See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/259651318

A comparison of apical root resorption after orthodontic treatment with

surgical exposure and traction of maxillary impacted canines versus that
without impactions

Article in The European Journal of Orthodontics · January 2014

DOI: 10.1093/ejo/cjt099 · Source: PubMed

CITATIONS READS

19 2,185

4 authors:

Evangelia Lempesi Nikolaos Pandis

Hellenic Air Force Universität Bern
5 PUBLICATIONS 98 CITATIONS 411 PUBLICATIONS 7,953 CITATIONS

SEE PROFILE SEE PROFILE

Padhraig S Fleming Maria Mavragani

Trinity College Dublin University of Bergen
221 PUBLICATIONS 4,822 CITATIONS 12 PUBLICATIONS 436 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

The anterior componant View project

Characterization of retrieved fixed retainer multistranded archwires View project

All content following this page was uploaded by Nikolaos Pandis on 24 May 2015.

The user has requested enhancement of the downloaded file.

European Journal of Orthodontics 36 (2014) 690–697 © The Author 2014. Published by Oxford University Press on behalf of the European Orthodontic Society.
doi:10.1093/ejo/cjt099 All rights reserved. For permissions, please email: [email protected]
Advance Access publication 9 January 2014

A comparison of apical root resorption after orthodontic

treatment with surgical exposure and traction of maxillary
impacted canines versus that without impactions
Evangelia Lempesi*, Nikolaos Pandis**,***, Padhraig S. Fleming**** and
Maria Mavragani*
*Orthodontic Department, School of Dentistry, University of Athens, Greece, **Private Practice, Corfu, Greece,
***Department of Orthodontics and Dentofacial Orthopedics, Dental School/Medical Faculty, University of Bern,
Switzerland, ****Institute of Dentistry, Queen Mary University of London, UK

Correspondence to: Evangelia Lempesi, Aristeidou 37, 15234 Chalandri, Athens, Greece. E-mail: [email protected]

Summary
Background/objectives: Orthodontic management of maxillary canine impaction (MCI), including forced
eruption, may result in significant root resorption; however, the association between MCI and orthodonti-
cally induced root resorption (OIRR) is not yet sufficiently established. The purpose of this retrospective
cohort study was to comparatively evaluate the severity of OIRR of maxillary incisors in orthodontically
treated patients with MCI. Additionally, impaction characteristics were associated with OIRR severity.
Subjects and methods: The sample comprised 48 patients undergoing fixed-appliance treatment—24
with unilateral/bilateral MCI and 24 matched controls without impaction. OIRR was calculated using pre-

Downloaded from by guest on January 29, 2015

and post-operative panoramic tomograms. The orientation of eruption path, height, sector location, and
follicle/tooth ratio of the impacted canine were also recorded. Mann–Whitney U-test and univariate and
multivariate linear mixed models were used to test for the associations of interest.
Results: Maxillary central left incisor underwent more OIRR in the impaction group (mean differ-
ence = 0.58 mm, P = 0.04). Overall, the impaction group had 0.38 mm more OIRR compared to the control
(95% confidence interval, CI: 0.03, 0.74; P = 0.04). However, multivariate analysis demonstrated no differ-
ence in the amount of OIRR between impaction and non-impaction groups overall. A positive association
between OIRR and initial root length was observed (95% CI: 0.08, 0.27; P < 0.001). The severity of canine
impaction was not found to be a significant predictor of OIRR.
Limitations: This study was a retrospective study and used panoramic tomograms for OIRR measurements.
Conclusions: This study indicates that MCI is a weak OIRR predictor. Interpretation of the results needs
caution due to the observational nature of the present study.

Introduction
With the exception of the third molars, maxillary canines genetic factors are thought to have a bearing on the sever-
are most likely to be impacted, with the reported frequency ity of resorption, with allelic associations (Hartsfield et al.,
of ectopic canines typically ranging from 0.8 to 3% (Ericson 2004) and various morphological characteristics, includ-
and Kurol, 1986). Maxillary impacted canines can manifest ing blunt, pipette-shaped roots and thin roots, shown to be
in a range of bucco-lingual, vertical, and antero-posterior more prone to resorption (Levander and Malmgren, 1988;
locations with the complexity and duration of treatment Sameshima and Sinclair, 2001, 2004).
believed to relate to the degree of displacement (Stewart Furthermore, both the total treatment time and the mag-
et al., 2001; Zuccatti et al., 2006; Fleming et al., 2009). nitude of tooth movement are implicated in the develop-
A degree of root resorption is an inevitable conse- ment of OIRR (Segal et al., 2004; Weltman et al., 2010).
quence of orthodontic treatment, with the maxillary lateral Treatment time with ectopic canines is known to be consid-
and central incisors typically mostly affected (Linge and erably extended (Mavreas and Athanasiou, 2008); similarly,
Linge, 1983, 1991; Levander and Malmgren, 1988, 2000; significant movement is necessary to correct the position
Mirabella and Årtun, 1995). While typically orthodontically of grossly displaced teeth. Additionally, anchorage demands
induced root resorption (OIRR) is inconsequential and an on adjacent incisors are increased because they are subject
incidental radiographic finding, resorption of an incisor by to considerable reactionary forces during mechanical erup-
5 mm or more has been estimated to occur in 5% of ortho- tion of the canine tooth, predisposing them to resorption
dontic patients (Killiany, 1999). Both morphological and (Woloshyn et al., 1994; Blake et al., 1995).
AFTER SURGICAL EXPOSURE–ORTHODONTIC TRACTION VERSUS RESORPTION WITHOUT IMPACTION 691

