INTRODUCTION

There are two concepts in orthodontic tooth movement in terms of alveolar bone remodeling. If the alveolar bone is remodeled with coordination of resorption and apposition, tooth movement and bone remodeling occur at a 1∶1 ratio, and the tooth remains in the alveolar housing. This kind of tooth movement is known as “with-the-bone.”1 However, if balance between resorption and apposition of the alveolar bone is not established during tooth movement, the tooth will move out of the alveolar housing, which is referred to as “through-the-bone.”2

Bialveolar protrusion is one of the most common chief complaints in Asian orthodontic patients. The conventional treatment modality is extraction of the first premolars and retraction of the anterior teeth with maximum anchorage.3,4 However, excessive retraction of the anterior teeth may result in iatrogenic sequelae such as root resorption, alveolar bone loss, dehiscence, fenestration, and gingival recession.5–8 Therefore, morphometric evaluation of the alveolar bone and roots of the anterior teeth after en masse retraction may be a good model with which to explain the therapeutic limitation of orthodontic tooth movement and to define the “with-the-bone” and “through-the-bone” concepts.

Conventional two-dimensional (2D) lateral cephalograms have several limitations in terms of investigating the changes in the alveolar bone and roots, especially in the anterior region, as a result of the midsagittal projection.5,9–11 The advent of cone-beam computed tomography (CBCT) has made it possible to qualitatively and quantitatively evaluate the height and thickness of the alveolar bone and the length and thickness of the root.8,12–15

To the authors' knowledge, few studies have investigated the alveolar bone and root of the individual teeth according to a customized reference plane using CBCT images. Therefore, the purpose of this study was to evaluate the morphometric changes in the alveolar bone and roots of the maxillary anterior teeth (MXAT) after en masse retraction with maximum anchorage (EMR-MA) using superimposition of individual teeth with CBCT images.

MATERIALS AND METHODS

This retrospective study was performed under the approval of the Institutional Review Board (IRB) of KyungHee University Dental Hospital (IRB No.: KHD IRB-1108-04). The samples comprised 37 female adult patients with Class I dentoalveolar protrusion (mean age  =  26.6 ± 8.5 years; treatment duration  =  1.8 ± 0.4 years; SNA, 81.4°; SNB, 77.8°; maxillary central incisor [MXCI] to SN, 127.7°; IMPA, 98.9°; Table 1). These patients were treated by a single orthodontist with extraction of the four first premolars and sliding mechanics using EMR-MA. Working wire was 0.019 × 0.025–inch stainless-steel wire without extra torque, and a 0.022-inch straight-wire appliance with Roth setup (Clippy-C, Tomy, Futaba, Fukushima, Japan) was used. Conventional anchorage such as TPA and/or headgear and elastic chains with a force of 200 g was used. The amount of retraction of MXCI according to every 1 mm of anchorage loss of the maxillary first molar (MX1 to MX6 ratio) was 3.67 ± 0.58 mm (Table 1).

Table 1.  Cephalometric Characteristics of the Samples (N  =  37,Females)^a

View Fullscreen

The inclusion criteria for sampling were as follows: skeletal and dental Class I relationships, mild anterior crowding (arch length discrepancy of ≤3mm), labioversion of MXCI (MXCI-SN > 120°), and controlled tipping movement of the MXCI from superimposition of the lateral cephalograms before treatment (T1) and after space closure (T2) (Table 1; Figure 1). The exclusion criteria included tooth size anomaly, periodontal problems, spacing or moderate to severe crowding, and root resorption of the MXAT before orthodontic treatment.

