Effects of reverse headgear on pharyngeal airway in patients with different vertical craniofacial features

The aim of this study was to investigate the effects of reverse headgear (RH) on pharyngeal airway morphology in two groups of Class III patients with different vertical craniofacial features in comparison with an untreated Class III group. Seventeen subjects (9 males, 8 females; mean age 11.3 ± 0.98 years) with optimum vertical growth and 17 subjects (10 males, 7 females, mean age 11.5 ± 1.1 years) with a vertical growth pattern treated with a removable intra-oral appliance and a Delaire type facemask were included. An untreated Class III control group of 11 subjects (8 males, 3 females, mean age 9.1 ± 1.1 years) was included to compare the treated groups. The paired t-test for intragroup and one-way ANOVA for intergroup comparisons were performed. The relationships between changes in the craniofacial morphology and airway were assessed by Spearman correlation analysis. The airway dimensions at the adenoid side and soft palate were increased in the treatment groups compared to the control group (p < 0.05). The nasopharyngeal area demonstrated a significant difference in normodivergent and control subjects (p < 0.05). No significant difference was found in the airway morphology due to different vertical features. The effect of RH treatment on the sagittal airway dimensions revealed no significant difference between different vertical craniofacial features in the short term.


Introduction
Maxillary advancement by reverse headgear (RH) has been a major treatment option in young skeletal Class III patients, 1,2 providing enhancement of maxillary growth and restraint and/or redirection of mandibular growth. 3The beneficial effects of RH on the upper airway dimensions have been demonstrated in previous studies. 4,5,6Airway analysis often included cephalograms, 7 and their reliability was found to be adequate. 8tudies have shown an association between craniofacial morphology and airway dimensions, 9,10 and there is limited knowledge due to the effect of a steep mandibular plane angle on airway size. 11,12,13Reduction in airway space after mandibular setback surgery in Class III patients was declared to be a causative factor in airway obstruction. 14Reduced airway dimensions are proposed to be related to an increased mandibular plane angle. 15,16Hence, there was an interest in whether different vertical craniofacial features evoke differences in airway dimensions.Thus, the aim of this study was to investigate the effects of RH treatment on airway morphology in Class III patients, demonstrating different vertical facial patterns (normodivergent versus hyperdivergent) in comparison with an untreated Class III control group.

Methodology
Materials for this retrospective study were sele cted f rom t he f i les of t he Depa r t ment of Orthodontics, Gazi Universit y, Dentistry Faculty.The study was approved by the Ethical Committee of Gazi University (protocol number: 2014/12544).Patients were included based on the following criteria: (1) RH treatment, (2) skeletal (ANB < 0°) and dental Class III malocclusion with maxillary retrusion (SNA < 82°), (3) anterior c rossbite, (4) opt i m a l or h ig h m a nd ibu la r plane angle (SN/GoGn:26-38°or > 38°), ( 5) no congenital anomalies, (6) presence of good quality cephalograms.Of the 102 patients who fit the above criteria, patients with chronological ages between 9-12 years who were in the prepubertal period (between the PP 2 and MP 3cap developmental stages) were chosen. 17Sample size was calculated with a statistical power of 0.80, and the number of patients per group was required to be at least 10.The final treatment groups included 34 patients (15 males, 19 females).
The Reverse Headgear normodivergent (RH-ND) group consisted of 17 subjects (9 males, 8 females; mean age 11.3 ± 0.98 years) demonstrating optimum vertical growth (SN/GoGn: 26-38°) who were treated with a removable intra-oral appliance with a thickness to open the bite to an edge-to-edge incisal position, with hooks between the lateral incisors and canines, and a Delaire type facemask (Figure 1).The mean treatment time was 9.7 ± 2.8 months.
The Reverse Headgear hyperdivergent (RH-HD) group consisted of 17 children (10 males, 7 females, mean age 11.5 ± 1.1 years) with increased vertical growth (SN/GoGn > 38°) who were treated with the same appliance for 10.6 ± 1.2 months.Both groups revealed a protraction force of 400-450 g per side delivered by elastics from the hooks, approximately 30° below the occlusal plane. 18Patients were asked to wear the appliances 14-16 hours per day.All were treated at least to a positive dental overjet before discontinuing treatment.
Control (CNT) group: Data on 11 untreated Class III subjects (8 males, 3 females, mean age 9.1 ± 1.1 years), which were collected for a previous study for 12 months, were used.
All cephalog rams were taken in nat ural head position using a Trophy Instrumentarium Cephalometer (Instrumentarium Imaging Co., Tuusula, Finland) at 70KVp, 16 mAs -1 , under standard conditions in the occlusal position, as after a usual swallow.The radiographs were traced on acetate papers and measured manually by the same researcher.Cephalometric measurements are shown in Figure 2A, 2B.Area measurements were performed by the NETCAD for Windows Software Programme (Ulusal Co., Ankara, Turkey), an engineering drawing program.The airway area was divided by the palatal line into the nasopharyngeal area (NA) and oropharyngeal area (OA) (Figure 3).The cephalograms were scanned (Epson Co., Nagano, Japan) using standard conditions.NA and OA were digitized according to the specified points by two different researchers to obtain maximum reliability and agreement.The numerical value of the areas was calculated as units.Fifteen randomly selected radiographs among all groups were retraced and remeasured by the same researcher 15 days after the first evaluation.Pearson correlation analysis was used to assess the reliability and reproducibility of the variables.The reliability coefficients (r 2 ) calculated for each parameter were between 0.91 and 0.97.Data were analyzed with SPSS for Windows, version 20.0 (SPSS Inc., Chicago, USA).The Shapiro-Wilk test was used to check the data normality.Intragroup comparisons were assessed by paired t-test, and the one-way analysis of variance (ANOVA) test was used for intergroup comparisons.The relationship between changes in the pharyngeal airway and craniofacial morphology was evaluated using Spearman correlation analysis.p < 0.05 was considered to be statistically significant.

