Fixed Retainers

SYSTEMATIC REVIEW
The effects of fixed orthodontic

retainers on periodontal health:
A systematic review
Marie-Laure Arn,a Konstantinos Dritsas,a Nikolaos Pandis,a and Dimitrios Kloukosa,b
Bern, Switzerland, and Athens, Greece
Introduction: The objective of this systematic review was to assess the available evidence in the literature for
the effects of fixed orthodontic retainers on periodontal health. Methods: The following databases were
searched up to August 31, 2019: Medline, EMBASE, the Cochrane Oral Health Group's Trials Register,
CENTRAL, ClinicalTrials.gov, the National Research Register, and Pro-Quest Dissertation Abstracts and
Thesis database. Randomized controlled trials (RCTs), controlled clinical trials, cohort studies of prospective
and retrospective design, and cross-sectional studies reporting on periodontal measurements of patients who
received fixed retention after orthodontic treatment were eligible for inclusion. The quality of the included
RCTs was assessed per the revised Cochrane risk of bias tool for randomized trials (RoB 2.0), whereas the
risk of bias of the included cohort studies was assessed using the Risk Of Bias In Nonrandomized Studies of
Interventions tool. A modified version of the Newcastle-Ottawa scale was used for cross-sectional studies.
Results: Eleven RCTs, 4 prospective cohort studies, 1 retrospective cohort study, and 13 cross-sectional
studies fulfilled the inclusion criteria. The quality of evidence was low for most of the included studies. In
contrast to the general consensus, 2 RCTs, 1 prospective cohort study, and 2 cross-sectional studies
reported poorer periodontal conditions in the presence of a fixed retainer. The results of the included studies
comparing different types of fixed retainers were heterogeneous. Conclusions: According to the currently avail-
able literature, orthodontic fixed retainers seem to be a retention strategy rather compatible with periodontal
health, or at least not related to severe detrimental effects on the periodontium. (Am J Orthod Dentofacial
Orthop 2020;157:156-64)
R
etention after treatment remains 1 of the greatest extent.4 There is no consensus for the ideal duration of
challenges in orthodontics. The etiology of retention2; however, the first 8-month posttreatment
orthodontic relapse is complex and multifacto- period, when the remodeling of the periodontal fibers
rial; identified related factors include tension from the occurs, appears to be critical.5 Most clinicians choose a
periodontal fibers, the final occlusion, pressure from retention period longer than 8 months and often
the soft tissues, growth, and age advancement.1,2 The recommend life-long retention for all patients.6,7
risk of relapse is unpredictable3 and concerns a high Fixed and removable retainers continue to be the
proportion of postorthodontic patients. There is most common retention methods. Adjunctive proced-
evidence that during the postretention period, ures like pericision (or circumferential supracrestal
70%-90% of the cases show some relapse in the lower fiberotomy), a surgical technique aimed at cutting the
arch; the upper arch is also affected but to a lesser periodontal fibers around the neck of the teeth, and
interproximal enamel reduction (or reproximation)
a
Department of Orthodontics and Dentofacial Orthopedics, School of Dentistry, have been proposed, but are rarely implemented.6,8
University of Bern, Bern, Switzerland.
b
Over the last 50 years, several kinds of bonded retainers
Department of Orthodontics and Dentofacial Orthopedics, 251 Hellenic Air
Force General Hospital, Athens, Greece.
of varying sizes, cross-section forms, and materials have
All authors have completed and submitted the ICMJE Form for Disclosure of been suggested.9 The first nonbanded fixed retainer
Potential Conflicts of Interest, and none were reported. appeared in the 1970s and consisted of a thick, round
Address correspondence to: Dimitrios Kloukos, Department of Orthodontics and
Dentofacial Orthopedics, School of Dentistry, University of Bern, Freiburgstrasse
stainless steel wire that was bonded only to the
7 Bern, 3010, Switzerland; e-mail, dimitrios.kloukos@zmk.unibe.ch. canines.10 Shortly after, the multistranded wire retainer
Submitted, April 2019; revised and accepted, October 2019. was introduced11 and proposed in 2 types: the relatively
0889-5406/$36.00
Ó 2019 by the American Association of Orthodontists. All rights reserved.
rigid and thick multistranded wire fixed to the canines12
https://doi.org/10.1016/j.ajodo.2019.10.010 and the flexible smaller size multistranded wire bonded
156
Arn et al 157
to each tooth in the anterior segment.13 The alternatives Types of participants. Patients of any age who
to wire retainers are the fiber-reinforced composite received fixed retention after orthodontic treatment.
retainers and the ceramic alumina retention elements Types of interventions. All kinds of fixed retainers
which, however, are less popular.14-18 were considered eligible, irrespectively of the wire type,
The flexible multistranded wire bonded to each bonding materials, and the number of teeth bonded.
anterior tooth is currently considered the gold Types of control groups. Groups of individuals
standard.19,20 The preference to this retainer is without fixed retainers (removable retainers accepted),
attributed to the assumed extra mechanical retention repeated measurements of the same patients, or groups
to the composite resin due to the strands, and the wire with different wires for fixed retention.
flexibility that permits physiological tooth mobility and Types of outcomes. Periodontal measurements (of
reduced tension in the composite.21 Bonding on all any type) during fixed retention, assessing periodontal
anterior teeth is also effective in preventing relapse of health or disease, with no limit on the observation
the incisors.22 The thick stainless steel wire bonded period.
only to the canines was proposed for its effectiveness
in maintaining the intercanine width22 and for the easier Search methods for study identification
cleaning of the incisors,23 whereas the fiber-reinforced Detailed search strategies were developed and
composite retainer was selected for its esthetics and appropriately revised for each database, considering
biocompatibility.20 the differences in controlled vocabulary and syntax
Fixed orthodontic retainers are compliance free, rules. The following electronic databases were searched:
invisible, and worn continuously. However, teeth are Medline (via Ovid and PubMed from 1946 to August 31,
more prone to plaque and calculus accumulation,24 2019; Appendix 1), EMBASE (via Ovid), the Cochrane
and appropriate oral hygiene procedures are more Oral Health Group's Trials Register and CENTRAL.
complex and require more time.6 Widespread use of Unpublished literature was searched on ClinicalTrials.
fixed retainers and the need for long-term wear gov, the National Research Register, and Pro-Quest
demonstrate the importance of assessing the effects of Dissertation Abstracts and Thesis database. The search
this increased accumulation of deposits on the attempted to identify all relevant studies irrespective of
periodontium. Previous systematic reviews on the language. The reference lists of all eligible studies were
subject were inconclusive and also, most importantly, hand-searched for additional studies.
did not include all available studies existing in the
literature.23,25-27 Therefore, the first aim of this review Data collection and analysis
was to perform a more recent evaluation of the
potentially deleterious effects of fixed orthodontic Selection of studies. Study selection was performed
retainers on periodontal health. The secondary independently and in duplicate by the first 2 authors
objective was to compare different kinds of fixed (M.A. and K.D.) of the review, who were not blinded to
retainers according to their effects on periodontal the identity of the authors of the studies, their
health, and if possible, to recommend 1 of them. institutions, or the results of their research. The study
selection procedure was comprised of a staged reading
MATERIAL AND METHODS of titles, abstracts, and full-text. After exclusion of not
eligible studies, the full report of publications considered
The Preferred Reporting Items for Systematic Reviews by either author eligible for inclusion was obtained and
and Meta-Analyses28 were followed in reporting this assessed independently. Disagreements were resolved by
systematic review. The protocol of this study was not discussion and consultation with a third author (D.K.). A
registered in a publicly assessable database. This review record of all decisions on study identification was kept.
was conducted per the Cochrane Handbook for Data extraction and management. The first 2 authors
Systematic Reviews of Interventions.29 (M.A. and K.D.) performed data extraction indepen-
dently and in duplicate. Disagreements were resolved
Criteria for considering studies for this review by discussion or the involvement of a third author
Types of studies. Randomized controlled trials (D.K.). The following data were collected on a
(RCTs), controlled clinical trials, cohort studies of customized data collection form: (1) author, title, and
prospective and retrospective design, and cross- year of study; (2) design of the study; (3) number, age,
sectional studies were considered eligible for inclusion and gender of participants in intervention or control
in this review. Case reports, review articles, and animal groups; (4) type of retainer, dimensions of wire;
studies were excluded. (5) number of teeth upon which the retainer was
American Journal of Orthodontics and Dentofacial Orthopedics February 2020 Vol 157 Issue 2
158 Arn et al
bonded; (6) observation period (follow-up of patients); Data synthesis. We planned to conduct meta-
and (7) method of outcome assessment. If stated, analyses if there were studies of similar comparisons
sources of funding, trial registration, and publishing of reporting the same outcomes. Risk ratios would have
the trial's protocol were recorded. been combined for dichotomous data using the
Measures of effect. For continuous outcomes, mean random-effect models. In the case of qualitative
differences and standard deviations were used to synthesis, it was planned to include all eligible studies
summarize the data from each study. For dichotomous in the synthesis but to emphasize their weight according
data, the number of participants with events, and the to their quality.
total number of participants in the experimental and
control groups were analyzed. We planned to assess RESULTS
outcomes at more than 1 time point in the follow-up Selection of studies
period, where applicable. All such assessments were
recorded, and decisions on which time-of-outcome Of the 395 studies initially identified through
assessment to use from each study were based on the database searching, 29 studies fulfilled our inclusion
most commonly reported timing of assessment among criteria and were included in this systematic review.
all included studies. The flow diagram for the study selection is shown in
Unit-of-analysis issues. In all cases, the unit of
Figure. Of the 29 studies, 11 were RCTs, 4 prospective
analysis was the patient. We anticipated that some of cohort studies, 1 retrospective cohort study, and 13
the included studies would present data from repeated cross-sectional studies.
observations on participants, which could lead to
Risk of bias assessment
unit-of-analysis errors. In this case, we followed the
advice provided in Section 9.3.4 of the Cochrane Randomized controlled trials. The quality of the
Handbook for Systematic Reviews of Interventions.29 included RCTs was assessed per the revised Cochrane
Dealing with missing data. In studies where data were risk of bias tool for randomized trials (RoB 2.0)30 and
unclear or missing, we contacted the principal is presented in Table I.
investigators or the corresponding author, or both. Five RCTs33-37 reported a proper randomization
Assessment of heterogeneity. We assessed clinical, process; Tacken et al38 used a predictable allocation
methodological, and statistical heterogeneity by sequence, whereas the rest of the RCTs39-43 did not
examining the characteristics of the studies, the report enough information to assess this domain.
similarity between the types of participants, the Because of the nature of the intervention, the
interventions, and the outcomes as specified in inclusion blinding of the outcome assessor(s) was challenging.
criteria for considering studies for this review. Nevertheless, a low risk of bias was awarded when the
Assessment of reporting biases. Reporting biases vested interest of the outcome assessor(s) regarding
arise when the reporting of research findings is affected the results of the study was estimated as low.
by the nature or direction of the findings themselves. We Two RCTs41,42 showed a high risk of selective
attempted to minimize potential reporting biases, reporting. A preregistered protocol is necessary to
including publication bias, multiple (duplicate reports) control any risk of selection of the reported results and
publication bias, and language bias in this review, by was available only for Torkan.37 Finally, 2 studies40,43
conducting an accurate and at the same time, a sensitive had a high percentage of missing data, without balanced
search of multiple sources with no restriction on reasons and proportions across the groups. Torkan
language. We also searched for ongoing trials. et al37 was judged overall to be at low risk of bias for
Quality assessment of included studies. The quality 1 outcome and unclear for the other 4.
of the included RCTs was assessed per the revised Co- Cohort studies. None of the 5 cohort studies fulfilled
chrane risk of bias tool for randomized trials (RoB 2.0).30 adequately the criteria dictated by the Risk Of Bias In
The risk of bias of the included cohort studies of the pro- Nonrandomized Studies of Interventions Tool in order
spective and retrospective design was assessed using the to be graded at a low risk of bias (Table II). The tool
Risk Of Bias In Nonrandomized Studies of Interventions requires, among other things, absence of confounding,
Tool.31 A modified version of the Newcastle-Ottawa scale blinding of the outcome(s) assessor(s), and a
(NOS) was used for cross-sectional studies.32 For the sam- preregistered protocol to consider a nonrandomized
ple selection process, a maximum of 3 stars was given, for study equivalent to a well-performed randomized trial.
the comparability of the groups, 2 stars, and the outcome Three trials44-46 had an overall moderate risk of bias,
domain, again, 2 stars. When scoring a total of 7 stars, the and therefore appear to provide sound evidence for
study was considered as “good.” nonrandomized studies. Juloski et al47 used a control
February 2020 Vol 157 Issue 2 American Journal of Orthodontics and Dentofacial Orthopedics
Arn et al 159
Records identified through database searching:

(n = 395)
Identification
Records remaining after title screening Additional records identified through other
(n =163) sources or hand-searching
(n =5)
Records excluded
Records before duplicate removal (n = 69)
Screening
(n = 168) 69 duplicates removed
Records excluded
Full-text articles (n = 65)
assessed for eligibility 65 studies excluded after
(n = 99) abstract reading stage
Eligibility
Full-text articles excluded,

with reasons
Studies included in (n = 5)
qualitative synthesis
(n = 29) 1 study presenting
a case-report
2 studies assessing
periodontal outcomes of
Included
fixed appliances
Studies included in 2 studies not assessing
quantitative synthesis periodontal outcomes
(n =0)
Fig. Studies' Flow diagram.
group with a different population (different era) than the Qualitative synthesis of the included studies
intervention group, which could involve uncontrollable Because of the great heterogeneity in study designs,
confounding. Furthermore, in G€ okçe and Kaya,46 types of wire used, comparisons made, outcomes
because of the retrospective design, the selection of reported (Supplementary Table I), and the overall low
participants was related to the intervention, which quality of the included studies, a meta-analysis was
implies selection bias. Finally, Wu et al48 was graded at not conducted. For descriptive reasons, the 29 included
serious risk of bias because of the absence of a control studies were divided into the following 5 categories. The
group. results of the included studies are summarized in
Cross-sectional studies. The 13 cross-sectional Supplementary Tables II and III.
studies were graded according to a modified version of Studies comparing fixed retainers to orthodontically
the NOS.32 None of the studies scored 7 stars (necessary treated or untreated controls without retainers. Ten
to be considered as a “good” study), mostly because studies compared individuals with fixed retainers to a con-
of the lack of blinding and adequate control of trol group without retainers. In contrast to the general
confounding factors. The scores ranged from 2 to 6 stars consensus, 1 RCT38 and 2 cross-sectional studies49,50
(Table III). concluded that the presence of a fixed retainer was
160 Arn et al
Table I. Risk of bias of the included RCTs

Bias arising Bias due to
from the deviations Bias in Bias in the
randomization from intended Bias due to measurement selection of the
No. Author(s)/y process interventions missing data of outcomes reported result Overall
1 Al-Moghrabi Low risk Low risk Some concerns Some concerns Some concerns Some concerns
et al 201833
2 Artun et al 198739 Some concerns Low risk Low risk Low risk Some concerns Some concerns
3 Artun et al 199740 Some concerns Low risk High risk Low risk Some concerns High risk
4 Bazargani et al Low risk Low risk Low risk Low risk Some concerns Some concerns
201234
5 Liu 201035 Low risk Low risk Low risk Some concerns Some concerns Some concerns
6 Rose et al 200241 Some concerns Low risk Low risk Some concerns High risk High risk
7 Storey et al 201836 Low risk Low risk Low risk Some concerns Some concerns Some concerns
8 St€
ormann Some concerns Low risk Low risk Some concerns High risk High risk
and Ehmer 200242
9 Tacken et al 201038 High risk Low risk Low risk Some concerns Some concerns High risk
10 Torkan et al 201437 Low risk Low risk Low risk Some concerns for Low risk Some concerns for the
the outcomes outcomes PI/CI/GI/BOP
PI/CI/GI/BOP Low risk for the outcome
Low risk for the “width of the PDL space”
outcome
“width of the
PDL space”
11 Xu et al 201143 Some concerns Low risk High risk Some concerns Some concerns High risk
PI, plaque index; CI, calculus index; GI, gingival index; BOP, bleeding on probing; PDL, periodontal ligament.
associated with poorer periodontal conditions. The 7 Studies assessing different time-points of the same pa-
other studies,45,47,51-55 1 prospective cohort study, 1 tients. Three RCTs35,37,42 and 1 prospective cohort study48
retrospective cohort study, and 5 cross-sectional studies, assessed the same patients at different time-points.
did not describe any periodontal complications related Liu35 observed a higher level of gingival inflammation
to fixed retainers. On the contrary, Booth et al52 and Ar- after the implementation of the interventions, whereas
tun51 reported that more gingival inflammation was pre- ormann and Ehmer42 and Wu et al48 did not find any
St€
sent in the lingual areas of participants without a fixed detrimental effects on gingival conditions. In Torkan
retainer, meaning that patients with a fixed retainer could et al,37 the fiber-reinforced composite retainer was related
have improved oral hygiene because of the more regular to an increased accumulation of plaque and gingivitis,
recalls for prophylaxis. Finally, Westerlund et al55 did whereas the results for the multistranded wire retainer
not find any difference in marginal bone levels between were unclear.
orthodontically treated patients with a fixed retainer and Studies comparing different types of fixed retainers.
orthodontically treated patients without a fixed retainer. Eight RCTs, 1 prospective cohort study, and 5
Studies comparing fixed retainers to removable cross-sectional studies compared different types of fixed
retainers. 5 RCTs,33,36,39,40,43 2 prospective cohort retainers.
studies,44,46 and 4 cross-sectional studies51,54,56,57 Artun published 2 RCTs39,40 that included the same
compared fixed retainers to removable retainers. Of the participants, but with different follow-up periods. In
11 studies, only G€ okçe and Kaya46 reported more both studies, spiral wires bonded to the canines or all
gingival inflammation in the presence of a wire retainer, anterior teeth did not seem more deleterious to
although no significant difference in plaque accumula- periodontal health when compared with plain wires
tion, probing depth, and marginal recession was re- bonded to the canines.
ported between the groups. Four RCTs35,37,38,41 compared fiber-reinforced
Some of the studies36,43,44,54,56 reported and increased composite retainers to multistranded wire retainers.
accumulation of deposits (plaque and calculus) in patients Two of them found no difference in gingival bleeding
wearing fixed retainers, but no related periodontal com- and periodontal pocket depth35 or calculus
plications. Therefore, the higher calculus accumulation accumulation41 between the groups. However, Tacken
described in Xu et al43 cannot be considered per se as proof et al38 and Torkan et al37 agreed that multistranded
of a negative effect on the periodontium. wires were better in terms of gingival inflammation.
Arn et al 161
Bazargani et al34 reported that the application of
No control group
resin during the bonding procedure reduces the
Moderate risk
Moderate risk
Moderate risk
Overall
ormann and Ehmer42
Serious risk
Serious risk
deposition of calculus, whereas St€
found similar periodontal outcomes for customized
wires when compared with prefabricated wires.
MGI, modified gingival index; BOP, bleeding on probing; PI, plaque index; CI, calculus index; GCFF, gingival crevicular fluid flow; PPD, pocket probing depth; GI, gingival index.
All cross-sectional studies51,57-60 compared smooth
Bias in selection
wires to multistranded wires. Three of them did not

of the reported
Moderate risk
Moderate risk
Moderate risk
Moderate risk
Moderate risk observe any difference in gingival conditions between

result
the 2 types of retainers, whereas Knaup et al60 and

Rody et al57 found a higher degree of gingivitis in the
presence of multistranded wire retainers. The results of
the accumulation of deposits were more in favor of the
Low risk for the biochemical and
Moderate risk for the outcomes
Moderate risk for the outcomes
smooth wire retainers. Artun51 described an increased

