Computational fluid dynamics simulation of upper airway flow and soft tissue deformation in a mouth-breathing patient: a case study

Abstract

Objective

The primary objective of this study was to establish a three-dimensional fluid-structure interaction (FSI) model to compare the biomechanical effects of three distinct breathing modes—nasal (NA), oronasal (ONA), and oral (OA) ventilation—on upper airway dynamics in mouth-breathing patients. A key aim was to investigate a potential biomechanical link between mode-specific hydrodynamic forces and the soft tissue remodeling associated with adenoidal facies.

Methods

Mouth-breathing volunteer was selected, and cone-beam computed tomography (CBCT) and magnetic resonance imaging (MRI) images were collected. Subsequently,3D (Three-Dimensional) finite element models of the upper airway, oral airway, soft palate, and tongue body were reconstructed. Numerical simulations of the flow fields of the three different breathing modes, named nasal ventilation, oral nose ventilation, and oral ventilation, were performed using finite element analysis software.

Results

The simulations revealed distinct, mode-specific biomechanical environments. Oral breathing (OA) generated significantly higher airflow velocity and stronger negative pressure in the palatopharyngeal and epiglottic regions compared to nasal and oronasal breathing. This led to pronounced posterior-inferior displacement of the tongue and increased deformation of the soft palate. In contrast, nasal breathing produced turbulence confined primarily to the nasal cavity. The model successfully captured the transition to turbulent flow in the pharynx during oral breathing.

Conclusion

This preliminary simulation suggests that, within the constraints of our model, oral breathing is associated with increased airflow velocity, higher negative pressure in the palatopharyngeal and epiglottic regions, and a greater posterior displacement of the tongue. These factors may collectively increase the susceptibility to pharyngeal airway collapse.

Introduction

Mouth-breathing habit is a condition in which the patient is forced to breathe partially or completely through the mouth for long periods due to abnormalities in the nasal passages, such as rhinitis, sinusitis, enlarged turbinates, Adenoid hypertrophy, or adenoids [1, 2].

Chinese epidemiological surveys showed that the prevalence of mouth-breathing among children aged 2–12 years is 5.3% [3]. Not only can mouth-breathing be harmful to the patient’s dental, craniometrical, and even general growth and development, but the resulting facial deformities may even cause severe psychological and psychiatric disorders [2, 4, 5]. The resultant facial deformities may even cause severe psychological and psychiatric disorders.

Doran Haran et al. showed that mouth-breathing children had narrow upper dental arches, high arched palatal covers, less arch width in the molar region, less palatal surface area and volume than normal breathing children, and that recession and downward rotation of the jaw was more pronounced in mouth-breathing patients [6]. Kulnis et al. concluded that snoring children have a paracentral rotation of the mandibular body, mandibular retraction, and lower hyoid position compared to normal children [7]. However, most current studies on the mechanism of misshapen formation in mouth-breathing patients have been conducted on factors such as mechanical and muscular function. As a part of the airway wall, the oral and maxillofacial region must be affected by respiratory airflow. The effects on the growth and development of the oral maxillofacial region and its mechanisms have not been clarified.

Recently, due to the complexity of the vivo flow field and the high cost of in vitro studies, computational fluid dynamics and fluid-solid coupling methods have been gradually applied to the medical field for fluid studies of the upper respiratory tract. More accurate models of the upper respiratory tract were first developed by Martonen, et al. [8]. Soo-Jin Sun et al. developed a finite element model of fluid-solid coupling in patients with obstructive sleep apnea syndrome (OASHS) using CT images to analyze the interaction between the upper airway fluid and the soft palate [9]. Chang et al. modeled the upper airway before and after orthognathic surgery based on the patient’s CT images and simulated and compared them using Computational Fluid Dynamics (CFD) and fluid-structure interaction (FSI), respectively [10]. The results of this study demonstrated that the fluid dynamics approach can simulate the natural flow field environment in the human body more realistically.

