VALUE ADDITION OF CBCT IN ORTHODONTIC TREATMENT PLANNING OF IMPACTED CANINES AND CONSIDERATIONS FOR DENTAL IMPLANTS

Eric Strouse¹, Meenakshi Vishwanath¹, Jonathan Feldman¹, Sunil Mutalik¹, Melissa Landin¹, Aditya Tadinada¹

University of Connecticut School of Dental Medicine
University of Nebraska Medical Center, College of Dentistry
University of Connecticut School of Dental Medicine
University of Connecticut School of Dental Medicine
University of Connecticut School of Dental Medicine
University of Connecticut School of Dental Medicine

Objectives: The aim of this study was to assess the value addition of cone-beam computed tomography (CBCT) in the orthodontic treatment planning of impacted canines and considerations for dental implants based on the diagnosis and treatment plan.

Materials and Methods: 35 impacted canines from 17 treated patients were evaluated. The first set of simulated images consisted of traditional 2-dimensional peri-apical radiographs generated on the Clark’s tube shift technique. The second set consisted of CBCT scans, with the evaluators being able to manipulate the scans in all three orthogonal planes. Four orthodontists with varying experience and one oral maxillofacial radiologist completed questionnaires for each impacted canine and the radiographic modality.

Results: The data showed that the evaluators produced varying diagnosis and treatment plans depending on the radiological modality used. The impacted canine was correctly localized with 2D radiography 46% of the time, while evaluators were unsure or incorrect 54% of the time. In identifying root resorption there was 63% uncertainty or disagreement between the methods. There was 25% disagreement or uncertainty in regards to the decision on whether the impacted canine could be brought into occlusion using orthodontic methods. There was 21% disagreement or uncertainty on whether the impacted canine would be treatment planned for extraction.

Conclusions: CBCT may provide a value addition over 2D radiography in more accurate diagnosis and more confident treatment planning, leading to improved patient outcomes.

Keywords: Cone-beam computed tomography, cuspid, impacted, teeth

Introduction

Impacted canines are not an uncommon problem in dental patients. Other than the third molars, maxillary canines are the second most frequently impacted teeth with a prevalence between 1% to 3%, with mandibular canines encountering impactions seen between 1% and 2%. ^{(i, ii, iii, iv, v)}. Due to this relatively common occurrence, managing impacted canines has become a regular task in orthodontic practices. Treatment of impacted canines can be both extended and difficult, and due to the high esthetic demands and functional considerations both surgical and orthodontic intervention is often necessary ^(vi). Impacted canines can also lead to resorption of neighboring permanent teeth, especially the lateral incisor ^(vii,viii).

Accurate diagnosis and fundamentally sound treatment planning are the cornerstone of successful orthodontic treatment ^(ix). Precise localization of the impacted canine in all three planes of space, and its surrounding structures is essential in the orthodontic-surgical management of presenting patients. Traditionally, the use of two-dimensional (2D) radiographic images (panoramic, periapical, occlusal, and cephalograms) have been used in orthodontics. With 2D imaging comes distortions, superimpositions and other inherent issues. Diagnosis and treatment planning can become difficult with conventional radiography due to these reasons. With its lower cost, high spatial resolution and reduced radiation dose compared to medical computed tomography (CT), the development of Cone Beam Computed Tomography (CBCT) is emerging as the imaging modality of choice in clinical cases requiring extensive 3D views.

Should orthodontic traction not be feasible, or the patient does not desire the treatment, surgical removal of the impacted canine is the only solution available ^(x,xi). Due to the resultant poor esthetics and occlusal asymmetry, implant placement is a common subsequent step ^(xii). If adequate residual bone remain after the extraction of the canines, an immediate implant is an option for the patient ^(xiii). These clinical scenarios require knowledge of bone quantity, quality and proximity to critical anatomical structures ^(xiv).

In considering impacted canines, CBCT has been previously shown to be useful for their diagnosis, as well as for the identification of associated sequelae ^(xv), such as root resorption in adjacent incisors. This study was conducted in order to assess the diagnosis and treatment planning of impacted maxillary and mandibular canines as well as considerations for dental implant placement, should the tooth be extracted. We hypothesize that three-dimensional imaging will provide detailed information on the location, presence of PDL space, diagnosis of root resorption and offer superior information in orthodontic treatment planning.

