Vol. 113 No.

2 February 2012

ORIGINAL ARTICLE
Three-dimensional assessment of impacted canines and root
resorption using cone beam computed tomography
Snehlata Oberoi, DDS,a and Stephanie Knueppel, BS,b San Francisco, CA
UNIVERSITY OF CALIFORNIA, SAN FRANCISCO

Objective. The objective of this study was to localize impacted canines in 3 dimensions and determine the most common
location of impaction using cone beam computed tomography (CBCT). We also assessed root resorption of adjacent teeth.
The cusp tip of each impacted canine was located and digitized using Dolphin 3D imaging, after viewing sagittal, coronal,
and axial views. The position on the occlusal plane where the normally erupted canine cusp tip should be located was used
as a control reference point. The degree of impaction was defined by the difference between the impacted canine cusp tip
and reference cusp tip positions.
Materials and Methods. CBCT scans of 29 consecutive individuals with impacted canines undergoing orthodontic treatment
at the UCSF Orthodontic Clinic were included in this study.
Results. The average degree of mesial impactions was 10.1 mm, of distal impactions was 4.2 mm, of facial impactions was
4.16 mm, of palatal impactions was 1.8 mm, and of gingival impactions was 10 mm; 40.4% had no root resorption, 35.7%
showed slight root resorption, 14.2% showed moderate resorption, and 4% showed severe root resorption of the adjacent
lateral incisor.
Conclusion. We reliably assessed the position of impacted canines in 3 dimensions using CBCT, thereby improving accuracy
of location and facilitating precise surgical and orthodontic management. In our study 40.4% had no root resorption, 35.7%
showed slight root resorption, 14.2% showed moderate resorption, and 4% showed severe root resorption of the adjacent
lateral incisor. The most frequent location of impacted canines was palatal, mesial, and gingival. (Oral Surg Oral Med Oral
Pathol Oral Radiol 2012;113:260-267)

Maxillary canines are the second-most frequently im-
pacted teeth after the third molars with prevalence from
1% to 3%.1-4 The maxillary canine, in contrast to the
third molar, is located in a highly demanding area both
functionally and esthetically.5 Most of the impactions
are palatal (85%); whereas 15% are labial.6-8 Impacted
canines are more commonly seen in females than males
and there is wide variation among different racial pop-
ulations.9,10 Impacted canines can lead to varying de-
grees of resorption of the adjacent teeth, particularly the
lateral incisor, up to 48%.11 Root resorption can be
difficult to diagnose with traditional 2-dimensional
(2D) radiography, particularly if the canine is in direct
palatal or facial position to the lateral incisor roots.12
The etiology of impacted canines is unclear. It is mul-
tifactorial, and genetic and local factors might play a role.
It has been suggested that adjacent peg-shaped or missing
a
Associate Professor, Center for Craniofacial Anomalies, Department
of Orofacial Sciences, School of Dentistry, University of California,
San Francisco.
b
Dental Student, School of Dentistry, University of California, San
Francisco.
Received for publication May 20, 2010; returned for revision Mar 4,
2011; accepted for publication Mar 18, 2011.
© 2012 Elsevier Inc. All rights reserved.
2212-4403/$ - see front matter
doi:10.1016/j.tripleo.2011.03.035
lateral incisors result in palatal impaction because of a
lack of proper guidance to the canine during eruption.10,13
Peck et al.10 suggested that palatal canine impaction is
genetic in origin, whereas facial impaction is because of
inadequate arch length or crowding.
Orthodontic management of the impacted canine re-
quires accurate localization to surgically expose and
retrieve the tooth most efficiently.2 Until recently, con-
ventional 2D radiographic imaging was commonly
used for localization of impacted canines and in treat-
ment planning and evaluation of posttreatment results.
Traditional radiographic imaging is limited because it is
2D and the superimposition of structures reduces the
clinician’s ability to visualize individual structures.
Plain radiographs have several confounding factors,
such as image enlargement and distortion, structure
overlap, limited identifiable landmarks, and positioning
problems, that can adversely affect image quality, thus
increasing the risk of misinterpretation.6,14 Medical
computerized tomography (CT) was an improvement in
that it eliminates overlapping structures and allows for
visualization of specific tissue volume. However, the
high radiation doses patients are exposed to in the
process of obtaining medical CT scans limits its clinical
utility.15 The advent of 3D cone beam computed to-
mography (CBCT) has reduced the radiation dose,
making it an advantageous tool in dentistry.16