View Figure

• Download as Powerpoint
• Download Figure

To evaluate the alveolar bone accurately, CBCT images were taken at the T1 and T2 stages (Implagraphy, Vatech, Seoul, Korea; 12 × 9–cm field of view, 90-kVp tube voltage, 4.0-mA tube current, 0.2-mm isotropic voxel size, and 24-second scan time). The obtained data were analyzed by InVivoDental (Anatomage, San Jose, Calif). To set an identical reference plane in the T1 and T2 stages, three-dimensional (3D) superimposition of the maxillary central incisors (MXCI, N  =  66), maxillary lateral incisors (MXLI, N  =  69), and maxillary canines (MXC, N  =  69) was performed as follows (Figure 2). First, the long axis of each anterior tooth was set on the sagittal image at the T1 stage. Next, 3D superimposition was performed by the best-fit method using two sets of homologous landmarks in each CBCT image and a manual refinement process. Then the T2 image was reoriented on the same coordinate axis as the T1 image.

View Figure

• Download as Powerpoint
• Download Figure

Definitions of landmarks, reference planes, and variables are given in Figure 3. The alveolar bone area (ABA) was measured at the cervical, middle, and apical levels, respectively. Trisection of the root length into the cervical, middle, and apex levels was duplicated in the T2 stage to guarantee the same slice levels as those in the T1 stage. Vertical bone level (VBL), root length (RL), root area (RA), and prevalence of dehiscence (PD) were measured both on the palatal and labial sides. The percentage of VBL to RL, the ratio of ABA change, and the ratio of palatal to labial ABA were calculated.

View Figure

• Download as Powerpoint
• Download Figure

All of the measurements were repeated by the same operator after 2 weeks. The difference ranged from 0.27 mm to 0.35 mm for linear measurements, from 0.25° to 0.47° for angular measurements, and from 0.31 mm² to 0.48 mm² for area measurements, according to Dahlberg's formula.16 The mean of the two measurements was used for this study. Independent t-test, paired t-test, one-way analysis of variance (ANOVA) with Duncan's multiple comparison test, and cross-tab analysis were performed for statistical analyses.

RESULTS

Amount and Ratio of Changes in the Labial and Palatal ABA (Table 2)

Table 2.  Comparison of the Amounts of the Alveolar Bone Area Between the Pretreatment (T1) and Postretraction (T2) Stages in Each Tooth and the Amounts of Change during the T1 and T2 Stages and Ratio Among the Maxillary Anterior Teeth

View Fullscreen

Table 2.  Extended

View Fullscreen

On the labial side, ABA increased in the middle level of MXCI (P < .001) and in the middle and total levels of MXLI (P < .001 and P < .05, respectively). However, on the palatal side, ABA decreased in all levels of MXCI (all P < .001), MXLI (all P < .001), and MXC (cervical, apical, total, P < .001; middle, P < .01).

Although MXCI and MXLI showed a significant decrease in the ratio of change in ABA (ΔABA ratio) compared to MXC on the palatal side (cervical, 78% and 80% vs 48%, P < .01; middle, 60% and 55% vs 18%, P < .05; apical, 42% and 47% vs 26%, P < .05; total, 50% and 52% vs 25%, P < .001), MXCI exhibited a greater amount of change in ABA (ΔABA amount) than did MXC only in the middle level (−4.3 mm² vs −2.3 mm², P < .05).

Palatal to Labial ABA Ratio (Table 3)

Table 3.  Comparison of the Palatal to Labial Alveolar Bone Area Ratio Between the Pretreatment (T1) and Postretraction (T2) Stages in Each Tooth and the Amounts of Change During the T1 and T2 Stages Among the Maxillary Anterior Teeth

View Fullscreen

Table 3.  Extended

View Fullscreen

Although palatal to labial ABA ratios (P/L-ABA ratios) decreased in all areas of MXCI (cervical, middle, total, P < .001; apical, P < .05) and in some areas of MXLI (cervical, P < .001; apical, P < .05), MXC did not show differences in any of the areas between the T1 and T2 stages. Differences in ΔP/L-ABA ratio among MXCI, MXLI, and MXC were not significant in any of the areas.

