INTRODUCTION

Rapid maxillary expansion (RME) was introduced at the end of 19th century,¹ but only in the mid-1960s, with the research of Haas,² did it gain scientific respect. Since then, several modifications in the original expander design have been proposed.^3–8 Substituting the original tissue-borne fixed splint acrylic maxillary palate expander² and toothborne all-metal framework hygienic expanders (Hyrax)⁹ became very popular among orthodontists.¹⁰ The safety of palatal expansion has been consistently reported in the orthodontic literature.^11–17 However, a high incidence of external root resorption (ERR) is strongly associated with abutment teeth after RME.^18–24

To improve the orthopedic effect, the original Haas and Hyrax expanders have bands in the maxillary first premolars and first molars.^2,9 However, to simplify the technique, a wire-supported anchorage replaced the bands in the first premolars in several expander models.^4,6,25 Banded teeth receive a different load compared with wire-supported teeth, and such a difference may have mechanical and biological effects that have not yet been fully examined. The decision about whether the first premolars needs to be banded as a preparation for RME should be based on scientific evidence not only on practice management.

Thus, the aim of the present study was to assess ERR when bands or wires are used as orthodontic anchorage during RME. The null hypothesis was that banded and nonbanded abutment first premolars develop the same pattern of ERR after RME.

MATERIALS AND METHODS

Sample

The Institutional Review Board (IRB) of the Pontifical Catholic University of Minas Gerais (Belo Horizonte, Brazil) approved the current study. The rights of the patients were protected, and their parents/guardians signed the IRB-approved informed consent.

The sample consisted of 18 maxillary first premolars and six mandibular first premolars from nine orthodontic patients seen at the orthodontic clinic of the Pontifical Catholic University of Minas Gerais, (four boys and five girls). The patients had a mean age of 15.2 ± 1.4 years (range = 12–16 years old). All patients presented with maxillary constriction, maxillary dental crowding, and/or maxillary biprotrusion with orthodontic indication for both RME and extraction of first premolars. None of the patients had systemic problems, history of periodontal disease, previous orthodontic treatment, and history of carious lesions in the first premolars. First premolars were in Nolla stages 9 and 10 of dental development.²⁶

From each extracted premolar, buccal, palatal, and interproximal radicular surfaces were histologically analyzed in two segments (cervical third and middle third). The cervical third was the segment between the cementoenamel junction and 4 mm apically; the middle third extended 4 mm to 8 mm apically from the cementoenamel junction. Thus, 108 sites were analyzed from the maxillary premolars and 36 sites from the mandibular premolars (Figure 1).

View Figure

• Download as Powerpoint
• Download Figure

Orthodontic Anchorage During RME

A modified 9-mm Hyrax expander (Morelli Ortodontia, Sorocaba, Brazil) was used in the present study (Figure 2). Orthodontic bands (American Orthodontics, Sheboygan, Wis) were attached to both permanent maxillary first molars and to only one maxillary first premolar, which was randomly selected based on a split-mouth design. The contralateral maxillary first premolar was anchored on the palatine surface with an orthodontic wire. The banded maxillary first premolars (n = 9) composed the band group (BG), and the maxillary first premolars whose expansion load was applied by the wire framework (n = 9) made up the wire group (WG). The screw was activated four turns (0.8 mm) on the installation day, followed by one turn twice a day (morning and night) during the next 14 days. Thus, the expansion screw was activated to 6.4 mm. All patients underwent 3 months of retention, and the maxillary first premolars were then extracted by the same oral surgeon. Contact of the forceps with root cementum was avoided. Six mandibular first premolars served as the control group (CG). They were extracted at the beginning of treatment, before any orthodontic movement had been performed.

View Figure

• Download as Powerpoint
• Download Figure

Histologic Preparation

After extraction and tap-water washing, a straight longitudinal shallow groove was prepared as an identification mark on the root’s palatal surface with a double-faced diamond disk under water-cooling (Figure 3). To improve access of the fixing solution into the root canal, 2 mm of the apical third was removed immediately after extraction. The teeth were stored and fixed in 10% neutral buffered formalin for 72 hours. All samples processed for histology were then rinsed with phosphate buffered saline wash buffer (pH 7.3), immersed in Planck’s solution for 36 hours, and subsequently immersed in 4.13% EDTA (ethylenediaminetetraacetic acid) decalcifying solution (Lenza Farmaceutica, Belo Horizonte, Brazil) for 4 weeks, with frequent changes. Following decalcification, the samples were processed for routine histology by paraffin embedding. Sagittal sections (6 μm) were cut and stained with hematoxylin and eosin for routine histology. Samples from the radicular cervical and middle thirds were prepared, taking into consideration the distance from the cementoenamel junction (cervical, 0–4 mm; middle, 4–8 mm).

