Long-term effects of functional appliances in treated versus untreated patients with Class II malocclusion: A systematic review and meta-analysis

Objective To assess the cephalometric skeletal and soft-tissue of functional appliances in treated versus untreated Class II subjects in the long-term (primarily at the end of growth, secondarily at least 3 years after retention). Search methods Unrestricted electronic search of 24 databases and additional manual searches up to March 2018. Selection criteria Randomised and non-randomised controlled trials reporting on cephalometric skeletal and soft-tissue measurements of Class II patients (aged 16 years or under) treated with functional appliances, worn alone or in combination with multi-bracket therapy, compared to untreated Class II subjects. Data collection and analysis Mean differences (MDs) and 95% confidence intervals (95% CIs) were calculated with the random-effects model. Data were analysed at 2 primary time points (above 18 years of age, at the end of growth according to the Cervical Vertebral Maturation method) and a secondary time point (at least 3 years after retention). The risk of bias and quality of evidence were assessed according to the ROBINS tool and GRADE system, respectively. Results Eight non-randomised studies published in 12 papers were included. Functional appliances produced a significant improvement of the maxillo-mandibular relationship, at almost all time points (Wits appraisal at the end of growth, MD -3.52 mm, 95% CI -5.11 to -1.93, P < 0.0001). The greatest increase in mandibular length was recorded in patients aged 18 years and above (Co-Gn, MD 3.20 mm, 95% CI 1.32 to 5.08, P = 0.0009), although the improvement of the mandibular projection was negligible or not significant. The quality of evidence was ‘very low’ for most of the outcomes at both primary time points. Conclusions Functional appliances may be effective in correcting skeletal Class II malocclusion in the long-term, however the quality of the evidence was very low and the clinical significance was limited. Systematic review registration CRD42018092139


Selection criteria
Randomised and non-randomised controlled trials reporting on cephalometric skeletal and soft-tissue measurements of Class II patients (aged 16 years or under) treated with functional appliances, worn alone or in combination with multi-bracket therapy, compared to untreated Class II subjects.

Data collection and analysis
Mean differences (MDs) and 95% confidence intervals (95% CIs) were calculated with the random-effects model. Data were analysed at 2 primary time points (above 18 years of age, at the end of growth according to the Cervical Vertebral Maturation method) and a secondary time point (at least 3 years after retention). The risk of bias and quality of evidence were assessed according to the ROBINS tool and GRADE system, respectively. PLOS

Rationale
Class II malocclusion is the most prevalent antero-posterior jaw problem in orthodontics, affecting one third of the population [1,2]. The majority of Class II patients exhibit mandibular skeletal retrusion [3,4]. Reduced mandibular size is also a major feature of Class II malocclusion patients [5]. As a result, there has been great interest in the use of 'functional appliances', designed primarily to influence the lower dentition and enhance the growth of the mandible [3]. These appliances promote forward posturing of the mandible, although their effects also impact on the upper jaw [6,7]. The potential that functional appliances could modify skeletal growth is of great importance for patients and orthodontists alike. Improving facial aesthetics is one of the main reasons for seeking orthodontic treatment [8] and it is associated with a high level of patient and parent satisfaction [9]. Mandibular retrusion has a negative impact on perceived attractiveness [10], self-esteem and oral health-related quality of life [11]. The magnitude of the retrusion is also an important factor in treatment decision-making. Small skeletal discrepancies may only need multi-bracket therapy for the correction of malocclusion and refinement of teeth alignment. On the other hand, greater discrepancies may require a surgical treatment to modify the position and length of skeletal structures and to attain better aesthetic results [12].
Post-pubertal growth has been shown to produce dramatic alterations in skeletal and dental relationships [13]. There is no consensus on the age at which growth ends [14][15][16][17][18]. Overall, growth continues up to mid-adulthood, with different patterns in the two genders. Males show an anterior rotation of the mandible, whereas females demonstrate a posterior mandibular rotation [17,18]. An alternative method to establish when growth comes to an end is through using indicators of the growth phase, such as the hand-and-wrist maturation method [19] or the cervical vertebral maturation method [20].
To fully understand the real effects of functional appliances on the growth of the jaws and profile, it is essential to study these effects at the completion of patient growth, when biases and confounding factors due to natural changes are negligible. The long-term stability of these changes is important too.
To date, most systematic reviews investigating the treatment effects of functional appliances in Class II malocclusion patients have synthesized studies evaluating the skeletal and soft-tissue changes at the end of the orthodontic treatment [6,7,[21][22][23][24][25][26]. Only two reviews systematically searched for scientific evidence concerning the long-term stability of treatment results achieved by Class II functional appliance therapy [27,28]. Another systematic review is ongoing [29]. No previous reviews determined the effects of removable and fixed functional appliances in patients with Class II malocclusion compared to untreated controls at growth completion.

