European Journal of Orthodontics, 2017, 1–12

doi:10.1093/ejo/cjw077

Systematic Review

Open versus closed surgical exposure of palatally

impacted maxillary canines: comparison of the
different treatment outcomes—a systematic review
Dimitrios Sampaziotis1, Ioannis A. Tsolakis2, Elias Bitsanis1 and
Apostolos I. Tsolakis1
Department of Orthodontics, School of Dentistry and 2Department of Experimental Surgery and Surgical Research,
1

School of Medicine, National and Kapodistrian University of Athens, Greece

Correspondence to: Apostolos I. Tsolakis, Department of Orthodontics, School of Dentistry, University of Athens, Str. Thivon
2, Goudi, Athens 11527, Greece. E-mail: apostso@otenet.gr

Summary
Objective: This study aims to compare the effectiveness of two different canine exposure
techniques (open and closed) regarding periodontal outcomes, duration of surgical treatment
and canine’s eruption, patient’s inconvenience, aesthetics, and orthodontic treatment
complications.
Search methods: Electronic database searches of published and unpublished literature were
performed. The reference lists of eligible studies were hand searched for additional studies
Selection criteria: Randomized clinical trials (RCTs), quasi-randomized clinical trials (Q-RCTs) and
non-randomized trials of prospective and retrospective design with patients of any age that compared
group with palatally impacted canines treated by open exposure to a similar group treated by closed
exposure technique were selected. There was not any restriction in language or year of publication.
Data collection and analysis: Study selection, data extraction, and risk of bias assessment were
performed individually and in duplicate.
Results: Search strategy resulted in 159 articles and nine articles were selected for the final analysis.
They were three non-randomized trials, one Q-RCT, and two reports of another Q-RCT and three
reports of one RCT. The level of reported evidence was high for the RCT and one Q-RCT but poorer
for the other trials. Four articles reported periodontal outcomes, three searched the duration of
surgical procedure, two the duration of canine eruption, two investigated patient’s inconvenience,
two reported on failure rates and two addressed aesthetic outcomes. The results are inconsistent
and there is considerable disagreement for the majority of the outcomes among studies.
Conclusion: According to existing articles we may conclude that there is no difference between the two
techniques regarding the periodontal outcomes and aesthetic appearance. The surgical procedure is
shorter in the open exposure group and the amount of postoperative pain during the first day is similar
between the open and closed surgical exposure patients. However, these conclusions are based on
two single trials with high level of evidence, while the rest of the studies present high risk of bias.

Introduction
Maxillary canine’s impaction is a problem that the orthodontist
faces relatively often, as the maxillary canines are the second more
frequent impacted teeth, after the third molars and present preva-
lence of impaction which ranges from 1 per cent to 3 per cent (1).
Palatal impaction of canines is more frequent than labial and the

© The Author 2017. Published by Oxford University Press on behalf of the European Orthodontic Society. All rights reserved.
1
For permissions, please email: journals.permissions@oup.com
2 European Journal of Orthodontics, 2017

relevant ratio ranges from 2:1 to 9:1 (1).There is a debate among
clinicians whether the open or the closed surgical exposure, is the
favourable treatment of choice for palatally displaced canines. In
2013, two point/counterpoint debate articles depict this controversy
clearly. Mathews and Kokich (2) advocate for the open technique
followed by free eruption of the canine, while Becker and Chaushu
(3) suggest the closed traction method being the more appropriate.
To our knowledge, there are only two literature reviews, which
try to address the previous question. In 1999, Burden et al. (4) failed
to express a definitive answer regarding the approach that results in
better periodontal outcomes and less treatment time. It was reported
that the requirement of repeated surgery is maybe more common
with the closed eruption technique. More recently, Parkin et al. (5) in
2008 performed an intervention review of related articles with strict
inclusion criteria but they did not find any studies that met their
demands. They concluded that there is a need for randomized clini-
cal trials (RCTs) to compare the two treatment alternatives.
The aim of our study is to systematically review the up-to-date
scientific literature in order to define the most favourable technique,
for palatally impacted maxillary canines, in terms of treatment out-
comes. Periodontal health, aesthetics, patient’s inconvenience after
operation, duration of surgical procedure and canine’s eruption and
orthodontic treatment complications will be judged and compared
between the two treatment alternatives. Prisma 2009 checklist (sup-
plementary material) was followed for our study.
Methods
Protocol and registration
Not available.
Types of studies considered in the review
RCTs, quasi-randomized clinical trials (Q-RCTs) and non-rand-
omized trials, without any restriction in language and time of publi-
cation, were considered in our study.
Comparisons
Comparison of the outcomes between the two groups
Outcomes
Any of the following: periodontal health, aesthetics, patient’s incon-
venience after operation, duration of surgical procedure and canine’s
eruption and orthodontic treatment complications.
Search strategy for identification of studies
A literature search was carried out by applying the Medline data-
base (via PubMed), Embase (via Ovid), Scopus, Cochrane Oral
Health Group’s Trials Register, CENTRAL and Google Scholar
using the following Medical Subject Heading (MeSH) terms;
‘tooth, impacted’, ‘cuspid’, ‘cuspid/surgery’, ‘cuspid therapy’ and
other free-text terms. Unpublished literature was searched on
ClinicalTrials.gov, the National Research Register, and Pro-Quest
Dissertation Abstracts and Thesis database. Finally hand searching
was performed.
Search strategy for PubMed is presented on Table 1.
Two review authors (DS, IAT) performed the study selection
independently and in duplicate. Any inconsistencies were solved by
discussion with the supervisor (AIT). Potentially appropriate studies
were identified by title at first. After this stage, the abstracts were
read and the inappropriate studies were eliminated. The references
of the eligible studies were hand searched in order to identify any
possible additional articles, which were not found already. The final
choice was done when the full texts were examined, based on the
exclusion and inclusion criteria (Τable 2).
Data extraction
Data were extracted with reference to participants, interventions,
study duration/observational period, outcomes, methods of outcome
assessment, and results by two reviewers (DS, ΙAΤ) independently
and any inconsistencies were solved by discussion with a third
reviewer (IΒ).

