The severity of a palatally impacted canine (PIC) is gauged radiographically on 2-dimensional and 3-dimensional positional components: eg, angulation and height. We hypothesized that the position of a PIC relative to its virtual alignment in the arch is a better indication of impaction severity and treatment requirements. The aims of this research were to evaluate variations in PIC location on 3-dimensional images and to determine positional components associated with impaction severity.
Linear and angular measurements of 38 PICs from 28 cone-beam computed tomography scans were made on the panoramic, coronal, sagittal, and axial sections. Measurements included angulation of the PIC to the virtually aligned canine, midline, and palatal plane; and distances between cusp tip and apex to various reference planes—eg, occlusal and midpalatal. Statistical assessments comprised t tests for group comparisons based on PIC and virtually aligned canine severity (cutoff at 30°) and Pearson product moment correlations for associations among variables.
Angulations of the PIC to the virtually aligned canine were 32.5° ± 15.5° (range, 9°-59°) and 19.6° ± 6.9° and 45.37° ± 9.6°, respectively, in the less severe and more severe groups ( P <0.001). Group differences were significant (0.023 < P <0.001) for the apex and cusp distances between PICs and virtually aligned canines and to the midline reference planes, and for PIC angulations to the palatal plane and midline. Correlations were highest (0.7 <r <0.9; P <0.001) among PIC angulations to virtually aligned canines and to midline planes (panoramic and coronal sections), and cusp to midline distances (panoramic and axial views).
A novel measurement of PIC inclination to its virtual aligned position indicates medial inclination of the most severe PIC with the crown farther from the alveolar crest and the apex more posterior. The crown varied over a wider range in the transverse plane; the apex varied over a comparatively narrower track anteroposteriorly.
Angulation of a palatally impacted canine to its virtual aligned position was measured.
The distance needed to align the impacted canine is an indicator of impaction severity.
Crown position determines severity better than location of the apex.
The delineation of a 30° cutoff for severity seems realistic.
The most frequently impacted teeth after the third molars are the maxillary canines (1%-3%), with most of them in a palatal position. The variations in buccopalatal, vertical, and anteroposterior locations of impaction define treatment complexity and duration. Although severity of impaction and associated treatment difficulty are primarily ascribed to this 3-dimensional (3D) assessment, other critical factors include the applied surgical procedure (also associated with the tooth position), the amount and quality of the covering bone, and the traction mechanics including active force components and anchorage setup.
The various methods to determine impaction severity and relate it to treatment difficulty originated because of 2 major therapeutic side effects: the extended duration of orthodontic treatment and the resorption of adjacent teeth, particularly lateral incisors, reported in nearly 50% of patients with a palatally impacted canine (PIC) and related to unduly sustained forces in a protracted treatment.
Based mostly on 2-dimensional (2D) radiographs (periapical, intraocclusal, and panoramic), PIC severity has been stratified on positional components (horizontality, angulation, height). Ericson and Kurol classified severity through cusp tip position in sectors drawn to the adjacent lateral and central incisors: impactions mesial to the lateral incisors are more severe than those in a more distal position. Stewart et al associated severity and ensuing treatment time with the vertical distance of the cusp tip to the occlusal plane at a threshold of 14 mm. Pitt et al determined that the canine’s horizontal position, vertical height, buccopalatal position, and the patient’s age projected severity and treatment difficulty. Crescini et al found that every increase of 5° in the angle between the PIC and the midline resulted in 1 additional week of treatment. These approaches have not been compared in controlled studies and have not provided consistently predictable outcomes in patients. In addition, 2D radiographs cause variable distortions of anatomic dimensions and overestimated measurements, particularly patient positioning errors during radiography.
Although concerns for excessive radiation initially limit the use of 3D imaging methods, including cone-beam computed tomography (CBCT), the risk is reduced by imaging the specific canine area. New indexes were developed to predict impaction potential. Kau et al calculated a “KPG” index by adding the scores assigned to cusp tip and root tip in the 3 planes of space on the CBCT panoramic and axial sections. Alqerban et al determined that the strongest predictors of impaction were the PIC’s angulation to the lateral incisor, the distance to the occlusal plane, and the crown position relative to the arch and adjacent teeth.
The 3D methods have not yielded more definitively predictive information. Haney et al determined that CBCT changed 2D-based diagnosis and treatment planning of impacted canines in 27% of the evaluations. Although occurring in a relatively low percentage, this difference may be critical for an individual patient and indicates the need to further explore variations of PICs in 3D imaging to improve the assessment of impaction severity and in the future its link with treatment outcomes.