Objective
The objective of this systematic review was therefore to assess the skeletal and soft-tissue effects measured on lateral cephalograms produced by functional appliances in treated versus untreated Class II subjects in the long-term (primarily at the end of growth, secondarily at least 3 years after retention).

Protocol and registration
The present systematic review was performed according to the guidelines of the Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [30], and is reported on the basis of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (S1 table [31]). The protocol was published in the International Prospective Register of Systematic Reviews (PROSPERO) on 03 April 2018 (registration number CRD42018092139).

Information sources
The search strategy covered 11 bibliographic databases, 10 non-bibliographic databases and 3 unpublished studies sources, from their launch to March 2018 [32][33][34][35]. Hand-searching of the most common orthodontic journals was performed as well. The Cochrane Master List was consulted to facilitate the identification of these journals [30,34,36]. The reference lists of the trials eligible for inclusion and systematic reviews concerning Class II malocclusion treatment were also checked. Information concerning the name of the search source, the date range that were searched, and, for electronic databases, the search platform or provider are presented in S2 table.

Additional analysis
Subgroup and sensitivity analyses were performed in order to explore the source of heterogeneity and test the overall robustness of the data, respectively. All subgroup and sensitivity analyses were pre-specified in the protocol.
For all outcomes, results were divided according to the type of functional appliance. For the most clinically important outcomes, subgroup analyses were based on the following: ■ Patient characteristics (gender); ■ Beginning of the functional appliance therapy according to age (early treatments, commencing in children aged between 7 and 11 years; late treatments, beginning in adolescents aged between 12 and 16 years); ■ Start of the treatment according to the cervical vertebral maturation method (early treatments, with patients presenting with Cervical Vertebral Maturation Stage [CVMS] 1 or 2 at the first observation; late treatments, with subjects presenting with CVMS 2 or 3); ■ Post-retention period duration (3-4, 5-10 years after active treatment with functional appliances); Sensitivity analysis was performed to examine the impact of the study quality assessment on the overall estimates of effect.

Risk of bias across studies
Outcome reporting bias and publication bias were evaluated. In order to determine whether reporting bias was present, the Clinical Trial Register was screened using the International Clinical Trials Registry Platform of the World Health Organisation (http://apps.who.int/ trialssearch). When protocols were identified, discrepancies between the outcomes planned in the protocol and those reported in the final manuscript were assessed. The potential for reporting bias was explored by funnel plots if 10 studies were available [40].
The quality of evidence for all outcomes at both primary time points was judged using the Grading of Recommendations Assessment, Development and Evaluation working group methodology [41].

Study selection
The results of the search are summarised in Fig 1. Among 3046 records, eight non-randomised studies published in 12 papers were identified for inclusion in this review [42][43][44][45][46][47][48][49]. Two authors were contacted to clarify whether duplicate data was used in their trials. Since the study by Pavoni et al. [43] contained partial data of previous studies [50][51][52] and has the greater sample size and subgroup analysis, it was considered the reference study of the other reports. The thesis by Wigal [47] with complete data of the subsequent published study [53] was included as well. Excluded studies with reasons are listed in supplementary files (S4 Table, S2 Appendix).