Therefore, this study aimed to develop a three-dimensional finite element model of the upper airway, oral cavity, and surrounding soft tissues in a mouth-breathing patient. Using a fluid-structure interaction (FSI) approach, we sought to simulate the flow field distribution and soft tissue deformation under three breathing conditions: nasal, oral, and oronasal ventilation. The goal was to investigate the potential role of hydrodynamic factors in the development of mouth-breathing-associated malformations.

Methods

Modeling process

We selected one adult male mouth-breathing patient with normal height and weight, a history of nasopharyngeal obstruction, no diastema between bilateral first molars, and no orthodontic or orthognathic treatment history. Imaging studies revealed a deviated nasal septum in this patient. The mouth breathing identification criteria in this study were: a long-term history of open-mouth breathing, sleep snoring, the presence of lip incompetence, and a positive fogging mirror test. Patient underwent CBCT and MRI after sitting for 30 min at room temperature, maintaining his natural mouth-breathing state, calm breathing, reduce swallowing, and not talking. The imaging data were obtained with the patient in the standard supine position, consistent with common clinical and sleep-study protocols. We recognize that this position induces gravitational posterior displacement of the soft tissues (tongue, soft palate), leading to a more restricted pharyngeal airway [11, 12]. This is directly relevant to our study objective, which is to simulate airflow dynamics under conditions that predispose to nocturnal obstruction or sleep-disordered breathing. The CBCT and MRI data was stored in DICOM format.

Cone-beam computed tomography (KaVo 3D eXam, Kavo Dental, USA) data of the volunteer was imported into Mimics (Mimics 21.0, Materialise, Belgium) (Fig. 1). Paranasal sinus images were removed, thresholds were set, and upper airway, oral airway, and soft palate models were reconstructed and saved as STL files (Fig. 2). The CBCT and MRI data of the volunteer was imported into a 3D-slicer (Version 4.10.2, open source, http://www.slicer.org/), and the CBCT and MRI images were fused using the transformed module. The MRI images were used to locate the contours of both sides of the tongue body, and the line between the chin spine and the hyoid bone was used as the inferior border of the tongue, build a model of the tongue body and save it as an STL file (Fig. 3). The STL file of the tongue was imported into Geomagic Studio 14.0 software (Geomagic, USA) to generate the solid model of each part, and the model was saved in its native format (Fig. 2). The developed surface model was imported into Unigraphics NX 8.5 software (Siemens PLM Software). The combined relationships between the various entities were established, and the processed model was exported and saved in the x-t format.

Fig. 1.

Screenshot of CBCT modelling of the upper airway, oral airway, and soft palate sain a patient with mouth-breathing

Fig. 2.

a 3D reconstructed model of the upper respiratory tract, oral airway, soft palate, and tongue b Surface format of the 3D reconstructed model

Fig. 3.

Tongue body model created by the fusion of CBCT and MRI in 3D-slicer

The x-t format file was imported into Ansys workbench (Ansys, America) for meshing the model (Fig. 4). The number of grids produced in the CFD domains was 2,630,699 (Fig. 4). The divisions were unstructured tetrahedral structures.

Fig. 4.

a Mesh model of the upper airway and oral airway. b Tetrahedral mesh at the airway constriction. c Sagittal schematic diagram of the upper airway model

Numerical methodology and simulation setup

Physiological basis and boundary condition derivation

Respiratory boundary conditions were defined based on standard adult physiological parameters from the literature [13, 14]: a tidal volume (Q) of 600 mL, a respiratory cycle period of 3 s, and an inspiratory-to-expiratory (I: E) time ratio of 3:2. This resulted in an inspiratory phase duration (t_insp) of 1.8 s and an expiratory phase duration (t_exp) of 1.2 s. The cross-sectional area (S) at the target coronal plane in the pharyngeal cavity was obtained from the 3D reconstruction (Fig. 5). The characteristic airflow velocity (*v*) for each phase was then calculated using the equation of continuity:

Fig. 5.