View Figure

• Download as Powerpoint
• Download Figure

In each sample, the section with the largest root resorption was selected for measurement. The sections were inspected using light microscopy and were imaged at 10× magnification (Leica EZ4, Wetzlar, Germany). For morphometric assessment, the apical third was not considered.

Histologic evaluations were performed using ImageJ software (National Institutes of Health, Bethesda, MD). The following measurements were taken: (1) resorption width (Figure 4a); (2) resorption depth (Figure 4a); (3) lesion total area (Figure 4b); and (4) repair area (Figure 4c).

View Figure

• Download as Powerpoint
• Download Figure

RESULTS

ERR at the Level of Cementum and Dentin as well as Cementum Repair Were Found in All Maxillary First Premolars Subjected to RME

Three months after RME, all maxillary first premolars showed cementum and dentin resorption (Figure 5) as well as repair with cementum cells (Figures 6b through 6d). However, of the 108 examined maxillary premolars samples, only one showed complete cementum repair (Figure 6c). In all cases, the contour of the resorption area could be easily identified in the histologic evaluation (Figure 5). The mean resorption depth was 177.59 µm, and the mean width was 856.59 µm. The mean area of repair in the total area of resorption was 37%. Histologic examination of the six untreated control mandibular premolars showed no root resorption (Figure 6a).

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

ERR Was Similar When Orthodontic Bands or Orthodontic Wire Were Used as Anchorage During RME

Qualitative analysis showed that the type of anchorage for the first premolar (orthodontic band or wire) did not affect the incidence of ERR (P > .05) (Table 1). Quantitative analysis demonstrated that the lesion total area, repair area, and depth of resorption were not associated with the type of anchorage (Table 2).

Table 1. Association Between ERR and Anchorage Type, Tooth Surface, and Analyzed Site (Cervical or Middle Third of the Root)^a

View Fullscreen

Table 2. Association Between Anchorage Type (BG vs WG) and ERR Measurements^a

View Fullscreen

Radicular Cervical and Middle Thirds Had the Same Pattern of ERR After RME

Tables 1 and 3 show that ERR in the middle and cervical thirds was similar. In the cervical third, 18 of the 54 analyzed samples showed ERR, and 19 of the 54 samples of middle third showed ERR. In addition, quantitative analysis (Table 3) showed that there was no association between the radicular height position and the extension of ERR (lesion total area, repair area, width, and depth of resorption) (P > .05).

Table 3. Association Between Tooth Root Third (Middle vs Cervical Thirds) and ERR Measurements^a

View Fullscreen

Radicular Buccal Surface Had a Higher Amount of ERR After RME

The buccal surfaces showed a greater incidence of ERR compared with the palatine and interproximal surfaces after RME (Table 1). The buccal surface also showed more severe resorptive extension and had a greater width and depth of resorption compared with the palatine surface. Greater repair area was also found in the buccal surface. However, no statistically significant difference was found between the buccal and interproximal surfaces (Table 4).

Table 4. Association Between Tooth Surface (Buccal vs Palatal vs Interproximal) and ERR Measurement^a

View Fullscreen

DISCUSSION

In the current investigation, all maxillary premolars subjected to RME showed ERR reaching the cementum and dentin. Our finding was consistent with those of previous reports, corroborating with evidence that the RME is an orthodontic technique with a high rate of root resorption of abutment teeth.^{20–24,27,28} Because both orthodontic bands and wires have been used as anchorage devices in the first premolars, and wire framework are routinely used to increase the anchorage in other maxillary posterior teeth, we aimed to evaluate if there are differences in the pattern of ERR if orthopedic heavy loads are applied in banded and nonbanded teeth. We wished to find out which of these two mechanisms was less biologically harmful to the roots. Most of the previous investigations on this topic found ERR in banded teeth,^20,22,23,29 but a recent article suggested that even nonbanded maxillary teeth are also exposed to the risk of ERR after RME.²⁴ We found that both systems (band and wire) were associated with ERR. However, no statistically significant difference was found between band and wire with regard to the pattern of ERR.

