Title
Author
DOI
Article Type
Special Issue
Volume
Issue
Long-term recurrence-free survival of orthodontic treatments for Class II malocclusion as the important supplement protocol to adenotonsillectomy in children with obstructive sleep apnea syndrome: a case-control retrospective study
1Department of Stomatology, Children’s Hospital affiliated to Capital Institute of Pediatrics, 100020 Beijing, China
*Corresponding Author(s):summer861202@163.com (Mengxing Wan)
| History | Submitted: 23 March 2024 | Accepted: 20 June 2024 | Published: 03 March 2025 |
| Copyright: | ©2025 The Author(s). Published by MRE Press. |
Background: To estimate the long-term recurrence-free-survival of three orthodontic treatments in the management of residual obstructive sleep apnea syndrome (OSAS) after adenotonsillectomy in children combined with Class II malocclusion taking conventional positive airway pressure (PAP) as control. Methods: A retrospective case-control cohort study, enrolled pediatric OSAS with Class II after adenotonsillectomy. The records of the patients were divided into a rapid maxillary expansion (RME), mandibular advancement appliances (MAA) and myofunctional therapy (MFT) group based on the therapy they received. Control group comprised patients used PAP at the same admission time. The primary endpoint was a long-term recurrence-free-survival. Secondary outcomes measures included the immediate post-operative success rate, improvements in the Epworth Sleepiness Scale (ESS) scores, craniofacial changes, s-electromyogram (s-EMG) changes of masticatory muscle and long-term OSAS sequelae. Results: A total of 1217 cases were analyzed. The estimated 1-, 3- and 5-year recurrence-free rate was 81.9%, 70.8% and 56.4% in PAP vs. 86.2%, 81.0% and 77.5% in RME vs. 85.5%, 78.0% and 74.7% in MAA vs. 83.0%, 74.9% and 58.3% in MFT, respectively. Significantly greater median of recurrence-free-survival was observed in RME, MAA and MFT group as opposed to PAP-control group (122.04 ± 3.04 (95% CI (confidence interval): 116.08, 127.99), 20.04 ± 4.21 (95% CI: 111.79, 128.28), 68.96 ± 4.95 (95% CI: 59.24, 78.68) vs. 54.96 ± 2.51 (95% CI: 50.05, 59.87)). The post-treatment ESS scores, craniofacial variables, s-EMG parameters greatly improved at post-treatment from the baseline value and sustained during long-term follow-up. No severe complications were observed. Conclusions: As opposed to conventional PAP, a better long-term RFS (recurrence free survival) was associated with the therapeutic trial using RME and MAAs, which provided a reasonable alternative for residual OSAS after adenotonsillectomy in children combined with Class II malocclusion. Their benefits appeared especially in children due to poor compliance to conventional PAP.
Cite this article
Tian Xia, Ruqian Luo, Mengxing Wan. Long-term recurrence-free survival of orthodontic treatments for Class II malocclusion as the important supplement protocol to adenotonsillectomy in children with obstructive sleep apnea syndrome: a case-control retrospective study. Journal of Clinical Pediatric Dentistry. 2025; 49(2): 178-187. doi: 10.22514/jocpd.2025.037
1. Introduction
Obstructive sleep apnea syndrome (OSAS) is the most serious pediatric disease with sleep disorder. It is characterized by episodes of hypopnea or obstructive apnea during sleep, resulting in abnormal ventilation and sleep pattern [1]. The prevalence ranges from 1% to 5% in children, with a peak incidence between 2 and 6 years of age [2]. If not treated in an early time point, it can lead to severe complications such as growth retardation, learning deficits, neurocognitive impairments, pulmonary hypertension and cardiovascular consequences, that persist into adulthood [3]. Analysis of large-scale in Chinese children verified that unlike the cases in adults, tonsillar hypertrophy and adenoid hypertrophy were identified as the primary factors associated with the development of pediatric OSAS [4, 5]. Given that the adenotonsillectomy (AT) is reported as the first-line surgical treatment and cures airway obstruction in 75% to 100% of the cases. However, residual OSAS may persist in a substantial proportion of children even after AT, which indicates that in addition to large tonsils and adenoids, other comorbidities are present. Children with OSAS often present the underlying craniofacial abnormalities such as narrow upper airway as a result of having mandibular retrusion, small mandible and Class II malocclusion, which influence the presence and/or the severity of OSAS [6, 7]. The constant use of continuous positive airway pressure (PAP) is usually considered as the gold standard treatment for residual OSAS despite AT. Nevertheless, adherence to PAP in youth is generally very poor, and might be limited due to the risk of developing craniofacial sequels over time [8, 9]. Based on the existing findings, OSAS patients may benefit from orthodontic treatment options, such as rapid maxillary expansion (RME), mandibular advancement appliances (MAA) and myofunctional therapy (MFT), depending on repositioning the mandible in a forward and downward position to relieve upper airway obstructions during sleep. As tied together with conventional AT as the multidisciplinary treatment, it is more effective rather than separately to treat by AT alone [10, 11].