The primary aim of this retrospective cohort study was To assess the angulation of the impacted canine, the angle
to compare the severity of root resorption of the maxillary between the long axis of the impacted canine and the upper
incisors during orthodontic treatment in patients under- midline (angle α) and the angle between the long axis of
going mechanical eruption of surgically exposed canines the impacted canine and the long axis of the adjacent lat-
versus resorption in patients without impactions. A sec- eral incisor (angle β; Ericson and Kurol, 1988; Figure 1)
ondary aim was to gauge the influence of radiographic were measured. The maxillary midline was defined from
position of the canine on severity of treatment-induced the following reference points depicted on the panoramic
resorption. radiographs: midpalatal suture, anterior nasal spine, and
nasal septum. The cusp tip of the canine was localized in the
transverse plane in one of five sectors (Ericson and Kurol,
Subjects and methods 1988; Figure 2). In order to determine the height of the
The study protocol was approved by the ethical committee impacted canine (h), the vertical distance from the canine
of the School of Dentistry, University of Athens (decision cusp tip to the occlusal plane was measured (Ericson and
number 189/01.11.12). Kurol, 1988). The occlusal plane for left and right sides was
The present study adopted a retrospective cohort design. determined independently based on a tangent to the incisal
A sample size calculation determined that a minimum of edge of the maxillary central incisor and the occlusal sur-
22 subjects per group would be sufficient to detect a differ- face of the maxillary first permanent molar (Figure 1). The
ence of at least 0.85 mm in root resorption (standard devia- width of the dental follicle was calculated from the ratio of
tion, SD = 1 mm) between the impaction and non-impaction the maximum width of the follicle to the width of the canine
groups (Pandis et al., 2008), with 0.05 significance level crown (Figure 3).
and 80% power (β = 0.2). To determine root length of each incisor, perpendicular
The sample comprised 48 patients undergoing treat- projections from the points representing the incisal edge,
cementoenamel junction (CEJ), and the apex on the long

Downloaded from by guest on January 29, 2015

ment in the Postgraduate Clinic, Orthodontic Department,
School of Dentistry, University of Athens in Greece. In the axis of the tooth (Figure 4) were used. The most distinct
experimental group, 24 patients having maxillary canines CEJ landmark, either mesial or distal, was used, with the
impacted unilaterally or bilaterally, either palatally or buc- same aspect being used both for pre- and post-treatment
cally, were included. The control group was selected to DPTs. A correction factor (C) was estimated in order for
match the experimental group in respect of age, gender, and the pre- and post-treatment panoramic views to be com-
angle classification. parable, ensuring differences were not attributable to une-
Complete orthodontic records, including pre-treatment ven magnification or distortion. Assuming that the crown
and post-treatment panoramic tomograms (DPTs), were pre-
sent for all subjects. All subjects underwent comprehensive
treatment with fixed appliances. Surgical exposure of one or
both maxillary canines was undertaken to facilitate forced
eruption of the canine(s) in the impacted canine group.
Exclusion criteria included craniofacial deformities or syn-
dromes and reshaping of the incisal edge of the maxillary
incisors before final radiographic examination. A range of
demographic and clinical data including gender, age, over-
jet, overbite, angle classification, trauma, habits, agenesis,
oral habits, extraction, elastics, treatment duration, contrac-
tion duration, and general factors were obtained from the
records of each participant.

