The respiratory structures are closely related to dentofacial growth and development.1 For example, mouth breathing caused by narrow airways may lead to “adenoid facies.”2 The pharynx is one of the major anatomical structures in the upper airway.1 It is divided into several segments, including the nasopharynx, velopharynx, oropharynx, and hypopharynx.1

Anterior open bite (AOB) can occur as a result of various etiologies, including excessive vertical growth, oral habits, and crowding.3 It shows a higher prevalence in growing children than in adults because the frequency of AOB caused by oral habits, such as thumb sucking and tongue thrust, is high in growing children.4 Although self-correction is possible by discontinuing these habits,5 most patients require orthodontic treatment.

Orthodontic treatment of AOB is considered difficult because of the relapsing tendency of vertical problems and the tongue-thrust habit.6 However, since AOB causes various problems, it is also categorized as a malocclusion, indicating the need for orthodontic treatment. In addition to esthetic problems, AOB can cause functional problems such as pronunciation difficulties and reduced masticatory performance.7 Moreover, individuals with AOB may have a different oral structure from those with normal occlusion,8,9 which may change the pharyngeal airway space (PAS). Therefore, evaluation of the PAS in patients with AOB is essential.

Although a previous study10 evaluated the correlation between AOB and PAS, it used two-dimensional (2D) images. However, 2D measurement of the PAS is limited because it visualizes three-dimensional (3D) structures in two dimensions, and because 2D images do not provide information regarding the volume or cross-sectional area. Thus, 3D cone-beam computed tomography (CBCT) is appropriate for analyzing the PAS.11 To the best of our knowledge, this study is the first attempt to analyze the PAS in patients with AOB using 3D imaging.

The purpose of this study was to three-dimensionally evaluate the PAS of patients with AOB by using CBCT and to compare the findings with those obtained in age- and sex-matched individuals with normal occlusion.

Participants

This retrospective study was conducted on individuals aged over 20 years who underwent CBCT for other purposes such as third molar impaction or oral disease at the Pusan National University Dental Hospital from 2015 to 2020. The study protocol was approved by the Institutional Review Board of Pusan National University Dental Hospital (PNUDH-202016).

The open bite group (OBG, n = 25) consisted of patients with an anterior overbite of –3 mm or less, while the control group (n = 25) consisted of age- and sex-matched individuals with an anterior overbite of 1–3 mm, Angle Class I malocclusion (1° ≤ point A-nasion-point B angle (ANB) ≤ 4°), and a normodivergent profile (22° ≤ Frankfort–mandibular plane angle ≤ 28°) (Table 1). The exclusion criteria were as follows: (1) previous airway disease or surgery; (2) maxillofacial syndromes; (3) obstructive sleep apnea; (4) a history of temporomandibular joint disorder; (5) previous orthodontic treatment; (6) recent onset of AOB; (7) Angle Class III malocclusion (ANB < 1°); (8) abnormal head posture (craniocervical angle in the second vertebra, < 90° or > 110°)12; and (9) high body mass index (BMI) (> 30 kg/m²).13

Table 1 . Descriptive statistics.

Parameters	Open bite group (n = 25)	Control group (n = 25)	P value
Age (yr)	23.8 ± 2.4	22.9 ± 2.6	0.843*
Male	13	12	0.768†
Overbite (mm)	–4.6 ± 1.2	1.9 ± 0.8	< 0.001*
ANB (°)	3.0 ± 1.9	2.5 ± 1.2	0.252*
FMA (°)	27.6 ± 5.1	25.1 ± 2.7	0.393*
Angle classification
Class I	17	25
Class II	8	0
Class III	0	0

Values are presented as mean ± standard deviation or number only..

ANB, Angle of nasion-A point line to nasion-B point line; FMA, angle of the Frankfort line to the mandibular plane; NS, not significant..

*Independent t test..

^†Chi-square test..

To detect a mean difference of 2.0 cm³ in the nasopharyngeal volume (NPV) with a standard deviation of 1.6 (obtained from a previous pilot study), a β error of 0.20, and an α error of 0.05, 25 participants were required for each group.14

CBCT data

CBCT (DCT pro; Vatech, Seoul, Korea) was performed using the protocol of the image center under the same conditions for participants of both groups. The participants were instructed to bite with maximum intercuspation in an upright position, not to swallow, and not to move the tongue while undergoing CBCT. The CBCT settings were as follows: field of view, 20 × 19 cm; scanning time, 24 seconds; tube current, 4 mA; tube voltage, 90 kVp; and voxel size, 0.3 mm. CBCT data were reconstructed using InVivo5.4 software (Anatomage, San Jose, CA, USA).