Results
The pre-and post-treatment and pre-and post-observation descriptive values and comparison of the changes within each group are presented in Table 1.
Treatment changes in RH-HD group: Increases in SNA and ANB (p < 0.01), Co-A and Co-Gn (p < 0.001, p < 0.01) were found.The SN-PP angle decreased at a significance level of 0.05.There were increases in facial height, nasopharyngeal height, NA and OA dimensions (Table 1).Observation period changes: Both jaws came forward, declaring a significant decrease in ANB (p < 0.05), and increases in Co-A (p < 0.05) and Go-Gn (p < 0.01) were found.The mandibular plane angle decreased significantly (p < 0.05).Only nasopharyngeal height increased significantly (S-PNS, Pm'-SPL; p < 0.05).
Comparison of changes between groups: The increase in SNB was more prominent in the control group, showing counterclockwise rotation of the mandible during the observation period compared to the treatment groups (Table 2).The treatment groups showed significant improvements in ANB in NA showed a significant difference between RH-ND and the control group (p < 0.05), and MPS showed a difference between the treatment and control groups (p < 0.05).The hyoid bone is positioned more forward in the control group in comparison to the RH-HD group (p < 0.05).
In the correlation analysis, the airway measurements were selected as dependent variables: SNA, SNB, SN-GoGn, H-MP, H-SN, C3-H, SN-CVT and SN-OPT were considered as independent variables.For the RH-ND group, the changes in H-MP and C3-H showed significantly positive effects on OA (r = 0.596, p = 0.012; r = 0.484, p = 0.049, respectively).In the RH-HD group, while the change in SN-GoGn had a significantly negative effect on NA (r = -0.545,p = 0.024), SN-CVT and SN-OPT have significantly positive effects on IPS (r = 0.585, p = 0.014; r = 0.516, p = 0.034, respectively).In the control group, SN-OPT showed a significantly positive effect on MPS (r = 0.606, p = 0.048).