To our knowledge, there are systemic reviews and meta-analyses, supported mainly by uncontrolled clinical trials or limited in terms of small sample sizes, reporting the effectiveness of different types of orthodontic and functional treatments for pediatric OSAS, but very few studies compared these modalities together with long-term rather than short-term follow-up, or even had conflicting results. We therefore firstly conducted a 10-year retrospective case-control study, the aim of which was to estimate the long-term recurrence-free survival (RFS) of the three common orthodontic treatments for residual OSAS after AT in growing children, taking PAP as control.
2. Methods
2.1 Study design and participant recruitment
From 01 January 2012 to 29 February 2023, the electronic medical records (EMRs) of children affected by mild to moderate OSAS, with Class II malocclusion patterns, and treated by orthodontic treatments were screened in our stomatology department following the the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [12]. Inclusion criteria were as follow: (1) patient diagnosed with OSAS based on the polysomnography (PSG) examination in compliance with the standard criteria of the American Academy of Sleep Medicine (AASM) international classification of sleep disorders [13], or the inclusion (or not) of excessive daytime sleepiness as a necessary diagnostic criterion [14]; (2) an obstructive apnea-hypopnea index (AHI) score between 1 and 9 [13]; (3) age between 5–10 years; (4) history of AT surgery before orthodontic treatments; (5) cephalometric standards of Class II malocclusion with overjet ≥5.0 mm and ANB (AB plane angle) angle >4°; (6) body mass index (BMI) <24 kg/m2; (7) complete medical data; (8) at least 1 years of follow-up. Patients were excluded if they had other sleep disorders rather than OSAS, i.e., asthma, craniofacial anomalies related to certain syndromes such as Down, Pierre Robin Syndromes, etc., or other concomitant anatomical factors including severe temporomandibular disorders, nasal stenosis, macroglossia, rhinosinusitis and cleft palate.
Subjects were separated into one of the following three case groups or control group according to their therapeutic modalities (Fig. 1). RME, MAA and MFT group respectively consisted of patients receiving RME, MAAs and MFT therapy for residual OSAS following AT surgery. The control group were selected in patients who did not undergo any orthodontic procedure after AT surgery, whereas, used positive airway pressure therapy (PAP) at the same admission time. They were individually matched to cases by same sex, similar age and hospital admission date.
Fig. 1.Recurrence-free survival (RFS) curves for PAP, RME, MAA, MFT group over a 10-year follow-up period. According to Log-rank test, significant higher long-term RFS was observed in RME group as compared to PAP (p < 0.001) and MFT group (p = 0.016), while comparable to MAA group (p = 0.217). PAP: positive airway pressure; MAA: mandibular advancement appliances; RME: rapid maxillary expansion; MFT: myofunctional therapy.
2.2 PAP therapy
All patients were prescribed continuous PAP used a ResMed AirSense 10 AutoSet machine (ResMed, San Diego, CA, USA) with the ability to auto-titrate positive airway pressure mode based on a proprietary algorithm. After initial inpatient-education and mask fitting program, patients underwent a nocturnal polysomnography with PAP titration. The minimal PAP pressure able to improve the AHI to <5 episodes/h and without desaturation <90% including supine rapid eye movement (REM) sleep was used for the therapeutic PAP pressure. Participants were instructed to wear the PAP ≥4 hours a night over a 1-year period after trial and fitting. Adherence data including the PAP usage time, 95th percentile air pressure, AHI and mask leak were automatically upload to ResMed AirView™ 10 (ResMed, San Diego, CA, USA) for review.
2.3 Orthodontic therapy
All treatments were conducted by the same skilled clinicians who had expertise in techniques for children with class II malocclusion in our department.
2.3.1 RME procedures
Patients in this group were treated with the Hyrax expander, using fixed devices to anchor in the second deciduous molars and weld to an expander screw in the palatal surfaces. All patients were treated based on the same expansion protocol. The device was active after cementation. The spiral expander was rotated every 1/4 rotation (90°), and the rapid expansion involved 2 to 4 times per day for a total of 15 days. Therefore, the maxillary arch would be expanded by 0.4 mm a day and 4 mm in 10 days. Patients were reviewed at intervals of two weeks. Subsequently, the device was kept in the mouth as a retainer for an average duration of 12 months after the active expansion period.
2.3.2 MMA (mandibular advancement appliances) procedures
The Twin-Block appliance was used for all patients in this group, which was custom-made from dental impressions. It comprised the upper and lower bite blocks with inclined planes, which are designed to fit together so that the mandible adopted an anterior position. All patients were treated based on the standard titration protocol, in which the titration was set at 50% of the maximal mandibular protrusion at baseline. Next, patients were evaluated and advanced in steps of 0.5 mm per week at each clinic visit until the maximum comfortable mandibular advancement. The duo-block device should be worn for a period of about 12 months.
