Bone-Borne Maxillary Expansion and Traditional Rapid Maxillary Expansion: A Systematic Review and Meta-Analysis

Introduction

Transverse maxillary deficiency is one of the common orthodontic problems that is usually accompanied by posterior cross-bite and dental crowding (1, 2).To eliminate posterior cross-bite and increase arch perimeter maxillary expansion should be considered(3). Various methods have been used for resolving this problem in adolescents. The most common method was performed by a hyrax (i.e., hygienic rapid palatal expander appliance) (Tooth-anchored)(4, 5).However, with this tooth-anchored device, the resulted expansion is both dental and skeletal. Other disadvantages include the risk of root resorption, gingival recession, and fenestrations (6, 7, 8).

An alternative method is bone-borne appliances. These appliances are anchored from the palatal surfaces of maxilla with Temporary Anchorage Devices (TADs). It has been claimed that these appliances caused less buccal tipping and more skeletal effects (9-13) however, there are other studies that claimed there is no difference between bone-borne and tooth-borne appliances (14-17). The aim of this study was to perform a systematic review and meta-analysis to compare the effects of bone-borne expansion (BBE) and tooth-borne expansion (TBE).

Methods and Materials

We conduct this systematic review and meta-analysis in line with the guidelines of the Cochrane Handbook for Systematic Reviews of Interventions (version 5.1.0) and PRISMA statement. Studies about different appliances for the treatment of posterior cross-bite were searched through different electronic databases, such as MEDLINE, EMBASE, and Cochrane Central Register of Controlled Trials. An electronic search for IADR conference proceedings was also conducted. The time frame for studies was from the beginning to February 2019. The search strategy was adjusted for different databases (Table 2). A hand search was also performed on eligible article reference lists for related articles. Any non-English articles were translated, and their data were extracted. Any data not mentioned in the article was obtained from the authors or the article was excluded from the review.

Selection of studies

The relevant clinical trials and cohort studies were selected with the following PICO format:

related human clinical trials and cohort studies, studies on patients with maxillary constriction (i.e., population), orthodontic treatment conducted with BBE appliance (i.e., intervention), orthodontic treatment conducted with a TBE appliance (i.e., comparison), results analyzed using cone-beam computed tomography image and cast analyses before and after treatment (i.e., outcomes measured), analyzed treatment effects not confounded by additional and concomitant procedures (e.g., surgically-assisted rapid palatal expansion, extractions, and fixed appliances)

The first duplicate articles were excluded from the study. Articles were excluded if they did not meet the inclusion criteria or if they were abstracts, laboratory studies, descriptive studies, case reports, case series, and reviews or meta-analyses. Title and abstract screening were performed by two authors (E.B. and H.S.). Full texts and abstracts of eligible studies were also individually reviewed by them. In addition, any disagreement was resolved by another author (A.M.). The studies with a risk of bias were assessed based on the study design using the Cochrane risk of bias tool for randomized clinical trials (RCTs) or Downs and Black scale for controlled clinical trials (CCTs).

The main outcome measures (i.e., primary dependent variables) were the changes in the transverse dimension based on different landmarks (e.g., apex of teeth, crown of teeth, alveolar bone in different regions, and skeletal changes of maxilla or nasal bone) after treatment using bone-borne or tooth-borne appliances. The secondary outcomes were the changes in the angulation of teeth after treatment. Because of the differences in the treatment modalities, age group, study design, and gender of patients in the base studies, the standardized mean difference and random-effects meta-analysis was performed and if it was possible, subgroup analysis based on the age group, gender, and study design was performed so that the effects of these confounding factors on the outcome measures could be assessed.

Results

Search strategy, date of search, and number of relevant articles for each database can be seen in Table 2. After removing duplicates, 186 articles remained out of 304 initial articles. A total of 176 articles were excluded on the basis of title and abstract. Out of the 10 remaining 10 articles, 6 papers were excluded after the evaluation of their full texts (Figure 1). Finally, four studies were selected for meta-analysis (11-14).

Table I. Characteristics of the studies considered for the final analysis

Figure 1. PRISMA flowchart for the evaluated studies

Study characteristics

Study characteristics of the final four trials included in the meta-analysis are summarized in Table I. All selected clinical trials examined treatment results between bone-borne and tooth-borne maxillary expansion appliances in patients who needed maxillary expansion. All the studies were performed in a university setting or an educational institution. The studies consisted of three prospective studies, including one RCT (14),two CCTs (11, 12),and one cohort study (15). The total number of patients treated with BBE appliance was 60, whereas the overall TBE appliance sample consisted of 57 untreated subjects. Two of studies included both male and female participants (13, 14), and the other two included only girls (11, 12).

There was an age discrepancy between the included studies; 3 trials (12-14) had samples with a mean age of 12 to 14 years, and 1 trial (11)included older patients with a mean age of 17 to 18 years. The comparison between appliances was homogeneous among the selected studies: Tooth-borne expander (i.e., banded hyrax with bands on the first premolar and molars) and bone-borne expander (i.e., supported by four TADs in canine-first premolar area and first molar-second premolar area). In one study (13), there was an additional bonded hyrax group, and in another study (14),there was a control group without using any expanders not included in the meta-analysis.

