Impact of orthognathic surgery on the prevalence of dehiscence in Class II and Class III surgical-orthodontic patients:A cone beam computed tomographic study

Alveolar bone dehiscence (ABD) is defined as an increase in the distance between the cemento-enamel junction (CEJ) and the alveolar crest margin (ACM), when it is 2 mm or greater on the lingual and/or buccal sides.^1–8  The prevalence of dehiscence varied significantly among studies^3,6,9–11  because of the variability of the methods and samples (8.19%,¹¹  40.4%,¹²  50%,^10,13  and 71.61%¹⁴). However, there is a consensus that the prevalence of dehiscence increases with age.^3,11,15–17  Bone dehiscence is a risk factor for gingival recession, which involves the exposure of the root at the cervical area.⁹  This may cause a series of undesirable conditions such as dentin sensitivity, restriction when brushing, esthetic concerns due to root discoloration and gingival irregularity, and root caries.¹⁰ 

The diagnosis of dehiscence can be performed through cone-beam computed tomography (CBCT), with excellent accuracy. CBCT provides three-dimensional images that facilitate measurements from the buccal and lingual bone plates, overcoming the limitations of conventional two-dimensional radiographs.¹⁸ 

The etiology of the ABD is unclear. It appears to be influenced by the thickness of the alveolar bone.^10,19  Thinner bone has a greater risk of bone loss. In addition, it appears that excessive inclination toward the buccal or lingual plate may negatively influence bone integrity.^18,19  The direction and intensity of the force applied during orthodontic movement influence bone remodeling, and teeth can be moved away from the mid-alveolus, leading to a thinner bone on one side.^7,8,16,19  Orthodontists must be attentive to the presence of ABD and the potential to create dehiscences in every clinical condition, making sure that tooth movement does not exceed the bone boundaries.^18,19 

Although previous studies have assessed the frequency of dehiscences in different types of malocclusions as well as in different orthodontic movements, there is no research on the impact of orthognathic surgery on the prevalence of such defects in surgical-orthodontic cases. Thus, the aim of this study was to test the null hypothesis that orthognathic surgery had no impact on the prevalence of alveolar dehiscence.

This retrospective longitudinal study was approved by the Research Ethics Committee of the State University of Maringa (863260 18.6.0000.0104).

The sample calculation was performed based on a pilot study. Using a test power of 80%, alpha of .05, a maximum error of 0.6 mm, and a standard deviation of 1.6 mm, there was a need for 20 CBCTs per group.

The study evaluated 230 records from the database of the Laboratory of Image and Clinical Research of the State University of Maringa (UEM), from 2014 to 2018. All records were from patients who had undergone surgical-orthodontic treatment. The inclusion criteria were CBCT images of adults (> 18 years of age), with Class II or III facial profile, obtained on average 30 days prior to (T0) and six months after (T1) traditional bimaxillary orthognathic surgery. Exclusion criteria were patients with cleft lip and/or palate, craniofacial syndromes, history of dentofacial trauma, periodontal disease, previous orthognathic surgery, and presence of radiological artifacts that made most measurements impossible. All surgeries were performed by the residents and their supervisors from the UEM surgery team.

The sample comprised 23 patients with a Class II profile (16 women and 7 men) and 22 patients with a Class III profile (13 women and 9 men), with a mean age of 31.5 ± 10.28 and 26.9 ± 8.46 years, respectively. All patients had orthodontic preparation for orthognathic surgery for an average period of 26 months, had preventive care to avoid active periodontal disease, and were required to follow a prophylactic protocol for 15 consecutive days using chlorhexidine 0.12% mouthwash after surgery.

All CBCT scans were obtained by the same dental radiologist following a standardized protocol. Patients were seated in the natural head position, with the tongue and lips at rest, centric occlusion, and without support of the chin and head in order to not deform the facial soft tissues, which could undermine virtual planning of orthognathic surgery.²⁰  Images were obtained by an i-CAT Next Generation (Imaging Sciences International, Hatfield, Penn) with a 14-bit gray scale and 0.5-mm focal point. The volumes were reconstructed with 0.30 mm of isometric voxel, with a field of view of 17 × 23 cm (ranging from the frontal region, 2 cm above the glabella, to just below the hyoid bone, 2 cm below the mandible), a tube voltage of 120 kVp, tube current of 3–8 mA, and the amount of radiation with at most two previews (measured by the device's own DAP function) of 891.4 mGy × cm².