Risk of bias within studies
The overall risk of bias ranged from moderate to critical in the included studies (Table 4). Most studies suffered bias in selection of participants and due to deviations from intended interventions [42][43][44][45][46][47][48][49]. The estimated effect can be predicted to be greater than the true effect estimate in studies with the observed selection bias [42,43,49]. Multi-bracket therapy, as well as retention appliances, could enhance the treatment effects of functional jaw orthopaedics or control their relapse [43,[47][48][49]. Table 3. Characteristics of included studies (timing).

Results of individual studies
The main results of the included studies are reported in S5-S6 Tables. Only one report [51] found that Bionator therapy was able to significantly alter the sagittal position of both the maxillary and mandibular soft tissue profile components. During the overall observation period, functional jaw orthopaedics with the Bionator, followed by multibracket appliances produced a restraining effect on the soft tissue A point (-1.8 mm, CI not reported) and a protrusive effect on the soft tissue Pg point (+2.6 mm, CI not reported).

Synthesis of results
Seven studies (10 subgroups [42,43,[45][46][47][48][49]) were included in the meta-analyses of 9 outcomes at 3 time points (Table 5). Subgroup analyses according to the type of functional appliance are presented together with their overall effects (Tables 6-7). The forest plots concerning the most clinically relevant results are reported in the main text. Other findings are set out in S3 Appendix.
The most clinically relevant maxillary effects were produced by fixed functional appliances: the Herbst appliance (Co-A distance at least 3 years after retention, MD -4.08 mm, 95% CI -6.03 to -2.12, P < 0.0001, I 2 = 0%, 2 studies [ Table 7]) and the Forsus device, in combination with multi-bracket therapy (A to N perpendicular distance above 18 years of age, MD -6.30 mm, 95% CI -7.01 to -5.59, P < 0.00001, I 2 = Not applicable, 1 study [ Table 7]). Mandibular/Lower jaw changes. Treated patients showed a statistically significant increase in the mandibular length (Co-Gn distance) compared to untreated subjects, at both primary time points. The increase in the mandibular growth was 3.20 mm in patients aged 18 years and above (95% CI 1.32 to 5.08, P = 0.0009, I 2 = 75%, 4 studies [Fig 3]) and 2.87 mm at the end of growth according to the CVM method (95% CI 0.47 to 5.26, P = 0.02, I 2 = 74%, 4 studies [Fig 4]).
Removable functional appliances produced greater treatment effects than fixed devices. The greatest significant increase in the mandibular growth (Co-Gn distance) above 18 years of age was observed in a single study [43], in which a mixed subgroup of patients was treated either with the Bionator or Activator during puberty (MD 5.10 mm, 95% CI 3.29 to 6.91, P < 0.00001, I 2 = Not applicable, 1 study [ Table 6]). This group also showed a statistically significant improvement of the sagittal projection of the mandible (Pg to N perpendicular distance, MD 2.90 mm, 95% CI 1.11 to 4.69, P = 0.001, I 2 = Not applicable, 1 study [ Table 6]), although the test for subgroup differences was not significant (P = 0.13, I 2 = 51.5%).

Additional analysis
Few statistically significant differences were found among the subgroups analysed (Tables 8-9, S3 Appendix). Early treatment with functional appliances (commencing in children aged between 7 and 11 years) produced a greater improvement of the angular antero-posterior position of the maxilla (SNA angle) and the relationship between the two jaws (ANB angle) than late treatment (beginning in adolescents aged between 12 and 16 years). Sensitivity analyses revealed that, if only studies with low and moderate risk of bias were considered, differences in the most clinically important outcomes (SNA angle, Co-Gn distance, ANB angle) were not statistically significant ( Table 9).

Risk of bias across studies
The protocol of the included studies was not retrieved in the Clinical Trial Register, thus outcome reporting bias could not be assessed. Due to the limited number of included studies, an   evaluation for the existence of reporting bias (including publication bias) was not possible [40].
The GRADE assessment for all the outcomes at primary time points were rated as being 'very low' (Table 10), except for the Co-A distance when patients were 18 or older ('low'), and Co-Gn/Co-A difference above the age of 18 ('low') and at the end of growth ('moderate'). Since the included studies were observational, evidence supporting estimates of the intervention effects started to be rated as low-quality. The evidence was down rated for most of the outcomes, as a direct result of the risk of bias and inconsistency of included trials [41].