Risk of bias assessment

The only RCT (14) had a low risk of bias (Table 2). The risks of bias in other studies (11-13)were scored with medium quality, with the score of 15, 18 and 18 of 32 respectively (Table 2), according to the Downs and Black scale. Because of the low number of studies that were included in our study (i.e., lower than 10), the assessment of publication bias using the visual inspection of the funnel plots could not be performed. In addition, the results of the regression test of Egger et al. showed P-values of 0.497 for transverse expansion of alveolar bone in the molar area.

Quantitative data synthesis

After performing the meta-analysis on four studies consisting of 57 patients treated with TBE appliance and 60 patients treated with BBE devices, the standardized mean difference for the treatment effects of these devices on transverse change of molar apices was -0.14 mm (95% CI, -0.55 mm to 0.28 mm; P=0.53; I²=0%) (Figure 2.A). Also, the standardized mean difference for the transverse changes after treatment for molar crowns was 0.35 mm (95% CI, -0.29 mm to 0.99 mm; P=0.29; I²=65%). The standardized mean difference for premolar apices was 0.22 mm (95% CI, -0.49 mm to 0.92 mm; P=0.55; I²=62%) (Figure 2.B). Moreover, for the premolar crowns, the standardized mean difference was 0.63 mm (95% CI, -0.27 mm to 1.53 mm; P=0.17; I²=74%).

Data synthesis of two clinical trials with 35 bone-borne and 34 tooth-borne appliances showed that the treatment effect of tooth-borne appliances versus bone-borne appliances for the transverse change of incisor crowns was -0.16 mm (95% CI, -0.64 mm to 0.31 mm; P=0.50; I²=0%). The meta-analysis also demonstrated that in 43 tooth-borne and 46 bone-borne patients in three included trials, the standardized mean difference between the treatment effect of tooth-borne appliance versus bone-borne appliance was 0.04 mm (95% CI, -0.93 mm to 1.01 mm; P=0.94; I²=79%) for transverse skeletal change in alveolar bone in premolar area (Figure 2.C).

The standardized mean difference for transverse skeletal change in alveolar bone in the molar area between TBE appliance and BBE appliance patients was 0.05 mm (95% CI, -0.60 mm to 0.70 mm; P=0.88; I²=66%). This finding was obtained from four trials that were analyzed with 60 BBE and 57 TBE patients (Figure 2.D). The analysis of treatment effects on 37 TBE versus 39 BBE patients from three trials was -0.20 mm (95% CI, -0.68 mm to 0.28 mm; P<0.001; I²=86%) for transverse skeletal change in nasal bone. Regarding the maxillary transverse expansion after the treatment with RPEs, there were no significant standardized mean differences between both groups. The standardized mean difference was -0.05 mm (95% CI, -0.47 mm to 0.37 mm; P=0.81; I²=0%) from analyzing three trials with 44 patients treated with tooth-borne and 45 subjects treated with bone-borne appliances.

Figure 2. Forest plot depicting the standardized mean difference between the effects of tooth-borne expansion versus bone-borne expansion; A) mean difference for transverse expansion of molar crowns; B) mean difference for transverse expansion of premolar crowns; C) mean difference of the alveolar bone in the premolar area; D) mean difference of the alveolar bone in the molar area

Based on the synthesis of three trials with 43 tooth-borne and 46 bone-borne appliance patients, the standardized mean difference of TBE group relative to the BBE group for the dental angulation of left molars was 0.80 degrees (95% CI, 0.01 mm to 1.59 mm; P=0.05; I²=67%) (Figure 3.A); for the dental angulation of right molars, it was 0.65 degrees (95% CI, -0.67 mm to 1.94 mm; P=0.33; I²=88%); for the dental angulation of left premolars, it was 1.69 degrees (95% CI, 1.19 mm to 2.18 mm; P<0.001; I²=0%) (Figure 3.B); for the dental angulation of right premolars, it was 1.01 degrees (95% CI, 0.11 mm to 1.90 mm; P=0.03; I²=72%).

Because of the differences in the study designs of the included studies, a sensitivity analysis was conducted based on the study design (i.e., randomized and non-randomized trials). The sensitivity analysis showed no significant differences among the RCT, prospective CCT, and cohort studies regarding the transverse expansion in molars (Chi-square, 0.97; df=1; P=033; I²=0%) (Fig. 4).

Figure 3. Forest plot depicting the standardized mean difference between the effects of tooth-borne expansion versus bone-borne expansion on dental angulation; A) mean difference for the tipping of the left molars; B) mean difference for the tipping of the left premolars

Figure 4. Sensitivity analysis comparing the standardized mean difference of bone-borne expansion and tooth-borne expansion on the molar crown expansion of randomized clinical trial, prospective controlled clinical trials, and cohort studies