260
OOOO
Volume 113, Number 2
A common application of CBCT is in the evaluation
and surgical planning of impacted teeth.17 The addi-
tional third dimension provided by CBCT increases the
information available for the surgeon while planning
exposure or removal. Also, adding a third dimension to
the radiographic information may notably alter the
prevalence of root resorption.18 Currently, protocols
describing the location of impacted canines using
CBCT are based on reference lines both in the horizon-
tal occlusal plane and in the midsagittal plane.4,15 A
limitation of this method is that it uses 2D to quantify
3D images.
To take full advantage of the information provided
by CBCT, it is necessary to interpret volumetric images
on a 3D scale. Such a technique would enable clinicians
to describe and evaluate pathologies, deformities, and
impactions with greater detail and accuracy.19 The 3D
model must be reproducible and easily applied to dif-
ferent individuals allowing the surgeon and orthodon-
tist to determine the best clinical approach to treating
impacted canines. For the surgeon, knowing the precise
location of the tooth would reduce the invasiveness of
surgery. An accurate spatial 3D location of the im-
pacted canine would help to direct the surgeon/ortho-
dontist as to what orthodontic force vector should be
applied to move the canine efficiently, without causing
damage to adjacent teeth.
The aim of our study was to localize impacted ca-
nines in 3 dimensions and determine the most common
location of impaction, using CBCT. We also assessed
root resorption of adjacent teeth.
MATERIAL AND METHODS
CBCT scans of 29 consecutive individuals with im-
pacted canines undergoing orthodontic treatment at the
University of California, San Francisco, Orthodontic
Clinic were included in this study (CHR approval #
H44601-32,871-01). Twenty-six had bilateral impac-
tions and 16 had unilateral impactions. Age range was
10.6 to 28.0 years with a mean age of 16.6 years. The
individuals were scanned in an upright position using
12-inch diameter field of view (FOV) scans by the
Hitachi MercuRay CBCT machine (Hitachi Medical
Corporation, Tokyo, Japan) with parameters of 120
kVp, 15 mA, 0.376-mm slice thickness, and a total of
512 slices in Digital Imaging in Communications in
Medicine (DICOM) format as part of diagnostic eval-
uation for orthodontic treatment.
The cone beam– generated DICOM files were im-
ported into Dolphin 3D Imaging (Chatsworth, CA,
USA) 10.5 and reoriented to set the X, Y, and Z planes
using internal landmarks and cranial base structures. In
the sagittal view, the X-axis was set along the sella-
nasion line (a line from the frontonasal suture to the
ORIGINAL ARTICLE
Oberoi and Chen 261
center of sella tursica). In the coronal view, the line
tangent to the clivus in the base of the skull was set as
the Y-axis. For the axial view, the Z-axis was defined
by the line bisecting the optic foramina. The reorienta-
tion was done only once.
After reorienting, the cusp tip of each impacted ca-
nine was selected in the volumetric view using the
digitizing landmarks tool. These points were verified in
the sagittal, coronal, and axial views (Fig. 1). The X, Y,
and Z coordinates of each digitized point were deter-
mined by the software and pasted onto an Excel spread-
sheet. To determine reliability and repeatability of this
method, measurements were taken twice by the same
observer (time interval ⫽ 1 day) on the original reori-
ented volumetric image. The position on the occlusal
plane where the normally erupted canine cusp tip
should be located was used as a control reference point.
The occlusal plane was defined as the plane touching
the incisal edge of a central maxillary incisor to the
mesio-buccal cusp tips of the first maxillary molars.
The distal line angle of the maxillary lateral incisor and
the mesial line angle of the first premolar (i.e., the teeth
adjacent to the impacted canine) were projected onto
the occlusal plane. The midpoint of a line joining the
points of intersection was used as the reference cusp tip
position. Axial, sagittal, and coronal views were ob-
served to confirm the reference point and its X, Y, and
Z coordinates recorded on the Excel spreadsheet.
The degree of impaction was defined by the differ-
ence in the position of the impacted canine cusp tip and