2.3.3 MFT procedures
Each subject in this group were treated by the speech pathologists of the same rank according to oropharyngeal exercise protocol including both isotonic and isometric exercises to strengthen mouth and pharynx muscles, which were mainly working on functions such as breathing, blowing, sucking, speaking, chewing and swallowing. The MFT training included (1) soft palate: oral vowel sound pronunciation; (2) tongue: brushing it along the superior and lateral teeth surfaces, placing its tip against the hard palate and sliding it backward, pressing entirely the tongue against the hard palate and sucking upward against the palate, forcing its back against the floor of the mouth and keeping its tip in contact with the inferior incisive teeth; (3) face: alternated contraction and relaxation of orbicular muscle movements and lateral movements of buccinators and mandible muscles; (4) stomatognathic functions: inhaling nasally and exhaling orally, in conjunction with specific chewing training exercise using the tongue in the palate closed teeth without perioral contraction [15]. Two-hour exercises should be guaranteed every day, and persisted for a total of 12 months.
2.4 Outcome measurement
Demographic details and clinical information were taken from the EMRs. Follow-up data were collected as part of routine contact tracing through clinic visits at intervals of 6 months for at least two years after initiation of treatments, and thereafter every 12 months on schedule through telephone interviews by the specially trained investigators for the long-term follow-up visits. All data were documented in our prospective database.
Daytime sleepiness was estimated by the Epworth Sleepiness Scale (ESS), which is an 8-item questionnaire scored on a scale of 0–3 with a maximum score of 24. Higher scores indicated a higher likelihood of falling asleep [16]. The severity of OSAS was clinically evaluated by AHI, defined as the number of apneas or hypopneas recorded during per hour of sleep using polysomnogram. Therefore, a successful responder was pre-defined as a reduction of AHI >50% from the initial AHI value was achieved. Recurrence was considered as the recurrence of clinical sleep-related symptoms or AHI by repeat polysomnogram in patients, who were treated by AT followed by orthodontic treatments. Consequently, RFS was defined as the interval of maintaining clinical remission from the complete of treatments to the date when OSAS was recurrent. Cephalometric variables were estimated by the lateral cephalograms.
The primary endpoint was a long-term success rate after treatment. The secondary outcomes measures included the immediate post-operative success rate, improvements in ESS scores, the craniofacial changes, s-EMG changes of masticatory muscle and long-term OSAS sequelae.
2.5 Sample size calculation
PASS software, version 22 (NCSS, LLC. Kaysville, UT, USA) was used to calculate sample size. Our hypothesis was that the RME or MAA treatment for OASA children with Class II malocclusion were superior to the conventional PAP therapy with regard to long-tern RFS. Based on the previous publications, we estimated that the probability of observing a recurrence of OASA was 0.5 to 0.7 in the conventional PAP group and 0.4 to 0.6 in the RME or MAA group [17, 18]. The margin of clinical superiority was set at 0.9 with actual Hazard Ratio at 0.4 to 0.6 by 0.1. Assuming a power of 90% and type I error of 5% in a two-sided test, we determined to include at least 161 patients in each group.
2.6 Statistical analysis
Statistics were performed with SPSS software, version 22.0 (SPSS Inc, Chicago, IL, USA). Statistical significance was set at the 5% level with 2-tailed test.
Normal distribution (age, BMI, OAHI (obstructive apnea hypopnea index), ESS scores) and categorical variables (gender, OSAHS (obstructive sleep apnea hypopnea syndrome) severity, successful responders, compliance rate, adverse events) were shown as mean ± standard deviation (SD) and proportion. Statistic differences among groups were determined using analysis of variance (ANOVA) and Chi-square test, respectively. The probability of RFS was evaluated by the Kaplan-Meier method, with differences among groups assessed by log-rank analysis with 95% confidence interval (CI). The censored point was defined as remission at the last clinical contact or the last follow-up assessment. Patients lost during follow-up were censored at the time of the last follow-up. Analysis of cephalometric changes and s-EMG changes was performed using ANOVA and Bonferroni post-hoc test.