Area of the pharyngeal cavity

where t_phase denotes t_insp for inspiration and t_exp for expiration.

Table 1; Fig. 6 showed different boundary condition settings for the breathing modes.

Table 1.

Boundary Setup: Inlet and Outlet

	Nasal breathing		Nasal and oral breathing		Oral breathing
Phase	Expiratory	Inspiratory	Expiratory	Inspiratory	Expiratory	Inspiratory
Inlet	Pharynx	Nostril	Pharynx	Nostril oral	Pharynx	Oral
Outlet	Nostril	Pharynx	Nostril oral	Pharynx	Oral	Pharynx

Fig. 6.

Boundary condition settings

Governing Equations, Rheology, and turbulence modeling

The airflow was modeled as an incompressible, Newtonian fluid with a constant density (ρ) of 1.225 kg/m³ and a dynamic viscosity (µ) of 1.789 × 10⁻⁵ Pa·s. The Reynolds-Averaged Navier-Stokes (RANS) equations were solved. Turbulence was accounted for using the realizable k-ε model with standard wall functions [10], a validated choice for simulating internal flows at physiological Reynolds numbers in upper airway studies. The material assignment was shown in Table 2 [15, 16].

Table 2.

Material Assignment

	Air	Soft palate	Tongue
Density	1.225kg/m3	920kg/m3	—
Elastic modulus(E)	—	7540Pa	5961Pa
Dynamic viscosity	1.7894N/S·m2	—	—
Poisson's ratio(v)	—	0.49	0.49

Solver Settings, boundary Conditions, and convergence criteria

Numerical simulations were performed using the pressure-based solver in ANSYS Fluent (v2020 R2). The SIMPLE algorithm was used for pressure-velocity coupling. While initial stabilization used a first-order upwind scheme, the final computations for data collection were executed with a first-order upwind scheme for momentum and turbulence equations to enhance accuracy.

•  Inspiration: The nasal/oral inlets were set as pressure inlets with a gauge pressure of 0 Pa (atmospheric reference). The pharyngeal outlet was a velocity outlet with the magnitude given by v_insp.

•  Expiration: The pharyngeal inlet was set as a velocity inlet (v_exp), and the nasal/oral outlets were pressure outlets at 0 Pa (gauge).

All airway walls (including the tongue) were defined as stationary, no-slip walls. The interface between the airway lumen and the soft palate was designated as the fluid-structure interaction (FSI) coupling surface.

Simulations were considered converged when a dual criterion was met for at least 50 consecutive time steps: (1) the normalized residuals for continuity and momentum equations dropped below 1 × 10⁻³, and (2) the relative mass flow rate imbalance between all inlets and outlets was less than 0.5% [15]. A maximum of 1000 iterations per time step was allowed.

Fluid-Structure interaction and Post-Processing

A one-way fluid-structure interaction analysis was conducted in Ansys Workbench. The transient pressure field from the CFD solution on the soft palate surface was mapped as a load onto a corresponding structural model to compute tissue deformation. This process was repeated for three distinct ventilation scenarios in mouth-breathing patients: exclusive nasal, oronasal, and exclusive oral ventilation.

Post-processing and visualization were performed using CFD-POST. To quantitatively analyze the flow field, several representative cross-sections (coronal and sagittal) were defined along the upper airway (Fig. 7). The area-weighted average values of static pressure and velocity magnitude were calculated for each cross-section to systematically evaluate spatial variations in airflow dynamics [17]. The final displacement field and pressure distribution of the soft palate and tongue body were visualized as contour plots within the ANSYS post-processor. Utilizing ANSYS Mechanical’s node solution data and probe tools, quantitative displacement and pressure values were extracted from the target regions. This enabled direct comparison of displacement variations and pressure differentials across different simulated breathing scenarios.

Fig. 7.

