The objective of this research was to observe changes in aerodynamics and anatomic characteristics of the upper airway after mini-implants assisted rapid maxillary expansion and to evaluate the correlation between the 2 changes of the upper airway in young adults.
Thirty consecutive patients (mean age, 23.82 ± 3.90 years; median, 24.5 years; 9 males, 21 females) were involved. Cone-beam computed tomography was taken before activation and over 3 months. Three-dimensional models of the upper airway were reconstructed on the basis of cone-beam computed tomography. The anatomic characteristics of the upper airway, including volume, area, transverse, and sagittal diameter, were measured. The aerodynamic characteristics of the upper airway were calculated on the basis of 3-dimensional models using computational fluid dynamics. The correlation between the changes in aerodynamics and anatomic characteristics of the upper airway was explored.
The enlargements of the volume of the total pharynx, nasopharynx, and oropharynx were found (9.99%, 20.7%, and 8.84%, respectively). The minimum cross-sectional area increased significantly (13.6%). The airway resistance (R) and maximum velocity (V max ) decreased significantly in both the inspiration and expiration phase (inspiration: R, −26.8%, V max , −15.7%; expiration: R, −24.7%, V max , −16.5%). The minimum wall shear stress reduced significantly only in the inspiration phase (−26.3%). The correlations between decreased R and increased volume and minimum cross-sectional area were observed.
Mini-implants assisted rapid maxillary expansion is an effective device for improving anatomic characteristics represented by the total volume of the upper airway and minimum cross-sectional area, which contributed to the respiratory function depending on the favorable changes of aerodynamic characteristics including resistance, velocity, and minimum wall shear stress.
Mini-implants assisted rapid maxillary expansion could increase the anatomy of the upper airway.
Mini-implants assisted rapid maxillary expansion could improve the respiratory function of patients in the supine position.
The changes in anatomy contributed to the improvement of respiratory function.
Obstructive sleep apnea (OSA) is a breathing disorder during sleep characterized by recurrent apnea and hypopnea, which is associated with narrow upper airway, obesity, hypertension, and other adverse clinical diseases. In addition, some studies reported craniofacial disharmonies such as midface hypoplasia and narrow dentition were predisposing factors in the occurrence and progress of OSA, according to the influence of upper airway and muscular function. , OSA was a multifactor disease, and orthodontic treatment was only an effective method in treating OSA that was mainly caused by structural stenosis of the upper airway because of skeletal craniofacial disharmony. In contrast, the effects of orthodontic treatment were limited for OSA caused by other factors such as heredity, overweight, and dyslipidemia.
Rapid maxillary expansion (RME) is a conventional treatment method correcting transverse maxillary deficiency. Some studies have reported the effects of RME could increase the volume of the upper airway, and Cistulli et al first suggested RME could be a therapy for patients with OSA in 1998. However, conventional tooth anchored RME had dental side effects, such as buccal tipping of maxillary first molars and the reduction of buccal attachment of alveolar bone, and was not suitable for nongrowing patients whose midpalatal suture had been fused.
To reduce the undesirable side effects, mini-implants assisted rapid maxillary expansion (MARME) was designed to provide skeletal expansion guaranteed by 4 mini-implants and minimize the dental side effects compared with RME. , Some studies have pointed out MARME could cause the enlargement of the upper airway , and be considered as an effective treatment to alleviate the symptoms of OSA in adults.
Computational fluid dynamics (CFD) is an engineering technique used to solve problems relating to fluid or airflow by simulating the access that the fluid or airflow is going through a specific tube. CFD has been widely used to simulate the airflow of the upper airway on the basis of 3-dimensional (3D) data from CBCT, which could assess the respiratory function of patients. It could provide quantitative parameters describing the flow process of air in the upper airway by exhibiting the contours of aerodynamics characteristics such as pressure drop, velocity, and wall shear stress.
The effects of MARME on the upper airway of 1 patient using CFD analysis was reported. , However, the changes of 1 patient might not illustrate the reliability and universality of the effects of MARME. In addition, previous studies evaluated the changes of the upper airway after MARME on the basis of CBCT in the standing position. Based on these studies, the changes in a group of patients on the basis of CBCT in supine position had not been analyzed. In the optional location of taking CBCT, the supine position was the position that could describe the relatively narrow state of the upper airway, and the standing position was the position that could describe the relatively unobstructed state of the upper airway. Both positions were valuable positions for taking CBCT. In our study, we have gathered a group of patients, including 30 subjects from which the CBCT data were of subjects in the supine position.