Types of participants Dealing with missing data

Orthodontic patients of any age with palatally impacted maxillary
canines (one or both).
Types of interventions
Appropriate studies should include two groups of patients. One that
is treated with open exposure technique and another one treated
with closed exposure technique.
Authors were contacted via email to request information where
missing. In case of no response or no access of the missing data, only
the available data were reported and analysed.
Bias assessment
The risk of bias within included articles was assessed by ACROBAT-
NRSI tool of Cochrane (6) for non-randomized trials and by

Table 1. Search strategy

PubMed

Search number Search Items found

1 (((‘Tooth, Impacted’[Mesh]) AND ‘Cuspid’[Mesh]) AND (‘Cuspid/surgery’[Mesh] OR ‘Cuspid/therapy’[Mesh]))) 350

2 ((((((((((impact* AND ((maxilla* OR up*) AND (canines OR cuspids))) AND ((open OR closed) AND (exposure OR 68
technique OR approach OR eruption OR method OR management OR treatment))))))
3 (((impact* AND (canines OR cuspids) AND ((open OR closed) AND (exposure OR technique OR approach OR 151
eruption OR method OR management OR treatment)))))))
4 ((((impact* AND ((maxilla* OR up*) AND (canines OR cuspids))) AND (exposure OR technique OR approach OR 1067
eruption OR method OR management OR treatment))))))
5 (((canines OR cuspids) AND (open closed) AND (exposure OR technique OR approach OR eruption OR method OR 380
management OR treatment))))
6 ((impact* AND (canines OR cuspids) AND (exposure OR eruption)))) 771
7 (((canines OR cuspids)) AND palat* AND impact* 347

Last search on 5 October 2016.

D. Sampaziotis et al.
Cochrane handbook for systematic reviews (chapter 8) (7) for RCTs
and Q-RCTs, by two independent authors (DS, IAT) and any incon-
sistencies were solved by discussion with the supervisor (AIT).
Concerning the randomized and quasi-randomized trials, seven
domains of bias were estimated: sequence generation, allocation
concealment, blinding of participants and investigators, blinding
of outcome assessors, incomplete outcome data, selective outcome
reporting, and other sources of bias. A judgment of ‘low’, ‘high’, or
‘unclear’ risk of bias was made for each of the seven domains, while
a final overall judgment was assessed based on the following:
Low risk of bias if all key domains of the study were at low risk
of bias.
Unclear risk of bias if one or more key domains of the study were
unclear.
High risk of bias if one or more key domains were at high risk of
bias.
Regarding the non-randomized trials, bias due to confounding, bias
in selection of participants, bias in measurement of interventions,
bias due to departures from intended interventions, bias due to miss-
ing data, bias in measurement of outcomes, and bias in selection of
the reported result were assessed for the qualitative evaluation of the
study. Possible results for each domain and hence the overall evalu-
ation of each study: ‘low’, ‘moderate’, ‘serious’, ‘critical’ risk of bias
and ‘no information’.
For each separate outcome several confounders were taken into
account for the assessment of risk of bias. The initial position of
canine (depth of impaction, orientation and inclination) was evalu-
ated as a confounding factor for all outcomes. Age was also con-
sidered as a confounder for all outcomes except for duration of
surgical procedure. Additional confounders were oral hygiene and
initial periodontal condition of patient for the periodontal out-
comes and gender for postoperative pain perception. Bias due to
co-intervention was considered if different surgeons had performed
the operations.
Results
Our search strategy resulted in 159 articles, which were selected
according to their title. After abstract reading stage and elimination
of duplicates (43 papers) 14 articles were gathered and read in full-
text. Finally, nine articles (three reports of one RCT, one Q-RCT,
and two reports of another Q-RCT and three non-randomized tri-
als) were retrieved for the final analysis according to the inclusion/
exclusion criteria (Table 2). Table 3 summarizes the data of the nine
included articles and the procedure of selection is presented on the
flow diagram (Figure 1).
All three reports of the RCT (8–10) were from the study of
Parkin et al. but different outcomes were addressed in each article.
Table 2. Inclusion–exclusion criteria
Inclusion criteria
Studies that compare group of palatally impacted maxillary canines
treated by open technique with group treated by closed technique for any
of the outcomes mentioned.
RCTs, Q-RCTs, non-randomized trials (retrospective or prospective)
Studies in humans
Studies with patients of any age
Studies of any language
3
In the first one (8), mean surgical duration, mean postopera-
tive patient’s discomfort and failure traction rates are compared
between two exposure techniques. In the second article (9), peri-
odontal outcomes are searched, while in the third article (10) the
aesthetic appearance of previously impacted canine is judged.
The two articles of Smailiene et al. (11, 12) are actually the same
study (a quasi-randomized trial, Q-RCT) that was published
in a Lithuanian journal (11) as well as in the European Journal
of Orthodontics (EJO) (12). Nevertheless, we have considered
them also as two separate articles because different outcomes are
reported in each publication. In the former article (11), periodontal
outcomes and aesthetic appearance of previously impacted canines
are compared between the two techniques, whereas in the article
of EJO (12) mean duration for eruption of canine and periodontal
outcomes is mentioned.
The level of evidence was high for the RCT (8–10) (all articles
presented low risk of bias except for the outcomes mean surgical
duration and postoperative pain which were judged to be in high
risk of bias in the first article that was published in 2012 (8).
Regarding the Q-RCTs, the study of Gharaibeh et al. (13) had
low risk of bias but the study of Smailiene et al. (11, 12) was in high
risk of bias except for the outcome mean duration for eruption of
canine which was in low risk of bias according to our bias assess-
ment method (risk of bias for RCTs and Q-RCTs is presented on
Table 4). Quasi-randomized trials do not follow the protocol of suc-
cessful randomization that is used in RCTs (e.g. computer generated
random numbers, dice throwing etc), but they use fallible methods
of randomization, like allocation according to date of patient’s birth
or random sequence generated by odd (e.g. Smailiene et al. (12):
‘in every second patient the open technique was used’) instead. The
study of Gharaibeh et al. (13) was considered as a Q-RCT because
the authors did not mention any method of concealed randomization.
The lack of successful randomization and of concealed allocation of
participants affects the overall risk of bias of a Q-RCT as it could
influence the allocation of characteristics at baseline that could affect
the result (like the significantly different initial periodontal condi-
tion between two groups in the study of Smailiene et al. (11, 12)
that surely could affect the periodontal outcomes). However, in
the study of Gharaibeh et al. (13) despite the fact that it is unclear
whether the randomization was computerized (or performed by
another totally random way) or not and although there is uncon-
cealed allocation of participants, the distribution of baseline charac-
teristics that could possibly affect the outcomes—age, gender, need
for bone removal—are similar among groups. Therefore, the study
was considered to be in low risk of bias overall (random sequence
generation and allocation concealment were not considered as key
bias domains in this study).
Regarding the non-randomized studies (14–16), all were in seri-
ous risk of bias (Table 5).
Exclusion criteria
Studies that also include teeth other than canines and do not address each
group of teeth separately
Studies that include labially impacted maxillary canines or mandibular
impacted canines
Case reports, case series, reviews, authors’ opinion
Clefts, syndromes
 4