3. Results
3.1 Patients’ recruitment and baseline characteristics
A total of 1367 children were assessed for eligibility from 2005 to 2022. Of these, 10.9% of cases were excluded following data review, including 86 cases not meeting the inclusion criteria, 27 meeting the exclusion criteria and 37 with missing data. Therefore, a total of 1217 cases were involved. Demographic data and clinical characteristic of subjects before orthodontic treatments were illustrated in Table 1. No significant differences were observed among the four groups.
| Parameter | PAP (n = 274) | RME (n = 628) | MAA (n = 233) | MFT (n = 82) | F/χ2 value | p | |
| Age (yr) | 6.69 ± 1.73 | 6.59 ± 1.70 | 6.61 ± 1.91 | 6.67 ± 1.69 | 0.065 | 0.978 | |
| Female, sex, n (%) | 144/274 (52.6%) | 292/628 (46.5%) | 113/233 (48.5%) | 44/82 (53.7%) | 3.660 | 0.301 | |
| BMI, kg/m2 | 21.21 ± 1.84 | 21.94 ± 3.33 | 21.53 ± 2.65 | 21.16 ± 3.29 | 0.377 | 0.770 | |
| OAHI (times/h) | 5.87 ± 0.74 | 6.21 ± 1.28 | 5.95 ± 1.00 | 6.28 ± 1.08 | 0.850 | 0.470 | |
| min SaO2 (%) | 87.45 ± 6.34 | 86.86 ± 6.68 | 87.66 ± 8.78 | 87.84 ± 6.23 | 0.136 | 0.938 | |
| OSAHS severity | |||||||
| Mild (1 ≤ AHI < 5 time/h) | 127 (46.4%) | 307 (48.9%) | 112 (48.1%) | 38 (46.3%) | 0.587 | 0.899 | |
| Moderate (5 ≤ AHI < 10 time/h) | 147 (53.6%) | 321 (51.1%) | 121 (51.9%) | 44 (53.7%) | |||
| Adherence rate, n (%) | 222 (81.0%) | 577 (91.9%) | 221 (94.8%) | 72 (87.8%) | 32.648 | <0.001 | |
| Successful responders (%) | 195/274 (71.2%) | 534/628 (85.0%) | 189/233 (81.1%) | 55/82 (67.1%) | 31.981 | <0.001 | |
| PAP: positive airway pressure; MAA: mandibular advancement appliances; RME: rapid maxillary expansion; MFT: myofunctional therapy; BMI: body mass index; SaO2: arterial oxygen saturation; OSAHS: obstructive sleep apnea hypopnea syndrome; AHI: obstructive apnea-hypopnea index. |
3.2 Short-term outcomes
Among the involved cohort, 81.1% of patients achieved immediate success after MAAs therapy with a ≥50% AHI reduction, which is similar to that after 12 months of RME treatment (85.0%). Whereas, it is greater compared with PAP-control (71.1%) and MFT treatment (67.1%). According to ANOVA analysis, the post-treatment ESS scores (mean ± SD) decreased greatly from the baseline value of 21.33 ± 3.67 in PAP, 22.50 ± 1.64 in RME, 22.56 ± 3.04 in MAA and 23.09 ± 2.59 in MFT.
The cephalometric analysis was revealed in Table 2 using ANOVA analysis with Bonferroni post-hoc test. The maxillary incisor angle between the upper incisor line and maxillary plane significantly decreased as results of RME, MAA and MFT treatment, when comparing to PAP control group, indicating a maxillary incisor retroclination. Mandibular measurements showed that the SNB (sella-nasion-B point) angle presenting the sagittal position of the mandible significantly increased, and the mandibular incisor angle between the lower incisor line and the mandibular plane significantly increased after three orthodontic treatments. In which, it indicated a proclination of the mandibular incisors. Additionally, an increased mandibular base was also observed. Regarding the intermaxillary relationships, there was a significant increase in the ANB-angle. Ultimately, the overjet and overbite had significantly decreased. Table 3 showed the s-EMG findings. Significant improvements were achieved in the AsI (asymmetry index) of TA and MM (masseter muscles) after three orthodontic interventions. In addition to the benefits of balanced occlusion, the mV (millivolt) of TA and MM also showed significant increases. The control group did not show any significant changes.
