## Introduction

The purpose of this article was to evaluate the pharyngeal space volume, and the size and shape of the mandible and the hyoid bone, as well as their relationships, in patients with different facial types and skeletal classes. Furthermore, we estimated the volume of the pharyngeal space with a formula using only linear measurements.

## Methods

A total of 161 i-CAT Next Generation (Imaging Sciences International, Hatfield, Pa) cone-beam computed tomography images (80 men, 81 women; ages, 21-58 years; mean age, 27 years) were retrospectively studied. Skeletal class and facial type were determined for each patient from multiplanar reconstructions using the NemoCeph software (Nemotec, Madrid, Spain). Linear and angular measurements were performed using 3D imaging software (version 3.4.3; Carestream Health, Rochester, NY), and volumetric analysis of the pharyngeal space was carried out with ITK-SNAP (version 2.4.0; Cognitica, Philadelphia, Pa) segmentation software. For the statistics, analysis of variance and the Tukey test with a significance level of 0.05, Pearson correlation, and linear regression were used.

## Results

The pharyngeal space volume, when correlated with mandible and hyoid bone linear and angular measurements, showed significant correlations with skeletal class or facial type. The linear regression performed to estimate the volume of the pharyngeal space showed an *R *of 0.92 and an adjusted *R *^{2 }of 0.8362.

## Conclusions

There were significant correlations between pharyngeal space volume, and the mandible and hyoid bone measurements, suggesting that the stomatognathic system should be evaluated in an integral and nonindividualized way. Furthermore, it was possible to develop a linear regression model, resulting in a useful formula for estimating the volume of the pharyngeal space.

## Highlights

- •
A linear regression model and formula were developed to estimate pharyngeal volume.

- •
Pharyngeal space, mandible, and hyoid bone were compared.

- •
Patients with different facial types and skeletal classes showed correlations.

- •
Stomatognathic system should be evaluated in an integral, nonindividualized way.

Craniofacial growth and occlusion are influenced, among other things, by the respiratory function. An impaired nasal respiratory function is associated with airway inadequacy that can result in the habit of mouth breathing. This change in breathing pattern leads to lowering of the mandible and the tongue, and an extended head posture. Changes in normal airway function during the active facial growth period can have a profound influence on facial development by the time a patient comes for orthodontic treatment.

Combined orthodontic and orthognathic surgical treatment has become a common modality for the correction of facial deformities. An important aspect of orthognathic surgery is the effect of skeletal movements in the surrounding structures. Maxillomandibular advancement leads to anterior movements of the soft palate, base of the tongue, hyoid bone, and anterior pharyngeal tissues, resulting in increases in the volumes of the nasopharynx, oropharynx, and hypopharynx, and therefore increasing the posterior airway space. Mandibular setback surgery can cause relative narrowing of the pharyngeal airway and a significant posterior movement of the hyoid bone.

The hyoid bone is connected to the pharynx, mandible, and cranium by muscles and ligaments. The hyoid bone and its connecting muscles are also part of the oropharyngeal complex. Without the hyoid bone, our facility for maintaining an airway, swallowing, preventing regurgitation, and maintaining the upright postural position of the head could not be controlled as carefully.

The use of 3-dimensional (3D) imaging in dentistry, more specifically cone-beam computed tomography (CBCT), has increased considerably in the last years, making possible the evaluation of anatomic structures and analysis of pharyngeal space morphology. Because of its high spatial resolution, adequate contrast between the soft tissues and empty space, and the relatively low radiation dose compared with multislice computed tomography, CBCT is an important tool in the study of craniofacial development.

Due to the close relationship between the pharynx, mandible, and hyoid bone and the fact that orthodontic or orthognathic interventions may affect the pharyngeal space, information regarding the influence of skeletal classes and facial types on these structures would improve the diagnosis and treatment of orthodontic patients.