Materials and Methods

Cone beam images were collected from 24 consecutive patients, who were referred for localization of either unilateral or bilateral impacted or ectopically erupting canines (all the images were collected from multiple private orthodontic practices that have been de-identified and archived at the University of Connecticut School of Dental Medicine. The patients were of mixed dentition, aged 11-14 and all identifying characteristics including age, gender or race were removed for purposes of the study. A total of 35 impacted or ectopically erupting canines were studied, including 30 maxillary impactions, 5 mandibular impactions, 8 bilateral impactions, 7 unilateral right impactions, 12 unilateral left impactions. Digital imaging and communications in medicine (DICOM) format of the traditional 2D diagnostic radiographs and CBCT scans were utilized for each patient.

The traditional 2D radiographs included two simulated periapical images using the Clark’s Method of tube shift technique. For each impacted canine, one standard periapical and one employing a distal shift was provided (Figure 1). Volumetric images of the patient’s dentition were obtained from the CBCT scan (Figure 2). Images were viewed on HP Pavillion ZE 2000 computer and 20-inch dual monitor display with a 1600 x 900-pixel resolution. All patient identifiers including name, age, sex and race were removed. The institutional review board of the University of Connecticut approved this study.

Figure 1. Perapical radiographs using Clark’s Method. A, PA of #6; B, Distal shift PA of #6. C, PA of #11; D, Distal shift of #11

Figure 2. A-E, Reconstructed models of #6 and #11 from InVivo CBCT software; G-H, Section views in coronal, axial and sagittal views.

Four orthodontists and one oral and maxillofacial radiologist participated in the study. One orthodontist with over 25 years’ clinical experience, two with 5-10 years’ clinical experience and one in the final year of orthodontic residency. The radiologist had 5 years’ clinical experience. The impacted canines were evaluated over the course of two sessions. Prior to beginning, all evaluators reviewed the questionnaire as a calibration.

The impacted canine cases were ordered randomly. The periapical radiographs using simulated Clark’s tube shift Technique were assembled onto a PowerPoint slide presentation showing the two periapicals for each impacted canine, a standard periapical along with a periapical utilizing a distal shift were presented. The CBCT scans were ordered randomly and evaluators were able to freely manipulate the scans using the InVivo software to simulate actual radiological clinical practices. The viewing conditions (room lighting and display monitor settings) were standardized. Clinicians analyzed the images and completed the questionnaires independently. A total of 70 questionnaires were completed by each evaluator, one for the 2D radiographs and one for the CBCT scan for each impacted canine. A board-certified Oral and Maxillofacial Radiologist with extensive clinical experience scored the 35 CBCT cases for the following three variables: location of the impacted canine, is the impacted canine being blocked by any adjacent teeth or anatomic structures and whether the PDL space of the impacted canine can be visualized.

Statistical Analysis

The data was analyzed after collected using IBM SPSS Statistics software version-22. Interrater reliability was performed with an Alpha Cronbach’s test for each variable (Table 1). Descriptive statistics with counts of agreement and ability to diagnose and treatment plan for each imaging modality were calculated. These were used as a measure of difference between CBCT and 2D imaging.

Table 1. Cronbach alpha scores for CBCT and 2D for each variable between the five evaluators.

Results

For identify the location of the impacted canine, the crown of the impacted tooth was used to make the diagnosis of its buccal/lingual location. The five evaluators were able to correctly identify the location 46% of the time for the 35 impacted canines(175 total responses), with a range of 28.6% to 62.8% (Figure 3) Evaluators were unable to make a diagnosis using 2D radiography 24% of the time, and produced an incorrect diagnosis 30% of the time.

In diagnosing whether the impacted canine was being blocked by any adjacent teeth or anatomic structures, the five evaluators were in agreement with their diagnosis between the two imaging modalities 52% of the time (Figure 4) 42% were unable to make a diagnosis using 2D radiography. In diagnosing whether the PDL space of the impacted canine could be visualized, the five evaluators were in agreement 42% of the time (Figure 5). 32% of the time, evaluators were unable to make a determination when viewing only the 2D radiography. There were differences in diagnosis between the ability to visualize the PDL space for the two modalities 19% of the time.