Data analysis
All preoperative panoramic radiographs were scanned
(Epson Scanner, Expression 1680 Pro) with eight-bit-
gray-scale analysis at 150 dpi (dots per inch) and stored
in uncompressed JPEG format. Following digitization, the
radiographs were viewed randomly with 600% zoom by
one examiner (EL). The dhal software Viewbox, version
4.0.0.105 (Kifissia, Greece), was used for measurement of
data. The position of the impacted canines as depicted on Figure 1 Assessment of canine position, including angle α, angle β, and
initial panoramic views was evaluated quantitatively. height h.
692 E. LEMPESI ET AL.

length was unchanged from initial to final radiographic OIRR = R1 − (R2 × C), where R1 = root length before
examination, the correction factor was calculated as the treatment and R2 = root length after treatment.
ratio of radiographic crown length before treatment (C1) All measurements were performed for each incisor sepa-
to radiographic crown length after treatment (C2). OIRR rately and rounded up to the nearest 0.01 mm. The primary
was defined as the difference in root lengths before and outcome (OIRR) was assessed per tooth in millimetres and
after treatment, after accounting for differences in magni- as a percentage of initial root length.
fication. The formula used to calculate the degree of OIRR The intra-examiner reliability of the measurements of
was as follows: OIRR and tooth location (including angular and linear
measurements) was assessed by re-examining 10 randomly
selected panoramic radiographs 3 weeks after initial evalu-
ation. To quantify random error, the Dahlberg formula
(Dalhberg, 1940) was used: τ = √(∑D2 /2N), where D is the
difference between duplicate measurements and N is the
remeasured sample size. The systematic error was evaluated
using intra-class correlation coefficient (ICC).

Statistical analysis
Statistical analyses were performed at patient and tooth
levels. Descriptive statistical analysis for quantitative
variables was performed and frequency tables for both
Figure 2 Antero-posterior assessment of canine position, based on the
study of Ericson and Kurol (1988).
impaction and control groups were performed for quali-
tative variables. Mann–Whitney U-test was performed to

Downloaded from by guest on January 29, 2015

compare the extent of OIRR between the two treatment
groups for each tooth separately. Additionally, univari-
ate and multivariate linear mixed models were used to
investigate the influence of maxillary canine impaction
(MCI) and other variables, such as age, gender, habits,
and the clinical and therapeutic characteristics (inde-
pendent variables), on OIRR (dependent variable) for
each patient.
The association of impaction variables, such as angle
α, angle β, height, and follicle/tooth ratio for OIRR in the
impaction group was investigated using a multiple regres-
sion model. The level of significance was set at α = 0.05.
Figure 3 Cropped panoramic view showing measurement of follicle/ All analyses were conducted with STATA, version 12.1
tooth ratio. (StataCorp LP, College Station, Texas, USA).

Figure 4 Cropped panoramic radiographs (pre-treatment and post-treatment) of the same patient. Measurement of root length before and after treatment
for maxillary central incisor right. IE, incisal edge; CEJ, cementoenamel junction; Α, apex. (Α) C1, crown length before treatment; R1, root length before
treatment; (Β) C2, crown length after treatment; R2, root length after treatment.
AFTER SURGICAL EXPOSURE–ORTHODONTIC TRACTION VERSUS RESORPTION WITHOUT IMPACTION 693

Results In addition, age, gender, malocclusion characteristics, treat-

ment variables, and the initial degree of displacement of the
The intra-examiner reliability of the method was found to be
canine (Table 5) were not shown to be reliable predictors
excellent (ICC > 0.90) for all parameters. The random error
of root shortening. However, a positive association between
ranged from 0.15 to 0.3 mm for crown length measurement
OIRR and initial root length was observed (95% CI: 0.08,
and from 0.03 to 0.6 mm for root length. Each patient was
0.27; P < 0.001).
considered as a cluster contributing four maxillary incisors,
with the exception of one patient having agenesis of a right
Discussion
lateral incisor and one patient who had the left lateral inci-
sor extracted for orthodontic reasons, i.e. in total, 190 teeth Severe root resorption is one of the most significant and
were investigated in terms of OIRR. In the impaction group, common potential adverse consequences of fixed-appliance
17 subjects had unilateral and 7 had bilateral impaction. orthodontic treatment, with root length loss of more than
The two treatment groups were well matched for all 20% of all four maxillary incisors shown in almost 3%
investigated risk factors of OIRR, with the exceptions of of orthodontic patients (Sameshima and Sinclair, 2004).
overjet and treatment duration (Table 1); the impaction A combination of biological and mechanical factors is impli-
group had a smaller overjet (2.16 mm, SD = 1.76) compared cated in inflammatory OIRR (Brezniak and Wasserstein,
to the mean of the control group (3.50 mm, SD = 2.89; 2002). Specifically, known biological predictors include
P = 0.05). The duration of treatment with fixed appliances genetic susceptibility (Al-Qawasmi et al., 2003; Viecilli
was significantly longer (P =0.001) for the MCI group et al., 2009), and important mechanical factors include the
(41.2 months, SD = 11.9) than was the case in the control degree of required tooth movement, force levels, nature and
group (29.5 months, SD = 11.9). direction of forces, torque movements, and prolonged treat-
The mean amount of root resorption during treatment ment (Segal et al., 2004; Weltman et al., 2010).
ranged from 0.48 to 1.17 mm (4.5–10% of initial root Although biological factors are beyond the control of a