Discussion
Recently, 3D images have become used for evaluating airway dimensions. 19There is no doubt that 3D images would be preferable, but we believe that archives of two dimensional (2D) cephalometrics are still reliable for evaluating treatment effects, and they are reproducible and cheaper.Due to the retrospective design of this study and to ethical and economic limitations, we used cephalometric data.The ADA Council on Scientific Affairs recommended scientists to follow the "As Low As Reasonably Achievable (ALARA)" principle, which includes taking radiographs based on the patients' needs. 20dditionally, the association between 2D and 3D measurements was demonstrated. 21ur results showed that both treatment groups induced significant advancement of the maxilla associated with larger amounts of maxillary growth and the inhibition of sagittal mandibular growth with respect to the control group, consistent with previous studies reporting maxillary protraction between 1-3 mm. 1,4The mandible showed significant counterclockwise rotation and protrusion in the control group compared to the treatment groups.
No significant changes in airway parameters were found between the normodivergent and hyperdivergent groups in short term.The differences in mandibular rotation, which might implicitly cause an adverse effect on sagittal airway dimensions, exhibited no significant difference, which was consistent with a previous study. 22Due to the counterclockwise rotation of the palatal plane in the high angle group, the expected vertical changes in the mandibular plane angle did not occur, which might explain why we were not able to find intergroup differences.Current correlation analysis showed an inverse correlation between NA and the mandibular plane angle in the RH-HD group, consistent with Freitas et al., 23 who found that vertical growth patterns influenced upper airway dimensions.Confirming this point, Celikoglu et al. 24 found that the nasopharyngeal and oropharyngeal airway volume were significantly higher in the low-and normal-angle groups than in the high-angle group.Ucar and Uysal 16 also reported a decrease in nasopharyngeal airway space from low-angle to normal-to high-angle cases.
There are conflicting results in the literature regarding the effects of maxillary protraction with or without maxillary expansion therapy on the sagittal airway dimensions. 2,4,5,6,16According to current results, the nasopharyngeal dimensions representing the adenoid area demonstrated significant increases in the treatment groups compared to the control group, as in previous studies. 2,5,25This point might be related to the physiological growth pattern of the lymphoid tissue. 26Based on the knowledge that the mean size of the adenoids continues to increase until puberty with a gradual decline afterwards, 27 the adenoid tissue was approximately at its growth peak at the ages of the treated subjects in this study.Therefore, the existence of a direct favorable effect of RH treatment on the nasopharyngeal airway dimensions could not be concluded, and the size of adenoids, which are age-dependent, should be considered when evaluating the results.
Present analysis revealed a positive correlation between head posture and oropharyngeal airway in the RH-HD and control groups, in line with a previous study. 4Yagci et al. 6 also found increased airway dimensions and cranial flexion of the natural head position after expansion and protraction of the maxilla.Again, a positive correlation between the hyoid position and the oropharyngeal area in the RH-ND group was found in this study, declaring a more forward hyoid positioning after protraction.Accordingly, Muto et al. 28 found correlations between the pharyngeal airway and the hyoid position, jaw sizes, maxillary and mandibular prognathism, and mandibular inclination.The hyoid position represents tongue posture and function, and hyoid structures are guided to an anterio-inferior position to avoid compromising the vital airway passage as a compensatory action. 29

Conclusion
No significant differences were found for pharyngeal airway dimensions among different vertical skeletal patterns in Class III patients after short-term maxillary protraction.
When compared with the control subjects, the nasopharyngeal airway at the adenoid region was increased after maxillary protraction, which should be considered based on age, and the oropharyngeal airway suddenly declared advancement, at the level of the soft palate.

Figure 2 .Figure 3 .
Figure 2. A: Cephalometric analysis of jaws and nasopharynx.(1) SNA; (2) SNB; (3) ANB; (4) Co-A, effective maxillary length; (5) Co-Gn, effective mandibular length; (6) SN-GoGn; (7) SN-PP, palatal plane inclination; (8) N-ANS, upper anterior facial height; (9) ANS-Me, lower anterior facial height; (10) N-Me, total anterior facial height; (11) S-PNS, nasopharyngeal height; (12) ad1-PNS, distance from PNS to lower adenoid tissue; (13) ad2-PNS, distance from PNS to upper adenoid tissue; (14) AA'-Pm', distance between the perpendicular intersections of the anterior atlas and pterygomaxillary line along the palatal line; (15) Pm'-SPL, sphenoid line tangent to lower border of sphenoid.B: Cephalometric analysis of oropharynx, hyoid bone and head posture.(1) AA-PNS; (2) SPS, distance from the midpoint of the line from PNS to tip of soft palate to the horizontal counterpart on the posterior pharyngeal wall along parallel line to Frankfurt horizontal (FH) plane; (3) ve-Pve, distance of velum palatinum to the horizontal counterpart on the posterior pharyngeal wall along parallel line to FH plane; (4) MPS, distance from the tip of the soft palate to the horizontal counterpart on the posterior pharyngeal wall along parallel line to FH plane; (5) IPS, distance from the intersection points on the anterior and posterior pharyngeal wall through Cv2ai along parallel line to FH plane; (6) eb-Peb, distance from vallecula epiglottis to the horizontal counterpart on the posterior pharyngeal wall along parallel line to FH plane; (7) H-MP, perpendicular distance from hyoid bone to the mandibular plane; (8) H-SN, perpendicular distance from hyoid bone to SN plane; (9) C3-H, distance from the anterio-inferior point of the third cervical vertebra to point H; (10) SN-CVT, downward angle between SN plane and the line through Cv2tg and Cv4ip; (11) SN-OPT, downward angle between SN plane and the line through Cv2tg and Cv2ip.

Table 1 .
Pre-and post-treatment and observation mean and standard deviation of variables and comparison of the changes.

Table 2 .
Comparison of the changes between treatment and control groups.to the control group (p < 0.01).The treatment also induced favorable increases in Co-A (p < 0.05).Facial height and mandibular plane angle increased in the RH-ND group in comparison to other groups.