Table 3. Data extraction table

Study duration\ Methods of outcome

Author (year) Study design Participants Interventions observational period Outcomes assessment Results Conclusions

Wisth (1976) RCS O: 34 patients’ sex: O: radical (open) 18 months for open group and O1: Mean pocket O1: measurement of O1: significantly Closed technique has
19 F, 15 M age: exposure, 22 months for closed group depth pocket depth deeper in C group better periodontal
15 years 9 months. orthodontic traction O2: Mean loss of O2: Measurement of O2: Significantly results but the differ-
C: 22 patients after 1 week attachment loss of attachment more palatally in ence is small compar-
mean age: 16 years C: moderate (closed) O3: Mean bone O3: Radiological O group ing to open exposure.
2 months. exposure, immediate support evaluation of distance O3: No differences
traction. between CEJ & bone
Bonding during margin (bone support)
surgery for both
techniques
Pearson (1997) RCS O: 52 patients O: open exposure, O: 31.7 months (on average) O1: Mean surgical O1:Measurement of O1: 12 min (9–22) Failure rates greater
Sex: 20 M, 32 F bonding and start of C: 24.8 months (on average) duration surgical and whole (O)/ 36 min (27–43) for closed technique,
Age: 13.9 (10.5–24) traction 6–9 months O2: Failure rate treatment duration (C) mainly due to frac-
C: 52 patients after surgery (from start of traction O2: Two times tured ligature wire
Sex: 32 M, 20 F C: closed exposure, until debond) greater for which therefore is not
Age: 14.5 traction 3 months O2: Evaluation of failure C group a reliable method.
(11.7–24.2) after surgery with a rates (failure of Shorter surgical dura-
ligature wire, bonding exposure, bonding or tion in open technique
during surgery eruption, extraction rates)
Iramaneerat (1998) RCS O: 25 patients O: Open exposure, On average 28.8 months Mean duration of Measurement of Non-significantly The duration of treat-
Sex: 17 F, 8 M, age: bracketing after (until debond) canine’s eruption duration for canine’s different between ment does not differ
13.9 years 10–14 days (until its well eruption two groups between closed and
(13.5–14.3) C: closed exposure, positioning in the open technique
C: 25 patients immediate bonding line of the arch)
Sex: 17 F, 8 M, age: of attachment
13.5 years (13–14) Not mentioned
clearly when traction
of cuspid starts for
either technique
Gharaibeh (2007) Q-RCT O: 16 patients sex: O: open exposure, Follow-up 7 days after O1: Postoperative O1: Subjective O1: Similar amount of Closed technique
14 F, 2 M bracketing and ini- operation pain, recovery’s perception of pain pain, but provides with faster
Age: 17.3 tiation of traction 1 duration categorized as severe, quicker recovery in recovery whereas
years ± 4.5 week after surgery O2: Mean surgical moderate, mild C group open procedure
C: 16 patients C: closed exposure, duration O2: Measurement of surgi- O2: Shorter for O requires less time for
Sex: 14 F, 2 M initiation of traction cal duration group operation
Age: 17.6 1 week after surgery
years ± 2.4
European Journal of Orthodontics, 2017
Table 3. Continued

Study duration\ Methods of outcome

Author (year) Study design Participants Interventions observational period Outcomes assessment Results Conclusions

Smailiene (2013a) Q-RCT 43 patients (35 F, 8 O: Open exposure, The final examination was done O1: Mean pocket O1: Measurement of O1, O2, O3: Post-treatment status
M) age: 15.81 free eruption on average 4.19 months after depth pocket depth Non-significantly did not differ between
years SD: 3.04) C: Closed exposure, debond O2: Mean bone O2: Radiological different two groups
D. Sampaziotis et al.