Downloaded from by guest on January 29, 2015

length). OIRR of the maxillary left central incisor was clinician, mechanical factors increase the susceptibility to
shown to be significantly more likely in the impaction root resorption during forced eruption of ectopic canines. In
group, with a mean difference in resorption of 0.57 mm particular, significant tooth movement is necessary, torque is
(P = 0.04; Table 2). Overall, subjects in the impaction group usually important, and treatment tends to be lengthy. In the
suffered from an average of 0.38 mm more resorption than present study, univariate analysis showed that the impaction
their counterparts without impacted teeth (95% confidence group experienced slightly more (0.38 mm) OIRR than the
interval, CI: 0.03, 0.74; P = 0.04; Table 3). However, the control group. However, in the multivariate analysis, canine
multivariate analysis revealed no difference in the amount impaction was found to be a weak risk predictor of OIRR
of OIRR between the impaction and non-impaction groups during orthodontics. This finding relates to the fact that the
overall (β = 0.25, 95% CI: 0.27, 0.78; P = 0.35). Similar maxillary incisors effectively act as anchorage units and are
findings were obtained when the percentage of initial subjected to high intrusive forces during canine eruption.
root length was used as a dependent variable (Table 4). Intrusive forces are believed to place higher compressive

Table 1 Comparison of the two groups in terms of variables associated with orthodontically induced root resorption.

Variables Unit/category Non-impaction Impaction P value*

(n = 24; mean (SD) or %) (n = 24; mean (SD) or %)

Age Years 20.4 (9.2) 20.0 (8.2) 0.90 NS

Gender Male 6 (25%) 7 (29.2%) 0.75 NS
Female 18 (75%) 17 (70.8%)
Angle malocclusion I 11 (45.8%) 9 (37.5%) 0.82 NS
II 10 (41.7%) 11 (45.8%)
III 3 (12.5%) 4 (16.7%)
Habits/trauma/general factors No 10 (41.7%) 7 (29.2%) 0.37 NS
Yes 14 (58.3%) 17 (70.8%)
Agenesis/extraction No 15 (62.5%) 19 (79.2%) 0.20 NS
Yes 9 (37.5%) 5 (20.8%)
Duration of treatment with fixed appliances Months 29.5 (11.9) 41.2 (11.9) 0.001
Initial root length Millimetres 11.33 (1.99) 11.02 (1.84) 0.57 NS
Overjet Millimetres 3.5 (2.89) 2.16 (1.76) 0.05
Overbite Millimetres 2.98 (2.79) 4.17 (2.24) 0.11 NS
Duration of elastics Months 5.45 (5.50) 6.67 (6.43) 0.49 NS
Duration of contraction Months 2.18 (3.76) 1.73 (2.41) 0.63 NS

NS, Not significant; SD, standard deviation. *P value for the comparison of means between the two groups with t-test or percentages with chi-square test
694 E. LEMPESI ET AL.

Table 2 Orthodontically induced root resorption in millimetres and in percentage reduction of initial root length in the two comparison
groups.

Maxillary incisor Maxillary canine impaction group Control group

Mean SD Minimum Maximum Median Range Mean SD Minimum Maximum Median Range

12 n = 23 n = 24
Millimetres — 0.93 1.24 –1.00 4.16 0.87 5.16 — 0.48 0.74 –0.56 2.77 0.29 3.33
% — 8.77 11.74 –9.40 39.58 7.53 48.98 — 4.49 7.23 –6.47 21.58 2.46 28.05
11 n = 24 n = 24
Millimetres — 1.08 1.29 –1.97 3.55 0.87 5.53 — 0.65 0.99 –1.64 2.98 0.45 4.62
% — 9.63 10.95 –16.39 27.84 8.36 44.23 — 5.42 8.29 –13.26 23.46 3.14 36.73
21 n = 24 n = 24
Millimetres — 1.17 1.18 –0.53 4.20 0.93 4.73 — 0.60 0.75 –0.37 2.43 0.35 2.80
% — 9.97 8.93 –3.80 28.28 8.39 32.07 — 5.33 7.11 –3.43 24.42 6.45 27.85
22 n = 23 n = 24
Millimetres — 0.86 1.47 –0.65 4.66 0.33 5.32 — 0.78 0.86 –0.33 3.71 0.65 4.03
% — 7.53 12.58 –7.16 36.40 3.41 43.56 — 7.50 7.76 –3.84 28.56 6.45 32.40

SD, standard deviation.

