Changes in the alveolar bone thickness of maxillary incisors after orthodontic treatment involving extractions — A systematic review and meta-analysis

Background Orthodontic treatment involving en-masse retraction of incisors following premolar extractions, may induce morphological alterations of the alveolar bone surrounding the anterior teeth. Objective To assess changes in alveolar bone thickness around the incisors of extraction patients measured with CBCT. Material and Methods An electronic search was conducted in PubMed, Scopus, Embase and Cochrane Library, using search terms, with no limitation on publication date, up to April 2018. The articles selected for analysis included randomized controlled trials, case-control studies and cohort studies of patients treated with fixed appliances and premolar extractions, which had measured alveolar bone thickness with CBCT before and after treatment. Changes in bone thickness were calculated and the heterogeneity of the studies was assessed using the I2 and Cochran’s Q tests. Results Of the 136 articles identified in the initial search, 19 were related to the review subject. After removing a further 14 that did not meet the inclusion criteria, 5 articles were selected for analysis. All five were retrospective studies of medium quality. The main changes in alveolar bone thickness were found in the labial cervical third of the central incisor, presenting increases of 0.4-0.64 mm. On the palatal side the results varied considerably. Conclusions A significant increase in alveolar bone thickness occurs in the labial cervical third of the central incisor. These changes may be influenced by incisor position and inclination, the orthodontic technique and mechanics employed, the timing of the final CBCT scan and the bone remodeling capacity during en-masse retraction. Key words:Cone-beam computed tomography, alveolar bone, orthodontics.


Introduction
Cone-beam computed tomography (CBCT) provides faithful 3D images of anatomical structures, which traditional 2D imaging distorts or cannot make visible. CBCT facilitates multidisciplinary approaches to treat-ment and is used with increasing frequency in different areas of dentistry. However, based on European and American clinical practice guides (1,2), CBCT has not been recommended as a standard diagnostic and treatment planning method e77 in the field of orthodontics; such recommendations will depend on stringent analyses of benefits to the patient that prove its usefulness. Traditionally, periapical radiography, panoramic radiography and lateral cephalograms have been used to detect maxillary alveolar bone levels. But these two-dimensional radiographic methods suffer distortion, fail to show overlapping structures clearly, and make it impossible to measure alveolar bone thickness. But CBCT can be considered a suitable diagnostic method for measuring bone levels and analyzing the changes that take place during orthodontic treatment (3). Very few published studies have used CBCT to analyze changes in alveolar bone thickness resulting from orthodontic treatment. Moreover, their results have been contradictory, owing to methodological differences (patients age, sample size, follow-up periods) and to the variety of orthodontic mechanics employed and the degree of incisor proclination (4). According to Melsen and Allais (5), tooth movement takes place within a balance of bone apposition and resorption in which the tooth always remains within the bone. When this balance is upset dehiscences can occur and part of the root can become exposed. Consequently, after orthodontic treatment involving en-masse retraction of incisors following therapeutic extractions, morphometric evaluation of the alveolar bone and the roots of the anterior teeth could be a good way of studying the limitations of tooth movement in order to avoid undesirable effects such as root resorption, alveolar bone loss, dehiscences, fenestrations, and gingival recession (6). For these reasons, the aim of this systematic literature review was to examine the changes in alveolar bone thickness around the upper incisors, measured by CBCT scans of patients before and after orthodontic treatments involving upper premolar extractions.