The aim of this study was to correlate the volume of the pharyngeal space, the size and shape of mandible and the hyoid bone in patients with different facial types and skeletal classes. Furthermore, we estimated the volume of the pharyngeal space with a formula using only linear measurements.

## Material and methods

This study was approved by the research ethics committee of Piracicaba Dental School, State University of Campinas, in Brazil with protocol number 092/2014.

This retrospective study was performed on a batch of previously taken CBCT volumes (i-CAT Next Generation; Imaging Sciences International, Hatfield, Pa) at 120 kV, 5 mA, 23 × 17-cm field of view, 0.4-mm voxel, and 40-second scanning time, with indication for orthodontic treatment or orthognathic surgery planning. The CBCT examinations were made with each subject sitting upright, and with the Frankfort horizontal plane parallel to the ground and the patient’s teeth occluding in maximum intercuspation.

A total of 161 CBCT volumes from 80 men and 81 women, aged between 21 and 58 years (mean age, 27 years), were included in this study. Patients younger than 21 years (due to incomplete development of their craniofacial structures), patients who had orthognathic surgery, and those with pathologies in the region of the head and neck or syndromes were excluded from the study.

Skeletal class (Classes I, II, and III) and facial type (brachycephalic, mesocephalic, and dolichocephalic) were determined by an orthodontist with 13 years of experience for each patient from multiplanar reconstructions (lateral cephalometric) derived from the CBCT images with the NemoCeph software (Nemotec, Madrid, Spain).

To determine the skeletal classes, classified as Class I, Class II, or Class III, we used the SNA, SNB, and ANB angle measures from the cephalometric analysis of Steiner. All patients selected as Class II had a greater value of SNA angle (accentuated development of the maxilla), whereas all Class III patients had an increased SNB angle (accentuated development of the mandible). The measurement of Jarabak and Fizzell for line connecting point A to the occlusal plane and line connecting point B to the occlusal plane was used to confirm the skeletal class classification.

With regard to facial type, differentiation into vertical groups (brachyfacial, mesofacial, dolichofacial) was determined by the VERT index (arithmetic average of 5 cephalometric measurements, angle of the facial axis, facial depth, mandibular plane angle, lower facial height, and mandibular arch) as calculated in the cephalometric analysis of Ricketts. With regard to the VERT index, a negative value corresponded to a dolichofacial type and a positive value to a brachyfacial type; if the value was zero, the patient was classified as mesofacial.

Linear and angular measurements were performed using the CS 3D Imaging software (version 3.4.3; Carestream Health, Rochester, NY). The linear ones were anterior nasal spine-posterior nasal spine distance (ANS-PNS distance), shortest distance of the pharyngeal space (shortest distance), anteroposterior distance of the pharyngeal space-C1 (APC1), latero-lateral distance of the pharyngeal space-C1 (LLC1), anteroposterior distance of the pharyngeal space-C2 (APC2), latero-lateral distance of the pharyngeal space-C2 (LLC2), anteroposterior distance of the pharyngeal space-C3 (APC3), latero-lateral distance of the pharyngeal space-C3 (LLC3), anteroposterior distance of the pharyngeal space-base of epiglottis (AP epiglottis), latero-lateral distance of the pharyngeal space-base of epiglottis (LL epiglottis), latero-lateral distance of mandible (LL mandible), anteroposterior distance of mandible (AP mandible), latero-lateral distance of hyoid bone (LL hyoid bone), and anteroposterior distance of hyoid bone (AP hyoid bone). The angular measurements were anteroposterior angle of mandible (AP angle of mandible), transverse angle of mandible (TA mandible), and transverse angle of hyoid bone (TA hyoid bone). The measurements are listed in Table I and Figures 1-4 .

