SYSTEMATIC REVIEW

Effects of lip bumper therapy on the

mandibular arch dimensions of children
and adolescents: A systematic review
Lucas Garcia Santana,a Esdras de Campos França,b Carlos Flores-Mir,c Lucas Guimara~es Abreu,b
a ^
Leandro Silva Marques, and Paulo Antonio Martins-Junior b

Diamantina and Belo Horizonte, Minas Gerais, Brazil, and Edmonton, Alberta, Canada

Introduction: The aim of this systematic review was to identify, evaluate, and provide a synthesis of the avail-
able literature on the effects of lip bumper (LB) therapy on the mandibular dental arch of children and adoles-
cents. Methods: MEDLINE, Scopus, Web of Science, Cochrane Library, and Lilacs were systematically
searched without restrictions up to May 2019. Risk-of-bias assessment was performed using Cochrane's tool
for randomized controlled trials (RCTs) and the Risk of Bias in Nonrandomized Studies of Interventions tool
for non-RCTs. The Grading of Recommendations, Assessment, Development and Evaluation tool was used
to assess the quality of the evidence. Results: After examination of the full texts, 6 studies were included.
One RCT presented unclear risk of bias, and 5 non-RCTs presented serious to moderate risk of bias. LB
therapy resulted in a buccal inclination of the incisors, distalization of the permanent first molars, and distal
inclination of the permanent first molars, which increased perimeter and arch length. An increase in the arch
width with greater gain in the interpremolar and/or deciduous molar distance and less gain in intercanine and
intermolar distances was also reported. LB therapy increased the risk of second molar impaction with
inclination .30
and the risk of ectopic eruption when treatment time was .2 years. The level of the
evidence was graded as very low for variable arch length and second molar eruption disturbances. All other
outcomes were graded as having low level of evidence. Conclusions: Owing to the low level of certainty iden-
tified, the conclusions should be considered cautiously. Increase in arch perimeter and width was attributed to
the proclination of the incisors, buccalization of the deciduous molar and premolar areas, and distal inclination of
the molars. However, there was an increased chance of impaction and ectopic eruption of permanent
second molar after treatment with LB. (Am J Orthod Dentofacial Orthop 2020;157:454-65)

L
ip bumper (LB) therapy may represent a manage- anterior teeth and the concomitant distal forces
ment alternative for the resolution of future space exerted on the permanent first molars (M1). Hence, the
deficiency in the mandibular dental arch reducing therapeutic effect of the LB will occur by the labial
the necessity for tooth extractions.1-3 The primary displacement of the incisors and the distal inclination
objective of LB therapy is to reduce anterior dental of the M1.4-6
crowding by increasing the length and width of the A few studies suggest that the LB can maintain the
mandibular dental arch. These alterations can be position of the M1 or distalize it, preserving or increasing
attributed to the removal of lip pressure on the the leeway space.7-9 However, if the LB distalizes the M1
crown while assisting in crowding resolution, it also
a
reduces the available distal space, altering the
Department of Pediatric Dentistry and Orthodontics, Federal University of Vales
do Jequitinhonha e Mucuri, Diamantina, Minas Gerais, Brazil.
physiological eruption of the permanent second molars
b
Department of Pediatric Dentistry and Orthodontics, Federal University of Minas (M2) and the available space for them.10,11 This may
Gerais, Belo Horizonte, Brazil.
c
leave many orthodontists reluctant to use this therapeu-
Department of Orthodontics, University of Alberta, Edmonton, Alberta, Canada.
All authors have completed and submitted the ICMJE Form for Disclosure of Po-
tic approach.
tential Conflicts of Interest, and none were reported. To our knowledge, only 1 systematic review12 has
Address correspondence to: Leandro Silva Marques, Department of Pediatric evaluated the effects of LB therapy on mandibular dental
Dentistry and Orthodontics, Federal University of Vales do Jequitinhonha e Mu-
curi, Diamantina, MG - 39100000, Brazil; e-mail, [email protected].
arch dimensions under 3 aspects (ie, distalization and/or
Submitted, June 2019; revised and accepted, October 2019. distoangulation of M1, labial inclination of the mandib-
0889-5406/$36.00 ular incisors, and arch width changes). However, that re-
Ó 2019 by the American Association of Orthodontists. All rights reserved.
https://doi.org/10.1016/j.ajodo.2019.10.014
view did not consider the effect on the eruption of M2,

