- Email: [email protected]
Associations between palatally displaced canines and maxillary lateral incisors
ORIGINAL ARTICLE
Associations between palatally displaced canines and maxillary lateral incisors Ing Wei Liuk,a Richard John Olive,b Mark Griffin,c and Paul Monsourd Brisbane and Herston, Queensland, Australia
Introduction: The purpose of this research was to investigate relationships among the location and orientation of palatally displaced canines and the dimension and orientation of the maxillary lateral incisor. Methods: An experimental group of 40 patients with 46 palatally displaced canines (20 from boys, 26 from girls; mean age of the subjects, 13.9 years; age range, 10.5-15.9 years) was selected from the records of patients referred to a radiology practice specializing in cone-beam volumetric tomography imaging. This group was age- and sexmatched with 30 normal subjects with 60 canines (26 from boys, 34 from girls; mean age of the subjects, 13.8 years; age range, 10.4-15.7 years). Digital cone-beam volumetric tomography images were imported into an imaging software and were reoriented and reconstructed into several radiographic images in coronal and sagittal sections; a maxillary arch occlusal view was also produced. The angular and linear variables of the canines and the maxillary lateral incisors were measured by using software measurement tools. Independent t tests or Mann-Whitney U tests were used accordingly based on normality of the data to compare the variables between the palatally displaced canine and the control groups. Multiple linear regressions were used to examine the relationships between the canine variables (dependent variables) and the maxillary lateral incisor variables together with confounding variables (independent variables). Results: The maxillary lateral incisors in the palatally displaced canine group tended to be more upright in the sagittal and coronal planes. Generally, the most significant independent variables according to backward examination of linear regression for canine variables (coronal and sagittal angulations, and distance to the occlusal plane of palatally displaced canines) were the coronal and sagittal angulations of the maxillary lateral incisors, the length and buccolingual root width of the maxillary lateral incisors, and age. Conclusions: The orientation and location of palatally displaced canines were associated with changes in the angulations of maxillary lateral incisors and small lateral incisors. (Am J Orthod Dentofacial Orthop 2013;143:622-32)
P
alatally displaced canines have been found to be an autosomal dominant trait with low penetrance and variable expressivity; environmental factors might also play an important role in the etiology of impacted canines.1 The guidance theory has been advocated to explain the cause of palatally displaced a Postgraduate student, Discipline of Orthodontics, School of Dentistry, University of Queensland, Brisbane, Queensland, Australia. b Adjunct associate professor, Discipline of Orthodontics, School of Dentistry, University of Queensland, Brisbane, Queensland, Australia. c Research fellow, School of Population Health, University of Queensland, Herston, Queensland, Australia. d Foundation professor, Discipline of DentoMaxilloFacial Radiology, School of Dentistry, University of Queensland, Brisbane, Queensland, Australia. The authors report no commercial, proprietary, or financial interest in the products or companies described in this article. Supported by a grant from the Australian Society of Orthodontists' Foundation for Research and Education. Reprint requests to: Ing Wei Liuk, Discipline of Orthodontics, School of Dentistry, University of Queensland, 200 Turbot St, Brisbane, Queensland 4000, Australia; e-mail, [email protected]. Submitted, February 2012; revised and accepted, November 2012. 0889-5406/$36.00 Copyright Ó 2013 by the American Association of Orthodontists. http://dx.doi.org/10.1016/j.ajodo.2012.11.025
622
canines, proposing that the palatal displacement of the canines was caused by genetically determined anomalies that caused an environmentally generated alteration in the eruption pattern of the canines.2,3 The distally tipped lateral incisor is a normal feature during the “ugly duckling stage.”4 Authors of a longitudinal study concluded that distal tipping together with proclination of the lateral incisor might indicate a buccal ectopic canine eruption, but distal tipping alone should not be an indication of canine eruption disturbances.5 Another study described how a palatally displaced canine might press on the root of the lateral incisor and push the root labially while the crown swings palatally.6 It was even suggested that palatally displaced canines can cause distal crown tipping of the lateral incisors as well as rotation.7 No study has previously investigated the relationships of the location and orientation of palatally displaced canines with the dimension and orientation of the maxillary lateral incisors. The aim of this study was to investigate the relationships between the variables of palatally displaced canines and maxillary lateral incisors
Liuk et al
623
Fig 1. Volumetric image was orientated with the occlusal plane horizontal.
together with confounding variables. The null hypothesis was that no relationship exists between these variables. MATERIAL AND METHODS
An experimental group (palatally displaced canines group) of 40 patients with 46 palatally displaced canines (20 from boys, 26 from girls), in the age range 10 to 15.9 years (mean ages were 13.9 years for boys, 13.8 years for girls, and 13.9 years overall), was selected from the records of patients referred to a radiology practice specializing in cone-beam volumetric tomography imaging. The palatally displaced canine group was age- and sex-matched with a control group comprising 30 normal subjects with 60 canines (26 from boys, 34 from girls; mean ages were 13.8 years for boys, 13.8 years for girls, and 13.8 years overall). The cone-beam volumetric tomography digital imaging and communications in medicine (DICOM) files were imported into Dolphin Imaging software (version 11.0; Dolphin Imaging & Management Solutions, Chatsworth, Calif). The volumetric image was reoriented as if the head was looking straight out of the computer screen (with the facial sagittal plane perpendicular to the computer screen), with the occlusal plane horizontal, touching the incisal edge of the maxillary central incisors and the mesiobuccal cusp tips of
the maxillary first molars (Fig 1). Radiographic images of the canine and the lateral incisor in the coronal and sagittal sections were reconstructed from the volumetric image, and a maxillary arch occlusal view was also produced. The mesiodistal position of the canine cusp tip,8 the vertical position of the canine cusp tip,9 the severity of lateral incisor resorption,10 the lateral incisor length (Lat_Lgth), the buccolingual width at the cementoenamel junction (CEJ) level (BL_CEJ), the buccolingual width at 4 mm apical to the CEJ level (BL_CEJ4), the buccolingual width at 8 mm apical to the CEJ level (BL_CEL8), the mesiodistal width at the CEJ level (MD_CEJ), the mesiodistal width at 4 mm apical to the CEJ level (MD_CEJ4), and the mesiodistal width at 8 mm apical to the CEJ level (MD_CEJ8) of the maxillary lateral incisor were acquired as described previously.11 The average of MD_CEJ, MD_CEJ4, and MD_CEJ8 was calculated as MD_Ave, and the average of BL_CEJ, BL_CEJ4, and BL_CEJ8 was calculated as BL_Ave, to allow the multiple linear regression analysis as discussed below. The angular and linear measurements were made with the Dolphin measurement tools with precision values of 0.1 and 0.1 mm, respectively. The images were magnified by 400% to allow a better view, yet not to produce blurred outlines or pixelated objects.
American Journal of Orthodontics and Dentofacial Orthopedics
May 2013 Vol 143 Issue 5
Liuk et al
624
Fig 2. Angulation of the canine and the lateral incisor (relative to the vertical line) in the sagittal view, and distance of the canine cusp tip to the occlusal plane.
Fig 3. Angulation of the canine and the lateral incisor (relative to the midline) in the coronal view.
Measurements of the canine variables were carried out as follows. 1.
The sagittal angulation of the canine was measured relative to a vertical line that was perpendicular to the occlusal plane on the reconstructed sagittal radiographic image (Fig 2). The coronal angulation of the canine was measured relative to a facial
May 2013 Vol 143 Issue 5
midline that was also perpendicular to the occlusal plane on the reconstructed coronal radiographic image (Fig 3). The angle values were recorded as positive if the crown was tipping mesially in the sagittal view or buccally in the coronal view, and as negative if the crown was tipping distally or palatally.
American Journal of Orthodontics and Dentofacial Orthopedics
Liuk et al
625
Fig 4. Rotation of the lateral incisor (relative to the midline) in the occlusal view.
2.
The distance of the canine cusp tip was measured perpendicularly to the occlusal plane (Fig 2). The distance was recorded as 0 if the canine had fully erupted and no negative value was recorded.
Measurements of the lateral incisor variables were carried out as follows. 1.
2.
3.
