INTRODUCTION

Maxillary impacted canines are commonly encountered in orthodontics; their frequency is well reported in the general population.1–4 Approximately 2.5% of maxillary canines become impacted. Maxillary impacted canines are second only to maxillary third molars in frequency of dental impactions.3,5–7 Incidence is greater in females,5,8–10 with palatal impactions occurring twice as often in females.9 Impactions occur in a palatal, mid-ridge, or labial position, with most becoming impacted in the palatal location.11–13 Displaced canines cannot be palpated either buccally or palatally 85% of the time and have been shown to cause resorption of adjacent teeth.5,14–16 Increased treatment time17,18 and compromised esthetic results19 are also associated with impacted maxillary canines.

A variety of causes for canine impactions are hypothesized.20 Palatal impactions are associated with various dental anomalies, including anomalies of upper lateral incisors, aplasia, and impaction of other teeth.21 Palatal impactions have sufficient space for eruption 85% of the time,13 while labial impactions have inadequate arch space.22 The genetic origin of palatally displaced canines appears to be a product of polygenic multifactorial inheritance.23 In addition to genetic influence, environmental factors, such as peg-shaped laterals or diminutive lateral incisors, are also contributory to canine impactions.24

Studies examining a relationship between maxillary impacted canines and transverse discrepancies are contradictory. A premaxillary skeletal deficiency is associated with labial impactions, while maxillary transverse excess is associated with palatally impacted canines.25 Anterior transverse deficiencies are significant in patients with canine impactions.26 Conversely, Langberg and Peck27 state that there is no difference in the anterior and posterior maxillary arch width of subjects with palatally displaced canines. Saiar et al.28 show that in the evaluation of postero-anterior cephalograms, no difference is found in the width ratio of the maxilla and mandible when comparing impacted maxillary canines to Class I and Class II malocclusion controls without impactions.

The treatment of impacted canines varies depending on the clinical situation. Extraction of the primary canine frequently allows the impacted permanent canine to erupt, but the probability of eruption decreases as the canine cusp tip moves more mesially.29 Expansion either of the arch or local area is necessary in some cases to create adequate space for the canine to erupt.30 Some cases require surgical exposure19 and/or extraction and expansion to facilitate eruption. The decisions concerning surgically exposed impacted canines have not been analyzed to determine what factors influence these decisions.

The literature is seemingly replete with research on maxillary impacted canines1–32; however, little research examines predictors for orthodontic treatment choice. This retrospective study sought to determine the relationship between the choices of orthodontic treatment of surgically exposed unilateral maxillary impacted canines and designated pretreatment variables. The clinical implications would extend to providing the orthodontist with a better understanding of the variables that influence or predict orthodontic treatment choices.

MATERIALS AND METHODS

The sample for this retrospective study came from a prior study that examined maxillary impacted canines in 207 subjects (137 females, 70 males) drawn from three private orthodontic offices. The original sample consisted of 153 unilateral maxillary impacted canine cases and 54 bilateral maxillary impacted canine cases. Complete pretreatment orthodontic records were identified for each subject. Initial inclusion criteria were (1) agreement by two independent examiners, using study models and radiographs, that the canines exhibited true ectopic impaction and (2) all canines required surgical exposure for retrieval. Patients with impacted canines that self-erupted after exfoliation or removal of the primary canines were not included.

In the initial study, pretreatment study models were evaluated to determine interarch molar relationships. Interarch molar relationship was recorded on both right and left sides as Class I (mesiobuccal cusp of the maxillary first permanent molar + or −<1 mm from the buccal groove of the mandibular first permanent molar), end-on Class II (mesiobuccal cusp of the maxillary first permanent molar between 1 mm anterior to the buccal groove of the mandibular first permanent molar and the embrasure between the mandibular first permanent molar and second premolar), full-step Class II (mesiobuccal cusp of the maxillary first permanent molar in or mesial to the embrasure between the mandibular first permanent molar and second premolar), or Class III (mesiobuccal cusp of the maxillary first permanent molar >1 mm distal to the central groove of the mandibular first permanent molar). Two examiners classified the 414 sides independently. Regarding interarch molar classification, 14 of 414 sides were initially disagreed on, representing a 96.62% agreement rate. Differences in classification were noted and reexamined by both examiners together. A classification was then agreed on, bringing agreement to 100%.