Material and Methods
-Search Strategy A systematic review of the bibliography was conducted in accordance with PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) recommendations (7). The review protocol has been registered in the PROSPERO register (number CRD42018078114). Searches were made in the PubMed, Scopus, Embase and Cochrane Library databases, using the same search terms in each, with no limitation on publication date, up to and including April 2018.
The key words employed in the database searches combined MesH and non-MesH terms, joined by the Boolean operators AND and OR. A first search used the terms "alveolar bone thickness" OR "alveolar bone density" AND "orthodontic treatment." A second search used "alveolar bone thickness" AND "orthodontic treatment" AND "CBCT," followed by "alveolar bone thickness" AND "tooth movement" AND "CBCT." -Inclusion criteria The review accepted articles in any language. Randomized controlled trials, case-control studies, and cohort studies were included, as were both retrospective and prospective studies. Systematic reviews, meta-analyses, case reports, case series, literature reviews and editorials were excluded. The inclusion criteria were articles investigating patients in permanent dentition treated with fixed appliances, who underwent extractions and en-masse incisor retraction. A further criterion was the availability of CBCT scans measuring alveolar bone thickness on the labial and palatal sides of the incisors both before and after treatment. Articles that included patients treated by orthognathic surgery or patients with congenital disorders and/or systemic diseases were excluded. Two reviewers independently assessed the titles and abstracts of all the articles identified in the initial searches. Reviewer agreement on article selection after reading the title and abstract was evaluated using the Kappa score (8). In cases of disagreement, a third reviewer was consulted. If the abstract did not contain sufficient information to reach a decision, the reviewers read the full article before taking the final decision. When a selection had been made on the basis of titles and abstracts, the reviewers then read the full texts and recorded the reasons for rejecting any that were excluded. -Data extraction and variables analysed The following variables were analyzed in each article: type of study (prospective or retrospective), sample size (gender, age), follow-up time (from initial CBCT to final CBCT), method (treatment performed, anchorage, variables measured), results obtained, commercial name of the CBCT unit employed, which teeth were measured (maxillary incisors), conclusions, and article quality ( Table 1). -Quality assessment The quality of the articles was classified according to the CONSORT criteria as adapted by Mattos et al. (9), which have been used by several authors in other systematic reviews (10)(11)(12). This adaptation assesses 9 of the 27 CONSORT criteria, evaluating the quality of the methodology, design, execution and analysis of each article and classifying them into three levels: low, medium or high quality (Table 2). -Quantitative analysis of the studies: meta-analyses The fixed effects model was used for the studies included in each meta-analysis; when heterogeneity was detected, the random effects model was used. The significance of the meta-analyses was assessed using the Z test. Heterogeneity was measured by the I2 test, which classifies heterogeneity as mild (I2 25%-49%), moderate (50%-74%) or high (>75%) and Cochran's Q test, in e78   Table 1: Summary of articles included in the review. R: retrospective; n: number; m/f: male/female; FOV: field of view. S1: cervical segment, S2: middle segment, S3: apical segment.

CBCT Conclusions
Ahn et al., The changes in alveolar bone thickness on the labial side were not significant.
On the palatal side, the changes were significant in S1 and S3. On the labial side, there were no significant changes in bone thickness. On the palatal side, a decrease in bone thickness was observed in S1 and S2. Labial alveolar bone thickness in S3 and total alveolar bone thickness in S1 both increased significantly.
which a threshold p-value of 0.1 was considered statistically significant.
For changes in bone thickness, the estimated effect size was the difference in means between the initial and final stages. The publication bias was measured by Rosenthal's fail-safe number, which estimates the number of studies that would be required for a meta-analysis with a significant result (p<0.05) to cease to be significant. Egger's regression intercept and its p-value were also determined. The meta-analyses were performed with Comprehensive Meta-analysis V.3 (Biostat, Inc) software.

-Selection of articles and flow diagram
The initial database search identified 136 articles: 33 in Pubmed, 46 in Embase, 49 in Scopus, and 8 in the Cochrane Library. Of these, 98 were found to be duplicates, leaving 38. When the titles and abstracts were read, 19 works were found to deal with the subject under review. Inter-assessor agreement obtained a Kappa score of 0.86. After critical reading of the full text, 14 out of the19 articles were rejected, as they did not meet the inclusion criteria. Finally, a total of 5 articles were included in this systematic review and 4 in the meta-analysis (Fig. 1).
-Qualitative synthesis All five articles were retrospective, and of moderate quality according to the criteria proposed by Mattos et al. (9). Four concerned Angle class I patients with double protrusion who underwent extraction of the four first premolars (6,13,14,15), while the remaining work stu-  Table 2: Quality assessment of the articles included in the review.  Table 1). Three of the studies had been approved by an ethics committee (13)(14)(15). The other two articles did not mention approval (3,6). The method used to measure alveolar bone thickness was the same in all five studies. Labial (Vb) and palatal (Pt) thicknesses were measured from CBCT images, dividing the root with parallel lines at 3 mm intervals from the cemento-enamel junction to the apex. In this way, the measurements were made in the cervical, middle and apical thirds and in the total tooth below the cemento-enamel junction (3,6,13-15). Ahn et al. (13) observed that bone thickness on the labial side increased in the middle third, by 0.27 mm for upper central incisors and by 0.65 mm for lateral incisors, with statistically significant differences (p<0.01), but decreased significantly on the palatal side at all levels. These authors did not relate bone thickness to the inclination or position of the maxillary incisors. But the other authors did identify an association. Picanço et al. (3), who divided the patients into two groups, found that in the group that underwent extraction of the maxillary first premolars (G1) and in the control group with no extractions (G2), the G1 group patients showed greater retraction of the maxillary incisor and a more vertical position, while the G2 patients showed greater labialization and protrusion of the incisors. The bone thickness in the labial cervical third was greater in G1 patients than in G2. Yodthong et al. (14) found that changes in the group of patients treated by retraction with tipping were greater in the palatal cervical third of the incisors (r=0.6; p=0.006), while the changes in alveolar bone thickness were more negative in the group treated by retraction with torque (r=-0.3; p=0.031). Sariyaka et al. (6) found that the change in maxillary labial bone thickness was not statistically significant. The width of the bone labial to the maxillary left lateral incisor decreased significantly in the middle segment (S2) (p<0.05). Regarding the upper bone thickness lingual to the incisors, the apical segment (S3) measurements showed minimal change, but the measurements at the cervical (S1) and middle (S2) segment levels differed significantly over time. -Quantitative synthesis Quantitative analysis compared changes in bone thickness at the level of the three labial and three palatal segments (S1 cervical, S2 middle, S3 apical in each case) of the upper central incisor, as this was the variable that the four studies had in common. Six meta-analyses were performed, one each for the three labial and three palatal upper central incisor segments.