PAS measurements

The 3D images were then re-oriented. The Frankfort horizontal (FH) plane was used as the horizontal reference plane. The plane vertical to the FH plane and crossing the nasion and basion was used as the vertical reference plane (VRP). The reference plane and landmarks were determined on the basis of previous studies (Figure 1 and Table 2).15 The sagittal slice of the VRP was chosen to allow the software to evaluate the airway. The threshold range of air was set between –1,000 and –587 Hounsfield units.16

Figure 1. Landmarks and reference planes. See Table 2 for the definitions of each landmark.

Table 2 . Landmarks and descriptions.

Landmarks	Description
Ba	Basion
S	Sella
So	Midpoint in the Ba-S line
PNS	Posterior nasal spine
UT	Tip of the uvula
ET	Tip of the epiglottis
CV4	Most anterior inferior point of body of fourth cervical vertebra
HY	The most anterosuperior point of the hyoid bone
Frankfort horizontal plane	Plane constructed on both sides of Porion and right Orbitale
PL1	Plane passing through So and PNS
PL2	Plane parallel to FH plane and passing through PNS
PL3	Plane parallel to FH plane and passing through UT
PL4	Plane parallel to FH plane and passing through ET
PL5	Plane parallel to FH plane and passing through CV4
PAL	Shortest distance from HY to PL2

The PAS was segmented into four parts and the volume of each part was measured. The upper limit of the NPV was the plane (PL1) connecting the posterior nasal spine (PNS) at the midpoint (So) of the sella and the basion. The lower limit of the NPV was the plane (PL2) parallel to the FH plane while passing through the PNS. The lower limit of the velopharyngeal volume (VPV) was the plane (PL3) that passed through the tip of the uvula (UT) parallel to the FH plane. The lower limit of the oropharyngeal volume (OPV) was the plane (PL4) parallel to the FH plane while passing through the tip of the epiglottis (ET). The lower limit of the hypopharyngeal volume (HPV) was the plane (PL5) parallel to the FH plane while passing through the most anteroinferior point of the body of the fourth cervical vertebra (CV4). The total airway volume (TAV), which was the volume of all four segments, was also measured. The pharyngeal airway length (PAL) was the vertical distance from PL2 to the most anterosuperior point of the hyoid bone (HY). In addition, the area, anteroposterior length (APL), and transverse width (TW) of the part showing minimal constriction in the PAS were measured (Figure 2).

Figure 2. Measurements of the pharyngeal airway space. A, Anteroposterior length and transverse width of the part showing minimal constriction. B, A three-dimensional airway model with total volume and the part showing minimal constriction.

Statistical analysis

All measurements were performed by a single investigator with more than 5 years of experience in using InVivo software. Intra-examiner reliability was measured using the intraclass correlation coefficient (ICC). Re-measurements were conducted after 1 month for all participants. Levene’s and Shapiro–Wilk tests were used to determine homogeneity and normality. An independent t test was conducted to verify significant differences between the two groups. Pearson’s correlation analysis was used to evaluate the correlation between changes in the PAS score and the amount of anterior overbite. SPSS version 19.0 (IBM Corp., Armonk, NY, USA) was used for all of the statistical analyses.

All ICCs were greater than 0.882 (Table 3). The OBG showed narrower volumes for all PAS segments, but only the NPV, HPV, and TAV of the two groups were significantly different (Table 4). All measurements of the part with minimal constriction were lower in the OBG, but only the area and TW showed significant differences. A statistically longer PAL was observed in the OBG. Pearson’s correlation analysis revealed no correlation between the amount of anterior overbite and changes in the PAS (Table 5).

Table 3 . Intraclass correlation coefficients for intra-examiner reliability.

Variables	Open bite group	Control group
Nasopharyngeal volume	0.921	0.911
Velopharyngeal volume	0.919	0.923
Oropharyngeal volume	0.932	0.882
Hypopharyngeal volume	0.927	0.915
Total airway volume	0.912	0.910
Area of most constriction part	0.934	0.923
APL of most constriction part	0.913	0.918
TW of most constriction part	0.902	0.892
Pharyngeal airway length	0.954	0.962

APL, anteroposterior length; TW, transverse width..

Table 4 . Comparison of the pharyngeal airway space.

Measures	Open bite group	Control group	P value
Nasopharyngeal volume (cm3)	8.47 ± 1.48	10.34 ± 1.86	0.041*
Velopharyngeal volume (cm3)	7.42 ± 2.26	9.40 ± 2.53	0.439
Oropharyngeal volume (cm3)	5.42 ± 2.06	6.21 ± 2.50	0.357
Hypopharyngeal volume (cm3)	5.52 ± 1.70	6.94 ± 2.48	0.045*
Total airway volume (cm3)	26.83 ± 6.06	32.79 ± 9.79	0.017*
Area of most constriction part (mm2)	165.00 ± 41.51	230.60 ± 57.08	0.039*
APL of most constriction part (mm)	9.94 ± 3.73	11.61 ± 2.96	0.122
TW of most constriction part (mm)	24.95 ± 5.12	29.72 ± 8.57	0.031*
Pharyngeal airway length (mm)	62.32 ± 5.17	58.21 ± 4.84	0.016*

Values are presented as mean ± standard deviation..