The location of each impacted canine was determined from the surgical report and/or the presence of an obvious canine bulge on the model and was recorded as facial, mid-ridge, or palatal. Canine angulation was measured on the pretreatment panoramic radiograph using a protractor with the center point positioned on the incisal tip of the canine and the reference axis parallel to the alveolar bone height between the first and second molars on the right and left sides. The angle of the ray from the incisal tip and the root apex to the reference axis was measured. Greater than 45° was considered a vertical impaction; 45° and less was considered a horizontal impaction (Figure 1).

View Figure

• Download as Powerpoint
• Download Figure

High-resolution images of the casts were obtained of the 207 maxillary and mandibular study casts using a Canon 5D camera with a 100-mm f2.8 macro lens and ring flash mounted on a tripod. The images were captured on Av (aperture preferred mode) at f32, ISO 500, and evaluative metering. Image quality was set to “fine.” Canon “remote capture” was used to enable computer screen viewing and keyboard capture of the images. A 20-mm calibration scale was placed at the occlusal plane of each image to compensate for magnification.

The inclusion criteria for this portion of the study was (1) unilateral maxillary impacted canine; (2) no prior orthodontic treatment; (3) white ethnicity only; (4) eruption of permanent maxillary and mandibular central incisors, lateral incisors, first molars, and permanent mandibular canines; (5) eruption of maxillary and mandibular first and second premolars and/or primary maxillary and mandibular first and second molars; and (6) eruption of the impacted canine after exposure. Eruption was defined as having both the mesial and distal contact points visible on the occlusal images. Instances in which a missing or unerupted tooth prevented digitization of a landmark and recording of a transverse dimension resulted in elimination of that measurement. Cusp tips constructed as part of a dental restoration resulted in exclusion of the patient's records. Of the 153 unilateral maxillary impacted canine patients, 97 patients (59 females and 38 males) aged 14.2 ± 1.9 years met the inclusion criteria.

For the purpose of this study, the following variables were evaluated: (1) gender, (2) age, (3) molar classification, (4) canine location, (5) canine position, (6) available canine space, (7) maxillary transverse dimensions, (8) mandibular transverse dimensions, (9) maxillary incisor angulation, (10) maxillary incisor position, (11) mandibular incisor angulation, (12) mandibular incisor position, (13) maxillary/mandibular skeletal relationship, and (14) lower incisor crowding.

Following calibration on the 20-mm scale, the points listed in Table 1 were digitized by one investigator (B.S.). Maxillary and mandibular anterior and posterior transverse dimensions, mandibular incisor crowding, and available space for the maxillary impacted canine were assessed using a custom J2SE Java application (Table 1; Figure 2). The digitization point for cusp tips was defined as the center of the cusp tip of interest. In instances in which occlusal wear prevented the cusp tip from being identified, the center of the wear facet was used. Incisor crowding was assessed as the summed displacement of adjacent anatomic contact points of the mandibular anterior teeth, as described by Little (Figure 3).33

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Table 1 Digitized Maxillary and Mandibular Landmarks and Parameters

View Fullscreen

Cephalometric radiographs coinciding with the orthodontic models were imported and traced by the same investigator (B.S.) using Dolphin Imaging, version 11.0 (Dolphin Imaging, Chatsworth, Calif; Figure 4). Maxillary and mandibular skeletal relationship, maxillary incisor inclination and position, and mandibular incisor inclination and position were determined as defined in Table 2 and Figure 4.

View Figure

• Download as Powerpoint
• Download Figure

Table 2 Cephalometric Landmarks and Measurements

View Fullscreen

Actual orthodontic treatment, based on treatment notes, was classified as (1) extraction, (2) expansion, (3) extraction and expansion, and (4) nonextraction and nonexpansion and was associated with the overseeing orthodontist. Maxillary expansion was defined as expansion achieved through use of a hyrax expander.

Intraexaminer reliability was examined a priori using two measurement sessions of 10 subjects each, with coded and randomized casts and cephalograms. Reliability was also assessed during and after primary data collection, using the same 10 subjects, for a total of four reliability measurement sessions.