The sagittal and coronal angulations of the lateral incisor were measured similarly to those for the canines (Figs 2 and 3). The rotation of the lateral incisor was measured relative to a palatal midline that was parallel to the midsagittal plane of the face on the maxillary arch occlusal view image (Fig 4). The rotational angle was recorded with the value starting from 0 (with a tangent to the buccal crown surface parallel to the palatal midline), and the value was increasing (positive) as the crown was rotating distobuccally or decreasing (negative) as the crown was rotating distopalatally. The tooth lengths and widths of the lateral incisors were acquired as described previously.11
Statistical analysis
All measurements were made by 1 investigator (I.W.L.). A pilot study comprising 16 subjects with 20 palatally displaced canines, matched to 10 normal subjects with 20 control canines was conducted. Since the sample size of the pilot study was too small to do an a priori power analysis for running a multiple linear regression, a post hoc power analysis was used to examine
the power of the data. Power analysis was carried out by using G*Power 3 software.12 Statistical analysis was performed with the SPSS Statistics Standard GradPack (version 19.0; IBM, Armonk, NY). The measurement data were analyzed for normality by using a normal probability plot and the Shapiro-Wilks test. Independent t tests and MannWhitney U tests were used accordingly to compare the palatally displaced canine and control groups. A significance level of 0.005 was used to account for multiple comparisons. A linear regression with backward examination was conducted to look for the most significant independent variables for each canine variable (Can_Sag, Can_Cor, Can_Occl). Fifteen independent variables (MDsect [mesiodistal position of the canine cusp tip], Vzone [vertical position of the canine cusp tip], Lat_Sag, Lat_Cor, Lat_Rot, Lat_Lgth, MD_CEJ, BL_CEJ, MD_CEJ4, BL_CEJ4, MD_CEJ8, BL_CEJ8, age, sex, and resorption) were included in the initial analysis. Because of the great colinearity between the variables and the large number of variables considered, a principal component analysis was used to analyze the relationships between the variables. After variable reduction procedures, multiple linear regressions were used to analyze the relationship between the canine variables and the lateral incisor variables, taking account the confounding variables. Linear regressions were used in this study rather than analyses with correlation matrices with bivariate correlations because of the multiple independent variables and also the correlation between the independent variables and the involvement of confounding variables. Since the variables MDsect and Vzone were not independent from
American Journal of Orthodontics and Dentofacial Orthopedics
May 2013 Vol 143 Issue 5
Liuk et al
626
Fig 5. Box plot of angular variables ( ) comparing the palatally displaced canine and the control groups. Can_Sag, Canine sagittal angulation; Can_Cor, canine coronal angulation; Lat_Sag, lateral incisor sagittal angulation; Lat_Cor, lateral incisor coronal angulation; Lat_Rot, lateral incisor rotational angulation.
the canine variables Can_Sag, Can_Cor, and Can_Occl, they were not included in the regression analysis. RESULTS
A post hoc power analysis was carried out to examine the linear regression on the sagittal angulation of the canine, and it was found that the power was 82.2%. We analyzed 46 palatally displaced canines and 60 control canines in this study. All data were normally distributed, except for those of the coronal angulation of the canine and the distance of the canine cusp tip to the occlusal plane from the control group. Thirteen canine and lateral incisor variables were compared between the palatally displaced canine and the control groups (Figs 5 and 6) with independent t tests (assuming unequal variance) and Mann-Whitney U tests accordingly (Table I). The canine variables between the 2 groups were significantly different (P \0.001). Also, the lateral incisor variables between the 2 groups were significantly different (P \0.005), except for the mesiodistal width at 4 mm apical to the CEJ level (P .0.05), whereas the mesiodistal width at 8 mm apical to the CEJ level and the average of the mesiodistal widths at the 3 levels were borderline significant (P \0.05).
May 2013 Vol 143 Issue 5
In this study, the maxillary lateral incisors in the palatally displaced canine group tended to be more upright in the sagittal (P \0.005) and coronal (P \0.001) planes, and more mesiolabially rotated (P \0.005) compared with the control group. The means in the palatally displaced canine group were 17.7 for the sagittal angulation of the lateral incisor, 9.8 for the coronal angulation of the lateral incisor, and 45.3 for the rotation of the lateral incisor; the measurements in the control group were 30.9 , 14.8 , and 57.1 , respectively. The mean of the sagittal angulation of the canine in the palatally displaced canine group was 36.0 , and the coronal angulation of the canine was 35.5 (palatally inclined); these measurements in the control group were 22.7 and 7.1 , respectively. Cohen13 suggested that the strength of correlation coefficients of 0.10 to 0.29 is small, 0.30 to 0.49 is moderate, and 0.50 to 1.0 is large or high. Patient age was found to be correlated moderately with the mesiodistal position of the canine cusp tip and the vertical position of the canine cusp tip (Table II). The sagittal angulation of the canine and the coronal angulation of the canine were highly correlated with each other, and also with the mesiodistal position of the canine cusp tip. The
American Journal of Orthodontics and Dentofacial Orthopedics
Liuk et al
627
Fig 6. Box plot of linear variables (mm) comparing the palatally displaced canine and the control groups. Can_Occl, Canine cusp tip to the occlusal plane distance; Lat_Lgth, lateral incisor length; MD_CEJ, mesiodistal width at the CEJ level; BL_CEJ, buccolingual width at the CEJ level; MD_CEJ4, mesiodistal width at 4 mm apical to the CEJ level; BL_CEJ4, buccolingual width at 4 mm apical to the CEJ level; MD_CEJ8, mesiodistal width at 8 mm apical to the CEJ level; BL_CEJ8, buccolingual width at 8 mm apical to the CEJ level.
distance of the canine cusp tip to the occlusal plane was moderately correlated with the coronal angulation of the canine and the vertical position of the canine cusp tip. The mesiodistal position of the canine cusp tip did not correlate with either the vertical position of the canine cusp tip or the distance of the canine cusp tip to the occlusal plane. Pearson's correlations between these variables are presented in Table III. According to the principal component analysis with varimax rotation, some variables actually correlate to each other and are clustered together as shown in the pattern matrix table (Table IV). The varimax rotation method was used, since the initial direct quartimin rotation method showed that the correlation coefficients between the components were lower than 0.32 (Tabachnick and Fidell,14 Pallant15). The variable can be said to load on a given component when the loading factor is 0.32 or greater for that component but not on other components.14 Table IV shows some groups of variables that are closely related to one another. The first component contains variables related to the buccolingual widths of the lateral incisors. The second component contains
variables related to the mesiodistal widths of the lateral incisors. The third component contains variables related to the coronal and sagittal angulations of the canines. It was found that the buccolingual width at the CEJ level, the buccolingual width at 4 mm apical to the CEJ level, and the buccolingual width at 8 mm apical to the CEJ level were correlated on the same principal component and thus can be reduced to a single variable; therefore, the 3 measurements can be averaged to give 1 measurement that is the average of the buccolingual widths at the 3 levels. Similarly, the mesiodistal width at the CEJ level, the mesiodistal width at 4 mm apical to the CEJ level, and the mesiodistal width at 8 mm apical to the CEJ level could be reduced to 1 variable that is the average of the mesiodistal widths at the 3 levels. The mesiodistal position of the canine cusp tip was also found to correlate with the sagittal angulation of the canine and the coronal angulation of the canine on the same principal component and hence will not be included in subsequent regression analysis. Likewise, the vertical position of the canine cusp tip was correlated with the distance of the canine cusp tip to the occlusal plane on the same principal component and was
American Journal of Orthodontics and Dentofacial Orthopedics
May 2013 Vol 143 Issue 5
Liuk et al
628
Table I. Results of independent t tests and Mann-Whitney U tests comparing the angular and linear variables between the palatally displaced canine (PDC) and control groups PDC group (n 5 46) Variable Can_Sag ( ) Can_Cor ( ) Can_Occl (mm) Lat_Sag ( ) Lat_Cor ( ) Lat_Rot ( ) Lat_Lgth (mm) MD_CEJ (mm) MD_CEJ4 (mm) MD_CEJ8 (mm) MD_Ave (mm) BL_CEJ (mm) BL_CEJ4 (mm) BL_CEJ8 (mm) BL_Ave (mm)
Mean 36.0 35.5 9.8 17.7 9.8 45.3 21.3 4.3 3.4 2.5 3.4 5.9 4.8 3.6 4.8
SD 8.7 13.1 1.5 6.6 8.7 19.8 1.8 0.4 0.5 0.4 0.4 0.8 0.9 0.9 0.8
Control group (n 5 60) Mean 22.7 7.1 1.8 30.9 14.8 57.1 23.3 4.6 3.5 2.7 3.6 6.6 5.4 4.3 5.4
SD 4.5 8.1 3.6 5.7 4.6 12.3 2.7 0.6 0.5 0.5 0.5 0.6 0.8 0.8 0.7
t test P value \0.001y \0.001y \0.001y \0.001y 0.001y 0.001y \0.001y 0.004y 0.228 0.013* 0.016* \0.001y \0.001y \0.001y \0.001y
Mean difference 13.3 42.6 8.0 13.2 5.0 11.7 2.1 0.3 0.1 0.2 0.2 0.7 0.6 0.7 0.7
Mann-Whitney U test P value \0.001y \0.001y \0.001y \0.001y 0.003y \0.001y \0.001y 0.003y 0.341 0.024* 0.041* \0.001y 0.003y 0.001y 0.001y
Can_Sag, Canine sagittal angulation; Can_Cor, canine coronal angulation; Can_Occl, canine cusp tip to occlusal plane distance; Lat_Sag, lateral incisor sagittal angulation; Lat_Cor, lateral incisor coronal angulation; Lat_Rot, lateral incisor rotational angulation; Lat_Lgth, lateral incisor length; MD_CEJ, mesiodistal width at the CEJ level; MD_CEJ4, mesiodistal width at 4 mm apical to the CEJ level; MD_CEJ8, mesiodistal width at 8 mm apical to the CEJ level; MD_Ave, average of MD_CEJ, MD_CEJ4, and MD_CEJ8; BL_CEJ, buccolingual width at the CEJ level; BL_CEJ4, buccolingual width at 4 mm apical to the CEJ level; BL_CEJ8, buccolingual width at 8 mm apical to the CEJ level; BL_Ave, average of BL_CEJ, BL_CEJ4, and BL_CEJ8. *Significant at the 0.05 level; ysignificant at the 0.005 level.
