In this systematic review, we identified and qualified the evidence of long-term reports on the effects of rapid maxillary expansion (RME) on airway dimensions and functions.
Electronic databases (Ovid, Scirus, Scopus, Virtual Health Library, and Cochrane Library) were searched from 1900 to September 2010. Clinical trials that assessed airway changes at least 6 months after RME in growing children with rhinomanometry, acoustic rhinometry, computed tomography, or posteroanterior and lateral radiographs were selected. Studies that used surgically assisted RME and evaluated other simultaneous treatments during expansion, systemically compromised subjects, or cleft patients were excluded. A methodologic-quality scoring process was used to identify which studies would be most valuable.
Fifteen articles fulfilled the inclusion criteria, and full texts were assessed. Three were excluded, and 12 were assessed for eligibility. Four articles with low methodologic quality were not considered. The remaining 8 were qualified as moderate. The posteroanterior radiographs showed that nasal cavity width increases; in the lateral radiographs, decreased craniocervical angulation was associated with increases of posterior nasal space. Cone-beam computed tomography did not show significant increases of nasal cavity volume. Rhinomanometry showed reduction of nasal airway resistance and increase of total nasal flow, and acoustic rhinometry detected increases of minimal cross-sectional area and nasal cavity volume.
There is moderate evidence that changes after RME in growing children improve the conditions for nasal breathing and the results can be expected to be stable for at least 11 months after therapy.
Rapid maxillary expansion (RME) is an effective orthopedic procedure that has been routinely used in growing patients in orthodontics. The goal of RME is to open the midpalatal suture, providing correct and stable maxillary width. Although this therapy is carried out to correct dental and skeletal maxillary transverse discrepancies, some authors showed that treatment outcomes could also increase nasopharyngeal airway dimensions and improve patients’ nasal breathing.
It has been hypothesized that, since the maxillary bones form half of the nasal cavity’s structures, when the midpalatal suture is open, the nasal cavity’s lateral walls are also displaced apart, and its volume increases, and upper airway resistance decreases over time. Head posture had also been associated with respiratory function, and increased craniocervical angulation was observed as a functional response to facilitate oral breathing to compensate for nasal obstruction. Once RME results in increased nasal airway patency and reduced nasal airway resistance (NAR), the airway flow increases, and the craniocervical angulation consequently is reduced. Another reported consequence after RME is higher tongue repositioning, which could increase airway volume. Several studies have been conducted to evaluate these effects, but there is still controversy about the authentic long-term influence of RME on nasal cavity dimensions and functions.
Differing measurement methods of nasal airway dimensions and function have been proposed and used, such as rhinomanometry (RMN), acoustic rhinometry (AR), radiography, and, recently, cone-beam computed tomography (CBCT). Each technique has its strengths and limitations. Both RMN and AR are objective tests for the assessment of nasal airway patency. RMN measures air pressure and airflow rate during breathing, calculating NAR, whereas AR uses a reflected sound signal to measure the cross-sectional area and nasal passage volume. Cephalometric radiographs are routinely used for orthodontic treatment evaluation. With posteroanterior (PA) headfilms, it is possible to measure nasal cavity width and, with lateral headfilms, airway length and craniocervical angulation. CBCT technology allows segmentation and visualization of the hollow airway in 3 dimensions and determines, in addition to lengths and angles, the airway volume and surface area.
A maxillary transverse deficiency is a common skeletal problem in the craniofacial region, and it is often found in children with abnormal breathing. Scientific evidence on the nasal airway would augment orthodontists’ information to patients that RME could not only produce dentoalveolar changes, but also have implications for the nasal complex. Previously, a meta-analysis and a systematic review on the skeletal effects after RME found a significant increase in nasal cavity width. However, none of these studies aimed to associate this skeletal change with breathing function. Recently, another systematic review evaluated airway changes with AR but did not confirm the clinical breathing benefit after the therapy. This study was not restricted to orthopedic expansion but included studies evaluating children and adults. Also, the follow-up period was not considered. So, there is still no evidence that children having orthopedic expansion can obtain any breathing benefit after a follow-up period.