RESULTS

All ICCs were ≥0.95, and reliability was considered excellent. Tables 3 and 4 provide descriptive statistics for the categorical and continuous variables, respectively. Approximately 75% of the subjects had no extraction and no expansion. Therefore, because of this limitation, the analyses grouped subjects with any extraction and/or expansion and compared them with subjects with no extraction or expansion.

Table 3 Descriptive Categorical Variables

View Fullscreen

Table 4 Descriptive Continuous Variables

View Fullscreen

Subjects with Class II end-on molars on the unaffected side were less likely to have extraction and/or expansion (Table 5). Subjects with extraction and/or expansion had decreased L1-MP, CS, MaPWI, MaPWII, MaMWI, and MaMWII and increased mandibular incisor Irregularity Index than nonextraction/nonexpansion subjects (Table 6).

Table 5 Categorical Variables' P Values for the Comparison Between Extraction and/or Expansion and Nonextraction and Nonexpansion

View Fullscreen

Table 6 Continuous Variables' P Values for the Comparison Between Extraction and/or Expansion and Nonextraction and Nonexpansion

View Fullscreen

Using a multiple-variable model, canine space was the single most important predictor of extraction and/or expansion (lower for extraction/expansion subjects), followed by maxillary molar transverse dimension (lower for extraction subjects) and mandibular incisor Irregularity Index (higher for extraction/expansion subjects). The odds of extraction/expansion were decreased by 0.58 for every 1-mm increase in canine space and 0.59 for every 1-mm in maxillary molar width. Conversely, the odds of extraction/expansion increased by 1.47 for every 1-mm increase in mandibular incisor Irregularity Index.

DISCUSSION

This study agrees with others5,8–10 that maxillary impacted canines occur more frequently in females than in males; 61% were female and 39% were males. Impactions also occurred with greater frequency in the palate (81%), agreeing with prior studies.11–13 No difference was noted in molar classification between affected and unaffected sides. Distribution was relatively even between occurrence on either the right or left, as well as molar classification on the affected side or nonaffected sided.

A limitation of this study resulted from the relatively small number of subjects who required extraction, expansion, or extraction and expansion. This resulted in the comparison of extraction and/or expansion to nonextraction and nonexpansion with the continuous and categorical variables.

Orthodontic treatment was not influenced by sex, the impacted side, location, position, or orthodontist (Table 5). However, subjects with Class II end-on molars on the nonaffected side were less likely to have extraction and/or expansion. One possible explanation may be that the rotation of the molar into a Class I relationship allowed for more space to resolve crowding.

While the aim of this study was not to compare opposing arch widths to evaluate whether they play a role in impacted canines,25–28 a constricted upper arch was significant when looking at extraction and/or expansion treatment. Contrary to our study, available canine space is not reported as an issue in palatal impactions.13 However, in certain instances, space may need to be gained through extraction and/or expansion when treating patients with unilateral maxillary impacted canines with decreased lower incisor to mandibular plane (L1-MP), available canine space, constricted maxillary premolars, and molar transverse dimensions and increased mandibular incisor Irregularity Index (Table 6).

Evaluation for predictors that might influence an orthodontist's decision to extract and/or expand was determined using a multivariable model. The outcome of treatment was not predicted, only the variables that influenced the choice of treatment. Canine space was the most important factor followed by maxillary molar transverse dimension and the mandibular incisor Irregularity Index. Every 1-mm increase in canine space and maxillary molar transverse dimension (MaMWII) decreased the odds of extraction and/or expansion (or increased the odds of nonextraction and nonexpansion). Every 1-mm increase in the mandibular incisor Irregularity Index increased the odds of extraction and/or expansion (or decreased the odds of nonextraction and nonexpansion). It should be noted that because of the relatively small number of subjects treated with extraction and/or expansion, a sample containing a larger number with this method of treatment should confirm this finding. However, the power of the sample size used in this study was adequate as defined.

CONCLUSIONS

• Available canine space, maxillary transverse dimension at the molars, and mandibular incisor Irregularity Index serve as indicators for extraction and/or expansion in patients involving unilateral maxillary impacted canines requiring surgical exposure.

• The results also suggest many of these patients are treated without extraction and/or expansion.