O: 22 patients initiation of traction support evaluation of distance O4: Non-significantly

(age at the time of 1 week after surgery O3: Mean gingival between CEJ & bone different
examination: recession margin
18.6 years (SD 3.45) O4: Aesthetic (bone support)
C: 21 patients (age at appearance O3: Measurement of
the time of gingival recession
examination: 19.7 O4: Visual assessment of
(SD: 4.37) canine’s appearance
(tooth position, inclina-
tion, colour, shape)
Smailiene (2013b) Q-RCT 43 patients (35 F, Ο: Open exposure. Follow-up on average O1: Mean pocket O1: Measurement of O1, O2, O3, O4: Periodontal status
8 M) Free eruption. 4.19 months (±1.44) after depth pocket depth Non-significantly was similar for two
O: 22 patients Surgery done debond O2: Mean gingival O3: Measurement of different groups.
Age: 15.46 ± 3.28 (at 1.55 months before recession gingival recession O5: Significantly Time for canine’s
the time of operation) the placement of fixed O3: Mean width of O3: Measurement of width shorter for O group extrusion shorter in
C: 21 patients appliances keratinized tissues of keratinized tissues open group
Age: 16.15 ± 2.79 (at C: Closed exposure, O4: Mean bone O4: Radiological evalu-
the time of operation) initiation of trac- support ation
tion 1 week after O5: Mean duration O5: Assessment of dura-
surgery. Surgery done for eruption of tion from surgery until
3.67 months after canine a bracket can be bonded
bonding of fixed ap- on the middle of canine’s
pliances labial surface
Parkin (2012) RCT O: 40 patients O: Open exposure, Follow-up 10 days after O1: Mean surgical O1: Evaluation of surgical O1: Non- No difference in
Sex: 27 F, 13 M surgical pack place- operation duration time significantly operating time and
Age: 14.3 years ment for 10 days, O2: Amount & O2: Questionnaire for different patient’s postoperative
(SD 1.3) bracketing and trac- duration of assessment of O2: discomfort between
Finally 35 included tion when adequate postoperative pain, postoperative Non-significantly two groups
in analysis amount of enamel difficulty in function discomfort different
C: 41 patients was present O3: failure rates O3: Evaluation of O3: 9.6% in
Sex:25 F, 16 M C: Closed exposure, required re-exposures O group and 2.9%
Age: 14.1 years (SD possibly immediate in C group
1.6) traction
Finally 36 included
in analysis
5
 6

Table 3. Continued

Study duration\ Methods of outcome

Author (year) Study design Participants Interventions observational period Outcomes assessment Results Conclusions

Parkin (2013) RCT O: 40 patients O: Open exposure, Examination 3 months after O1: Level of O1, O2, O3: Clinical O1, O2, O3, O4: There are not
Sex: 27 F, 13 M surgical; pack debond attachment periodontal Non-significantly any differences in
Age: 14.3 years placement for O2: Gingival examination different between periodontal health for
(SD 1.3) 10 days, bracketing recession O4: Radiographic evalua- two groups. canines treated with
Finally 33 patients and traction when O3: Crown height tion of bone levels open technique and
included in analysis adequate amount of O4: bone support for those treated with
C: 41 patients enamel was present O1, O2, O3, O4: closed technique
Sex: 25 F, 16 M C: Closed exposure, Mean difference of
Age: 14.1 years possibly immediate previously impacted
(SD 1.6) traction canine and its normal
Finally 29 patients contralateral
included in analysis
Parkin (2015) RCT O: 40 patients O: Open exposure, Aesthetic judgment Aesthetic appear- Judgment of aesthetic Similar aesthetic Non-significant differ-
Sex: 27 F, 13 M surgical; pack 3 months after debond ance of previously appearance of previously appearance between ences in aesthetic ap-
Age: 14.3 years placement for impacted canines impacted canines by lay- canines of two pearance of previously
(SD 1.3) 10 days, bracketing people and orthodontists groups, similar impacted canines
Finally 34 patients and traction when frequency of correct between open and
included in analysis adequate amount of identification of closed groups
C: 41 patients enamel was present previously impacted
Sex: 25 F, 16 M C: Closed exposure, canine between two
Age: 14.1 years possibly immediate groups
(SD 1.6) traction
Finally 33 included
in analysis