Selection of different cross-sections

Results

Changes in upper airway pressure of different breathing patterns

Pressure changes of exhaled phase

The ranges of pressure variation in the upper airway during nasal ventilation, oronasal ventilation, and oral ventilation were − 16.41 to 16.98 Pa, -8.34 to 8.94 Pa, and − 234.9 to 173.8 Pa, respectively (Fig. 8). The pressure on both sides of the nasal floor was inconsistent during nasal and oral-nasal ventilation. The pressure in right inferior nasal passage was more significant than that in the left (Fig. 9).

Fig. 8.

a Nose ventilation expiratory phase pressure cloud (b) Oral-nasal ventilation Exhalation phase pressure cloud (c) Oral ventilation Exhalation phase pressure cloud

Fig. 9.

Display of pressure contour map comparison between the left and right nasal cavities located on the same coronal plane.

During the exhalation phase, the pressure values for all breathing modes were greater than one standard atmosphere. However, during nasal and oronasal ventilation, no substantial changes in overall pressure were observed. In contrast, during oral ventilation, the airflow at the anterior tooth row pressure increased more significantly than in the other respiratory modes. Additionally, pressure in the intrinsic oral cavity was considerably greater during mouth ventilation than in the two ways of breathing (Fig. 10).

Fig. 10.

Trends in exhalation phase pressure for different breathing patterns

Pressure changes in the Suction phase

The pressure variation range in the inspiratory phase during nasal ventilation was − 74.43 to -6.52pa during nasal ventilation. The pressure variation range was − 37.05pa to 0.35pa; during oral ventilation, the pressure variation range was − 178.4pa to 1.58pa. In the mode of nasal and oral-nasal ventilation, the pressure distribution in the airway were uniform (Fig. 11).

Fig. 11.

a Nasal ventilation inspiratory phase pressure distribution (b) Oral-nasal ventilation Inspiratory phase pressure distribution (c) Oral ventilation Inspiratory phase pressure distribution

Overall, upper airway pressure was negative in the inspiratory phase, with minor pressure changes, during nasal and oral nose ventilation. During mouth ventilation, a significant increase in negative pressure was observed. In contrast, oral ventilation produces significant pressure changes at the junction of the intrinsic oral cavity and the oral vestibule (Fig. 12).

Fig. 12.

Trends in inspiratory phase pressure for different respiratory modes

Characteristics of flow filed distribution of different breathing patterns

Flow field of the expiratory phase

During nasal ventilation, the airflow primarily entered through the nasal cavity with minimal entry into the oral cavity. Turbulence predominantly occurred at the soft palate, upper nasal tract, and middle nasal cavity. During oral ventilation, there was less airflow into nasal cavity, and turbulence occured mainly at the soft palate and epiglottis (Fig. 13).

Fig. 13.

a Nasal ventilation upper airway flow velocity vector (b) Oral-nasal ventilation upper airway flow velocity vector (c) Oral ventilation upper airway flow velocity vector

During oral ventilation, the velocity of airflow in the airway was higher than in the other two respiratory modes, and the variation in flow rate was more pronounced (Fig. 14).

Fig. 14.

Variation in the exhalation phase flow rate for different respiratory modes

Flow field of the inspiratory phase

In all three respiratory modes, the maximum velocity of airflow occurred at the narrowing of the airway. The maximum flow rate of oral ventilation was greater than those of the other two breathing modes (Fig. 15).

Fig. 15.

a Nasal ventilation inspiratory phase flow velocity vector (b)Oral-nasal ventilation inspiratory phase flow velocity vector (c) Oral ventilation inspiratory phase flow velocity vector

During nasal and oral-nasal ventilation, there was no significant change in flow velocity in the inherent oral cavity and oropharynx. Under all three breathing modes, the velocity of airflow from the lower edge of the epiglottis to the pharyngeal cavity showed a trend of decrease (Fig. 16).

Fig. 16.