Summary of evidence
The results demonstrated that functional appliances, worn alone or in combination with multi-bracket therapy, produced an improvement of the maxillo-mandibular relationship at almost all time points. The improvement was around -1 degree for the angular measurement (ANB angle) and between -3.5 and 2.5 mm for the linear outcomes (Wits appraisal, Co-Gn/ Co-A difference). The decrease in the ANB angle and Wits appraisal was consistent with that reported in previous systematic reviews on the effects of functional appliances in the short- [6,21,22,24,26,28] and long-term [28].
In agreement with previous reviews [7,21,24], a restraint of maxillary growth (SNA angle, -1 degree) was observed in included studies. Above 18 years of age or at the end of growth  according to the cervical vertebral maturation method [20], the increase in the mandibular length (Co-Gn distance) was approximately 3 mm greater in the treated patients compared to that in untreated subjects. Similar results were found in the subgroups of adolescents studied by Perinetti et al. [6,22]. However, the improvement of the position of the mandible was negligible or not significant, as inferred from results of its measurements (SNB angle, Pg to N perpendicular). During growth, the mandible is translated downward and forward, while at the same time it increases in size by growing upward and backward [12,14]. Vertical growth can reduce the effects of the increase in mandibular length on its projection.
According to the GRADE Working Group, the quality of evidence was 'very low' for most of the outcomes at both primary time points. Most of the studies received a very low rating, because of their risk of bias and inconsistency [41].
Overall, the clinical significance of these findings was limited. Several approaches were described to establish if the 'statistically significant' differences were also 'clinically important'. The small or minimal clinical important, moderate and large effects were conventionally defined as half, one, and two standard deviations of the normal values, respectively [54]. According to these thresholds, functional appliances produced only small clinically significant changes in the linear maxillo-mandibular measurements (Wits appraisal, Co-Gn/Co-A difference) and in the mandibular length (Co-Gn distance).

Strengths and limitations
Strengths of the present systematic review were in the efforts made to respect rigorous standards for quality and reduce risk of bias: original research question; unrestricted electronic search of 24 databases and additional manual searches; pre-defined and unambiguous eligibility criteria with rationale; adjustment for magnified linear measurements; 3 time points evaluated with rationale; pre-defined and broad additional analyses. However, limitations occurred at some levels. Although both randomised and non-randomised controlled studies were sought, only retrospective controlled clinical trials were retrieved with negative consequences on the quality of evidence of the effect estimates. It needs to be noted that only long-term studies were considered eligible. The whole observational periods of included trials ranged from 4.7 to 10.2 years.
Participants were eligible regardless of their baseline disease severity. The antero-posterior relationship between the two arches or jaws affects the amount of advancement produced by functional appliances, therefore this could influence the treatment effects. The greater the space created between the upper and lower front teeth is, the more protruded position of the  mandible can be achieved. Different classifications of malocclusion also bring into question the applicability of results. Any type of functional appliance, worn alone or in combination with multi-bracket therapy, was included. As anticipated, multi-bracket therapy, as well as retention appliances, could enhance the treatment effects of functional jaw orthopaedics or control their relapse. Moreover, trials with historical untreated controls from growth studies showed larger treatment effects compared to trials with untreated controls from clinical archives [55].
Other limitations concerned the evaluated outcomes. The present systematic review mainly assessed cephalometric skeletal measurements which can be considered as 'clinically important outcomes'. The effects of functional appliances on the soft-tissue facial structures were searched, but few results were found. Multiple related outcomes were also analysed. In fact, the ANB angle is defined as the difference between the SNA and SNB angles, whilst the Co-Gn/ Co-A difference is defined as the total mandibular length (Co-Gn) minus Co-A distance. The greater the number of outcomes, the higher the chance of finding a false positive result [56]. Cephalometric magnification was not reported or retrieved in 2 studies [42,44]. Linear measurements of these studies were excluded from meta-analyses. The impact of dental movements on the skeletal measurements cannot be examined further, as the objective of this systematic review was to assess the skeletal effects produced by functional appliances in the long-term.
With regards to time points, two alternative methods were used to define the completion of growth. Each of these methods is affected by some limitations. The age threshold of 18 years, as reported in one included trial [48], was chosen to maximise the data available. In studies of long duration with several periods of follow-up, the Cochrane Collaboration recommends to select a single time point and analyse only data at this time [30]. Some investigations reported that growth continues up to 21 years of age [15] or more [16][17][18]. However, above 18 years of age, most changes in the mandibular growth (Co-Gn distance) appear to be as non-clinically significant (mean change = 0.1 mm per year [17,18]). None of the included trials evaluated the treatment effects of functional appliances in patients aged at least 21 years old. The cervical vertebral maturation method was also employed. The accuracy of this method is questionable. No skeletal maturity indicator may be considered to have a full diagnostic reliability in the identification of the phases of mandibular growth [57]. All the studies had a post-retention period of at least 3 years, so that a sufficient post-retention period after the functional appliance therapy could be guaranteed [42][43][44][45][46][47][48][49].