Low risk for the outcome GCFF
Bias in measurement
accumulation of plaque and calculus along the wire in

bacteriologic analysis
the plain wire group, whereas the other cross-sectional

of outcomes
PPD/BOP/GI/PI/CI
studies reported more plaque deposition in the

MGI/BOP/PI/CI
multistranded wire group. Corbett et al59 observed that

Moderate risk
Moderate risk
Moderate risk
the wave-type wire was related to a higher frequency

and easiness of flossing; nevertheless, even so,
self-reported flossing compliance did not improve the
periodontal status in this group.
Studies assessing different vertical wire positions.
missing data
Bias due to
One prospective cohort study45 and 1 cross-sectional

Low risk
Low risk
Low risk
Low risk
Low risk
study49 observed that positioning a fixed retainer more

incisally or more gingivally does not influence the
periodontal outcomes.
deviations from
Other studies. A cross-sectional study61 compared

interventions
Bias due to
intended
the short- and long-term effects of fixed retainers on

Low risk
Low risk
Low risk
Low risk
Low risk
periodontal health. The long-term retention group

showed higher calculus accumulation, greater marginal
recession, and increased probing depth. However, the
classification of
difference in age between the 2 groups could be the

interventions
Bias in the
cause of these findings and not the longer presence of

a fixed retainer itself.
Low risk
Low risk
Low risk
Low risk
Low risk
DISCUSSION
Table II. Risk of bias of the included cohort studies
This systematic review aimed to evaluate the

Serious risk
Selection
potentially deleterious effects of fixed retainers on

Low risk
Low risk
Low risk
Low risk
bias
periodontal health. As a secondary aim, different kinds

of fixed retainers were compared according to their
effects on the periodontium. Eleven RCTs and 18
Not applicable
Confounding
Moderate risk
Moderate risk
Moderate risk
nonrandomized studies fulfilled the inclusion criteria.

Serious risk
Most of the included studies seem to indicate that

fixed retainers do not affect periodontal health. In
contrast to the general consensus, 2 RCTs,35,38 1 pro-
spective cohort study,46 and 2 cross-sectional
Juloski et al 201747
Heier et al 199744
studies49,50 reported poorer periodontal conditions in

Kaji et al 201345
Wu et al 201448
okçe and Kaya
Author(s)/y
the presence of a fixed retainer. One RCT37 showed

201946
that fiber-reinforced composite retainers could be asso-

ciated with periodontal complications, but found un-
G€
clear results for multistranded wire retainers. However,

no severe detrimental effects on periodontal health
No.
1
3
4
were reported in any of the included studies.
162 Arn et al
Table III. Risk of bias of the included cross-sectional studies

Selection Comparability Outcome(s)
Comparability
Ascertainment factors and
Representativeness of the exposure confounding Assessment of
No. Author(s)/y of the sample Nonrespondents (risk factor) factors the outcome(s) Statistical test NOS score
1 Al-Nimri et al b* a* a* a* d a* 6*
200958 b*
2 Artun 198451 b* c a* No star a* a* 4*
3 Booth et al a* a* a* No star d a* 4*
200852
4 Cerny et al d c a* No star d a* 2*
201056
5 Neto et al c a* a* No star d a* 3*
201050
6 Corbett et al a* a* a* No star d a* 4*
201559
7 Dietrich et al a* a* a* No star d a* 4*
201453
8 Knaup et al a* a* a* No star d a* 4*
201960
9 Levin et al a* a* a* No star d a* 4*
200849
10 Pandis et al b* c a* No star d a* 3*
200761
11 Rody et al b* c a* a* d a* 4*
201154
12 Rody et al b* c a* No star d a* 3*
201657
13 Westerlund b* c a* a* d a* 4*
et al 201755
Note. See Supplementary Data for the meaning of a-d for each domain.
The results of the studies comparing smooth round Three prospective cohort studies44-46 were graded as
wires to multistranded wires39,40,51,57-60 were not good with a moderate risk of bias; the other 2 cohort
unanimous, although most of them reported no studies47,48 showed; however, a serious risk of bias.
difference between the interventions. On the other Included cross-sectional studies were also unsatisfactory
hand, fiber-reinforced composite retainers could be with a NOS score of 3 or 4 stars for most of them.
more deleterious to the periodontium than wire Therefore, a meta analysis could unfortunately not be
retainers.37,38 Two studies45,49 also showed that conducted, and despite the significant amount of
positioning the retainer more incisally or more gingivally included studies, clear recommendations for usual prac-
does not seem to influence the periodontal outcomes. tice according to the currently available literature were
Nevertheless, because of the limitations of this review, difficult to draw.
recommendations on the best type of fixed retainer to Another limitation of this systematic review was the
use cannot be given. observation period of the included studies. Many
The current systematic review presents several limita- practitioners indeed recommend life-long retention,
tions that have to be highlighted. The first limitation was but relevant evidence on the potential side effects after
the high amount of methodological heterogeneity in such a long time is missing. The follow-ups of the
study designs, types of wire used, comparisons made, included RCTs and cohort studies ranged from 4 months
and outcomes reported, which could be identified across to 4 years and from 2 months to 5 years, respectively.
the included studies. A second limitation was the lack of Some of the included cross-sectional studies had long
high-quality evidence. All included RCTs were indeed at follow-up periods52,55,56,61; for example, Booth et al52
unclear or high risk of bias. Most of them had issues with assessed a gingival index after 20-29 years. Nevertheless,
the blinding of the outcome(s) assessor(s), which will be their quality was questionable.
a recurrent problem in prospective studies since the The findings of this review are consistent with
intervention appraised is evident and easy to detect. previous systematic reviews.23,25-27 However, the
Arn et al 163
number of included studies was substantially lower, 12. Artun J, Zachrisson B. Improving the handling properties of a
ranging from 1 to 7 trials evaluating periodontal composite resin for direct bonding. Am J Orthod 1982;81:269-76.
13. Zachrisson BU. The bonded lingual retainer and multiple spacing
outcomes. The paucity of high-quality evidence was
of anterior teeth. Swed Dent J Suppl 1982;15:247-55.
also pointed out, and none of them was capable of 14. Diamond M. Resin fiberglass bonded retainer. J Clin Orthod 1987;
providing clinical guidance according to the available 21:182-3.
literature. The more recent review,27 which included 4 15. Orchin JD. Permanent lingual bonded retainer. J Clin Orthod 1990;
RCTs and 3 prospective cohort studies, highlighted the 24:229-31.
16. Geserick M, Ball J, Wichelhaus A. Bonding fiber-reinforced lingual
relatively minor changes in periodontal parameters but
retainers with color-reactivating flowable composite. J Clin Orthod
concluded that more prospective studies with matched 2004;38:560-2.
controls incorporating baseline assessment are needed 17. Amundsen OC, Wisth PJ. Clinical pearl: LingLock—the flossable
to develop useful conclusions. fixed retainer. J Orthod 2005;32:241-3.
18. Annousaki O, Zinelis S, Eliades G, Eliades T. Comparative analysis
of the mechanical properties of fiber and stainless steel
CONCLUSIONS
multistranded wires used for lingual fixed retention. Dent Mater
According to the currently available literature, 2017;33:e205-11.
orthodontic fixed retainers seem to be a retention 19. Labunet AV, Badea M. In vivo orthodontic retainer survival—a
review. Clujul Med 2015;88:298-303.
strategy rather compatible with periodontal health, or
20. Oshagh M, Heidary S, Dehghani Nazhvani A, Koohpeima F, Koohi
at least not related to severe detrimental effects on the Hosseinabadi O. Evaluation of histological impacts of three types
periodontium. However, a lack of high-quality evidence of orthodontic fixed retainers on periodontium of rabbits. J Dent
is indisputable. Well-performed RCTs, comparing fixed (Shiraz) 2014;15:104-11.
retainers to untreated subjects and different kinds of 21. Dahl EH, Zachrisson BU. Long-term experience with direct-
bonded lingual retainers. J Clin Orthod 1991;25:619-30.
fixed retainers, during prolonged periods of follow-up,
22. Butler J, Dowling P. Orthodontic bonded retainers. J Ir Dent Assoc
are needed to provide definitive conclusions about the 2005;51:29-32.
effects of fixed retainers on periodontal health and 23. Littlewood SJ, Millett DT, Doubleday B, Bearn DR,
clinical recommendations. Worthington HV. Retention procedures for stabilising tooth
position after treatment with orthodontic braces. Cochrane
Database Syst Rev 2016;1:CD002283.
SUPPLEMENTARY DATA
24. Jepsen S, Deschner J, Braun A, Schwarz F, Eberhard J. Calculus
Supplementary data associated with this article can removal and the prevention of its formation. Periodontol 2000
be found, in the online version, at https://doi.org/10. 2011;55:167-88.
25. Littlewood SJ, Millett DT, Doubleday B, Bearn DR,
1016/j.ajodo.2019.10.010.
Worthington HV. Orthodontic retention: a systematic review.
J Orthod 2006;33:205-12.
REFERENCES 26. Westerlund A, Daxberg E, Liljegren A. Stability and side effects of
orthodontic retainers—a systematic review. Dentistry 2014;04:258.
1. Melrose C, Millett DT. Toward a perspective on orthodontic 27. Al-Moghrabi D, Pandis N, Fleming PS. The effects of fixed and
retention? Am J Orthod Dentofacial Orthop 1998;113:507-14. removable orthodontic retainers: a systematic review. Prog Orthod
2. Johnston CD, Littlewood SJ. Retention in orthodontics. Br Dent J 2016;17:24.
2015;218:119-22. 28. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred
3. Lopez-Areal L, Gandıa JL. Relapse of incisor crowding: a visit to the Reporting Items for Systematic Reviews and Meta-Analyses: the
Prince of Salina. Med Oral Patol Oral Cir Bucal 2013;18:e356-61. PRISMA statement. J Clin Epidemiol 2009;62:1006-12.
4. Kaan M, Madlena M. Retention and relapse. Review of the 29. Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews
literature: [in Hungarian]. Fogorv Sz 2011;104:139-46. of Interventions. Chichester: John Wiley & Sons; 2011.
5. Reitan K. Clinical and histologic observations on tooth movement 30. Higgins JPT, Savovic J, Page MJ, Sterne JAC, et al. Revised Co-
during and after orthodontic treatment. Am J Orthod 1967;53: chrane risk of bias tool for randomized trials (RoB 2.0), updated
721-45. 20 October 2016. Available from http://www.riskofbias.info.
6. Littlewood SJ, Kandasamy S, Huang G. Retention and relapse in 31. Sterne JA, Hernan MA, Reeves BC, Savovic J, Berkman ND,
clinical practice. Aust Dent J 2017;62(Suppl 1):51-7. Viswanathan M, et al. ROBINS-I: a tool for assessing risk of bias in
7. Little RM, Riedel RA, Artun J. An evaluation of changes in non-randomised studies of interventions. BMJ 2016;355:i4919.
mandibular anterior alignment from 10 to 20 years postretention. 32. Wells GA, Shea B, O’Connell D, Peterson J, Welch V, Losos M,
Am J Orthod Dentofacial Orthop 1988;93:423-8. Tugwell P. The Newcastle-Ottawa Scale (NOS) for assessing the
8. Boese LR. Fiberotomy and reproximation without lower retention, quality of nonrandomised studies in meta-analyses 2017. Avail-
nine years in retrospect: part I. Angle Orthod 1980;50:88-97. able from http://www.ohri.ca/programs/clinical_epidemiology/
9. Bearn DR. Bonded orthodontic retainers: a review. Am J Orthod oxford.asp.
Dentofacial Orthop 1995;108:207-13. 33. Al-Moghrabi D, Johal A, O’Rourke N, Donos N, Pandis N, Gonzales-
10. Knierim RW. Invisible lower cuspid to cuspid retainer. Angle Marin C, et al. Effects of fixed vs removable orthodontic retainers on
Orthod 1973;43:218-20. stability and periodontal health: 4-year follow-up of a randomized
11. Zachrisson BU. Clinical experience with direct-bonded orthodontic controlled trial. Am J Orthod Dentofacial Orthop 2018;154:
retainers. Am J Orthod 1977;71:440-8. 167-74.e1.
164 Arn et al
34. Bazargani F, Jacobson S, Lennartsson B. A comparative evaluation 48. Wu HM, Zhang JJ, Pan J, Chen D. Clinical evaluation of glass
of lingual retainer failure bonded with or without liquid resin. fiber-reinforced composites for fixed orthodontic lingual retainers:
Angle Orthod 2012;82:84-7. [in Chinese]. Shanghai Kou Qiang Yi Xue 2014;23:80-2.
35. Liu Y. Application of fiber-reinforced composite as fixed lingual 49. Levin L, Samorodnitzky-Naveh GR, Machtei EE. The association of
retainer. Hua Xi Kou Qiang Yi Xue Za Zhi 2010;28:290-3. orthodontic treatment and fixed retainers with gingival health. J
36. Storey M, Forde K, Littlewood SJ, Scott P, Luther F, Kang J. Periodontol 2008;79:2087-92.
Bonded versus vacuum-formed retainers: a randomized controlled 50. Cesar Neto JB, Regio MRS, Martos J, Spautz F, Moraes GBD.
trial. Part 2: periodontal health outcomes after 12 months. Eur J Analysis of the periodontal status of patients with mandibular-
Orthod 2018;40:399-408. bonded retainers. Rev odonto ciênc (Online) 2010;25:132-6.
37. Torkan S, Oshagh M, Khojastepour L, Shahidi S, Heidari S. Clinical 51. Artun J. Caries and periodontal reactions associated with
and radiographic comparison of the effects of two types of fixed long-term use of different types of bonded lingual retainers. Am
retainers on periodontium—a randomized clinical trial. Prog J Orthod 1984;86:112-8.
Orthod 2014;15:47. 52. Booth FA, Edelman JM, Proffit WR. Twenty-year follow-up of
38. Tacken MP, Cosyn J, De Wilde P, Aerts J, Govaerts E, Vannet BV. patients with permanently bonded mandibular canine-to-canine
Glass fibre reinforced versus multistranded bonded orthodontic retainers. Am J Orthod Dentofacial Orthop 2008;133:70-6.
retainers: a 2 year prospective multi-centre study. Eur J Orthod 53. Dietrich P, Patcas R, Pandis N, Eliades T. Long-term follow-up of
2010;32:117-23. maxillary fixed retention: survival rate and periodontal health. Eur
39. Artun J, Spadafora AT, Shapiro PA, McNeill RW, Chapko MK. J Orthod 2015;37:37-42.
Hygiene status associated with different types of bonded, 54. Rody WJ Jr, Akhlaghi H, Akyalcin S, Wiltshire WA,
orthodontic canine-to-canine retainers. A clinical trial. J Clin Wijegunasinghe M, Filho GN. Impact of orthodontic retainers on
Periodontol 1987;14:89-94. periodontal health status assessed by biomarkers in gingival
40. Artun J, Spadafora AT, Shapiro PA. A 3-year follow-up study of crevicular fluid. Angle Orthod 2011;81:1083-9.
various types of orthodontic canine-to-canine retainers. Eur J 55. Westerlund A, Oikonomou C, Ransj€ o M, Ekestubbe A, Bresin A,
Orthod 1997;19:501-9. Lund H. Cone-beam computed tomographic evaluation of the
41. Rose E, Frucht S, Jonas IE. Clinical comparison of a multistranded long-term effects of orthodontic retainers on marginal bone levels.
wire and a direct-bonded polyethylene ribbon-reinforced resin Am J Orthod Dentofacial Orthop 2017;151:74-81.
composite used for lingual retention. Quintessence Int 2002;33: 56. Cerny R, Cockrell D, Lloyd D. Long-term results of permanent
579-83. bonded retention. J Clin Orthod 2010;44:611-6: quiz 22.
42. St€ormann I, Ehmer U. A prospective randomized study of different 57. Rody WJ Jr, Elmaraghy S, McNeight AM, Chamberlain CA, Antal D,
retainer types. J Orofac Orthop 2002;63:42-50. Dolce C, et al. Effects of different orthodontic retention protocols
43. Xu XC, Li RM, Tang GH. Clinical evaluation of lingual fixed retainer on the periodontal health of mandibular incisors. Orthod Craniofac
combined with Hawley retainer and vacuum-formed retainer: [in Res 2016;19:198-208.
Chinese]. Shanghai Kou Qiang Yi Xue 2011;20:623-6. 58. Al-Nimri K, Al Habashneh R, Obeidat M. Gingival health and
44. Heier EE, De Smit AA, Wijgaerts IA, Adriaens PA. Periodontal relapse tendency: a prospective study of two types of lower fixed
implications of bonded versus removable retainers. Am J Orthod retainers. Aust Orthod J 2009;25:142-6.
Dentofacial Orthop 1997;112:607-16. 59. Corbett AI, Leggitt VL, Angelov N, Olson G, Caruso JM. Periodontal
45. Kaji A, Sekino S, Ito H, Numabe Y. Influence of a mandibular fixed health of anterior teeth with two types of fixed retainers. Angle
orthodontic retainer on periodontal health. Aust Orthod J 2013; Orthod 2015;85:699-705.
29:76-85. 60. Knaup I, Wagner Y, Wego J, Fritz U, J€ager A, Wolf M. Potential
46. G€okçe B, Kaya B. Periodontal effects and survival rates of different impact of lingual retainers on oral health: comparison between
mandibular retainers: comparison of bonding technique and wire conventional twistflex retainers and CAD/CAM fabricated nitinol
thickness. Eur J Orthod 2019. retainers: a clinical in vitro and in vivo investigation. J Orofac
47. Juloski J, Glisic B, Vandevska-Radunovic V. Long-term influence Orthop 2019;80:88-96.
of fixed lingual retainers on the development of gingival recession: 61. Pandis N, Vlahopoulos K, Madianos P, Eliades T. Long-term
a retrospective, longitudinal cohort study. Angle Orthod 2017;87: periodontal status of patients with mandibular lingual fixed
658-64. retention. Eur J Orthod 2007;29:471-6.
Supplementary Table I. Characteristics of the included studies
American Journal of Orthodontics and Dentofacial Orthopedics
Arn et al
Study Setting Sample size, sex,
Author(s)/y design and date and age of participants Intervention(s) Periodontal outcome(s) Follow-up period(s)
Al-Moghrabi RCT Queen Mary University n 5 42 In the lower arch: On the lower anterior teeth: 4 y (mean retention
et al 201833 of London 10 male, 32 female 0.0175" Modified gingival index period 5 4.16 6 0.35 y)
Date not reported Mean age 5 21.15 6 2.41 y Coaxial wire from canine (Lobene 1986)
to canine (n 5 21; Calculus and plaque
3 male, 18 female; mean levels
age 5 21.54 6 3.06 y) Clinical attachment level
Vacuum-formed Bleeding on probing
retainer, wear time
progressively reduced
(n 5 21; 7 male,
14 female; mean
age 5 20.77 6 1.49 y)
Artun et al RCT One private practice n 5 44 In the lower arch: Along the wire and the 0/4 mo
198739 Date not reported Sex and age not reported 0.032" Plain wire composite:
bonded to the canines Modified plaque index
(n 5 11) (L€
oe 1967)
0.032" Spiral wire Modified calculus
bonded to the canines index (Ramfjord 1959)
(n 5 11)
0.025" Spiral wire Along the gingival margin of
bonded to all anterior the lower anterior teeth
teeth (n 5 11) (ID/L):
Plaque index (L€oe 1967)
Removable retainer
(n 5 11) Calculus index (Ramfjord
1959)
Gingival index (L€
oe
1967)
Gingival bleeding (GI
score 2)
Artun et al RCT One private practice n 5 49 In the lower arch: Along the wire and the 0/3 y
February 2020 Vol 157 Issue 2
199740 Date not reported Sex and age not reported 0.032" Plain wire bonded composite:
to the canines (n 5 11) Modified plaque index
0.032" Spiral wire (L€
oe 1967)
bonded to the canines Modified calculus
(n 5 13) index (Ramfjord 1959)
0.025" Spiral wire
bonded to all anterior Along the gingival margin of
the lower anterior teeth
teeth (n 5 11)
removable retainer (ID/L):
(n 5 14) Plaque index (L€oe 1967)
Calculus index
(Ramfjord 1959)
164.e1
Gingival index (L€
oe
1967)
Probing attachment level
(L only)
Supplementary Table I. Continued
164.e2
Bazargani RCT Postgraduate Dental n 5 52 In the lower arch: Calculus accumulation 2 y (mean retention
et al 201234 Education Center, 26 male, 26 female 0.0195" Multistranded (present or not) adjacent to period 5 24.4 6 4.7 mo)
Orebro, Kingdom of Mean age 5 18.3 6 1.3 y wire bonded to all the wire and the composite
Sweden anterior teeth with
Between 2008 and 2010 composite only (n 5 26;
12 male, 14 female; mean
age 5 18.2 6 1.6 y)
0.0195" Multistranded
wire bonded to all
anterior teeth with 2-step
bonding resin
and composite (n 5 26; 14
male, 12 female;
mean age 5 18.4 6 1.6 y)
Liu 201035 RCT Hospital of Stomatology, n 5 60 In the lower arch: On the lingual aspect of the 0/6/12 mo
Peking University School Sex and age not reported 0.0295" Fiber-reinforced lower anterior teeth
December 2007 to April 2008 composite retainer (Ml/L/DL):
bonded to all anterior Bleeding
teeth (n 5 30) index

