Effects of Herbst Appliance Treatment on the Mandibular Incisor Segment: A 3-D Analysis

Background: Three-dimensionally evaluation of the treatment changes of a Herbst appliance using a lower anchorage unit not touching the lingual surface of the lower incisors. Methods: 23 Class II:1 patients (12 males, 11 females) with a mean age of 15.7±1.7 years treated with a Flip-Lock Herbst ® (TP Orthodontics, Inc., La Porte, IN, USA) appliance. The lower anchorage unit for the Herbst appliance consisted of two anchor bands connected by a lingual arch with 3mm distance from the incisor’s lingual surface. Treatment changes in mandibular incisor inclination, overjet and overbite were evaluated by means of cone beam computed tomography images (i-CAT ® Classic unit, Imaging Sciences International, Hateld, PA, USA) obtained before and after treatment with the Herbst appliance. Results: On average, there was a statistically signicant increase in mandibular incisor inclination (2.6+1.8°) and a reduction in overjet (3.2+2.2mm) and overbite (1.3+0.9mm). Genders did not differ signicantly. Incisor proclination was however only seen in 74% of the patients. The changes in mandibular incisor inclination were associated with the changes in overjet (/r/ = 0.1 to 0.5) and overbite (/r/ = 0.3 to 0.7). Conclusion: A Herbst appliance with a mandibular anchorage unit distant from the incisor’s lingual surface results in smaller amounts of mandibular incisor proclination compared to literature. However, as it induces canine anchorage loss, the decreased amount of proclination may not prevail after multibracket treatment. Trial study


Background
Lower incisor proclination is an undesired side effect of Herbst treatment. Although, so far, no investigation could prove a clinically signi cant negative impact of incisor proclination on periodontal health, neither short-nor long-term [1][2][3][4][5], a proclination may hinder the achievement of a Class I occlusal relationship by increasing lower dental arch length and is said to be more prone to relapse. Up to now, regardless of the type of anchorage unit used, not even with additional skeletal anchorage, could the position of the mandibular incisor be satisfactorily controlled during Herbst treatment [6][7][8][9]. To what extent, a lower arch anchorage unit not touching the lingual surface of the lower incisor could prevent or reduce the amount of proclination has not been studied yet.
All aforementioned mentioned studies have used two-dimensional radiographs for the analysis of Herbst appliance treatment effects, which hinder a tooth speci c evaluation of changes. In contrast, cone beam computed tomography (CBCT) images allow to assess a single tooth three-dimensionally (3D) and to study tooth inclination changes induced by different orthodontic appliances [10,11] as well as their impact on alveolar bone support [5].
The aim of the present study was to evaluate individually for each incisor by means of CBCT the mandibular incisor inclination changes as well as overjet and overbite changes induced by a Herbst appliance with a lower anchorage unit not touching the lingual surface of the lower incisors.

