Validated reconstructions of geometries of nasal cavities from CT scans

An adjoint-based approach for the surgical correction of nasal septal deviations

Deviations of the septal wall are widespread anatomic anomalies of the human nose; they vary significantly in shape and location, and often cause the obstruction of the nasal airways. When severe, septal deviations need to be surgically corrected by ear-nose-throat (ENT) specialists. Septoplasty, however, has a low success rate, owing to the lack of suitable standardized clinical tools for assessing type and severity of obstructions, and for surgery planning. Moreover, the restoration of a perfectly straight septal wall is often impossible and possibly unnecessary. This paper introduces a procedure, based on advanced patient-specific Computational Fluid Dynamics (CFD) simulations, to support ENT surgeons in septoplasty planning. The method hinges upon the theory of adjoint-based optimization, and minimizes a cost function that indirectly accounts for viscous losses. A sensitivity map is computed on the mucosal wall to provide the surgeon with a simple quantification of how much tissue removal at each location would contribute to easing the obstruction. The optimization procedure is applied to three representative nasal anatomies, reconstructed from CT scans of patients affected by complex septal deviations. The computed sensitivity consistently identifies all the anomalies correctly. Virtual surgery, i.e. morphing of the anatomies according to the computed sensitivity, confirms that the characteristics of the nasal airflow improve significantly after small anatomy changes derived from adjoint-based optimization.

Computational Fluid Dynamics Modeling of Nasal Obstruction and Associations with Patient-Reported Outcomes

BACKGROUND

Nasal obstruction is a common problem, with significant impact on quality of life. Accurate diagnosis may be challenging because of the complex and dynamic nature of the involved anatomy. Computational fluid dynamics modeling has the ability to identify specific anatomical defects, allowing for a targeted surgical approach. The goal of the current study is to better understand nasal obstruction as it pertains to disease-specific quality of life by way of a novel computational fluid dynamics model of nasal airflow.

METHODS

Fifty-three patients with nasal obstruction underwent computational fluid dynamics modeling based on computed tomographic imaging. Nasal resistance was compared to demographic data and baseline subjective nasal patency based on Nasal Obstructive Symptom Evaluation scores.

RESULTS

Mean Nasal Obstructive Symptom Evaluation score among all patients was 72.6. Nasal Obstructive Symptom Evaluation score demonstrated a significant association with nasal resistance in patients with static obstruction (p = 0.03). There was a positive correlation between Nasal Obstructive Symptom Evaluation score and nasal resistance in patients with static bilateral nasal obstruction (R2 = 0.32) and poor correlation in patients with dynamic bilateral obstruction caused by nasal valve collapse (R2 = 0.02). Patients with moderate and severe bilateral symptoms had significantly higher nasal resistance compared to those with unilateral symptoms (p = 0.048).

CONCLUSIONS

Nasal obstruction is a multifactorial condition in most patients. This study shows correlation between simulated nasal resistance and Nasal Obstructive Symptom Evaluation score in a select group of patients. There is currently no standardized diagnostic algorithm or gold standard objective measure of nasal airflow; however, computational fluid dynamics may better inform treatment planning and surgical techniques on an individual basis.

CLINICAL QUESTION/LEVEL OF EVIDENCE

Risk, V.

Three-dimensional structure of the basal lamella of the middle turbinate

The middle turbinate’s basal lamella (3BL) is a variable landmark which needs to be understood in endoscopic transnasal skull base surgery. It comprises an anterior frontal and a posterior horizontal part and appears in its simplest depiction to be “L”-shaped, when viewed laterally. In this study we analyzed its 3D morphology and variations focusing on a precise and systematic description of the anatomy. CBCTs of 25 adults, 19 cadavers and 6 skulls (total: 100 sides) were investigated with the 3DSlicer software, creating 3D models of the 3BL. We introduced a novel geometrical classification of the 3BL’s shape, based on segments. We analyzed their parameters and relationship to neighboring structures. When viewed laterally, there was no consistent “L”-shaped appearance of the 3BL, as it is frequently quoted. A classification of 9 segment types was used to describe the 3BL. The 3BLs had in average of 2.95 ± 0.70 segments (median: 3), the most frequent was the horizontal plate (23.05% of all segments), next a concave/convex plate (22.71%), then a sigma plate (22.37%). Further types were rare. We identified a horizontal plate in 68% of all lateral views whilst 32% of the 3BLs were vertical. A sigma–concave/convex–horizontal trisegmental 3BL was the most common phenotype (27%). Globally, the sigma–concave/convex pattern was present in 42%. The 3BL adhered the ethmoidal bulla in 87%. The segmenting method is eligible to describe the 3BL’s sophisticated morphology.

Modelling nasal airflow coefficients: an insight into the nature of airflow

BACKGROUND

There has been considerable discussion and conflicting views regarding the presence of laminar or turbulent flow within the nose. The aim of this study was to investigate how the modelling of variable flow coefficients can assist in the evalua- tion of the characteristics of flow in the resistive segments of the nose.

METHODOLOGY

A comparison was made between the flow coefficient for the nasal valve, obtained from a mathematical model, and resistive flow components such as a Venturi meter and orifice tube. Also, a variable loss coefficient was formulated for the whole (unilateral) nose which, by utilising the intersection of the laminar and turbulent asymptotes, provided an estimation for the critical Reynolds number (Rcrit).

RESULTS

The results show that the flow resistance of the nasal valve is considerably greater than that for both a Venturi meter and an orifice tube implying turbulent or turbulent-like flow for much of nasal inspiration. Regarding the loss coefficient for the whole (unilateral) nose, normal respiration flowrates are displaced well away from the laminar asymptote. The critical Reynolds number was estimated to be 450.

CONCLUSIONS

A novel method of determining the flow characteristics of the nose, particularly the critical Reynolds number, is presented. The analysis indicates a higher degree of turbulence than is assumed from a simple traditional calculation using a hy- draulic diameter and flow through straight tubes. There are implications for computational fluid dynamics (CFD) modelling where either the entire nasal airflow is assumed to be laminar or a low turbulence model implemented.

Rhinomanometry Versus Computational Fluid Dynamics: Correlated, but Different Techniques

BACKGROUND

Past studies reported a low correlation between rhinomanometry and computational fluid dynamics (CFD), but the source of the discrepancy was unclear. Low correlation or lack of correlation has also been reported between subjective and objective measures of nasal patency.

OBJECTIVE

This study investigates (1) the correlation and agreement between nasal resistance derived from CFD (R_CFD) and rhinomanometry (R_RMN), and (2) the correlation between objective and subjective measures of nasal patency.

METHODS

Twenty-five patients with nasal obstruction underwent anterior rhinomanometry before and after mucosal decongestion with oxymetazoline. Subjective nasal patency was assessed with a 0-10 visual analog scale (VAS). CFD simulations were performed based on computed tomography scans obtained after mucosal decongestion. To validate the CFD methods, nasal resistance was measured in vitro (R_EXPERIMENT) by performing pressure-flow experiments in anatomically accurate plastic nasal replicas from 6 individuals.

RESULTS

Mucosal decongestion was associated with a reduction in bilateral nasal resistance (0.34 ± 0.23 Pa.s/ml to 0.19 ± 0.24 Pa.s/ml, p = 0.003) and improved sensation of nasal airflow (bilateral VAS decreased from 5.2 ± 1.9 to 2.6 ± 1.9, p < 0.001). A statistically significant correlation was found between VAS in the most obstructed cavity and unilateral airflow before and after mucosal decongestion (r = -0.42, p = 0.003). Excellent correlation was found between R_CFD and R_EXPERIMENT (r = 0.96, p < 0.001) with good agreement between the numerical and in vitro values (R_CFD/R_EXPERIMENT = 0.93 ± 0.08). A weak correlation was found between R_CFD and R_RMN (r = 0.41, p = 0.003) with CFD underpredicting nasal resistance derived from rhinomanometry (R_CFD/R_RMN = 0.65 ± 0.63). A stronger correlation was found when unilateral airflow at a pressure drop of 75 Pa was used to compare CFD with rhinomanometry (r = 0.76, p < 0.001).

CONCLUSION

CFD and rhinomanometry are moderately correlated, but CFD underpredicts nasal resistance measured in vivo due in part to the assumption of rigid nasal walls. Our results confirm previous reports that subjective nasal patency correlates better with unilateral than with bilateral measurements and in the context of an intervention.

Narrowed Posterior Nasal Airway Limits Efficacy of Anterior Septoplasty

Background: Predicting symptomatic relief after septoplasty has been difficult. Minimal cross-sectional area (mCSA) measured by acoustic rhinometry and airflow resistance (R) measured by rhinomanometry have been used to select surgical candidates with mixed success. An important assumption is that mCSA and resistance are tightly coupled, but studies have reported weak or no correlation. Recently, we proposed the Bernoulli Obstruction Theory as an explanation, where tight coupling between mCSA and R is only predicted below a critical mCSA (A_crit). Methods: The nasal airway and septum of 10 healthy subjects were reconstructed from computed tomography scans. Simulated anterior septal deviations of increasing severity were created. Computational fluid dynamics simulations were performed to quantify mCSA, resistance, and flow in the healthy septum model and four simulated septal deviation models for each subject (total of 50 models). Results: A tighter coupling between mCSA and resistance was found below A_crit, estimated to be 0.20 cm² (a very severe deviation). Above A_crit, enlarging the mCSA had a smaller effect in patients with narrower cross-sectional area in the postvalve region (CSA_PV). Conclusions: Two patterns of flow increase are expected with septoplasty. Below A_crit, enlarging mCSA predictably increases flow. Above A_crit, the effect size of increasing mCSA depends on CSA_PV. Unrecognized small CSA_PV may explain persistent sensation of nasal obstruction after septoplasty. Our data suggest that inferior turbinate reduction ipsilateral to a septal deviation may amplify airflow benefits after septoplasty in patients with a narrow CSA_PV.

Characterization of the Airflow within an Average Geometry of the Healthy Human Nasal Cavity

This study's objective was the generation of a standardized geometry of the healthy nasal cavity. An average geometry of the healthy nasal cavity was generated using a statistical shape model based on 25 symptom-free subjects. Airflow within the average geometry and these geometries was calculated using fluid simulations. Integral measures of the nasal resistance, wall shear stresses (WSS) and velocities were calculated as well as cross-sectional areas (CSA). Furthermore, individual WSS and static pressure distributions were mapped onto the average geometry. The average geometry featured an overall more regular shape that resulted in less resistance, reduced WSS and velocities compared to the median of the 25 geometries. Spatial distributions of WSS and pressure of the average geometry agreed well compared to the average distributions of all individual geometries. The minimal CSA of the average geometry was larger than the median of all individual geometries (83.4 vs. 74.7 mm²). The airflow observed within the average geometry of the healthy nasal cavity did not equal the average airflow of the individual geometries. While differences observed for integral measures were notable, the calculated values for the average geometry lay within the distributions of the individual parameters. Spatially resolved parameters differed less prominently.