The objective of this study was to compare the maxillary transverse dimensions between subjects with impacted maxillary canines and subjects without canine impactions, with similar vertical and sagittal features.
In this retrospective study, 86 cone-beam computed tomography images of subjects with impacted maxillary canines (45 unilateral, 41 bilateral) and 67 images of subjects without dental impactions (control group) matched by similar vertical (NSAr, SArGo, ArGoMe) and sagittal (ANB, SNA, APDI) skeletal characteristics, were analyzed. The maxillary width was measured at 4 levels: first molar basal width, first molar alveolar width, first premolar basal width, and first premolar alveolar width. Group comparisons were performed with analysis of variance and post-hoc Scheffé tests. The influence of group features on the transverse dimensions was evaluated by a multiple linear regression analysis.
Groups with unilateral and bilateral impacted maxillary canines showed significantly smaller first molar basal widths, first molar alveolar widths, and first premolar alveolar widths compared with the control group ( P = 0.030, P <0.001, and P <0.001, respectively). First premolar basal widths were not significantly different among the groups.
Subjects with unilateral or bilateral impacted maxillary canines have smaller maxillary transverse dimensions than do subjects without impaction. Orthodontists should consider the relationship of maxillary width and canine impaction during diagnosis and treatment planning.
This was a CBCT study of subjects with unilateral and bilateral impacted maxillary canines.
Subjects without canine impaction were used as the control group.
Maxillary transverse dimensions were compared between the 3 groups.
Maxillary impacted canines subjects show smaller maxillary transverse dimensions.
A common clinical finding in orthodontics is impacted maxillary canines. This dental eruption anomaly could be associated with morphologic variations in the maxillofacial and dentoalveolar structures. The maxillary canine is the third most impacted tooth after the maxillary and mandibular third molars. The prevalence of this condition varies depending on the evaluated population and has a range of 0.9% to 4.7%. The etiology of palatally impacted canines is mainly associated with 2 theories: growth direction and the genetic theory, whereas the etiology of labially impacted canines appears to be dental crowding.
Different studies have focused on evaluating the dental characteristics of subjects with impacted maxillary canines and have found that they are related to maxillary lateral incisor anatomy and agenesis, among other findings. Other authors have studied the skeletal sagittal pattern of subjects with and without impacted maxillary canines and reported no significant differences between groups. However, Cernochova and Izakovicova-Holla found greater prevalences of Class I skeletal sagittal pattern in subjects with palatally displaced canines and Class III skeletal sagittal pattern in subjects with labially displaced canines. On the other hand, Basdra et al associated canine impaction with Class II Division 2 malocclusion. Regarding the associated vertical pattern, some studies found a greater tendency of hypodivergence, and other authors mentioned normal patterns. In transverse measurements, there are differences among studies. McConnell et al associated a transverse deficiency of the maxilla with palatally displaced canines, contrary to other studies reporting a relationship between greater maxillary transverse dimensions and canine impaction. However, these studies did not match the groups to make true comparisons. Other authors did not find significant differences in maxillary width, but these studies did not have an adequate control group. For these reasons, it seems necessary to clarify whether impacted maxillary canines have a relationship with the transverse skeletal dimensions of the maxilla.
Currently, cone-beam computed tomography (CBCT) is the most complete and efficient imaging tool for diagnosis and planning of impacted tooth treatment. By applying the ALARA principle, the analysis can be carried out in 3 dimensions, with a minimal dosage and high precision. The size, position, and potential effects of this eruptive anomaly can be analyzed with greater advantages over 2-dimensional images.
Measurement of the maxillary transverse dimension is still controversial. There are no studies comparing these dimensions between subjects with unilateral and bilateral impacted maxillary canines and subjects without dental impactions, with similar vertical and sagittal characteristics, using CBCT. Therefore, the objective of this study was to compare the maxillary transverse dimensions of subjects with unilateral and bilateral impacted maxillary canines and a control group without dental impactions, with similar vertical and sagittal characteristics, using CBCT. The null hypothesis was that there are no differences in the maxillary transverse dimensions in subjects with impacted maxillary canines compared with a matched control group without impaction.
Material and methods
This retrospective study was approved by the institutional ethics committee of Universidad Científica del Sur, Lima, Perú. The sample consisted of CBCT images of subjects from 2 private diagnostic centers in Lima, Perú. Sample size was calculated considering a mean difference of 3.6 mm in the intermolar distance as a clinically relevant difference between the bilateral impacted maxillary canine group and the control group, using a standard deviation of 5.5 (obtained from a previous pilot study) with a 2-sided significance level of 0.05 and power of 80%. Although a minimum of 37 subjects per group was required, pretreatment records of 86 subjects with at least 1 impacted maxillary canine (45 unilateral, 41 bilateral; Fig 1 ) were used for the study groups, and 67 subjects with similar vertical and sagittal characteristics but without dental impactions were used as the control group. The selection criteria required that subjects have 2 mm or less of anterior dental crowding (measured between the central and lateral incisors). Subjects with previous orthodontic treatment, cleft lip or palate, craniofacial anomalies, head and neck syndromes, tumors, trauma or history of trauma, absence or dental agenesis, or other maxillary lesions were excluded.