the reference cusp tip. Presence of the primary canine
and degree of root resorption were also recorded. All 3
views of the DICOM images were observed to catego-
rize root resorption severity based on the following
scale: none (intact root surfaces) to slight (up to half the
dentine thickness to the pulp), moderate (midway to the
pulp or more, pulp lining unbroken), and severe (pulp
exposed).11,20
To determine the reliability and reproducibility of
this method, all measurements were made twice by the
same observer with a time interval of 1 day. X, Y, and
Z values for the 2-days’ measurements of the impacted
cusp tips were plotted against each other. The plots
form a near 45-degree linear relationship. To further
confirm consistency between the 2 sets of measure-
ments, the mean axis value of each point was plotted
against the difference between the 2 readings21 (Fig. 2).
To further study the error we also computed Euclidean
distance between the 2 measurements in the X-, Y-, and
Z-axes. The mean difference and SD of the differences
for the 2 point sets in the X-axis were 0.31 and 0.92,
respectively, in the Y-axis were 0.23 and 1.09, respec-
tively, and in the Z-axis were 0.23 and 1.20, respec-
tively (Fig. 2).
ORAL AND MAXILLOFACIAL RADIOLOGY
262 Oberoi and Chen
Fig. 1. Re-orientation of the volume in coronal, sagittal and
axial planes of view.
RESULTS
The method was found to be reproducible, with differ-
ences between each measurement within 2 SDs of the
mean of the difference (Fig. 2). Most of the individuals
with impacted canines were female (76%). Palatal im-
pactions were more common (60%).
OOOO
February 2012
In the X-axis, 93% (n ⫽ 39) of the impactions were
mesial by 10 to 15 mm; 7% (n ⫽ 3) were distal
impactions by 10 to 15 mm. In the Y-axis, all impacted
cusp tips were located closer to the occlusal plane. In
the Z-axis, 60% (n ⫽ 25) of the impactions were palatal
by 2 to 4 mm; 40% (n ⫽ 16) were facial impactions by
0 to 4 mm. The average degree of mesial impactions
was 10.1 mm (range – 0.4 to 18.5 mm); the average
degree of the distal impaction was 4.2 (range –1.8 to
5.0 mm). The average degree of the facial impaction
was 4.16 mm (range 0.1 to 7.1 mm); the average degree
of palatal impaction was 1.8 mm (range – 0.4 to 8.9
mm). The average degree of the gingival impactions
was 10 mm (range – 0.1 to 21.4 mm).
In our study, 40.4% (n ⫽ 17) had no root resorption,
35.7% (n ⫽ 15) showed slight root resorption, 14.2%
(n ⫽ 6) had moderate resorption, and 4.0% (n ⫽ 4) had
severe root resorption of the adjacent lateral incisor. In
only one individual was there resorption of the central
incisor in addition to the lateral incisors (Fig. 3). In this
individual with bilateral canine impactions, the left
central incisor showed severe signs of resorption,
whereas the right central incisor showed slight root
resorption. Sixteen had retained primary canines (38%).
DISCUSSION
Our main objective was to reliably assess the position
of impacted canines in 3 dimensions and determine the
most common position of the impacted canine incisal
tip, using CBCT. We used a reproducible method that
describes the location of the canine in all 3 planes of
space. The most frequent locations of impacted canines
were palatal, mesial, and gingival.
Studies have shown that the prevalence of maxillary
canine impaction varies from 0.9% to 3.0% in the
general population. In European subjects, palatal canine
impaction was found to be 2 to 3 times more frequent
than facial, whereas in the Asian population, facial or
midalveolar impactions were found to be more com-
mon. The ratio for palatal impaction versus facial im-
paction between European and Asian populations was
found to be 5:1. Palatal impactions were more common
(60%) in our study of a mixed racial population. Other
studies have also found palatal impactions to be more
common.4,6,7 In our study, 76% of the individuals with
impacted canines were female, similar to other studies.4
Distal impactions are rarely mentioned in the literature.
In our study, 7% had distal impactions.
Traditionally, the position of the impacted canine
has been assessed in 2 dimensions: mesial/distal and
buccal/palatal. Studies have shown that 2D imaging
with panoramic x-rays is not a reliable method for
OOOO ORIGINAL ARTICLE
Volume 113, Number 2 Oberoi and Chen 263