Implications for practice
Based on results of this review, weak recommendations can be provided on the long-term effects of functional appliances in treated versus untreated Class II subjects. There is a very low quality evidence that functional appliance therapy produced an improvement of skeletal Class II malocclusion at the end of growth and at least 3 years after retention. Treated patients exhibited an increase in the mandibular length compared to untreated subjects, although with marginal clinical significance.

Implications for research
Further high quality primary studies are needed to confirm or reject the findings of this review. Randomised controlled trials comparing treated patients to untreated subjects (no historical controls) should be carried out. A consensus should be formed on the clinically important measurements to be used for the inclusion in the study and assessment of the effects. Few linear measurements for the position of the maxilla and mandible, the relationship between these jaws, seem to be more appropriate because of their influence on the soft tissue measurements. Patient important outcomes, such as perceived attractiveness, self-esteem and oral healthrelated quality of life, should be assessed as well.

Conclusions
Functional appliances, worn alone or in combination with multi-bracket therapy, may be effective in correcting skeletal Class II malocclusion in the long-term. The increase in the mandibular length may contribute to the improvement of the maxillo-mandibular relationship, although it brought about a negligible or non-significant improvement of the mandibular projection. The quality of evidence was 'very low' for most of the outcomes at both primary time points; the clinical significance of these findings was limited. Further randomised controlled trials evaluating clinically and patient important outcomes are needed to confirm or reject the findings of this review.

Differences between protocol and review
The data extracted were not preliminarily annualised to minimize heterogeneity related to the observation period variability. Annualised changes (mean differences divided by the duration of the whole observational period) seemed to be inappropriate to evaluate the treatment effects in the long-term. If an appliance produced a certain amount of improvement in a given period (reported as degrees/year or mm/year), it does not mean that the device could cause the established improvement for each year of treatment.
An adjustment for magnified linear measurements was introduced to avoid distorted analyses.
Supporting information S1 Study selection 9 State the process for selecting studies (i.e., screening, eligibility, included in systematic review, and, if applicable, included in the meta-analysis).
8 Data collection process 10 Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and any processes for obtaining and confirming data from investigators.

8
Data items 11 List and define all variables for which data were sought (e.g., PICOS, funding sources) and any assumptions and simplifications made. 8

Risk of bias in individual studies
12 Describe methods used for assessing risk of bias of individual studies (including specification of whether this was done at the study or outcome level), and how this information is to be used in any data synthesis.

9
Summary measures 13 State the principal summary measures (e.g., risk ratio, difference in means). 9

Synthesis of results
14 Describe the methods of handling data and combining results of studies, if done, including measures of consistency (e.g., I2) for each meta-analysis.

9
Risk of bias across studies 15 Specify any assessment of risk of bias that may affect the cumulative evidence (e.g., publication bias, selective reporting within studies).