In this article, we introduce a new scheme for severity assessment based on projected treatment outcome. Accurate determination dictated the reliance on 3D CBCT images. We hypothesized that the assessment of the PIC relative to its virtual posttreatment correction would better reflect the severity of impaction by personalizing the impaction to the patient. To this end, we defined the “virtually aligned canine” (VAC) as the simulated aligned tooth in its final posttreatment position in the arch to determine its position in all planes of space ( Fig 1 ). Accordingly, the aim of this study was to determine, based on 3D images of PICs, the positional components associated with impaction severity in relation to treatment objective, not only diagnostic features. Treatment objective was defined as the simulated end position of the canine after treatment. See Supplemental Materials for a short video presentation about this study.
Material and methods
Our material comprised CBCT scans of 28 patients (mean age, 16.06 ± 4.9 years; 16.9 ± 4.9 years for male subjects, 15.7 ± 4.9 years for female subjects) who had 38 PIC s (18 unilateral, 10 bilateral) and sought orthodontic treatment at the American University of Beirut Medical Center in Beirut, Lebanon. The scans were prescribed for accurate localization of the impacted canines after a clinical examination that included an initial diagnostic panoramic or periapical radiograph. This retrospective study was approved by the institutional review board.
CBCT scans were selected according to the following criteria.
Presence of unilateral or bilateral PIC. Canines had been considered at higher potential for impaction when they have not erupted into the oral cavity by the age of 13 years (1 year after the normal maxillary permanent canine eruption age range of 11-12 years ) and at the clinical examination, they were not palpable in the vestibule, prompting further radiographic confirmation. Within this scheme, 2 girls whose initial regional CBCT scans were taken at ages 10 and 11 were included in the study; they were treated nearly 1 year later with exposure of the canine and then with orthodontic traction into the arch. A subsequent preexposure CBCT scan was not taken to minimize radiation; the tooth was followed with periapical radiography.
CBCT scans of good quality and sufficient field of view covering at least half of the maxilla.
The exclusion criteria were craniofacial anomalies or syndromes and x-rays with limited field of view or low resolution that precluded accurate measurements.
Linear and angular measurements, recorded using the Ez3D Plus 3D CDViewer software (version 22.214.171.124; Vatech Global, Gyeonggi-do, Korea), included the following.
On the panoramic section ( Fig 2 ), PIC/VAC angle, defined by the intersection of the axis of the impacted canine and the simulated aligned tooth between the adjacent teeth (lateral incisor and first premolar), determined by drawing a vertical line parallel to these teeth or along the long axis of the primary canine if present. Other measurements included the cusp tip to occlusal plane vertical distance, distances between cusp tip and apex to the midline, and the PIC angulation to the midline.
On the coronal section ( Fig 3 , A ), the PIC angulation to the midline.
On the axial section ( Fig 3 , B ), cusp tip and apex deviations, the transverse projections of the respective distances between the cusp tips and apices of the PIC and VAC; distances between the cusp tip and apex to the midpalatal plane and distance between the mesial aspect of the first molar to prosthion (most anterior point on the alveolar bone crest).
On the sagittal section ( Fig 3 , C ), PIC angulation to palatal plane, cusp tip to occlusal plane (vertical projection), and the anterior projection of the cusp tip to the frontal plane through prosthion.
On the axial and sagittal views involving the apex and cusp tip, whereby both were not simultaneously visible relative to the reference lines, the measurements were obtained by scrolling to the section that contained the apex or the crown tip. The projection of the apical point or the cusp tip was marked on the pertinent section once identified; the distances between the targeted points were automatically provided by the software.
Two subgroups of 18 and 20 teeth ( Table I ) each were categorized based on the severity of the angle PIC/VAC. The cutoff was set at a realistic threshold of 30°, nearly one third of the angle (about 80°) between a normally inclined canine (about 11°) and the most severe possibility of a totally horizontal impaction (0° to the palatal plane).