Table 3 Results of univariate and multivariate regression analyses of the influence of the independent variables maxillary canine
impaction (0 = no, 1 = yes), gender (0 = male, 1 = female), age (years), treatment duration (months), initial root length (millimetres),
overjet (millimetres), and overbite (millimetres) on the dependent variable orthodontically induced root resorption (millimetres).

Downloaded from by guest on January 29, 2015

Unit/category Univariate Multivariate

β 95% CI P value β 95% CI P value

Impaction No Baseline — — Baseline — —

Yes 0.38 0.03, 0.74 0.04 0.25 −0.27, 0.78 0.35
Gender Male Baseline — — Baseline — —
Female 0.01 −−0.42, 0.43 0.97 0.24 −0.23, 0.71 0.31
Age Years 0.01 −0.02, 0.03 0.57 0.01 −0.01, 0.03 0.46
Treatment duration Months 0.01 −0.01, 0.02 0.28 0.01 −0.01, 0.03 0.21
Initial root length Millimetres 0.14 0.05, 0.22 0.002 0.17 0.08, 0.27 <0.001
Overbite Millimetres 0.02 −0.05, 0.10 0.50 0.04 −0.05, 0.13 0.43
Overjet Millimetres −0.02 −0.09, 0.06 0.66 −0.01 −0.11, 0.09 0.87

CI, confidence intervals.

forces on the periodontal ligaments in the apical region, Age, gender, malocclusion characteristics, and treatment
risking root resorption (Han et al., 2005; Harris et al., 2006; variables were not found to be significant predictors of OIRR;
Weltman et al., 2010). Furthermore, during orthodontic however, a positive association between OIRR and initial root
alignment of impacted canines, torque is required to align length was observed. Based on the adjusted analysis, for each
the canine; and torquing moments are therefore transmit- millimetre increase in tooth length, 0.17 mm greater OIRR
ted to the maxillary incisors. Another factor implicated in can be expected. This finding is in agreement with previ-
the increase in OIRR in the presence of canine impaction is ous studies (Mirabella and Årtun, 1995; Sameshima and
the requirement for more prolonged orthodontic treatment Sinclair, 2001) and may stem from the possible requirement
(Mavreas and Athanasiou, 2008). In view of the limited dif- for heavier forces to move teeth with longer roots, and the
ference observed between the groups in the present study, fact that the magnitude of displacement of the root apex is
it may be that certain genetic profiles that may override the larger during tipping or torquing when teeth are longer. While
importance of the afore-mentioned factors exist, thus con- apical shortening arising in teeth with initially shorter roots
cealing the deleterious effect of treatment-related factors. may be of greater concern, neither clear pattern nor increased
It has previously been identified that genetic factors may propensity has emerged from previous research (Levander
account for almost two-thirds of the observed variability in and Malmgren, 1988; Lund et al., 2012). However, in other
the extent of OIRR (Harris et al., 1997; Hartsfield et al., studies an increased susceptibility to resorption has been
2004). described (Taithongchai et al., 1996; Harris et al., 1997).
AFTER SURGICAL EXPOSURE–ORTHODONTIC TRACTION VERSUS RESORPTION WITHOUT IMPACTION 695

Table 4 Results of univariate and multivariate regression analyses of the influence of the independent variables maxillary canine
impaction (0 = no, 1 = yes), gender (0 = male, 1 = female), age (years), treatment duration (months), initial root length (millimetres),
overjet (millimetres), and overbite (millimetres) on the dependent variable percentage reduction of initial root length (%).

Unit/category univariate Multivariate

β 95% CI P value β 95% CI P value

Impaction No Baseline — — Baseline — —

Yes 0.30 0.08,6.51 0.045 2.15 −2.50,6.80 0.37
Gender Male Baseline — — Baseline — —
Female 0.75 −3.03,4.54 0.7 1.78 −2.37,5.93 0.40
Age Years 0.08 −0.11,0.28 0.41 0.10 −0.11,0.31 0.36
Treatment duration Months 0.09 −0.03,0.22 0.15 0.10 −0.07,0.26 0.26
Initial root length Millimetres 0.55 −0.21,1.32 0.17 0.84 0.01,1.67 0.05
Overbite Millimetres 0.17 −0.48,0.82 0.61 0.25 −0.55,1.06 0.54
Overjet Millimetres −0.2 −0.88,0.48 0.57 −0.06 −0.96,0.84 0.90

CI, confidence intervals.