CBCTs were ordered and imported in DICOM format to OsiriX software (version 3.3 32-bit, Pixmeo, Geneve, Switzerland). The images were analyzed in two-dimensional multiplanar reconstruction, which allowed for visualization of the tomographic image in the three orthogonal planes: coronal, sagittal, and axial (Figure 1). The Window Level − Window Width (WL/WW) instrument, which enabled variation of the gray scale for tissue density, was standardized at WL 600,000 and WW 3200,000, respectively.

The distance from the CEJ to the ACM was measured on the buccal and lingual surfaces of the central incisor, lateral incisor, canine, first premolars, and first molars of both arches. Angular cephalometric measurements (SN.GoGn, ANB, IMPA, and 1.PP) were obtained for sample characterization.

For anterior teeth, the measurements were performed in the sagittal view. The reference cursor was positioned centrally on the tooth in the axial plane, and the orientation line followed the midpoint of the buccal surface. In the coronal plane, the cursor was positioned on the first third of the root, and the orientation line followed the long axis of the tooth from the cusp to the apex (Figure 1). For the posterior teeth, the measurements were taken in the coronal view using the same orientation of the tooth long axis (Figure 2), except for multiroot teeth. In two-rooted teeth (lower first molar), the cursor was oriented centrally to the mesial root, and in three-rooted teeth (upper first molar), the cursor was centrally positioned on the mesiobuccal root with the orientation line going through the buccal and palatal root for buccal and lingual surface measurements (Figure 2). One calibrated operator performed all measurements.

Fifteen percent of the samples were randomly selected and remeasured 30 days after the first measurements. The reliability was assessed by the intraclass correlation (IC) and paired t-test. Shapiro-Wilk, McNemar, and Wilcoxon tests were performed at a significance level of 5% using Bioestat 5.0 software (BioEstat, Belém, Brazil).

All patients who met the eligibility criteria were included in the study. However, not all the sample units for each research participant were included. Among the images taken from 1800 sites, 468 had imprecise visualization of the CEJ or alveolar crest due to radiologic artifacts or teeth that were absent. Thus, 1332 sites were measured (644 in the Class II group and 688 in the Class III group). Demographic and cephalometric information is shown in Table 1.

The IC showed good concordance both for the Class II and Class III groups (0.71; 95% IC 0.54, 0.83, and 0.74; 95% IC 0.58, 0.85, respectively). Student t-test revealed no statistical differences between the first and second measurements (P < .05), with mean differences of 0.0059 mm for the Class II group and 0.0048 mm for the Class III group.

The data were not normally distributed (Shapiro-Wilk test, P < .05); thus, medians and percentiles were compared. Data from both sides were pooled, as no significant differences were found between the right and the left sides.

The prevalence of ABD increased slightly in both groups: 4% for the buccal and 6% for the lingual sides in the Class II group, and 3% for the buccal and 6% for the lingual sides in the Class III group (Tables 2 and 3).

ABD median increased significantly in the Class II group only for the lower first molar on the lingual side, whereas ABD increased significantly in the Class III group only for the upper canine on the buccal side (Tables 4 and 5).

When the Class II and Class III groups were compared, the Class II group displayed more ABD than the Class III group, at both T0 and T1 (Tables 6 and 7).

