Functional appliance treatment for mandibular fractures: A systematic review with meta‐analyses

Abstract Objectives Mandibular collum fractures among growing patients can lead to abnormal growth, function, esthetics and ultimately quality of life. Among the proposed treatment alternatives, orthopaedic treatment with functional appliances has been suggested, with encouraging results. Aim of the present systematic review was to critically appraise existing evidence on the outcome of functional appliance treatment among growing patients with mandibular collum fractures. Materials and methods Eight databases were searched up to October 2020 for randomised and non‐randomised clinical studies assessing functional appliance treatment outcome for children with mandibular fractures. After duplicate study selection, data extraction and risk of bias assessment, random effects meta‐analyses of mean differences (MD) and their 95% confidence intervals (CIs) were performed, followed by assessment of the quality of evidence with GRADE. Results A total of 8 unique studies (one prospective and nine retrospective non‐randomised) with 223 children could be identified. Functional appliance treatment was associated with greater anteroposterior condyle dimensions of the injured condyle compared with the contralateral healthy condyle (3 studies; MD = 0.87 mm; 95% CI = 0.30 to 1.45 mm; p = .003). No difference was found in the mesiodistal condyle size between the injured and the contralateral healthy joint (3 studies; MD = −0.05 mm; 95% CI = −1.05 to 0.95 mm; p = .92), but collum length was smaller at the injured side compared with the contralateral one (1 study; MD = −2.89 mm; 95% CI = −5.29 to −0.49 mm; p = .02). Treatment outcome might be influenced by patient age, patient sex and severity/localisation of the fracture, but the quality of evidence for all analyses was very low due to methodological limitations leading to bias. Conclusions While some evidence exists that functional appliances might lead to good clinical rehabilitation of fractured mandibular condyles, including considerable bone remodelling, available studies are small and have methodological weaknesses.


| Rationale
The mandibular condyle is one of the most common sites of facial skeleton that is subject to injury 1 and, due to its role in mandibular form and function, may adversely affect growth and development of the stomatognathic system. Among growing children, acute injuries to the mandibular condyle might lead to serious adverse effects such as temporomandibular dysfunction, disturbed mandibular growth and temporomandibular joint ankylosis. 2 Several therapeutic approaches for condylar fractures exist, including conservative treatment (observation, soft diet, analgetic use), intermaxillary fixation, functional appliance therapy, surgical treatment or a combination thereof. 2 Particularly among children, surgical treatment might not be the first treatment of choice due to the possibility of external scars, nerve damage and abnormal postsurgical growth. 3,4 Conservative treatment with or without intermaxillary fixation seems to often result in good mandibular function, but with the condyles not being completely remodelled in large portion of the patients and with possible late complications such as ankylosis, disturbances of facial growth or functional disorders of the temporomandibular joint. [3][4][5][6][7] On the other hand, orthopaedic treatment with functional appliances that reposition the mandible has been reported to aid stabilisation and rehabilitation, while minimising morbidity. [8][9][10] Functional appliances used for Class II malocclusion due to retrognathic mandibles have been associated with adaptations of the condyle and the glenoid fossa, 11 and early studies in the 90s have shown encouraging results for condylar fractures-including very good remodelling rates. 8,12 However, evidence on the efficacy and safety of orthopaedic rehabilitation for collum fractures remains inconclusive.

| Objective
Therefore, aim of this systematic review was to assess the evidence from clinical studies on humans undergoing orthopaedic treatment with functional appliances after any kind of mandibular condyle fracture.

| Protocol and registration
This review's protocol was made a priori and registered in Open Science Framework (https://osf.io/8ry6p/), and any deviations were noted (Supplement 1). This review is conducted and reported according to Cochrane Handbook 13 and PRISMA statement, 14 respectively.

| Eligibility criteria
According to the Participants, Intervention, Comparison, Outcome, Study design (PICOS) schema and due to the scarcity of randomised clinical trials on this subject, included were randomised and nonrandomised clinical studies on human patients with growing potential (<18 years of age), of any sex, ethnicity or malocclusion with any kind of condylar fractures treated with any kind of functional appliances. No limitations concerning language, publication year or status were applied. Excluded were animal studies, case reports and non-clinical studies. The primary outcome for this review was the restoration of function (treatment success) as reported by the patient. Secondary outcomes included dimensions of the condyle, morphology of the condyle, joint pain, joint sounds during mouth opening and midline deviations in occlusion.

| Information sources and search
Eight electronic databases were searched systematically without any restrictions for publication date, language or type from inception up to 1 October 2020 (Supplement 2), while Directory of Open Access Journals (DOAJ), Digital Dissertations, metaRegister of Controlled Trials, WHO and Google Scholar, as well as the reference/citation lists of eligible articles or existing systematic reviews were manually searched for any additions.

