Abstract
A systematic review and meta-analysis was performed to assess the clinical outcomes of the following four methods for the management of temporomandibular joint (TMJ) ankylosis: gap arthroplasty (GA), interpositional gap arthroplasty (IPG) using the temporalis muscle, reconstruction of the TMJ using a costochondral graft (CCG), and alloplastic joint reconstruction (AJR) of the TMJ. A comprehensive electronic and manual search of the literature without date or language restriction was performed in December 2013 to identify randomized controlled trials, controlled clinical trials (CCTs), and retrospective studies with the aim of comparing the four surgical modalities for TMJ ankylosis. Sixteen publications were included: seven were CCTs and nine were retrospective. A significant difference was found between GA and IPG in maximal inter-incisal opening (MIO) and recurrence rate ( P = 0.04 and P = 0.02, respectively). A significant difference was found between IPG and CCG reconstruction in MIO ( P = 0.01), but no significant difference with regard to the recurrence rate ( P = 0.71). There was a significant difference between costochondral joint (CCJ) and AJR for MIO and pain ( P = 0.04 and P = 0.03, respectively). The results of the meta-analysis showed that IPG results in a significant improvement in MIO and lower recurrence rate when compared to GA. Also, IPG shows a greater improvement in MIO and comparable recurrence rate when compared to CCG reconstruction. GA and CCG reconstruction have a comparable recurrence rate. Lastly, CCJ provides greater MIO when compared to AJR, whereas AJR is superior to CCJ in reducing pain.
Ankylosis of the temporomandibular joint (TMJ) is a disabling condition that causes problems in mastication, digestion, speech, function, cosmesis, and maintenance of oral hygiene. It can also cause disturbances of facial growth and acute compromise of the airway, which invariably results in physical and psychological disability. TMJ ankylosis usually results from injury (13–100%), local or systemic infection (10–40%), or systemic disease (10%), such as ankylosing spondylitis, rheumatoid arthritis, and psoriasis. It may also result from surgery to the TMJ. The hypothesis proposed for trauma cases is that intra-articular haematoma, scarring, and the formation of excessive bone gives rise to hypomobility.
Ankylosis of the TMJ may be classified using a combination of the site (intra-articular or extra-articular), type of tissue involved (bony, fibrous, or fibro-osseous), and extent of fusion (complete or incomplete). Tripathy et al. have classified ankylosis as true or false. In true ankylosis there is bony or fibrous adhesion between the surfaces of the joint within the capsule, whereas in false ankylosis the problems lie in the surrounding structures.
Ankylosis of the TMJ usually develops before the age of 10 years, but can be found at any age; the incidence peaks in patients aged 20–30 years. It is slightly more common in boys than girls, with a ratio of 1.4:1.2. Unilateral ankylosis has been reported to be more common than bilateral, with a ratio of 1.5:1. Patients present with a limitation of mouth opening and a maximal inter-incisal distance of between 0 and 20 mm. The condition causes aesthetic defects to the face, malocclusion, inability to enjoy eating, and malnutrition, particularly when it develops in childhood.
Various procedures for the treatment of TMJ ankylosis have been described in the literature. These include gap arthroplasty (GA), interpositional gap arthroplasty (IPG), and/or joint reconstruction using autogenous grafts or alloplastic materials. Patients whose joints are reconstructed with alloplastic materials may experience infection and inflammatory problems as the immune system reacts to a foreign body. Autogenous interpositional materials, such as temporalis muscle, dermis–fat graft, and auricular cartilage, have been used in several studies to prevent re-ankylosis. Some authors have suggested using the remaining TMJ disc, after its repositioning. Several autologous bone grafts are used to reconstruct the ramus–joint complex, including costochondral grafts (CCG) and clavicular bone grafts. The resected elongated coronoid process and excised ankylotic mass have also been tried. Distraction osteogenesis has been used successfully before and after the release of the joint ankylosis, to correct secondary facial asymmetry. To avoid relapses, wide bone resection of the ankylosis remains the most important aspect of the surgical protocol, regardless of the technique used.
The treatment of TMJ ankylosis to restore full mouth opening and normal oral function remains a significant challenge to clinicians and patients (children and adults) due to technical difficulties and a high incidence of recurrence.
Although a variety of techniques for the treatment of TMJ ankylosis have been described in the literature, there is no published consensus regarding the best treatment and no single method has produced uniformly successful results.
The aim of the present study was to identify the clinical outcomes of the following four methods for the management of TMJ ankylosis: (1) GA, (2) IPG using the temporalis muscle, (3) reconstruction of the TMJ using a CCG, and (4) alloplastic joint reconstruction (AJR) of the TMJ.
Methods
Inclusion and exclusion criteria
Randomized or quasi-randomized controlled trials (RCTs), controlled clinical trial (CCTs), and clinical series comparing the various surgical techniques for the treatment of TMJ ankylosis (GA, IPG using temporalis muscle, reconstruction of the TMJ using a CCG, and AJR of the TMJ) and reporting maximal inter-incisal opening (MIO), pain, and the incidence of recurrence were included.