Methods
This retrospective study was reviewed and approved by the Ethics Committee of Araraquara Dental School, Paulista State University (FOAr-UNESP), project number 62/2010. A total of 30 patients meeting the inclusion criteria were invited to participate in the study. Five patients refused to participate. Two patients were excluded because of appliance breakage. The nal sample consisted of 23 consecutively treated (12 male, 11 female; mean age 15.7 ± 1.7 years) patients with Class II division 1 malocclusion. Hand wrist radiographs were used to assess the skeletal maturity according to Hägg and Taranger [12].
The percentage of cases within the different skeletal maturity groups before treatment were: MP3-G (4%), The inclusion criteria were bilateral Class II canine and molar relationship ≥½ cusp, overjet > 5 mm, complete permanent dentition (except third molars), convex pro le, straight nasolabial angle and short mentocervical line. Exclusion criteria were syndromic patients, increased vertical facial height, previous orthodontic treatment and need for maxillary expansion.
All patients were treated with a Herbst appliance (Fig. 1). The anchorage unit for the Herbst appliance consisted of upper rst molar bands connected by a transpalatal arch (1.2 mm steel wire) with 2 mm distance from the palate as well as two occlusal extensions to reduce rst molar intrusion and prevent second molar overeruption. In the lower arch two anchor bands were connected by a lingual arch (1.2 mm stainless steel wire) with 3 mm distance from the incisor's lingual surface to prevent incisor proclination.
The labial cantilever was connected to the lingual arch at the level of the interproximal area between the canine and rst premolar on both sides. The telescopic mechanism used was a Flip-Lock Herbst® (TP Orthodontics, Inc., La Porte, IN, USA). No additional appliances were used. The average treatment duration with the Herbst appliance was 8.5 ± 0.7 months.
CBCT images were obtained before (T0) and after (T1) treatment. Patients were scanned in an upright position with maximum intercuspation using a tomographic i-CAT® Classic unit (Imaging Sciences The CBCT images were analyzed metrically. In the rst step, the reference plans were de ned. The plane that includes the superior tip of the odontoid process of the axis, the tip of the anterior nasal spine and the nasion point is de ned as the midsagittal plane (MSP) and the sagittal plane was oriented to coincided with the MSP [10,11], (Fig. 2). In the coronal and axial view, the cursors are set to intersect in the center of the mandibular incisor of interest [10,11], (Fig. 3).
The following measurements were performed in sagittal multiplanar reconstruction: mandibular incisor inclination, overjet and overbite. The mandibular incisor inclination measurement was angular between three points, the rst at the incisal edge, the second at the root apex and the third at the intersections of the axial and coronal cursors, positioned at the incisal edge and root apex, respectively (Fig. 4). Overjet measurement was linear, parallel to the axial cursor, between two points, one at the incisal edge of maxillary incisor and the other one at the buccal surface of mandibular incisor (Fig. 4). Overbite measurement was linear, parallel to the coronal cursor, between two points, one at the incisal edge of mandibular incisor and other at the intersections of the axial and coronal cursors, positioned at the incisal edge of maxillary incisor and incisal edge of mandibular incisor, respectively (Fig. 4).

Statistical Analysis
All measurements were performed twice by a single examiner with a minimum interval of at least two weeks between the measurements. The error of the method was evaluated by Intraclass Correlation Coe cient (ICC) and indicated excellent reliability for mandibular incisor inclination (ICC = 0.965), overjet (ICC = 0.970) and overbite (ICC = 0.958).
After the data had been tested for normality with the Shapiro-Wilk Test, Student's t-Test and Wilcoxon t-Test were used to compare dependent samples in parametric and non-parametric cases, respectively. Student's independent t-Tests was used for gender comparison and Pearson's and Spearman's rank correlation analyses were performed to determine the relationship of changes in mandibular incisor inclination, overjet and overbite.
A distinction was made between the following correlations: strong (/r/ ≥0.8), moderate (/r/ 0.4-0.8) and weak (/r/ <0.4). If there was a weak insigni cant correlation, it was denoted as no correlation. Statistical analysis was performed using SPSS® (SPSS Inc, Chicago, IL, USA) and GraphPad Prism® (GraphPad Prism Inc, San Diego, CA, USA) and the results were considered at a signi cance level of 5%.

Results
Comparisons of mandibular incisor inclination, overjet and overbite between genders did not show any statistical differences (Table 1), therefore the data were pooled for further evaluation. The individual mean changes for each patient are shown in Fig. 5. T0-before treatment; T1-after treatment; xmean; SD-standard deviation; Plevel of signi cance.
There was a statistical signi cant moderate correlation between mandibular incisor inclination changes and overjet changes for the mandibular left lateral incisor and mandibular right central incisor (Table 5). There were statistical signi cant moderate correlation between mandibular incisor inclination changes and overbite changes for the mandibular left lateral incisor, mandibular right central incisor, mandibular right lateral incisor and total (Table 6). Table 5 Pearson's and Spearman's rank correlation analysis between mandibular incisor inclination changes and overjet changes.