| Study selection
Two authors (CS and SNP) screened the titles and/or abstracts of studies retrieved from the searches to identify articles that potentially meet the inclusion criteria, before moving to their full texts.
Any differences between the two authors were resolved by discussion with a third author (TE).

| Data collection process and items
Data collection from the identified reports was conducted using pre-defined and piloted forms covering: (a) study characteristics (design, clinical setting, country), (b) patient characteristics (age, sex), (c) characteristics of the mandibular fracture (affected sides, Spiessl & Schroll categorisation 15 ), (d) functional appliance used, and (e) follow-up period. Data were extracted by two authors (CS and SNP) with the same way to resolve discrepancies as above.

| Risk of bias of individual studies
The risk of bias of included randomised studies was to be assessed according to Cochrane guidelines with the RoB 2.0 tool for randomised trials. 16 The risk of bias of non-randomised studies was assessed with a customised checklist based on the ROBINS-I ('Risk Of Bias In Non-randomised Studies-of Interventions') tool for nonrandomised studies. 17 Assessment of the risk of bias of included studies was likewise performed independently by two authors (CS and SNP), with the same way to resolve discrepancies consulting a third author (TE).

| Data synthesis and summary measures
An effort was made to include all existing trials in the analysis; where data were missing, they were calculated by ourselves or extracted from graphs (Supplement 1). As the outcome of orthopaedic treatment with functional appliances is bound to be affected by patient-and treatment-related characteristics, a random-effects model was deemed appropriate to calculate the average distribution of true effects, based on clinical and statistical reasoning, 18 and a restricted maximum-likelihood random-effects model was used according to recent guidance. 19 Mean differences (MDs) for continuous outcomes and odds ratios (ORs) for binary outcomes and their corresponding 95% confidence intervals (CIs) were calculated as effect sizes.
The extent and impact of between-study heterogeneity was assessed by inspecting the forest plots and by calculating the τ 2 (absolute heterogeneity) and the I 2 statistics (relative heterogeneity; inconsistency), respectively. I 2 defines the proportion of total variability in the result explained by heterogeneity, and not chance, and we considered arbitrarily I 2 over 75% to represent considerable heterogeneity, while also considering the heterogeneity's direction (localisation on the forest plot) and uncertainty intervals around heterogeneity estimates. 20 Ninety-five per cent predictive intervals were calculated for meta-analyses of ≥3 trials to incorporate existing heterogeneity and provide a range of possible effects for a future clinical setting, which are crucial for the correct interpretation of random-effects meta-analyses. 21

| Additional analyses and risk of bias across studies
Possible sources of heterogeneity were a priori planned to be sought through subgroup analyses and random-effects meta-regression in meta-analyses of at least 5 trials but could ultimately not be performed (Supplement 1). Likewise, reporting biases were planned but ultimately not assessed, due to the limited number of meta-analysed trials.
The overall quality of meta-evidence (ie, the strength of clinical recommendations) was rated using the Grades of Recommendations, Assessment, Development and Evaluation (GRADE) approach 22 following recent guidance on combining randomised with nonrandomised studies. 23 Robustness of the results was planned a priori to be checked with sensitivity analyses based on (a) inclusion/exclusion of non-randomised studies, (b) inclusion/exclusion of trials with methodological shortcomings and (c) improvement of the GRADE classification. In the end, no sensitivity analysis could be conducted due to the limited number of studies and all of them having methodological insufficiencies.
All analyses were run in Stata version 14.0 (StataCorp LP) by one author (SNP), and the data set was openly provided 24 with a 5% significance level.

| Study selection
The electronic literature search yielded 834 results, while another two were manually identified from the reference/citation lists of identified papers ( Figure 1). After duplicate removal and screening the titles/abstracts of identified reports, the full texts of 144 papers were checked against the eligibility criteria (Supplement 3).
Ultimately, 9 papers pertaining to 8 unique studies (1 prospective and 7 retrospective non-randomised) were included, which were published as journal papers.

| Study characteristics
The included studies were conducted in university clinics in Austria, China and Germany (Table 1) Orthopaedic treatment was done with activator appliances in six of the studies and with occlusal splints in the remaining two studies.
One study (Kahl and Gerlach, 1990) 8 also compared three different groups: one with only functional appliance treatment, one with intermaxillary fixation and then functional appliance treatment, and one with intermaxillary fixation and functional exercises. Follow-up after condylar fracture ranged from 1 to 10 years, and patients were evaluated clinically or radiographically.

| Risk of bias within studies
The included non-randomised studies presented several issues that increased their risk for bias ( Table 2). Seven of the 8 studies were retrospective, while the patient's sex and skeletal maturation age were often inadequately described. In half of the included studies (50%), selection bias could exist, as patient selection was based on factors that could be associated with the outcome of treatment. No study blinded the outcome assessor, and only 3 studies (38%) included a relatively adequate sample with at least 25 patients. Follow-up periods were relatively adequate, and only one study (13%) did not have a minimum follow-up of a year. From the two studies that made comparisons between groups (Kahl and Gerlach, 1990; Kahl-Nieke and Fischbach, 1998), 8,25 matching according to patient characteristics was judged to be adequate only for one of them, while the observation period was not common across groups for both of the studies.