0.0354" Multistranded Pocket depth
stainless steel wire
bonded to all anterior
teeth (n 5 30)
Rose et al RCT Freiburg University, Federal n 5 20 In the lower arch: Amount of calculus in the 3/6/9/12/15/18/21/24 mo
200241 Republic of Germany 12 male, 8 female 0.0393" Plasma-treated gingival area on the lingual
Date not reported Mean age 5 22.4 6 9.7 y woven polyethylene surface of the incisors
ribbon bonded to all
anterior teeth (n 5 10)
stainless steel wire
teeth (n 5 10)
Storey RCT St Luke's Hospital, Bradford n 5 60 In both arches: On all erupted teeth 0/3/6/12 mo
et al 201836 and York Hospital, and 30 male, 30 female 0.0195" Three-stranded in both arches (B/L/M/D):
Leeds Dental Institute, Mean age not reported twistflex stainless steel Plaque index
United Kingdom of Great wire bonded to all (Silness and L€
oe 1964)
Britain and Northern anterior teeth (n 5 30; Gingival index (L€
oe
Ireland 15 male, 15 female; 1967)
March 2012 to median age 5 16 y) Calculus index (Greene
Arn et al
September 2013 Vacuum-formed retainer and Vermillion 1964)
at night (n 5 30; 12
male, 18 female;
median age 5 17 y)
Arn et al
St€
ormann and RCT Not reported n 5 103 In the lower arch: Plaque index (Lange 1/3/2012/24 mo
Ehmer 200242 Sex not reported 0.0195" Customized wire 1977)
Age range 5 13-17 y bonded to all anterior Bleeding on probing
teeth (n 5 31)
0.0215" Customized wire
teeth (n 5 38)
Prefabricated retainer
bonded to the canines
(n 5 34)
Tacken et al RCT Three private practices n 5 184 In the upper arch, bonded to At 3 sites per tooth (M/C/D): 6/12/2018/24 mo
201038 Date not reported 90 male, 94 female all incisors, and in the Plaque index (Quigley
Mean age 5 14 y lower arch, bonded to all and Hein 1962)
anterior teeth: Modified gingival index
GF-reinforced retainer - (Lobene et al 1986)
500 unidirectional GF Bleeding on probing
(n 5 45; 23 male, (Cosyn and Verelst 2006)
22 female; mean
age 5 14.8 6 1.3 y)
GF-reinforced retainer -
1000 unidirectional GF
(n 5 48; 23 male, 25
female; mean age 5
14.6 6 2.7 y)
wire (n 5 91; 44
male, 47 female; mean
age 5 15.0 6 1.3 y)
Untreated subjects
(n 5 90; 45 male, 45
female; mean
age 5 14.1 6 1.1 y)
Torkan RCT Shiraz University of Medical n 5 40 In both arches: On the lingual aspect of the 0/6 mo
et al 201437 Sciences, Islamic Republic After exclusion and attrition: Fiber-reinforced upper and lower anterior
of Iran 10 male, 20 female composite retainer teeth:
September 2008 to January bonded to all anterior Plaque index (L€oe 1967)
2010 teeth (n 5 20; after Calculus index (Ramfjord
exclusion/attrition: 6 male 1959, Greene and
and 9 female, mean age Vermillion 1960)
5 16.2 6 1.9 y) Gingival index (L€
oe
0.0175" Multistranded 1967)
164.e3
stainless steel wire Bleeding on probing
164.e4
bonded to all anterior On the upper and lower
teeth (n 5 20; after anterior teeth:
exclusion/attrition: 4 Width of the PDL space
male and 11 female, (on periapical
mean age 5 15.7 6 radiographs)
2.1 y)
Xu et al 201143 RCT Ninth People's Hospital, n 5 45 Combined retainers: Calculus index of the 6/12 mo
Shanghai Jiao Tong 16 male, 29 female 0.0195" multistranded upper and lower arches
University NiTi wire bonded
School of Medicine 3-3 1 Hawley retainer at
Date not reported night (n 5 20; 6 male, 14
female; mean age 5
15.2 y)
Vacuum-formed
retainer, for both
arches, worn the whole
day (n 5 25; 10 male,
15 female; mean
age 5 13.6 y)
G€
okçe and Prospective cohort Baskent University, Ankara, n 5 100 In the lower arch: On the lower anterior teeth: 0/1 week-1/3/6 mo
Kaya 201946 Republic of Turkey 39 male, 61 female 0.0215" Multistranded Plaque index (L€oe 1967)
Orthodontic treatment Median age range 5 wire direct bonded to all (B/L)
completed between 16.5-18.0 y anterior teeth (n 5 20; 8 Gingival index (L€
oe
February and November male, 12 female; 1967) (B/L/M/D)
2016 median age 5 17.5 y) Probing depth (Ml/L/DL)
0.0215" Multistranded Marginal recession
wire indirect bonded to (Ml/L/DL)
all anterior teeth (n 5 20; Bleeding on probing
7 male, 13 female; (MB/B/DB/Ml/L/DL)
median age 5 18 y)
wire direct bonded to
all anterior teeth (n 5 20;
6 male, 14 female;
median age 5 16.5 y)
wire indirect bonded to
all anterior teeth
(n 5 20; 11 male, 9
female; median
Arn et al
age 5 17.5 y)
Arn et al
Removable Essix
appliance (1-mm
thickness) to wear day
and night (n 5 20; 7
male, 13 female;
median age 5 16.5 y)
Heier Prospective cohort One private practice and the n 5 36 In both arches (majority of On the upper and lower 0/1/3/6 mo
et al 199744 dental clinic at the Free Sex not reported participants): anterior teeth:
University of Brussels Age range 5 12.8-21.1 y; 0.0175" Multistranded Modified plaque index
Date not reported mean age 5 16.3 y wire bonded to all (Quigley and Hein 1962)
anterior teeth (n 5 22) (B/L)
Removable retainer Calculus index (Volpe
(n 5 14) et al 1967) (L€oe 1967)
Modified gingival index
(Lobene et al 1986, 1989)
(B/L/ID)
Bleeding on probing
(B/L/ID)
On 11M/B, 23D/L, 31D/L,
43M/B:
GCF flow
Kaji Prospective cohort One private practice n 5 31 In the lower arch: On the mesiolingual area of 0/1/4/8 weeks
et al 201345 Date not reported 6 male, 25 female Stainless steel wire the lower left canine:
Mean age 5 27.5 6 7.0 y bonded to all anterior Plaque index (Silness and
teeth (n 5 17; 4 male, L€
oe 1964)
13 female; mean Calculus index (Greene
age 5 28.6 6 7.8 y) and Vermillion 1960)
No fixed retainer, no Gingival index (L€ oe and
evidence of visual Silness 1963)

malalignment between Pocket probing depth
the lower bicuspids Bleeding on probing
(n 5 14; 2 male, GCF collection (volume,
12 female; mean elastase activity, protein
age 5 26.1 6 5.9 y) content, quantity of IL-
1b and PGE2)
Subgroups based on wire
Stimulated saliva [LDH
position: activity, F-Hb, counts of
Wire within the P.g and T.f]
upper one-third of
clinical crown (n 5 7)
164.e5
164.e6
Wire within the lower
two-thirds of
clinical crown (n 5 10)
Wu et al 201448 Prospective Shanghai Stomatological n 5 12 Glass fiber-reinforced On the lingual aspect of the 0/6/12/18/24 mo
noncontrolled Disease Center 5 male, 7 female composite lower anterior teeth:
cohort April 2011 to August 2013 Age range 5 17-25 y retainer bonded to all Gingival index
lower anterior teeth Sulcus bleeding index
Juloski Retrospective cohort University of Oslo, n 5 144 Past orthodontic On the lower anterior teeth: before treatment/after/5 y
et al 201747 Kingdom of Norway 65 male, 79 female treatment and fixed Gingival recession posttreatment
Orthodontic treatment mandibular retainer Calculus accumulation Mean retention/observation
between 2008 and 2015 (n 5 48; 24 male, 24 period:
female; mean TR1 group:
age 5 12.78 6 1.36 y) 5.24 6 0.66 y
Past orthodontic TR- group:
treatment, no form of 5.28 6 0.53 y
retention in the mandible UnT group:
(n 5 48; 25 6.47 6 0.31 y