This study showed that there was a slight increase in ABD after orthognathic surgery. Kramer et al.²¹  had a similar outcome, finding a significant increase in gingival recession after orthognathic surgery, which implied alveolar dehiscence.⁵ 

Le Fort I and sagittal osteotomies are used to correct most skeletal problems in orthognathic surgery.²²  These procedures are quite invasive and can cause a number of postoperative complications, including vertical loss of alveolar bone. Le Fort I osteotomy can have an impact on the vascular supply of the upper teeth. After incisions and subsequent osteotomy, periosteal detachment is performed, which promotes bone exposure at the site. One possible explanation for the dehiscences observed on the teeth evaluated in this study may be related to the tissue trauma and decreased blood flow caused by the surgery.²³  This injury to the medullary nutritional system may cause transient ischemia and insipid osteonecrosis, which may be related to dehiscence formation. Gingival recession was reported²²  in 0.8% of patients after Le Fort I, due to a postoperative complication involving partial necrosis of the maxillary segment. In addition, alveolar bone loss may also be related to damage caused by overheating of the alveolar bone with the use of reciprocating saws, drills, and chisels.²⁴ 

In the present study, most sites diagnosed with dehiscence occurred at the lingual surfaces, although surgical incisions were performed in the buccal region. The surgical procedure for both Class II and Class III involved osteotomies that were close to the lower molars and may have caused inferior alveolar vessel-nerve distension, laceration, or section at the time of the osteotomy and fixation, resulting in nutritional deficiency that may have affected the teeth in the lower arch. Foushee et al.²⁵  reported a direct relationship between mandibular advancement surgery associated with genioplasty and the appearance of gingival recession in the lower anterior teeth in 42% of patients. In the present sample, 50% of the patients underwent genioplasty, but there was no significant increase in dehiscences in the anterior region of the mandible.

Carroll et al.²⁶  suggested that rigid internal maxillary fixation during orthognathic surgery may be related to ABD formation; however, they did not find significant results in their study. The data found in the current study seemed to show no relationship between semi-rigid fixation plates and ABD.

Orthodontic-surgical treatment is not performed in patients with active periodontal disease. Weinspach et al.²²  analyzed the subgingival periodontal pathogenic microbiota and concluded that they were present in low concentrations for up to 6 weeks postoperatively, without considering the relationship of poor oral hygiene among patients during the healing period. Interestingly, however, the lingual/palatal areas underwent major changes in the present study. Because of postoperative fixation and limitations in opening, access to perform hygiene on the lingual surfaces may have been disrupted, despite the use of chlorhexidine, and this factor requires further research.

In the present study, the Class II group had more hyperdivergent patients than the Class III group, and the prevalence of dehiscence was higher in the Class II group. The literature is not clear about the influence of skeletal pattern on dehiscence prevalence,^1,2  and such influence needs more study.

The present study evaluated images taken after the surgical postoperative period of up to 6 months when orthodontic treatment was resumed, so it was not possible to clarify whether the outcomes of the study arose only from the surgery, the orthodontic movement, or a combination of both. In addition, the lack of biomechanics records from orthodontic procedures during such intervals is an issue in this discussion. Such a limitation was also observed in another retrospective study,²²  where the author was also unable to establish a causal relationship for gingival recession definitively. It is important to note that preoperatory orthodontic procedures were not taken into account in the present study, but the actual surgical event per se, as the main possible factor influencing ABD formation.

Some limitations in accuracy have been reported for 0.3-mm voxel CBCT image evaluations, although they allow for precise measurements.^27–29  Alveolar bone with a thickness less than 0.3 mm may wrongly be diagnosed as ABD²⁸  and still be overestimated in a false-positive result in very thin bony plate regions.²⁸  Peterson et al.³⁰  also reported an increase in the prevalence of ABD when measured using CBCTs as compared with clinical measurements. On the other hand, it is necessary to analyze the cost-benefit balance based on the principles of ALARA.²⁹  Higher spatial resolution of the image would require a smaller voxel size and consequently a higher dose of radiation.

More clinical studies should be carried out to clarify the prevalence of bone dehiscence in patients after orthognathic surgery.

• Considering the limitations of the study, the null hypothesis was rejected, as both Class II and III patients exhibited a slight increase (about 5%) in the incidence of alveolar dehiscence after orthognathic surgery.

• A temporary reduction in vascular supply and difficulties in oral hygiene may explain this result; however, more studies are needed.