therefore excluded from subsequent regression analysis. As a result, 15 initial independent variables were reduced to 9 independent variables that were then examined in the regression analysis. It was also found from the rotated component matrix table that the lateral incisor length actually loaded on both principal components 1 and 2, denoting that lateral incisor length was correlated with buccolingual width at the CEJ level, buccolingual width at 4 mm apical to the CEJ level, and buccolingual width at 8 mm apical to the CEJ level, and also with mesiodistal width at the CEJ level, mesiodistal width at 4 mm apical to the CEJ level, and mesiodistal width at 8 mm apical to the CEJ level. Nine independent variables—sagittal angulations of the lateral incisor, coronal angulations of the lateral incisor, rotation of the lateral incisor, the lateral incisor length, average of mesiodistal widths at the 3 levels, average of buccolingual widths at the 3 levels, age, sex, and resorption—were included in the initial model of linear regression (Table V). It was found that 30.8% (R2 5 0.308) of the variation in the sagittal angulation of the canine and 30.6% (R2 5 0.306) of the variation in the coronal angulation of the canine can be explained by the 9 independent variables. The most significant independent variables according to the backward examination of linear regression for the sagittal angulation of the canine were the coronal angulations of the lateral
May 2013 Vol 143 Issue 5
Table II. Correlation of MDsect and Vzone with age
and sex Age Variable MDsect Vzone
Pearson correlation 0.400 0.325
Sex P value 0.006* 0.028*
Pearson correlation 0.007 0.165
P value 0.961 0.272
MDsect, Mesiodistal position of canine (sectors I to IV); Vzone, vertical position of canine (zones 1 to 5). *Significant at the 0.05 level.
incisor; for the coronal angulation of the canine, they were lateral incisor length and age. Also, 32.9% (R2 5 0.329) of the variation in the distance of the canine cusp tip to the occlusal plane can be explained by the 9 independent variables, and the most significant variables were the sagittal angulations of the lateral incisor, the coronal angulations of the lateral incisor, and the average of the buccolingual widths at the 3 levels. DISCUSSION
The differences between the palatally displaced canine and control groups for the sagittal and coronal angulations and displacement from the occlusal plane of the canine describe the canine ectopia and are consistent with other studies.16-19
American Journal of Orthodontics and Dentofacial Orthopedics
Liuk et al
629
Table III. Correlation between canine variables, MDsect, and Vzone in the palatally displaced canine group Can_Sag Variable Can_Sag Can_Cor Can_Occl MDsect Vzone
Can_Cor
Pearson correlation
P value
0.518 0.251 0.548 0.063
\0.001y 0.093 \0.001y 0.678
Pearson correlation 0.518
P value \0.001y
0.394 0.612 0.271
0.007* \0.001y 0.069
Can_Occl Pearson correlation 0.251 0.394 0.244 0.438
P value 0.093 0.007* 0.102 0.002y
MDsect Pearson correlation 0.548 0.612 0.244 0.207
Vzone
P value \0.001y \0.001y 0.102
Pearson correlation 0.063 0.271 0.438 0.207
P value 0.678 0.069 0.002y 0.169
0.169
Can_Sag, Canine sagittal angulation; Can_Cor, canine coronal angulation; Can_Occl, canine cusp tip to occlusal plane distance; MDsect, mesiodistal position of the canine (sectors I to IV); Vzone, vertical position of the canine (zones 1 to 5). *Significant at the 0.05 level; ysignificant at the 0.005 level.
Table IV. Rotated component matrix showing the 6 components extracted by the principal component analysis Component/Total variance explained Variable BL_CEJ4 BL_CEJ Lat_Cor BL_CEJ8 Lat_Rot Lat_Lgth MD_CEJ4 MD_CEJ8 MD_CEJ Can_Cor MDsect Can_Sag Age Vzone Can_Occl Lat_Sag Sex Resorption
1/30% 0.904* 0.884* 0.794* 0.762* 0.543* 0.535* 0.194 0.153 0.105 0.058 0.317 0.297 0.010 0.005 0.048 0.277 0.079 0.020
2/14% 0.227 0.192 0.152 0.337* 0.117 0.511* 0.914* 0.835* 0.800* 0.047 0.103 0.011 0.263 0.200 0.150 0.107 0.597* 0.014
3/12% 0.078 0.020 0.131 0.292 0.151 0.178 0.101 0.224 0.085 0.839* 0.762* 0.746* 0.608* 0.127 0.283 0.003 0.070 0.011
4/8% 0.001 0.078 0.227 0.040 0.132 0.200 0.041 0.166 0.243 0.208 0.169 0.045 0.102 0.793* 0.739* 0.571* 0.019 0.106
5/7% 0.024 0.125 0.096 0.028 0.535* 0.206 0.026 0.030 0.211 0.049 0.000 0.315 0.472* 0.090 0.306 0.179 0.623* 0.051
6/6% 0.063 0.157 0.096 0.045 0.056 0.010 0.089 0.027 0.104 0.024 0.253 0.185 0.087 0.235 0.020 0.459* 0.029 0.895*
Can_Sag, Canine sagittal angulation; Can_Cor, canine coronal angulation; Can_Occl, canine cusp tip to occlusal plane distance; Lat_Sag, lateral incisor sagittal angulation; Lat_Cor, lateral incisor coronal angulation; Lat_Rot, lateral incisor rotational angulation; Lat_Lgth, lateral incisor length; MD_CEJ, mesiodistal width at the CEJ level; MD_CEJ4, mesiodistal width at 4 mm apical to the CEJ level; MD_CEJ8, mesiodistal width at 8 mm apical to the CEJ level; MD_Ave, average of MD_CEJ, MD_CEJ4, and MD_CEJ8; BL_CEJ, buccolingual width at the CEJ level; BL_CEJ4, buccolingual width at 4 mm apical to the CEJ level; BL_CEJ8, buccolingual width at 8 mm apical to the CEJ level; BL_Ave, average of BL_CEJ, BL_CEJ4, and BL_CEJ8; MDsect, mesiodistal position of the canine (sectors I to IV); Vzone, vertical position of the canine (zones 1 to 5). *Loading factor with coefficient $0.32.
In this study, on average, the maxillary lateral incisors of the palatally displaced canine group were more retroclined by 13.2 in the sagittal plane and more upright by 5 in the coronal plane compared with those of the control group. A cover bite, defined as a deep overbite with or caused by retroclination of the maxillary incisors, has been found to be prevalent in a palatally displaced canine group.20 Another study on Class II Division 2 patients reported that 7.5% of the patients had unilateral or bilateral peg-shaped lateral incisors, whereas 33.5% had unilateral or bilateral palatally displaced canines.21
The maxillary lateral incisors of the palatally displaced canine group in this study were more mesiolabially rotated by 11.7 compared with those of the control group. A previous study reported that the maxillary lateral incisors were more mesiolabially rotated when they were adjacent to palatally displaced canines with a more severe impaction.22 Some authors claimed that palatally displaced canines were not associated with rotation of the lateral incisors, but they only recorded incisor rotation that was more than 45 in their study.5
American Journal of Orthodontics and Dentofacial Orthopedics
May 2013 Vol 143 Issue 5
Liuk et al
630
Table V. Results of multiple linear regressions for the dependent canine variables in the palatally displaced canine
group Initial model from standard linear regression Unstandardized coefficients Dependent Independent variable variable Can_Sag (Constant) Lat_Sag Lat_Cor Lat_Rot Lat_Lgth MD_Ave BL_Ave Age Sex Resorption Can_Cor (Constant) Lat_Sag Lat_Cor Lat_Rot Lat_Lgth MD_Ave BL_Ave Age Sex Resorption Can_Occl (Constant) Lat_Sag Lat_Cor Lat_Rot Lat_Lgth MD_Ave BL_Ave Age Sex Resorption
B 45.97 0.25 0.32 0.08 1.71 2.19 0.09 1.59 0.01 0.72 12.91 0.28 0.02 0.02 3.97 5.90 3.83 4.55 1.99 2.13 9.04 0.06 0.09 0.01 0.19 0.15 1.13 0.09 0.28 0.12
SE 23.92 0.21 0.19 0.07 1.06 4.29 2.61 0.88 3.06 2.68 36.33 0.32 0.28 0.11 1.60 6.51 3.96 1.33 4.65 4.08 4.04 0.04 0.03 0.01 0.18 0.72 0.44 0.15 0.52 0.45
Final model from stepwise linear regression Unstandardized coefficients
Beta 0.19 0.32 0.19 0.35 0.10 0.01 0.28 0.00 0.04 0.14 0.01 0.02 0.54 0.18 0.24 0.53 0.08 0.08 0.24 0.55 0.13 0.23 0.04 0.62 0.10 0.10 0.04
P value R2 0.063 0.308 0.249 0.096 0.271 0.114 0.613 0.971 0.079 0.998 0.790 0.724 0.306 0.394 0.946 0.894 0.018* 0.371 0.340 0.002y 0.671 0.604 0.032 0.329 0.135 0.005y 0.426 0.295 0.835 0.015* 0.530 0.586 0.792
B 39.72
SE 1.81
Beta
0.38
0.14
0.38
15.72
16.30
3.69
1.17
0.43
7.35 0.07 0.09
1.35 0.03 0.03
0.30 0.52
0.94
0.32
0.52
P value \0.001
R2 0.144
0.009*
0.340
0.185
0.003y
\0.001 0.262 0.038* 0.004y
0.005y
B, Unstandardized coefficients; Beta, standardized coefficients; SE, standard error; R2, coefficient of determination or squared correlation coefficient; Can_Sag, canine sagittal angulation; Can_Cor, canine coronal angulation; Can_Occl, canine cusp tip to occlusal plane distance; Lat_Sag, lateral incisor sagittal angulation; Lat_Cor, lateral incisor coronal angulation; Lat_Rot, lateral incisor rotational angulation; Lat_Lgth, lateral incisor length; MD_CEJ, mesiodistal width at the CEJ level; MD_CEJ4, mesiodistal width at 4 mm apical to the CEJ level; MD_CEJ8, mesiodistal width at 8 mm apical to the CEJ level; MD_Ave, average of MD_CEJ, MD_CEJ4, and MD_CEJ8; BL_CEJ, buccolingual width at the CEJ level; BL_CEJ4, buccolingual width at 4 mm apical to the CEJ level; BL_CEJ8, buccolingual width at 8 mm apical to the CEJ level; BL_Ave, average of BL_CEJ, BL_CEJ4, and BL_CEJ8. *Significant at the 0.05 level; ysignificant at the 0.005 level.