Table 5 Results of multivariate regression analysis of the particularly either case–control or cohort studies in which the
influence of the independent variables angle α (degrees), angle β unavoidable possibility of selection bias issues exist.
(degrees), height (millimetres), follicle/tooth ratio, and sector (1, In the impaction group, treatment time was considerably
2, 3, 4, 5) on the dependent variable orthodontically induced root longer than that in the control group; this is in agreement with

Downloaded from by guest on January 29, 2015

resorption (millimetres).
other studies. Stewart et al. (2001) compared retrospectively
treatment time between young patients (aged 20 years or
Unit/category β 95% CI P value younger) with palatally displaced canines and a control group
without impactions. Orthodontic treatment was 5.9 months
Angle α Degrees 0.01 −0.004, 0.03 0.12 longer on average in the impaction group. In addition, ortho-
Angle β Degrees 0.01 −0.003, 0.03 0.11 dontic treatment of cases with bilaterally impacted canines was
Height Millimetres 0.04 −0.06, 0.13 0.43
Follicle/tooth ratio Ratio 0.4 −0.78, 1.57 0.51 6.5 months longer on average versus cases with unilaterally
Sector 1 Reference Baseline — impacted canines. Similar findings were observed in the pre-
2 0.47 −0.24, 1.18 0.2 sent study, with the mean between-groups difference in treat-
3 0.6 −0.23, 1.42 0.16
4 0.67 −0.63, 1.4 0.07 ment time being almost 12 months. Both unilateral (n = 17) and
5 0.47 −0.36, 1.31 0.27 bilateral (n = 7) impaction cases were considered in the present
research to increase sample size and external validity. A pro-
CI, confidence intervals. spective study including only cases of unilateral impaction may
be a better approach because retrospective studies have several
Selection of the control group was performed by match- limitations when undertaken in an available sample.
ing the impaction group in respect of age, gender, and angle When studying OIRR, it is important to distinguish between
malocclusion. Statistical evaluation revealed a shorter treat- the patient-level and the tooth-level analysis. Measurements
ment duration and larger pre-treatment overjet in the control of different teeth derived from the same subject are correlated
group. Although the treatment duration could have been used with each other. It would be intuitive to expect similar levels
as a covariate in the analysis because it may have an effect of resorption of contralateral incisors to occur during ortho-
on the extent of resorption, covariate matching methods have dontic treatment on the same patient. Based on the tooth-level
two major practical limitations: the number of confounding analysis, it was found that the maxillary central left incisor
variables must be relatively small in respect to the sample underwent more OIRR in the impaction group than the same
size (24 impacted patients) and participants must be available tooth type in the non-impaction group. This finding should
for both groups. Thus matching more than three confounding be interpreted with caution because these statistical tests do
variables was considered prohibitive. Multivariate analysis not account for the clustering effects of correlated data, lead-
accounted for these between-groups differences, thus decreas- ing to the possibility of finding statistically significant results
ing possible confounding effects. Ideally, a randomized clini- that are not genuine (Koletsi et al., 2012). Therefore, further
cal trial would be most appropriate to eliminate any known statistical analysis that considered the correlated nature of the
and unknown confounders; however, such a design to address data was implemented .
this research question is unfeasible. Therefore, knowledge Panoramic radiographs were used to assess OIRR in the
of this topic can only be based upon observational studies, present study. This technique is in keeping with previous
696 E. LEMPESI ET AL.