In diagnosing whether the crown of the impacted was in contact with the adjacent incisor a cutoff of less than 1mm was used. There was a 30% agreement in diagnosis between the two modalities (Figure 6). The evaluators were unable to make a diagnosis using 2D radiography 46% of the time. Assessing the state of root development produced the same diagnosis between modalities among the five evaluators 57% of the time (Figure 7). Evaluators were unsure of a diagnosis using 2D radiography 8% of the time, and 45% of the time produced differing diagnosis.

In detecting root resorption, the five evaluators were in agreement on their diagnosis 38% of the time for the 35 canines (Figure 8). 43% were not sure in making a diagnosis using 2D radiography. For 14% of the cases, evaluators did not detect the presence of root resorption using 2D radiography, but were able to diagnose root resorption using CBCT. In 6% of the cases, evaluators diagnosed the presence of root resorption with 2D radiography, but did not with CBCT.

In predicting whether the impacted canine would erupt unassisted, evaluators were in agreement between 2D and CBCT 79% of the time for the 35 impacted canines (Figure 9). Evaluators were not sure in making a diagnosis using 2D radiography 15% of the time. The evaluators were asked if they would make the decision to bring the impacted canine into occlusion using orthodontics. In 75% of the cases, they were in agreement with their treatment plan between 2D and the CBCT (Figure 10). In 15% of the cases, they were not sure using 2D radiography, but did make a treatment planning decision using the CBCT.

The evaluators were prompted additionally if they would choose to extract the impacted canine from the radiographic imaging provided. For 79% of the cases, evaluators were in agreement for both the 2D and CBCT imaging (Figure 11). In 15% of the cases, they were not sure using 2D radiography, but did make a decision using the CBCT. There was one case for which three of the evaluators all noted that they would consider extracting the lateral incisor, and consider placing a dental implant in the edentulous zone. They all said they would request a CBCT for treatment planning purposes.

Questions for dental implant considerations were answered if there was a possibility of implant placement (either “yes” or “not sure” for treatment planning extraction of the impacted canine). There were 20 total cases in which evaluators said they would not consider planning a dental implant from viewing the CBCT, but were “not sure” from the 2D radiographs (Figure 12). Furthermore, when dental implants were considered, evaluators were asked whether they could visualize various aspects of anatomy from the various radiographic modalities. In 100% of the cases, evaluators said they could visualize bone height, trabecular pattern and proximity to critical structures using both 2D and CBCT (Figure 13). In 100% of the cases, evaluators said they could visualize bone width, buccal and lingual cortical plates with CBCT, while 0% said they could visualize these qualities from the 2D radiography.

Discussion

Orthodontic treatment planning for impacted canines depends on a multitude of factors. Being able to accurately localize the impacted canine is very important in regards to treatment planning. Additionally, impacted canines can present other difficulties including root resorption of adjacent teeth which can have significant consequences. This study builds on some early research comparing various radiographic modalities in planning impacted canines. Periapical radiographs are traditionally a common choice as a means for locating objects buccal-lingually using two-dimensional imaging with the Clark’s shift technique.

Due to the complexity of impacted canines, accurate diagnosis is highly important for proper management. The ability to precisely locate an impacted canine in space makes for improved surgical and orthodontic planning. Clinicians need to visualize an impacted canine is in relation to its neighboring teeth and other anatomic structures in order to plan for future treatment. Visualization of other variables, such as the presence of an intact PDL space and root development can also play important roles in diagnosing the eruption potential of these teeth. By possessing superior diagnostic information, clinicians can plan with more confidence, leading to fewer mismanaged cases. Our study had an overall findings suggesting a decreased ability to make confident diagnosis and treatment using 2D imaging alone. Uniformly across the variables in both diagnosis and treatment planning, there was a clear benefit in the use of CBCT.