Variation of flow rate in each cross-section with different respiration methods

During nasal and oronasal inspiration, laminar flow predominates posterior to the soft palate (Fig. 17). In contrast, oral inspiration generates a prominent vortex region with flow separation behind the soft palate, accompanied by a substantial turbulence-induced recirculation zone in the oropharynx (Fig. 18).

Fig. 17.

a Laminar flow behind the soft palate during nasal inspiration (b) Laminar flow behind the soft palate during oral-nose inhalation (c) Turbulence behind the soft palate during oral inhalation

Fig. 18.

a Velocity vector diagram of the oropharynx during nasal inhalation (b) Velocity vector diagram of the oropharynx during oral-nasal inhalation (c) Velocity vector diagram of the oropharynx during oral inhalation

Displacement and pressure of soft tissues in different breathing patterns

Displacement and pressure of the soft palate

The maximum displacement of the soft palate occurred at the uvula during both the expiratory and inspiratory phases (Fig. 19). During nasal and oral nose ventilation, the soft palate displacement was slight; underlying oral ventilation, the overall removal of the uvula, and the pressure on the surface of the uvula were more significant (Table 3).

Fig. 19.

Deformation variations during expiratory phase under different breathing patterns

Table 3.

Displacement and pressure on the soft palate

	Expiratory phase		Aspirated phase
	Deformation(mm)	Pressure (pa)	Deformation (mm)	Pressure (pa)
Nasal ventilation	0.2635	10	0.6389	25
Nose and oral ventilation	0.1399	5	0.4004	13
Oral ventilation	3.544	135	1.4052	62

In both the expiratory and inspiratory phases of oral ventilation, the uvula was pushed toward back and up by the airflow close to the posterior pharyngeal wall. During nasal ventilation, the uvula was pushed toward forward and downward by the airflow close to the posterior tongue.

Displacement and pressure of the tongue

Under the three ventilation conditions, the maximum displacement of the tongue occurred in the posterior tongue (Fig. 20). During oral breathing, the downward and backward displacements of the tongue were significant, and the downward pressure on the tongue was also greater compared to the other ventilation modes (Table 4).

Fig. 20.

Deformation variations during expiratory phase under different breathing patterns

Table 4.

Displacement and pressure on the tongue

	Expiratory phase		Inspiratory phase
	Deformation(mm)	Pressure(pa)	Deformation(mm)	Pressure(pa)
Nv	1.13	15	2.78	38
ONV	0.5	7	1.29	20
OV	12.8	199	6.96	108

Discussion

Discussion on modeling methods and material assignment

In medical finite element analysis, fluid-solid coupling methods were mainly uesd for studying OASHS mechanism [9]. Previous studies have shown that the difficulty of tongue modeling lies in the lack of contrast between soft tissues. Soft tissue boundaries are challenging to visualize on CBCT and MRI images. The difficult that construct the tongue was distinguish the lateral surface of the tongue and the mandibular mucosa [18]. In this experiment, artifacts from tongue movements could not be avoided due to a long time during the MRI taken by volunteers. Therefore, this study used the transformed module in the 3D slicer software to fuse the CBCT and MRI images. This alignment is based on the line connecting the mental spine and the superior border of the hyoid body.

The tongue body was manually tracked layer by layer to locate the lateral edge of the tongue on both sides. The line between the chin spine and the superior border of the hyoid body was used as the demarcation line between the tongue body and the floor of mouth. The 3D model was saved and imported into the editing module, which can simulate the tongue morphology more realistically and accurately [19].

For the soft palate and tongue, we employed a linear elastic material model. We recognize that biological tissues exhibit viscoelastic and hyperelastic behaviors [20]. While a hyperelastic model (e.g., Ogden or Mooney-Rivlin) would more accurately capture large-strain behavior, our primary focus was on comparing relative differences in displacement and stress patterns between breathing modes under comparable loading [21]. The linear elastic model provides a consistent and computationally efficient basis for this comparative analysis, and the observed trends are expected to hold qualitatively.