studies (Apajalahti and Peltola, 2007; Pandis et al., 2008; References

Dudic et al., 2009; Huang et al., 2010). The study of external Al-Qawasmi R A et al. 2003 Genetic predisposition to external apical
lateral root resorption as a consequence of impaction was root resorption. American Journal of Orthodontics and Dentofacial
beyond the scope of this investigation, which was limited to Orthopedics 123: 242–252
root shortening during orthodontic treatment. The latter may Alqerban A, Jacobs R, Souza P C, Willems G 2009 In-vitro comparison
of 2 cone-beam computed tomography systems and panoramic imaging
introduce three-dimensional root resorption of the apex. It for detecting simulated canine impaction-induced external root resorp-
is accepted that panoramic views may be insensitive to very tion in maxillary lateral incisors. American Journal of Orthodontics and
minor changes in root lengths and may be less accurate than Dentofacial Orthopedics 136: 764.e1–764.e11
periapical radiographs in studying the severity of OIRR. Apajalahti S, Peltola J S 2007 Apical root resorption after orthodontic
treatment—a retrospective study. European Journal of Orthodontics 29:
However, the validity of the periapical films in accurately 408–412
depicting OIRR has been also questioned because of the Blake M, Woodside D G, Pharoah M J 1995 A radiographic comparison of
influence of tooth shape and morphology (Katona, 2006), apical root resorption after orthodontic treatment with the edgewise and
inconsistent film and patient positioning, and bending of the Speed appliances. American Journal of Orthodontics and Dentofacial
Orthopedics 108: 76–84
film intra-orally. In a direct comparison of panoramic and
Brezniak N, Wasserstein A 2002 Orthodontically induced inflammatory
periapical films of OIRR, however, the differences accru- root resorption. Part I: The basic science aspects. Angle Orthodontist
ing between the recorded lengths of the maxillary incisors 72: 175–179
was less than 0.2 mm (Sameshima and Asgarifar, 2001). Dalhberg G 1940 Errors of estimation. In: Statistical methods for medical
It is, therefore, reasonable to infer that the use of repeated and biological students. George Allen and Unwin, London, pp. 122–132
panoramic films resulted in a valid assessment of incisor Dudic A, Giannopoulou C, Leuzinger M, Kiliaridis S 2009 Detection of
apical root resorption after orthodontic treatment by using panoramic
length in the present investigation. Comparing DPT with radiography and cone-beam computed tomography of super-high reso-
cone-beam computed tomography (CBCT) in orthodontic lution. American Journal of Orthodontics and Dentofacial Orthopedics
patients, OIRR was found to be underestimated on pano- 135: 434–437
ramic films (Dudic et al., 2009). Alqerban et al. (2009) in Ericson S, Kurol J 1986 Longitudinal study and analysis of clinical super-

Downloaded from by guest on January 29, 2015

vision of maxillary canine eruption. Community Dentistry and Oral
their study demonstrated high sensitivity and specificity for Epidemiology 14: 172–176
CBCT. Although CBCT is considered a more precise tool Ericson S, Kurol J 1988 Early treatment of palatally erupting maxil-
in terms of registration of landmarks and OIRR estima- lary canines by extraction of the primary canines. European Journal
tion, there is still the concern of high radiation exposure, Orthodontist 10: 283–295
particularly in adolescence (SEDENTEXCT, 2012; http:// Fleming P S, Scott P, Heidari N, Dibiase A T 2009 Influence of radio-
graphic position of ectopic canines on the duration of orthodontic treat-
www.sedentexct.eu/files/radiation_protection_172.pdf)]. ment. Angle Orthodontist 79: 442–446
In the present study, the aim was to compare the extent of Han G, Huang S, Von den Hoff J W, Zeng X, Kuijpers-Jagtman A M 2005
OIRR in two groups rather than to determine the absolute Root resorption after orthodontic intrusion and extrusion: an intraindi-
values of root shortening; consequently, panoramic views vidual study. Angle Orthodontist 75: 912–918
were deemed appropriate for this purpose. Harris D A, Jones A S, Darendeliler M A 2006 Physical properties of
root cementum: part 8. Volumetric analysis of root resorption craters
The occurrence of negative values for OIRR could be after application of controlled intrusive light and heavy orthodontic
attributed either to immaturity of certain teeth or to meas- forces: a microcomputed tomography scan study. American Journal of
urement issues relating to DPTs (Mavragani et al., 2002). Orthodontics and Dentofacial Orthopedics 130: 639–647
The youngest patients were 10.3 years old, having root Harris E F, Kineret S E, Tolley E A 1997 A heritable component for exter-
nal apical root resorption in patients treated orthodontically. American
growth potential at the start of treatment. Another source Journal of Orthodontics and Dentofacial Orthopedics 111: 301–309
of potential inaccuracy could relate to differences in incisor Hartsfield J K, Everett E T, Al-Qawasmi R A 2004 Genetic factors in exter-
torque. Planned alteration of incisor inclination may alter nal apical root resorption and orthodontic treatment. Critical Reviews in
apparent tooth lengths radiographically (Apajalahti and Oral Biology and Medicine 15: 115–122
Peltola, 2007). Baseline imbalances in overjet between the Huang Y, Wang X X, Zhang J, Liu C 2010 Root shortening in patients
treated with two-step and en masse space closure procedures with slid-
groups may be indicative of differences in tooth inclinations, ing mechanics. Angle Orthodontist 80: 492–497
although skeletal relationships are also important. However, Katona T R 2006 Flaws in root resorption assessment algorithms: role
multivariate analysis involving multiple predictor variables of tooth shape. American Journal of Orthodontics and Dentofacial
was undertaken to account for this potential difference. Orthopedics 130: 698.e19–698.e27
Killiany D M 1999 Root resorption caused by orthodontic treatment: an evi-
dence-based review of literature. Seminars in Orthodontics 5: 128–133
Conclusions Koletsi D, Pandis N, Polychronopoulou A, Eliades T 2012 Does published
orthodontic research account for clustering effects during statistical data
Patients with at least one impacted maxillary canine ortho- analysis? European Journal of Orthodontics 34: 287–292
dontically treated with surgical exposure and traction Levander E, Malmgren O 1988 Evaluation of the risk of root resorp-
experienced similar levels of root resorption compared tion during orthodontic treatment: a study of upper incisors. European
Journal of Orthodontics 10: 30–38
to orthodontic patients with normally erupting canines.
Levander E, Malmgren O 2000 Long-term follow-up of maxillary incisors
Further research is required to validate the findings from with severe apical root resorption. European Journal of Orthodontics
this investigation. 22: 85–92
 AFTER SURGICAL EXPOSURE–ORTHODONTIC TRACTION VERSUS RESORPTION WITHOUT IMPACTION 697