VBL of the Alveolar Bone (Table 4)

Table 4.  Comparison of the Amount and Ratio of the Vertical Bone Level of the Alveolar Bone Between the Pretreatment (T1) and Postretraction (T2) Stages in Each Tooth and the Amounts of Change During the T1 and T2 Stages Among the Maxillary Anterior Teeth

View Fullscreen

Table 4.  Extended

View Fullscreen

At the T1 stage, the amounts and ratios of VBL (VBL amount and VBL ratio, respectively) were greater on the labial side than on the palatal side in MXAT (MXCI and MXLI, P < .001; MXC, P < .01). At T2 stage, the VBL amount and VBL ratio of MXAT increased both on the palatal and labial sides. Although MXCI and MXLI exhibited a greater ΔVBL amount and ΔVBL ratio on the palatal side than on the labial side, MXC did not exhibit differences in ΔVBL amount between the labial and palatal sides (ΔVBL amount; MXCI and MXLI, P < .001; ΔVBL ratio; 1.6% vs 29.3%, 4.9% vs 36.1%, 8.3% vs 15.3%, MXCI and MXLI, P < .001; MXC, P < .05).

In comparison of ΔVBL on the palatal side, MXCI and MXLI showed higher values than MXC in terms of the amount and ratio (all P < .001). However, on the labial side, MXC had a greater ΔVBL amount than did MXCI (P < .05).

Root Resorption (Table 5)

Table 5.  Comparison of Root Resorption in Terms of Root Length and Root Area Between the Pretreatment (T1) and Postretraction (T2) Stages in Each Tooth and the Amounts of Change During the T1 and T2 Stages Among the Maxillary Anterior Teeth

View Fullscreen

Table 5.  Extended

View Fullscreen

Although significant root resorption occurred in MXCI, MXLI, and MXC (RL, all P < .001; RA, all P < .001), the amounts of decreases in RL and RA did not differ among MXCI, MXLI, and MXC.

PD in the Cervical Area (Table 6)

Table 6.  Comparison of the Prevalence of Dehiscence in the Cervical Area Between the Labial and Palatal Sides in Each Tooth and Among the Maxillary Anterior Teeth on Each Side at the postretraction (T2) Stage^a

View Fullscreen

Root exposure (dehiscence) in the cervical area occurred more frequently on the palatal side than on the labial side (MXCI and MXLI, P < .001; MXC, P < .01). In addition, there was a higher percentage of dehiscence in MXLI and MXC than in MXCI on the labial side (14% and 12% vs 2%, P < .05) and in MXCI and MXLI than in MXC on the palatal side (67% and 68% vs 32%, P < .001).

DISCUSSION

The findings that ABA increased in the middle level of MXCI (P < .001) and in the middle and total levels of MXLI (P < .001 and P < .05, respectively) on the labial side and decreased in all the levels of MXAT (P < .001; middle of MXC, P < .01) on the palatal side (Table 2) imply that en masse retraction of the MXAT might result in tooth movement “through-the-bone.” This implication was confirmed by the fact that P/L-ABA ratios decreased in all areas of MXCI (cervical, middle, total, P < .001; apical, P < .05) and in some areas of MXLI (cervical, P < .001; apical, P < .05) after retraction (Table 3).

The increase in ΔABA amount on the labial side was much lower than the decrease in ΔABA amount on the palatal side (4% vs −50% in MXCI, 24% vs −52% in MXLI; Table 2), which means that bone apposition in the tension area of the inner side of the labial alveolar bone was not sufficient and/or bone resorption occurred on the outer side of the labial alveolar bone. Sarikaya et al.8 reported that the apposition process in the inner cortical plate of the labial alveolar bone is somewhat slower than the resorption process in the outer cortical plate of the labial alveolar bone.