All other types of study design, such as uncontrolled studies, case reports, technical reports, animal studies, and reviews, were excluded.
Search methods for the identification of studies
This systematic review and meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)-E 2012 checklist. A comprehensive electronic search without date or language restriction was performed in December 2013 using the following electronic databases: PubMed, Cochrane Database of Systematic Reviews, the Cochrane Central Register of Controlled Trials (CENTRAL), EMBASE, MEDLINE, CINAH, SPORTDiscus, and Electronic Journal Center. One or a combination of the following search terms was used: ‘TMJ ankylosis’ AND/OR ‘resection ankylotic mass’ AND/OR ‘gap arthroplasty’ AND/OR ‘interpositional arthroplasty’ AND/OR ‘TMJ arthrotomy’ AND/OR ‘mandibular range of motion’ AND/OR ‘re-ankylosis/relapse’ AND/OR ‘maximal mouth opening’, ‘alloplastic reconstruction’, ‘total joint replacement’, ‘autogenous reconstruction’.
A manual search of oral and maxillofacial surgery journals, including the International Journal of Oral and Maxillofacial Surgery, British Journal of Oral and Maxillofacial Surgery, Journal of Oral and Maxillofacial Surgery, Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontology, Journal of Cranio-Maxillofacial Surgery, Journal of Craniofacial Surgery, Journal of Maxillofacial and Oral Surgery, and the Turkish Journal of Pediatrics, was also performed.
The reference lists of identified studies and relevant reviews on the subject were also scanned for possible additional studies. Moreover, online databases providing information about clinical trials in progress were checked ( http://clinicaltrials.gov ; http://www.centerwatch.com/clinical-trials ; http://www.clinicalconnection.com ).
Study selection and data collection process
The authors assessed the eligibility of all studies retrieved from the databases independently. Disagreements concerning the selected studies were resolved by discussion. From the studies included in the final analysis, the following data were extracted (when available): authors, study design, year of publication, number of patients, gender (male:female ratio), mean age in years, pre- and postoperative MIO, recurrence, and the follow-up period (mean). Authors were contacted for possible missing data.
Synthesis of results
The primary outcome variable was MIO, which was compared for GA vs. IPG, IPG vs. CCG, and AJR vs. costochondral joint (CCJ).
The mean change from preoperative to postoperative MIO/pain for both comparisons was either directly reported in the studies or could be calculated from the means of MIO/pain for all patients in each group using the following formula:
We used a conservative approach to estimate the standard deviation (SD) of the change from baseline for those studies in which the SD of the mean difference between baseline and post-treatment was not directly available, as follows:
SD ( change from baseline ) = ( preoperativ SD ) 2 + ( postoperative SD ) 2 2
The secondary outcomes assessed were the recurrence rate (re-ankylosis) and pain, as follows: GA vs. IPG (recurrence rate); GA vs. CCG (recurrence); IPG vs. CCG (recurrence); AJR vs. CCJ (pain).
For continuous outcomes, the weighted mean difference (WMD) and 95% confidence interval (CI) were used to summarize the data for each group. For binary outcomes, we planned to calculate a standard estimation of the odds ratio (OR) by random effects model if heterogeneity was detected, otherwise with a fixed effects model, with 95% CI.
The data were analysed using Review Manager statistical software (version 5.2.6; The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark).
Assessment of heterogeneity
The significance of any discrepancies in the estimates of the treatment effects of the different trials was assessed by means of Cochran’s test for heterogeneity and the I 2 statistic, which describes the percentage total variation across studies that is due to heterogeneity rather than chance. Heterogeneity was considered statistically significant if P < 0.1. A rough guide to the interpretation of I 2 given in the Cochrane handbook is as follows: (1) from 0 to 40%, the heterogeneity might not be important; (2) from 30% to 60%, it may represent moderate heterogeneity; (3) from 50% to 90%, it may represent substantial heterogeneity; (4) from 75% to 100%, there is considerable heterogeneity.
Investigation of publication bias
A funnel plot (plot of the effect size versus standard error) was drawn. Asymmetry of the funnel plot may indicate publication bias and other biases related to sample size, although asymmetry may also represent a true relationship between trial size and effect size.
Sensitivity analysis
If there were sufficient included studies, we planned to conduct a sensitivity analysis to assess the robustness of the review results by repeating the analysis with the following adjustments: exclusion of studies with short follow-up periods and with a high risk of bias.
Risk of bias in individual studies
The quality of the articles was scored using the checklist proposed by Rangel et al. and Versteegh et al. The checklist comprises three subscales containing 30 items in total. The three subscales are (1) potential clinical relevance, (2) quality of study methodology, and (3) quality of discussion and stated conclusions. The maximum total score is 45 points. Scores ranging from 0 to 15 indicate a study of poor quality, studies scoring from 16 to 30 points are considered to be fair, and scores of 31 points or higher indicate a qualitatively good study. All studies of poor quality (scoring less than 16 points) were excluded.