| Results of individual studies and data synthesis
The included studies reported on a wide variety of outcomes after functional appliance treatment. Kahl   All included studies reported aggregate data in the paper, while for one study 29 data were extracted from boxplots. Two studies 8,25 also provided individual patient data in tables, which were extracted and re-analysed.
The study of Kahl  Direct comparisons within and across studies could be made in only a handful of instances (Table 3). Meta-analysis of three studies 10,25,27 indicated that functional appliance treatment was associated with greater anteroposterior condyle dimensions of the injured condyle compared with the contralateral healthy condyle (MD = 0.87 mm; 95% CI = 0.30 to 1.45 mm; p = .003; Figure 2).

| Additional analyses, risk of bias across studies, sensitivity analyses and quality of evidence
Several subgroup analyses, meta-regressions, assessments for reporting biases and sensitivity analyses were originally planned in the review's protocol but could ultimately not be performed due to limited data and inadequate reporting (Supplement 1).
The quality of evidence according to GRADE was very low for all meta-analyses, since only non-randomised and especially retrospective clinical studies with many methodological issues were available.
This means that further research in terms of well-designed studies is very likely to have an important impact, which is likely to change our current estimates of effect. The identified studies indicated that the functional appliances might lead to good clinical rehabilitation of fractured mandibular condyles in growing patients. The rationale behind this approach is that early controlled mobilisation of the mandible using a functional appliance results in a re-instatement of an organised functional condyle, 6 while late complications such as ankylosis, facial growth disturbances or functional joint disorders are prevented. 31 Meta-analysis of three studies indicated that after functional appliance treatment the condyle of the affected size was significantly larger on the anteroposterior dimension than the condyle of the non-affected side (MD = 0.87 mm; p = .003). This might indicate that the anterior repositioning of the mandible acts as a stimulant that ultimately lead to bone remodelling in the condylar area. This was shown decades ago to be feasible in animal studies 32,33 and was confirmed by subsequent human studies on Class II malocclusion. 11

| Summary of evidence
However, no correlation has been found between an increased size of the condylar head and the functional status of the stomatognathic system. 34 This new bone growth is associated at the same time with an 10%-70% decrease in muscle volume 30 and an overcompensation of function from the contralateral healthy side, where a 20%-40% increase in the volume of the lateral pterygoid muscle can be seen.
The prognosis of the condylar rehabilitation seems to be mainly affected by the type of the mandibular fracture-namely its localisation and the existence of dislocation. Low fracture types and fractures with dislocation might be associated more with shortening of the condylar process and excessive bone overgrowth that high fractures and fractures without dislocation. 25 Re-analysis of that study's data (Supplement 5) indicated that patients with luxated fractures had significantly smaller odds of having a condylar head of size (±10%) to the healthy unaffected condyle compared with patients without luxation. Similar results have been given by reports of cases, 34 where high fractures showed a high degree of remodelling, while some presented a notching of the lateral surface of the condylar head and a slight medial deviation of the condylar head.
Low fractures with luxation on the other side presented often unfavourable remodelling, with irregular condylar morphology and altered topography related to the glenoid fossa, but still with muscular adaptation and no functional disturbances. However, especially in patients with low condylar fractures, residual facial asymmetries and malocclusions can be seen, 35 which might even need to be corrected with orthognathic surgery.
Remodelling prognosis of the fractured condyle seems also to be closely associated with the age of the patient. 25

| Strengths and limitations
This systematic review has several strengths, comprising an a priori registered protocol, 39  Some limitations also do exist in the present review. For one, methodological issues existed for all included studies that might influence conclusions, and this is especially the case for included retrospective non-randomised studies. 41,42 Furthermore, all metaanalyses were based predominantly on small trials, which might affect the precision of the estimates. 43 Additionally, the small number of studies with limited samples that were ultimately included in the meta-analyses and their incomplete reporting of results and potential confounders precluded the conduct of many analyses for subgroups and meta-regressions that might enable identification of patient subgroups for which functional appliances might be most effective.

| CON CLUS IONS
There is currently very limited evidence on the treatment of growing patients with mandibular fractures using orthopaedic functional appliances. Some data indicate that functional appliance treatment is associated with partial or complete remodelling of the fractured condyle and the temporomandibular joint with acceptable clinical results. However, existing studies are mostly old, single-group cohort studies without control groups and with many methodological issues and uncertainty still exist around the long-term outcomes of functional appliances for condylar fractures.

CO N FLI C T O F I NTE R E S T
None to declare.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/joor.13178.

DATA AVA I L A B I L I T Y S TAT E M E N T
The full data set is openly provided through Zenodo (http://doi.