11
Additional analyses 16 Describe methods of additional analyses (e.g., sensitivity or subgroup analyses, meta-regression), if done, indicating which were pre-specified. 10

Results
Study selection 17 Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions at each stage, ideally with a flow diagram.

11
Study characteristics 18 For each study, present characteristics for which data were extracted (e.g., study size, PICOS, follow-up period) and provide the citations. ("Class II malocclusion" or "Class II" or Class II div*).af. 65204 #2 ("prominent upper front teeth" or "prominent upper teeth" or "prominent teeth" (Functional or orthopedic or orthopaedic or interceptive or preventive or bite jump* or (mandib* and (advanc* or enhanc* or postur* or protract* or reposition*)) or Activator or Andresen or Bass or Bionator or Bimler or Frankel or Fraenkel or "Functional magnetic system" or Harvold or Monoblock or "Twin block" or Herbst or "Mandibular anterior repositioning appliance" or MARA or "Eureka spring" or Forsus or "Jasper jumper" or "Sabbagh spring" or "Twin force"

Search Query Hits Google Scholar
"Class II malocclusion" AND "functional appliances" AND "long term" 1510

Search Query Hits Turning Research into Practice (TRIP) database
"Class II malocclusion" AND "functional appliances" AND "long term" 14

Search Query Hits ProQuest Dissertation & Theses Global
"Class II malocclusion" AND "functional appliances" AND "long term" 112

Search Query Hits ProQuest Dissertations and Theses -UK & Ireland / Index to Theses
"Class II malocclusion" AND "functional appliances" AND "long term" 0

OpenGrey -formerly System for Information on Grey Literature (SIGLE)
"Class II malocclusion" AND "functional appliances" AND "long term" 0

Search Query Hits Current controlled trials metaRegister of Controlled Trials (mRCT) -active and archived registers
"Class II malocclusion" AND "functional appliances" 2

International prospective register of systematic reviews (PROSPERO)
Class II malocclusion AND functional appliances S4 Diff, difference; 95% CI, 95% confidence intervals; P, P value reported by the original study.

Treated Control
Results S1 Appendix. Eligibility criteria with rationale.

Study designs
The following study designs were included: randomised controlled trials (RCTs), controlled (nonrandomised) clinical trials (CCTs), controlled before-after (CBA) studies, and case-control or nested case-control studies. Prospective and retrospective cohort studies, cross-sectional studies, case series, and case reports were excluded.
Since the aim of this review was to compare Class II malocclusion patients treated with functional appliances to untreated subjects, only experimental and observational studies with a comparison group were included [1]. The decision to evaluate both randomised and non-randomised controlled trials was made, in order to collect a wide range of studies. A limited number of trials assessing any type of outcome in the long-term was found in previous systematic reviews [2][3][4][5].

Participants
Children and adolescents (aged 16 years or under) receiving orthodontic treatment to correct Class II malocclusion were included. Active treatment with functional appliances had to be completed by the age of 16 years, to allow for a sufficient post-retention period at growth completion. Studies were considered eligible regardless of how the baseline disease was measured (e.g. dental casts, lateral cephalograms) and its severity (e.g. full or half Class II molar relationship, depending on whether the lower molars were placed in a completely or partially posterior position relative to the upper molars, respectively).
Given the potential of functional appliances in modifying the patient growth, they are commonly used in childhood and adolescence [6]. Thus, the analysis of the treatment effects of these appliances on adults was considered to be of minor relevance.
Although there is no agreement on the definitions of childhood and adolescence, in a recent systematic review of the Cochrane Collaboration 'children' were defined as subjects aged from 7 to 11 years, whereas 'adolescents' were defined as subjects aged from 12 to 16 years [6]. This practical categorisation is also used in other systematic reviews [7], trials [8], and some national health services (e.g. United Kingdom and Italy). Alternative methods to establish the growth phase, such as the hand-and-wrist maturation method [9] or the cervical vertebral maturation method [10] were not chosen as inclusion criteria, due to them not being globally accepted [11].