To produce a true gold standard would have required an anatomical dissection. This was not a viable option, both practically and ethically. With its 1:1 anatomical accuracy, along with a lack of distortion or artifact, CBCT was used as the gold standard for the purposes of localization. This was confirmed by a very high interrater reliability score as evaluators diagnosed the location using the CBCT, the high agreement confirming the accuracy of the imaging. Surprisingly, the percentages of accurate diagnosis was lower than other studies comparing localization using periapical radiographs. Ericson and Kurol ⁽¹⁶⁾ demonstrated only 8% of the impacted canines not being able to be accurately localized labiopalatally using periapical images. Despite the lack of a true gold standard, interrater agreement was high for each of the variables for the CBCT scans. The 2D imaging produced lower agreement. We attribute this to the true accuracy of CBCT. Evaluators were able to make more consistent and true diagnoses, and therefore more appropriate treatment plans. The lower kappa scores for 2D images were indicative of a decreased ability to make orthodontic diagnoses using this imaging alone. Improper localization has a direct relationship to patient outcomes. One study analyzed the treatment of 37 impacted canines that were taken as “failed”. Of the failures, 40.5% were due to mistaken localization⁽¹⁷⁾

Additionally, although evaluators were standardized in regards to their understanding of the questionnaire, they all came from various backgrounds with different training and treatment philosophies. One orthodontist serving as an evaluator, with over 25 years of clinical experience, was not trained with three-dimensional imaging and does not use it in his private practice. For this reason, he was less comfortable manipulating the three-dimensional software which may have impacted his responses to our survey. Other evaluators were trained in various countries, with different treatment philosophies and methods, taking away from the uniformity of the study. This variety of evaluators does, however, provide a wide mix of philosophies brought together and averaged for the purposes of this study. There will be inherent disagreement in many aspects of diagnosis and treatment planning in orthodontics. With improved diagnostic imaging, we found diagnosis and treatment planning to be more consistent. This was evident in our study, causing difficulty in producing detailed statistical measures.

Several authors have confirmed that when using three-dimensional imaging root resorption is more prevalent than it is assumed to be, and conventional radiographic imaging has been shown to be inadequate in detecting root resorption (Alqerban et al. 2011). CBCT eliminates the problems with conventional imaging and allowing for an incraesed ability to detect root resorption. Our study had similar had similar findings, in only 38% of the cases was the detection of root resorption similar between CBCT and 2D imaging. There was a large percentage of cases (43%) in which a presence or lack-thereof root resorption was made with CBCT but evaluators were not sure when viewing 2D imaging. Haney et al. had similar findings, in which 64% of judges were in agreement between 2D and CBCT, with a wide range of 36% to 86% in the judge’s agreement. This study builds on that by increasing the sample size from 25 to 35, surpassing the minimal sample size for a normal bell curve distribution. Early detection of root resorption is of great importance. If not detected early, the loss of an incisor is a possibility, making the increased diagnostic ability of CBCT to be very beneficial.

Due to their difficulty and possibility of various sequelae, impacted canines have been studied extensively in the literature. It is known that CBCT has the capability to produce more accurate diagnoses due to its inherent properties. Our study showed that there was a 79% agreement between the two imaging modalities in the decision to treatment plan orthodontic treatment. The remaining 21% produced varying treatment plans between the imaging types or an inability to make a plan using the 2D imaging alone, a large proportion of the cases. With such uncertainty in using 2D imaging alone and the potential difficulty of these cases, it may be a suggestion for use of CBCT in treatment planning of impacted canines. Other images that are traditionally taken, such as a panoramic, lateral cephalogram and more are also able to be generated from a CBCT, aiding the provider in the treatment planning process. Further, with an assessment of the entire craniofacial complex, better assessing the airway, TMJ and other possible incidental findings, allowing for the best possible patient outcomes.