This study aimed to assess the changes of anatomic and aerodynamics characteristics of the group of patients on the upper airway and to evaluate the correlation between aerodynamics characteristics calculated by CFD and the anatomy of the upper airway.
Material and methods
From 2019 to 2020, patients who were diagnosed with transverse deficiency of maxillary and have undergone MARME were consecutively enrolled in this retrospective study. During the collection process, 2 patients were excluded because of loose mini-implants, and 3 patients were excluded because of the defection of CBCT data. Ultimately, 30 patients were involved in this study. (mean age, 23.82 ± 3.90 years; median, 24.5 years; 9 males, 21 females; range, 18-33 years). The study was approved by the relative ethical commission, and the informed consent was signed by each patient.
The inclusion criteria for this study were shown as followed: (1) aged >18 years; (2) maxillomandibular skeletal transverse discrepancy 3 mm or greater ( Fig 1 ); (3) no history of expansion treatment or orthognathic surgery; and (4) no severe dentofacial anomalies such as a cleft lip or palate.
Each patient was treated by maxillary skeletal expansion type II (BioMaterials Korea, Seoul, Korea) developed by Brunetto et al at the University of California Los Angeles, which consisted of 2 stainless steel arms soldered to the bands on the maxillary first molars. After the bands were bonded to the maxillary first molars, 4 mini-implants (diameter, 1.5 mm; length, 11 mm; Mplant Series, BioMaterials Korea) were placed along with guided slots in the midpalatal region under local infiltration anesthesia ( Fig 1 ). The jackscrew was orientated on the midpalatal region generally, which was activated one sixth of a turn (0.13 mm) each day. The amount of expansion was set depending on the severity of each patient, which ranged from 40-60 turns, and the duration of expansion ranged from 40 to 60 days. The retention after activation was 3 months, allowing bone formation in the separated maxillary suture.
CBCT scans (5G; NewTom, Verona, Italy) were obtained before activation (T0) and over 3 months (T1). The CBCT device was set at 110 kVp; 7.33 mA; 0.3-mm voxel size; scan time, 4.8 seconds; and field of view of 18 cm × 16 mm. To assure the reliability of the measurements from the airway segmentation, we checked the position of each patient when taking CBCT. Each patient was scanned in the supine position in which the Frankfort horizontal plane (FHP) was perpendicular to the floor, keeping the teeth in centric occlusion, the tongue in the position at the end of swallowing (against the palate), breathing smoothly, no swallowing and the duration of the scan was about 15 seconds which was not difficult for our patients to keep constant. We had tried our best to ensure the uniformity of the head position and posture of our patients during CBCT before and after the treatment. The uniformity assured the reliability of the measurements to the greatest extent. The CBCT data were saved as Digital Imaging and Communications in Medicine format.
The Digital Imaging and Communications in Medicine data were imported into Dolphin Imaging software (version 11.8; Dolphin Imaging and Management Solutions, Chatsworth, Calif). The images were reoriented along the palatal suture, tangent to the nasal floor and parallel to the FHP. The upper airway was divided into 3 segments: nasopharynx, oropharynx, and hypopharynx, and all the descriptions and definitions of anatomic parameters were shown in Figures 2 and 3 and Table I .