RCS, retrospective cohort study; PCS, prospective cohort study; Q-RCT, quasi-randomized clinical trial; RCT, randomized clinical trial; F, female; M, male; O, open technique; C, closed technique; O1, first outcome; O2,
second outcome; O3, third outcome; O4, fourth outcome; O5, fifth outcome; O6, sixth outcome.
European Journal of Orthodontics, 2017
D. Sampaziotis et al.
Clinical heterogeneity among studies (different outcome assess-
ment, variable age of patients and different follow-up duration), and
the high risk of bias in general precluded the quantitative synthesis
of results in meta-analysis.
Periodontal outcome
Four articles report on periodontal outcomes (9, 11, 12, 14). The
retrospective study of Wisth et al. (14) concludes that closed expo-
sure technique provides with better periodontal results as loss of
attachment was found to be greater for the canines treated with open
technique especially on their palatal side. On the other hand, the
results from the two articles of Smailiene et al. (11, 12) show that
there is no statistical difference for periodontal outcomes in terms
of mean pocket depth, gingival recession, bone support, and width
of keratinized gingiva between closed and open exposure technique.
Similarly, the first report of the RCT of Parkin et al. (9) did not
find different periodontal status after treatment for the previously
impacted canines between two techniques. More specifically, level of
attachment, crown height, bone support and gingival recession were
investigated in terms of comparisons of mean differences between
previously impacted canines and their normal contralaterals for
closed and open eruption technique. Τhe statistical tests of Parkin’s
et al. work, proved that there is no statistical difference between
mean differences in open and those in closed exposure technique.
Although there are conclusions for periodontal outcomes in the
aforementioned articles (9, 11, 12, 14) the three of them (11, 12, 14)
presented high risk of bias while only the article of Parkin et al. (9)
was considered to be in low risk of bias regarding this specific
outcome (post-treatment periodontal condition).
Duration of surgical procedure
Regarding the duration of surgical procedure, two studies (13, 15)
found significantly less time in the operation room for the open tech-
nique while Parkin et al. (8) report that there is not any statistically
significant difference between the two treatment alternatives. Two
(8, 15) out of three articles (8, 13, 15) that report on duration of
surgical procedure present high risk of bias, whereas only one study
(13) presents low risk of bias for this outcome.
Duration of canine’s eruption
Iramaneerat et al. (16) found that the time needed for canine’s erup-
tion, more specifically the duration from the surgical exposure of the
canine until it was well positioned in the line of the arch does not dif-
fer between the two exposure techniques. Iramaneerat et al. (16) had
considered that proper positioning of the canine was attained when
a 0.018 inch steel archwire, or equivalent, was fully engaged on the
canine. On the other hand, Smailiene et al. (12) reported that the
eruption of impacted canine was quicker for the group treated with
open technique. These investigators assessed the duration from sur-
gery until a bracket can be bonded on the middle of canine’s labial
surface. The risk of bias for the duration of canine’s eruption was
judged to be high for the study of Iramaneerat et al. (16), while in
the article of Smailiene et al. (12) was low.
Orthodontic treatment complications
Two articles (8, 15) estimated the orthodontic treatment complica-
tions between the two techniques and presented inconsistent results.
Pearson et al. (15) found over twofold failure rates for the group
treated with closed exposure technique in comparison with the group
that had undergone the open exposure technique. Parkin et al. (8) on
7
the other hand, reported that 9.6 per cent of the patients treated
with open technique required re-exposure of the canine, while the
corresponding rate for the closed technique was only 2.9 per cent.
However, there was not any statistical test to prove any significant
difference. Risk of bias was judged to be low in the article of Parkin
et al. (8), while it was high in the study of Pearson et al. (15) regard-
ing the specific outcome (failure rates).
Pain perception
Perception of pain after surgical exposure of canines is investigated
in two articles (8, 13) which reported that there is no difference in
the amount of pain between closed and open technique. A moderate
degree of discomfort was observed after the procedure, which disap-
peared few days later. However, risk of bias was high in the article of
Parkin et al. (8) for this outcome. The study of Gharaibeh et al. (13)
presented low risk of bias.
Aesthetic outcomes
Two papers (10, 11) agree that there is no difference in the aesthetic
outcome between the two different techniques. Only one of them
(10) presents low risk of bias whereas the other (11) was judged to
be in high risk of bias for this outcome. More specifically, Smailiene
et al. (11) assessed the inclination, the shape and colour of treated
canines between two groups and did not find any differences. In the
work of Parkin et al. (10) on the other hand, blinded laypeople and
orthodontists were recruited to judge the aesthetic appearance of
the maxillary canines in all participants. The one upper canine was
treated either with closed or open exposure technique and its con-
tralateral was physiologically erupted. The authors report that the
frequency of correct identification of the operated canine and the fre-
quency that the previously impacted tooth was considered aestheti-
cally better than its contralateral did not differ significantly between
closed and open exposure technique groups either for the panel of
orthodontists or for that of laypeople.
Discussion
Different outcomes were searched among studies, the quality of
which was also variable, according to risk of bias assessment.
The three articles of Parkin et al. (8–10), which are parts of a
single RCT, provide generally high level of evidence. Based on the
Cochrane tool for bias assessment for RCTs (7), risk of bias is low
for the periodontal outcomes, aesthetic appearance, and failure rates
while postoperative pain perception and duration of surgical pro-
cedure, present high risk of bias according to our evaluation. Two
of the participants in the aforementioned study (Parkin et al. (8))
who had been exposed to open canine exposure were also undergone
first molar extraction and frenectomy at the same time. In addition,
in the same article (8) 27 cases of simultaneous premolar extrac-
tion were included and it is not clarified whether their distribution
between the two groups was similar. Therefore, the conclusion that
open and closed exposure technique does not differ in terms of post-
operative pain and duration of surgical procedure is not a reliable
one. It is obvious though that the closed technique requires more
time in the operation room as the procedures of attachment bond-
ing and wound suturing add in the total operation time. However,
the extra appointment needed for attachment bonding in the case of
open exposure technique must be taken into consideration.
The quasi-randomized trial of Gharaibeh et al. (13), presents
low risk of bias for both reported outcomes (duration of surgical
8 European Journal of Orthodontics, 2017
Figure 1. Flow diagram illustrating the identification and selection of studies used in this review.
procedure and postoperative pain). Although the investigators do not
provide concealed allocation of participants, they state that the need
for bone removal, age and gender distribution was similar between
the two groups and the surgeon was the same for all operations. For
this reason the statements that the surgical procedure is shorter in
the open exposure group and that the amount of postoperative pain
during the first postoperative day is similar between the two groups
are considered conclusions with high level of evidence. Concerning
the progressive disappearance of pain Gharaibeh et al. (13) found
faster recovery after closed exposure technique while Parkin et al.
(8) did not find any difference. The study of Chaushu et al. (17) was
excluded from our review because it included impacted incisors and
canines without providing separate results for canines. However, it
was reported that the disappearance of pain was faster following the
closed technique. This is in agreement with the study of Gharaibeh
(13). Healing by secondary intention in the open exposure technique
may play a significant role here. The inconsistency with the results
of Parkin et al. (8) might be due to the fact that the prementioned
investigators used an ‘extensive’ surgical technique and state that
they remove bone to expose the largest diameter of the impacted
canine crown.
The articles of Smailiene et al. (11, 12) report on periodontal
outcomes (pocket depth, loss of attachment, bone support, width
of keratinized gingiva), aesthetic appearance (inclination, position,
colour, and shape of the treated canine) and the duration for canine’s
eruption. Regarding the periodontal outcomes, it was found that
they were similar between closed and open exposure techniques, but
the study presents high risk of bias as there is no report for the initial
periodontal condition at baseline for the two groups. High risk of
bias is also present in considering aesthetic appearance, because it
is not mentioned whether the assessors who evaluated the inclina-
tion, the position, the colour and shape of the canine were blinded.
Despite the unconcealed allocation, the outcome duration of canine’s
eruption is probably in low risk of bias as the protocol of open and
closed technique is mentioned clearly, any possible confounders (age,
depth of impaction) have similar distribution at baseline among
groups and the definition of eruption is objective. However, these
authors state that although non-significant, the difference between
initial depth of impaction may have affected this result.
All three non-randomized trials (14–16) are in serious risk of bias
for all reported outcomes. Being more specific, the study of Wisth
et al. (14) which states that the closed exposure technique results in
more favourable periodontal outcomes (reduced loss of attachment),
is a retrospective trial which presents high risk of bias in selection of
participants and bias due to confounding (initial periodontal condi-
tion), bias due to departures from the intended intervention (possibly
different surgeons) and bias in measurement of the outcomes (not
blinded assessors). The retrospective study of Pearson et al. (15) was
considered with high risk of bias because it presented high risk of
selection bias, different depth of impaction and different surgeons
Table 4. Risk of bias assessment for RCTs, Q-RCTs