The decrease in ABA of MXCI and MXLI was more significant in the cervical region of the palatal side (−78% and −80%, respectively; Table 2) because the MXAT of the samples showed a controlled tipping movement pattern, leading to a greater accumulation of pressure in the alveolar crest region on the palatal side. However, on the labial side, the middle area of MXCI and MXLI showed a greater increase in ABA than did the cervical area, where more tension existed (27% vs −10%, 65% vs 4%, respectively; Table 2). The reason that the cervical area did not show an increase in ABA on the labial side seems to be an inflammatory periodontal response concentrated in the cervical area, resulting in loss of VBL in spite of the greater tensional force. Therefore, the entire alveolar housing, not merely the bone in the apical zone, should be considered when a clinician tries to define the therapeutic limits for orthodontic tooth movement.8

At the T1 stage, VBL on the labial side was greater than that on the palatal side in all of the MXAT (MXCI and MXLI, P < .001; MXC, P < .01; Table 4). These findings are consistent with those of Nahm et al.,13 who reported VBL ratios of 19.3% and 15.0% for the labial and palatal aspects of the MXAT, respectively. The finding that at the T2 stage, MXCI and MXLI showed a greater VBL and a higher PD on the palatal side than on the labial side compared to MXC (Table 4) might result from the discrepancy in the direction of tooth movement in relation to the labiolingual long axis of the roots among MXCI, MXLI, and MXC (Figure 4).

View Figure

• Download as Powerpoint
• Download Figure

Dehiscence is a major bony defect that is difficult to discern on conventional 2D radiographs.17 A previous CBCT study18 defined dehiscence as the absence of cortical bone around the root in at least three sequential views. In our study, dehiscence was determined when the ABA covering the cervical root was zero. In spite of this strict definition of PD, a strikingly higher PD in the cervical area on the palatal side was shown after space closure (67% of MXCI, 68% of MXLI, and 32% of MXC, respectively; Table 6). Because the upper parts of the roots of MXAT are supported by thin alveolar bone13 they are vulnerable to dehiscence during retraction. However, a potential limitation of this result can exist in the risk of overestimating dehiscence. A previous study19 reported that the positive predictive value of dehiscence was only 0.51 and that alveolar bone height could be measured with an accuracy of 0.6 mm using CBCT with a voxel size of 0.38 mm at 2 mA. Moreover, since additional bone remodeling continues slowly after active tooth movement, the results from this study should be interpreted conservatively when one is applying them to the clinical situation.

Since patients with dentoalveolar protrusion usually have thin and elongated anterior alveoli and/or a bony defect before treatment,8,11,13 pushing the tooth against the thin cortical bone may cause root resorption and/or an alveolar bone defect (Figure 5). If the bracket prescription with excessive root torque is used in the MXAT, excessive root movement can cause a greater risk of root resorption and PD of the labial or palatal cortical plate.7 For these patients, retraction of the anterior teeth using absolute anchorage with orthodontic mini-implants may not always be the right answer. If labioversion of the incisors is excessive and the alveolus is thin, retraction of the anterior teeth combined with corticotomy of the alveolar bone can offer an effective alternative with which to minimize the risk of uncontrolled movements of the anterior teeth.3,4,11

View Figure

• Download as Powerpoint
• Download Figure

Another important issue is whether or not the repair of alveolar bone loss is possible after space closure and during the retention period. In this study, some patients who were able to undergo CBCT at the debonding stage did not show spontaneous bone apposition on the labial and palatal sides (Figure 6). Previous studies6,8,20,21 reported that once the cortical plate had been penetrated by the root, recovery of the well-defined dense cortical plate would not occur. Therefore, longitudinal studies are needed to investigate whether repair of the alveolar bone defect takes place after space closure and during the retention period and to identify discriminative factors for good and poor capacity for alveolar bone remodeling.

View Figure

• Download as Powerpoint
• Download Figure

CONCLUSION

• During EMR-MA in patients with Class I dentoalveolar protrusion, the ABA and VBL on the palatal side and the RL and RA of the MXCI and MXLI were significantly decreased.