European Journal of Orthodontics, 2016, 113–126

doi:10.1093/ejo/cjv034
Advance Access publication 19 May 2015

Systematic Review

Treatment effects of fixed functional appliances

in patients with Class II malocclusion: a
systematic review and meta-analysis

Downloaded from https://academic.oup.com/ejo/article-abstract/38/2/113/2599887 by guest on 20 December 2018

Vasileios F. Zymperdikas*, Vasiliki Koretsi**, Spyridon N. Papageorgiou
***,****,***** and Moschos A. Papadopoulos******
*Dental Unit, Medical Company, 71st Airmobile Brigade, Nea Santa, Greece, **Department of Orthodontics, School of
Dentistry, University Medical Centre Regensburg, Germany, Departments of ***Orthodontics and ****Oral Technology,
University of Bonn, Germany, *****Clinical Research Unit 208, University of Bonn, Germany, ******Department of
Orthodontics, School of Health Sciences, Faculty of Dentistry, Aristotle University of Thessaloniki, Greece

Correspondence to: Moschos A. Papadopoulos, Department of Orthodontics, School of Health Sciences, Faculty of Den-
tistry, Aristotle University of Thessaloniki, GR-54124 Thessaloniki, Greece. E-mail: [email protected]

Summary
Objective:  To assess the treatment effects of fixed functional appliances (FFAs) in treated versus
untreated Class II patients by means of lateral cephalometric radiographs.
Search methods:  Unrestricted electronic search of 18 databases and additional manual searches
up to October 2014.
Selection criteria: Prospective randomized and non-randomized controlled trials reporting on
cephalometric angular measurements of Class II patients treated with FFAs and their matched
untreated controls.
Data collection and analysis:  Skeletal, dental, and soft tissue cephalometric data were annualized
and stratified according to the time of evaluation in effects. Following risk of bias evaluation, the
mean differences (MDs) and 95 % confidence intervals (CIs) were calculated with random-effects
models. Patient- and appliance-related subgroup analyses and sensitivity analyses were performed
with mixed-effects models.
Results:  Nine studies were included (244 patients; mean age: 13.5 years and 174 untreated controls;
mean age: 12.8 years) reporting on cephalometric effects directly after the removal of FFAs. FFAs
were found to induce a small reduction of SNA angle (MD = −0.83 degree/year, 95 % CI: −1.17 to
−0.48), a small increase of SNB angle (MD = 0.87 degree/year, 95 % CI: 0.30–1.43), and moderate
decrease of ANB angle (MD = −1.74 degree/year, 95 % CI: −2.50 to −0.98) compared to untreated
Class II patients. FFA treatment resulted in significant dentoalveolar and soft tissue changes. Several
patient- or appliance-related factors seem to affect the treatment outcome. Long-term effectiveness
of FFAs could not be assessed due to limited evidence.
Conclusions:  According to existing evidence, FFAs seem to be effective in improving Class II malocclusion
in the short term, although their effects seem to be mainly dentoalveolar rather than skeletal.

Introduction development of Class  II malocclusion could be attributed to sev-

eral factors; hence, accurate diagnosis is important for the selection
Rationale
of the corresponding treatment plan. Among these factors, man-
Class  II malocclusion is a frequent condition and a rather com- dibular retrognathism is considered a dominant one (1, 3). In this
mon reason for patients seeking orthodontic treatment (1, 2). The respect, mandibular advancement through the use of functional

© The Author 2015. Published by Oxford University Press on behalf of the European Orthodontic Society. All rights reserved.
113
For permissions, please email: [email protected]
114 European Journal of Orthodontics, 2016, Vol. 38, No. 2

appliances is considered as a popular treatment approach in grow- Materials and methods

ing individuals.
Protocol and registration
While some researchers posit favourable treatment outcomes
based on mandibular growth, attributed either as a mandibular The protocol for the present systematic review was constructed a
length augmentation or effective condyle growth (4–7), others dis- priori according to the Cochrane Handbook for Systematic Reviews
pute the magnitude of these effects (8, 9). Furthermore, a restriction of Interventions 5.1.0 (31) and is available upon request (no registra-
effect on the maxilla has been likewise supported by some research- tion was performed). The systematic review is reported on the basis
ers (10–14) and questioned by others (5, 15). Moreover, existing of the PRISMA statement (32) and its extension for abstracts (33).
evidence indicates that the dentoalveolar changes produced by func-
tional treatment outweigh the skeletal changes attained (13, 15–17). Information sources and search
Functional appliances can be categorized into either removable or Eighteen electronic databases were systematically and unrestrictedly
fixed ones (FFAs). An important discriminating factor between them is searched up to November 2011 and updated in October 2014. In
the need for patient compliance, which is considered to be a possible an effort to screen as many eligible articles as possible and not to
influence on the treatment outcomes (18, 19). As a consequence, it is miss any pertinent studies, the search strategy covered initially all

Downloaded from https://academic.oup.com/ejo/article-abstract/38/2/113/2599887 by guest on 20 December 2018

essential to assess those two types of functional appliances separately in functional appliances used for Class II treatment (both removable
order to investigate their clinical effectiveness and to reveal any existing and fixed) and then was cautiously limited to FFAs. MESH terms
differences. Previously published systematic reviews on the subject pre- and the respective keywords were used properly to fit each database
sented methodological limitations (9, 20–30). The current systematic (Supplementary Table 1). The search strategy included no limitations
review on FFAs supplements a previously published systematic review concerning language, publication year, or status. The reference lists
focused exclusively on removable functional appliances (17). of the included trials and relevant reviews were manually searched
as well. Grey literature was also assessed through proper registers
Objectives and databases. When considered necessary, authors were contacted
This study aims to summarize current evidence only from rand- for complementary data or clarifications. The search was performed
omized controlled trials (RCTs) and prospective controlled clini- independently by two authors (VFZ and VK).
cal trials (pCCTs) assessing by means of lateral cephalometric
radiographs the clinical effectiveness of FFAs for the treatment of Eligibility criteria and study selection
patients with Class II malocclusion in comparison with untreated The eligibility criteria were pre-determined (Table 1). A study was
individuals, as well as to identify any factors affecting the treat- considered eligible when it reported on at least one treatment arm
ment outcomes. with a FFA and simultaneously all of the inclusion and none of

Table 1.  Eligibility criteria used for the study selection.

Category Inclusion criteria Exclusion criteria

Participant characteristics Studies on human patients with Class II malocclusion of Patients with craniofacial syndromes and/or cleft lip palate
any age or gender Patients with temporomandibular joint disorders
Animal studies
Intervention Orthodontic treatment with fixed functional appliances Patients with Class II malocclusion treated with extrac-
tions, Class II elastics, orthognathic surgery, or removable
functional appliances
Comparison Untreated patients with Class II malocclusion matched Studies without an untreated Class II control group
for age and gender
Outcome Studies providing angular skeletal, dentoalveolar and soft Studies providing only linear cephalometric measurements
tissue cephalometric measurements from lateral cephalo- Electromyographic evaluation
metric analysis Evaluation employing 3D imaging techniques
Cost-benefit analysis
Study design Randomized controlled clinical trials Unsupported opinion of expert
Prospective controlled clinical trials Editor’ s choices
Replies to the author/editor
Interviews
Commentaries
Books’/conferences’ abstracts
Summaries
Cross-sectional surveys
Case series without a control
Case reports
Case-control observational studies
Cohort studies
Retrospective clinical trials
Narrative reviews*
Systematic reviews*
Meta-analyses*

*After checking the reference lists for relevant articles.

V. F. Zymperdikas et al. 115

the exclusion criteria were fulfilled. In order to investigate only methodological limitations were judged to exist when a pCCT col-
the effects of FFAs, data concerning any previous or subsequent lected less than 17 points on the modified checklist (41).
phases with fixed appliances were not included, since fixed appli-
ances are likely to alter the effects caused by functional treatment Risk of bias across studies
(30). After the elimination of duplicates, the decision for the selec- If a sufficient number of trials were identified (n > 10), reporting
tion was made by taking into consideration the title, abstract, biases (small-study effects or publication bias) were planned to be
and, when it was considered necessary, the full text of the respec- assessed through the inspection of a contour-enhanced funnel plot
tive articles. Multiple reports pertinent to the same trial/patient (42), Begg’s rank correlation test (43), and Egger’s weighted regres-
cohort were grouped together. When the same trial was published sion test (44). If the tests hinted towards the existence of publica-
in various languages, the English version was preferred. Finally, tion bias, the Duval and Tweedie’s trim and fill procedure (45) was
articles including at least one treatment arm with FFAs were planned to be performed.
selected. The overall quality of evidence (confidence in effect estimates)
for each of the main outcomes was rated by using the Grades of
Data collection process and data items Recommendation, Assessment, Development, and Evaluation

Downloaded from https://academic.oup.com/ejo/article-abstract/38/2/113/2599887 by guest on 20 December 2018