The aim of our systematic review was to identify and qualify the evidence of long-term reports evaluating changes in airway dimensions and functions in patients having RME during the growth period. Studies using RMN, AR, radiography, and CBCT were considered for this purpose. The focused questions were the following. What are the effects on airway, nasal cavity, and NAR in children who underwent RME therapy? Are these changes stable in the long term? Do children undergoing RME therapy to correct a transverse discrepancy have any long-term benefit in breathing function?
Materials and methods
The method for this systematic review was based on the PRISMA guidelines ( www.prisma-statement.org ) recommended in the American Journal of Orthodontics and Dentofacial Orthopedics . To identify relevant studies (from 1900 to the third week of September 2010), irrespective of language, a detailed search was conducted in the following electronic databases: Ovid Medline, Scirus, Scopus, Virtual Health Library, and Cochrane Library. The search strategy included appropriate changes in the key words and followed each database’s syntax rules ( Table I ).
|Rapid maxillary expansion OR rapid palatal expansion OR maxillary disjunction OR palatal disjunction OR Palatal Expansion Technique AND airway OR nasal OR respiration OR breathing||1950 to September, week 3, 2010
|Scirus (Medline/PubMed; Science Direct; PubMed Central; BioMed)
|“Rapid maxillary expansion” OR “rapid palatal expansion” OR “maxillary disjunction” OR “palatal disjunction” OR “Palatal Expansion Technique” AND “oropharyngeal airway” OR “nasal airway” OR “nasal cavity” OR “nasal volume” OR “respiration” OR “breathing”||1900-2011
Information type-abstracts, articles
|Rapid maxillary expansion OR rapid palatal expansion OR maxillary disjunction OR palatal disjunction OR Palatal Expansion Technique AND airway OR nasal OR respiration OR breathing||Articles, title, abstract, keyword
All years to present
|VHL (LILACS, IBECS, Medline, Scielo)
|Palatal Expansion Technique AND Respiration (MeSH)||–|
|Rapid and maxillary and expansion and nasal||–|
Table II outlines the populations, interventions, comparisons, and outcomes (PICO format) and the null hypothesis used for this systematic review. For the full articles to be selected from the abstracts, they had to satisfy the following inclusion criteria: human controlled clinical trial; follow-up of at least 6 months after RME therapy; subjects during their growth period; and the use of RMN, AR, radiography, or CBCT to measure airway differences. The exclusion criteria were surgical or other simultaneous treatment during the active expansion phase; surgical treatment that could affect RME effects during the evaluation period; and systemically compromised subjects or cleft patients used as subjects.
|Population||Subjects during growth period with transverse maxillary deficiency|
|Comparison||Paired age and sex subjects who did not undergo to RME therapy|
|Outcome||Changes in airway dimension or function|
|There was no long-term difference in airway changes between subjects who had RME and those who did not|
The initial selection excluded all titles and abstracts not related to the topic or that involved any exclusion criteria. Theses, annals, reviews, and case reports were also excluded. The next step was a detailed review of the selected abstracts to screen those that respected all inclusion and exclusion criteria. Two researchers (C.B. and M.A.Jr.) made independent selections, and their results were compared to identify discrepancies. If the abstract contained insufficient information for a decision of inclusion or exclusion, the full article was obtained and reviewed before a final decision. Titles with no abstract available that suggested a relationship to the objectives of this review were selected to screen the full text. The reference lists of the retrieved articles were also hand searched for additional relevant publications that could have been missed in the databases.