Maintenance of the radicular resorptive process was observed even when the expansion was terminated. Our histologic evaluations performed 3 months after the end of RME showed active signs of ERR. Such a finding is in accordance with previous data. According to Barber and Sims²⁰ and Langford,²⁸ long-term ERR is due to the accumulated loads in the appliance, even after the activation period. Nevertheless, Zimring and Isaacson¹⁷ have reported that the expander loads are completed after 5 to 7 weeks.

With regard to the most resorbed root surface after RME, previous studies have reported that the buccal surface is the most affected in extension, depth, and length,^{20,22–24,27} while the palatine surface is the least affected.^20,21 Our findings were consistent with these results. The greater compression loads transferred onto the buccal surface during RME may account for the greater ERR on the buccal surface. However, we also found that the buccal surface showed greater repair. Thus, we can infer that the resorbed sites are likely to be repaired in the short term.

Clinicians routinely make a radiographic diagnosis of ERR. However, radiographs cannot detect the microscopic changes on the root surfaces compared with a histologic examination.^29–31 Thus, radiography should be used with caution in scientific investigations on ERR. In the present study, a histologic evaluation of the topographic changes on radicular surfaces was performed to provide precise and reliable information regarding the differences between the two types of orthodontic anchorage. The sample size was adequate for histologic analysis.

In our investigation, the mandibular first premolar controls showed no signs of ERR. Such a finding was consistent with those of previous studies.^20,21 However, Chan et al.³² found root resorption in nonorthodontically treated control teeth. These contrasting results might be due to differences in the method of measurement because they were performed using scanning electronic microscopy analysis. We used an expander model that was very similar to the one proposed by Barber and Simms,²⁰ but their expander had only one banded first premolar. The maxillary contralateral tooth served as an unloaded control. They reported no signs of ERR in control teeth.

The duration of the retention period is relatively controversial in the orthodontic field. Periods ranging from 3 to 12 months have been recommended.^2,17 Apparently, the longer the retention time, the greater the stability. According to Thorne et al.,³³ a minimum of 3 months should be expected. We selected a 3-month retention period because we aimed to evaluate the histologic changes in the short term. Langford and Sims²² found no relationship between the length of retention and ERR in their study. However, Barber and Sims²⁰ reported that the ERR was more significant between the end of the expansion and after 3 months of retention. Vardimon et al.³⁴ observed that the cementum was rapidly laid down during the retention period, but the increase in the formation of Sharpey’s fibers occurred during the postretention period, thereby restricting the length of this phase. Nevertheless, it is fair to infer that these heterogeneous findings may be obtained with different retention periods. Thus, we suggest that a similar research design should be used with longer periods of retention. We propose that additional areas of repair will be formed and that the depth and width of the resorptive areas will be smaller.

The implications of ERR in terms of the longevity of the teeth is uncertain.²⁰ Orban³⁵ reported that orthodontically induced root resorption is transient and results in no definitive damage, or it may only exhibit a small radicular defect that does not affect the health and physiologic function of the teeth in the long term. However, there is little scientific evidence to support this statement. In the current study, we found consistent evidence that cells of the cementum lay down in areas of resorption after the end of the active treatment of RME. However, complete repair of the resorption area by cementum cells was only observed in one of the 108 samples, 3 months after the end of RME. Moreover, this full radicular cementum repair was not necessarily synonymous with the re-ligation of the main periodontal fibers in the restored area.²⁰

Thus, the present findings indicate that clinicians need to consider the risks of RME. Unfortunately, a dentist cannot precisely estimate the extension of root resorption caused by RME in the short term. Furthermore, we have demonstrated for the first time that the use of an orthodontic band as an anchorage does not cause significantly more ERR compared with an orthodontic wire.

CONCLUSIONS

• ERR at the level of the dentin and cementum and partial cementum repair were found in all first premolars 3 months after RME.

• The type of anchorage (orthodontic band or wire) had no effect on ERR.