In this study, as the lateral incisor angulation in the coronal view decreased, the canine angulation in the sagittal view increased (negatively correlated). Thus, clinically, the lateral incisors adjacent to palatally displaced canines will appear to be less proclined or more upright. Impacted canines have been associated with both proclined and palatally tipped lateral incisors, depending on their buccal or palatal impaction, respectively.5,6 Ericson and Kurol5 suggested that lateral incisors with distal crown tipping and proclination might hint at buccally displaced canines. However, they did not draw any conclusions about the clinical appearance of
May 2013 Vol 143 Issue 5
lateral incisors in relation to palatally displaced canines. Jacobs suggested that a buccally directed force from the palatally displaced canine might cause the incisor root to tip labially and its crown palatally.6 This scenario cannot be discounted by our findings. In this study, as the length of the maxillary lateral incisor decreased, the canine tended to incline more palatally. The lateral incisor root length could be a critical factor in the progression of canine impaction, and it has been previously suggested that a short root compromised the canine guidance provided by the lateral incisor.23
American Journal of Orthodontics and Dentofacial Orthopedics
Liuk et al
In a previous study, there was a tendency for younger subjects to have less severe sectors of impaction.22 In our study, canine impaction was generally more severe with increasing age, as demonstrated by increased crown angulation toward the palatal side. This is contrary to another study reporting that younger patients had more severe impactions.24 In this study, when the lateral incisor crown displayed reduced inclination in the sagittal or coronal plane (the lateral incisor crown tipped more palatally or mesially), the canine would be displaced at a higher level from the occlusal plane. A recent study of normal children in the mixed dentition reported a positive correlation between the distance of the canine cusp tip from the occlusal plane and the angulation of the lateral incisor in the coronal view, but not the angulation of the lateral incisor in the sagittal view.25 This seems to disagree with our study, but comparison between these 2 studies is difficult, since the sample groups and study designs were different. In this study, the average buccolingual root widths of the maxillary lateral incisors in the palatally displaced canine group were significantly smaller than those of the control group. Nevertheless, in the palatally displaced canine group, the average maxillary lateral incisor buccolingual root width was found to be associated positively with the canine cusp tip distance from the occlusal plane. This means that palatally displaced canines with higher vertical displacements in this study were associated with greater buccolingual maxillary lateral incisor root widths; this seems to be contrary to our finding that the average buccolingual root width of the maxillary lateral incisors in the palatally displaced canine group was significantly smaller than those of the control group. No previous study has addressed this issue. The lateral incisor might play an obstructive role, and it has been advocated that anomalous and late-developing lateral incisors could redeflect the canine, causing second-stage impaction as 1 of the 5 elements in the guidance theory.3 To assimilate both aforementioned findings, it seems that maxillary lateral incisors with significantly smaller buccolingual root widths might cause first-stage impaction of the canine in the guidance theory. However, among these diminutive lateral incisors, those with greater buccolingual root widths appear to exert a greater effect in causing second-stage impaction of the canine. This could explain the finding in this study that the palatally displaced canines with higher vertical displacements were associated with maxillary lateral incisors with greater buccolingual root widths. The intrabony migration of canines might be a result of teeth competing for space and searching for the
631
path of least resistance.26 It has been stated that teeth erupt in the path of least resistance, and this is why a displaced canine can correct itself after removal of the deciduous canine.26 This study supports the hypothesis that the root of the lateral incisor acts as a natural barrier to the canine's initial mesial and palatal movement.2 The most significant independent variables in the model of linear regression denoted the associations between the canine and lateral incisor variables. However, linear regression will not confirm the causal relationships among these variables. CONCLUSIONS
1.
2.
3.
Maxillary lateral incisors associated with palatally displaced canines on average were more retroclined in the sagittal plane by 13.2 , more upright in the coronal plane by 5 , and more rotated mesiolabially by 11.7 than were those adjacent to normal canines. Regression analyses showed that palatally displaced canines were tipped more mesially when the lateral incisors were more upright in the coronal plane, palatally displaced canines became more palatally tipped with increasing age and in relation to shorter lateral incisor root lengths, and the vertical displacements of palatally displaced canines from the occlusal plane were greater when the lateral incisor was more upright in the coronal and sagittal planes, and when the buccolingual width of the lateral incisor was increased. We concluded from this study that the orientation and the location of palatally displaced canines were associated with changes in the angulations of the maxillary lateral incisors and with small lateral incisors.
REFERENCES 1. Camilleri S, Lewis CM, McDonald F. Ectopic maxillary canines: segregation analysis and a twin study. J Dent Res 2008;87: 580-3. 2. Becker A. In defense of the guidance theory of palatal canine displacement. Angle Orthod 1995;65:95-8. 3. Becker A. The orthodontic treatment of impacted teeth. 2nd ed. New York: Thieme; 2007. 4. Broadbent BH. Ontogenetic development of occlusion. Angle Orthod 1941;11:223-41. 5. Ericson S, Kurol J. Longitudinal study and analysis of clinical supervision of maxillary canine eruption. Community Dent Oral Epidemiol 1986;14:172-6. 6. Jacobs SG. Localization of the unerupted maxillary canine: how to and when to. Am J Orthod Dentofacial Orthop 1999;115:314-22. 7. Shapira Y, Kuftinec MM. Early diagnosis and interception of potential maxillary canine impaction. J Am Dent Assoc 1998; 129:1450-4.
American Journal of Orthodontics and Dentofacial Orthopedics
May 2013 Vol 143 Issue 5
Liuk et al
632
8. Lindauer SJ, Rubenstein LK, Hang WM, Andersen WC, Isaacson RJ. Canine impaction identified early with panoramic radiographs. J Am Dent Assoc 1992;123:91-2:95-7. 9. Kau CH, Pan P, Gallerano RL, English JD. A novel 3D classification system for canine impactions—the KPG index. Int J Med Robot 2009;5:291-6. 10. Liu DG, Zhang WL, Zhang ZY, Wu YT, Ma XC. Localization of impacted maxillary canines and observation of adjacent incisor resorption with cone-beam computed tomography. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008;105:91-8. 11. Liuk IW, Olive RJ, Griffin M, Monsour P. Maxillary lateral incisor morphology and palatally displaced canines: a case-controlled cone-beam volumetric tomography study. Am J Orthod Dentofacial Orthop 2013;143:522-6. 12. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 2007;39:175-91. 13. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale, NJ: L. Erlbaum Associates; 1988. 14. Tabachnick BG, Fidell LS. Using multivariate statistics. 5th ed. Boston: Pearson/Allyn & Bacon; 2007. 15. Pallant J. SPSS survival manual: a step by step guide to data analysis using SPSS. 4th ed. Crows Nest, New South Wales: Allen & Unwin; 2011. 16. Warford JH Jr, Grandhi RK, Tira DE. Prediction of maxillary canine impaction using sectors and angular measurement. Am J Orthod Dentofacial Orthop 2003;124:651-5.
May 2013 Vol 143 Issue 5
17. Ericson S, Kurol J. Early treatment of palatally erupting maxillary canines by extraction of the primary canines. Eur J Orthod 1988; 10:283-95. 18. Diamanti J, Townsend GC. New standards for permanent tooth emergence in Australian children. Aust Dent J 2003;48:39-42. 19. Berkovitz BK. How teeth erupt. Dent Update 1990;17:206-10. 20. Leifert S, Jonas IE. Dental anomalies as a microsymptom of palatal canine displacement. J Orofac Orthop 2003;64:108-20. 21. Basdra EK, Kiokpasoglou M, Stellzig A. The Class II Division 2 craniofacial type is associated with numerous congenital tooth anomalies. Eur J Orthod 2000;22:529-35. 22. Olive RJ. Factors influencing the non-surgical eruption of palatally impacted canines. Aust Orthod J 2005;21:95-101. 23. Becker A, Zilberman Y, Tsur B. Root length of lateral incisors adjacent to palatally-displaced maxillary cuspids. Angle Orthod 1984; 54:218-25. 24. Stewart JA, Heo G, Glover KE, Williamson PC, Lam EW, Major PW. Factors that relate to treatment duration for patients with palatally impacted maxillary canines. Am J Orthod Dentofacial Orthop 2001;119:216-25. 25. Baratieri C, Canongia AC, Bolognese AM. Relationship between maxillary canine intra-alveolar position and maxillary incisor angulation: a cone beam computed tomography study. Braz Dent J 2011;22:146-50. 26. Ngan PW, Wolf T, Kassoy G. Early diagnosis and prevention of impaction of the maxillary canine. ASDC J Dent Child 1987;54: 335-8.