A methodologic-quality scoring process was used to identify which selected studies would be most valuable. Our scoring process was a modified version of one previously used in a systematic review by Lagravère et al. The full texts of articles selected for eligibility were assessed on the basis of study design, study measurements, and statistical analyses ( Table III ). When the article fulfilled satisfactorily 1 methodologic criterion, the maximum of the point was checked (1 or 2); when it partially fulfilled the criterion, half of a pointwas checked; and when it did not fulfill the methodologic criterion, 0 was checked. Before the assessment of the studies, 2 researchers (C.B. and M.A.Jr.) discussed all the criteria analyzed to reach consensus about their content. The most ambiguous topic was to define adequate descriptions of the population (item A, Table III ) and the RME therapy (item G, Table III ). In these 2 items, the criterion was considered fulfilled when all 3 selected aspects were described, and fulfilled partially when only 2 were described. No point was checked when just 1 aspect was described. Item C (sample size) was the only criterion with a maximum score of 2 points. It was scored as 2 points when both groups, treated and control, were larger than or equal to 30 subjects; when just 1 group was larger than or equal to 30 and the other was more than 20 and less than 30 subjects, 1.5 points was scored. When the sample was larger than or equal to 20 and less than 30 for both groups, 1 point was scored. When 1 group had less than 20 subjects, 0.5 point was scored. The methodologic-quality assessment scores ranged from 0 to 14 points. Studies were qualified as having high (score, >12), moderate (scores, ≥7 and ≤12), or low (score, < 7) methodologic quality.
|I. Study design (8)|
|A. Population adequately described (age, sex, brief medical history) (1)|
|B. Selection criteria described (1)|
|C. Sample size: ≥20/group (1) or ≥30/group (2)|
|D. Control with no orthodontic treatment (1)|
|E. Timing prospective (1)|
|F. Randomization stated (1)|
|G. RME adequately described (appliance, activation, retention) (1)|
|II. Study measurements (3)|
|H. Measurement method appropriate to the article objective (1)|
|I. Blinding: examiner and statistician (1)|
|J. Reliability described and adequate (1)|
|III. Statistical analysis (3)|
|K. Statistical test appropriate for data ( 1 )|
|L. Confounders stated: radiography and CT evaluation (standardization of head and tongue position); RMN and AR (use of nasal decongestant) (1)|
|M. Significance: P value stated and confidence intervals provided ( 1 )|
A total of 232 titles or abstracts were identified in the electronic databases used ( Fig ). Duplicate records appearing in more than 1 database search were considered only once. From the titles, we excluded all records not related to the review topic, that used surgically assisted RME therapy, that evaluated other simultaneous treatments during expansion; that were not human studies, that evaluated systemically compromised subjects or cleft patients; and theses, annals, reviews, and case reports. From the 100 abstracts left, 15 fulfilled the inclusion criteria, and the full texts were assessed. Furthermore, 3 articles, those by Tecco et al, Franchi et al, and Hartgerink and Vig, were excluded because the same data were reported in the finally selected articles: Tecco et al, Cameron et al, and Hartgerink et al, respectively. The articles by Baccetti et al and Cameron et al also used the same sample of subjects, but both studies were included because the authors ued different analyses. Then, 12 were assessed for eligibility ( Table IV ) and qualified according to Table III . From these articles, 4 had low methodologic quality and were not considered. Only 8 articles fulfilled all selection criteria and had adequate evidence to be considered in this systematic review.
|McGuinness and McDonald||1||1||2||1||1||0||0.5||1||0||1||1||0||1||10.5||Moderate|
|Monini et al||0.5||1||2||1||1||0||1||1||0||0||1||0.5||1||10||Moderate|
|Tecco et al||1||1||1||1||1||1||1||1||0||0.5||1||0||0||9.5||Moderate|
|Baccetti et al||0.5||1||1.5||1||0||0||1||1||0||1||1||1||0.5||9.5||Moderate|
|Cameron et al||0.5||1||1.5||1||0||0||1||1||0||1||1||1||0.5||9.5||Moderate|
|Compradretti et al||1||1||1||1||1||0||1||1||0||0||0.5||0.5||0.5||8.5||Moderate|
|Zhao et al||1||1||1||0||0||0||1||1||0.5||1||1||0||0.5||8||Moderate|
|De Felippe et al||1||1||1||1||0||0||0||1||0||1||0.5||0||0.5||7||Moderate|
|Altug-Atac et al||1||1||0||1||0||0||0.5||1||0||1||1||0||0||6.5||Low|
|Chiari et al||1||1||0||1||1||0||1||1||0||0||0.5||0||0||6.5||Low|
|Warren et al||0.5||0||1||0||1||0||0.5||1||0||0||1||0||0||5||Low|
|Hartgerink et al||0||0||1.5||0||1||0||0||1||0||0||1||0||0||4.5||Low|
As shown in Table IV , all studies included were classified as having moderate evidence. None met all requirements in our specific methodologic scoring protocol. Only Tecco et al stated the randomization of their sample. Blinding of the statistician was not reported in any studies, and only Zhao et al had the examiner blinded. Samples larger than 30 children per group, treated and control, were used by McGuinness and McDonald and Monini et al. Baccetti et al and Cameron et al evaluated more than 30 children only in their treated groups and had 20 children in the groups without treatment; both scored 1.5 points for the sample-size requirement.