|Parameters||Descriptions and definitions|
|CSA min||The minimum cross-sectional area of the upper airway|
|Transverse diameter of CSA min||The transverse line on the greatest transverse dimension at CSA min|
|Sagittal diameter of CSA min||The sagittal line on the greatest sagittal dimension at CSA min|
|Total volume||The anterior border is the line passing through PNS and S; the inferior border is the line parallel to the FHP passing through C4 (the anterior and inferior point of the fourth cervical vertebra), and the posterior border is the pharyngeal posterior wall|
|Superior boundary||The line passing through PNS and S|
|Nasopharyngeal volume||The anterior border is the line passing through PNS and S, the inferior border is the line parallel to the FHP passing through PNS, and the posterior border is the pharyngeal posterior wall|
|Interface1||The line parallel to the FHP passing through PNS|
|Transverse diameter of interface1||The transverse line on the greatest transverse dimension at interface1|
|Sagittal diameter of interface1||The sagittal line on the greatest sagittal dimension at interface1|
|Oropharyngeal volume||The superior border is the line parallel to the FHP passing through PNS, the inferior border is the line parallel to the FHP passing through the top of the epiglottis, and the posterior border is the pharyngeal posterior wall|
|Interface2||The line parallel to the FHP passing through the top of the epiglottis|
|Transverse diameter of interface2||The transverse line on the greatest transverse dimension at interface2|
|Sagittal diameter of interface2||The sagittal line on the greatest sagittal dimension at interface2|
|Hypopharyngeal volume||The superior border is the line parallel to the FHP passing through the top of the epiglottis, the inferior border is the line parallel to the FHP passing through C4 (the anterior and inferior point of the fourth cervical vertebra), and the posterior border is the pharyngeal posterior wall|
|Inferior boundary||The line parallel to the FHP passing through C4 (the anterior and inferior point of the fourth cervical vertebra)|
The CBCT data were imported into Mimics software (version 19.0; Materialise, Leuven, Belgium). The upper airway was highlighted by setting the threshold between −1024 and −360 Hounsfield Units. The anterior upper boundary was the line passing through PNS and S, the upper boundary was the roof of the nasopharynx, and the lower boundary was the plane across the C4 point (the most anterior inferior point of the fourth cervical vertebra) parallel to the FHP. All 3D models were exported as stereolithography files ( Fig 4 ).
All the models of the upper airway were loaded into software (ICEM 16.0; ANSYS, Canonsburg, Pa) to generate tetrahedral volume mesh. According to the complexity of the model of the upper airway, a typical grid consisted of approximately 2 million tetrahedral cells ( Fig 5 ).
After mesh generation, the 3D mesh was imported into software (FLUENT 16.0; ANSYS) for airflow simulation. The steady-state Reynolds Averaged Navier-Stokes formulation together with the laminar model was used to model aerodynamic characteristics of the upper airway. Second-order discretization schemes were used, and the coupling between velocity and pressure was achieved using the SIMPLE algorithm. The density and the viscosity of the air were set as 1.225 kg/m 3 and 1.79 × 10 −05 kg/m/s, respectively, which was acquiescent in the software. An inlet volume flow rate of 166 mL s −1 (10 l min −1 ) was set in the airflow simulation, and the standard atmospheric pressure of 0 Pa was set for the inlet. The air within the upper airway was thought to be adiabatic. In the inspiration phase, the inlet boundary was set at the line passing through PNS and S, and the outlet boundary was set at the plane across the C4 point parallel to the FHP. Conversely, the expiration phase was simulated by setting an inlet at the plane across the C4 point parallel to the FHP and outlet at the line passing through PNS and S. The iteration numbers were 400 steps.
Airway resistance (R) was calculated by the following formula: R = ΔP/Q. The total pressure drop between the inlet and outlet of the upper airway (ΔP) was computed by P max − P min , and Q was the volume flow rate, which was a constant. All the parameters were measured again by 1 researcher (H.T.) after 1 week, and the average value was applied in this study.
All the data were measured repeatedly after 1 week by 1 operator (H.T.), and the intraclass correlation coefficient was 0.91-0.97, indicating repeat agreement regarding all measurements.
Statistical analysis was performed with SPSS software (version 21; SPSS Inc, Chicago, Ill). The normality of data distribution and the homogeneities of variances were checked by the Shapiro-Wilk test and the Levene test, respectively. A paired t test was used for the comparison of normally distributed data between T0 and T1 and the Wilcoxon test for the comparison of nonnormally distributed data. Pearson correlation test was used to analyze the correlation of normally distributed data, and Spearman correlation test was used to assess the correlation of nonnormally distributed data. A P value <0.05 was determined as a statistical significance.
We observed significant changes in both anatomic and aerodynamics characteristics.
The changes of anatomic parameters of the upper airway between T0 and T1 were shown in Table II . Significant increases in total volume (V Tot ), minimum cross-sectional area (CSA min ) along with the transverse diameter and sagittal diameter of CSA min were found. The 3 segments of the upper airway, volume, cross-sectional area, and transverse diameter of nasopharynx displayed significant increases from T0 to T1 ( P <0.01). In addition, the volume of the oropharynx showed a significant increase ( P = 0.043), and other parameters of the 3 segments of the upper airway did not reveal significant changes.