APL, anteroposterior length; TW, transverse width..

*P < 0.05..

Table 5 . Pearson’s correlation analysis between the amount of anterior overbite and the measured variables in the pharyngeal airway space.

Variables	r	P value
Nasopharyngeal volume	0.189	0.189
Velopharyngeal volume	0.180	0.212
Oropharyngeal volume	0.222	0.122
Hypopharyngeal volume	0.266	0.062
Total airway volume	0.250	0.080
Area of most constriction part	0.221	0.123
APL of most constriction part	0.213	0.138
TW of most constriction part	0.214	0.136
Pharyngeal airway length	–0.234	0.101

APL, anteroposterior length; TW, transverse width; r, Pearson correlation coefficient..

The present study demonstrated smaller volumes in all the pharyngeal segments in the OBG. We speculated that individuals with AOB had a different oral structure from those with normal occlusion,8,9 which may have changed the PAS. Laganà et al.8 reported that patients with AOB had a high and narrow palate, which is related to the tongue-thrust habit. In addition, the tongue positions of patients with AOB may have changed. Since patients with AOB often show a hyperdivergent profile,9 the mandible rotates clockwise, and the base of their tongues moves to the posteroinferior side. This tongue position may narrow the PAS. However, statistical significance was only observed for NPV, HPV, and TAV. This may be owing to the fact that VPV and OPV are affected by changes in soft tissues such as the uvula and epiglottis.

A significantly longer PAL was observed in the OBG. The PAL response is considered to compensate for the narrow airway.17,18 Thus, the position of the hyoid bone may reflect the efficiency of breathing. If breathing is difficult, the hyoid bone moves down to increase breathing capacity. This was consistent with the results of a previous study10 using 2D analysis. The part showing minimal constriction was also significantly narrower in the OBG. The amount of anterior overbite and changes in PAS were not correlated. Thus, the PAS did not depend on the amount of overbite but was related to the presence of AOB. Additionally, we assumed that individual variations existed within each group.

Since PAS is affected by head posture,12 sex,19 BMI,13 and age,15 the control group was selected by matching age and sex. Participants with a high BMI or abnormal head posture were excluded. Many studies have compared the differences in PAS in relation to the anteroposterior skeletal pattern, but their results are controversial. Some studies20,21 reported no differences in PAS in relation to the anteroposterior skeletal pattern, while another study15 reported that Class III malocclusion was associated with a wider PAS than Class II malocclusion. Therefore, we excluded patients with Class III malocclusion to minimize the effects of the anteroposterior skeletal pattern. The difference in PAS in relation to the vertical skeletal pattern is also a topic of controversy. One study20 reported no differences in PAS, whereas another study22 reported a difference.

At present, there is no consensus regarding the criteria for PAS segmentation.23 This study used the segmentation criteria proposed by Claudino et al.15. The nasal cavity was excluded from the measurement because it contains a variety of anatomical structures that cannot be clearly segmented.

Because the PAS has a 3D and irregular shape, it cannot be accurately measured using 2D linear measurements. In a previous study,24 3D CBCT was shown to have higher accuracy than 2D lateral cephalograms in PAS evaluation. Moreover, the InVivo5 program used in this study reduced measurement errors because it automatically measured the volume and the part showing minimal constriction only if the upper and lower limits of PAS were provided with a threshold.25 Nevertheless, there remain concerns regarding the measurement of airways using CBCT.26

A thorough understanding of the growth of the craniofacial area is crucial for orthodontists. Orthodontists must understand the growth mechanisms in patients with AOB to provide complete care to these patients.

This study had some limitations. First, consistent control of the position of soft tissues, such as the tongue, over the prolonged CBCT scan times was challenging. Second, this study only involved anatomic evaluations; however, dynamic evaluations, such as functional assessment of airflow, are also necessary. Finally, the sample size of this study was small. A larger sample size is required to obtain more reliable results. Thus, further studies on the association between AOB correction and PAS are warranted.

The limited evidence obtained in this study suggests that patients with AOB have a narrower PAS and smaller part showing minimal constriction. However, no correlation was observed between the amount of anterior overbite and the changes in the PAS.

This work was supported by a 2-year research grant of Pusan National University.

Conceptualization: SHK. Data curation: YIK. Formal analysis: YIK. Funding acquisition: SSK. Methodology: SHK. Project administration: SBP. Visualization: SBP. Writing–original draft: SSK. Writing–review & editing: SHK.

No potential conflict of interest relevant to this article was reported.