Discussion
Few studies have investigated the use of CBCT to register the changes in alveolar bone thickness around the incisors that take place in cases treated with extractions and en-masse retraction of the incisors. One reason for this is the ethical problem of radiation exposure, as this requires submitting the patient to two radiographic examinations; another is the methodological diversity between the few studies that have been published. Work is currently underway to reduce the radiation emitted by CBCT machines, as many of the patients examined by this method are children, who are more susceptible to the harmful effects of radiation. For this reason, it is important to maintain an optimal balance between the need for adequate image quality and radiation dose. Although CBCT scanners can capture a precise 3D image of the dentoalveolar complex, it is important to select cases that really will benefit from CBCT examination and to assess borderline cases carefully. These cases include alveolar bone phenotypes that are clinically too narrow to accommodate labio-lingual displacement; patients with periodontal disease; cases that require tooth movement beyond the alveolar limits, including borderline cases that require a decision as to whether or not to extract teeth; and cases of transposed or impacted teeth (16 (17) showed that quantitative assessment of alveolar cortical bone using computed tomography (CT) is feasible above a minimum bone thickness of 0.5 mm, and obtained results that were statistically similar to histological measurements. Numerous authors have found that palatal movement of the incisors narrows the alveolar bone on the palatal side (6,15,18). Some authors have even found a 1 mm reduction in alveolar bone thickness between pre-treatment and post-treatment me-e83 asurements (15 (6). These differences could lead to differences in the end results, as some authors, such as Vardimon et al. (18), have argued that the cortical plate of alveolar bone can undergo remodeling during treatment, changing its shape and position. This contrasts with the hypothesis put forward by Handelman, who argued that there are limitations on tooth movement caused by the cortical plates and showed that bone remodeling is possible during tooth movement, induced by biological forces (19). Dentoalveolar anatomy establishes the limits of orthodontic tooth movement, and the bone's capacity for adaptation during tooth movement, as well as its morphology once the teeth have reached their final position (16). In our opinion, determining whether bone accompanies teeth during retraction and whether the bone is capable of remodeling or not (with or without the possible undesirable results) justifies using CBCT both before and after orthodontic treatment in cases treated with extractions and incisor retraction. The results demonstrated that palatal movements of the maxillary incisors reduced the palatal alveolar bone. This finding disagrees with De Angelis (20), who claimed that alveolar bone has a bending capacity. In the present study, the maxillary bone thickness did not remain the same but decreased. This finding of reduced alveolar bone thickness in the direction of tooth movement agrees with the results obtained by Wainwright (21), Vardimon et al. (18) and Wehrbein et al. (22). Only one of the controlled studies divided the sample into differentiated groups: one group in which treatment included extracting first premolars, and the control group, in which extractions were not performed (3). Further research is needed with longitudinal controlled and blinded studies to observe cases from the start of treatment and examine changes in the bone before and after en-masse retraction of the incisors. Cases must be selected to form groups with similar characteristics in terms of malocclusion, anteroposterior position, and inclination of the incisors, crowding, and the mechanics used. The present review suffered some limitations: the small number of controlled studies; their small sample sizes (only 10-25 patients); and the varying methods used to measure bone thickness.

Conclusions
Despite the methodological variations between the studies reviewed, it may be stated that a significant increase in alveolar bone thickness occurs in the cervical third on the labial side of the central incisor after orthodontic treatment involving extractions. On the palatal side, the findings vary. These changes may be influenced by factors such as incisor position and inclination before and after treatment, the technique and mechanics employed, the timing of the final CBCT scan, and the bone's capacity for remodeling during en-masse retraction of the incisors.