Results
Study selection
The study selection process is summarized in Fig. 1 . The electronic search resulted in 1534 entries. After the initial screening of titles and abstracts, 1458 articles were excluded for being cited in more than one search of terms. After retrieving full texts of pertinent articles ( n = 76), 60 were excluded for the following reasons: letter to the editor ( n = 1), abstract only ( n = 3), absence of comparator ( n = 24), review article ( n = 4), technical report ( n = 2), did not report pre- and postoperative MIO for each group ( n = 5), discussion to article ( n = 1), did not mention in which group re-ankylosis cases occurred ( n = 3), and did not assess the outcomes of interest ( n = 17). Thus, a total of 16 articles were included in the review. Among the 16 articles, eight compared GA to IPG groups for MIO and recurrence rate ; four compared IPG to CCG groups for MIO and recurrence rate ; two articles compared the GA to CCG groups with regard to recurrence rate ; and three compared CCJ to AJR with regard to MIO and pain. One study (Tanrikulu et al. ) was used in two comparisons (GA vs. IPG and IPG vs. CCG) because it assessed these three groups for MIO and recurrence.
Characteristics of studies included
GA versus IPG
Detailed data of the eight studies included that compared GA and IPG are presented in Table 1 . Five CCTs and three retrospective studies (RS) were included in the meta-analysis in terms of MIO and recurrence. In total, 263 patients were enrolled in eight studies comparing GA ( n = 127) to IPG ( n = 136). The maximum follow-up varied between 6 months and 15 years.
| Authors | Study design | Year | Patients ( n ) | Gender, M:F ratio | Mean age, years | MIO (mm) | Recurrence | Follow-up | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GA group | IPG with TF group | GA group | IPG group | |||||||||||
| Pre-op. | Post-op. | Gain in MIO | Pre-op. | Post-op. | Gain in MIO | |||||||||
| Tanrikulu et al. | RS | 2005 | 8 | 9 | 5:3 GA 4:5 IPG |
7–15 (mean 12.4) | 2.5 | 31 | 28.5 | 3 | 32.2 | 29.2 | 0 GA 1 IPG | 12–180 (12.5) months |
| Ramezanian and Yavary et al. | CCT | 2006 | 22 | 26 | 21 M, 27 F | 19.5 ± 8.9 | 8.7 | 32.1 | 24.33 | 10.3 | 33.9 | 23.6 | 7 GA 0 IPG |
12–24 and 48 months |
| Danda et al. | CCT | 2009 | 8 | 8 | 4:4 GA 5:3 IPG |
GA: 8.7 IPG: 10.3 |
3.75 | 31.12 | 27.37 | 3.5 | 31.43 | 27.93 | 1 GA 1 IPG |
3, 6, 9, and 12 months |
| Zhi et al. | RS | 2009 | 24 | 17 | 20 M, 22 F | 5–55 (mean 22.25) | 7 | 25.58 | 18.58 | 9 | 29.57 | 20.57 | 3 GA 0 IPG |
1–11 years |
| Elgazzar et al. a | RS | 2010 | 11 | 14 | NM | NM | 05.3 | 29.1 | 23.8 | 05.3 | 30.7 | 25.4 | 2 GA 0 IPG (?) |
1 week, 6 months, 1 year, and yearly |
| Holmlund et al. | CCT | 2013 | 14 | 22 | 1:13 GA 3:19 IPG |
22–88 (66) | 19.3 | 30.9 | 11.6 | 21 | 63.7 | 15.7 | 0 GA 0 IPG |
1 week, 1, 3, and 6 months, 1 year, and yearly |
| Mansoor et al. | CCT | 2013 | 30 | 30 | 36 M, 24 F | 13.33 (±4.85) | 6.47 | 30.80 | 24.33 | 8.43 | 32.20 | 23.77 | 0 GA 1 IPG |
Up to 6 months |
| Shaikh et al. | CCT | 2013 | 10 | 10 | 6 M, 24 F | 15.15 | 3.90 | 33.30 | 29.4 | 0.50 | 33.40 | 32.9 | 0 GA 0 IPG |
Up to 1 year |
a Personal communication: the author was contacted to obtain missing data.
Surgical approaches to the TMJ were the same in all of the studies, through the modified pre-auricular incision of Al-Kayat and Bramley with temporal extension shaped like a question mark. Concerning the IPG group, a temporalis myofascial muscle flap (TMF) was the material (tissue) that interposed between the condyle stump and the base of the skull.
IPG versus CCG
Details of the four studies that compared IPG to CCG are presented in Table 2 . Two CCTs and two RS were included in the meta-analysis with regard to MIO and recurrence rate. In total, 83 patients were enrolled in the four studies, with 44 patients in the IPG group and 39 in the CCG group. The maximum follow-up varied between 1 and 15 years.