Agreement of measurements
5

Mean + 3
2SD
2
Difference

Mean
1

0
-70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50
-1
Mean -
2SD
-2

-3

-4

-5

Average

Comparision of Z-axis Values

50
S e c o n d M e a s u re m e n t

40

30

20
20 30 40 50

First Measurement
Fig. 2. Plot to show agreement of measurements.

localization of impacted canines.22 For accurate lo- CBCT allows description in all planes of space.
cation, the position of the canine must be measured Previously, spatial relationship of the impacted canines
in 3 dimensions. was described using 2D images obtained from CBCT4;
ORAL AND MAXILLOFACIAL RADIOLOGY OOOO
264 Oberoi and Chen February 2012

Fig. 2. (Continued).
OOOO
Volume 113, Number 2
Fig. 3. A panoramic radiograph of a 10.5 year old female with bilateral canine impactions with no sign of root resorption of the
adjacent lateral or central incisors. 3D CBCT images showing axial and sagittal slices that were used to identify the impacted
canines and the slight root resorption on the right central incisor and severe root resorption of the left central incisors.
however, it has been shown that it is feasible to quan-
tify 3D tooth displacements from CBCT data with
sufficient accuracy.23 Three-dimensional volumetric
imaging might provide information for improved diag-
nosis and treatment plans, and ultimately result in more
successful treatment outcomes and improved patient
care.24 Kau et al.25 stated that in complex orthodontic
cases, such as in canine impactions and cleft lip and
palate, 3D imaging is mandatory and CBCT is the
imaging of choice. We have described the 3D assessment
of the eruption path of the canine in individuals with
bone-grafted alveolar clefts using CBCT. We found that
most canines on both the cleft and noncleft sides moved
incisally, facially, and mesially.26 In this study, we have
described the location of the impacted canine in all 3
dimensions and found palatal, mesial, and gingival loca-
tions to be the most common. We have described the 3D
relationship of the impacted canine to its desired final
position in relation to the adjacent teeth.
CBCT, which provides a lower-dose, lower-cost
alternative to conventional CT, is being used with
increasing frequency in the practice of orthodontics
and oral and maxillofacial radiology. Ludlow et al.27
found the calculated doses for a 12-FOV scan in mSv
were NewTom3G (Elmsford, NY, USA; 45, 59), I-CAT
(Hartfield, PA, USA; 135, 193), and CB MercuRay (477,
ORIGINAL ARTICLE
Oberoi and Chen 265
558). These are 4 to 42 times greater than comparable
panoramic examination doses (6.3 mSv, 13.3 mSv). Re-
ductions in dose were seen with reduction in field size and
mA and kV technique factors.27 Based on the diagnostic
requirements, the FOV, mA, and kV can be altered to
reduce the radiation dosage.
Individuals with severe lateral root resorption present
treatment challenges to both orthodontists and oral max-
illofacial surgeons with increased treatment time and ex-
pense. Previous studies on canine impaction–related root
resorption were mainly 2D studies and date back more
than 10 years.28 Although our study supports previous
studies showing that ectopic eruption of the impacted
canine may cause root resorption of the maxillary incisors,
most commonly the lateral incisors, in our study, 40.4%
had no root resorption, 35.7% showed slight root resorp-
tion, 14.2% showed moderate resorption, and 4% showed
severe root resorption of the adjacent lateral incisor. This
is similar to the findings of Ericson et al.,11 where 9% had
moderate resorption but very different from the findings of