454
Santana et al 455

and only 1 study was included at that time, possibly the Controlled-trials Database of clinical trials (http://
because of language restriction and exclusion of uncon- www.controlled-trials.com), the Clinical Trials-US Na-
trolled studies. No analysis of risk of bias and level of cer- tional Institute of Health (http://www.clinicaltrials.
tainty supporting the conclusions were considered. gov), and the National Institute for Health and Clinical
Additional studies on LB therapy have been published Excellence (http://www.nice.org.uk) were consulted to
in recent years. check for possible ongoing studies. There was no restric-
For these reasons, the aim of this systematic review tion of language or year or status of publication for in-
was to identify, evaluate, and provide a synthesis of clusion.
the available literature on the effects of LB therapy on
the mandibular dental arch of children and adolescents. Search strategy
The search strategy used was as follows: lip bumper
MATERIAL AND METHODS OR lip-bumper OR lip bumpers OR buccal shield OR
Protocol and registration buccal shields. The search strategy was originally
The report of this systematic review followed the planned for PubMed and subsequently adapted for the
guidelines of the Preferred Reporting Items for System- other databases (Supplementary Table).
atic Review and Meta-Analyses statement.13 The study
protocol was registered on the International Prospective Study selection
Register of Systematic Reviews (CRD4201911937). The selection of the studies consisted of 2 phases.
During the first phase, 2 authors (LGS and ECF) indepen-
Eligibility criteria dently examined the titles and/or abstracts. Those refer-
For this systematic review, a set of questions ences that met the eligibility criteria were included. Full
following the population, intervention, comparison, texts of references with insufficient information in the
outcomes, and study design structure was established: title and/or abstract for a decision on inclusion or exclu-
sion were retrieved for evaluation in phase 2. During the
(1) Population: children and adolescents (under the second phase of article selection, the same authors inde-
age of 18 years) with mixed or permanent denti- pendently evaluated the full texts. Those studies that
tion. met the eligibility criteria were included. In both phases,
(2) Intervention: orthodontic treatment with LB in the differences were resolved by consensus. If necessary, a
mandibular arch. third author decided whether or not to include the study
(3) Comparison: control group of individuals who were (PAMJ). For the second phase, the reasons for exclusion
not submitted to any treatment, a group of individ- of the studies were also recorded.
uals submitted to orthodontic treatment with other
appliances, or measurements before LB. Data extraction and items extracted
(4) Primary outcome: arch perimeter, arch width, arch
The data extraction of the included articles was per-
length, incisor angulation, distalization and tip of
formed independently and in duplicate by 2 authors
the M1 and stability or relapse of those changes;
(LGS and ECF). A standardized table was used to extract
Secondary outcome: M2 eruption disturbances.
the data. The following data were extracted: authors,
(5) Study design: RCT or non-RCT.
year of publication, study setting, description of groups
The exclusion criteria were case reports, review articles, (sex and age of patients, control group, sample size,
abstracts and discussions, animal studies, studies with or- dentition at therapy onset), characteristics of the LB
thodontic and/or orthopedic approaches performed used, and outcomes. Data were compared for accuracy,
concomitant with LB in the mandibular arch or maxillary and any discrepancy was resolved through the re-
arch, treatment with tooth extraction, or any other surgi- examination of the original study until a consensus
cal procedure, studies evaluating individuals with cranio- was reached.
facial deformities, syndromes, cleft lip palates, and studies
with a sample smaller than 10 individuals. Assessment of bias risk within studies
The risk of bias in RCTs was assessed using the Co-
Information sources chrane tool.14 The following items were evaluated: (1)
Electronic searches in MEDLINE (via PubMed and random sequence generation, (2) allocation conceal-
OVID), Web of Science, Cochrane Library, Scopus, and ment, (3) blinding of participants and personnel, (4)
Lilacs were conducted up to May 2019. In addition, blinding of the outcome assessor, (5) incomplete

American Journal of Orthodontics and Dentofacial Orthopedics April 2020  Vol 157  Issue 4
456 Santana et al

outcome data, (6) selective outcome reporting, and (7) Synthesis of results
other sources of bias. To evaluate the last item of this Data collected were synthetized in a descriptive table.
tool, the authors considered the following characteristics A meta-analysis was planned if there was relative homo-
as sources of bias: absence of sample size calculation; geneity among included studies.
inclusion and exclusion criteria incompatible with the
objectives of the study; inadequate statistical test; RESULTS
absence of standard deviation and confidence interval
Study selection
(CI); and observation period incompatible with the ob-
jectives of the study. For each item, the included study The search strategy yielded a total of 553 studies
could be awarded with low risk of bias (plausible bias from the electronic databases. After the removal of du-
that would not seriously alter results), unclear risk of plicates and application of the eligibility criteria, 63 arti-
bias (plausible bias that raises some doubts about the re- cles were considered for full-text evaluation. The reasons
sults), or high risk of bias (plausible bias that seriously for excluding studies after full-text assessment are pro-
impacts the confidence in results). vided in Table I. At the end of the final eligibility evalu-
The risk of bias of non-RCTs was assessed using the ation phase, only 6 articles were included in this
Risk of Bias in Nonrandomized Studies of Interventions systematic review.3,5,10,17-19 A complete search
tool.15 The following domains were evaluated: bias for flowchart is provided in Figure 1.
confounding factors, selection bias, intervention bias,
bias for lack of data, and bias in the measurement of Study characteristics
outcomes and selective reporting. Bias in each domain Table II provides the descriptive characteristics of the
and the risk of overall bias for the non-RCTs were judged 6 studies included in this systematic review. Only 1 study
as low, moderate, serious, critical, or no information. was an RCT,3 whereas 5 studies were non-RCTs.5,10,17-19
A total of 34 individuals participated in the RCT.3
Evaluation of the level evidence (risk of bias across Regarding the non-RCTs,5,17-19 175 individuals
studies) participated in 4 studies in which a primary outcome
The level of evidence was assessed using the Grading of this systematic review was evaluated. A total of 395
of Recommendations, Assessment, Development and individuals participated in the study10 in which a sec-
Evaluation (GRADE) Pro software (GRADEpro Guideline ondary outcome was evaluated.
Development Tool, available online at gradepro.org).16 The mean age of participants at baseline ranged from
For each outcome examined, the GRADE assesses the 10 to 12 years.5,10 In 2 studies, participants began treat-
number of studies included, the studies' designs, risk of ment in the mixed dentition.3,18 In 1 study, participants
bias, inconsistency, indirectness, imprecision, and other began treatment in mixed and permanent dentition.19 In
considerations (such as publication bias). Depending on 1 study, the criterion for participants to begin LB therapy
the seriousness of the limitation in each one of these do- was the noneruption of the permanent mandibular
mains, the evidence could be downgraded by 1 or 2 levels. M2.10 In 2 studies, only the age of the participants
Based on this assessment, the certainty of the evaluation beginning treatment was provided.5,17 The mean time
of the outcome could be very low, low, moderate, or high
quality. The level of evidence was assessed separately for
clinical and observational studies, and the higher level of Table I. Number of excluded studies with reasons af-
certainty was reported. When the level was the same for ter full-text evaluation
evidence from clinical trials and observational studies,
Reasons for exclusion Number of studies
the same outcome was presented in 2 different lines, sum-
Description of clinical technique 4
marizing the findings from the separate sets of evidence Use of other appliances and/or 23
for each class of studies. techniques in conjunction
with LB therapy
Patent registration 3
Summary measurements
Abstract in conference proceedings 4
Measurements for the primary outcome were based Case reports or case series 10
on continuous data (millimeters or degrees from dental Investigation not relevant to 9
the subject of this study
casts, tomography, or radiographies), but measurements
No appropriate data provided 1
for the secondary outcome were based on nominal or Narrative reviews 2
ordinal data (impaction or lack of space from dental Commentaries 1
casts or radiographies). Sum 57