A summary of the participants, interventions, comparisons, and study design characteristics from each included study in the qualitative synthesis is shown in Table V . When specific data were necessary that were not specified in the article, the authors were contacted to obtain the required additional information.
|Authors||Participants||Intervention||Comparison (Subjects with no RME therapy)||Study design|
|Total (female/male)||Mean age (range)||Brief medical history||RME
(expander type, activation protocol, retention time)
|Total (female/male)||Mean age (range)||Brief medical history||Follow-up (mean)||Evaluation method|
|McGuinness and McDonald||39 (23/16)||10-16 years||General and dental health, no nasal obstruction history||Bonded expander; 2 turns per day (mean 3 weeks); no time retention given + orthodontic treatment||36 (24/12)||10-16 years||No orthodontic treatment||12 months||Lateral radiograph|
|Monini et al||38||7.85 years
|Primary snoring and nasal respiratory obstruction||Hyrax; 1 or 2 turns per day until overcorrection; 1 year||50||Aimilar||Without nasal abnormalities and pathologic occlusions||12 months||RNM
|Tecco et al ∗||23
|Reduced nasopharyngeal airway adeqacy (cephalometrically) and mouth breathing||R.E.P. (Dentaurum Italia s.r.l.); 4 turns (0.8 mm) on the first day followed by 2 turns per day until the required expansion was achieved; mean, 4.7 months||22 (only female)||8.1 years
|Reduced nasopharyngeal airway adeqacy (cephalometrically) and mouth breathing; no orthodontic treatment||12 months||Lateral radiograph|
|Cameron et al||42 (25/17)||11.8 years||Not reported||Haas; 2 turns a day (0.5 mm) until 10.5 mm (mean, 3 weeks); mean 2 months + fixed orthodontic appliance||20 (9/11)||11.8 years||No orthodontic treatment||At least 5 years||PA
|Baccetti et al †||ETG-29 (18/11)
|ETG -11 years
|Not reported||Haas; 2 turns a day (0.5 mm) until 10.5 mm (mean, 3 weeks); mean, 2 months + fixed orthodontic appliance||ECG-11 (2/9)
LCG- 9 (7/2)
|No orthodontic treatment||At least 5 years||PA
|Compradretti et al||27 (14/13)||9.5 years
|Presenting maxillary constriction and any cause for nasal obstruction was excluded||Hyrax; 2 turns a day (0.5 mm) during 2 weeks; 3 months||24 (16/8)||10.2 years
|Presenting maxillary constriction and any cause for nasal obstruction was excluded||11 months||RNM
|Zhao et al||24 (18/6)||12.8 years
|Healthy||Hyrax; 1 or 2 turns per day until the required expansion with slight overcorrection; at least 3 months + fixed orthodontic appliance||24 (18/6)||12.8 years (8.6-15.8)||Orthodontic treatment||15 months||CBCT|
|De Felippe et al||25 (14/11)||Initial: 13.16 years
Final: 18.28 years
|No history of nasal congestion and infection or cold during evaluation||Haas, hyrax, and bonded; 2 turns per day (50% of the sample), 1 turn per day (42%), and 1 turn every other day (8%), until 2 to 3 mm of overexpansion wasachieved; 3-6 months + full orthodontic treatment (95%)||25 (14/11)||Only compared the final data: 18.48 years (12-22)||No history of nasal congestion and infection or cold during evaluation; no orthodontic treatment||60 months||AR|
† Same sample used by Cameron et al however, the treated group was divided into early treated group ( ETG ) and late treated group ( LTG ), and the control group was divided into early control group ( ECG ) and late control group ( LCG ).