Random sequence Allocation

Author (year) Outcomes generation concealment Performance bias Detection bias Attrition bias Selective reporting Other Overall

Gharaibeh O1: Postoperative Unclear for all High for all Low for all Low for O1 (al- Low for all Low for all Low for all outcomes Low for all outcomes
(2007) pain outcomes (‘…ran- outcomes (not outcomes (same though subjective outcomes outcomes (all the (no possible risk of (not concealed
O2: Mean surgical domly selected’) not mentioned surgeon) there is the same (no missing prespecified bias from other allocation but baseline
D. Sampaziotis et al.

duration possible to conclude concealment of questionnaire for outcome data) variables are sources) characteristics that
if randomization was allocation, probably both groups) measured) could possibly affect
successful not done) Low for O2 the outcomes—age,
(objective outcome) gender, need for bone
removal—similar
among groups)
Smailiene O1: Mean pocket High for all High for all Low for all High for Low for all out- Low for all Low for all outcomes High for all outcomes
(2013a) depth outcomes (‘…every outcomes (not outcomes (same O1, O3, O4 comes outcomes (all the (no possible risk of (not blinded assessors
O2: Mean bone second patient’) not mentioned surgeon) (subjective (no missing prespecified bias from other and initial periodontal
support successful concealment, outcomes—not outcome data) variables are sources) condition not
O3: Mean gingival randomization probably not done) blinded examiners) measured) mentioned—not
recession Low for O2 (blinded successful
O4: Aesthetic examiners) randomization)
appearance of
previously impacted
canine
Smailiene O1: Mean pocket High for all High for all Low for all Low for O1, O2, O3, Low for all out- Low for all Low for all outcomes High for O1, O2, O3, O4
(2013b) depth, O2: mean outcomes (‘…every outcomes (not outcomes (same O4 (blinded comes outcomes (all the (no possible risk of (not successful
gingival recession second patient’) not mentioned surgeon) assessors) (no missing out- prespecified bias from other randomization could
O3: Mean width of successful concealment, Low for O5 come data) variables are sources) have affected the
keratinized tissues randomization probably not done) (objective) measured) outcomes as initial
O4: Mean bone periodontal condition
support is not mentioned)
O5: Mean duration Low for O5
for eruption of
canine
Parkin (2012) O1: Mean surgical Low for all out- Low for all Low for all out- Low for O1, O3 Unclear for Low for all High for O1, O2 High for O1, O2 Low
duration comes (‘…computer outcomes (‘sealed comes (all surgeons (objective) low O1 (outcomes outcomes (all the (frenectomy, premo- for O3
O2: Postoperative generated random opaque envelopes’) had adequate for O2 (subjective from 12 cases prespecified lar’s and first molar’s
pain numbers’) successful successful experience using outcome but same missing—imbal- variables are extraction performed
O3: Failure rates randomization concealment of the two questionnaire for anced number measured) simultaneously with
allocation techniques) both groups) across groups); canine’s exposure in
Low for O2, O3 27 patients); Low for
(missing outcome O3 (no possible risk
data balanced in of bias from other
number across sources)
groups—reasons
not related to
outcome)
9
 10