60% severe resorption. Other studies have shown root
resorption to vary from 38%29 to 67%.4 CBCT allows
visualization of the roots in all projections and is pre-
sumed to yield a more accurate assessment than has been
obtained in the studies cited previously. It is possible that
we have underestimated root resorption owing to inade-
ORAL AND MAXILLOFACIAL RADIOLOGY
266 Oberoi and Chen
quate visualization, because the large FOV diminishes the
resolution of the image (pixel size 0.377 in a 12-inch scan
vs 0.292 in a 9-inch scan). In our study, the relatively low
incidence of root resorption was probably because of the
low resolution of the images obtained with the CB Mer-
cuRay, which does not allow for clear depiction of resorp-
tion craters. In a study to compare the diagnostic accuracy
of CBCT images of different voxel resolutions to detect
simulated small internal resorption cavities, ultra- and
high-resolution Iluma (Imtec, Ardmore, OK, USA) and
Accuitomo (J. Morita, Kyoto, Japan) CBCT images per-
formed similarly and better than low-resolution Iluma
images in the detection of simulated internal resorption ex
vivo.30
In our study, 38% had primary canines retained. The
percentage of retained primary canines is in accordance
with previous studies of 33.3% to 48.1%.4,15
CONCLUSIONS
We reliably assessed the position of impacted canines
in 3 dimensions using CBCT, thereby improving accu-
racy of location and facilitating precise surgical and
orthodontic management. In our study, 40.4% had no
root resorption, 35.7% showed slight root resorption,
14.2% showed moderate resorption, and 4% showed
severe root resorption of the adjacent lateral incisor.
The most frequent locations of impacted canines were
palatal, mesial, and gingival. This method can also be
applied in clinical studies to evaluate tooth movement
between time points.
REFERENCES
1. Mason C, Papadakou P, Roberts GJ. The radiographic localiza-
tion of impacted maxillary canines: a comparison of methods.
Eur J Orthod 2001;23:25-34.
2. Preda L, La Fianza A, Di Maggio EM, Dore R, Schifino MR,
Campani R, et al. The use of spiral computed tomography in the
localization of impacted maxillary canines. Dentomaxillofac Ra-
diol 1997;26:236-41.
3. 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 Dentofac
Orthop 2001;119:216-25.
4. Walker L, Enciso R, Mah J. Three-dimensional localization of
maxillary canines with cone-beam computed tomography. Am J
Orthod Dentofac Orthop 2005;128:418-23.
5. Fernandez E, Bravo LA, Canteras M. Eruption of the permanent
upper canine: a radiologic study. Am J Orthod Dentofac Orthop
1998;113:414-20.
6. Elefteriadis JN, Athanasiou AE. Evaluation of impacted canines
by means of computerized tomography. Int J Adult Orthodon
Orthognath Surg 1996;11:257-64.
7. Ericson S, Kurol J. Resorption of maxillary lateral incisors
caused by ectopic eruption of the canines. A clinical and radio-
graphic analysis of predisposing factors. Am J Orthod Dentofac
Orthop 1988;94:503-13.
8. Shapira Y, Kuftinec MM. Early diagnosis and interception of
potential maxillary canine impaction. J Am Dent Assoc 1998;
129:1450-4.
OOOO
February 2012
9. Leifert S, Jonas IE. Dental anomalies as a microsymptom of
palatal canine displacement. J Orofac Orthop 2003;64:108-20.
10. Peck S, Peck L, Kataja M. The palatally displaced canine as
a dental anomaly of genetic origin. Angle Orthod 1994;64:
249-56.
11. Ericson S, Kurol PJ. Resorption of incisors after ectopic
eruption of maxillary canines: a CT study. Angle Orthod
2000;70:415-23.