Measurement | Description | Reconstruction | Figure |
---|---|---|---|

Pharyngeal space dimensions: | |||

Anterior nasal spine-posterior nasal spine distance | Line from the most anterior to the most posterior point of hard palate | Sagittal | 1A |

Shortest distance | Horizontal line on the greatest constriction of pharyngeal space | Sagittal | 1B |

C1- latero-lateral distance | Horizontal line on the greatest latero-lateral dimension of pharyngeal space oriented at the level of the most inferior point of C1 | Axial | 2A |

C1-anteroposterior distance | Vertical line on the greatest anterior-posterior dimension of pharyngeal space oriented at the level of the most inferior point of C1 | Axial | 2A |

C2-latero-lateral distance | Horizontal line on the greatest latero-lateral dimension of pharyngeal space oriented at the level of the most inferior point of C2 | Axial | 2B |

C2-anteroposterior distance | Vertical line on the greatest anteroposterior dimension of pharyngeal space oriented at the level of the most inferior point of C2 | Axial | 2B |

C3-latero-lateral distance | Horizontal line on the greatest latero-lateral dimension of pharyngeal space oriented at the level of the most inferior point of C3 | Axial | 2C |

C3-anteroposterior distance | Vertical line on the greatest anteroposterior dimension of pharyngeal space oriented at the level of the most inferior point of C3 | Axial | 2C |

Epiglottis-latero-lateral distance | Horizontal line on the greatest latero-lateral dimension of pharyngeal space oriented at the level of the most concave point of epiglottis base | Axial | 2D |

Epiglottis-anteroposterior distance | Vertical line on the greatest anteroposterior dimension of pharyngeal space oriented at the level of the most concave point of epiglottis base | Axial | 2D |

Mandible dimensions: | |||

Anterior-posterior angle of mandible | Angle between the most posterior point of the mandibular condyle, the gonion point and the most inferior border of the mandible body | Sagittal (MIP) | 1C |

Transverse angle of mandible | Angle between the most anterior point of the mandibular symphysis and the gonion point on right and left sides of the mandible | Axial | 3A |

Latero-lateral distance of mandible | Line between the right and left gonion points | Axial | 3B |

Anteroposterior distance of mandible | Perpendicular line from the most anterior point on the lingual surface of the symphysis to a line between the right and left gonion points | Axial | 3C |

Hyoid bone dimensions: | |||

Transverse angle of hyoid bone | Angle between the projections of the lines crossing the lesser and greater horns of right and left sides of hyoid bone | Axial | 4A |

Latero-lateral distance of hyoid bone | Line between the right and left greater horns | Axial | 4B |

Anteroposterior distance of hyoid bone | Perpendicular line from the most anterior point in the concavity of the body of the hyoid bone to a line between the right and left greater horns | Axial | 4C |

To perform the linear measurements, all CBCT examinations were oriented according to the structures to be measured. For pharyngeal space, the vertical reference line was positioned in the median sagittal plane, and the horizontal line was positioned from the anterior nasal spine to the posterior nasal spine in the axial and sagittal reconstructions. For the mandible, the horizontal line was tangentially positioned on the lower edge of the mandible in the sagittal reconstruction and then moved superiorly to the genial tubercle. For the hyoid bone, the software orientation line was positioned, in the sagittal reconstruction, on the long axis of this bone.

The analysis of the pharyngeal space volume (PS volume) was carried out from the 3D model. The reconstruction of the 3D model was established with the semiautomatic segmentation mode of the software Insight ITK-SNAP (version 2.4.0; Cognitica, Philadelphia, Pa), which measured the volume of the structure in cubic millimeters. The volume measured in this study corresponded to the union between oropharynx and hypopharynx. For this, we followed the anatomic delimitations described by Park et al establishing a superior reference line traced from the posterior nasal spine to the lowest point of the first cervical vertebra, and an inferior reference line traced on the most inferior point of the fourth cervical vertebra perpendicular to the medial sagittal plane ( Fig 5 ).

One calibrated examiner (Y.N.) performed all software operations and measurements in a subdued and quiet room. The examiner executed all 18 measures 10 times, with a 1-day interval, to assess the reproducibility of the method. The intraclass correlation coefficient was determined to assess the investigator’s reproducibility on the measurements.