This is one of the first studies to look at the consideration of treatment planning dental implants in the situation of impacted canines. Some patients may not desire orthodontic treatment, and bringing an impacted canine into occlusion may sometimes not be feasible. A case study from Garcia et al. investigated immediate implants after the removal of impacted maxillary canines in nine patients. Practitioners will be interested in what situations implant placement is an option, and what radiographic modalities will be necessary. Our study found 20 cases in which the evaluator was not sure about placing an implant when viewing 2D imaging, but would not consider an implant when viewing the CBCT of the same case. In other cases practitioners made varying decisions between imaging modalities. The evaluators noted that they were considering placing an implant, they would request for a CBCT. This indicates that consideration of extraction and implant placement will have to be taken on a case by case basis. It is well known that CBCT provides superior diagnostic capabilities in the treatment planning and placement of dental implants (Tyndall et al. 2012). Our study supported this, with evaluators responding that they were able to visualize important variables for implant placement not visualized with 2D imaging. While evaluators did state they were able to view the proximity to critical structures using 2D imaging, it has been shown that in the placement of mini implants CBCT significantly reduces root perforation compared to other imaging, indicating that avoiding anatomical structures may require more accurate imaging⁽¹⁸⁾.

Although there is little in published literature regarding the financial cost of CBCT, the cost of low radiation three-dimensional imaging using CBCT is becoming more affordable. It is important for the benefits of this imaging modality to become available for practitioners. In addition to the technology becoming more affordable, radiation doses of three-dimensional CBCT are becoming lower. CBCT imaging for comprehensive orthodontic patients is approximately 65 μSv compared with about 26 μSv for a lateral cephalogram and a panoramic image taken with a digital machine with subsequent low-dose protocols for orthodontics 35-40 μSv. This amount is lower than a full-mouth series of intra-oral radiographs to assess periodontal status⁽¹⁹⁾. This, combined with the superior diagnostic information provided by CBCT, may make this modality a valuable option for orthodontists in the diagnosis and treatment planning of impacted canines. A randomized clinical trial to further analyze these same questions is not viable option, as it is unethical to expose patients randomly to additional radiation solely for the purpose of a research study, without a clinical or radiological basis.

Conclusions

These results suggest that three-dimensional CBCT may produce improved accuracy in the diagnosis of impacted canines. When compared with 2D imaging, inconclusive to erroneous treatment plans can be produced. Three-dimensional imaging may have the ability to offer advantages in proper diagnosis and treatment planning due to its superior information, allowing for improved overall outcomes and patient care.

Funding

This research received no specific grant or financial support from any funding agency in the public, commercial or not-for-profit sectors.

1 Preda L, La Fianza A, Di Maggio EM, Dore R, Schifino MR, Campani R. The use of spiral computed tomography in the localization of impacted maxillary canines. Dentomaxillofacial Radiology, 26, 236-41, 1997

2 Walker L, Enciso R, Mah J. Three-dimensional localization of maxillary canines with cone-beam computed tomography. Am J Orthod Dentofacial Orthop, 128, 418-23, 2005.

3 Chawla, Sonia et al. Impacted Canines: Our Clinical Experience. International Journal of Clinical Pediatric Dentistry, 4.3, 207–212, 2011.

4 Yavuz MS, Aras MH, Büyükkurt MC, Tozoglu S. Impacted Mandibular Canines. J Contemp Dent Pract, (8)7, 078-085, 2007.

5 Shah RM,Boyd MA,Vakil TF. Studies of permanent tooth anomalies in 7,886 Canadian individuals. II: congenitally missing, supernumerary and peg teeth. Dent J, 44:265-8, 1978.

6 Alqerban, R. Jacobs, P. Lambrechts, G.Loozen, G.Willems, Root resorption of the maxillary lateral incisor caused by impacted canine: a literature review. Clin Oral Investig, 13, 247-255, 2009.

7 Ericson S, Kurol PJ. Resorption of incisors after ectopic eruption of maxillary canines: a CT study. Angle Orthodontics, 70(6):415-23, 2000.

8 Cuminetti F, Boutin F, Frapier L. Predictive factors for resorption of teeth adjacent to impacted maxillary canines. Int Orthod, (1):54-68, 2017.

9 Tadinada A, Marckaz A, Yadav S and Mukherjee P. Applications of Cone Beam Computed Tomography in Orthodontics: A Review. Turkish Journal of Orthodontics. 29, 73-79, 2016.

10 Demarosi F, Varoni E, Rimondini L, Carrassi A, Leghissa GC. Immediate Implant Placement After Removal of Maxillary Impacted Canine Teeth: A Technical Note. Int J Oral Maxillofac Implants, 31(1), 191-4, 2016.