Table 4. Continued

Random sequence Allocation

Author (year) Outcomes generation concealment Performance bias Detection bias Attrition bias Selective reporting Other Overall

Parkin (2013) O1: Level of Low for all out- Low for all out- Low for all out- Low for all Low for all Low for all Low for all Low for all outcomes
attachment comes (‘…computer comes comes (all surgeons outcomes (masked outcomes (ITT outcomes (all the outcomes (no
O2: Gingival generated random (‘sealed opaque had adequate examiners) analysis, reasons prespecified possible risk of bias
recession numbers’) successful envelopes’) experience using for missing data variables are from other sources)
O3: Crown height randomization successful the two techniques) not related to measured)
O4: Bone support concealment of outcome, level of
O1, O2, O3, O4: allocation dropout was
Mean difference of accounted for in
previously impacted the analysis)
canine and its
normal
contralateral
Parkin (2015) Aesthetic Low (‘…computer Low (‘sealed Low (all surgeons Low (subjective Low (missing Low (all the Low for all Low
appearance of generated random opaque envelopes’) had adequate outcome but outcome data prespecified outcomes (no
previously numbers’) successful successful experience using blinded assessors) balanced in variables are possible risk of bias
impacted canines randomization concealment of the two techniques) number across measured) from other sources)
allocation groups—reasons
not related to
outcome)

O1, first outcome; O2, second outcome; O3, third outcome; O4, fourth outcome; O5, fifth outcome; O6, sixth outcome; Q-RCT, quasi-randomized clinical trial; RCT, randomized clinical trial.
European Journal of Orthodontics, 2017
Table 5. Risk of bias assessment for non-randomized trials

Bias due to
D. Sampaziotis et al.

Bias in selection Bias in departures from

Bias due to of participants measurement intended Bias due to missing Bias in measurement Bias in selection of
Author (year) Outcomes confounding into the study of interventions interventions data of outcomes the reported result Overall bias

Wisth (1976) O1: Mean pocket Serious for all Serious for all Serious for all Serious for all out- Serious for all out- Serious for all out- Low for all Serious for
depth outcomes (at least outcomes (selection outcomes comes (there are comes (missing comes (the outcome outcomes (all all outcomes
O2: Mean loss of one critically into the study was (intervention possible data—baseline measure was reported results cor- (the study
attachment important domain related to status not well co-interventions) characteristics; the subjective and was respond to is judged to
O3: Mean bone not appropriately intervention and defined) risk of bias cannot be assessed by intended outcomes) be in serious
support measured or not possibly to out- removed through outcome asses- risk of bias in
O1, O2, O3: For adjusted for) come) appropriate analysis) sors aware of the at least one
previously im- intervention received domain)
pacted canine by study participants,
methods of outcome
assessment not
comparable across
groups)
Pearson (1997) O1: Mean surgical Serious for all Serious for all Serious for all Serious for all Serious for all out- Low for O1, O2 (ob- Low for all outcomes Serious for
duration, O2: outcomes (at least outcomes outcomes outcomes (there comes (missing jective method (all reported results all outcomes
Failure rate one critically (selection into (intervention are possible data—baseline of outcome assess- correspond to in- (the study
important domain the study was status not well co-interventions) characteristics; the ment, comparable tended outcomes) is judged to
not appropriately related to defined) risk of bias cannot be across intervention be in serious
measured or not intervention and removed through ap- groups; any error is risk of bias in
adjusted for) possibly to propriate analysis) unrelated to interven- at least one
outcome) tion status) domain)
Iramaneerat Mean Low Low (selection into Serious Serious (there are Low (data were reason- Serious (subjec- Low (all reported Serious (the
(1998) duration of (no confounding the study (intervention possible ably complete) tive outcomes—the results correspond to study is
canine’s eruption expected) was unrelated to status not well co-interventions) methods of outcome intended outcomes) judged to be
this outcome) defined) assessment were not in serious
comparable across risk of bias in
groups) at least one
domain)