In experiments, the mechanical properties of biological soft tissues (such as the Young’s modulus of the soft palate) exhibit inter-individual variability, often requiring the use of approximate values in computational models. Our choice of E = 7.5 kPa falls within the commonly cited range of 1–20 kPa [22]. To assess the influence of this parameter, a sensitivity analysis was performed where E was varied by ± 50%. The results indicated that while the absolute magnitude of displacement scaled proportionally with the change in stiffness, the relative differences between the three breathing patterns and the spatial distribution of deformation remained qualitatively consistent. This suggests that our primary comparative conclusions regarding the impact of ventilation mode are robust to reasonable variations in this material property. The magnitude and distribution of soft palate displacement computed in our one-way FSI analysis are consistent with the findings of prior dedicated fluid-structure interaction studies of the upper airway, lending credibility to our coupled physics approach [22].

Fluid-solid coupling studies

Preceding studies mainly focused on the study of airway hydrodynamics and ignored airflow’s interaction with the surrounding soft tissues. Based on previous research findings, airflow within the maxillary sinus is minimal during respiration, while the sphenoid, ethmoid, and frontal sinuses exhibit virtually no airflow [23]. Consequently, the influence of airflow within the paranasal sinuses on the flow field within the upper respiratory tract can be considered negligible. Therefore, to simplify the model for computational analysis, this experiment excluded the maxillary, sphenoid, and frontal sinuses during upper airway modeling, retaining only the nasal cavity and pharyngeal structures.

Turbulence in the oral airway determined the overall airflow pattern throughout the airway [24]. In a previous study, Zhao et al. conducted a fluid dynamic analysis of the upper airway and reported that increased total nasal resistance during nasal congestion exacerbates negative pressure in the palatopharyngeal cavity, potentially aggravating pharyngeal collapse in OSA patients [25]. Chen et al. reported that Class II malocclusion patients showed elevated nasal resistance and greater negative inspiratory pharyngeal pressure than Class I/III counterparts, potentially due to mandibular retraction [26]. The significant negative airway pressure may be related to their mandibular retraction. The low pressure in the velopharyngeal had an adsorptive effect on the surrounding soft tissues. This study indicated that oral ventilation establishes a distinct and more aerodynamically stressful environment in the pharynx. This mode generated the highest levels of turbulence and the most pronounced pharyngeal negative pressure among all tested patterns. The pharyngeal pressure drop predicted by our model under resting breathing conditions falls within the characteristic range (approximately 5–20 Pa) reported in comprehensive reviews of upper airway CFD studies [27].

From a fluid mechanics perspective, the elevated flow velocity observed during oral breathing (as per the continuity equation, Q = v·A) directly leads to a local reduction in static pressure within constricted regions, as described by the Bernoulli principle. This explains the significantly greater (more negative) pharyngeal pressure measured. Concurrently, the high-velocity jet entering from the oral cavity inherently possesses greater shear and momentum, which destabilizes the flow upon interaction with the complex pharyngeal geometry, leading to enhanced turbulence generation and flow separation. The complex flow patterns predicted in our simulations—such as turbulent flow formation in the oropharynx and pharyngeal negative pressure generated during inspiration—exhibit high consistency with flow structures visualized in detailed experimental studies using particle image velocimetry (PIV) in similar geometries (e.g., research by Omid Bafkar et al., published in the Journal of Biomechanics, 2021) [28].

Anatomically, the pharyngeal airway is a collapsible conduit surrounded by compliant soft tissues (tongue, lateral walls). The strong, fluctuating negative pressure acts as an external collapsing force on the airway walls. According to the Starling resistor model of collapsible tubes, when the intraluminal pressure falls sufficiently below the surrounding tissue pressure, the tube becomes susceptible to narrowing or collapse.