Interventions
Any type of functional appliance, defined as a removable or fixed orthodontic appliance that postures the mandible forward [12]. Functional appliances had to be worn alone or in combination with multi-bracket therapy so as to be included. When functional appliances were worn alone, this therapy could also take place after the functional appliance treatment. A concurrent or subsequent phase with multi-bracket appliances to align teeth is the most common clinical pathway in Orthodontics [2,13].
Conversely, association with other Class II devices designed primarily to restrain the maxilla (e.g. headgear) was set as an exclusion criterion. Mechanics opposite to those employed during the functional appliance therapy were kept out, so as to reduce co-intervention bias [14].
Only functional appliances worn for 6 months or longer were considered eligible. The duration of treatment with functional appliances is usually from 6 to 18 months, followed by night-time insertion of the appliance, or though the use of a stabilization plate [5,6,12,13,15,16]. A wider spectrum of treatment period was considered to be valid, in order to include as many eligible studies as possible.

Comparators
Class II malocclusion patients treated with functional appliances were compared only to untreated Class II subjects. No other type of orthodontic appliance or brace was considered as a comparator.
Patterns of mandibular growth in subjects with untreated Class II malocclusion differ from those of untreated subjects with normal occlusion (Class I). The deficiency in mandibular growth in Class II subjects is significant at the growth spurt, and it is maintained at the post-pubertal stage. Thus, the use of untreated Class II comparators in studies or reviews on the effectiveness of dentofacial orthopaedics on mandibular growth is recommended [17].
For this comparison, groups had to be of similar ages at the commencement of the observational period (age differences between the treated and untreated groups less than 18 months).

Outcomes
The following clinically important outcomes were recorded: § Cephalometric skeletal measurements evaluating the antero-posterior position of the maxilla and mandible, the total mandibular length or length of its parts (ramus and corpus), the mutual relationship between the two jaws. § Soft tissue changes of both lips and chin, measured on lateral cephalograms.
Measurements derived from any cephalometric analysis were included. Due to possible variation in outcome definitions over time, outcomes were collected as reported. Definitions of outcomes as reported in individual studies were extracted as well.
It is not possible to establish the true nature of a malocclusion without information on the underlying skeletal relationships. Cephalometric analysis still remains the most widespread, safest and most precise method of measuring changes to skeletal structures [18]. The use of alternative methods, such as the cone-beam computer tomography (CBCT), should not be implemented for this purpose [19]. According to the 'Guidelines on CBCT for dental and maxillofacial radiology', large volume CBCT should not be used as a standard diagnosis method in Orthodontics. In comparison to conventional radiograph, CBCT has higher radiation doses and, having so stated, its use may be justified in treatment planning, solely for complex cases of skeletal abnormality, particularly those requiring combined orthodontic/surgical management [19].
Lateral cephalograms can also be useful for analysing soft-tissue changes. At this time, alternative methods, such as two-dimensional or three-dimensional photographs, are not widespread as much as lateral cephalograms in orthodontic practice and research.

Timing
Studies were selected for inclusion based on the duration of follow-up of outcomes. Studies should have measured outcomes at the end of growth, defined by age or using indicators of the growth phase. Otherwise, studies should have a post-retention period of at least 3 years.
Contrary to the age threshold established when selecting the inclusion of participants, no age criteria was used to define the end of growth. Literature disagrees on the completion of the maxillofacial unit growth [20][21][22][23][24].
Since the real and stable results produced by functional appliances are the areas of interest, a minimum post-retention period after functional jaw orthopaedics was imposed. There is no recognised duration for retainers to be worn after multi-bracket appliances. It has been shown that if patients stop wearing retainers for between 1 and 2 years after correction of teeth positions there is a risk of long-term relapse [25]. There is no definitive agreement on the retention protocol after functional appliance therapy either [15,26,27]. Nevertheless, it is clinically unlikely that a treatment initiated in adolescence and skeletally stable after a 3 year follow up could relapse. For these reasons, a post-retention period of at least 3 years as eligibility criteria was set.