O1, first outcome; O2, second outcome; O3, third outcome; O4, fourth outcome; O5, fifth outcome; O6, sixth outcome.
11
12
performing the exposure operations. Apart from that, the over two-
fold failure rate for the closed technique which was reported by the
investigators is exclusively due to the fractured wires used for the
orthodontic traction in the closed exposure group and is not cor-
related with the type of surgical technique. Finally, in the study of
Iramaneerat et al. (16) it is stated that there is not any difference
between open and closed exposure technique regarding the dura-
tion for eruption of the canine. However, the time in which the fixed
appliances were placed, the time that the traction was started and
the type of used biomechanics are not mentioned. In addition, the
definition of eruption is problematic, as the canine was considered
fully erupted when a certain archwire was engaged in the bracket of
the tooth.
Conclusion
Only a few studies provide high level of evidence whether open or
closed exposure technique is more favourable treatment of choice in
terms of the post-treatment periodontal status, duration of surgical
procedure, duration for canine’s eruption, failure rates, postopera-
tive pain perception and aesthetic appearance. We used the GRADE
approach in order to interpret the results of this review.
The RCT of Parkin et al. (8–10) is a well-conducted study and pro-
vides the best evidence regarding periodontal outcomes and aesthetic
appearance, which do not present any difference for the two exposure
techniques. The study of Gharaibeh et al. (13) presents also low risk
of bias for the outcomes regarding duration of surgical treatment and
postoperative pain. According to the prementioned article the surgi-
cal procedure is shorter in the open exposure group and the amount
of postoperative pain during the first day is similar between the two
groups. However, all patients have faster recovery from postoperative
pain after the closed exposure technique. Overall no strong conclusion
for a specific outcome can rely on a single trial. The rest of the included
studies present high risk of bias and thus fail to provide safe results.
To conclude, more RCTs should be performed in the future for
stronger evidence and clarification of some controversial issues.
Supplementary material
Supplementary material is available at European Journal of
Orthodontics online.
Funding
No funding was obtained for this Systematic Review.
Conflict of interest
None to declare.
References
1. Stewart, J. A., Heo, G., Glover, K. E., Williamson, P. C., Lam, E. W. and
Major, P. W. (2001) Factors that relate to treatment duration for patients
with palatally impacted maxillary canines. American Journal of Ortho-
dontics and Dentofacial Orthopedics, 119, 216–225.
European Journal of Orthodontics, 2017
2. Mathews, D. P. and Kokich, V. G. (2013) Palatally impacted canines: the
case for preorthodontic uncovering and autonomous eruption. American
Journal of Orthodontics and Dentofacial Orthopedics, 143, 450–458.
3. Becker, A. and Chaushu, S. (2013) Palatally impacted canines: the case for
closed surgical exposure and immediate orthodontic traction. American
Journal of Orthodontics and Dentofacial Orthopedics, 143, 451–459.
4. Burden, D. J., Mullally, B. H. and Robinson, S. N. (1999) Palatally ectopic
canines: closed eruption versus open eruption. American Journal of
Orthodontics and Dentofacial Orthopedics, 115, 640–644.
5. Parkin, N., Benson, P. E., Thind, B. and Shah, A. (2008) Open ver-
sus closed surgical exposure of canine teeth that are displaced in the
roof of the mouth. The Cochrane Database of Systematic Reviews, 8,
CD006966.
6. Sterne, J. A. C., Higgins, J. P. T. and Reeves, B. C.; on behalf of the Devel-
opment Group for ACROBAT-NRSI. (2014) A Cochrane Risk of Bias
Assessment Tool: for NonRandomized Studies of Interventions (ACRO-
BAT-NRSI), Version 1.0.0 [Internet]. http:// www. riskofbias. info (2 Feb-
ruary 2016, date last accessed).
7. Higgins, J. P. T. and Green, S. (eds). (2011) Cochrane Handbook for Sys-
tematic Reviews of Interventions Version 5.1.0 [Internet]. The Cochrane
Collaboration, London, UK. http://handbook.cochrane.org (2 February
2016, date last accessed).
8. Parkin, N. A., Deery, C., Smith, A. M., Tinsley, D., Sandler, J. and Benson,
P. E. (2012) No difference in surgical outcomes between open and closed
exposure of palatally displaced maxillary canines. Journal of Oral and
Maxillofacial Surgery, 70, 2026–34
9. Parkin, N. A., Milner, R. S., Deery, C., Tinsley, D., Smith, A. M., Germain,
P., Freeman, J. V., Bell, S. J. and Benson, P. E. (2013) Periodontal health of
palatally displaced canines treated with open or closed surgical technique:
a multicenter, randomized controlled trial. American Journal of Ortho-
dontics and Dentofacial Orthopedics, 144, 176–184.
10. Parkin, N. A., Freeman, J. V., Deery, C. and Benson, P. E. (2015) Esthetic
judgments of palatally displaced canines 3 months postdebond after surgi-
cal exposure with either a closed or an open technique. American Journal
of Orthodontics and Dentofacial Orthopedics, 147, 173–181.
11. Smailienė, D., Kavaliauskienė, A. and Pacauskienė, I. (2013) Posttreatment
status of palatally impacted maxillary canines treated applying 2 different
surgical-orthodontic methods. Medicina (Kaunas), 49, 354–360.
12. Smailiene, D., Kavaliauskiene, A., Pacauskiene, I., Zasciurinskiene, E.
and Bjerklin, K. (2013) Palatally impacted maxillary canines: choice of
surgical-orthodontic treatment method does not influence post-treatment
periodontal status. A controlled prospective study. European Journal of
Orthodontics, 35, 803–810.
13. Gharaibeh, T. M. and Al-Nimri, K. S. (2008) Postoperative pain after sur-
gical exposure of palatally impacted canines: closed-eruption versus open-
eruption, a prospective randomized study. Oral Surgery, Oral Medicine,
Oral Pathology, Oral Radiology, and Endodontics, 106, 339–342.
14. Wisth, P. J., Norderval, K. and Bøoe, O. E. (1976) Comparison of two
surgical methods in combined surgical-orthodontic correction of impacted
maxillary canines. Acta Odontologica Scandinavica, 34, 53–57.
15. Pearson, M. H., Robinson, S. N., Reed, R., Birnie, D. J. and Zaki, G. A.
(1997) Management of palatally impacted canines: the findings of a col-
laborative study. European Journal of Orthodontics, 19, 511–515.
16. Iramaneerat, S., Cunningham, S. J. and Horrocks, E. N. (1998) The effect
of two alternative methods of canine exposure upon subsequent duration
of orthodontic treatment. International Journal of Paediatric Dentistry, 8,
123–129.
17. Chaushu, S., Becker A., Zeltser, R., Branski, S., Vasker, N. and Chaushu, G.
(2005) Patients perception of recovery after exposure of impacted teeth:
a comparison of closed- versus open-eruption techniques. Journal of Oral
and Maxillofacial Surgery, 63, 323–329.