| Parameters | PAP (n = 195) | RME (n = 534) | MAA (n = 189) | MFT (n = 55) | ||||||||
| T0 | T1 | T2 | T0 | T1 | T2 | T0 | T1 | T2 | T0 | T1 | T2 | |
| Angular (°) | ||||||||||||
| SNA | 80.93 ± 1.35 | 77.21 ± 1.48* | 78.35 ± 1.36* | 80.44 ± 1.37 | 80.65 ± 1.36 | 81.42 ± 1.17 | 79.50 ± 1.29 | 80.25 ± 1.16 | 80.89 ± 1.25 | 79.75 ± 1.25 | 80.00 ± 1.41 | 80.75 ± 1.01 |
| SNB | 74.24 ± 1.52 | 74.53 ± 1.64^#■ | 74.58 ± 1.33^#■ | 74.35 ± 1.16 | 76.65 ± 1.26*• | 77.00 ± 2.04*• | 74.51 ± 1.29 | 76.75 ± 1.28*• | 76.93 ± 1.94*• | 73.50 ± 0.99 | 75.75 ± 0.96*• | 75.13 ± 1.79* |
| ANB | 5.71 ± 1.14 | 5.53 ± 1.21^#■ | 5.43 ± 1.02^#■ | 5.76 ± 1.19 | 3.04 ± 1.06*• | 3.12 ± 1.04*• | 5.80 ± 1.24 | 3.01 ± 1.31*• | 3.10 ± 1.05*•■ | 5.70 ± 1.06 | 3.40 ± 0.89*• | 3.56 ± 2.04*•^# |
| Maxillary incisor angle Is-as/Max | 102.71 ± 1.64 | 102.36 ± 1.40^#■ | 102.12 ± 1.23^#■ | 101.65 ± 1.39 | 96.02 ± 2.91*• | 95.26 ± 2.26*• | 101.80 ± 1.92 | 95.81 ± 2.68*• | 95.38 ± 2.63*• | 101.81 ± 1.52 | 97.81 ± 1.47*• | 96.44 ± 1.70*• |
| Mandibular incisor angle ii-ai/Man | 90.22 ± 1.07 | 90.92 ± 1.35^#■ | 92.80 ± 1.63^#■ | 90.50 ± 1.19 | 97.44 ± 1.70*• | 97.12 ± 1.47*• | 90.40 ± 1.14 | 97.82 ± 1.30*• | 97.63 ± 1.52*• | 90.80 ± 1.10 | 95.38 ± 1.22*• | 96.04 ± 1.49*• |
| Linear (mm) | ||||||||||||
| Maxillary base Art-A | 86.06 ± 4.05 | 84.12 ± 4.84*^#■ | 83.01 ± 5.20*^#■ | 86.72 ± 4.74 | 88.81 ± 4.64*• | 89.01 ± 4.66*• | 87.11 ± 4.55 | 88.94 ± 4.59*• | 88.92 ± 3.83*• | 86.80 ± 4.44 | 88.80 ± 4.28*• | 88.17 ± 3.69* |
| Mandibular base Art-Pog | 68.12 ± 3.25 | 69.34 ± 3.33^#■ | 68.27 ± 3.63^#■ | 67.05 ± 3.24 | 73.14 ± 3.70*• | 73.81 ± 3.94*• | 66.83 ± 3.03 | 73.21 ± 3.70*• | 73.63 ± 3.83*•■ | 68.11 ± 3.63 | 71.87 ± 3.33*• | 70.18 ± 3.58*•^# |
| Overjet | 7.35 ± 1.48 | 7.38 ± 1.54^#■ | 7.44 ± 1.55^#■ | 7.39 ± 1.66 | 3.38 ± 1.17*• | 3.12 ± 1.44*• | 7.40 ± 1.89 | 3.44 ± 1.71*• | 3.20 ± 1.65*•■ | 7.33 ± 1.75 | 3.58 ± 1.86*• | 3.65 ± 1.49*•^# |
| Overbite | 6.26 ± 1.26 | 6.19 ± 1.56^#■ | 6.63 ± 1.41^#■ | 6.17 ± 1.05 | 3.68 ± 1.59*• | 3.70 ± 1.96*• | 6.10 ± 1.58 | 3.63 ± 1.51*• | 3.78 ± 2.09*• | 6.25 ± 1.66 | 3.61 ± 1.14*• | 4.03 ± 1.25*•^# |
| SNA: sella-nasion-A point; SNB: sella-nasion-B point; ANB: AB plane angle; PAP: positive airway pressure; RME: rapid maxillary expansion; MAA: mandibular advancement appliances; MFT: myofunctional therapy. *p < 0.05 compared with T0, ^p < 0.05 compared with RME group, #p < 0.05 compared with MAA group, ■p < 0.05 compared MFT group, •p < 0.05 compared with PAP group. |
| Variables | PAP (n = 274) | RME (n = 628) | MAA (n = 233) | MFT (n = 82) | ||||||||
| T0 | T1 | T2 | T0 | T1 | T2 | T0 | T1 | T2 | T0 | T1 | T2 | |
| mV of MM | 85.93 ± 13.51 | 81.21 ± 14.80^#■ | 81.35 ± 13.69^#■ | 84.65 ± 13.74 | 113.91 ± 14.84*• | 111.42 ± 11.71*• | 88.60 ± 10.48 | 109.64 ± 21.64*• | 110.89 ± 11.25*• | 86.75 ± 14.25 | 108.00 ± 12.41*• | 109.75 ± 10.01*• |
| mV of TA | 64.24 ± 15.23 | 74.53 ± 16.49^#■ | 74.58 ± 1.33^#■ | 65.63 ± 15.75 | 89.91 ± 17.14*• | 90.00 ± 18.61*• | 66.98 ± 14.88 | 88.55 ± 19.96*• | 89.93 ± 16.94*• | 63.50 ± 14.99 | 85.75 ± 15.41*• | 84.13 ± 11.79*• |
| AsI of MM | 79.90 ± 5.17 | 81.97 ± 7.20^#■ | 80.65 ± 7.66^#■ | 78.94 ± 5.85 | 91.18 ± 4.85*• | 90.88 ± 4.33*•■ | 77.59 ± 4.02 | 92.03 ± 3.54*• | 91.00 ± 4.64*•■ | 78.34 ± 5.67 | 88.23 ± 4.57*• | 86.09 ± 3.57*•^# |
| AsI of TA | 76.26 ± 4.97 | 78.33 ± 3.74^#■ | 76.63 ± 3.41^#■ | 76.95 ± 5.11 | 90.49 ± 4.22*• | 93.70 ± 4.96*•■ | 78.08 ± 4.90 | 91.63 ± 4.03*• | 91.78 ± 4.09*•■ | 77.25 ± 5.66 | 90.42 ± 5.14*• | 83.83 ± 5.25*•^# |