American Journal of Orthodontics and Dentofacial Orthopedics
Associations between palatally displaced canines and maxillary lateral incisors Ing Wei Liuk,a Richard John Olive,b Mark Griffin,c and Paul Monsourd Brisbane and Herston, Queensland, Australia
Introduction: The purpose of this research was to investigate relationships among the location and orientation of palatally displaced canines and the dimension and orientation of the maxillary lateral incisor. Methods: An experimental group of 40 patients with 46 palatally displaced canines (20 from boys, 26 from girls; mean age of the subjects, 13.9 years; age range, 10.5-15.9 years) was selected from the records of patients referred to a radiology practice specializing in cone-beam volumetric tomography imaging. This group was age- and sexmatched with 30 normal subjects with 60 canines (26 from boys, 34 from girls; mean age of the subjects, 13.8 years; age range, 10.4-15.7 years). Digital cone-beam volumetric tomography images were imported into an imaging software and were reoriented and reconstructed into several radiographic images in coronal and sagittal sections; a maxillary arch occlusal view was also produced. The angular and linear variables of the canines and the maxillary lateral incisors were measured by using software measurement tools. Independent t tests or Mann-Whitney U tests were used accordingly based on normality of the data to compare the variables between the palatally displaced canine and the control groups. Multiple linear regressions were used to examine the relationships between the canine variables (dependent variables) and the maxillary lateral incisor variables together with confounding variables (independent variables). Results: The maxillary lateral incisors in the palatally displaced canine group tended to be more upright in the sagittal and coronal planes. Generally, the most significant independent variables according to backward examination of linear regression for canine variables (coronal and sagittal angulations, and distance to the occlusal plane of palatally displaced canines) were the coronal and sagittal angulations of the maxillary lateral incisors, the length and buccolingual root width of the maxillary lateral incisors, and age. Conclusions: The orientation and location of palatally displaced canines were associated with changes in the angulations of maxillary lateral incisors and small lateral incisors. (Am J Orthod Dentofacial Orthop 2013;143:622-32)
P
alatally displaced canines have been found to be an autosomal dominant trait with low penetrance and variable expressivity; environmental factors might also play an important role in the etiology of impacted canines.1 The guidance theory has been advocated to explain the cause of palatally displaced a Postgraduate student, Discipline of Orthodontics, School of Dentistry, University of Queensland, Brisbane, Queensland, Australia. b Adjunct associate professor, Discipline of Orthodontics, School of Dentistry, University of Queensland, Brisbane, Queensland, Australia. c Research fellow, School of Population Health, University of Queensland, Herston, Queensland, Australia. d Foundation professor, Discipline of DentoMaxilloFacial Radiology, School of Dentistry, University of Queensland, Brisbane, Queensland, Australia. The authors report no commercial, proprietary, or financial interest in the products or companies described in this article. Supported by a grant from the Australian Society of Orthodontists' Foundation for Research and Education. Reprint requests to: Ing Wei Liuk, Discipline of Orthodontics, School of Dentistry, University of Queensland, 200 Turbot St, Brisbane, Queensland 4000, Australia; e-mail, [email protected]. Submitted, February 2012; revised and accepted, November 2012. 0889-5406/$36.00 Copyright Ó 2013 by the American Association of Orthodontists. http://dx.doi.org/10.1016/j.ajodo.2012.11.025
622
canines, proposing that the palatal displacement of the canines was caused by genetically determined anomalies that caused an environmentally generated alteration in the eruption pattern of the canines.2,3 The distally tipped lateral incisor is a normal feature during the “ugly duckling stage.”4 Authors of a longitudinal study concluded that distal tipping together with proclination of the lateral incisor might indicate a buccal ectopic canine eruption, but distal tipping alone should not be an indication of canine eruption disturbances.5 Another study described how a palatally displaced canine might press on the root of the lateral incisor and push the root labially while the crown swings palatally.6 It was even suggested that palatally displaced canines can cause distal crown tipping of the lateral incisors as well as rotation.7 No study has previously investigated the relationships of the location and orientation of palatally displaced canines with the dimension and orientation of the maxillary lateral incisors. The aim of this study was to investigate the relationships between the variables of palatally displaced canines and maxillary lateral incisors
Liuk et al
623
Fig 1. Volumetric image was orientated with the occlusal plane horizontal.
together with confounding variables. The null hypothesis was that no relationship exists between these variables. MATERIAL AND METHODS
An experimental group (palatally displaced canines group) of 40 patients with 46 palatally displaced canines (20 from boys, 26 from girls), in the age range 10 to 15.9 years (mean ages were 13.9 years for boys, 13.8 years for girls, and 13.9 years overall), was selected from the records of patients referred to a radiology practice specializing in cone-beam volumetric tomography imaging. The palatally displaced canine group was age- and sex-matched with a control group comprising 30 normal subjects with 60 canines (26 from boys, 34 from girls; mean ages were 13.8 years for boys, 13.8 years for girls, and 13.8 years overall). The cone-beam volumetric tomography digital imaging and communications in medicine (DICOM) files were imported into Dolphin Imaging software (version 11.0; Dolphin Imaging & Management Solutions, Chatsworth, Calif). The volumetric image was reoriented as if the head was looking straight out of the computer screen (with the facial sagittal plane perpendicular to the computer screen), with the occlusal plane horizontal, touching the incisal edge of the maxillary central incisors and the mesiobuccal cusp tips of
the maxillary first molars (Fig 1). Radiographic images of the canine and the lateral incisor in the coronal and sagittal sections were reconstructed from the volumetric image, and a maxillary arch occlusal view was also produced. The mesiodistal position of the canine cusp tip,8 the vertical position of the canine cusp tip,9 the severity of lateral incisor resorption,10 the lateral incisor length (Lat_Lgth), the buccolingual width at the cementoenamel junction (CEJ) level (BL_CEJ), the buccolingual width at 4 mm apical to the CEJ level (BL_CEJ4), the buccolingual width at 8 mm apical to the CEJ level (BL_CEL8), the mesiodistal width at the CEJ level (MD_CEJ), the mesiodistal width at 4 mm apical to the CEJ level (MD_CEJ4), and the mesiodistal width at 8 mm apical to the CEJ level (MD_CEJ8) of the maxillary lateral incisor were acquired as described previously.11 The average of MD_CEJ, MD_CEJ4, and MD_CEJ8 was calculated as MD_Ave, and the average of BL_CEJ, BL_CEJ4, and BL_CEJ8 was calculated as BL_Ave, to allow the multiple linear regression analysis as discussed below. The angular and linear measurements were made with the Dolphin measurement tools with precision values of 0.1 and 0.1 mm, respectively. The images were magnified by 400% to allow a better view, yet not to produce blurred outlines or pixelated objects.
American Journal of Orthodontics and Dentofacial Orthopedics
May 2013 Vol 143 Issue 5
Liuk et al
624
Fig 2. Angulation of the canine and the lateral incisor (relative to the vertical line) in the sagittal view, and distance of the canine cusp tip to the occlusal plane.
Fig 3. Angulation of the canine and the lateral incisor (relative to the midline) in the coronal view.
Measurements of the canine variables were carried out as follows. 1.
The sagittal angulation of the canine was measured relative to a vertical line that was perpendicular to the occlusal plane on the reconstructed sagittal radiographic image (Fig 2). The coronal angulation of the canine was measured relative to a facial
May 2013 Vol 143 Issue 5
midline that was also perpendicular to the occlusal plane on the reconstructed coronal radiographic image (Fig 3). The angle values were recorded as positive if the crown was tipping mesially in the sagittal view or buccally in the coronal view, and as negative if the crown was tipping distally or palatally.
American Journal of Orthodontics and Dentofacial Orthopedics
Liuk et al
625
Fig 4. Rotation of the lateral incisor (relative to the midline) in the occlusal view.
2.
The distance of the canine cusp tip was measured perpendicularly to the occlusal plane (Fig 2). The distance was recorded as 0 if the canine had fully erupted and no negative value was recorded.
Measurements of the lateral incisor variables were carried out as follows. 1.
2.
3.