Therefore, during oral ventilation, the fluid-dynamically driven low pressure reduces the supportive distending force on the pharyngeal tissues, while the associated turbulent fluctuations may induce tissue vibration (manifesting as snoring) and exacerbate flow instability. This synergistic effect necessitates greater respiratory effort to maintain patency, providing a mechanistic explanation for the subjective sensation of dyspnea and the elevated risk of flow-limited breathing or obstructive events associated with this ventilation pattern.

Mouth breathing aggravates the adenoid face

The normal growth and development of the craniofacial region depend on appropriate functional stimuli. Classical theories, such as Moss’s Functional Matrix Hypothesis and the homeostasis theory of dentofacial development, posit that physiological nasal breathing provides the optimal mechanical environment: the tongue naturally rests against the hard palate, jointly guiding the normal descent of the palate and dental arch development [2, 29]. In contrast, chronic mouth breathing—often a compensatory response to upper airway obstruction (e.g., adenoid hypertrophy)—disrupts this equilibrium. Clinically, it is closely associated with the “adenoid facies,” characterized by a high-arched palate, dental crowding, and mandibular retraction [30, 31]. However, a key question remains unclear: through what specific biomechanical mechanisms does mouth breathing alter the local microenvironment to drive these morphological changes?

This study, for the first time, provides a quantitative, parallel comparison of the aerodynamic environment and mechanical loading on tissues within the airway under nasal, oronasal, and exclusive oral breathing patterns using computational fluid dynamics (CFD) simulations. We found that exclusive oral breathing establishes a fundamentally different flow regime. Its average airflow velocity (20.18 m/s) was more than three times higher than during nasal (6.01 m/s) or oronasal (4.25 m/s) breathing. This high-velocity jet directly leads to a significant increase in turbulence intensity and kinetic energy in the pharynx. Correspondingly, the altered flow field exerts distinctly different mechanical loads on the tongue, a key functional matrix. During the expiratory phase, the average downward pressure on the tongue during oral breathing (199 Pa) was an order of magnitude greater than during nasal (15 Pa) or oronasal (7 Pa) breathing. A similarly significant difference was observed during the inspiratory phase (108 Pa vs. 38 Pa/20 Pa).

These quantitative findings can be interpreted through fundamental fluid-structure interaction principles. First, according to Bernoulli’s principle, the high flow velocity during oral breathing causes a significant local reduction in static pressure, resulting in a higher transmural pressure difference across the compliant tongue and pharyngeal walls. This sustained load, directed downward and posteriorly, forces the tongue to adapt from its physiological, palate-supporting rest position to a lower, more retruded posture [15], directly disrupting a key determinant of normal dental arch form. Second, the increased turbulence indicates greater flow instability and energy dissipation within the airway. This not only necessitates increased respiratory effort but also subjects the surrounding soft tissues to high-frequency oscillatory and fluctuating shear stresses, potentially affecting neuromuscular tone and tissue remodeling.

Of particular importance, our data suggest a possible “threshold effect.” During oronasal breathing, when oral airflow constituted only approximately 30% of the total, the aerodynamic parameters (flow velocity, pressure) remained closer to those of pure nasal breathing. This implies that preserving even partial nasal patency may be sufficient to mitigate the most adverse aerodynamic loads, offering a mechanistic explanation for the spectrum of clinical severity.

Integrating our results with previous longitudinal evidence [32], we propose a mechanistic pathway hypothesis: chronic, exclusive oral breathing leads to altered upper airway aerodynamics (high velocity, low pressure, high turbulence), which in turn causes sustained abnormal mechanical loading on the tongue and pharyngeal soft tissues. To adapt to this load, the posture and function of the tongue change adaptively, progressively disrupting the homeostasis of dentofacial growth, and may ultimately contribute to the development of a high-angle, mandibular retrusion facial morphology [33, 34].