| s-EMG: surface electromyography; TA: anterior temporalis; MM: masseter muscles; mV: maximum voltage; AsI: asymmetry index; PAP: positive airway pressure; RME: rapid maxillary expansion; MAA: mandibular advancement appliances; MFT: myofunctional therapy. *p < 0.05 compared with T0, ^p < 0.05 compared with RME group, #p < 0.05 compared with MAA group, ■p < 0.05 compared MFT group, •p < 0.05 compared with PAP group. |
3.3 Long-term outcomes
The median duration of follow-up was 105.00 months (IQR (inter-quartile range): 32, 178) (range: 24, 162) in PAP group, 96.50 (IQR: 30.5, 162.5) (range: 24, 178) in RME, 95.00 (IQR: 27, 163) (range: 24, 159) in MAA and 99 (IQR: 27, 171) (range: 24, 145) in MFT, respectively. No differences were observed with p = 0.842. Fig. 2 showed the RFS curves. A total of 29.7%, 26.6%, 22.2% and 38.2% of patients in PAP, RME, MAA and MFT experienced an event of OSAS recurrence during the long-term follow-up period. According to the Kaplan-Meier analysis, the rate of RFS was achieved as 81.9% vs. 86.2% vs. 85.5% vs. 83.0% in PAP, RME, MAA and MFT group, of 1 year, which were significantly higher than that in the matched-control group. The estimated 3-year recurrence-free of PAP, RME, MAA and MFT was 70.8%, 81.0%, 78.0% and 74.9%, respectively. Compared to those in control group, patients had a significantly higher rate of long-term remission of OSAS. As expected, this difference remained highly significant on the estimated 5-year in REM, MMA and MFT as opposed to control group (77.5% vs. 74.7% vs. 58.3% vs. 56.4%). As a matter of fact, the median of RFS was significantly longer in patients involved in RME (122.04 ± 3.04 (95% CI: 116.08, 127.99)), MAA group (120.04 ± 4.21 (95% CI: 111.79, 128.28)) than those in PAP-control (54.96 ± 2.51 (95% CI: 50.05, 59.87)) and MFT group (68.96 ± 4.95 (95% CI: 59.24, 78.68)) (RME vs. PAP, p = 0.001; MAA vs. PAP, p = 0.002; RME vs. MFT, p = 0.016; MAA vs. MFT, p = 0.048). But it did not differ between RME and MAA (p = 0.217), and between PAP and MFT (p = 0.408).
Fig. 2.ANOVA analysis showed that PAP, RME, MAA and MFT treatment were associated with significant greater decreases in the mean of ESS scores at post-treatment and long-term follow-up. Significant lower ESS scores were observed in three orthodontic groups as opposed to PAP-control group. *p < 0.05 compared with PAP, ^p < 0.05 compared with RME, #p < 0.05 compared with MAA, +p < 0.05 compared with MFT. PAP: positive airway pressure; MAA: mandibular advancement appliances; RME: rapid maxillary expansion; MFT: myofunctional therapy; TV: television; ESS: Epworth Sleepiness Scale.
As far as the long-term cephalometric variables were concerned, significant differences were found among RME, MMA and MFT in respect to ANB angle, mandibular base, overjet and overbite during the long-term follow-up based on the ANOVA analysis and Bonferroni post-hoc test. Subjects in PAP-control group showed a negative change in SNA (sella-nasion-A point) and maxillary base in response to the development of retrusion in midface (Table 2). Additionally, the results of long-term follow-up also confirmed the improved s-EMG parameters of masticatory muscles. However, significantly decreased AsI of TA and MM occurred after long-term of MFT treatment as compared to RME and MAA (Table 3).