## Statistical analysis

Data normality and homoscedasticity of variances were accessed by Shapiro-Wilk and Levene tests, respectively. Data analysis was carried out using statistical software (version 15.8; MedCalc Software, Ostend, Belgium). Analysis of variance and Tukey tests with a significance level of 0.05 (alpha, 5%) were used to compare the groups (facial type and skeletal class), and the Pearson correlation test was used to identify correlations between the volume of the pharyngeal space and the other variables analyzed. A linear regression was performed to create a formula for estimating the volume of the pharyngeal space.

## Results

The intraclass correlation coefficients were 0.988 for the linear measurements and 0.99 for the angular measurements.

The distributions of skeletal class and facial type are shown in Table II . There were no statistically significant differences between sexes for facial type or skeletal class.

Female (n = 81) | Male (n = 80) | Total (N = 161) | |
---|---|---|---|

Class I | 37 (45.7%) | 23 (28.8%) | 60 (37.3%) |

Class II | 31 (38.3%) | 29 (36.2%) | 60 (37.3%) |

Class III | 13 (16.0%) | 28 (35.0%) | 41 (25.4%) |

Brachycephalic | 35 (43.2%) | 35 (43.8%) | 70 (43.5%) |

Dolichocephalic | 17 (21.0%) | 21 (26.2%) | 38 (23.6%) |

Mesocephalic | 29 (35.8%) | 24 (30.0%) | 53 (32.9%) |

With regard to skeletal class, transverse angle of mandible, AP mandible, LL mandible, APC2, LL epiglottis, shortest distance, and PS volume showed statistically significant differences ( Table III ).

Class I (n = 60) | Mean (±SEM) | Class III (n = 41) | |
---|---|---|---|

Class II (n = 60) | |||

Transverse angle of hyoid bone | 41.7 (±1.26) | 43.5 (±1.38) | 41 (±1.38) |

AP hyoid bone | 23.2 (±0.55) | 23.5 (±0.43) | 25 (±0.72) |

LL hyoid bone | 36 (±0.6) b | 38.5 (±0.65) a | 37.5 (±0.84) ab |

Transverse angle of mandible | 62.5 (±0.44) a | 63.3 (±0.5) a | 60.4 (±0.72) b |

AP mandible | 56.4 (±0.66) b | 57.8 (±0.73) b | 62.8 (±1.09) a |

LL mandible | 80.5 (±0.66) b | 83.9 (±0.71) a | 85.5 (±0.9) a |

AP angle of mandible | 132 (119) | 131 (118.5) | 129 (12.5) |

APC1 | 13.1 (10.1) | 15 (11.15) | 13.4 (4.65) |

APC2 | 10.5 (±0.4) b | 9.5 (±0.45) c | 12.3 (±0.67) a |

APC3 | 11.7 (±0.45) | 11.1 (±0.59) | 13.1 (±0.65) |

AP epiglottis | 14.9 (±0.31) | 14.4 (±0.39) | 15.3 (±0.49) |

LLC1 | 29.2 (±0.85) | 30.7 (±0.92) | 31.2 (±1.14) |

LLC2 | 24.6 (±0.74) | 26.2 (±0.93) | 27.5 (±1.3) |

LLC3 | 29.6 (±0.54) | 29.9 (±0.69) | 30.9 (±0.72) |

LL epiglottis | 34.2 (±0.56) b | 36.2 (±0.49) a | 36.3 (±0.68) a |

ANS-PNS distance | 54.1 (±0.6) | 54.9 (±0.54) | 54.5 (±0.59) |

Shortest distance | 7.8 (±0.35) ab | 6.9 (±0.45) b | 8.8 (±0.43) a |

PS volume | 14560.2 (±660.32) b | 16110.8 (±910.7) ab | 18840.8 (±970.38) a |