations has the potential to improve surgical outcomes for nasal airway obstruction (NAO) patients, but the benefits of virtual surgery planning must outweigh the risks of radiation exposure. Cone beam computed tomography (CBCT) scans represent an attractive imaging modality for virtual surgery planning due to lower costs and lower radiation exposures compared with conventional CT scans. However, to minimize the radiation exposure, the CBCT sinusitis protocol sometimes images only the nasal cavity, excluding the nasopharynx. The goal of this study was to develop an idealized nasopharynx geometry for accurate representation of outlet boundary conditions when the nasopharynx geometry is unavailable. Anatomically-accurate models of the nasopharynx created from thirty CT scans were intersected with planes rotated at different angles to obtain an average geometry. Cross sections of the idealized nasopharynx were approximated as ellipses with cross-sectional areas and aspect ratios equal to the average in the actual patient-specific models. CFD simulations were performed to investigate whether nasal airflow patterns were affected when the CT-based nasopharynx was replaced by the idealized nasopharynx in 10 NAO patients. Despite the simple form of the idealized geometry, all biophysical variables (nasal resistance, airflow rate, and heat fluxes) were very similar in the idealized vs. patient-specific models. The results confirmed the expectation that the nasopharynx geometry has a minimal effect in the nasal airflow patterns during inspiration. The idealized nasopharynx geometry will be useful in future CFD studies of nasal airflow based on medical images that exclude the nasopharynx.

Sensitivity of nasal airflow variables computed via computational fluid dynamics to the computed tomography segmentation threshold

“…Computer simulations of nasal airflow have many applications, such as quantifying the regional doses of nasal sprays [ 26 – 29 ] and virtual surgery planning for patients with nasal airway obstruction [ 6 , 7 , 9 – 12 , 30 , 31 ]. Many studies have evaluated how CFD-derived airflow variables are affected by numerical methods, including inlet boundary conditions [ 32 ], outlet boundary conditions [ 33 , 34 ], flow regime (laminar or turbulent) [ 24 ], and assumption of transient or steady-state flow [ 35 ]. One underlying assumption of CFD models that is rarely considered in the literature is the accuracy of the anatomic model [ 13 ].…”

Estimates of nasal airflow at the nasal cycle mid-point improve the correlation between objective and subjective measures of nasal patency

“…The pre-surgery outlet pressure was such that the transnasal pressure drop (nostrils to choana) was the same in the pre- and post-surgery models. This is important because the soft palate can have different configurations in the pre- and post-surgery scans leading to different nasopharynx resistances in the pre- and post-surgery models, which can confound the results when the same outlet pressure is imposed on all models (Borojeni et al, 2016; Kim et al, 2013). In other words, inhalation rates were different in the pre- and post-surgery models, but the pressure drop (nostrils to choana) was the same in each subject.…”

Relationship between degree of obstruction and airflow limitation in subglottic stenosis

“…The steady‐state inhalation rate was computed as two times the minute volume under the assumption that inspiration and expiration have the same duration (Table SI, available online). For the adult model, a steady‐state inhalation rate of 15 L/minute was used, which is the most common inhalation rate used for adults in the computational fluid dynamics (CFD) literature . Airway resistance was defined as , where is the pressure drop from nostrils to carina and is the inhalation rate.…”

“…For the adult model, a steady-state inhalation rate of 15 L/minute was used, which is the most common inhalation rate used for adults in the computational fluid dynamics (CFD) literature. 16 Airway resistance was defined as R5DP=Q, where DP is the pressure drop from nostrils to carina and Q is the inhalation rate. The same outlet pressure was used in all simulated SGS models (40%-90% obstruction) of each patient such that the desired inhalation rate (Table SI, available online) was obtained only in the original healthy model (0% obstruction), but the pressure drop DP was the same in all simulated SGS models of a given subject.…”