3.4 Adverse events
There were no major complications observed in any of the orthodontic groups. The minor side-effects were reported in 52.1% of the patients receiving RME and MAA, with temporomandibular joint pain (10.3%), myofascial pain, mouth dryness, excessive drooling, gingiva irritation (18.6%), discomfort and tenderness of the teeth being the most frequent (23.2%). But these symptoms were transient and resolved in a few months post-treatment. Facial flattening and maxillary retrusion were observed in 65.3% and 34.3% of cases in the PAP group. Other side effects including nasal congestion, oronasal dryness, epistaxis, eye irritation from air leak, facial pain and skin abrasion affected 23.1% of the patients.
4. Discussion
To the best of our knowledge, the present study was firstly presented a 10-year longitudinal comparison of RFS of the common orthodontic modalities as a part of comprehensive treatment in a large group of pediatric patients with residual OSAS following AT. Our main result indicated that the use of RME and MMA as an adjuvant therapy for craniofacial alterations in persist OSAS after AT, resulted in significantly longer RFS than MFT and conventional PAP, with fewer new adverse events identified.
According to preliminary studies, many adolescences or adults begin snoring and experience OSAS in childhood, therefore, the importance of early diagnosis and measurement to prevent long-term behavioral and emotional problems in adult life should be stressed [19]. AT surgery is a well-established treatment option for pediatric OSAS. However, pediatric OSAS presents a multifactorial etiology leading to residual apnea when AT was only submitted. Interestingly, radiological study proved that craniofacial morphological factors were associated with a narrow upper airway and sleep-disordered breathing [20]. Epidemiological literature pointed out that the prevalence of craniofacial abnormalities in children with OSAS was 89.9% as compared to a healthy pediatric population [21], and various in persistent OSAS despite AT surgery [22]. Although PAP therapy is an effective treatment modality in this setting, effectiveness is limited by poor adherence [23]. If the etiology of residual OSAS is retrognathia or maxillary constriction, nowadays, there are orthodontic treatment options, such as RME, MAAs and MFT, suggested to complete correction of the malocclusion and to gain long-term stability during growth [24].
RME correction was first proposed as an orthodontic treatment in young patients. Recently, it was widely adopted in clinical practice to correct a narrow upper jaw since it first successfully linked to SDB (sleep-disordered breathing). It could reduce the risk of obstruction by decreasing nasal resistance and allowing tongue repositioning, which contributed to OSAS. As a consequence, the meta-analysis confirmed the short-term efficacy of RME with the evidence of a significant improvement in the AHI in children with slight or moderate OSAS as early as symptoms appear [18, 25]. The functional MAAs aiming to correct mandibular retrognathia through redirecting and stimulating of anterior mandibular growth has been used as an efficient therapy for mild-to-moderate pediatric OSAS with a reduction of at least 50% in the AHI [26]. MFT proposals is a multi-component treatment including several combinations of isotonic and isometric exercises, which involves several combinations of oropharyngeal muscles. The primary mechanism of action of MFT is to improve the function of upper airway dilator muscles that are essential to maintain pharyngeal patency, while muscular endurance exercises aim to improve the mobility of oropharyngeal muscles to reduce airway closures during sleep. Based on previous findings, it probably reduces AHI by 62% in children, daytime sleepiness and snoring in the short-term as opposed to sham therapy or no treatment [15]. Consistent with previous literatures, 85.0%, 81.1% and 67.1% of child patients achieved a 50% reduction in AHI immediately after RME, MAA and MFT for the treatment of residual OSAS following AT. In addition, we found that the ESS cores were significantly reduced from the baseline value in all the three orthodontic groups. Furthermore, RME, MAA and MFT were significantly associated with good adherence as compared to the conventional PAP, which was consistent to previous study [27]. Nowadays, elastodontic appliances or clear aligners have sustained their upward trend of being utilized more frequently as technology advances, which also showed good adherences [28]. However, their effects on OSAS patients are highly debatable with very limit evidences.
In the present study, at the end of orthodontic treatments, subjects showed a corrected positive overjet and overbite with better s-EMG results, which reflected the improvement in facial profile and occlusion. To have a better explanation of these results, some authors reported that after the mid-palatal suture opening using RME, not only transversal, but also vertical and anteroposterior changes occurred [29]. Previous study also revealed evidence that orofacial muscle training protocols on the correction of myofunctional and musculoskeletal problems in developing dentition, particular facilitating anterior open bite correction, which resulted in a positive overjet and overbite [30].