age 5 12.39 6 1.52 y)
Untreated subjects
(n 5 48; 16 male, 32
female; mean
age 5 11.73 6 0.36 y)
Al-Nimri Cross-sectional Jordan University of Science n 5 62 In the lower arch: On the lower anterior teeth: .12 mo
et al 200958 and Technology 18 male, 44 female 0.036" Round stainless Plaque index (Silness and Mean retention period:
Between 2003 and 2005 steel wire bonded to the L€
oe 1964) (B/L/M/D) RW group:
canines (n 5 31; 9 male, gingival index (L€oe and 21.31 6 8.97 mo
22 female; mean Silness 1963) (B/L) MW group:
age 5 20.23 6 3.8 y) Oral Hygiene Index (with 19.35 6 6.67 mo
0.015" Multistranded the exclusion of the
wire bonded lower anterior segment)
to all anterior teeth (Greene and Vermillion
(n 5 31; 9 male, 22 1964)
female; mean age 5
19.97 6 4.2 y)
Artun 198451 Cross-sectional Setting and date 1st comparison: n 5 74 First comparison, in the lower Along the wire: First comparison—mean
not reported Sex not reported arch: Plaque along Wire Index retention period:
0.032" Calculus along Wire SW group: 1.4 6 0.7 y
Spiral wire bonded to Index PW group: 5.4 6 2.3 y
Arn et al
the canines (n 5 31; NR group: 2.7 6 0.8 y
mean age 5 Along the gingival margin of
the anterior teeth:
17.10 6 3.11 y)
Arn et al
0.032" Plain wire bonded Plaque index (Silness and Second comparison—mean
to the canines (n 5 18; L€oe 1964) (ID/L) retention period:
mean age 5 19.4 6 3.0 y) Calculus index (Ramfjord SW group: 2.3 6 1.10 y
No fixed retainer 1959) (ID/L) RP group: 1.2 6 0.7 y
(n 5 25; mean Modified gingival index
age 5 17.2 6 0.8 y) (L€oe and Silness 1963)
(ID/L)
Second comparison, in the
Loss of attachment
upper arch: (Ramfjord 1959) (L)
0.0195" Spiral wire Crevice depth
teeth (n 5 14; mean Interdentally:
age 5 21.6 6 3.8 y) Papillitis (NBP system,
Jensen retention plate Dolles and Gjermo 1980)
(n 5 20; mean
age 5 16.10 6 0.9 y)
Booth Cross-sectional One private practice n 5 60 In the lower arch: Gingival index (L€oe and 20-29 y
et al 200852 Date not reported Sex and age not reported Fixed retainer bonded to Silness 1963) of the upper
the canines: majority of and lower arches from first
0.025" stainless steel molar to first molar on the
wire, minority of 0.032" buccal and lingual aspects
multistranded wire
(n 5 45)
No more fixed retainer
(n 5 15)
Cerny Cross-sectional Setting and date n 5 61 In one arch (maxilla or Plaque index (Greene PBR group: .15 y
et al 201056 not reported Sex and age not reported mandible) or in both: and Vermillion 1960) RR group: 2 y
Permanent Calculus index (Greene
bonded retainer (n 5 46) and Vermillion 1960)
Removable retainer Gingival recession index

(n 5 43) Modified gingival index
Subgroup: PBR 1 RR (Lobene 1989)
(n 5 28; Alveolar bone index
the patients who had (Adams and Nystrom
worn both retainers were 1986)
included in both
groups as appropriate)
Neto et al Cross-sectional Setting and date n 5 40 Past orthodontic On the lower anterior teeth 2-8 y
201050 not reported 14 male, 26 female treatment and fixed (MB/B/DB/Ml/L/DL):
Age not reported mandibular retainer Plaque index
(n 5 20; 5 male,
164.e7
Bleeding on probing
15 female) Gingival recession
164.e8

Untreated subjects Clinical attachment level
(n 5 20; 9 male, 11 Probing depth
female)
Corbett et al Cross-sectional One private practice n 5 74 In both arches: 0.022" On the upper and lower 2-4 y
201559 Date not reported 30 male, 54 female round Blue Elgiloy anterior teeth: Mean retention period:
Age range 5 13-22 y wave-type wire bonded plaque index (L€oe 1967) WR group:
to all anterior teeth calculus index (Greene 31.6 6 3.2 mo
(n 5 39; 13 male, 26 and Vermillion 1964) SR group: 42.3 6 2.4 mo
female; mean pocket probing depth
age 5 16.9 6 0.96 y) (MB/B/DB/Ml/L/DL)
Maxilla: 0.0215" bleeding on probing
multistranded gingival recession (B/L)
stainless steel straight
wire bonded to all On teeth 11 and 41:
GCF volume (B/L)
anterior teeth; mandible:
0.0315" multistranded
stainless steel straight

wire bonded to the
canines and central
incisors (n 5 35; 17
age 5 18.3 6 1.3 y)
Dietrich et al Cross-sectional University of Zurich n 5 41 0.016 3 0.016" Stainless On the upper incisors: Retention period:
201453 Date not reported 16 male, 25 female steel wire Plaque index (Silness and Range 5 5 y and 2 mo -
Age not reported bonded to all upper L€
oe 1964) (B/P) 11 y and 7 mo
incisors Gingival index (L€
oe and Mean 5 7 y and 5 mo
Silness 1963) (B/P)
Probing depth (MB/B/
DB/MP/P/DP)
Bleeding on probing
(MB/B/DB/MP/P/DP)
Knaup Cross-sectional University Aachen, Federal n 5 61 In the lower arch: On the lower anterior teeth: Mean retention
et al 201960 Republic of Germany Age range 5 22 - 56 y 0.0175" Twistflex Plaque index (L€oe 1967) period 5 7.2 6 0.8 mo
September 2014 to June Sex not reported stainless steel (B/L/M/D)
2015 wire bonded to all Gingival index (L€
oe
anterior teeth (n 5 31) 1967) (B/L/M/D)
0.014 3 0.014" Probing depth (MB/B/
computer-aided DB/Ml/L/DL)
Arn et al
design/computer-aided
manufacturing
Arn et al

electropolished nitinol Bleeding on probing
wire bonded to all (Ainamo and Bay 1975)
anterior teeth (n 5 30) (MB/B/DB/Ml/L/DL)
Marginal recessions (B/L)
Levin Cross-sectional Military dental clinic of the n 5 92 Past orthodontic On the upper and lower Mean retention
et al 200849 Israeli Defense Forces 46 male, 46 female treatment 1 1 or 2 anterior teeth (6 sites per period 5 4.57 6 2.2 y
May to August 2007 Age range 5 18-26 y; fixed retainer(s) (n 5 25) tooth):
mean age 5 20.6 6 1.7 y Subgroups: Plaque index (Silness and
o Fixed retainer(s) L€
oe 1964)
placed in a Gingival index (L€oe and
more gingival Silness 1963)
position Probing depth
o Fixed retainer(s) Bleeding on probing
placed in a Gingival recession
more incisal
position
No fixed retainer, with or

without past orthodontic
treatment (n 5 67)
Subgroup: past
orthodontic
treatment 1 no fixed
retainer (n 5 39)
Pandis Cross-sectional Long-term group: n 5 64 In the lower arch: On the lower anterior teeth: Retention period:
et al 200761 one private practice 21 male, 43 female 0.195" Multistranded Plaque index (L€oe 1967) LT group: range 5 9 -
Short-term group: wire bonded at least 9 y (B/L) 11 y; mean 5 9.65 y

not reported ago (n 5 32; 11 male, 21 Calculus index (Greene ST group: 3-6 mo
Date not reported female; mean and Vermillion 1960)
age 5 25 6 1.29 y) Gingival index (L€oe
Similar fixed retainer 1967) (M/L/B/D)
bonded for 3-6 mo Probing depth (MB/B/
(n 5 32; 10 male, 22 DB/Ml/L/DL)
female; mean age 5 Marginal recession
16.4 6 1.26 y) Bone level (distance
between the CEJ to the
alveolar crest on
radiographs)
164.e9
164.e10

Rody Cross-sectional Not reported n 5 31 In the lower arch: On the lingual aspect of a Retention period:
et al 201154 17 male, 14 female 0.028" Round stainless lower central incisor and of Range 5 4 - 10 y
Age range 5 20-35 y steel wire bonded to a second premolar: Mean 5 5.6 y
the canines (n 5 10; 3 Plaque accumulation
male, 7 female; mean Probing depth
age 5 28 6 4.9 y) Bleeding on probing
Hawley retainer (n 5 11; GCF collection (volume,
9 male, 2 female; biomarkers of
mean age 5 24 6 3.6 y) inflammation and
No retainer (n 5 10; 5 periodontal remodeling)
age 5 26.9 6 4.22 y)
Rody Cross-sectional University of Florida n 5 36 In the lower arch: On the lower anterior teeth (6 At least 6 mo
et al 201657 Date not reported 13 male, 23 female 0.028" Round smooth sites per tooth): Mean retention period:
Age range 5 18-45 y wire bonded to the Plaque score SW group:
canines (n 5 12; 4 male, Gingivitis score 76 6 63.62 mo
8 female; mean Probing depth MW group:

age 5 28.75 6 7.1 y) Bleeding on probing 44.50 6 26.80 mo
0.016"-0.0195" Round Gingival recession RR group:
multistranded wire 43.67 6 26.85 mo
bonded to all anterior On the lingual aspect of the
teeth (n 5 12; 5 male, tooth 31:
GCF collection (volume,
7 female; mean
age 5 28.75 6 6.9 y) biomarkers)
Removable retainer at
night (n 5 12; 4 male, 8
female; mean age 5
23.25 6 4.69 y)
Westerlund Cross-sectional Clinic of Orthodontics, n 5 62 Past orthodontic On the lower anterior teeth: 10 y
et al 201755 Public Dental Service, 21 male, 41 female treatment 1 0.0195" distance of the
Gothenburg, Mean age 5 27 y multistranded wire cementoenamel junction
Kingdom of Sweden bonded to all anterior to the marginal bone crest
Date not reported teeth in the lower arch (D/M/B/L surfaces)
(n 5 34) measured by cone beam
Past orthodontic computed tomography
treatment without
fixed retainer (n 5 14)
Untreated subjects
Arn et al
(n 5 14)
ID, interdental; L, lingual; GI, gingival index; ML, mesiolingual; DL, distolingual; B, buccal; M, mesial; D, distal; GF, glass fiber; DB, distobuccal; PDL, periodontal ligament; NiTi, nickel titanium;
LDH, lactate dehydrogenase; F-Hb, free hemoglobin; Pg, Porphyromonas gingivalis; Tf, Tannerella forsythia; GCFF, gingival crevicular fluid flow, TR+, orthodontic treatment and fixed retainer; TR-,
orthodontic treatment and no retention; UnT, untreated; RW, round wire; MW, multistranded wire; NBP, non-bleeding papillae; SW, spiral wire; PW, plain wire; NR, no fixed retainer; PBR, per-
manent bonded retainer; RR, removable retainer; CEJ, cementoenamel junction; LT, long-term; ST, short-term.
Arn et al
Supplementary Table II. Results of the included studies assessing the effects of fixed retainers on periodontal health
Effects on
periodontal
Author(s)/y Study design Intervention(s) health Description of periodontal results
Al-Moghrabi RCT In the lower arch: NO No significant difference in gingival inflammation, calculus and
et al 201833 0.0175" Coaxial wire from canine to canine plaque levels, clinical attachment level, and bleeding on
Vacuum-formed retainer, wear time progressively reduced probing between the groups.
Artun et al 198739 RCT In the lower arch: NO No significant difference in plaque accumulation, calculus
0.032" Plain wire bonded to the canines accumulation, and gingival inflammation between the groups.
0.032" Spiral wire bonded to the canines
0.025" Spiral wire bonded to all anterior teeth
Removable retainer
Artun et al 199740 RCT In the lower arch: NO No significant difference in plaque accumulation, calculus
0.032" Plain wire bonded to the canines accumulation, gingival inflammation, and probing attachment
0.032" Spiral wire bonded to the canines level between the groups.
0.025" Spiral wire bonded to all anterior teeth
Removable retainer
Liu 201035 RCT In the lower arch: YES Both retainer types increased the gingival bleeding but did not
0.0295" Fiber-reinforced composite retainer bonded to all influence the pocket depth measurement.
anterior teeth
0.0354" Multistranded stainless steel wire bonded to all
anterior teeth
Storey et al 201836 RCT In both arches: NO The bonded retainers group showed a higher accumulation of
0.0195" Three-stranded twistflex stainless steel wire bonded plaque and calculus and a minimally worse gingival
to all anterior teeth inflammation, but no clinically significant, adverse periodontal
Vacuum-formed retainer at night health problems.
St€
ormann and RCT In the lower arch: NO Plaque accumulation increased with time in all groups, but
Ehmer 200242 0.0195" Customized wire bonded to all anterior teeth gingivitis occurred in sporadic cases.
0.0215" Customized wire bonded to all anterior teeth
Prefabricated retainer bonded to the canines

Tacken et al 201038 RCT In the upper arch, bonded to all incisors, and in the lower arch, YES All retainer groups showed periodontal complications; they
bonded to all anterior teeth: accumulated more plaque and presented a higher level of
Glass fiber-reinforced retainer-500 unidirectional GF gingival inflammation compared with the untreated subjects.
Glass fiber-reinforced retainer-1000 unidirectional GF
0.0215" Multistranded wire
Untreated subjects
Torkan et al 201437 RCT In both arches: YES/NO The fiber-reinforced composite retainer had apparent side effects
Fiber-reinforced composite retainer bonded to all anterior on periodontal health; gingival inflammation and plaque
teeth accumulation increased in both arches; calculus accumulation
0.0175" Multistranded stainless steel wire bonded to all increased only in the mandible.
164.e11
anterior teeth The results for the multistranded wire retainer were less
categorical; gingival inflammation remained unchanged in the
164.e12
Supplementary Table II. Continued
Effects on
periodontal
mandible, although it increased in the maxilla; the plaque
accumulation did not change in both arches, and the calculus
accumulation increased only in the lower jaw.
Xu et al 201143 RCT Combined retainers: 0.0195" multistranded NiTi wire (YES) The calculus deposition was higher with combined retainers than
bonded 3-3 1 Hawley retainer at night with vacuum-formed retainers.
Vacuum-formed retainer, for both arches, worn the whole
day
G€
okçe and Kaya Prospective In the lower arch: YES No significant difference in plaque accumulation, probing depth,
201046 cohort 0.0215" Multistranded wire direct bonded to all anterior and marginal recession between the groups.
teeth The Essix group showed less gingival inflammation than the
0.0215" Multistranded wire indirect bonded to all anterior 0.0215" Indirect group at 1 month and 3 months, than the
teeth 0.0175" Indirect group at 3 months and than the 0.0215"
0.0175" Multistranded wire direct bonded to all anterior Direct group at 6 months. At 1 month, the bleeding on probing
teeth scores was significantly different between the 0.0215" Indirect
0.0175" Multistranded wire indirect bonded to all anterior and Essix groups.
teeth
Removable Essix appliance (1-mm thickness) to wear day
and night
Heier et al Prospective In both arches (majority of participants): NO The groups showed a comparable limited gingival inflammation.
199744 cohort 0.0175" Multistranded wire bonded to all anterior teeth Slightly more calculus and plaque were observed on the lingual
Removable retainer surfaces in the fixed retainer group; for the calculus
deposition, this difference was already present at baseline.
Kaji et al 201345 Prospective cohort In the lower arch: NO No significant difference in plaque accumulation, calculus
Stainless steel wire bonded to all anterior teeth accumulation, gingival inflammation, pocket depth, and GCF
No fixed retainer, no evidence of visual malalignment volume between the groups.
between the lower bicuspids A difference between the groups was present for the change in
elastase activity and protein content in GCF from baseline to
8 weeks. There was no significant difference in salivary
biochemical or bacteriologic test items between the groups.
Wu et al 201448 Prospective Glass fiber-reinforced composite retainer bonded to all lower NO Glass fiber-reinforced composite retainer did not cause gingivitis.
noncontrolled anterior teeth
cohort
Juloski et al Retrospective cohort Past orthodontic treatment and fixed mandibular retainer NO Significantly more calculus accumulation observed in the retainer
201747 Past orthodontic treatment, no form of retention in the group but no significant difference in the development of
mandible gingival recession between the groups.
Untreated subjects
Arn et al
Arn et al
Effects on
periodontal
51
Artun 1984 Cross-sectional First comparison, in the lower arch: NO First comparison:
0.032" Spiral wire bonded to the canines No significant difference in plaque accumulation, calculus
0.032" Plain wire bonded to the canines accumulation, crevice depth, and papillitis between the
No fixed retainer groups. When compared with the combined scores of the fixed
retainer groups, the reference group showed a higher level of
Second comparison, in the upper arch:
gingival inflammation on the lingual surfaces.
0.0195" Spiral wire bonded to all anterior teeth Second comparison:
Jensen retention plate The spiral wire group presented deeper crevice depth in
interproximal areas than the retention plate group, but no
pockets of 4 mm or more were registered; there was no
significant difference in plaque accumulation, calculus
accumulation, gingival inflammation and papillitis between
the groups.
Booth et al 200852 Cross-sectional In the lower arch: NO In the mandible, no significant difference in gingival
Fixed retainer bonded to the canines: majority of 0.025" inflammation between the groups. The group without a fixed
stainless steel wire, minority of 0.032" multistranded wire retainer presented a higher level of gingivitis in the maxillary
No more fixed retainer lingual areas, suggesting possible better hygiene and more
regular recalls for prophylaxis in the patients with a lower fixed
retainer.
Cerny et al 201056 Cross-sectional In one arch (maxilla or mandible) or both: NO The accumulation of dental plaque and calculus was higher in the
Permanent bonded retainer PBR group but no significant difference in gingival recession,
Removable retainer gingivitis, and alveolar bone loss between the groups.
Neto et al 201050 Cross-sectional Past orthodontic treatment and fixed mandibular retainer YES No significant difference in gingival recession and bleeding on
Untreated subjects probing between the groups. However, the treated subjects
showed a higher accumulation of plaque, deeper probing
depths, and greater attachment loss.
Dietrich et al 201453 Cross-sectional 0.016 3 0.016" stainless steel wire bonded to all upper incisors NO 46% of the patients had at least one site with a probing depth of
more than 3 mm; however, plaque index, gingival index, and

bleeding on probing were equal or inferior to demographic and
historical data.
Levin et al 200849 Cross-sectional Past orthodontic treatment 1 1 or 2 fixed retainer(s) YES When compared with either reference groups, fixed retainers were
No fixed retainer, with or without past orthodontic associated with a higher incidence of plaque retention,
treatment bleeding on probing and gingival recession on the anterior
Subgroup: past orthodontic treatment 1 no fixed lingual surfaces. No significant difference in probing depth
retainer and buccal gingival recession between the groups. Results for
the gingival index not reported.
Pandis et al 200761 Cross-sectional In the lower arch: YES/NO The long-term retention group showed higher calculus
0.195" Multistranded wire bonded at least 9 years ago accumulation, greater marginal recession, and increased
164.e13
Similar fixed retainer bonded for 3-6 months probing depth.
There was no significant difference in plaque accumulation,
gingival inflammation and bone level between the groups.
164.e14
Effects on
periodontal
Rody et al 201154 Cross-sectional In the lower arch: NO The fixed retainer group presented more plaque accumulation in
0.028" Round stainless steel wire bonded to the canines the incisor region; however, no significant difference in
Hawley retainer probing depth, bleeding on probing, and GCF volume between
No retainer the groups.
The concentration of MMP-9 was higher in the incisor sites of the
fixed retainer group, which could suggest a subclinical
inflammation, but further studies are needed to validate MMP-
9 as a prognostic factor for periodontitis in healthy individuals.
The concentrations of IFN-Y and IL-10 were higher in the
premolar sites of the removal retainer group.
Rody et al 201657 Cross-sectional In the lower arch: NO The multistranded wire group exhibited higher plaque
0.028" Round, smooth wire bonded to the canines accumulation. Compared with the smooth wire group, this
0.016"-0.0195" Round multistranded wire bonded to all group also showed increased gingivitis. No significant
anterior teeth difference in probing depth, bleeding on probing, gingival
Removable retainer at night recession, and GCF volume between the groups.
A significant difference was found between the groups for the
level of RANKL, OPG, OPN, M-CSF, MMP-3, and MMP-9. The