The sagittal and coronal angulations of the lateral incisor were measured similarly to those for the canines (Figs 2 and 3). The rotation of the lateral incisor was measured relative to a palatal midline that was parallel to the midsagittal plane of the face on the maxillary arch occlusal view image (Fig 4). The rotational angle was recorded with the value starting from 0 (with a tangent to the buccal crown surface parallel to the palatal midline), and the value was increasing (positive) as the crown was rotating distobuccally or decreasing (negative) as the crown was rotating distopalatally. The tooth lengths and widths of the lateral incisors were acquired as described previously.11
Statistical analysis
All measurements were made by 1 investigator (I.W.L.). A pilot study comprising 16 subjects with 20 palatally displaced canines, matched to 10 normal subjects with 20 control canines was conducted. Since the sample size of the pilot study was too small to do an a priori power analysis for running a multiple linear regression, a post hoc power analysis was used to examine
the power of the data. Power analysis was carried out by using G*Power 3 software.12 Statistical analysis was performed with the SPSS Statistics Standard GradPack (version 19.0; IBM, Armonk, NY). The measurement data were analyzed for normality by using a normal probability plot and the Shapiro-Wilks test. Independent t tests and MannWhitney U tests were used accordingly to compare the palatally displaced canine and control groups. A significance level of 0.005 was used to account for multiple comparisons. A linear regression with backward examination was conducted to look for the most significant independent variables for each canine variable (Can_Sag, Can_Cor, Can_Occl). Fifteen independent variables (MDsect [mesiodistal position of the canine cusp tip], Vzone [vertical position of the canine cusp tip], Lat_Sag, Lat_Cor, Lat_Rot, Lat_Lgth, MD_CEJ, BL_CEJ, MD_CEJ4, BL_CEJ4, MD_CEJ8, BL_CEJ8, age, sex, and resorption) were included in the initial analysis. Because of the great colinearity between the variables and the large number of variables considered, a principal component analysis was used to analyze the relationships between the variables. After variable reduction procedures, multiple linear regressions were used to analyze the relationship between the canine variables and the lateral incisor variables, taking account the confounding variables. Linear regressions were used in this study rather than analyses with correlation matrices with bivariate correlations because of the multiple independent variables and also the correlation between the independent variables and the involvement of confounding variables. Since the variables MDsect and Vzone were not independent from
American Journal of Orthodontics and Dentofacial Orthopedics
May 2013 Vol 143 Issue 5
Liuk et al
626
Fig 5. Box plot of angular variables ( ) comparing the palatally displaced canine and the control groups. Can_Sag, Canine sagittal angulation; Can_Cor, canine coronal angulation; Lat_Sag, lateral incisor sagittal angulation; Lat_Cor, lateral incisor coronal angulation; Lat_Rot, lateral incisor rotational angulation.
the canine variables Can_Sag, Can_Cor, and Can_Occl, they were not included in the regression analysis. RESULTS
A post hoc power analysis was carried out to examine the linear regression on the sagittal angulation of the canine, and it was found that the power was 82.2%. We analyzed 46 palatally displaced canines and 60 control canines in this study. All data were normally distributed, except for those of the coronal angulation of the canine and the distance of the canine cusp tip to the occlusal plane from the control group. Thirteen canine and lateral incisor variables were compared between the palatally displaced canine and the control groups (Figs 5 and 6) with independent t tests (assuming unequal variance) and Mann-Whitney U tests accordingly (Table I). The canine variables between the 2 groups were significantly different (P \0.001). Also, the lateral incisor variables between the 2 groups were significantly different (P \0.005), except for the mesiodistal width at 4 mm apical to the CEJ level (P .0.05), whereas the mesiodistal width at 8 mm apical to the CEJ level and the average of the mesiodistal widths at the 3 levels were borderline significant (P \0.05).
May 2013 Vol 143 Issue 5
In this study, the maxillary lateral incisors in the palatally displaced canine group tended to be more upright in the sagittal (P \0.005) and coronal (P \0.001) planes, and more mesiolabially rotated (P \0.005) compared with the control group. The means in the palatally displaced canine group were 17.7 for the sagittal angulation of the lateral incisor, 9.8 for the coronal angulation of the lateral incisor, and 45.3 for the rotation of the lateral incisor; the measurements in the control group were 30.9 , 14.8 , and 57.1 , respectively. The mean of the sagittal angulation of the canine in the palatally displaced canine group was 36.0 , and the coronal angulation of the canine was 35.5 (palatally inclined); these measurements in the control group were 22.7 and 7.1 , respectively. Cohen13 suggested that the strength of correlation coefficients of 0.10 to 0.29 is small, 0.30 to 0.49 is moderate, and 0.50 to 1.0 is large or high. Patient age was found to be correlated moderately with the mesiodistal position of the canine cusp tip and the vertical position of the canine cusp tip (Table II). The sagittal angulation of the canine and the coronal angulation of the canine were highly correlated with each other, and also with the mesiodistal position of the canine cusp tip. The
American Journal of Orthodontics and Dentofacial Orthopedics
Liuk et al
627
Fig 6. Box plot of linear variables (mm) comparing the palatally displaced canine and the control groups. Can_Occl, Canine cusp tip to the occlusal plane distance; Lat_Lgth, lateral incisor length; MD_CEJ, mesiodistal width at the CEJ level; BL_CEJ, buccolingual width at the CEJ level; MD_CEJ4, mesiodistal width at 4 mm apical to the CEJ level; BL_CEJ4, buccolingual width at 4 mm apical to the CEJ level; MD_CEJ8, mesiodistal width at 8 mm apical to the CEJ level; BL_CEJ8, buccolingual width at 8 mm apical to the CEJ level.
distance of the canine cusp tip to the occlusal plane was moderately correlated with the coronal angulation of the canine and the vertical position of the canine cusp tip. The mesiodistal position of the canine cusp tip did not correlate with either the vertical position of the canine cusp tip or the distance of the canine cusp tip to the occlusal plane. Pearson's correlations between these variables are presented in Table III. According to the principal component analysis with varimax rotation, some variables actually correlate to each other and are clustered together as shown in the pattern matrix table (Table IV). The varimax rotation method was used, since the initial direct quartimin rotation method showed that the correlation coefficients between the components were lower than 0.32 (Tabachnick and Fidell,14 Pallant15). The variable can be said to load on a given component when the loading factor is 0.32 or greater for that component but not on other components.14 Table IV shows some groups of variables that are closely related to one another. The first component contains variables related to the buccolingual widths of the lateral incisors. The second component contains
variables related to the mesiodistal widths of the lateral incisors. The third component contains variables related to the coronal and sagittal angulations of the canines. It was found that the buccolingual width at the CEJ level, the buccolingual width at 4 mm apical to the CEJ level, and the buccolingual width at 8 mm apical to the CEJ level were correlated on the same principal component and thus can be reduced to a single variable; therefore, the 3 measurements can be averaged to give 1 measurement that is the average of the buccolingual widths at the 3 levels. Similarly, the mesiodistal width at the CEJ level, the mesiodistal width at 4 mm apical to the CEJ level, and the mesiodistal width at 8 mm apical to the CEJ level could be reduced to 1 variable that is the average of the mesiodistal widths at the 3 levels. The mesiodistal position of the canine cusp tip was also found to correlate with the sagittal angulation of the canine and the coronal angulation of the canine on the same principal component and hence will not be included in subsequent regression analysis. Likewise, the vertical position of the canine cusp tip was correlated with the distance of the canine cusp tip to the occlusal plane on the same principal component and was
American Journal of Orthodontics and Dentofacial Orthopedics
May 2013 Vol 143 Issue 5
Liuk et al
628
Table I. Results of independent t tests and Mann-Whitney U tests comparing the angular and linear variables between the palatally displaced canine (PDC) and control groups PDC group (n 5 46) Variable Can_Sag ( ) Can_Cor ( ) Can_Occl (mm) Lat_Sag ( ) Lat_Cor ( ) Lat_Rot ( ) Lat_Lgth (mm) MD_CEJ (mm) MD_CEJ4 (mm) MD_CEJ8 (mm) MD_Ave (mm) BL_CEJ (mm) BL_CEJ4 (mm) BL_CEJ8 (mm) BL_Ave (mm)
Mean 36.0 35.5 9.8 17.7 9.8 45.3 21.3 4.3 3.4 2.5 3.4 5.9 4.8 3.6 4.8
SD 8.7 13.1 1.5 6.6 8.7 19.8 1.8 0.4 0.5 0.4 0.4 0.8 0.9 0.9 0.8
Control group (n 5 60) Mean 22.7 7.1 1.8 30.9 14.8 57.1 23.3 4.6 3.5 2.7 3.6 6.6 5.4 4.3 5.4
SD 4.5 8.1 3.6 5.7 4.6 12.3 2.7 0.6 0.5 0.5 0.5 0.6 0.8 0.8 0.7
t test P value \0.001y \0.001y \0.001y \0.001y 0.001y 0.001y \0.001y 0.004y 0.228 0.013* 0.016* \0.001y \0.001y \0.001y \0.001y
Mean difference 13.3 42.6 8.0 13.2 5.0 11.7 2.1 0.3 0.1 0.2 0.2 0.7 0.6 0.7 0.7
Mann-Whitney U test P value \0.001y \0.001y \0.001y \0.001y 0.003y \0.001y \0.001y 0.003y 0.341 0.024* 0.041* \0.001y 0.003y 0.001y 0.001y
Can_Sag, Canine sagittal angulation; Can_Cor, canine coronal angulation; Can_Occl, canine cusp tip to occlusal plane distance; Lat_Sag, lateral incisor sagittal angulation; Lat_Cor, lateral incisor coronal angulation; Lat_Rot, lateral incisor rotational angulation; Lat_Lgth, lateral incisor length; MD_CEJ, mesiodistal width at the CEJ level; MD_CEJ4, mesiodistal width at 4 mm apical to the CEJ level; MD_CEJ8, mesiodistal width at 8 mm apical to the CEJ level; MD_Ave, average of MD_CEJ, MD_CEJ4, and MD_CEJ8; BL_CEJ, buccolingual width at the CEJ level; BL_CEJ4, buccolingual width at 4 mm apical to the CEJ level; BL_CEJ8, buccolingual width at 8 mm apical to the CEJ level; BL_Ave, average of BL_CEJ, BL_CEJ4, and BL_CEJ8. *Significant at the 0.05 level; ysignificant at the 0.005 level.