CBCT scans were acquired with Picasso Master 3D (Vatech, Hwaseong, South Korea) set to 8 mA, 90 Kv, and exposure time of 20 seconds, with a flat panel detector 25 × 20 cm, and a field of view of 20 × 19 cm. CBCT synthesized cephalograms were obtained to match the groups by vertical and sagittal characteristics. Vertical growth pattern (mesofacial, brachyfacial, dolichofacial) was evaluated by measuring the nasion-sella-articulare angle (NSAr), sella-articulare-gonion angle (SArGo), and articulare-gonion-menton angle (ArGoMe). Skeletal sagittal relationship (Class I, Class II, Class III) was evaluated by ANB and APDI angles. Maxillary sagittal position (normal, retrusive, protrusive) was evaluated by measuring sella-nasion-A point angle (SNA), palatal plane (PP)/anterior cranial base (ACB), and ratio (PP/ACB) ( Table I ; Fig 2 ).
|Vertical growth pattern parameters|
|NSAr||The angle between nasion (N), sella (S), and articulare (Ar). Used for Bjork-Jarabak polygon sum.|
|SArGo||The angle between sella (S), articulare (Ar), and gonion (Go). Used for Bjork-Jarabak polygon sum.|
|ArGoMe||The angle between articulare (Ar), gonion (Go), and menton (Me). Used for Bjork-Jarabak polygon sum.|
|Facial biotype angle||The algebraic sum of the angles NSAr, SArGo, and ArGoMe.|
|Skeletal sagittal relationship parameters|
|ANB||The angle between A, N, and B.|
|APDI||The anteroposterior dysplasia indicator was obtained from the algebraic sum of the angles N-Pg-FH (facial plane) plus/minus the angle AB-facial plane (positive when point B is ahead of point A and negative when point A is ahead of point B) and plus/minus the angle FH-PP (palatal plane) (negative when PP is tilted upward and positive when tilted down).|
|Maxillary sagittal position parameters|
|SNA||The angle between sella (S), nasion (N), and subnasal (A).|
|Ratio (PP/ACB)||The ratio of the palatal plane (PP) to the anterior cranial base (ACB).|
|Transversal parameters. (maxillary width measurements)|
|MBW||The maxillary first molar basal width dimension was measured over the line formed from the outer edges of the right and left sides of the maxillary base (lateral limits) along the nasal floor reference plane.|
|MAW||The maxillary first molar alveolar width dimension was measured between the most occlusal points of the maxillary alveolar process on the first molar coronal slice.|
|PMBW||The maxillary first premolar basal width dimension was measured over the line formed from the outer edges of the right and left sides of the maxillary base (lateral limits) along the nasal floor reference plane.|
|PMAW||The maxillary first premolar alveolar width was measured between the most occlusal points of the maxillary alveolar process on the first premolar coronal slice.|
For maxillary width measurements, DICOM files were imported into OnDemand 3D software (version 1.0; Cybermed, Seoul, South Korea) that was used to orient the CBCT scans and measure all data, based on the method of Podesser et al. Maxillary transverse dimensions were measured at 4 levels: first molar basal width (MBW), first molar alveolar width (MAW), first premolar basal width (PMBW), and first premolar alveolar width (PMAW) ( Fig 3 ). The measurements were made on slices showing the maxillary first premolars and first molars. Molar measurements were made on the most anterior coronal slice showing the buccal root furcation with the palatal plane horizontal in the CBCT scan. Landmarks were placed at the most inferior point on the right and left nasal floor to draw a nasal floor reference plane passing through these 2 landmarks. Premolar measurements were made on the coronal slice showing the center of the root canal, with the same landmark placement and reference lines considered for molar measurements. The definitions of the cephalometric and CBCT measurements are shown in Table I .
Twenty records were reanalyzed by the same examiner (N.A-A.) after a 30-day interval. Intraexaminer reliability was assessed by the intraclass correlation coefficient. Additionally, random error of reproducibility was calculated with Dahlberg’s formula.
Statistical analyses were performed using SPSS software for Windows (version 22; IBM, Armonk, NY). Data distribution normality was evaluated by Shapiro Wilk tests. Group comparability regarding sex and age were evaluated with chi-square and analysis of variance (ANOVA) tests, respectively. Group comparabilities regarding the vertical growth pattern (mesofacial, brachyfacial, dolichofacial), skeletal sagittal relationship (Class I, Class II, Class III), and maxillary sagittal position (normal, retrusive, protrusive) were evaluated with chi-square tests as well. Group comparisons regarding vertical and sagittal skeletal characteristics and maxillary transverse dimensions were performed with ANOVA and post hoc Scheffé tests. Multiple linear regression analyses were included to estimate the influence of group characteristics in the maxillary transverse dimension measurements using the overfit method, where an initial multiple regression analysis with all the variables followed by a second regression analysis with only variables showing P values smaller than 0.25 were performed for each transverse dimension. Statistical significance was set at P <0.05 for all the tests.