n = 38
n = 18
n = 20
|Patient age (y)||16.1||4.9||14.7||3.0||17.3||6.0||0.103|
|PIC/VAC (°)||32.5||15.5||19.0||6.6||43.6||9.9||<0.001 ∗|
|Vertical (mm) a||9.5||2.3||8.5||2.2||10.3||2.1||0.016 ∗|
|Cusp to midline (mm) b||6.8||3.9||9||3.5||4.8||3.2||<0.001 ∗|
|Apex to midline (mm) c||19.4||2.6||18.3||2.0||20.3||2.7||0.011 ∗|
|PIC to midline (°) d||35.1||15.4||22.6||9.8||46.4||9.6||<0.001 ∗|
|Vertical (mm) a||10.4||2.3||9.7||2.2||11.0||2.3||0.098|
|PIC to palatal plane (°)||110.6||13.1||103.6||9.0||116.9||13.3||0.001 ∗|
|Cusp deviation (mm)||11.1||3.3||9.9||2.8||12.2||3.2||0.023 ∗|
|Apex deviation (mm)||8.2||2.3||7.1||2.0||9.2||2.2||0.003 ∗|
|Cusp to midline (mm) b||4.9||3.0||6.5||2.6||3.5||2.7||0.001 ∗|
|Apex to midline (mm) c||14.8||1.7||14.5||1.4||15.1||2 .0||0.333|
|First molar to midpalatal plane (mm)||16.6||2.0||16.5||1.8||16.6||2.3||0.900|
|First molar to prosthion (mm)||34.5||2.8||35.5||2.3||33.5||2.7||0.027 ∗|
|PIC to midline (°) d||27.4||15.18||17.2||11.9||36.7||11.6||<0.001 ∗|
Statistics included descriptive computations, based on the angulation severity of the subgroups, as well as bivariate and multivariate analyses. The Pearson correlation coefficient was used to explore associations among variables. To test which variables would predict the severity of the PIC/VAC angulation, associations between the latter, positional variables, and age were evaluated using multivariate regression analysis. To determine intraexaminer reliability, an author (K.Z.) repeated all measurements on 10 CBCT radiographs at least 14 days after the initial assessment. The measurements were evaluated with the 2-way mixed-effects intraclass correlations for absolute agreement on single measures. Statistical significance was set at P <0.05. SPSS software (version 20.0; IBM, Armonk, NY) and Stata software (version 11.1; StataCorp, College Station, Tex) were used for all statistical analyses.
The intraclass correlation coefficients gauging reliability of repeated measurements were high, ranging from 0.91 to 0.997, except for the anterior distance in the sagittal plane (r = 0.80).
The male-to-female ratio of patients with PIC was 1:2.5. The range of the PIC/VAC angle was 9° to 59°, and its mean in the total sample was 32.47° ± 15.46° (19° ± 6.6° and 43° ± 9.9° in the lower and higher severity subgroups, respectively; P <0.001; Table I ). Differences between these subgroups were statistically significant for various parameters: the inclination to the palatal plane (sagittal, P <0.001), apex and cusp distances between PIC and VAC (axial plane, P = 0.023 and 0.003, respectively), apex to midline distance (panoramic, P = 0.011), and inclination to midline (coronal, P <0.001; panoramic, P <0.001) were all greater in the high severity group.
The cusp to midline distance was smaller in the higher severity group (axial, P = 0.001; panoramic, P <0.001), reflecting the more medial position of the crown in this group. The height of the cusp was on average similar between the severity groups on the sagittal view ( P = 0.098) but different on the panoramic view ( P = 0.016), although the difference between the group means was less than 1 mm. The distance between the anteroposterior position of the first molar and the interincisal point in the axial view was smaller in the severe group compared with the less severe group ( P = 0.027; Table I ).
The measurements of similar positions on the panoramic view and other views—ie, the cusp tip to occlusal plane vertical distances, distances between the cusp tip and apex to the midline, and the PIC angulation to the midline—were statistically significantly different (0.015 < P <0.001); only the cusp tip to occlusal plane vertical distances between the severity groups measured on the panoramic and sagittal views were not different ( Table I ).
The highest correlations were observed between the angulations of PIC to VAC and the angulations to the midline on the panoramic (r = 0.9) and coronal (r = 0.85) views, as well as PIC/VAC to the position of its cusp tip on the panoramic (r = −0.7) and axial (r = −0.7) views ( Table II ). The canine angulation to midline was significantly correlated to cusp tip deviation (r = 0.67) and cusp tip to midline (r = −0.80 on panoramic and r = −0.86 on axial views) ( P <0.001). Moderate correlations (0.48 <r <0.57) were noted between PIC/VAC and the deviations of the canine cusp and apex. The highest correlations for cusp deviation were with the cusp to midline distance (r = −0.7) and PIC angulation to midline on the axial view. The highest correlation for apex deviation was with the PIC angulation to the palatal plane (r = 0.62).