12. Ericson S, Kurol J. CT diagnosis of ectopically erupting maxil-
lary canines—a case report. Eur J Orthod 1988;10:115-21.
13. Becker A, Smith P, Behar R. The incidence of anomalous max-
illary lateral incisors in relation to palatally-displaced cuspids.
Angle Orthod 1981;51:24-9.
14. Waitzman AA, Posnick JC, Armstrong DC, Pron GE. Craniofa-
cial skeletal measurements based on computed tomography: part
II. Normal values and growth trends. Cleft Palate Craniofac J
1992;29:118-28.
15. 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.
16. Boeddinghaus R, Whyte A. Current concepts in maxillofacial
imaging. Eur J Radiol 2008;66:396-418.
17. Scarfe WC, Farman AG, Sukovic P. Clinical applications of
cone-beam computed tomography in dental practice. J Can Dent
Assoc 2006;72:75-80.
18. Alqerban A, Jacobs R, Lambrechts P, Loozen G, Willems G.
Root resorption of the maxillary lateral incisor caused by im-
pacted canine: a literature review. Clin Oral Investig 2009;
13:247-55.
19. Lou L, Lagravere MO, Compton S, Major PW, Flores-Mir C.
Accuracy of measurements and reliability of landmark identifi-
cation with computed tomography (CT) techniques in the max-
illofacial area: a systematic review. Oral Surg Oral Med Oral
Pathol Oral Radiol Endod 2007;104:402-11.
20. Maini A, Durning P, Drage N. Resorption: within or without?
The benefit of cone-beam computed tomography when diagnos-
ing a case of an internal/external resorption defect. Br Dent J
2008;204:135-7.
21. Bland JM, Altman DG. Statistical methods for assessing agree-
ment between two methods of clinical measurement. Lancet
1986;1:307-10.
22. Nagpal A, Pai KM, Setty S, Sharma G. Localization of impacted
maxillary canines using panoramic radiography. J Oral Sci
2009;51:37-45.
23. Chen J, Li S, Fang S. Quantification of tooth displacement from
cone-beam computed tomography images. Am J Orthod Dento-
fac Orthop 2009;136:393-400.
24. Haney SJ, Nicoll R, Mansueto M. Functional impressions for
complete denture fabrication. A modified jump technique. Tex
Dent J 2010;127:377-84.
25. Kau CH, Bozic M, English J, Lee R, Bussa H, Ellis RK, et al.
Cone-beam computed tomography of the maxillofacial re-
gion—an update. Int J Med Robot 2009;5:366-80.
26. Oberoi S, Gill P, Chigurupati R, Hoffman WY, Hatcher DC,
Vargervik K. Three-dimensional assessment of the eruption path
of the canine in individuals with bone-grafted alveolar clefts
using cone beam computed tomography. Cleft Palate Craniofac J
2010;47:507-12.
27. Ludlow JB, Davies-Ludlow LE, Brooks SL, Howerton WB.
Dosimetry of 3 CBCT devices for oral and maxillofacial radiol-
ogy: CB Mercuray, NewTom 3G and i-CAT. Dentomaxillofac
Radiol 2006;35:219-26.
OOOO ORIGINAL ARTICLE
Volume 113, Number 2 Oberoi and Chen 267

28. Alqerban A, Jacobs R, Fieuws S, Willems G. Comparison of two Reprint requests:

cone beam computed tomographic systems versus panoramic im-
Snehlata Oberoi, DDS
aging for localization of impacted maxillary canines and detection
Associate Clinical Professor
of root resorption. Eur J Orthod 2011;33:93-102.
Division of Orthodontics
29. Ericson S, Kurol J. Radiographic assessment of maxillary canine
eruption in children with clinical signs of eruption disturbance. Department of Orofacial Sciences
Eur J Orthod 1986;8:133-40. School of Dentistry
30. Kamburoglu K, Kursun S. A comparison of the diagnostic ac- University of California, San Francisco
curacy of CBCT images of different voxel resolutions used to S747 Parnassas Avenue
detect simulated small internal resorption cavities. Int Endod J San Francisco, CA 94143-044
2010;43:798-807. sneha.oberoi@ucsf.edu