therefore excluded from subsequent regression analysis. As a result, 15 initial independent variables were reduced to 9 independent variables that were then examined in the regression analysis. It was also found from the rotated component matrix table that the lateral incisor length actually loaded on both principal components 1 and 2, denoting that lateral incisor length was correlated with buccolingual width at the CEJ level, buccolingual width at 4 mm apical to the CEJ level, and buccolingual width at 8 mm apical to the CEJ level, and also with mesiodistal width at the CEJ level, mesiodistal width at 4 mm apical to the CEJ level, and mesiodistal width at 8 mm apical to the CEJ level. Nine independent variables—sagittal angulations of the lateral incisor, coronal angulations of the lateral incisor, rotation of the lateral incisor, the lateral incisor length, average of mesiodistal widths at the 3 levels, average of buccolingual widths at the 3 levels, age, sex, and resorption—were included in the initial model of linear regression (Table V). It was found that 30.8% (R2 5 0.308) of the variation in the sagittal angulation of the canine and 30.6% (R2 5 0.306) of the variation in the coronal angulation of the canine can be explained by the 9 independent variables. The most significant independent variables according to the backward examination of linear regression for the sagittal angulation of the canine were the coronal angulations of the lateral
May 2013 Vol 143 Issue 5
Table II. Correlation of MDsect and Vzone with age
and sex Age Variable MDsect Vzone
Pearson correlation 0.400 0.325
Sex P value 0.006* 0.028*
Pearson correlation 0.007 0.165
P value 0.961 0.272
MDsect, Mesiodistal position of canine (sectors I to IV); Vzone, vertical position of canine (zones 1 to 5). *Significant at the 0.05 level.
incisor; for the coronal angulation of the canine, they were lateral incisor length and age. Also, 32.9% (R2 5 0.329) of the variation in the distance of the canine cusp tip to the occlusal plane can be explained by the 9 independent variables, and the most significant variables were the sagittal angulations of the lateral incisor, the coronal angulations of the lateral incisor, and the average of the buccolingual widths at the 3 levels. DISCUSSION
The differences between the palatally displaced canine and control groups for the sagittal and coronal angulations and displacement from the occlusal plane of the canine describe the canine ectopia and are consistent with other studies.16-19
American Journal of Orthodontics and Dentofacial Orthopedics
Liuk et al
629
Table III. Correlation between canine variables, MDsect, and Vzone in the palatally displaced canine group Can_Sag Variable Can_Sag Can_Cor Can_Occl MDsect Vzone
Can_Cor
Pearson correlation
P value
0.518 0.251 0.548 0.063
\0.001y 0.093 \0.001y 0.678
Pearson correlation 0.518
P value \0.001y
0.394 0.612 0.271
0.007* \0.001y 0.069
Can_Occl Pearson correlation 0.251 0.394 0.244 0.438
P value 0.093 0.007* 0.102 0.002y
MDsect Pearson correlation 0.548 0.612 0.244 0.207
Vzone
P value \0.001y \0.001y 0.102
Pearson correlation 0.063 0.271 0.438 0.207
P value 0.678 0.069 0.002y 0.169
0.169
Can_Sag, Canine sagittal angulation; Can_Cor, canine coronal angulation; Can_Occl, canine cusp tip to occlusal plane distance; MDsect, mesiodistal position of the canine (sectors I to IV); Vzone, vertical position of the canine (zones 1 to 5). *Significant at the 0.05 level; ysignificant at the 0.005 level.
Table IV. Rotated component matrix showing the 6 components extracted by the principal component analysis Component/Total variance explained Variable BL_CEJ4 BL_CEJ Lat_Cor BL_CEJ8 Lat_Rot Lat_Lgth MD_CEJ4 MD_CEJ8 MD_CEJ Can_Cor MDsect Can_Sag Age Vzone Can_Occl Lat_Sag Sex Resorption
1/30% 0.904* 0.884* 0.794* 0.762* 0.543* 0.535* 0.194 0.153 0.105 0.058 0.317 0.297 0.010 0.005 0.048 0.277 0.079 0.020
2/14% 0.227 0.192 0.152 0.337* 0.117 0.511* 0.914* 0.835* 0.800* 0.047 0.103 0.011 0.263 0.200 0.150 0.107 0.597* 0.014
3/12% 0.078 0.020 0.131 0.292 0.151 0.178 0.101 0.224 0.085 0.839* 0.762* 0.746* 0.608* 0.127 0.283 0.003 0.070 0.011
4/8% 0.001 0.078 0.227 0.040 0.132 0.200 0.041 0.166 0.243 0.208 0.169 0.045 0.102 0.793* 0.739* 0.571* 0.019 0.106
5/7% 0.024 0.125 0.096 0.028 0.535* 0.206 0.026 0.030 0.211 0.049 0.000 0.315 0.472* 0.090 0.306 0.179 0.623* 0.051
6/6% 0.063 0.157 0.096 0.045 0.056 0.010 0.089 0.027 0.104 0.024 0.253 0.185 0.087 0.235 0.020 0.459* 0.029 0.895*
Can_Sag, Canine sagittal angulation; Can_Cor, canine coronal angulation; Can_Occl, canine cusp tip to occlusal plane distance; Lat_Sag, lateral incisor sagittal angulation; Lat_Cor, lateral incisor coronal angulation; Lat_Rot, lateral incisor rotational angulation; Lat_Lgth, lateral incisor length; MD_CEJ, mesiodistal width at the CEJ level; MD_CEJ4, mesiodistal width at 4 mm apical to the CEJ level; MD_CEJ8, mesiodistal width at 8 mm apical to the CEJ level; MD_Ave, average of MD_CEJ, MD_CEJ4, and MD_CEJ8; BL_CEJ, buccolingual width at the CEJ level; BL_CEJ4, buccolingual width at 4 mm apical to the CEJ level; BL_CEJ8, buccolingual width at 8 mm apical to the CEJ level; BL_Ave, average of BL_CEJ, BL_CEJ4, and BL_CEJ8; MDsect, mesiodistal position of the canine (sectors I to IV); Vzone, vertical position of the canine (zones 1 to 5). *Loading factor with coefficient $0.32.
In this study, on average, the maxillary lateral incisors of the palatally displaced canine group were more retroclined by 13.2 in the sagittal plane and more upright by 5 in the coronal plane compared with those of the control group. A cover bite, defined as a deep overbite with or caused by retroclination of the maxillary incisors, has been found to be prevalent in a palatally displaced canine group.20 Another study on Class II Division 2 patients reported that 7.5% of the patients had unilateral or bilateral peg-shaped lateral incisors, whereas 33.5% had unilateral or bilateral palatally displaced canines.21
The maxillary lateral incisors of the palatally displaced canine group in this study were more mesiolabially rotated by 11.7 compared with those of the control group. A previous study reported that the maxillary lateral incisors were more mesiolabially rotated when they were adjacent to palatally displaced canines with a more severe impaction.22 Some authors claimed that palatally displaced canines were not associated with rotation of the lateral incisors, but they only recorded incisor rotation that was more than 45 in their study.5
American Journal of Orthodontics and Dentofacial Orthopedics
May 2013 Vol 143 Issue 5
Liuk et al
630
Table V. Results of multiple linear regressions for the dependent canine variables in the palatally displaced canine
group Initial model from standard linear regression Unstandardized coefficients Dependent Independent variable variable Can_Sag (Constant) Lat_Sag Lat_Cor Lat_Rot Lat_Lgth MD_Ave BL_Ave Age Sex Resorption Can_Cor (Constant) Lat_Sag Lat_Cor Lat_Rot Lat_Lgth MD_Ave BL_Ave Age Sex Resorption Can_Occl (Constant) Lat_Sag Lat_Cor Lat_Rot Lat_Lgth MD_Ave BL_Ave Age Sex Resorption
B 45.97 0.25 0.32 0.08 1.71 2.19 0.09 1.59 0.01 0.72 12.91 0.28 0.02 0.02 3.97 5.90 3.83 4.55 1.99 2.13 9.04 0.06 0.09 0.01 0.19 0.15 1.13 0.09 0.28 0.12
SE 23.92 0.21 0.19 0.07 1.06 4.29 2.61 0.88 3.06 2.68 36.33 0.32 0.28 0.11 1.60 6.51 3.96 1.33 4.65 4.08 4.04 0.04 0.03 0.01 0.18 0.72 0.44 0.15 0.52 0.45
Final model from stepwise linear regression Unstandardized coefficients
Beta 0.19 0.32 0.19 0.35 0.10 0.01 0.28 0.00 0.04 0.14 0.01 0.02 0.54 0.18 0.24 0.53 0.08 0.08 0.24 0.55 0.13 0.23 0.04 0.62 0.10 0.10 0.04
P value R2 0.063 0.308 0.249 0.096 0.271 0.114 0.613 0.971 0.079 0.998 0.790 0.724 0.306 0.394 0.946 0.894 0.018* 0.371 0.340 0.002y 0.671 0.604 0.032 0.329 0.135 0.005y 0.426 0.295 0.835 0.015* 0.530 0.586 0.792
B 39.72
SE 1.81
Beta
0.38
0.14
0.38
15.72
16.30
3.69
1.17
0.43
7.35 0.07 0.09
1.35 0.03 0.03
0.30 0.52
0.94
0.32
0.52
P value \0.001
R2 0.144
0.009*
0.340
0.185
0.003y
\0.001 0.262 0.038* 0.004y
0.005y
B, Unstandardized coefficients; Beta, standardized coefficients; SE, standard error; R2, coefficient of determination or squared correlation coefficient; Can_Sag, canine sagittal angulation; Can_Cor, canine coronal angulation; Can_Occl, canine cusp tip to occlusal plane distance; Lat_Sag, lateral incisor sagittal angulation; Lat_Cor, lateral incisor coronal angulation; Lat_Rot, lateral incisor rotational angulation; Lat_Lgth, lateral incisor length; MD_CEJ, mesiodistal width at the CEJ level; MD_CEJ4, mesiodistal width at 4 mm apical to the CEJ level; MD_CEJ8, mesiodistal width at 8 mm apical to the CEJ level; MD_Ave, average of MD_CEJ, MD_CEJ4, and MD_CEJ8; BL_CEJ, buccolingual width at the CEJ level; BL_CEJ4, buccolingual width at 4 mm apical to the CEJ level; BL_CEJ8, buccolingual width at 8 mm apical to the CEJ level; BL_Ave, average of BL_CEJ, BL_CEJ4, and BL_CEJ8. *Significant at the 0.05 level; ysignificant at the 0.005 level.