April 2020  Vol 157  Issue 4 American Journal of Orthodontics and Dentofacial Orthopedics
Santana et al 457

Identification
Records iden fied through electronic databases searching: Ovid/Medline
(n=91), PubMed (n=109), Web of Science (n=114),
Screening Cochrane (n=17), Lilacs/Bireme (n=117), SCOPUS (n=105)

Records screened
(n=553)

Records excluded based on

tle and available abstract,
and duplicate (n=490)

Studies considered for full text

Eligibility

reading (n=63)

Full-text ar cles excluded

because of not matching the
inclusion criteria (n=57)

Studies included in
Included

qualita ve synthesis
(n=6)

Studies included in
Analyzed

quan ta ve synthesis (meta-

analysis)
(n=0)

Fig 1. Preferred reporting items for systematic review and meta-analyses flowchart of article retrieval.

interval between onset and end of LB therapy varied In the RCT study (Fig 2), domains of random
substantially and ranged from 6 to 28 months.3,10 sequence generation and allocation concealment were
In most studies, the LB with lip shield in the region of not applicable and were set as unclear by default.3 The
incisors and canines, with coverage of acrylic5 or plastic domain blinding of participants was considered as hav-
was used.3,17 Other types included LB with acrylic in the ing a high risk of bias, even though it is not possible to
region of canines and premolars and a plastic tube bias the participant in this type of intervention. No infor-
placed at the anterior segment extending between the mation was provided regarding blinding of outcome
2 canines,18 LB covered with plastic tube,17 and LB assessment. The other bias domain received a high-risk
with wire covered with a layer of plastic shrink tube.5 classification because no sample size calculation was re-
In 2 studies, details of the LB model used were not pro- ported. Overall, this study was identified as having a high
vided.10,19 risk of bias.
For the non-RCT (Table III), 3 studies5,10,18 received a
Risk of bias within studies serious risk-of-bias assessment and 2 studies17,19
The methodological appraisal of the included studies received moderate risk-of-bias assessment in the domain
is reported in Figure 2 and Table III. confounding. The domain deviation from intended

American Journal of Orthodontics and Dentofacial Orthopedics April 2020  Vol 157  Issue 4
April 2020  Vol 157  Issue 4

458
Table II. Summary of study characteristics and results of the included studies
Results, P value
Study design, Descriptions of LB characteristics and
Authors, year local setting Age, dentition groups (n) intervention protocol Measurements investigated D1 D2 P value
Mandibular arch changes
Nevant et al (1991)5 non-RCT, POC G1: 11 y and G1: patients with G1: LB with wire covered with M1 apex movement (mm) 1.2 6 2.1 1.2 6 2.1 NS
G2: 12 y, NR mandibular arch a layer of plastic shrink M1 cusp movement 0.1 6 1.0 1.5 6 1.9 NS
length deficiency tubing was activated at the M1-occlusal plane (
) 2.7 6 4.9 8.0 6 8.3 \0.05*
(4-8 mm) were adjustment loops to remain Total arch length (mm) 2.6 6 2.6 7.4 6 4.2 \0.05*
treated with LB approximately 2-3 mm in C-C (mm) 1.3 6 0.7 2.8 6 2.9 \0.05*
(n 5 20) front of the mandibular 1PM-1PM (mm) 2.0 6 0.9 4.7 6 2.4 \0.05*
G2: patients with same incisors at the level of the M1-M1 (mm) 0.7 6 1.5 4.2 6 0.8 \0.05*
characteristics as gingiva. It was set 4-5 mm
G1, were treated away from the buccal
with another type of segments and expanded
LB (n 5 20) approximately 2 mm at the
M1. The mean treatment
time was 1.4 y.
G2: LB with an acrylic from C
to C was placed
American Journal of Orthodontics and Dentofacial Orthopedics

approximately 2 mm in
front of the mandibular
incisors; vertically, the top
of the shield was positioned
7 mm from the incisal edge.
The LB was placed, on
average, 4 mm away from
the 1PM and 4-5 mm away
from the M1. The mean
treatment time was 1 y.
Davidovitch RCT, NR 10.2 y, mixed G1: white ethnicity LB with plastic shield from C M1 angulation CT (
) 6.3 6 1.2 2.1 6 1.3 \0.05*
et al (1997)3 patients, with to C was used and M1 angulation Cph (
) 3.3 6 3.6 0.7 6 1.7 NS
3-8 mm mandibular positioned approximately M1 movement of Cres CT 1.6 6 0.5 0.6 6 0.5 \0.05*
arch length 1.5-2-mm labial to the (mm)
deficiency, presence gingival third of the M1 movement of Cres Cph 0.6 6 1.1 0.3 6 0.7 NS
of the mandibular mandibular incisors. The (mm)
deciduous M2, and appliance was inserted in a Incisor inclination (
) 3.1 6 2.4 0.1 6 1.7 \0.05*
Class I malocclusion passive state, and Incisor apex movement (mm) 0.6 6 0.5 0.2 6 0.5 NS
treated with LB continuous wear was M2-M2 deciduous (mm) 1.8 6 1.3 0.3 6 0.6 \0.01*
(n 5 16) assured by ligating the LB C-C (mm) 1.8 6 0.4 0.2 6 0.9 \0.01*
CG: same to the mandibular M1 Perimeter (mm) 4.1 6 2.0 1.7 6 1.3 \0.01*

Santana et al
characteristics as bands. The LB were used for Arch length (mm) 2.1 6 0.8 1.1 6 1.0 \0.01*
G1, untreated 6 mo.
(n 5 18)
 American Journal of Orthodontics and Dentofacial Orthopedics

Santana et al
Table II. Continued

Results, P value
Study design, Descriptions of LB characteristics and
Authors, year local setting Age, dentition groups (n) intervention protocol Measurements investigated D1 D2 P value
Ingervall non-RCT, NR G1:10.9 y G1: patients were LB was anchored in buccal Lower incisor:
and Th€ uer (1998)17 and G2: 10.8 y, treated with a LB, tubes on mandibular M1 Labial surface (
) 0.5 0.4 \0.001y
NR which was covered and adjusted and Incisal edge (
) 0.8 0.7 \0.001y
with plastic tubing positioned 6 mm below the Inclination (
) 2.0 3.0 \0.001y
(n 5 20) edges of the mandibular Arch length (mm) 1.3 2.2 \0.001y
G2: patients were incisors, and to lie 3 mm
treated with a LB, away from the labial
which had an acrylic surfaces of the incisors,
shield in the labial C and from the buccal
fold below canine surfaces of the premolars.
and incisors The LB were used for 8 mo.
(n 5 19)
Solomon et al (2006)19 non-RCT, POC 11.6 y, mixed G1: patients treated Prefabricated LB was adjusted C-C (mm) 2.1 — \0.001z
and permanent nonextractionwith a to rest in the vestibule at the 1PM (mm) 4.4 — \0.001z
LB (n 5 51) level of the free gingival 2PM (mm) 4.1 — \0.001z
margins 1-2 mm facial to M1 (mm) 3.8 — \0.001z
the teeth and tied in the Arch length (mm) 4.5 — \0.001z
buccal tube of the M1 to
continuous wear. Average
time of LB use was 16.40
(64.99) mo.
Moin and non-RCT, POC NR, mixed G1: consecutively LB with acrylic in the region of C-C cusp tip (mm) 2.4 6 2.0 0.6 6 0.9 \0.001*
Bishara (2007)18 patients treated canines and premolars, with C-C cervical (mm) 2.3 6 1.8 — —
with LB by the same welded hooks to tie the LB. M1 deciduous cusp (mm) 5.0 6 2.2 0.1 6 0.7 \0.001*
orthodontist. A piece of plastic tubing M1 deciduous cervical (mm) 4.7 6 2.6 — —
Inclusion criteria was slipped onto the wire M1 deciduous fossae (mm) 3.4 6 2.2 0.4 6 0.8 \0.001*
were the presence of and placed at the anterior M2 deciduous cervical (mm) 3.7 6 2.0 — —
at least 4 segment extending M1-M1 permanent fossae 2.4 6 2.6 — —
April 2020  Vol 157  Issue 4

mandibular incisors between the C. LB was (mm)

and 2 permanent adjusted 2 mm away from M1-M1 permanent cervical 3.1 6 2.4 — —
M1 (n 5 45) the incisor area and 4, 8, (mm)
CG: mean values and 2 mm wider in the C, Arch length (mm) 1.6 6 2.1 — —
obtained from 2PM, and M1 areas, IMPA (
) 1.2 6 4.1 — —
untreated patients respectively. The mean
from another study treatment time was 15.7
(Moorrees, 1959)24 (67.6) mo.

459
April 2020  Vol 157  Issue 4

460
Table II. Continued

Results, P value
Study design, Descriptions of LB characteristics and
Authors, year local setting Age, dentition groups (n) intervention protocol Measurements investigated D1 D2 P value
Permanent M2 eruption disturbances
Ferro et al (2011)10 non-RCT, POC 10.2 y, NR G1 patients with 2 mm LB was kept gingival in the Impaction of M2: 0.027*
or more of anterior vertical plane, and a Bilateral n (%) 9 (3.5) 1 (0.7)
crowding treated distance of 1-2 mm from Unilateral n (%) 9 (3.5) 1 (0.7)
with LB. Exclusion the incisor was kept in the None n (%) 242 (93) 133 (98)
criteria were sagittal plane. Patients Ectopic eruption of M2: \0.001*
permanent M2 could remove the LB by Bilateral n (%) 15 (6.0) 1 (0.8)
eruption and themselves but were asked Unilateral n (%) 26 (10.4) 1 (0.8)
possible factors to wear it 24 h/d, taking it None n (%) 210 (83) 132 (98)
predisposing or out only for meals. The OR for M2 impaction 9 (2-45) 1 0.007*
impeding M2 mean treatment time was OR for M2 ectopic 18 (4-82) 1 \0.001*
impaction, such as 28 mo. OR for M2 impaction 2 (0-7) — NS
agenesis, dental (treatment duration, .2 y)
inclusions, OR for M2 ectopic (treatment 2.6 (1-6) — 0.04z
destroying caries, duration, .2 y)
and previous dental Inclination of M2 for
American Journal of Orthodontics and Dentofacial Orthopedics

extractions impaction
(n 5 260) \10
(OR) 1 —
CG: same 21
-30
(OR) 1 (0-4) — NS
characteristics as .30
(OR) 10 (2-43) — 0.001z
G1, untreated Inclination of M2 for
(n 5 135) ectopic
\10
(OR) 1 —
21
-30
(OR) 1 (0-2) — NS
.30
(OR) 3 (0-13) — NS

POC, Private orthodontic clinic; G1, group 1; G2, group 2; C, canine; NR, not reported; 1PM, first premolar; NS, not significant; CG, control group; CT, computer tomography; Cph, cephalometric;
Cres, center of resistance; 2PM, second premolars; OR, odds ratio; IMPA, incisor mandibular plane angle.
Note. Negative values indicate distal tip, lingual tip, distal movement, or decrease of linear measurements. Positive values indicate mesial tip, buccal tip, mesial movement, or increase of linear mea-
surements. D1 is the difference between baseline and G1 treatment and D2 is the difference between baseline and G1 treatment, or baseline and CG follow-up.
*Concerning D1-D2 difference.
y
Concerning baseline-D1 and baseline-D2.
z
Concerning baseline-D1.

Santana et al
Santana et al 461

Results of individual studies

A summary of the variables assessed in the included
studies can be found in Table II.
In the RCT, a substantial increase in the arch perim-
eter (4.1 6 2.0 mm) in individuals treated with LB
compared with untreated individuals (1.7 6 1.3 mm)
was observed.3 In the same study, an increase in the
length of the mandibular arch of the individuals treated
with LB (2.1 6 0.8 mm) was also observed, whereas
among the untreated individuals this measure was
reduced (1.1 6 1.0 mm). In 4 other studies,5,17-19
the increase in the length of the mandibular arch was
confirmed, with the largest increase (7.4 mm),5 achieved
by individuals who had worn LB with canine-to-canine
acrylic, for an average time of 1 year.
Arch width was evaluated through the intercanine
distance,3,5,18,19 interpremolar width,5,19 first and sec-
Fig 2. Summary of the risk-of-bias assessment accord- ond deciduous molars interarch width,3,18 and the inter-
ing to the Cochrane collaboration tool, used to record molar distance.5,18,19 All individuals treated with LB
the risk of bias of 1 article classified as an RCT.
presented an improvement in these measurements,
with the most substantial improvement in the area of de-
intervention (at intervention) and missing data pre- ciduous molars and their succeeding premolars (ranging
sented low risk of bias in all studies. Therefore, the over- from 3.4 to 5.0 mm). Compared with untreated individ-
all risk of bias was considered moderate in 2 studies17,19 uals, there was an improvement of 2.0 mm (P \0.05) in
and serious in 3 studies,5,10,18 indicated by poor report- the intercanine distance after 6 months of LB therapy.3
ing and experimental design. Three studies3,17,18 evaluating the mandibular incisor
position indicated that there was an increased buccal
inclination, with inclination up to 3.1
in the treated
Synthesis of results group of individuals vs 0.1
in the untreated individ-
Because of the substantial amount of clinical, uals.3 There was no significant linear movement from
methodological, and statistical heterogeneity, a meta- the apex of the incisor to the vestibular (0.6 mm).3 In
analysis was not justifiable. Identified sources of hetero- the RCT, the analysis demonstrated that the M1 dis-
geneity were distinct LB designs used, different placed to the distal end after LB therapy.3 Among the
landmark references identified to evaluate the outcome, untreated individuals, this tooth presented mesialization
absence of a control group in some studies, and the short (P\0.05). In addition, LB therapy resulted in distal tip of
follow-up duration. Hence, only a descriptive compari- the M1, with a maximum value of 8
, against 2
mesial
son was reported. tip of this tooth among untreated individuals.3,5

Table III. Evaluation of risk of bias of the included studies using ROBINS-I
Domain

Selecting Deviations from Selecting Overall

participants for Classifying intended Missing Measuring reported risk-of-bias
Study Confounding the study the interventions intervention data outcomes result judgment
Nevant et al (1991)5 Serious Moderate Low Low Low Moderate Moderate Serious
Ingervall Moderate No reported Moderate Low Low Moderate Moderate Moderate
and Th€ uer (1998)17
Solomon et al (2006)19 Moderate Moderate Low Low Low Moderate Low Moderate
Moin and Serious Moderate Low Low Low Serious Low Serious
Bishara (2007)18
Ferro et al (2011)10 Serious Low Low Low Low Moderate Low Serious

ROBINS-I, Risk of Bias in Nonrandomized Studies of Interventions.

American Journal of Orthodontics and Dentofacial Orthopedics April 2020  Vol 157  Issue 4
462 Santana et al

Disorders in the eruption of the permanent M2 were derotation of the M1, it has been proposed that the
found after treatment with LB, which increased M2 greatest lateral gain would occur in the posterior2,8,9
impaction (odds ratio 9.0; 95% CI, 1.8-45.2) and ectopic or canine area.20 This systematic review suggests that
eruption (odds ratio 18.5; 95% CI, 4.1-82.4). An initial the greatest improvement occurs in the region of the de-
mesial inclination of the M2 .30
was associated with ciduous molars and their succeeding premolars. This can
a higher risk of impaction compared with a mesial be explained because the width in the premolar area
angulation \10
(P 5 0.001). Duration of treatment shows a steady increase throughout adolescence,
with LB .2 years relatively increased the risk of ectopic whereas intercanine width increases rapidly until the pri-
eruptions 2.6 times (95% CI, 1.0-6.5).10 mary canines are shed and then decreases continu-
ously.19,21 LB therapy in the included studies began in
Assessment of the certainty of evidence the late mixed dentition or early permanent dentition;
The certainty of evidence was evaluated according to therefore, it is expected that the greatest intercanine
the GRADE approach (Table IV). Reasons for downgrad- improvement has already occurred. Nevertheless, LB
ing the evidence are detailed there. therapy in the included studies resulted in little improve-
For the outcomes intercanine distance, arch width, ment in the intercanine width by removal of lip pressure
arch perimeter, incisor inclination, and tipping of M1, from this area.
the certainty levels were graded as low. For the outcomes Another factor to consider is the wearing of the same
arch length and M2 eruption disturbances, the certainty device with different activation protocols. The amount
levels were graded as very low. of activation of the appliance in the molar region (range
2 to 5 mm),5,18 premolars or primary molars (range 3 to
DISCUSSION 8 mm),17,18 canines (range 3 to 5 mm),5,17 and incisor
(range 1 to 3 mm)17,19 were different. The region where
Summary of evidence the appliance was farthest from the buccal surface of the
Owing to the increased interest in LB as an alternative teeth was in the premolars or primary molars, which may
method to minimize the risk of tooth extraction in cases be related to the greater transversal improvement in this
of mild mandibular crowding in children and adoles- area after LB therapy. These factors were considered
cents, many studies have been conducted to determine when a decision was made not to execute a meta-anal-
the LB effects on dental arch dimensions. To minimize ysis.
the influence of other simultaneous treatments on the In summary, it has been suggested that the wearing
findings of LB therapy, the included studies had to of LB increases the mandibular arch length by the distal
have the LB as the only therapy applied to directly affect movement of the molars and the labial movement of the
the mandibular arch. Most of the included studies had incisors5,17,18 increasing the arch perimeter, thereby
low to moderate methodological quality and low level reducing crowding,3,4,8 which was supported by this sys-
of certainty for the assessed outcomes. tematic review. Three studies3,17,18 showed that there
Despite the fact that the GRADE tool suggested an was proclination of the lower incisors after wearing the
overall low level of certainty, the results of this system- LB, but with no marked apex movement.3 With reduced
atic review consistently suggest that the LBs are effective body movement of the incisors, some issues such as the
appliances for gaining mandibular arch circumference in stability by excessive buccal inclination movement and
mild to moderately crowded arches. A combination of the likely compromised periodontal response remain
M1 distoangulation, proinclination of the incisors, and difficult concerns to account for.
buccal tipping of the deciduous molars and/or premolars Another potential action of the LB is the prevention
seems to explain increases in width, length, and perim- of the mesial migration of the mandibular molars during
eter of the mandibular arch. There is consistency in those the late mixed dentition phase. This occurs by distal
movements, but the high level of uncertainty implies displacement of the teeth,3 distal tipping,5 or both,
that the probable amount for each type of movement thus preserving the E space,22 providing adequate space
is unpredictable. The potential adverse side effects to solve crowding that is present in the mixed dentition
include increased chances of M2 impaction or ectopic in borderline cases.
eruption. Although the early creation of adequate space in the
This may happen because of the LB holding the anterior part of the arch is usually considered a priority
cheeks away from the buccal surfaces of the teeth, allow- for crowding relief, less attention is sometimes given
ing the tongue pressure to act on the opposite direction to what happens distally in the arch. One of the included
to increase transverse arch dimensions. Owing to the pe- studies10 reported a greater chance of impacting the per-
riodic expansion of the bumper associated with a manent M2 after wearing the LB. Negative predictive

April 2020  Vol 157  Issue 4 American Journal of Orthodontics and Dentofacial Orthopedics
Santana et al 463

Table IV. GRADE evidence profile table about efficiency of LB therapy on mandibular arch
Certainty assessment

Overall
Study Other certainty of
No. of studies design Risk of bias Inconsistency Indirectness Imprecision consideration Impact evidence
Arch perimeter
1 CT Very serious* Not serious Not serious Not serious None An increase in the arch 44
perimeter is LOW
expected after LB
therapy compared
with untreated
patients
Arch length
3 CT Serious* Not serious Very seriousy Not serious None The use of LB resulted 4
in an increase in the VERY LOW
length of the
mandibular arch
2 OS Very serious* Not serious Very seriousz Not serious None 4
VERY LOW

Arch width (permanent M1)

1 CT Not Serious* Not serious Very seriousz Not serious None A single study showed 44
an increase in LOW
arch width in the
M1 area

Intercanine distance
2 CT Serious* Not serious Seriousy Not serious None All studies showed an 44
increase in LOW
arch width in the
canine area

Incisor angulation
2 CT Serious* Not serious Seriousy Not serious None All studies showed a 44
buccal inclination LOW
of the
mandibular incisors

Distalization of M1
2 CT Serious* Not serious Seriousy Not serious None One study showed 44
distal movement of LOW
the center of
resistance of the
M1. Another study
showed movement
toward the distal
cusp while the apex
moved to mesial
Tip of M1
2 CT Serious* Not serious Seriousy Not serious None All studies showed 44
distal tip of LOW
the M1 after the LB
therapy

American Journal of Orthodontics and Dentofacial Orthopedics April 2020  Vol 157  Issue 4
464 Santana et al

Table IV. Continued

Certainty assessment

Overall
Study Other certainty of
No. of studies design Risk of bias Inconsistency Indirectness Imprecision consideration Impact evidence
M2 eruption disturbances
1 OS Serious* Not serious Serious§ Not serious None Treatment with LB 4
increased the VERY LOW
impact and eruption
ectopic of M2. The
initial crowding, M2
inclination
and treatment time
include the risk of
eruption
disturbance.

CT, Clinical trial; OS, observational studies.

*Based on the bias of risk assessment tool.
y
Some of the studies had no control group.
z
Absence of control group.
§
There were differences between groups at baseline.

factors include pretreatment M2 inclination .30
, and dental arch changes, with multiple measurement ap-
LB treatment longer than 2 years. However, the risk of proaches being used made it impossible to carry out a
M2 eruption disturbances seems to be more related to meta-analysis. The recommendations for wearing an
their previous position than the amount of distal move- LB were also different, with instructions ranging from
ment of M1. Thus, it is important to consider the M2 po- wearing the appliance part-time to appliances tied in
sition and the available space in the posterior molar the molars intended to be worn all the time. Moreover,
region during the decision process. Otherwise, improve- fundamental questions such as phase of dentition at
ments in the anterior dental arch spaces come at the cost the beginning of treatment, initial crowding, and time
of lack of space in the posterior region. of wearing LB varied substantially and were a source
Finally, building a young patient's willingness to of heterogeneity among the included studies. Finally,
cooperate steadily also becomes a key factor for LB for including non-RCTs, greater heterogeneity was ex-
treatment effectiveness. Lips appear quite bulged when pected than a systematic review of an RCT. This is due
an LB is in the mouth, and this may generate a social to the increased potential for methodological diversity
integration problem. The maximum duration of LB- through variation among primary studies in their risk
wearing in the included studies in this systematic review of selection bias, variation in the way confusion is
was 28 months. An alternative to make up for a lack of considered in the analysis, and increased risk of other
cooperation was to tie the LB.3,19 Longer treatments biases owing to poor design and execution.
should be discouraged because they have shown to in- The included studies did not blind the participants or
crease M2 ectopic eruptions, the correction of which the clinician, although that was not a option. In addi-
would require additional orthodontic intervention in tion, the RCT failed to provide details on the sample
any case. size, allocation concealment, sample randomization,
and blinding outcome assessor and statistician. In
Limitations and future directions non-RCTs,5,10,17-19 although selection bias cannot be
Data extractions and qualitative analyses, usually avoided, recruitment of consecutively treated
performed in qualitative systematic descriptions, present participants could, at least in part, prevent some
marked drawbacks in comparison with meta-analytic re- selection bias. Pairwise sampling with a group of
views because it becomes quite challenging to weigh the untreated individuals would allow the study to provide
data coming from individual studies. The use of the more reliable conclusions on the effects of LB therapy
GRADE tool should have taken this into consideration. when normal growth does not take place.
The inclusion of only 1 controlled study3 that re- Future randomized and controlled clinical studies are
sponded to the primary outcome of this review, com- recommended, with well-designed methodology,
bined with the wide variation in measurement of following guidelines for clinical trials such as

April 2020  Vol 157  Issue 4 American Journal of Orthodontics and Dentofacial Orthopedics
Santana et al 465

Consolidated Standards of Reporting Trials,23 so that a 7. Osborn WS, Nanda RS, Currier GF. Mandibular arch perimeter
high level of evidence and certainty on the effects of changes with lip bumper treatment. Am J Orthod Dentofacial Or-
thop 1991;99:527-32.
therapy with LB can be secured.
8. Werner SP, Shivapuja PK, Harris EF. Skeletodental changes in the
adolescent accruing from use of the lip bumper. Angle Orthod
CONCLUSIONS 1994;64:13-20: discussion 21-2.
9. Ferris T, Alexander RG, Boley J, Buschang PH. Long-term stability
Current evidence suggests a low to very low level of
of combined rapid palatal expansion-lip bumper therapy followed
certainty (GRADE assessment) regarding the effects of by full fixed appliances. Am J Orthod Dentofacial Orthop 2005;
LB therapy on the mandibular arch of children and ado- 128:310-25.
lescents. Well-designed research is needed to guide 10. Ferro F, Funiciello G, Perillo L, Chiodini P. Mandibular lip bumper
treatment decisions on this topic. treatment and second molar eruption disturbances. Am J Orthod
Dentofacial Orthop 2011;139:622-7.
Therefore, weak evidence suggests that with the use
11. Baccetti T. Tooth anomalies associated with failure of eruption of
of an LB an increase in arch perimeter and width were first and second permanent molars. Am J Orthod Dentofacial Or-
attributed to the proclination of the incisor, buccaliza- thop 2000;118:608-10.
tion of the deciduous molar and/or premolar areas, 12. Hashish DI, Mostafa YA. Effect of lip bumpers on mandibular
and distal inclination of the molars. arch dimensions. Am J Orthod Dentofacial Orthop 2009;135:
106-9.
However, simultaneously, an increased chance of
13. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group.
impaction and ectopic eruption of permanent mandib- Preferred reporting items for systematic reviews and meta-
ular M2 after treatment with LB was noted. analyses: the PRISMA statement. J Clin Epidemiol 2009;62:
1006-12.
ACKNOWLEGMENT 14. Higgins JP, Altman DG, Gøtzsche PC, J€ uni P, Moher D, Oxman AD,
et al. The Cochrane Collaboration's tool for assessing risk of bias in
This work was financed in part by CAPES (Coor- randomised trials. BMJ 2011;343:d5928.
denaç~ao de Aperfeiçoamento de Pessoal de Nıvel Supe- 15. Sterne JA, Hernan MA, Reeves BC, Savovic J, Berkman ND,
rior - Code: 88881.336658/2019-01) for supporting Viswanathan M, et al. ROBINS-I: a tool for assessing risk of bias in
non-randomised studies of interventions. BMJ 2016;355:i4919.
academic and professional development.
16. Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-
Coello P, et al. GRADE: an emerging consensus on rating quality
SUPPLEMENTARY DATA of evidence and strength of recommendations. BMJ 2008;336:
924-6.
Supplementary data associated with this article can be
17. Ingervall B, Th€ uer U. No effect of lip bumper therapy on the pres-
found, in the online version, at https://doi.org/10. sure from the lower lip on the lower incisors. Eur J Orthod 1998;
1016/j.ajodo.2019.10.014. 20:525-34.
18. Moin K, Bishara SE. An evaluation of buccal shield
REFERENCES treatment. A clinical and cephalometric study. Angle Orthod
2007;77:57-63.
1. Cetlin NM, Ten Hoeve A. Nonextraction treatment. J Clin Orthod 19. Solomon MJ, English JD, Magness WB, McKee CJ. Long-term sta-
1983;17:396-413. bility of lip bumper therapy followed by fixed appliances. Angle Or-
2. Ferro F, Perillo L, Ferro A. Non extraction short-term arch changes. thod 2006;76:36-42.
Prog Orthod 2004;5:18-43. 20. Vargo J, Buschang PH, Boley JC, English JD, Behrents RG,
3. Davidovitch M, McInnis D, Lindauer SJ. The effects of lip bumper Owen AH 3rd. Treatment effects and short-term relapse of
therapy in the mixed dentition. Am J Orthod Dentofacial Orthop maxillomandibular expansion during the early to mid
1997;111:52-8. mixed dentition. Am J Orthod Dentofacial Orthop 2007;131:
4. Bjerregaard J, Bundgaard AM, Melsen B. The effect of the mandib- 456-63.
ular lip bumper and maxillary bite plate on tooth movement, oc- 21. Williams RE, Ceen RF. Craniofacial growth and the dentition. Pe-
clusion and space conditions in the lower dental arch. Eur J diatr Clin North Am 1982;29:503-22.
Orthod 1980;2:257-65. 22. Gianelly AA. Leeway space and the resolution of crowding in the
5. Nevant CT, Buschang PH, Alexander RG, Steffen JM. Lip bumper mixed dentition. Semin Orthod 1995;1:188-94.
therapy for gaining arch length. Am J Orthod Dentofacial Orthop 23. Moher D, Hopewell S, Schulz KF, Montori V, Gøtzsche PC,
1991;100:330-6. Devereaux PJ, et al. CONSORT 2010 explanation and elaboration:
6. Raucci G, Pach^eco-Pereira C, Elyasi M, d'Apuzzo F, Flores-Mir C, updated guidelines for reporting parallel group randomised trials.
Perillo L. Predictors of postretention stability of mandibular dental BMJ 2010;340:c869.
arch dimensions in patients treated with a lip bumper during mixed 24. Moorrees CFA. The Dentition of the Growing Child: A Longitudinal
dentition followed by fixed appliances. Angle Orthod 2017;87: Study of Dental Development Between 3 and 18 Years of Age.
209-14. Cambridge, Mass: Harvard University Press; 1959.

American Journal of Orthodontics and Dentofacial Orthopedics April 2020  Vol 157  Issue 4
465.e1 Santana et al

Supplementary Table. Search strategy in the

different databases
Electronic database Set of terms (key terms)
PubMed ((((("lip"[MeSH Terms] OR "lip"[All Fields])
AND bumper[All Fields]) OR (("lip"[MeSH
Terms] OR "lip"[All Fields]) AND
bumper[All Fields])) OR lip-bumpers
[All Fields]) OR (buccal[All Fields] AND
shield[All Fields])) OR (buccal[All Fields]
AND shields[All Fields])
Web of Sciences TOPIC: (lip bumper OR lip bumper
OR lip-bumpers OR buccal
shield OR buccal shields)
SCOPUS (TITLE-ABS-KEY (lip bumper) OR
TITLE-ABS-KEY (lip bumper) OR
TITLE-ABS-KEY (lip-bumpers) OR
TITLE-ABS-KEY (buccal shield) OR
TITLE-ABS-KEY (buccal shields))
The Cochrane Library (lip bumper):ti,ab,kw OR (lip bumpers)
:ti,ab,kw OR (lip-bumper):ti,ab,kw
OR (buccal shield):ti,ab,kw OR
(buccal shields):ti,ab,kw
Lilacs (tw:(lip bumper)) OR (tw:(lip bumpers))
OR (tw:(lip-bumper)) OR
(tw:(buccal shield))
OR (tw:(buccal shields))

April 2020  Vol 157  Issue 4 American Journal of Orthodontics and Dentofacial Orthopedics