Data were extracted independently on predefined and piloted (GRADE) approach (46).
forms by two authors (VFZ and VK). Any ambiguities were The minimal clinical important, large, and very large effects were
resolved after discussion with the last author (MAP). In an effort conventionally defined (47) as half, one, and two standard devia-
to investigate the clinical effects of FFAs in Class II treatment, only tions of the Caucasian cephalometric norm plus 1 degree, respec-
angular cephalometric measurements on lateral cephalometric tively, to allow for method error. The optimal information size (i.e.
analyses were considered as primary outcomes due to the fact that required meta-analysis sample size) was calculated for each outcome
linear measurements are prone to magnification bias (34, 35). Due independently for α = 5 per cent and β = 20 per cent.
to the variability of the terms used among the authors for identi- The search strategy, study selection, data extraction, and
cal variables, all equivalent terms pertaining to the same variable within-studies risk of bias assessment were performed indepen-
were grouped (Supplementary Table  2) and one term was used dently without blinding (48) by two review authors (VFZ and VK),
throughout the review. across-studies risk of bias assessment by a third author (SNP), and
If the same variable was reported in at least two included trials, the any disagreement was resolved by discussion with the last author
respective data were extracted and categorized as skeletal (sagittal and (MAP). Authors were contacted when necessary to resolve ambi-
vertical), dental, soft tissue variables and ratios. Reported outcomes guities or provide complementary data. Inter-reviewer agreement
were stratified based on the time of evaluation in effects: 1. after the for the three stages was evaluated with the unweighted Cohen’s
removal of the corresponding FFA and 2. after the retention phase. kappa (49).
Several factors were determined a priori to be investigated via
subgroup analyses for their possible effect on the FFA treatment out- Summary measures and synthesis of results
comes, if at least five studies reported on the corresponding factor. Data were summarized and considered suitable for pooling if
These factors were divided into patient-related (i.e. specific patients’ similar control groups of untreated Class  II patients were used
characteristics) and appliance-related (i.e. specific features of the and if the same cephalometric angular outcomes were reported.
appliance design or the treatment plan). The patient-related factors In an effort to account for the different follow-up periods of the
involved 1.  patients’ gender ratio (male patients/female patients), included studies, the treatment and the observational changes
2. patients’ skeletal growth stage (pre-peak and peak or post-peak), were annualized. Mean Differences (MDs) and their correspond-
and 3. patients’ growth pattern (horizontal, vertical, or average). The ing 95 % confidence intervals (CIs) were calculated. The random-
discrimination between the various stages of skeletal maturation effects model as proposed by DerSimonian and Laird (50) was
relied exclusively on the cervical vertebral maturation index or on chosen a priori as the primary method to estimate all pooled
hand-wrist radiographs, which are considered as effective methods estimates, appropriately supported by both clinical and statisti-
for the identification of skeletal age (36, 37). Studies providing data cal reasoning (51), as described previously (17). The extent and
concerning only the chronological and/or dental age of the patients impact of between-study heterogeneity was assessed by inspect-
were excluded from the evaluation of skeletal maturity, since these ing the forest plots and by calculating the τ2 and the I2 statistics,
methods are not considered reliable for the discrimination of skeletal respectively. The 95 % CIs around I2 were calculated according
growth stages (38, 39). If the growth pattern was clearly stated in to the non-central chi-square approximation of Q (52). For meta-
the included articles, then it was categorized accordingly. When no analyses with greater than or equal to three trials, 95 % predic-
such data were available, the discrimination was based on the mean tion intervals (53, 54) were calculated to quantify of treatment
values of either the SN-ML or FH-ML angles as reported on the effects of FFAs in a future trial.
patients’ baseline characteristics. All P values were two sided with α = 5 %, except for the test of
Further, the appliance-related factors included 1. appliance used between-studies or between-subgroups heterogeneity (α = 10 %) (55).
(i.e. the exact type and design of the respective appliance) and 2. con-
struction bite (single step versus stepwise mandibular advancement).
Additional analyses
Possible sources of heterogeneity in meta-analyses with five or more
Risk of bias in individual studies studies were sought through pre-specified mixed-effects subgroup
The Cochrane Collaboration’s tool (31) and a modified Downs analyses and random-effects meta-regression with the Knapp–
and Black checklist (40) were used to assess the risk of bias in Hartung adjustment (56).
RCTs and pCCTs, respectively. The overall risk of bias was judged Robustness of the results was a priori to be checked with sensi-
as ‘high’, ‘low’, or ‘unclear’ for randomized studies, while serious tivity analyses based on 1. the duration of the FFA treatment, 2. the
116 European Journal of Orthodontics, 2016, Vol. 38, No. 2

Records identified through Additional records identified

database searching (n = 9107) through other sources (n = 8)

Records after duplicates removed (n = 6342)

Records screened on the basis of title and abstract (n = 6342)

Records excluded (n = 6087)

Full-text articles assessed for eligibility (n = 255)

Downloaded from https://academic.oup.com/ejo/article-abstract/38/2/113/2599887 by guest on 20 December 2018

Articles excluded with reasons (n = 245)
Animal studies (n = 1)
Patients treated with Class II elastics, orthognathic surgery or
extractions (n = 8)
Studies without or inappropriate control group (n = 73)
Studies irrelevant to the subject of this study (n = 39)
Studies employing 3D imaging techniques (n = 6)
Retrospective clinical trials (n = 80)
Ongoing studies (n = 2)
No angular measurements included (n = 10)
Studies reporting on removable functional appliances (n = 22)
Studies without discrimination between fixed functional appliance
treatment and treatment with fixed orthodontic appliances (n = 4)

Studies included in the qualitative synthesis (systematic review) (n = 10)

Studies included in the quantitative synthesis (meta-analysis)

(n = 10; datasets = 9)

Figure 1.  PRISMA flow diagram for the selection of studies.

Table 2.  Number of excluded studies with reasons.

Reason for exclusion Excluded articles on the basis of Excluded articles on the
title and abstract basis of full text

Animal studies 1 1
Patients treated with extractions, Class II elastics, or orthognathic surgery 89 8
Studies without or inappropriate comparison group 144 73
Investigation not relevant to the subject of this study 5604 39
Evaluation employing 3D imaging techniques 7 6
Books’/conferences’ abstracts 72 —
Cross-sectional surveys 4 —
Case reports 62 —
Retrospective clinical trials 68 80
Narrative reviews 9 —
Systematic reviews 23 —
Meta-analyses 4 —
Ongoing studies — 2
No angular cephalometric measurements — 10
Removable functional appliances — 22
No discrimination between treatment with fixed functional appliances and fixed — 4
orthodontic appliances
Sum 6087 245
Table 3.  Characteristics of the studies included in the current systematic review. CCT†, prospective controlled trial with historical control; CCT, prospective controlled trial; RCT, randomized
controlled trial; Gp1/2, treatment group; Ctr, control group; ME, maxillary expansion; M/F, males/females; m, months; NR, not reported; TMJ, temporomandibular joint.

Treatment
Skeletal growth time* Risk of Conflict of
A/A Study Design Setting Characteristics of patients Interventions No. of patients (M/F) Age in years (SD) stage (m) Outcomes bias interest

1 Alali (57) CCT University; Syria Class II/1 malocclusion, overjet > 4 mm, ANB > 4° Fixed lingual Gp: 21 (10/11); Ctr: Gp: 13.2 (0.9); Peak Gp: 8.0; Skeletal, dental 16 NR
and APg/NL < 80º, SNB < 76º, pubertal growth mandibular growth 17 (7/10) Ctr: 12.5 (2.1) Ctr: 8.0
spurt peak modificator
V. F. Zymperdikas et al.

2 Baysal and Uysal RCT University; ANB > 4°, SNB < 78°, overjet ≥ 5 mm, SN-GoGn Cast splint Herbst Gp: 20 (9/11); Ctr: 20 Gp: 12.7 (1.4); Pre-peak and Gp: 15.8; Skeletal, dental, High Internal
(58) Turkey 32° ± 6°, crowding in dental arches ≤ 4 mm, (ME) (11/9) Ctr: 12.2 (1.5) peak Ctr: 15.6 soft tissue risk
bilateral Class II molar and canine relationship
≥ 3.5 mm
3 de Almeida et al. CCT† University; Class II/1 malocclusion with bilateral distal molar Herbst (modified) Gp: 30 (15/15)**; Ctr: Gp: 9.8 (NR)***; Pre-peak and Gp: 12.0; Skeletal, dental, 23 NR
(59, 60) Brazil relationship > one-half cusp, presence of mandibu- 30 (15/15) Ctr: 9.7 (NR) peak Ctr: 12.0 soft tissue
lar deciduous second molars, ANB ≥ 4.5°
4 Gunay et al. (61) CCT† University; Skeletal and dental Class II malocclusion due Forsus Fatigue Resist- Gp: 15 (6/9); Ctr: 12 Gp: 15.0 (1.2); Post-peak Gp: 5.3; Skeletal, dental, 18 NR
Turkey to retrognathic mandible, normal or low-angle ant Device (3/9) Ctr: 14.1 (1.4) Ctr: 6.0 soft tissue
growth pattern, post-peak growth period, no
extracted or congenitally missing permanent teeth,
minimum crowding in the lower dental arch
5 Karacay et al. (62) CCT Military Active growth period, normal or mildly prog- Gp1: Forsus nitinol Gp1: 16 (9/7); Gp2: Gp1: 13.6 (1.2); NR Gp1: 5.3; Skeletal, dental 22 NR
Medical Acad- nathic maxilla, retrognathic mandible, horizontal flat spring; Gp2: Jas- 16 (10/6); Ctr: 16 Gp2: 14.0 (1.9); Gp2: 5.2;
emy; Turkey or normal growth pattern, Class II molar per jumper (NR) Ctr: 13.8 (1.4) Ctr: 6.0
relationship, overjet < 7 mm, minimum crowding,
permanent dentition
6 Latkauskienė (63) CCT† University; Lithu- ≥ End-to-end Class II molar relationship bilater- Stainless steel crown Gp: 40 (20/20); Ctr: Gp: 13.6 (1.3); Peak and post- Gp: 12.0; Skeletal, dental 25 NR
ania ally or more severe, permanent dentition, no Herbst 18 (11/7) Ctr: 13.9 (1.6) peak Ctr: 12.0
active hard tissue lesions, no previous orthodontic
treatment or tooth extractions, no bone level
problems, no TMJ complaints, no tooth size,
form, and number anomalies, no facial develop-
ment or mental syndromes, no pregnancy
7 Oztoprak et al. (64) CCT† University; Class II malocclusion due to retrognathic mandi- Gp1: Sabbagh Gp1: 20 (9/11); Gp2: Gp1: 15.3 (1.2); Post-peak Gp1: 5.2; Skeletal, dental, 23 NR
Turkey ble, SN-MP = 25°–35°, post-peak growth period, universal spring Gp2: 20 (8/12); Ctr: 19 Gp2: 15.1 (1.0); Gp2: 5.2; soft tissue
no extracted or congenitally missing permanent Forsus FRD (5/14) Ctr: 14.8 (1.3) Ctr: 6.0
teeth, minimum crowding in the lower arch
8 Phelan et al. (65) CCT† Private practice; Class II/1 malocclusion of a half or full cusp, Sydney magnoglide Gp: 31 (19/12); Ctr: 30 Gp: 13.5 (1.2); Pre-peak, peak, Gp: 12.0; Skeletal, dental 22 External; non-
Germany overjet ≥ 6 mm, ANB > 3.5°, and non-extraction (15/15) Ctr: 13.0 (1.6) and post-peak Ctr: 12.0 profit
treatment plan
9 Uyanlar et al. (66) CCT University; SNB < 80°, SN-ML ≤ 32°, post-peak growth Sabbagh universal Gp: 15 (7/8); Ctr: 12 Gp: 15.2 (1.1); Post-peak Gp: 5.2; Skeletal, dental, 19 NR
Turkey period, no extracted or congenitally missing spring (3/9) Ctr: 14.1 (1.4) Ctr: 6.0 soft tissue
permanent teeth, minimum crowding in the lower
dental arch

*Control group received no intervention; it refers to the observation time.

**Only for soft tissue measurements—Gp: 29 (15/14) and Ctr: 28 (14/14).
***Only for soft tissue measurements—Gp: 9.9 (NR).
117

Downloaded from https://academic.oup.com/ejo/article-abstract/38/2/113/2599887 by guest on 20 December 2018

Table 4.  Details of the performed meta-analyses with tests on heterogeneity. FA, functional appliance group; Ctr, control group; MD, mean difference; CI, confidence interval; PI, prediction
118

interval; NA, not applicable.

Effect size Heterogeneity

No. Variable Studies Post–pre in FA* Post–pre in Ctr* MD 95% CI P value 95% PI P value τ2 I2 (95% CI)

1 SNA 9** ↓ (−0.57) ↑ (0.13) −0.83 −1.17, −0.48 <0.001 −1.58, −0.08 0.202 0.069 27% (0, 66)
2 SNB 9** ↑ (1.07) ↑ (0.11) 0.87 0.30, 1.43 0.003 −0.84, 2.57 <0.001 0.435 72% (32, 84)
3 SN-Pg*** 2 ↑ (1.45) ↑ (0.24) 1.29 −0.52, 3.09 0.162 NA <0.001 1.588 94% (NA)
4 ANB**** 9** ↓ (−1.95) ↓ (−0.13) −1.74 −2.50, −0.98 <0.001 −4.30, 0.82 <0.001 1.019 87% (78, 92)
5 NA-Apg**** 5** ↓ (−3.01) ↓ (−0.96) −1.86 −5.06, 1.34 0.254 −13.68, 9.95 <0.001 11.120 90% (78, 94)
6 SGo:Nme (%) 4** ↑ (1.31) ↑ (1.29) 0.71 −0.09, 1.51 0.080 −1.04, 2.46 0.424 0.000 0% (0, 68)
7 SN-ML 8** ↑ (0.16) ↓ (−0.43) 0.48 0.04, 0.92 0.031 −0.07, 1.03 0.949 0.000 0% (0, 56)
8 NL-ML 2** ↑ (0.28) ↑ (0.15) 0.13 −0.95, 1.20 0.818 NA 0.776 0.000 0% (NA)
9 SN-NL 6** ↑ (0.41) ↓ (−0.29) 0.54 −0.23, 1.32 0.168 −1.30, 2.39 0.181 0.284 34% (0, 73)
10 SN-OP 4** ↑ (8.44) ↓ (−1.58) 10.09 7.22, 12.96 <0.001 −0.96, 21.13 0.099 4.442 52% (0, 82)
11 y axis*** 2** ↑ (3.65) ↑ (0.34) 3.06 −2.18, 8.29 0.252 NA 0.002 12.881 90% (NA)
12 1s-SN**** 6** ↓ (−6.49) ↑ (0.93) −7.50 −10.88, −4.11 <0.001 −18.32, 3.33 <0.001 12.209 79% (44, 89)
13 1i-ML**** 9** ↑ (8.20) ↑ (0.25) 7.99 3.56, 12.42 <0.001 −8.34, 24.32 <0.001 42.595 96% (95, 97)
14 1s-1i***** 3** ↓ (−4.59) ↓ (−1.48) −8.32 −13.38, −3.25 0.001 −57.04, 40.41 0.190 8.038 40% (0, 82)
15 1s-NA 2 ↓ (−3.38) ↑ (1.16) −4.24 −6.09, −2.40 <0.001 NA 0.233 0.572 30% (NA)
16 1i-NB 2 ↑ (4.60) ↑ (0.36) 4.20 2.48, 5.91 <0.001 NA 0.195 0.684 40% (NA)
17 1i-VL 3** ↑ (18.67) ↓ (−1.13) 19.78 15.50, 24.06 <0.001 −24.38, 63.94 0.129 7.319 51% (0, 85)
18 N′SnPg′ 2 ↑ (1.88) ↓ (−0.19) 2.01 1.05, 2.96 <0.001 NA 0.622 0.000 0% (NA)
19 Nasolabial angle 5** ↑ (0.49) ↑ (0.49) 0.03 −2.39, 2.45 0.979 −4.83, 4.89 0.347 0.802 10% (0, 68)
20 Mentolabial angle 2 ↑ (10.00) ↓ (−5.19) 14.99 8.09, 21.88 <0.001 NA 0.233 7.337 30% (NA)
21 H angle 4** ↓ (−2.14) ↓ (−2.01) −1.95 −3.16, −0.74 0.002 −4.61, 0.70 0.908 0.000 0% (0, 68)
22 ANSMe:Nme****** 3** ↑ (0.89) ↑ (0.63) Omitted
23 Gonial ratio 3** ↑ (1.99) ↑ (0.23) 1.62 −0.66, 3.90 0.164 −23.17, 26.41 0.080 2.452 60% (0, 87)
24 S-Ar/Ar-Go 2** ↓ (−3.86) ↓ (−4.77) 0.49 −3.48, 4.46 0.809 NA 0.667 0.000 0% (NA)

Bold values indicate statistically significant differences between the functional appliance and the control groups at the 5% level.
*Results from random-effects meta-analysis of the post–pre differences in each group to provide an overview of the effect’s direction.
**Pooled trial arms included.
***High heterogeneity identified; however, our confidence in the calculation of heterogeneity is limited due to the small number of studies. Furthermore, it would affect only the estimation of the effect magnitude, not its
direction (i.e. all studies lie on the same side of the forest plot).
****High heterogeneity identified; however, heterogeneity is explained by differences between subgroups. Caution is warranted on the interpretation of the overall effect estimate; estimates for subgroups are to be pre-
ferred.
*****Initial analysis included four studies (MD = −3.14; 95% CI = −15.95 to 9.67; P = 0.631; τ2 = 160.559; I2 = 95%), but the study of Alali (57) (fixed lingual mandibular growth modificator) was omitted to achieve
homogeneity.
******High heterogeneity identified, which remained unexplained; meta-analysis of three studies (MD = 0.13; 95% CI = −2.17, 2.42; P = 0.915; τ2 = 3.472; I2 = 85%) was omitted, as studies were distributed on both sides
of the forest plot and elimination of a single study was not straightforward.
European Journal of Orthodontics, 2016, Vol. 38, No. 2

Downloaded from https://academic.oup.com/ejo/article-abstract/38/2/113/2599887 by guest on 20 December 2018

V. F. Zymperdikas et al. 119

method error of the cephalometric analysis (where reported), 3. the Study characteristics and risk of bias within studies
improvement of the GRADE classification, and 4. the study design. The characteristics of the 10 included studies (9 datasets) are
presented in Table  3. Seven of them took place at a university,
one at a military academy, and one at a private practice, includ-
Results
ing 418 subjects, from whom 244 were patients with a mean
Study selection age of 13.5 years that received a FFA treatment and 174 were
From the initially identified 9115 records, 6342 remained after exclu- untreated individuals with a mean age of 12.8 years that did not
sion of duplicates and 6087 additional records were excluded on the receive any treatment at all. The majority of the patients were
basis of screening (Table 2). A total of 255 full texts were assessed treated with the original design of the corresponding FFA, while
for eligibility (Figure 1), with 10 articles (57–66) having at least one in two studies, the FFAs were either modified or incorporated
treatment arm with a FFA and being included in the systematic review. additional elements for maxillary expansion. All included studies
Two articles reported (59, 60) data from the same study/cohort and provided data on skeletal and dentoalveolar changes, while five
were grouped together. Thus, 9 unique datasets were finally included reported additionally on soft tissue cephalometric outcomes and
in the qualitative and quantitative synthesis of this systematic review. three studies reported on cephalometric ratios.

Downloaded from https://academic.oup.com/ejo/article-abstract/38/2/113/2599887 by guest on 20 December 2018

In total, 2 authors were contacted; one of them responded, while the According to the Cochrane Collaboration’s tool, the only
other one did not respond (communication details available upon one RCT identified (58) was judged to be in high risk of bias
request). The kappa scores before reconciliation for the selection, (Supplementary Table 3a). The eight pCCTs (57, 59–66) scored an
data extraction, and risk of bias assessment procedures were 0.855, average of 22.4 points on the modified Downs and Black tool, while
0.923, and 0.891, respectively (with asymptotic standard errors none of them was considered to present serious methodological lim-
0.103, 0.088, and 0.097), indicating almost perfect agreement. itations (Supplementary Table 3b).

Figure 2.  Forest plot of the mean difference of the SNA, SNB, and ANB angles (Post-Pre increments in °/year) between FFAs and control groups based on the
random-effects model together with the 95% confidence interval (CI) and the 95% prediction interval (PI).
120 European Journal of Orthodontics, 2016, Vol. 38, No. 2

Table 5.  GRADE summary of findings table for the main outcomes of the systematic review directly after treatment with fixed functional
appliances. CI, confidence interval; Ctr, untreated control group; mo, month; FFA, fixed functional appliance; GRADE, Grades of Recom-
mendation, Assessment, Development, and Evaluation.

Illustrative comparative risks (95% CI)

Assumed risk Corresponding risk

No. of patients Quality of evi-
Outcomes Untreated (Ctr) patients FFA patients (trials) dence (GRADE) Comments

Annualized SNA change The SNA increased on average The mean SNA decreased in the 418 (9) ⊕⊕⊖⊖ Low —
from baseline (follow-up: by 0.13° per year in the Ctr FFA groups by 0.83° per year
5.2–15.8 months)* groups (range −0.52° to 0.80°) (95% CI: 0.48°–1.17° decrease)
compared to the Ctr groups
Annualized SNB change The SNB increased on average The mean SNB increased in the 418 (9) ⊕⊕⊖⊖ Low Effect magnitude af-
from baseline (follow-up: by 0.11° per year in the Ctr FFA groups by 0.87° per year fected by appliance

Downloaded from https://academic.oup.com/ejo/article-abstract/38/2/113/2599887 by guest on 20 December 2018

5.2–15.8 months)* groups (range −0.30° to 1.16°) (95% CI: 0.30°–1.43° increase)
compared to the Ctr groups
Annualized ANB change The ANB decreased on average The mean ANB decreased in the 418 (9) ⊕⊕⊖⊖ Low Effect magnitude
from baseline (follow-up: by 0.13° per year in the Ctr FFA groups by 1.74° per year affected by (i)
5.2–15.8 months)* groups (range −1.26° to 0.60°) (95% CI: 0.98°–2.50° decrease) patient’s gender and
compared to the Ctr groups (ii) appliance
Annualized SN-ML The SN-ML decreased on aver- The mean SN-ML increased in 360 (8) ⊕⊕⊖⊖ Low —
change from baseline age by 0.43° per year in the Ctr the FFA groups by 0.48° per year
(follow-up: 5.2– groups (range −1.92° to 0.30°) (95% CI: 0.04°–0.92° increase)
15.8 months)* compared to the Ctr groups
Annualized 1s-SN change The 1s-SN increased on average The mean 1s-SN decreased in the 262 (6) ⊕⊕⊕⊕ Very Effect magnitude
from baseline (follow-up: by 0.93° per year in the Ctr FFA groups by 7.50° per year high** affected by (i)
5.2–15.8 months)* groups (range 0.40° to 3.06°) (95% CI: 4.11°–10.88° decrease) patient’s skeletal
compared to the Ctr groups growth stage and
(ii) construction bite
Annualized 1i-ML change The 1i-ML increased on average The mean 1i-ML increased in the 418 (9) ⊕⊕⊕⊕ Very Effect magnitude
from baseline (follow-up: by 0.25° per year in the Ctr FFA groups by 7.99° per year high** affected by (i)
5.2–15.8 months)* groups (range −0.80° to 1.00°) (95% CI: 3.56°–12.42° increase) patient’s skeletal
compared to the Ctr groups growth stage and
(ii) appliance
Annualized nasola- The nasolabial angle increased The mean nasolabial angle 210 (5) ⊕⊕⊖⊖ Low —
bial angle change from on average by 0.49° per year in increased in the FFA groups by
baseline (follow-up: the Ctr groups (range −1.34° 0.03° per year (95% CI: 2.39°
5.2–15.8 months)* to 1.81°) decrease–2.45° increase) compared
to the Ctr groups

Patients: receiving orthodontic treatment to improve Class II malocclusion. Settings: university clinics (Brazil, Lithuania, Syria, Turkey), private practice (Ger-
many), and military academy (Turkey). Intervention: FFAs (fixed lingual mandibular growth modificator, Forsus fatigue resistant device, Forsus nitinol flat spring,
Herbst, Jasper jumper, Sabbagh universal spring, Sydney magnoglide). Comparison: untreated patients from follow-up or historical controls. All judgements start
from ‘low’ due to the vast inclusion of non-randomized studies.
*From cephalometric analysis.
**Upgraded by two for effect magnitude; very large effect (cephalometric norm + 2 standard deviations + 1° for method error), which was included in the
mean effect, the confidence interval, and the prediction interval, while no serious limitations were found. Furthermore, magnitude of incisor inclination change
significantly associated with duration of functional appliance treatment (dose-response effect).

Results of individual studies, synthesis of results, 0.83 degree/year was induced by FFAs. The effect of FFAs on the
and risk of bias across studies skeletal relationships of the maxilla to the mandible was favour-
Effectiveness of FFA treatment directly after appliance removal able, with the ANB angle being on average 1.74 degree decreased
Meta-analyses could be performed regarding only the short-term annually (P < 0.001) compared to the untreated group, indicating
effectiveness of FFAs (i.e. from the time point of placement of the a moderate improvement of the skeletal Class II jaw relationships.
corresponding FFAs until immediately after their removal) com- Finally, as far as the vertical skeletal relationships are concerned,
pared to natural growth (as indicated by the data of the untreated no significant effects could be found, except for annual increases
control individuals) for 24 cephalometric variables, including 11 of the SN-ML and SN-OP angles by 0.48 and 10.09 degree/year,
skeletal (5 sagittal and 6 vertical), 6 dental, 4 soft tissue variables, respectively. The later indicates a clinically significant effect on the
and 3 ratios (Table 4). In short, many skeletal, dental, and soft tissue inclination of the occlusal plane during mandibular advancement.
variables were found to be significantly affected by FFA treatment. With regard to the dentoalveolar changes, treatment effects were
With regard to the skeletal changes in the sagittal plane, the evident on all variables corresponding to the upper and lower dental
skeletal growth of the mandible was slightly affected by FFAs, arches. Significant retroclination of the upper incisors was observed
with the SNB angle being on average 0.87 degree/year greater compared to the untreated group, as seen from the 1s-SN (−7.50
than the untreated group (Figure  2). Further, a statistically sig- degrees/year) and 1s-NA (−4.24 degrees/year) angles. Additionally,
nificant slight restriction effect on the maxillary growth of about the lower incisors were significantly proclined, as seen from the
Table 6.  Details of the performed subgroup analyses. MD, mean difference; CI, confidence interval; PSG, P values for difference between subgroups; FFRD, Forsus fatigue resistant device;
FLMGM, fixed lingual mandibular growth modificator; FNFS, Forsus nitinol flat spring; JJ, Jasper jumper; SUS, Sabbagh universal spring; SM, Sydney magnoglide; SS, stainless steel.
V. F. Zymperdikas et al.

SNA (n = 11) SNB (n = 11) ANB (n = 11) NA-APg (n = 6) SN-ML (n = 10)

n MD (95% CI) PSG n MD (95% CI) PSG n MD (95% CI) PSG n MD (95% CI) PSG n MD (95% CI) PSG

Patient related
  Gender ratio 11 0.190 11 0.710 11 0.080 6 0.002 10 0.368
  Skeletal growth stage
   Pre-peak and peak 3 −0.92 (−1.35, −0.50) 0.316 3 1.23 (−0.12, 2.57) 0.401 3 −2.05 (−3.43, −0.67) 0.218 2 −4.89 (−9.21, −0.57) 0.135 3 0.46 (−0.09, 1.00) 0.440
  Post-peak 4 −0.11 (−1.81, 1.59) 4 0.48 (−0.31, 1.27) 4 −0.49 (−2.03, 1.06) 4 0.72 (−3.06, 4.50) 4 0.95 (−0.07, 1.98)
Appliance related
 Appliance
  FFRD 2 1.31 (0.00, 2.62) 0.284 2 0.43 (−0.65, 1.49) 0.082 2 0.94 (−0.42, 2.30) 0.017 2 4.08 (1.38, 6.78) 0.060 2 1.41 (−0.39, 3.21) 0.881
  FLMGM 1 −0.90 (−1.62, −0.18) 1 2.84 (1.98, 3.70) 1 −3.59 (−4.35, −2.83) 1 −7.16 (−8.93, −5.39) 1 0.30 (−0.92, 1.52)
  FNFS 1 −1.48 (−7.22, 4.26) 1 4.37 (−1.02, 9.76) 1 −7.19 (−9.67, −4.71) 1 −0.43 (−9.76, 8.90)
  Herbst 3 −0.75 (−1.27, −0.23) 3 0.55 (0.24, 0.86) 3 −1.22 (−1.58, −0.86) 1 −2.75 (−3.79, −1.71) 2 0.50 (−0.12, 1.11)
  JJ 1 −1.27 (−6.82, 4.28) 1 2.62 (−2.40, 7.64) 1 −4.32 (−6.41, −2.23) 1 −1.83 (−9.22, 5.56)
  SUS 2 −1.64 (−3.18, −0.09) 2 0.55 (−0.63, 1.72) 2 −1.77 (−2.90, −0.64) 2 −2.54 (−5.13, 0.05) 2 0.74 (−0.51, 1.98)
  SM 1 −1.30 (−2.01, −0.59) 1 0.60 (−0.15, 1.35) 1 −1.60 (−2.18, −1.02) 1 0.20 (−0.75, 1.15)
  Herbst design
  SS crown 1 −0.50 (−1.08, 0.08) 0.095* 1 0.70 (0.22, 1.18) 0.607* 1 −1.10 (−1.55, −0.65) 0.236*
  Cast splint 1 −1.17 (−1.61, −0.74) 1 0.35 (−0.15, 0.85) 1 −1.63 (−2.20, −1.06) 1 0.56 (−0.23, 1.35) 0.802*
  Modified 1 −0.40 (−1.20, 0.40) 1 0.60 (−0.06, 1.26) 1 −1.00 (−1.54, −0.46) 1 0.40 (−0.57, 1.37)
  Construction bite
  Single step 2 −0.86 (−1.60, −0.12) 0.444 2 0.44 (0.04, 0.84) 0.452 2 −1.31 (−1.92, −0.69) 0.765 1 −2.75 (−3.79, −1.71) 0.436 2 0.50 (−0.12, 1.11) 0.542
  Stepwise 7 −0.27 (−1.20, 0.66) 7 0.68 (0.27, 1.09) 7 −1.89 (−3.37, −0.40) 4 0.72 (−3.06, 4.50) 6 0.89 (−0.12, 1.89)
121

Downloaded from https://academic.oup.com/ejo/article-abstract/38/2/113/2599887 by guest on 20 December 2018

 122

Table 6.  Continued

SN-NL (n = 8) 1s-SN (n = 8) 1i-ML (n = 11) Nasolabial angle (n = 6)

n MD (95% CI) PSG n MD (95% CI) PSG n MD (95% CI) PSG n MD (95% CI) PSG

Patient related
  Gender ratio 8 0.168 8 0.497 11 0.377 6 0.104
  Skeletal growth stage
   Pre-peak and peak 2 0.06 (−0.47, 0.59) 0.052 1 −4.24 (−6.19, −2.30) 0.035 3 0.16 (−5.12, 5.43) 0.010 2 0.51 (−5.05, 6.07) 0.824
  Post-peak 4 1.93 (0.70, 3.17) 4 −10.92 (−13.92, −7.92) 4 18.89 (12.50, 25.28) 4 −0.44 (−3.64, 2.76)
Appliance related
 Appliance
  FFRD 2 1.60 (0.08, 3.11) 0.397 2 −10.60 (−15.89, −5.30) 0.151 2 24.51 (21.59, 27.43) 0.002 2 −2.75 (−6.45, 0.96) 0.348
  FLMGM 1 0.00 (−0.63, 0.63) 1 −6.72 (−9.88, −3.56)
  FNFS 1 −1.79 (−6.55, 2.97) 1 −11.54 (−17.93, −5.15) 1 11.03 (5.47, 16.59)
  Herbst 1 0.20 (−0.78, 1.18) 1 −4.24 (−6.19, −2.30) 3 3.92 (2.25, 5.59) 2 0.51 (−5.05, 6.07)
  JJ 1 0.03 (−4.21, 4.27) 1 −12.03 (−20.39, −3.67) 2 16.15 (6.07, 26.23)
  SUS 2 2.59 (0.46, 4.72) 2 −11.07 (−14.71, −7.44) 2 13.31 (9.95, 16.67) 2 3.01 (−1.71, 7.73)
  SM 1 −2.20 (−4.24, −0.16) 1 0.40 (−1.05, 1.85)
  Herbst design
  SS crown 1 5.40 (3.38, 7.42) 0.102* 0.084*
  Cast splint 1 2.74 (1.33, 4.15) 1 −2.15 (−6.26, 1.96)
  Modified 1 4.00 (1.58, 6.42) 1 3.54 (−1.45, 8.53)
  Construction bite
  Single step 1 0.20 (−0.78, 1.18) 0.135 1 −4.24 (−6.19, −2.30) 0.009 2 3.06 (1.84, 4.28) 0.057 2 0.51 (−5.05, 6.07) 0.824
  Stepwise 6 1.57 (0.42, 2.72) 6 −11.13 (−13.71, −8.55) 7 15.34 (8.39, 22.30) 4 −0.44 (−3.64, 2.76)

Bold values indicate statistically significant differences between subgroups.

*Mixed-effects subgroup analysis not possible due to the small sample. Conventional inverse variance subgroup analysis reported.
European Journal of Orthodontics, 2016, Vol. 38, No. 2

Downloaded from https://academic.oup.com/ejo/article-abstract/38/2/113/2599887 by guest on 20 December 2018

V. F. Zymperdikas et al. 123

1i-ML (7.99 degrees/year), 1i-NB (4.20 degrees/year), and 1i-VL Discussion

(19.78 degrees/year) angles. Consequently, a statistically significant
Summary of evidence
decrease in the interincisal angle was also noted (−8.32 degrees/year).
The influence of FFAs on the soft tissues was significant for almost This systematic review included data from 418 subjects (244 Class
all available outcomes, with the mentolabial angle providing the more II patients and 174 untreated individuals) and 10 RCTs and pCCTs,
evident change (14.99 degrees/year). Further, the H-angle was slightly which assessed angular cephalometric changes induced by Class II
decreased (−1.95 degree/year), while the N′SnPg′ angle was slightly treatment with FFAs. With regard to the ANB angle, the results
increased (2.01 degrees/year) compared to untreated patients. Finally, from the random-effects meta-analyses indicated that FFAs had a
no significant changes were observed regarding the cephalometric statistically significant contribution in the improvement of skeletal
ratios investigated. Class II relationship (as seen by the average annualized decrease of
1.74 degree of the ANB angle of the treated patients in comparison
to untreated controls). This improvement was accomplished with
Risk of bias across studies approximately equal contributions from mandibular growth aug-
The GRADE assessments for the main outcomes after appliance mentation (0.87 degree/year) and restriction of maxillary growth

Downloaded from https://academic.oup.com/ejo/article-abstract/38/2/113/2599887 by guest on 20 December 2018

removal ranged from low to high (Table 5). The quality of clinical (0.83 degree/year). However, the skeletal contribution in the sagittal
recommendations was upgraded due to the magnitude of treatment plane to the correction of Class II malocclusion can be considered
effects for the upper incisor retroclination (via the 1s-SN angle) and clinically small in concordance with previous studies (8, 14, 15).
the proclination of the mandibular incisors (via the 1i-ML angle). Furthermore, in agreement with previous findings (67–69), the FFA
All judgements made for the GRADE analysis of each outcome are treatment was associated with a significant increase of inclination of
presented in detail in Supplementary Table 4. the occlusal plane relative to anterior cranial base.
The effects of FFAs were more pronounced on the dentition,
Additional analyses where the maxillary incisors were significantly retroclined and the
The results of the included studies varied considerably and hetero- mandibular incisors were significantly proclined, resulting, thus, in
geneity influenced the results of FFAs according to various patient- a greater interincisal angle, agreeing with previous evidence (13,
related factors (Table  6). Skeletal correction and facial convexity 69–71). Contradictory results regarding the effect of FFAs on soft
(via the ANB and NA-APg angles, respectively) were significantly tissues have been reported (13, 23, 69, 72). Based on the results of
associated with patient’s gender. Post-peak patients showed a this study, soft tissues were significantly affected in favour of profile
greater dentoalveolar effect with a greater emphasis on SN-NL and improvement by FFAs. Finally, no significant changes were observed
1i-ML angles and a greater reduction in 1s-SN angle compared with in the investigated cephalometric ratios.
patients at pre-peak and peak skeletal growth stage. Finally, the Several patient-related factors might explain the considerable
impact of growth pattern on the treatment outcome was planned variation in the treatment outcome among the included studies. The
to be assessed yet was not possible due to insufficient data reported impact of patient’s gender on the respective treatment could not be
in the included trials. formally assessed in this study with separate sub-populations of
With regard to the appliance-related factors, for mandibular sag- male and female patients due to the incomplete reporting of data in
ittal growth (via the SNB angle) and skeletal Class II correction (via the original articles. The results of the meta-regression with the male/
the ANB angle), no statistical differences were observed, whereas female ratio from each study indicate that the Class II correction and
the Forsus™ Fatigue Resistant Device was associated with the great- the skeletal facial convexity (through the ANB and NA-APg angles,
est proclination of the mandibular incisors. In addition, stepwise respectively) directly after the appliance removal might be different
mandibular advancement was associated with greater retroclination between boys and girls, but further research is required to confirm or
of the upper incisors and greater proclination of the lower incisors refute this. However, in a retrospective study, Rizell et al. (73) failed
compared to single step advancement. to identify a statistically significant association between gender and
Due to the limited number of included studies, an evaluation for outcome of functional treatment.
the existence of reporting bias (including publication bias) was not The investigation of the patients’ growth pattern and its influence
possible to be performed. on the FFA treatment results was originally planned but was not pos-
Sensitivity analysis on the basis of treatment duration sible due to the lack of data.
(Supplementary Table  5) indicated that the duration of FFA treat- When the skeletal growth stage is taken into account, outcomes
ment was significantly associated with the inclination of the upper of FFA treatment were found to differ between patients before or
and the lower incisors (via the 1s-SN and 1i-ML angles, respec- during and patients after the growth peak. According to a previous
tively). Furthermore, the sensitivity analysis based on the design of study, dentoalveolar effects seem to increase in post-peak patients
the included studies (Supplementary Table 6) found no statistically (74). Moreover, skeletal effects seem to be more pronounced in
significant differences between pCCTs and RCTs for the investi- patients treated before (74, 75) or during the growth peak (8). The
gated angular measurements. Sensitivity analysis on the basis of the results of the present study indicate a trend towards more favourable
reported method error for each outcome could not be performed due skeletal sagittal changes in the pre-peak/peak than in the post-peak
to incomplete reporting from the included studies. Finally, sensitivity growth stage although no statistical inference could be confirmed.
analyses on the basis of the GRADE quality of recommendations With regard to appliance-related factors, considerable differences
were not feasible, as the only reason for downgrading the quality of were found among the seven different FFAs used in the included stud-
recommendations was the inclusion of pCCTs over RCTs. ies. Among them, the Forsus™ Fatigue Resistant Device was associ-
ated with the greatest proclination of the lower incisors (via the 1i-ML
Effectiveness of FFA treatment after retention angle). A comparison of the effectiveness of the included FFAs is pre-
Due to inadequate number of identified studies, no meta-analyses sented in Supplementary Figures 1–3 regarding three skeletal variables
could be performed concerning the changes induced by FFAs after and in Supplementary Figures 4 and 5 regarding two dentoalveolar
the retention phase. variables. However, these comparisons should be interpreted with
124 European Journal of Orthodontics, 2016, Vol. 38, No. 2

caution due to the presence of considerable heterogeneity and their planned patient- and appliance-related factors in this review while
indirect nature. In a comparative evaluation of the Forsus™ Nitinol reporting biases could not be formally assessed (80).
Flat Spring and the Jasper Jumper (15), the superiority of the latter
in advancing the mandible was reported. Finally, contrary to the pre-
sent study, a previous systematic review (8) reported that the Herbst Conclusions
appliance was found to produce more favourable results in mat- According to existing evidence, the following conclusions can be
ters of increased mandibular growth compared to other functional drawn on the short-term effectiveness of FFAs:
appliances. However, only the Herbst appliance and the Mandibular
Anterior Repositioning Appliance were included from FFAs, while no 1. The treatment effects of FFAs on the skeletal tissues in patients
quantitative synthesis was conducted. with Class  II malocclusion excluding the effects of normal
The three different designs of the Herbst appliance that were growth were small and probably of minor clinical importance.
used in the included studies were compared separately, with some 2. The treatment of Class II malocclusion with FFAs was associated
significant differences among them (Table  5). Burkhardt et  al. with small stimulation of mandibular growth, small inhibition of
(76) compared acrylic-splint and stainless-steel crown Herbst and maxillary growth, and with more pronounced dentoalveolar and

Downloaded from https://academic.oup.com/ejo/article-abstract/38/2/113/2599887 by guest on 20 December 2018

reported that both the appliances produced similar skeletal changes. soft tissue changes.
According to a previous assessment of the efficacy of three different 3. Patient- and appliance-related factors seem to influence the treat-
Herbst anchorage systems (77), greater lower incisor proclination ment outcomes, yet complementary research is required to thor-
was observed with the cast-splint design than in the banded designs oughly investigate the respective effects.
although none of the designs could prevent mandibular anchorage 4. The long-term effects of FFAs could not be properly assessed
loss. In the last years, attempts have been made to prevent anterior because of insufficient number of relative trials at present.
anchorage loss during mandibular protraction by utilizing minis- Taking into account the clinical recommendations derived from
crew implants, however, without particular success (78). the GRADE framework, high GRADE assessments could be drawn
Moreover, the comparison between stepwise activation and regarding the 1s-SN and 1i-ML angles exclusively. Particularly:
maximum mandibular advancement did not reveal any significant
differences, with the exception of the greater maxillary incisor ret- 1. Clinicians should confidently expect an average reduction in the
roclination that was observed with the former. Contrary to these 1s-SN angle of 7.50 degree/year with the use of FFAs compared
results, a controlled trial investigating the effects of the Herbst appli- to untreated patients.
ance (72), reported that stepwise mandibular advancement produced 2. Clinicians should confidently expect an average increase in the
greater skeletal changes compared to single step advancement, while 1i-ML angle of 7.99 degree/year with the use of FFAs compared
the dentoalveolar changes were more pronounced when the activa- to untreated patients.
tion was performed at a single step.
Recommendations concerning the effectiveness of FFA treatment on
the restriction of maxillary growth, advancement of the mandible,
Strengths and limitations
correction of skeletal Class II malocclusion, mandibular plane, and
The strengths of the present review include the predefined protocol, nasolabial angles are weaker and future research could affect them.
the thorough and unrestricted literature search, and the strict meth- Treatment of Class II malocclusion with FFAs seems to be not as
odology that was carefully followed during every stage of it follow- effective as believed in matters of skeletal correction. Additional stud-
ing specific and detailed guidelines (31–33, 46). In addition, the 10 ies are required for a thorough assessment of the skeletal, dental, and
included studies enabled adequately powered meta-analyses for many soft tissue outcomes of FFAs in the long term. The provision of detailed
important treatment outcomes, for which the required meta-analysis data from these studies regarding patients’ characteristics (gender,
sample size was fulfilled almost exclusively. Since a random-effects growth pattern, and skeletal maturation), particular features of the
model was used for data synthesis, the results of the present study used functional appliance (the exact appliance design and possible
provide the average of the FFA effects across the included studies. incorporation of additional elements), as well as the followed retention
Heterogeneity was explained in most cases by the predefined sub- scheme should be considered. Finally, in order to enable also the assess-
group analyses, while sensitivity analyses indicated that the results ment of linear variables, the magnification factor of the lateral cepha-
were fairly robust. In addition, data from a minimum of five studies lometric radiographs should be reported in each of the respective trials.
were considered as adequate to perform a subgroup analysis in order
to minimize multiple testing. Although most of the included trials
took place at university settings, their findings could possibly be gen- Supplementary material
eralized to the average patient due to their broad inclusion criteria.
Supplementary material is available at European Journal of
Finally, the vast majority of the included trials (89 %) were pCCTs,
Orthodontics online.
which represent more ‘realistic’ situations in matters of the daily clini-
cal practice, compared to RCTs, and thus possibly strengthening the
applicability of the outcomes. Acknowledgements
Nevertheless, although no serious methodological limitations
We thank the following author for providing clarifications for the systematic
were found in the included original studies, their quality could have
review: O. Dalci (Discipline of Orthodontics, Faculty of Dentistry, University
been better. Moreover, none of the included studies provided results
of Sydney, Sydney, Australia).
concerning the number and the experience of the respective clini-
cians, which could introduce proficiency bias (79). Unfortunately,
treatment results concerning the long-term effects of FFAs were not References
reported in the selected studies, precluding an assessment of the 1. Pancherz, H., Zieber, K. and Hoyer, B. (1997) Cephalometric characteris-
results’ stability in current evaluation. Finally, the limited number tics of class II division 1 and class II division 2 malocclusions: a compara-
of the eligible studies prevented the investigation of all the originally tive study in children. The Angle Orthodontist, 67, 111–120.
V. F. Zymperdikas et al. 125

2. Kim, Y.H. (1979) A comparative cephalometric study of class II, Division 22. Flores-Mir, C. and Major, P.W. (2006) Cephalometric facial soft tissue
1 nonextraction and extraction cases. The Angle Orthodontist, 49, 77–84. changes with the twin block appliance in class II division 1 malocclusion
3. McNamara, J.A., Jr (1981) Components of class II malocclusion in chil- patients. A systematic review. The Angle Orthodontist, 76, 876–881.
dren 8–10 years of age. The Angle Orthodontist, 51, 177–202. 23. Flores-Mir, C., Major, M.P. and Major, P.W. (2006) Soft tissue changes
4. Paulsen, H.U., Karle, A., Bakke, M. and Herskind, A. (1995) CT-scanning with fixed functional appliances in class II division 1. The Angle Ortho-
and radiographic analysis of temporomandibular joints and cephalometric dontist, 76, 712–720.
analysis in a case of Herbst treatment in late puberty. European Journal of 24. Antonarakis, G.S. and Kiliaridis, S. (2007) Short-term anteroposterior

Orthodontics, 17, 165–175. treatment effects of functional appliances and extraoral traction on class II
5. McNamara, J.A., Jr, Howe, R.P. and Dischinger, T.G. (1990) A comparison malocclusion. A meta-analysis. The Angle Orthodontist, 77, 907–914.
of the Herbst and Fränkel appliances in the treatment of class II malocclu- 25. Harrison, J.E., O’Brien, K.D. and Worthington, H.V. (2007) Orthodontic
sion. American Journal of Orthodontics and Dentofacial Orthopedics, 98, treatment for prominent upper front teeth in children. Cochrane Database
134–144. of Systematic Reviews, 3, CD003452.
6. Franchi, L., Baccetti, T. and McNamara, J.A. Jr (1999) Treatment and 26. Barnett, G.A., Higgins, D.W., Major, P.W. and Flores-Mir, C. (2008) Imme-
posttreatment effects of acrylic splint Herbst appliance therapy. American diate skeletal and dentoalveolar effects of the crown- or banded type
Journal of Orthodontics and Dentofacial Orthopedics, 115, 429–438. Herbst appliance on class  II division 1 malocclusion. The Angle Ortho-

Downloaded from https://academic.oup.com/ejo/article-abstract/38/2/113/2599887 by guest on 20 December 2018

7. Woodside, D.G., Metaxas, A. and Altuna, G. (1987) The influence of func- dontist, 78, 361–369.
tional appliance therapy on glenoid fossa remodeling. American Journal of 27. Bertl, M.H., Mandl, C. and Crismani, A.G. (2011) Do functional ortho-
Orthodontics and Dentofacial Orthopedics, 92, 181–198. dontic appliances stimulate mandibular growth in class  II division 1
8. Cozza, P., Baccetti, T., Franchi, L., De Toffol, L. and McNamara, J.A., Jr patients? International Journal of Stomatology & Occlusion Medicine, 4,
(2006) Mandibular changes produced by functional appliances in class II 45–53.
malocclusion: a systematic review. American Journal of Orthodontics and 28. Niu, Y. and Zhou, H. (2011) Effect on functional appliances on mandib-
Dentofacial Orthopedics, 129, 599.e1–e12. ular growth on skeletal class  II malocclusion: a systematic review. West
9. Flores-Mir, C., Ayeh, A., Goswani, A. and Charkhandeh, S. (2007) Skeletal China Journal of Stomatology, 29, 384–388.
and dental changes in class II division 1 malocclusions treated with splint- 29. Perillo, L., Cannavale, R., Ferro, F., Franchi, L., Masucci, C., Chiodini, P.
type Herbst appliances. A systematic review. The Angle Orthodontist, 77, and Baccetti, T. (2011) Meta-analysis of skeletal mandibular changes dur-
376–381. ing Frankel appliance treatment. European Journal of Orthodontics, 33,
10. Covell, D.A., Jr, Trammell, D.W., Boero, R.P. and West, R. (1999) A cepha- 84–92.
lometric study of class II Division 1 malocclusions treated with the Jasper 30. Perinetti, G., Primožič, J., Furlani, G., Franchi, L. and Contardo, L. (2014)
Jumper appliance. The Angle Orthodontist, 69, 311–320. Treatment effects of fixed functional appliances alone or in combination
11. Croft, R.S., Buschang, P.H., English, J.D. and Meyer, R. (1999) A cephalo- with multibracket appliances: a systematic review and meta-analysis. The
metric and tomographic evaluation of Herbst treatment in the mixed den- Angle Orthodontist. First published on September 4, 2014.
tition. American Journal of Orthodontics and Dentofacial Orthopedics, 31. Higgins, J.P.T. and Green, S. (2011) Cochrane handbook for system-

116, 435–443. atic reviews of interventions [version 5.1.0, updated March  2011]. The
12. Hägglund, P., Segerdal, S. and Forsberg, C.M. (2008) The integrated
Cochrane Collaboration. www.cochrane-handbook.org (15 March 2015,
Herbst appliance—treatment effects in a group of adolescent males with date last accessed).
class II malocclusions compared with growth changes in an untreated con- 32. Liberati, A., Altman, D.G., Tetzlaff, J., Mulrow, C., Gøtzsche, P.C., Ioan-
trol group. European Journal of Orthodontics, 30, 120–127. nidis, J.P., Clarke, M., Devereaux, P.J., Kleijnen, J. and Moher, D. (2009)
13. Küçükkeleş, N., Ilhan, I. and Orgun, I.A. (2007) Treatment efficiency in The PRISMA statement for reporting systematic reviews and meta-anal-
skeletal class II patients treated with the Jasper Jumper. The Angle Ortho- yses of studies that evaluate health care interventions: explanation and
dontist, 77, 449–456. elaboration. Journal of Clinical Epidemiology, 62, e1–e34.
14. Lai, M. and McNamara, J.A., Jr (1998) An evaluation of two-phase treat- 33. Beller, E.M., Glasziou, P.P., Altman, D.G., Hopewell, S., Bastian, H., Chal-
ment with the Herbst appliance and preadjusted edgewise therapy. Semi- mers, I., Gøtzsche, P.C., Lasserson, T., Tovey, D. and PRISMA for Abstracts
nars in Orthodontics, 4, 46–58. Group (2013) PRISMA for abstracts: reporting systematic reviews in jour-
15. Darda, M., Goel, S. and Gupta, R. (2010) A cephalometric comparison of nal and conference abstracts. PLoS Medicine, 10, e1001419.
the dentoskeletal changes in class II malocclusion by using Jasper Jumper 34. Adams, J.W. (1940) Correction of errors in cephalometric roentgeno-

and Forsus-A clinical study. International Journal of Contemporary Den- grams. The Angle Orthodontist, 10, 3–13.
tistry, 1, 79–86. 35. Athanasiou, A.E. (1995) Orthodontic Cephalometry. Mosby-Wolfe, Lon-
16. Cope, J.B., Buschang, P.H., Cope, D.D., Parker, J. and Blackwood, H.O., don, UK.
III (1994) Quantitative evaluation of craniofacial changes with Jasper 36. Gandini, P., Mancini, M. and Andreani, F. (2006) A comparison of hand-
Jumper therapy. The Angle Orthodontist, 64, 113–122. wrist bone and cervical vertebral analyses in measuring skeletal matura-
17. Koretsi, V., Zymperdikas, V.F., Papageorgiou, S.N. and Papadopoulos, M.A. tion. The Angle Orthodontist, 76, 984–989.
(2014) Treatment effects of removable functional appliances in patients 37. Rasool, G., Bashir, U. and Kundi, I. (2010) Comparative evaluation

with class II malocclusion: a systematic review and meta-analysis. European between cervical vertebrae and hand-wrist maturation for assessment of
Journal of Orthodontics. First published on November 13, 2014. skeletal maturity orthodontic patients. Pakistan Oral & Dental Journal,
18. Schäfer, K., Ludwig, B., Meyer-Gutknecht, H. and Schott, T.C. (2015) 30, 85–95.
Quantifying patient adherence during active orthodontic treatment with 38. Alkhal, H.A., Wong, R.W. and Rabie, A.B. (2008) Correlation between
removable appliances using microelectronic wear-time documentation. chronological age, cervical vertebral maturation and Fishman’s skeletal
European Journal of Orthodontics, 37, 73–80. maturity indicators in Southern Chinese. The Angle Orthodontist, 78,
19. Sahm, G., Bartsch, A. and Witt, E. (1990) Micro-electronic monitoring of 591–596.
functional appliance wear. European Journal of Orthodontics, 12, 297–301. 39. Bala, M., Pathak, A. and Jain, R.L. (2010) Assessment of skeletal age using
20. Chen, J.Y., Will, L.A. and Niederman, R. (2002) Analysis of efficacy of MP 3 and hand-wrist radiographs and its correlation with dental and
functional appliances on mandibular growth. American Journal of Ortho- chronological ages in children. Journal of the Indian Society of Pedodon-
dontics and Dentofacial Orthopedics, 122, 470–476. tics and Preventive Dentistry, 28, 95–99.
21. Flores-Mir, C. and Major, P.W. (2006) A systematic review of cephalo- 40. Downs, S.H. and Black, N. (1998) The feasibility of creating a checklist for
metric facial soft tissue changes with the Activator and Bionator appli- the assessment of the methodological quality both of randomised and non-
ances in class II division 1 subjects. European Journal of Orthodontics, 28, randomised studies of health care interventions. Journal of Epidemiology
586–593. and Community Health, 52, 377–384.
126 European Journal of Orthodontics, 2016, Vol. 38, No. 2

41. Harmelink, K. (2010) Effects of neurolytic blocks (botulinum toxin type 62. Karacay, S., Akin, E., Olmez, H., Gurton, A.U. and Sagdic, D. (2006) For-
A and motor branch block) in patients with a Stiff Knee Gait: a systematic sus nitinol flat spring and Jasper jumper corrections of class II division 1
review. Master thesis, Utrecht University. malocclusions. The Angle Orthodontist, 76, 666–672.
42. Peters, J.L., Sutton, A.J., Jones, D.R., Abrams, K.R. and Rushton, L. (2008) 63. Latkauskienė, D. (2012) Treatment of angle class  II malocclusion with the
Contour-enhanced meta-analysis funnel plots help distinguish publication crown Herbst appliance. Doctoral dissertation, Kauno medicinos universitetas.
bias from other causes of asymmetry. Journal of Clinical Epidemiology, 64. Oztoprak, M.O., Nalbantgil, D., Uyanlar, A. and Arun, T. (2012) A cepha-
61, 991–996. lometric comparative study of class II correction with Sabbagh Universal
43. Begg, C.B. and Mazumdar, M. (1994) Operating characteristics of a rank Spring (SUS(2)) and Forsus FRD appliances. European Journal of Den-
correlation test for publication bias. Biometrics, 50, 1088–1101. tistry, 6, 302–310.
44. Egger, M., Smith, G.D., Schneider, M. and Minder, C. (1997) Bias in meta- 65. Phelan, A., Tarraf, N.E., Taylor, P., Hönscheid, R., Drescher, D., Baccetti,
analysis detected by a simple, graphical test. British Medical Journal, 315, T. and Darendeliler, M.A. (2012) Skeletal and dental outcomes of a new
629–634. magnetic functional appliance, the Sydney Magnoglide, in class II correc-
45. Duval, S. and Tweedie, R. (2000) Trim and fill: a simple funnel-plot-based tion. American Journal of Orthodontics and Dentofacial Orthopedics,
method of testing and adjusting for publication bias in meta-analysis. 141, 759–772.
Biometrics, 56, 455–463. 66. Uyanlar, A., Nalbantgil, D. and Arun, T. (2014) Evaluation of dentofacial

Downloaded from https://academic.oup.com/ejo/article-abstract/38/2/113/2599887 by guest on 20 December 2018

46. Guyatt, G.H., Oxman, A.D., Schünemann, H.J., Tugwell, P. and
changes caused with Sabbagh universal spring in mandibular retrognathic
Knottnerus, A. (2011) GRADE guidelines: a new series of articles in the patients: a pilot study. Dentistry, 4, 237.
Journal of Clinical Epidemiology. Journal of Clinical Epidemiology, 64, 67. Alvares, J.C.D.C., Cançado, R.H., Valarelli, F.P., Freitas, K.M.S.D. and
380–382. Angheben, C.Z. (2013) class  II malocclusion treatment with the Herbst
47. Sloan, J., Symonds, T., Vargas-Chanes, D. and Fridley, B. (2003) Practical appliance in patients after the growth peak. Dental Press Journal of
guidelines for assessing the clinical significance of health-related quality of Orthodontics, 18, 38–45.
life changes within clinical trials. Drug Information Journal, 37, 23–31. 68. Siara-Olds, N.J., Pangrazio-Kulbersh, V., Berger, J. and Bayirli, B. (2010)
48. Morissette, K., Tricco, A.C., Horsley, T., Chen, M.H. and Moher, D. (2011) Long-term dentoskeletal changes with the Bionator, Herbst, Twin Block,
Blinded versus unblended unblinded assessments of risk of bias in stud- and MARA functional appliances. The Angle Orthodontist, 80, 18–29.
ies included in a systematic review. The Cochrane Database of Systematic 69. Nalbantgil, D., Arun, T., Sayinsu, K. and Fulya, I. (2005) Skeletal, dental
Reviews, 9, MR000025. and soft-tissue changes induced by the Jasper Jumper appliance in late
49. Cohen, J. (1960) A coefficient of agreement for nominal scales. Educa- adolescence. The Angle Orthodontist, 75, 426–436.
tional and Psychological Measurement, 20, 37–46. 70. Heinig, N. and Göz, G. (2001) Clinical application and effects of the For-
50. DerSimonian, R. and Laird, N. (1986) Meta-analysis in clinical trials. Con- sus spring. A study of a new Herbst hybrid. Journal of Orofacial Orthope-
trolled Clinical Trials, 7, 177–188. dics, 62, 436–450.
51. Papageorgiou, S.N. (2014) Meta-analysis for orthodontists: Part I—How 71. Cacciatore, G., Ghislanzoni, L.T., Alvetro, L., Giuntini, V. and Franchi, L.
to choose effect measure and statistical model. Journal of Orthodontics, (2014) Treatment and posttreatment effects induced by the Forsus appli-
41, 317–326. ance: a controlled clinical study. The Angle Orthodontist, 84, 1010–1017.
52. Orsini, N., Bottai, M., Higgins, J. and Buchan, I. (2006) Heterogi: Stata 72. Purkayastha, S.K., Rabie, A.B. and Wong, R. (2008) Treatment of skeletal
module to quantify heterogeneity in a meta-analysis. Statistical Software class II malocclusion in adults: stepwise vs single-step advancement with
Components. www.EconPapers.repec.org/RePEc:boc:bocode:s449201 (15 the Herbst appliance. World Journal of Orthodontics, 9, 233–243.
March 2015, date last accessed). 73. Rizell, S., Svensson, B., Tengström, C. and Kjellberg, H. (2006) Functional
53. Higgins, J.P., Thompson, S.G. and Spiegelhalter, D.J. (2009) A re-evalua- appliance treatment outcome and need for additional orthodontic treat-
tion of random effects meta-analysis. Journal of the Royal Statistical Soci- ment with fixed appliance. Swedish Dental Journal, 30, 61–68.
ety: Series A (Statistics in Society), 172, 137–159. 74. Frye, L., Diedrich, P.R. and Kinzinger, G.S. (2009) class II treatment with
54. Graham, P.L. and Moran, J.L. (2012) Robust meta-analytic conclusions fixed functional orthodontic appliances before and after the pubertal
mandate the provision of prediction intervals in meta-analysis summaries. growth peak—a cephalometric study to evaluate differential therapeutic
Journal of Clinical Epidemiology, 65, 503–510. effects. Journal of Orofacial Orthopedics, 70, 511–527.
55. Ioannidis, J.P. (2008) Interpretation of tests of heterogeneity and bias in 75. Aras, A., Ada, E., Saracoğlu, H., Gezer, N.S. and Aras, I. (2011) Com-
meta-analysis. Journal of Evaluation in Clinical Practice, 14, 951–957. parison of treatments with the Forsus fatigue resistant device in relation to
56. Knapp, G. and Hartung, J. (2003) Improved tests for a random-effects skeletal maturity: a cephalometric and magnetic resonance imaging study.
meta-regression with a single covariate. Statistics in Medicine, 22, 2693– American Journal of Orthodontics and Dentofacial Orthopedics, 140,
2710. 616–625.
57. Alali, O.H. (2014) A prospective controlled evaluation of class II division 76. Burkhardt, D.R., McNamara, J.A., Jr and Baccetti, T. (2003) Maxillary
1 malocclusions treated with fixed lingual mandibular growth modificator. molar distalization or mandibular enhancement: a cephalometric compar-
The Angle Orthodontist, 84, 527–533. ison of comprehensive orthodontic treatment including the pendulum and
58. Baysal, A. and Uysal, T. (2011) Soft tissue effects of Twin Block and Herbst the Herbst appliances. American Journal of Orthodontics and Dentofacial
appliances in patients with class  II division 1 mandibular retrognathy. Orthopedics, 123, 108–116.
European Journal of Orthodontics, 35, 71–81. 77. Weschler, D. and Pancherz, H. (2005) Efficiency of three mandibular

59. de Almeida, M.R., Henriques, J.F., de Almeida, R.R., Weber, U. and McNa- anchorage forms in Herbst treatment: a cephalometric investigation. The
mara, J.A., Jr (2005) Short-term treatment effects produced by the Herbst Angle Orthodontist, 75, 23–27.
appliance in the mixed dentition. The Angle Orthodontist, 75, 540–547. 78. Bremen, J.V., Ludwig, B. and Ruf, S. (2014) Anchorage loss due to Herbst
60. de Almeida, M.R., Flores-Mir, C., Brandão, A.G., de Almeida, R.R. and de mechanics-preventable through miniscrews? European Journal of Ortho-
Almeida-Pedrin, R.R. (2008) Soft tissue changes produced by a banded- dontics. First published on December 1, 2014.
type Herbst appliance in late mixed dentition patients. World Journal of 79. Krishna, R., Maithreyi, R. and Surapaneni, K.M. (2010) Research bias: a
Orthodontics, 9, 121–131. review for medical students. Journal of Clinical and Diagnostic Research,
61. Gunay, E.A., Arun, T. and Nalbantgil, D. (2011) Evaluation of the imme- 4, 2320–2324.
diate dentofacial changes in late adolescent patients treated with the For- 80. Papageorgiou, S.N. (2014) Meta-analysis for orthodontists: Part II—is all
sus(™) FRD. European Journal of Dentistry, 5, 423–432. that glitters gold? Journal of Orthodontics, 41, 327–336.