Discussion
This CBCT study evaluated three-dimensionally the mandibular incisor inclination changes as well as overjet and overbite changes induced by a Herbst appliance with a lower anchorage unit not touching the lingual surface of the lower incisors.
3D CBCT images allow an assessment of the buccolingual inclination of individual teeth with good accuracy in any given plane [13,14]. Regarding the acquisition of tomographic images, the accuracy of CBCT with different voxel resolutions (0.2 and 0.4 mm) to linear measurement was evaluated and there was no signi cant statistical difference between these voxel protocols [15,16].
Up to now, regardless of the type of anchorage unit used, not even with additional skeletal anchorage, could the position of the mandibular incisor during Herbst treatment be satisfactorily controlled nor be predicted on an individual level [4,[6][7][8][17][18][19][20][21]. It is however known, that the length of the anchorage unit [19,20,22], the severity of the malocclusion [8], the incremental mandibular advancement [23,24], the skeletal anchorage [6,7] and the skeletal maturity [25][26][27] of the patients in uence the amount of incisor proclination to a certain extent. Thus, comparisons with literature are di cult and very anchorage sensitive and the corresponding conclusions have to be drawn with care.
In interpreting the results, it must be kept in mind that the changes induced by the Herbst appliance are a summation effect of treatment and dentofacial growth.
The skeletal maturity evaluation of the present sample showed that 92% of the patients were in the postpubertal period, a developmental stage during which more dentoalveolar than skeletal changes are seen during a Class II treatment with a Herbst appliance [27]. The mandibular incisors of patients treated after the pubertal growth peak have been shown to procline more than those of patients treated pre-peak [25,26].
There were no statistical differences in mandibular incisor inclination, overjet and overbite between genders neither before nor after treatment, which is in concordance with previous cephalometric studies [8,28].
The mean increase of mandibular incisor inclination from T0 to T1 (2.6°) was smaller when compared to previous studies using different types of Herbst appliance anchorage such as a cast splint appliances (6.7° to 12°) [8], and even skeletal anchorage (4.8°), [6,7]. One reason for this smaller amount of proclination could be the design of the lingual arch with 3 mm distance from the incisor's lingual surface upon insertion, thus reducing the anchorage load on the incisors. According Pancherz and Hansen [19], a reduced anchorage load on the incisors by means of a lingual acrylic pelotte reduces the amount of incisor proclination.
The mean reduction of overjet (-3.2 mm) was rather small in comparison to the above mentioned Herbst studies, thus resulting in a reduced risk for anchorage loss [8,19,29]. However, Martin and Pancherz [8] evaluated mandibular incisor proclination during Herbst treatment with cast splint anchorage including a lingual arch touching the lower incisors and found an average lower incisor proclination of 6.7°, which is twice as much as in the present study.
Another important point regarding the mandibular anchorage unit used in the present study is the connection of the labial cantilever with the lingual arch at the level of the interproximal area between the canine and rst premolar on both sides. This contact with the distal surface of the canines favors a canine anchorage loss which induces a crowding in the lower incisor/canine area (Fig. 6). As in the modern Herbst therapy, the orthopedic phase is followed by a multibracket appliance (MB) phase, the crowding will result in secondary incisor proclination. Thus, whether the amount of proclination using an anchorage unit not touching the lingual surface of the lower incisors is really smaller after the end of active orthodontic treatment (Herbst plus subsequent MB) cannot be answered yet.
There was a marked interindividual variation in treatment effects in the present study with respect to the amounts of mandibular incisor proclination, overjet and overbite reduction. This large variation has also been described in previous studies and the reasons for it remain largely unknown [4,6,19,30]. A retroclination of the lower incisors in certain patients has also been reported in previous Herbst publications [6,8,21].
Overall, there were some limitations due the study design which should be considered when interpreting the results. These limitations include the small sample size, absence of a control group, length of observation period (only Herbst phase) and tomography images acquisitions protocols (voxel size and eld of vision). These weaknesses will be considered in a future research project.

Conclusion
The 3D evaluation of the mandibular incisor segment by means of CBCT scans revealed that Herbst appliance treatment with a mandibular anchorage unit distant from the incisor's lingual surface results in smaller amounts of mandibular incisor proclination compared to literature. However, as it induces canine anchorage loss, the decreased amount of proclination may not prevail after MB treatment.

CONSENT FOR PUBLICATION
Written informed consent to publish individual person's data (images) were obtained.

AVAILABILITY OF DATA AND MATERIALS
The datasets used during the current study are available from the corresponding author on reasonable request.

COMPETING INTERESTS
The authors declare that they have no competing interests. AUTHORS' CONTRIBUTIONS JPS contributed to the conception and project design, data analysis, statistical analyses and wrote the manuscript. TBR contributed to the conception and project design. DBR contributed to the conception and project design, data analysis and wrote the manuscript. SR contributed to the conception and project design, data analysis and wrote the manuscript.