In this study, as the lateral incisor angulation in the coronal view decreased, the canine angulation in the sagittal view increased (negatively correlated). Thus, clinically, the lateral incisors adjacent to palatally displaced canines will appear to be less proclined or more upright. Impacted canines have been associated with both proclined and palatally tipped lateral incisors, depending on their buccal or palatal impaction, respectively.5,6 Ericson and Kurol5 suggested that lateral incisors with distal crown tipping and proclination might hint at buccally displaced canines. However, they did not draw any conclusions about the clinical appearance of
May 2013 Vol 143 Issue 5
lateral incisors in relation to palatally displaced canines. Jacobs suggested that a buccally directed force from the palatally displaced canine might cause the incisor root to tip labially and its crown palatally.6 This scenario cannot be discounted by our findings. In this study, as the length of the maxillary lateral incisor decreased, the canine tended to incline more palatally. The lateral incisor root length could be a critical factor in the progression of canine impaction, and it has been previously suggested that a short root compromised the canine guidance provided by the lateral incisor.23
American Journal of Orthodontics and Dentofacial Orthopedics
Liuk et al
In a previous study, there was a tendency for younger subjects to have less severe sectors of impaction.22 In our study, canine impaction was generally more severe with increasing age, as demonstrated by increased crown angulation toward the palatal side. This is contrary to another study reporting that younger patients had more severe impactions.24 In this study, when the lateral incisor crown displayed reduced inclination in the sagittal or coronal plane (the lateral incisor crown tipped more palatally or mesially), the canine would be displaced at a higher level from the occlusal plane. A recent study of normal children in the mixed dentition reported a positive correlation between the distance of the canine cusp tip from the occlusal plane and the angulation of the lateral incisor in the coronal view, but not the angulation of the lateral incisor in the sagittal view.25 This seems to disagree with our study, but comparison between these 2 studies is difficult, since the sample groups and study designs were different. In this study, the average buccolingual root widths of the maxillary lateral incisors in the palatally displaced canine group were significantly smaller than those of the control group. Nevertheless, in the palatally displaced canine group, the average maxillary lateral incisor buccolingual root width was found to be associated positively with the canine cusp tip distance from the occlusal plane. This means that palatally displaced canines with higher vertical displacements in this study were associated with greater buccolingual maxillary lateral incisor root widths; this seems to be contrary to our finding that the average buccolingual root width of the maxillary lateral incisors in the palatally displaced canine group was significantly smaller than those of the control group. No previous study has addressed this issue. The lateral incisor might play an obstructive role, and it has been advocated that anomalous and late-developing lateral incisors could redeflect the canine, causing second-stage impaction as 1 of the 5 elements in the guidance theory.3 To assimilate both aforementioned findings, it seems that maxillary lateral incisors with significantly smaller buccolingual root widths might cause first-stage impaction of the canine in the guidance theory. However, among these diminutive lateral incisors, those with greater buccolingual root widths appear to exert a greater effect in causing second-stage impaction of the canine. This could explain the finding in this study that the palatally displaced canines with higher vertical displacements were associated with maxillary lateral incisors with greater buccolingual root widths. The intrabony migration of canines might be a result of teeth competing for space and searching for the
631
path of least resistance.26 It has been stated that teeth erupt in the path of least resistance, and this is why a displaced canine can correct itself after removal of the deciduous canine.26 This study supports the hypothesis that the root of the lateral incisor acts as a natural barrier to the canine's initial mesial and palatal movement.2 The most significant independent variables in the model of linear regression denoted the associations between the canine and lateral incisor variables. However, linear regression will not confirm the causal relationships among these variables. CONCLUSIONS
1.
2.
3.
Maxillary lateral incisors associated with palatally displaced canines on average were more retroclined in the sagittal plane by 13.2 , more upright in the coronal plane by 5 , and more rotated mesiolabially by 11.7 than were those adjacent to normal canines. Regression analyses showed that palatally displaced canines were tipped more mesially when the lateral incisors were more upright in the coronal plane, palatally displaced canines became more palatally tipped with increasing age and in relation to shorter lateral incisor root lengths, and the vertical displacements of palatally displaced canines from the occlusal plane were greater when the lateral incisor was more upright in the coronal and sagittal planes, and when the buccolingual width of the lateral incisor was increased. We concluded from this study that the orientation and the location of palatally displaced canines were associated with changes in the angulations of the maxillary lateral incisors and with small lateral incisors.
REFERENCES 1. Camilleri S, Lewis CM, McDonald F. Ectopic maxillary canines: segregation analysis and a twin study. J Dent Res 2008;87: 580-3. 2. Becker A. In defense of the guidance theory of palatal canine displacement. Angle Orthod 1995;65:95-8. 3. Becker A. The orthodontic treatment of impacted teeth. 2nd ed. New York: Thieme; 2007. 4. Broadbent BH. Ontogenetic development of occlusion. Angle Orthod 1941;11:223-41. 5. Ericson S, Kurol J. Longitudinal study and analysis of clinical supervision of maxillary canine eruption. Community Dent Oral Epidemiol 1986;14:172-6. 6. Jacobs SG. Localization of the unerupted maxillary canine: how to and when to. Am J Orthod Dentofacial Orthop 1999;115:314-22. 7. Shapira Y, Kuftinec MM. Early diagnosis and interception of potential maxillary canine impaction. J Am Dent Assoc 1998; 129:1450-4.
American Journal of Orthodontics and Dentofacial Orthopedics
May 2013 Vol 143 Issue 5
Liuk et al
632
8. Lindauer SJ, Rubenstein LK, Hang WM, Andersen WC, Isaacson RJ. Canine impaction identified early with panoramic radiographs. J Am Dent Assoc 1992;123:91-2:95-7. 9. Kau CH, Pan P, Gallerano RL, English JD. A novel 3D classification system for canine impactions—the KPG index. Int J Med Robot 2009;5:291-6. 10. Liu DG, Zhang WL, Zhang ZY, Wu YT, Ma XC. Localization of impacted maxillary canines and observation of adjacent incisor resorption with cone-beam computed tomography. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008;105:91-8. 11. Liuk IW, Olive RJ, Griffin M, Monsour P. Maxillary lateral incisor morphology and palatally displaced canines: a case-controlled cone-beam volumetric tomography study. Am J Orthod Dentofacial Orthop 2013;143:522-6. 12. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 2007;39:175-91. 13. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale, NJ: L. Erlbaum Associates; 1988. 14. Tabachnick BG, Fidell LS. Using multivariate statistics. 5th ed. Boston: Pearson/Allyn & Bacon; 2007. 15. Pallant J. SPSS survival manual: a step by step guide to data analysis using SPSS. 4th ed. Crows Nest, New South Wales: Allen & Unwin; 2011. 16. Warford JH Jr, Grandhi RK, Tira DE. Prediction of maxillary canine impaction using sectors and angular measurement. Am J Orthod Dentofacial Orthop 2003;124:651-5.
May 2013 Vol 143 Issue 5
17. Ericson S, Kurol J. Early treatment of palatally erupting maxillary canines by extraction of the primary canines. Eur J Orthod 1988; 10:283-95. 18. Diamanti J, Townsend GC. New standards for permanent tooth emergence in Australian children. Aust Dent J 2003;48:39-42. 19. Berkovitz BK. How teeth erupt. Dent Update 1990;17:206-10. 20. Leifert S, Jonas IE. Dental anomalies as a microsymptom of palatal canine displacement. J Orofac Orthop 2003;64:108-20. 21. Basdra EK, Kiokpasoglou M, Stellzig A. The Class II Division 2 craniofacial type is associated with numerous congenital tooth anomalies. Eur J Orthod 2000;22:529-35. 22. Olive RJ. Factors influencing the non-surgical eruption of palatally impacted canines. Aust Orthod J 2005;21:95-101. 23. Becker A, Zilberman Y, Tsur B. Root length of lateral incisors adjacent to palatally-displaced maxillary cuspids. Angle Orthod 1984; 54:218-25. 24. Stewart JA, Heo G, Glover KE, Williamson PC, Lam EW, Major PW. Factors that relate to treatment duration for patients with palatally impacted maxillary canines. Am J Orthod Dentofacial Orthop 2001;119:216-25. 25. Baratieri C, Canongia AC, Bolognese AM. Relationship between maxillary canine intra-alveolar position and maxillary incisor angulation: a cone beam computed tomography study. Braz Dent J 2011;22:146-50. 26. Ngan PW, Wolf T, Kassoy G. Early diagnosis and prevention of impaction of the maxillary canine. ASDC J Dent Child 1987;54: 335-8.
American Journal of Orthodontics and Dentofacial Orthopedics