Linge B O, Linge L 1983 Apical root resorption in upper anterior teeth. SEDENTEXCT 2012 Cone beam CT for dental and maxillofacial radi-
European Journal of Orthodontics 5: 173–183 ology. Evidence-based guidelines. European Commission’s Radiation
Linge L, Linge B O 1991 Patient characteristics and treatment variables asso- Protection No 172. (http://www.sedentexct.eu/files/radiation_protec-
ciated with apical root resorption during orthodontic treatment. American tion_172.pdf) (13 December 2013, date last accessed)
Journal of Orthodontics and Dentofacial Orthopedics 99: 35–43 Segal G R, Schiffman P H, Tuncay O C 2004 Meta analysis of the treat-
Lund H, Gröndahl K, Hansen K, Gröndahl H G 2012 Apical root resorp- ment-related factors of external apical root resorption. Orthodontics and
tion during orthodontic treatment. A prospective study using cone beam Craniofacial Research 7: 71–78
CT. Angle Orthodontist 82: 480–487 Stewart J A, Heo G, Glover K E, Williamson P C, Lam E W, Major P
W 2001 Factors that relate to treatment duration for patients with pala-
Mavragani M, Boe O E, Wisth P J, Selvig K A 2002 Changes in root length
tally impacted maxillary canines. American Journal of Orthodontics and
during orthodontic treatment: advantages for immature teeth. European
Dentofacial Orthopedics 119: 216–225
Journal of Orthodontics 24: 91–97
Taithongchai R, Sookkorn K, Killiany D M 1996 Facial and dentoalveolar
Mavreas D, Athanasiou A E 2008 Factors affecting the duration of ortho-
structure and the prediction of apical root shortening. American Journal
dontic treatment: a systematic review. European Journal of Orthodontics
of Orthodontics and Dentofacial Orthopedics 110: 296–302
30: 386–395
Viecilli R F, Katona T R, Chen J, Hartsfield J K Jr, Roberts W E Jr 2009
Mirabella A D, Årtun J 1995 Prevalence and severity of apical root resorp-
Orthodontic mechanotransduction and the role of the P2X7 receptor.
tion of maxillary anterior teeth in adult orthodontic patients. European
American Journal of Orthodontics and Dentofacial Orthopedics 135:
Journal of Orthodontics 17: 93–99
694.e1–694.e16
Pandis N, Nasika M, Polychronopoulou A, Eliades T 2008 External api- Weltman B, Vig K W, Fields H W, Shanker S, Kaizar E E 2010 Root resorp-
cal root resorption in patients treated with conventional and self- tion associated with orthodontic tooth movement: a systematic review.
ligating brackets. American Journal of Orthodontics and Dentofacial American Journal of Orthodontics and Dentofacial Orthopedics 137:
Orthopedics 134: 646–651 462–476
Sameshima G T, Asgarifar K O 2001 Assessment of root resorption and root Woloshyn H, Årtun J, Kennedy D B, Joondeph D R 1994 Pulpal and perio-
shape: periapical vs panoramic films. Angle Orthodontist 71: 185–189 dontal reactions to orthodontic alignment of palatally impacted canines.
Sameshima G T, Sinclair P M 2001 Predicting and preventing root resorp- Angle Orthodontist 64: 257–264
tion: part II. Treatment factors. American Journal of Orthodontics and Zuccatti G, Ghobadlu J, Nieri M, Clauser C 2006 Factors associated with
Dentofacial Orthopedics 119: 511–515 the duration of forced eruption of impacted maxillary canines: a ret-
Sameshima G T, Sinclair P M 2004 Characteristics of patients with severe rospective study. American Journal of Orthodontics and Dentofacial

Downloaded from by guest on January 29, 2015

root resorption. Orthodontics and Craniofacial Research 7: 108–114 Orthopedics 130: 349–356

View publication stats