This model emphasizes that the pathogenicity of mouth breathing may be dose-dependent, related to the completeness of nasal obstruction. In clinical practice, our findings suggest that any intervention capable of restoring even partial nasal airflow (e.g., adenoidectomy, allergy management) may help normalize the local biomechanical environment, thereby mitigating adverse growth patterns. Future research should correlate these simulated loading conditions with long-term imaging data to validate this proposed causal chain.

There are still some limitations in this study. First, it lacks a formal mesh convergence analysis, which could further refine the quantitative accuracy of absolute flow and pressure values. Second, the findings have not been validated against experimental or clinical measurements. Future work will address these by performing a systematic mesh independence study and correlating simulation results with in-vivo or benchtop data to strengthen the model’s predictive credibility [35].

Conclusion

In summary, this case study provides preliminary insights into hydrodynamic mechanisms that may contribute to the pathogenesis of adenoidal facies in mouth-breathing patients. Under physiological conditions, the tongue maintains an optimal position within the palatal space, with its morphology coordinated with the palate. When oropharyngeal obstruction necessitates a shift to oral breathing, the airflow dynamics are markedly altered.

From a fluid dynamics perspective, oral breathing creates a significantly more turbulent and higher-energy airflow environment. During oral exhalation, the maximum flow velocity in the oropharynx was approximately 1600% higher than during nasal breathing and about 375% higher than during oronasal breathing. Oral ventilation also produced the most pronounced negative pressure in the pharynx. The time-averaged minimum static pressure in the palatopharyngeal region reached approximately − 234.9 Pa during OA, compared to -74.43 Pa during NA and − 37.05 Pa during ONA. This represents an increase in suction force magnitude of approximately 227% relative to nasal breathing. The tongue body exhibited a maximum posterior-inferior displacement of 0.13 mm during OA, which was 1200% greater than the maximum displacement observed under NA. Similarly, the maximum deformation of the soft palate increased by approximately 1270% in the OA mode compared to the NA baseline.

In conclusion, compared to nasal or oronasal breathing, oral breathing generates higher airflow velocities, stronger negative pharyngeal pressure, and greater displacement of key airway tissues. These specific biomechanical factors provide a plausible mechanism for the increased airway collapsibility and potential long-term musculoskeletal adaptations—such as those associated with adenoidal facies—linked to chronic mouth breathing.

Abbreviations

E

Young’s modulus (elastic modulus)

ν

Poisson’s ratio

ε

Strain

p, P

Pressure (static)

Δ*p*

Pressure difference/drop

v, U

Flow velocity (vector, magnitude)

Re

Reynolds number

ρ

Density

Q

a tidal volume

*t*

Time

µ

Dynamic viscosity

ω

Angular frequency

max, min

Maximum, minimum value

NA, ONA, OA

Nasal, Oronasal, Oral breathing mode

t_insp

inspiratory phase duration

t_exp

expiratory phase duration

3D

Three-Dimensional

FSI

Fluid-Structure Interaction

CFD

Computational Fluid Dynamics

FEA/FEM

Finite Element Analysis/ Method

CBCT

Cone-Beam Computed Tomography

MRI

Magnetic Resonance Imaging

OSA

Obstructive Sleep Apnea

Authors’ contributions

Conceptualization: Linkun Zhang, Investigation: Wei Ji, Wenting Xie, Yangyang Cui, Formal analysis: Yaqin XingWriting – original draft: Wei Ji, Zhongfang ZhangWriting – review & editing: Wei Ji, Yuan Luo, Linkun ZhangAll authors read and approved the final manuscript.

Funding

This work was supported by Tianjin Key Laboratory of Oral and Maxillofacial Function Reconstruction [grant numbers 2023KLQN02].

Data availability

All data generated or analyzed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

Approval was obtained from the Institutional Review Board of Nankai University. The procedure used in this study adhere to the tenets of the Declaration of Helsinki. Informed consent for participation was obtained from a single individual participating in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Conflict of interest

All authors certified that they have no affiliations with or involvement in any orgainzation or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript.