Comparative Analysis of Spreader Grafts and Spreader Flaps on Intranasal Drug Delivery Efficiency to Posterolateral Nasal Wall

Abstract

Background: This study compared the impact of spreader grafts (SG) and spreader flaps (SF) on the transport of intranasal drug delivery to target the posterolateral nasal wall. Method: SG and SF were each performed in sequence on two cadaveric specimens after soft tissue elevation technique. Computed tomography scans were acquired following each procedure to generate anatomic models for computational fluid dynamics simulation of intranasal sprays under the following conditions: inhalation rate (15 and 30 L/min), spray velocity (1, 5, and 10 m/s), spray released location (center, lateral, medial, top, and bottom), head position (upright, tilted-forward, tilted-backward, and supine), and particle diameter (1-100 µm). Percentage of particles deposited on the posterolateral nasal wall were calculated. Results: For Specimen 1, highest posterolateral wall depositions were Pre-Op: left = 74%, right = 74%; SF: left = 53%, right = 22%; SG: left = 60%, right = 61%. For Specimen 2, highest posterolateral wall depositions were Pre-Op: left = 25%, right = 83%; SF: left = 29%, right = 76%; SG: left = 14%, right = 72%. In general, posterolateral wall deposition was higher at 30 L/min inhalation rate and at 1 m/s spray velocity. Conclusions: Drug delivery targeting the posterolateral nasal wall appears to be dependent on many factors. However, midvault nasal reconstruction does not increase drug delivery to the posterolateral nasal wall.

Keywords

computational fluid dynamics intranasal sprays rhinoplasty spreader flap and spreader graft

Introduction

Topical intranasal medications are the first-line treatment for a wide variety of rhinologic conditions ranging from allergic rhinitis to chronic rhinosinusitis.¹ Symptomatic relief requires effective delivery of drug particles to relevant regions of the nasal cavity, namely, the highly vascularized mucosa of the turbinates along the posterolateral nasal wall. There are several methods for administering topical intranasal medications including sprays, nebulized inhalations, and irrigated solutions.² Intranasal sprays are the most commonly used method, and they rely heavily on proper application technique in order to be effective. This is thought to be primarily due to factors including head position, nozzle insertion depth, and patient inhalation rate, on the spatial deposition of drug particles within the nasal cavity.³ Additional factors associated with manufacturing of the intranasal spray also play a role—spray release velocity, spray plume angle, and drug particle size.⁴

For certain patients, anatomic factors such as septal deviation, turbinate hypertrophy, and compromised nasal valve may contribute to altered airflow dynamics associated with nasal obstruction, as well as mucosal conditions such as allergic rhinitis. In these cases, surgical correction of the anatomic obstruction may allow for improved airflow, symptomatic relief, and enhanced delivery of topical medications.⁵ Isolated septoplasty and inferior turbinate reduction are common procedures that can alleviate symptoms from septal deviation and turbinate hypertrophy, but they do not address the internal nasal valve; rather, rhinoplasty with midvault nasal reconstruction is used to address internal nasal valve collapse.

One common technique for correction of internal nasal valve dysfunction involves placement of autologous cartilage or allografts between the dorsal septum and the upper lateral cartilage (ie, spreader grafts) to create a wider internal nasal valve.^6,7 Alternatively, the spreader flap technique involves folding the upper lateral cartilages medially to widen the midvault and provide structural support for the internal nasal valve without the use of a graft.^8-11 Of note, these procedures are commonly associated with dorsal hump reduction.

Recent studies have characterized the impact of nasal midvault reconstruction on nasal airflow patterns using computational methods, but the effects of rhinoplasty for internal nasal valve dysfunction on intranasal drug delivery have not been extensively studied.^12,13 An improved understanding of how midvault reconstruction techniques can impact drug particle deposition patterns post-operatively could inform surgical decision-making to maximize symptom relief for each patient based on individual anatomy. Studying the effects of different techniques for administering intranasal medications on particle deposition patterns could also identify the best characteristics for administering intranasal medications.

In that regard, the present study aims to (1) determine differences in intranasal drug delivery to the posterolateral nasal wall after spreader flap and spreader graft techniques for nasal midvault reconstruction, and (2) determine the best spray administration characteristics that would maximize posterolateral nasal wall drug deposition after each surgical procedure.

Methods

Specimen Preparation

Two unembalmed fresh frozen cadaveric specimens with no known history of head or neck trauma, surgery, or cancer (United Tissue Network, Norman, Oklahoma, USA) were selected for the study. The specimens were prepared using Duke University validated methods for clearing airway debris and rehydrating the nasal mucosa as described by Coan et al.^12,14 In brief, frozen cadaveric specimens were defrosted to 65°F. Nasal passages were irrigated with cool water for 3 minutes to remove debris, then filled with isotonic saline and allowed to equilibrate for 15 minutes to rehydrate the nasal mucosa. The nasal passages were dried with compressed air at 15 L/min for 2 minutes.

In preparation for rhinoplasty, soft tissue elevation was done with marginal and transcollumellar incisions. The soft tissue envelope was elevated to the level of the nasal bones, deep to the superficial musculoaponeurotic system. The nasal soft tissue was re-draped, the skin incisions closed, and a computed tomography (CT) scan obtained to serve as the pre-rhinoplasty (pre-operative) airways. Spreader graft (SG) was then performed using fresh frozen cartilage sheet allografts 2 mm in width (MTF Biologics, Edison, NJ, USA) and approximately 4 mm in height, which were cut to match the length of the midvault from the anterior septal angle to the caudal aspect of the nasal bones. The grafts were placed 1 mm below the dorsal septal plane with two horizontal mattress sutures using 4-0 PDS bilaterally. The upper lateral cartilages were then placed over the grafts and sutured to the septum with simple interrupted sutures. The soft tissues were re-draped, skin incisions closed, and a CT scan obtained. The skin was then reopened, the sutures in the upper lateral cartilage removed, and the cartilage grafts removed. Hemostats were used to fold the medial margins of the upper lateral cartilages inwards to create spreader flaps (SF). The medial edges were secured with two horizontal mattress sutures of 4-0 PDS bilaterally. The soft tissues were re-draped, skin incisions closed, and a CT scan obtained.

Nasal Cavity Reconstruction

CT images obtained before and after each procedure were read into the imaging analysis software Avizo Lite™ 9.5.0 (Thermo Fisher Scientific, Waltham, Massachusetts, USA) for creation of anatomically realistic 3D models of the nasal cavity from nostrils to nasopharynx (Figure 1).^12,15-17 Nasal cavity models were segmented with subsequent manual editing as needed, similar to our previously published models.^16,18,19

Nasal cavity models were imported into the computer-aided design and mesh-generating software package ICEM-CFD™ 19.0 (ANSYS, Canonsburg, Pennsylvania, USA) for defining anatomical regions to track drug deposition and meshing using similar techniques described in our prior publications.^18,20-25 The anterior and posterior portions of the nasal passageways were separated using a plane at the most anterior portion of the inferior turbinate bilaterally. Each unilateral anterior and posterior region was further separated mediolaterally into septal wall and lateral wall, excluding the olfactory cleft (Figure 2A).

Drug Delivery Simulation

To solve the discretized governing equations of particle fluid dynamics, hybrid tetrahedral-prism meshes were generated in the airway using ICEM-CFD™ with approximately 4 million unstructured tetrahedral elements. A four-layer prism-element was created near the airway wall. Mesh quality analysis was performed to ensure that the mesh aspect ratio was adequately smoothed to prevent distorted elements from impacting the accuracy of numerical simulations. Mesh refinement analysis was not performed because the choice of mesh density chosen was consistent with a detailed mesh sensitivity analysis reported by Frank-Ito et al²⁶ The simulation setup mirrored the delivery of intranasal topical drugs using nasal spray devices under 2 inspiratory conditions: gentle inspiration at 15 L/min and deep inspiration at 30 L/min. Simulations of drug particle administration were performed in 2 uncoupled phases: airflow phase and particle transport phase.

To simulate delivery of intranasal sprays at physiologically realistic inspiratory flow of 15 L/min, steady laminar incompressible conditions were imposed in nasal models using the computational fluid dynamics modeling software, Fluent™ 19.0 (ANSYS, Inc., Canonsburg, Pennsylvania, USA). Appropriate “mass-flow-outlet” boundary conditions were specified at the outlet (end of nasopharynx) to target 0.0003 kg/s (15 L/min). Atmospheric conditions were specified with zero gauge pressure at the nostril inlet. The nasal wall was considered stationary with no-slip boundary conditions. For simulations of 30 L/min, the airflow phase involved steady turbulent flow under physiologically realistic inspiratory conditions for delivery of intranasal sprays. Boundary conditions were similar with those described for laminar conditions, except the outlet “mass-flow-outlet” condition was specified to target 0.0006 kg/s (30 L/min).

The intranasal drug transport phase involved simulating particle trajectories in the nasal cavity during inspiration. Dispersed particles exchanged momentum and mass with inhaled airflow. Drug particle trajectories were calculated using the Euler-Lagrange approach via the Lagrangian discrete phase model in Fluent™, assuming unit density, spherical particles with particle density of 1000 kg/m³, and ignoring particle-to-particle interactions.²⁷ Spray plume angle was 68°. Three spray velocities were simulated: 1, 5, and 10 m/s, consistent with the approximate range (1.5-14.7 m/s) observed in commercial products.^28,29 Micron particles from 1 to 100 µm with 1 µm increments were simulated. For each particle size, 3500 monodisperse particles were released into the unilateral nasal cavity for a total of 350,000 particles.

Four different head positions were simulated: upright (head in a neutral upright position), backward (head tilted back at 45°), forward (head tilted forward at 45°), and supine (head in supine position, akin to lying on the back). Simulated particles were released evenly across the spray nozzle orifice from an insertion depth of 15 mm from nostril alar into the airway (Figure 2B). Five spray release positions per head position were simulated (Figure 2B): bottom (aiming inferiorly), top (aiming superiorly), center (aiming toward centrally), lateral (aiming laterally), and medial (aiming medially).

Results

Parameter Combinations for Maximal Deposition

Table 1 shows parameter combinations that produced the top 5 highest depositions on the posterolateral wall (Figure 2A). For Specimen 1 Pre-Op, top 5 depositions on the left posterolateral wall ranged 54% to 74%, and the optimal parameters were 30 L/min inhalation (4 out of 5), 1 m/s spray velocity (5 out of 5), and lateral release position (5 out of 5). Similarly, right posterolateral wall depositions ranged 54% to 74%, and the optimal parameters were 30 L/min inhalation (5 out of 5), 1 m/s spray velocity (5 out of 5), but medial release position (4 out of 5). No head position showed overall dominance in improving drug delivery to the posterolateral nasal wall.

Table 1.

Five maximum posterolateral wall particle deposition for specimens after each procedure: Pre-Op (PRE), spreader flap (SF), and spreader graft (SG), with respective inhalation flow rate, aerosol spray velocity, release position, and head position.

	Left side					Right side
	Deposition (%)	Flow Rate (L/min)	Spray Speed (m/s)	Release Position	Head Position	Deposition (%)	Flow Rate (L/min)	Spray Speed (m/s)	Release Position	Head Position
Specimen 1
Pre-Op	74.4	30	1	Lateral	Back	74.1	30	1	Medial	Supine
	74.3	30	1	Lateral	Upright	73.9	30	1	Medial	Back
	74.3	30	1	Lateral	Supine	73.5	30	1	Medial	Upright
	74.1	30	1	Lateral	Forward	73.1	30	1	Medial	Forward
	53.9	15	1	Lateral	Upright	54.2	30	1	Top	Supine
Spreader Flap	53.2	30	1	Lateral	Supine	21.6	30	1	Bottom	Upright
	53.0	30	1	Lateral	Back	21.6	30	1	Bottom	Forward
	52.3	30	1	Lateral	Upright	21.3	30	1	Bottom	Back
	52.0	30	1	Lateral	Forward	20.9	30	1	Bottom	Supine
	43.7	15	1	Lateral	Supine	9.2	30	1	Medial	Supine
Spreader Graft	59.7	30	1	Top	Back	61.3	30	1	Medial	Supine
	59.5	30	1	Top	Supine	61.0	30	1	Medial	Back
	59.3	30	1	Top	Upright	59.9	30	1	Medial	Upright
	58.5	30	1	Top	Forward	58.6	30	1	Medial	Forward
	46.6	15	1	Top	Back	51.0	30	1	Center	Supine
Specimen 2
Pre-Op	24.5	30	1	Lateral	Forward	83.4	30	1	Bottom	Supine
	24.4	30	1	Lateral	Upright	82.7	30	1	Bottom	Back
	24.1	30	1	Lateral	Back	81.8	30	1	Bottom	Upright
	24.1	30	1	Lateral	Supine	81.5	30	1	Bottom	Forward
	14.1	15	1	Lateral	Supine	71.2	30	1	Medial	Back
Spreader Flap	28.7	30	1	Center	Forward	75.6	30	1	Medial	Back
	28.6	30	1	Center	Upright	75.3	30	1	Medial	Supine
	28.2	30	1	Center	Back	73.8	30	1	Medial	Upright
	28.1	30	1	Center	Supine	71.1	30	1	Medial	Forward
	26.3	30	1	Lateral	Forward	68.5	15	1	Medial	Supine
Spreader Graft	14.3	30	1	Bottom	Forward	72.3	30	1	Medial	Back
	13.9	30	1	Bottom	Upright	72.0	30	1	Medial	Supine
	13.7	30	1	Bottom	Supine	70.3	30	1	Medial	Upright
	13.6	30	1	Bottom	Back	66.9	30	1	Medial	Forward
	11.6	15	1	Lateral	Upright	61.5	15	1	Medial	Back

For Specimen 1 SF, top 5 depositions on the left posterolateral wall ranged 44% to 53%, and optimal parameters were 30 L/min inhalation rate (4 out of 5), 1 m/s particle velocity (5 out of 5) and lateral released position (5 out of 5). Top 5 depositions on the right posterolateral wall were much lower ranging 9% to 22%, occurring mainly at 30 L/min inhalation rate (5 out of 5), 1 m/s particle velocity (5 out of 5), and bottom release position (4 out of 5). No head position showed overall dominance in improving drug delivery to the posterolateral nasal wall.

For Specimen 1 SG, top 5 depositions on the left posterolateral wall ranged 48% to 60%, and optimal parameters were 30 L/min inhalation rate (4 out of 5), 1 m/s particle velocity (5 out of 5), and top release position (4 out of 5). Top 5 depositions on the right posterolateral wall were much lower ranging 9% to 22%, occurring mainly at 30 L/min inhalation rate (5 out of 5), 1 m/s particle velocity (5 out of 5), and medial release position (4 out of 5). No head position showed overall dominance in improving drug delivery to the posterolateral nasal wall.

For Specimen 2 Pre-Op, top 5 depositions on the left posterolateral wall ranged 14% to 25%, and optimal parameters were 30 L/min inhalation rate (4 out of 5), 1 m/s particle velocity (5 out of 5), and lateral release position (5 out of 5). Top 5 depositions on the right posterolateral wall were much higher ranging 71% to 83%, occurring mainly at 30 L/min inhalation rate (5 out of 5), 1 m/s particle velocity (5 out of 5), and bottom release position (4 out of 5). No head position showed overall dominance in improving drug delivery to the posterolateral nasal wall.

For Specimen 2 SF, top 5 depositions on the left posterolateral wall ranged 26% to 29%, and optimal parameters were 30 L/min inhalation rate (5 out of 5), 1 m/s particle velocity (5 out of 5), and center release position (4 out of 5). Top 5 depositions on the right posterolateral wall were much higher ranging 69% to 76%, occurring mainly at 30 L/min inhalation rate (5 out of 5), 1 m/s particle velocity (5 out of 5), and bottom release position (4 out of 5). No head position showed overall dominance in improving drug delivery to the posterolateral nasal wall.

For Specimen 2 SG, top 5 depositions on the left posterolateral wall ranged 12% to 14%, and optimal parameters were 30 L/min inhalation rate (4 out of 5), 1 m/s particle velocity (5 out of 5), and bottom release position (4 out of 5). Top 5 depositions on the right posterolateral wall were much higher ranging 62% to 72%, occurring mainly at 30 L/min inhalation rate (4 out of 5), 1 m/s particle velocity (5 out of 5), and medial release position (5 out of 5). No head position showed overall dominance in improving drug delivery to the posterolateral nasal wall.

Administration Technique Comparisons

Figure 3A reports maximal posterolateral wall depositions for gentle (15 L/min) versus deep (30 L/min) inspiratory rates. In both specimens, for left and right sides and across all procedure types (Pre-Op, SF, and SG), the 30 L/min inspiratory rate produced higher maximum deposition than 15 L/min. Figure 3B reports maximal posterolateral wall depositions across four different head positions: upright (U), tilted-forward (TF), tilted backward (TB), and supine (S). Posterolateral wall depositions did not appear to be sensitive to head position. Figure 3C reports maximal posterolateral wall depositions comparing different intranasal spray release positions: center (C), lateral (L), medial (M), top (T), and bottom (B). There was no single drug particle released position that was uniformly dominant for each side of both specimens. Figure 3D reports the effects of spray particle velocity. Across the board, the slowest spray velocity of 1 m/s was associated with maximal deposition in both specimens on each side, regardless of procedure type.

Particle Size Range

For Specimen 1 Pre-Op left side, the particle size ranges with the highest posterolateral wall deposition were 21 to 30, 31 to 40, and 41 to 50 µm with 100% deposition. For Specimen 1 SF left side, the particle size range with the highest left posterolateral wall depositions was 21 to 30 µm with 96% deposition, and for Specimen 1 SG, the ranges were 11 to 20 µm and 21 to 30 µm with 100% deposition.

For Specimen 1 Pre-Op right side, the particle size ranges with the highest posterolateral wall deposition were 6 to 10, 11 to 20, 21 to 30, 31 to 40, and 41 to 50 µm with 100% deposition. For Specimen 1 SF right side, the particle size range with the highest right posterolateral wall deposition was 11 to 20 µm with 88% deposition, and for Specimen 1 SG, the ranges were 21 to 30 µm and 31 to 40 µm with 100% deposition. The spatial deposition patterns in the nasal cavity for Specimen 1 STE, SF, and SG are presented in Figure 4 for relevant particle size ranges (11-40 µm) and highest posterolateral wall depositions.

For Specimen 2 Pre-Op left side, the particle size range with the highest posterolateral wall deposition was 6 to 10 µm with 100% deposition. For Specimen 2 SF left side, the particle size range with the highest left posterolateral wall depositions was 21 to 30 µm with 100% deposition, and for Specimen 2 SG, the range was 11 to 20 µm with 98% deposition.

For Specimen 2 Pre-Op right side, the particle size ranges with the highest posterolateral wall deposition were 11 to 20, 21 to 30, 31 to 40, 41 to 50, and 51 to 60 µm, and 61 to 70 µm with 100%. For Specimen 2 SF right side, the particle size ranges with the highest right posterolateral wall deposition were 21 to 30 µm and 31 to 40 µm with 100% deposition, and for Specimen 2 SG, the range were 21 to 30 µm and 31 to 40 µm with 100% deposition.

Discussion

Prior work by our laboratory using computational modeling has suggested that SF and SG are associated with decreased nasal resistance and improved airflow. Similarly, a study by Shadfar et al³⁰ demonstrated that SG with or without flare suture combination improved nasal airflow and reduced nasal resistance when compared with Pre-Op, and a study by Brandon et al³¹ found that butterfly grafts and SG result in modest nasal airflow improvement and reduction in nasal resistance. However, to our knowledge, the present study is the first to investigate the interactions between nasal midvault reconstruction and intranasal drug delivery.

Our results, albeit based on a small sample size, suggest that SG and SF do not confer an advantage when targeting the posterolateral wall with intranasal sprays. Rather, in the majority of simulated cases (when varying spray administration parameters), the pre-operative preparation of soft tissue elevation was often associated with improved particle deposition to the posterolateral wall. The implications of this finding is that nasal midvault reconstruction procedures are unlikely to be of assistance in improving drug delivery for treating conditions that may rely on intranasal spray delivery, such as allergic rhinitis or sinusitis.

Of note, both specimens included in this study had middle to posterior nasal septal deviation. Specimen 1 had caudal deflection of the inferior septum to the right and a rightward posterior septal spur. Contrary to expectations, the presence of a rightward septal deviation in Specimen 1 did not create an overt imbalance in intranasal particle deposition between the left and right posterolateral wall for PRE and SG, although the deviation was associated with a twofold higher posterolateral wall deposition on the left in SF. The deviation in Specimen 2 was primarily to the left with a septal spur. Consistent with expectations, there was a gross imbalance between left and right posterolateral wall deposition in Specimen 2 for all procedures: PRE, SF, and SG, with deposition on the right posterolateral wall up to fivefold higher than on the left wall.

Nasal septal deviation has previously been reported to impact intranasal spray penetration toward the posterior nasal cavity.^32-34 According to Frank et al³³ sprayed particle depositions on the posterolateral wall were significantly lower on the deviated nasal side compared to the contralateral side across different head positions. Furthermore, surgical correction of the deviated nasal septum improved spray penetration to posterior regions of the nose on the deviated side, as well as leads to a more evenly posterior region deposition amount between the left and right nasal cavities. The discrepancy between expectations and the findings for Specimen 1 are likely due to the complexities of nasal airflow dynamics, which are thought to be highly dependent on locoregional as well as global nasal anatomy based on prior and ongoing work by our laboratory.

Under the majority of the conditions tested, our results suggest that a spray velocity of 1 m/s maximizes posterolateral wall deposition, and that an inspiratory rate of 30 L/min is superior to 15 L/min. Optimal release position and head position varied by specimen and surgical technique, suggesting a possible dependence on individual specimen anatomy.

Limitations to this preliminary study are primarily due to a small sample size, owing to the extensive computational resources required to perform the necessary simulations. Reconstruction of the different midvault procedures and undoing the sutures also induced post-mortem air trapping, which can mask any changes in airway patency, thus compromising segmentation of the nasal airway if not performed meticulously. Furthermore, static reconstruction of nasal anatomy from CT images does not capture the dynamic behavior of the nasal valve during respiration, which inevitably is an important feature of trans-valvular airflow that was not modeled in this work. Lastly, cadaveric specimens included in this study may not be characteristic of typical patients who undergo midvault reconstruction procedures; moreover, cadavers lack the mucosal thickness and tissue turgor in live humans. While the true effects of SF or SG on posterolateral wall deposition for this population may be mitigated by these limitations, the comparisons performed in this study cannot be done without cadaveric specimens. Future studies evaluating the impact of SF and SG should be in combination with surgical maneuvers that address other anatomic causes of nasal obstruction, such as septal deviation and turbinate hypertrophy.

Conclusion

Our pilot study using computational fluid dynamics modeling suggests that nasal midvault reconstruction does not confer an advantage when targeting intranasal spray medications to the posterolateral wall. However, deep inhalation during spray administration, and lower spray velocity were associated with improved posterolateral wall delivery, which can inform patient education regarding administration technique, as well as intranasal spray design and manufacturing.

Footnotes

Acknowledgements

We would like to thank the American Academy of Facial Plastic and Reconstructive Surgery for the opportunity to present this research as poster presentation at the virtual AAFPRS Annual Meeting in September 2020. In addition, special thanks to ANSYS, ANSYS Global Academic Program, and Dr Paolo Maccarini (Duke University) for support and strategic donation.

Data Availability

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Jeffrey R. Marcus, M.D., receives royalties from Thieme Medical Publishing and royalties for intellectual property for the SmartLock Hybrid MMF from Stryker. The other authors do not have any disclosures.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The Esthetic Surgery Education and Research Foundation (ASERF) provided funding in support of this research. Funds were used for procurement of cadaveric specimens, computer software for computational analysis, computed-tomography scanning, and other direct study costs. The content of this manuscript is the responsibility of the authors and does not necessarily represent the views of ASERF or the National Institutes of Health.

Informed Consent

Due to the nature of the cadaveric study, IRB approval was not required and informed consent not attained.

ORCID iD

Michelle H. Kim

References

Agnihotri

McGrath

Allergic and nonallergic rhinitis. Allergy Asthma Proc. 2019;40(6):376-379.

Harvey

Snidvongs

Kalish

Oakley

Sacks

Corticosteroid nasal irrigations are more effective than simple sprays in a randomized double-blinded placebo-controlled trial for chronic rhinosinusitis after sinus surgery. Int Forum Allergy Rhinol. 2018;8(4):461-470.

Gao

Shen

Mao

Factors influencing drug deposition in the nasal cavity upon delivery via nasal sprays. J Pharm Investig. 2020;50(3):251-259.

Foo

Cheng

Y-S

Donovan

MD.

The influence of spray properties on intranasal deposition. J Aerosol Med. 2007;20(4):495-508.

Friedman

Cekic

Gunel

Functional rhinoplasty. Facial Plast Surg Clinics. 2017;25(2):195-199.

Rudy

Moubayed

Most

SP.

Midvault reconstruction in primary rhinoplasty. Facial Plast Surg. 2017;33(2):133-138.

Sheen

JH.

Spreader graft: a method of reconstructing the roof of the middle nasal vault following rhinoplasty. Plast Reconstr Surg. 1984;73(2):230-237.

Gruber

Park

Newman

Berkowitz

Oneal

The spreader flap in primary rhinoplasty. Plast Reconstr Surg. 2007;119(6):1903-1910.

Oneal

Berkowitz

RL.

Upper lateral cartilage spreader flaps in rhinoplasty. Aesthet Surg J. 1998;18(5):370-371.

10.

Seyhan

Method for middle vault reconstruction in primary rhinoplasty: upper lateral cartilage bending. Plast Reconstr Surg. 1997;100(7):1941-1943.

11.

Byrd

Meade

Gonyon

Jr.

Using the autospreader flap in primary rhinoplasty. Plast Reconstr Surg. 2007;119(6):1897-1902.

12.

Avashia

Martin

Frank-Ito

, et al Computational analyses of physiologic effects after midvault repair techniques in rhinoplasty. FACE. 2023;4(1):22-32.

13.

Brandon

Austin

Fleischman

, et al Comparison of airflow between spreader grafts and butterfly grafts using computational flow dynamics in a cadaveric model. JAMA Facial Plast Surg. 2018;20(3):215-221.

14.

Coan

Neff

Mukundan

Jr Marcus

JR.

Validation of a cadaveric model for comprehensive physiologic and anatomic evaluation of rhinoplastic techniques. Plast Reconstr Surg. 2009;124(6):2107-2117.

15.

Martin

Marcus

Frank-Ito

DO.

Analysis of nasal air conditioning in subjects with unilateral cleft lip nasal deformity. Respir Physiol Neurobiol. 2021;291:103694.

16.

Popper

Martin

Shah

Sicard

Hodges

Frank-Ito

DO.

Intranasal spray characteristics for best drug delivery in patients with chronic rhinosinusitis. Laryngoscope. 2023;133(5):1036-1043.

17.

Tillis

Shah

Martin

Allori

Marcus

Frank-Ito

DO.

A systematic analysis of surgical interventions for the airway in the mature unilateral cleft lip nasal deformity: a single case study. Int J Comput Assist Radiol Surg. 2022;17(1):41-53.

18.

Choi

Jang

Ellison

Frank-Ito

DO.

Characterizing airflow profile in the postoperative maxillary sinus by using computational fluid dynamics modeling: a pilot study. Am J Rhinol Allergy. 2016;30(1):29-36.

19.

Sicard

Shah

Frank-Ito

DO.

Analyses on the influence of normal nasal morphological variations on odorant transport to the olfactory cleft. Inhal Toxicol. 2022;34(11-12):350-358.

20.

Frank-Ito

Kimbell

Borojeni

AAT

Garcia

GJM

Rhee

JS.

A hierarchical stepwise approach to evaluate nasal patency after virtual surgery for nasal airway obstruction. Clin Biomech. 2019;61:172-180.

21.

Frank-Ito

Sajisevi

Solares

Jang

DW.

Modeling alterations in sinonasal physiology after skull base surgery. Am J Rhinol Allergy. 2015;29(2):145-150.

22.

Patki

Frank-Ito

DO.

Characterizing human nasal airflow physiologic variables by nasal index. Respir Physiol Neurobiol. 2016;232:66-74.

23.

Frank-Ito

Carpenter

Cheng

, et al Computational analysis of the mature unilateral cleft lip nasal deformity on nasal patency. Plast Reconstr Surg Glob Open. 2019;7(5):e2244.

24.

Ramprasad

Frank-Ito

DO.

A computational analysis of nasal vestibule morphologic variabilities on nasal function. J Biomech. 2016;49(3):450-457.

25.

At Borojeni

Frank-Ito

Kimbell

Rhee

Garcia

GJM.

Creation of an idealized nasopharynx geometry for accurate computational fluid dynamics simulations of nasal airflow in patient-specific models lacking the nasopharynx anatomy. Int J Numer Methods Biomed Eng. 2017;33(5):e2825.

26.

Frank-Ito

Wofford

Schroeter

Kimbell

JS.

Influence of mesh density on airflow and particle deposition in sinonasal airway modeling. J Aerosol Med Pulm Drug Deliv. 2016;29(1):46-56.

27.

Dong

Shang

Tian

Inthavong

Detailed deposition analysis of inertial and diffusive particles in a rat nasal passage. Inhal Toxicol. 2018;30(1):29-39.

28.

Cheng

Holmes

Gao

, et al Characterization of nasal spray pumps and deposition pattern in a replica of the human nasal airway. J Aerosol Med. 2001;14(2):267-280.

29.

Kimbell

Segal

Asgharian

, et al Characterization of deposition from nasal spray devices using a computational fluid dynamics model of the human nasal passages. J Aerosol Med. 2007;20(1):59-74.

30.

Shadfar

Shockley

Fleischman

, et al Characterization of postoperative changes in nasal airflow using a cadaveric computational fluid dynamics model: supporting the internal nasal valve. JAMA Facial Plast Surg. 2014;16(5):319-327.

31.

Brandon

Stepp

Basu

, et al Nasal airflow changes with bioabsorbable implant, butterfly, and spreader grafts. Laryngoscope. 2020;130(12):E817-E823.

32.

Frank

Kimbell

Pawar

Rhee

JS.

Effects of anatomy and particle size on nasal sprays and nebulizers. Otolaryngol Head Neck Surg. 2012;146(2):313-319.

33.

Frank

Kimbell

Cannon

Pawar

Rhee

JS.

Deviated nasal septum hinders intranasal sprays: a computer simulation study. Rhinology. 2012;50(3):311-318.

34.

Frank

Kimbell

Cannon

Rhee

JS.

Computed intranasal spray penetration: comparisons before and after nasal surgery. Int Forum Allergy Rhinol. 2012;3(1):48-55.