fixed retainer groups presented a higher RANKL/OPG ratio
than the removable retainer group. The association between
retention groups and GCF biomarker levels should be further
explored in longitudinal studies.
Westerlund et al 201755 Cross-sectional Past orthodontic treatment 1 0.0195" multistranded wire NO No significant difference in marginal bone levels between the
bonded to all anterior teeth in the lower arch orthodontically treated patients with a fixed retainer, and the
Past orthodontic treatment without a fixed retainer orthodontically treated patients without a fixed retainer.
Untreated subjects
GCF, gingival crevicular fluid; PBR, permanent bonded retainer; MMP, matrix metallopeptidase; IFN-g, interferon-gamma; IL, interleukin; RANKL, receptor activator of nuclear factor kappa-B
ligand; OPG, osteoprotegerin; OPN, osteopontin; M-CSF, macrophage colony-stimulating factor.
Arn et al
Arn et al
Supplementary Table III. Results of the included studies comparing different types of fixed retainers
Author(s)/y Study design Interventions Effects on periodontal health Description of periodontal results
Artun et al 198740 RCT In the lower arch: 5 No significant difference in plaque
0.032" Plain wire bonded to the canines accumulation, calculus accumulation and
0.032" Spiral wire bonded to the canines gingival inflammation between the groups.
0.025" Spiral wire bonded to all anterior
teeth
Artun et al 199740 RCT In the lower arch: 5 No significant difference in plaque
0.032" Plain wire bonded to the canines accumulation, calculus accumulation,
0.032" Spiral wire bonded to the canines gingival inflammation and probing
0.025" Spiral wire bonded to all anterior attachment level between the groups.
teeth
Bazargani et al 201234 RCT In the lower arch: Two-Step bonding \ Composite only Application of resin during bonding
0.0195" Multistranded wire bonded to all resin1 composite procedure reduced calculus accumulation.
anterior teeth with composite only
0.0195" Multistranded wire bonded to all
anterior teeth with 2-step bonding resin
and composite
Liu 201035 RCT In the lower arch: 5 No significant difference in gingival bleeding
0.0295" Fiber-reinforced composite and pocket depth measurement between
retainer bonded to all anterior teeth the groups.
0.0354" Multistranded stainless steel
wire bonded to all anterior teeth
Rose et al 200241 RCT In the lower arch: 5 No significant difference in amount of

0.0393" Plasma-treated woven calculus between the groups.
polyethylene ribbon bonded to all
anterior teeth
0.0175" Multistranded stainless steel
wire bonded to all anterior teeth
St€
ormann and RCT In the lower arch: 5 No significant difference in plaque
Ehmer 200242 0.0195" Customized wire bonded to all accumulation between the groups.
anterior teeth
0.0215" Customized wire bonded to all
anterior teeth
Prefabricated retainer bonded to the
canines
Tacken et al 201038 RCT In the upper arch, bonded to all incisors, and Multistranded wire \ GFR500 retainer 5 Both glass fiber-reinforced retainers induced
in the lower arch, bonded to all anterior GFR1000 retainer a higher level of gingival inflammation
teeth: than the multistranded wire retainer,
Glass fiber-reinforced retainer - 500 although plaque accumulation did not
164.e15
unidirectional GF differ between the groups.
Glass fiber-reinforced retainer - 1000 No significant difference between the 2 GFR
unidirectional GF groups for any outcomes.
0.0215" Multistranded wire
164.e16
Supplementary Table III. Continued
Torkan et al 201437 RCT In both arches: Multistranded \ FRC retainer In both arches, the fiber-reinforced composite
Fiber-reinforced composite retainer wire retainer retainer caused a higher level of gingival
bonded to all anterior teeth inflammation and plaque accumulation
0.0175" Multistranded stainless steel than the multistranded stainless steel wire
wire bonded to all anterior teeth retainer.
G€
okçe and Prospective cohort In the lower arch: 5 No significant difference in plaque index,
Kaya 201046 0.0215" Multistranded wire direct gingival index, probing depth, marginal
bonded to all anterior teeth recession, and bleeding on probing
0.0215" Multistranded wire indirect between the groups.
bonded to all anterior teeth
0.0175" Multistranded wire direct
0.0175" Multistranded wire indirect
Kaji et al 201345 Prospective cohort Wire within the upper one-third of 5 No significant difference in gingival
clinical crown inflammation, pocket depth, GCF volume,
Wire within the lower two-thirds of calculus accumulation and plaque
clinical crown accumulation between the groups.

No significant difference in GCF biochemical
test items between the groups. A significant
difference between the groups was present
for the changes in salivary F-Hb
concentration from baseline to 4 weeks
after and from baseline to 8 weeks after.
Al-Nimri et al Cross-sectional In the lower arch: z No significant difference in gingival condition
200958 0.036" Round stainless steel wire bonded between the groups. The multistranded
to the canines wire group showed more plaque on the
0.015" Multistranded wire bonded to all distal surfaces of the lower anterior teeth
anterior teeth than the round wire group.
Artun 198450 Cross-sectional In the lower arch: z The plain wire group accumulated more
0.032" Spiral wire bonded to the canines plaque and calculus gingivally along the
0.032" Plain wire bonded to the canines wire than the spiral wire group; however,
no significant difference in plaque and
calculus accumulation along the gingival
margin. No significant difference in
gingival inflammation, crevice depth and
papillitis between the groups.
Arn et al
Arn et al
Supplementary Table III. Continued
Corbett et al 2015 59
Cross-sectional In both arches: 0.022" round Blue Elgiloy z No significant difference in calculus
wave-type wire bonded to all anterior accumulation, gingival inflammation,
teeth gingival recession and GCF volume
Maxilla: 0.0215" multistranded stainless between the groups. Small difference in
steel straight wire bonded to all anterior plaque accumulation (teeth 23 and 33 with
teeth; mandible: 0.0314" multistranded more plaque in the straight wire group) and
stainless steel straight wire bonded to the pocket depth (tooth 41 with a deeper
canines and central incisors pocket measurement in the wave-type wire
group).
Knaup et al 201960 Cross-sectional In the lower arch: Nitinol wire \ Twistflex stainless The nitinol wire was associated with better
0.0175" Twistflex stainless steel wire retainer steel wire retainer periodontal health indices (plaque index,
bonded to all anterior teeth gingival index, probing depth, and
0.014 3 0.014" computer-aided design/ bleeding on probing) than the twistflex
computer-aided manufacturing stainless steel wire.
electropolished nitinol wire bonded to all
anterior teeth
Levin et al 200849 Cross-sectional Fixed retainer placed in a more gingival 5 No significant difference between the groups.
position
Fixed retainer placed in a more incisal
position
Rody et al 201657 Cross-sectional In the lower arch: Smooth wire \ Multistranded The multistranded wire exhibited a higher
0.028" Round, smooth wire bonded to retainer wire retainer level of plaque accumulation and gingivitis
the canines than the smooth wire. No significant
0.016"-0.0195" Round multistranded difference in probing depth, bleeding on
wire bonded to all anterior teeth probing, gingival recession, and GCF
volume between the groups.
The symbol “5” means “same effects,” the symbol “\” means “fewer effects than,” and the symbol “z” means “effects quite similar.”
GF, glass fiber; GFR, glass fiber retainer; FRC, fiber-reinforced composite; GCF, gingival crevicular fluid; F-Hb, free hemoglobin.
164.e17

Fixed Retainers

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Fixed Retainers

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fixed Retainers

Uploaded by

Copyright:

Available Formats

SYSTEMATIC REVIEW

The effects of ﬁxed orthodontic

Records identified through database searching:

(n = 168) 69 duplicates removed

Full-text articles excluded,

Fig. Studies' Flow diagram.

Table I. Risk of bias of the included RCTs

Bazargani et al34 reported that the application of

wires to multistranded wires. Three of them did not

Moderate risk observe any difference in gingival conditions between

the 2 types of retainers, whereas Knaup et al60 and

Moderate risk for the outcomes

smooth wire retainers. Artun51 described an increased

accumulation of plaque and calculus along the wire in

the plain wire group, whereas the other cross-sectional

studies reported more plaque deposition in the

multistranded wire group. Corbett et al59 observed that

the wave-type wire was related to a higher frequency

One prospective cohort study45 and 1 cross-sectional

study49 observed that positioning a ﬁxed retainer more

Other studies. A cross-sectional study61 compared

the short- and long-term effects of ﬁxed retainers on

periodontal health. The long-term retention group

difference in age between the 2 groups could be the

cause of these ﬁndings and not the longer presence of

This systematic review aimed to evaluate the

potentially deleterious effects of ﬁxed retainers on

periodontal health. As a secondary aim, different kinds

nonrandomized studies fulﬁlled the inclusion criteria.

Most of the included studies seem to indicate that

studies49,50 reported poorer periodontal conditions in

the presence of a ﬁxed retainer. One RCT37 showed

that ﬁber-reinforced composite retainers could be asso-

clear results for multistranded wire retainers. However,

were reported in any of the included studies.

Table III. Risk of bias of the included cross-sectional studies

teeth (n 5 30) index

evidence of visual Silness 1963)

male, 23 female; mean

Removable retainer Gingival recession index

Study Setting Sample size, sex,

stainless steel straight

Study Setting Sample size, sex,

No ﬁxed retainer, with or

Short-term group: wire bonded at least 9 y (B/L) 11 y; mean 5 9.65 y

Study Setting Sample size, sex,

8 female; mean Probing depth MW group:

Prefabricated retainer bonded to the canines

more than 3 mm; however, plaque index, gingival index, and

level of RANKL, OPG, OPN, M-CSF, MMP-3, and MMP-9. The

Rose et al 200241 RCT In the lower arch: 5 No signiﬁcant difference in amount of

clinical crown accumulation between the groups.

You might also like