The intraclass correlation coefficient value was higher than 0.90 (95% confidence interval, 0.65-0.99). Dahlberg errors were smaller than 0.65° and 0.8 mm for angular and linear measurements, respectively.
The groups were comparable regarding sex. The bilateral group was the youngest group compared with the unilateral and control groups ( P <0.001; Table II ). The groups had similar distributions regarding vertical growth patterns, skeletal sagittal relationships, and maxillary sagittal positions ( P >0.05; Table III ). Of the 45 subjects in the unilateral group, 20 had labial impaction and 25 had palatal impaction. Of the 41 subjects in the bilateral group, 21 had bilateral labial impaction, 7 had bilateral palatal impaction, and 13 had combined labial and palatal impaction.
|Group||Sex ∗||Age †|
|Control||25||42||67||26.5 (6,05) A|
|Unilateral||17||28||45||23.1 (10,6) A|
|Bilateral||14||27||41||17.1 (10,7) B|
|Control (n = 67)||Unilateral (n = 45)||Bilateral (n = 41)||Total||P|
|Vertical growth pattern|
|Mesofacial||42 (63%)||30 (67%)||28 (68%)||100||0.539|
|Brachyfacial||10 (15%)||10 (22%)||6 (15%)||26|
|Dolichofacial||15 (22%)||5 (11%)||7 (17%)||27|
|Skeletal sagittal relationship||0.779|
|Class I||31 (46%)||22 (49%)||18 (44%)||71|
|Class II||23 (34%)||13 (29%)||17 (41%)||53|
|Class III||13 (20%)||10 (22%)||6 (15%)||29|
|Maxillary sagittal position||0.218|
|Normal||26 (39%)||19 (42%)||14 (34%)||59|
|Retrusion||7 (10%)||9 (20%)||11 (27%)||27|
|Protrusion||34 (51%)||17 (38%)||16 (39%)||67|
There were significant differences in the NSAr angle; it was greater for the unilateral group (126.5°) compared with the control (122.6°) and bilateral (124.1°) groups ( P = 0.003). There were no differences between the control and bilateral groups. Significant differences were also found for the facial biotype angle; it was greater in the bilateral group (396.1°) with differences of 3.9° compared with the control group (392.2°) and 1.5° compared with the unilateral group (394.6°) ( P = 0.008). There were no differences between the control and unilateral groups regarding this variable. No differences were found for other vertical and sagittal measurements ( P >0.05; Table IV ).
|Control (n = 67)||Unilateral (n = 45)||Bilateral (n = 41)||P|
|Vertical growth pattern parameters|
|NSAr||122.6 A||5.7||126.5 B||5.9||124.1 A||5.7||0.003|
|Facial biotype angle||392.2 A||6.7||394.6 A||6.1||396.1 B||6.1||0.008|
|Sagittal skeletal relationship parameters|
|Maxillary sagittal position parameters|
Regarding the transverse maxillary dimensions, there were statistically significant differences for MBW, MAW, and PMAW measurements; these were greater in the control group compared with the unilateral and bilateral groups. There were no significant differences between the 2 groups with impacted canines for any transverse measurements. For the MBW measurements, there were differences of 2.5 mm between the control group (66.7 mm) and the unilateral group (64.2 mm), and 2.6 mm between the control and bilateral groups (64.1 mm), P = 0.030. In the MAW measurement, there were differences of 3 mm between the control group (57.2 mm) and the unilateral group (54.2 mm), and 3.9 mm between the control and bilateral groups (53.3 mm), P <0.001. For the PMAW measurement, there were also significant differences between the control group and unilateral and bilateral groups (3.8 and 3.5 mm, respectively; P <0.001). There were no significant differences in the PMBW measurements among the 3 groups ( P = 0.063; Table V ). The transverse measurement comparisons considering the position of the impacted canines were similar between groups with unilateral labial impaction, unilateral palatal impaction, bilateral labial impaction, bilateral palatal impaction, and combined labial and palatal impaction (ANOVA, P >0.05; Table VI ).
|Control (n = 67)||Unilateral (n = 45)||Bilateral (n = 41)|
|MBW||66.7 A||5.5||64.2 B||6.1||64.1 B||5.5||0.030|
|MAW||57.2 A||2.7||54.2 B||4.2||53.3 B||3.7||<0.001|
|PMAW||45.2 A||2.3||41.4 B||4.2||41.7 B||3.4||<0.001|
|Unilateral labial (n = 20)||Unilateral palatal (n = 25)||Bilateral labial (n = 21)||Bilateral palatal (n = 7)||Bilateral combined labial and palatal (n = 13)||P|
The multiple linear regression analyses showed that MBW measurement was influenced by sex ( P <0.001), ANB ( P = 0.022), facial biotype angle ( P = 0.025), and group variables ( P = 0.028). MAW was influenced by sex ( P <0.001), SNA ( P = 0.004), and group variables ( P <0.001). PMBW was influenced by sex ( P = 0.010), age ( P = 0.005), ANB ( P <0.001), SNA ( P = 0.009), and facial biotype angle ( P <0.001). The PMAW measurement was influenced by sex ( P <0.001), facial biotype angle ( P <0.001), and group variables ( P <0.001) ( Table VII ).