11 García B, Boronat A, Larrazabal C, Peñarrocha M, Peñarrocha M. Immediate implants after the removal of maxillary impacted canines: a clinical series of nine patients. Int J Oral Maxillofac Implants,24(2), 348-52, 2009.

12 Cardaropoli D, Debernardi C, Cardaropoli G. Immediate placement of implant into impacted maxillary canine extraction socket. Int J Periodontics Restorative Dent, 27(1), 71-7, 2007.

13 Peñarrocha M, Peñarrocha M, García-mira B, Larrazabal C. Extraction of impacted maxillary canines with simultaneous implant placement. J Oral Maxillofac Surg, 65(11), 2336-9, 2007.

14 Tyndall DA,Price JB,Tetradis S. American Academy of Oral and Maxillofacial Radiology. Position statement of the American Academy of Oral and Maxillofacial Radiology on selection criteria for the use of radiology in dental implantology with emphasis on cone beam computed tomography.Oral Surg Oral Med Oral Pathol Oral Radiology, 113:817-26, 2012.

15 Alqerban, Ali et al. The Effect of Using CBCT in the Diagnosis of Canine Impaction and Its Impact on the Orthodontic Treatment Outcome. Journal of Orthodontic Science, 3.2, 34–40, 2014.

16 Ericson S, Kurol J. Radiographic assessment of maxillary canine eruption in children with clinical signs of eruption disturbance. Eur J Orthod, 8, 133-40, 1986.

17 Becker A, Chaushu G, Chaushu S. Analysis of failure in the treatment of impacted maxillary canines. Am J Orthod Dentofacial Orthop, 137(6), 743-54, 2010.

18 Landin M, Jadhav A, Yadav S, Tadinada A. A comparative study between currently used methods and Small Volume-Cone Beam Tomography for surgical placement of mini implants. Angle Orthod. 85, (3):446-53, 2015.

19 Larson BE. Cone-beam computed tomography is the imaging technique of choice for comprehensive orthodontic assessment. Am J Orthod Dentofacial Orthop, 141(4):402, 404, 406, 408, 410, 2012.

Table and Figure Legends

Table 1. Cronbach alpha scores for CBCT and 2D for each variable between the five evaluators.

Figure 1. Perapical radiographs using Clark’s Method. A, PA of #6; B, Distal shift PA of #6. C, PA of #11; D, Distal shift of #11

Figure 2. A-E, Reconstructed models of #6 and #11 from InVivo CBCT software; G-H, Section views in coronal, axial and sagittal views.

Figure 3. Percentages of agreement and differences for the location of the impacted canine (buccal, lingual/palatal, mid-alveolar, Not Sure) between the 2D traditional radiographs and the 3D CBCT.

Figure 4. Percentages of agreement between the 2D traditional radiographs and the 3D CBCT on whether or not the impacted canine is being blocked by any adjacent teeth or anatomic structures (B, blocked; NB, not blocked; NS, not sure.

Figure 5. Percentages of agreement between the 2D traditional radiographs and the 3D CBCT on whether or not the PDL space of the impacted canine can be visualized (NS, not sure).

Figure 7. Percentages of agreement between the 2D traditional radiographs and the 3D CBCT on the root development of the impacted canine.

Figure 8. Percentages of agreement between the 2D traditional radiographs and the 3D CBCT on the ability to detect the presence of root resorption (Y, yes; N, no; NS, not sure).

Figure 9. Percentages of agreement between the 2D traditional radiographs and the 3D CBCT on whether or the impacted canine will erupt unassisted (Y, yes; N, no; NS, not sure).

Figure 10. Percentages of agreement between the 2D traditional radiographs and the 3D CBCT on the decision to perform orthodontic treatment to bring the impacted canine into occlusion (Y, yes; N, no, NS, not sure).

Figure 11. Percentages of agreement between the 2D traditional radiographs and the 3D CBCT on whether to extract the impacted canine (Y, yes; N, no, NS, not sure).

Figure 12. Description of cases in which implants were considered, with the number of cases show for each situation.

Figure 13. Percent of practitioner’s ability to visualize anatomic variable considered in the placement of dental implants for both 2D and 3D CBCT imaging.

Table and Figure Legends