To our knowledge, only a few studies, even with small-size samples, have evaluated the long-term effects of these orthodontic treatments in pediatric OSAS, and their results were disappointing, as they did not address the etiological causes. To investigate the long-term stability, this study estimated the long-term effect of orthodontic treatments as adjunctive therapies to AT in children with Class II Malocclusion during a 10-year follow-up. According to our results, after a median of 90 months (IQR: 20, 160) follow-up, a 122.04 (95% CI: 116.07, 127.99), 120.04 (95% CI: 111.79, 128.28) and 68.96 (95% CI: 59.24, 78.68) months of RFS was observed in RME, MAA and MFT group, respectively, as opposed to 54.95 (95% CI: 50.05, 59.87) in PAP-control group. The estimated 3-year and 5-year recurrence-free of PAP, REM, MMA was 70.8% vs. 81.0% vs. 78.0% vs. 74.9% and 56.4% vs. 77.5% vs. 74.7%% vs. 58.3%, respectively. This was comparable to previous data reporting that the AHI decline maintained as indicated by follow-up tests ranging from 3 months to 14 years following RME [18]. The AHI significantly decreased at the long-term of 6.6 ± 2.8 years evaluation after MAA with 46.7% of patients obtaining an AHI <5 and 83.4% of patients attaining an AHI ≤15 events [31]. A longitudinal follow-up study taken from a randomized control trial containing 103 patients with OSAS revealed that both PAP and MAA therapy demonstrated good and stable treatment effects after a 1-, 2- and 10-year follow-up. PAP was better in terms of lowering AHI values after a decade of treatment, when both therapies are compared. However, it was limited in high rate of dropout due to nonadherence, died or lost to follow-up [17]. Conversely, this study found that patients had a significantly higher rate of long-term remission of OSAS as compared to PAP-control, for the correction of craniofacial abnormalities. Significant changes of overjet and overbite were found not only in short-term after a 12-month use of orthodontic appliances but also during the 10-year long-term follow-up. This results again mirrored findings by a meta-analysis of 12 studies revealing a significant improvement of the function or morphology of the craniofacial surface occurred over time with at least one year use of orthodontic appliances or MFT, and the impact becomes more pronounced as the duration of the intervention lengthens [31]. To the best of our knowledge, we firstly found that significant improvements about s-EGM factors illustrating balanced occlusion were achieved after long-term of three orthodontic interventions as opposed to control group, that did not show any significant changes.
With regard to long-term sequelae, this study showed that patients in three orthodontic groups did not experience a significant number of adverse events. Only a few minor post-treatment side effects occurred, which gradually resolved in a few months after therapy and not resulted in any significant long-term complications. Whereas, subjects in PAP control group experienced significant midfacial retrusion as opposed to other groups in our findings, which was consistent with previous evidence that nasal mask pressure inhibited midface growth or actively pushed midfacial structure backward during the growth phase [32].
The present study had several limitations. First, there might be the undetected confounders and probable bias due to the nature of retrospective analysis with observational data. Second, investigators who reviewed data were unable to be blinded to patients’ cohort. Thirdly, only a few cases were recruited for MFT group, this limited our ability to detect effects of long-term RFS. Future large-scale, long-term randomized studies were needed to validate our findings.
5. Conclusions
In conclusion, therapeutic trial using RME and MAAs showed a better long-term RFS as opposed to conventional PAP, based on our Kaplan-Meier analysis, and provided a reasonable alternative for residual OSAS after AT in children combined with Class II malocclusion. Their benefits appeared especially in children due to poor compliance to conventional PAP.
Abbreviations
RFS, recurrence-free survival; OSAS, obstructive sleep apnea syndrome; AT, adenotonsillectomy; PAP, positive airway pressure; RME, rapid maxillary expansion; MAA, mandibular advancement appliances; MFT, myofunctional therapy; ESS, Epworth Sleepiness Scale; EMG, electromyogram; STROBE, Observational Studies in Epidemiology; EMRs, electronic medical records; PSG, polysomnography; AASM, American Academy of Sleep Medicine; AHI, apnea-hypopnea index; BMI, body mass index; SD, standard deviation; IQR, interquartile range; CI, confidence interval; ANOVA, analysis of variance; OAHI, obstructive apnea hypopnea index; OSAHS, obstructive sleep apnea hypopnea syndrome; AsI, asymmetry index; MM, masseter muscles; mV, millivolt.
Availability of data and materials
The datasets generated during and/or analyzed during the current study are not publicly available, but are available from the corresponding author (Mengxing Wang) on reasonable request.
Author contributions
MXW—was involved in the conception and design, analysis and interpretation of the data; the drafting of the paper, revising it critically for intellectual content; and the final approval of the version to be published. TX, RQL—was involved in the conception and design, analysis and interpretation of the data; the drafting of the paper, revising it critically for intellectual content. And all authors agreed to be accountable for all aspects of the work.
Ethics approval and consent to participate
The ethics was approved by the Ethics Examining Committee of Human Research of Children’s Hospital affiliated to Capital Institute of Pediatrics (approval number 2023-012-17). Written informed consent was obtained from all participants.
Acknowledgment
Not applicable.
Funding
This research received no external funding.
Conflict of interest
The authors declare no conflict of interest.
