Abstract
The objective of this study was to undertake a systematic review to assess the efficacy of botulinum toxin therapy (BTX) for temporomandibular joint disorders (TMDs). A comprehensive search of major databases through PubMed, EMBASE, and Cochrane CENTRAL was conducted to locate all relevant articles published from inception to October 2014. Eligible studies were selected based on inclusion criteria and included English language, peer-reviewed publications of randomized controlled trials comparing BTX versus any alternative intervention or placebo. Quality assessment and data extraction were done according to the Cochrane risk of bias tool and recommendations. The entire systematic search and selection process was done independently by two reviewers. Five relevant study trials were identified, involving 117 participants. Two trials revealed a significant between-group difference in myofascial pain reduction, another trial that compared BTX with fascial manipulation showed equal efficacy of pain relief on TMDs, while the remaining two trials showed no significant difference between the BTX and placebo groups. Because of considerable variations in study methods and evaluation of results, a meta-analysis could not be performed. Based on this review, no consensus could be reached on the therapeutic benefits of BTX on TMDs. A more rigorous design of trials should be carried out in future studies.
Temporomandibular joint disorder (TMD) is a collective term for a group of clinical problems affecting the temporomandibular joint (TMJ), the masticatory muscles, and the associated structures. The prevalence of ‘treatment need for TMD’ in the general adult population was estimated at about 15–16%, based on a systematic review and meta-analysis of 17 population-based, non-patient studies of 9454 adults, and the prevalence among women is higher than among men. The symptoms and signs of TMD include chronic or acute facial pain, tenderness of the masticatory muscles, TMJ pain, TMJ clicking or crepitus during motion, jaw deviation, and functional limitation of jaw opening.
According to the American Academy of Orofacial Pain, TMD is classified into two groups – myogenous TMD, which is more related to masticatory muscle disorders, and arthrogenous TMD, which is more related to the TMJ itself. The Research Diagnostic Criteria (RDC), on the other hand, categorize TMD into three groups according to the common factors among conditions: group I is myofascial TMD, group II is disc displacement, and group III is other TMD including arthralgia, osteoarthritis, and osteoarthrosis. For the majority of TMD patients seeking medical treatment, group I and group II RDC/TMD present most often. The aetiology of TMD is hard to clarify, and many factors may play important roles in causing TMD, for example, trauma, adverse loading of the masticatory system, parafunctional habits, systemic factors (such as hormones), anatomical factors, and also psychosocial factors.
Various modes of treatment have been suggested for TMD, from patient education, pharmacological therapy, and psychological therapy, as well as non-invasive interventions including physiotherapy and the application of a splint, to further surgical interventions. To date, there is lack of consensus on the most efficacious treatment approach for TMD because of its multifactorial etiological nature. Nevertheless, the goal of TMD treatment is not to ‘cure the disease’, but to fulfil the patient’s expectation of symptom relief (especially pain) and to regain jaw function. Since bruxism and teeth clenching are two significant risk factors for TMD symptoms, especially myofascial pain, strategies targeted towards muscle relaxation may be a consideration for the treatment of these patients.
Botulinum toxin (BTX) is a potent neurotoxin synthesized by the Gram-negative, anaerobic, spore-forming bacterium Clostridium botulinum . It blocks the presynaptic release of acetylcholine (Ach) into the end-plate of the neural junction, thus leading to reduced activity of the muscles or glands. Its first medical use was to treat strabismus in the late 1970s, and botulinum toxin serotypes A and B (BTX-A and BTX-B) are now both approved by the US Food and Drug Administration for clinical use.
Although BTX was first used as a therapeutic agent for the treatment of neuromuscular disorders, its pain relieving effects were also observed when treating patients with cervical dystonia, and the clinical uses of BTX were expanded to include the management of pain disorders of musculoskeletal origin, including myofascial pain syndrome. Until recently, increasing studies have suggested an independent action of BXT on peripheral nociceptors by blocking the release of neurotransmitters including pain and inflammatory mediators. Due to its muscle activity reducing and pain relief effects, BTX has gained an emerging role as a potential therapy for TMD, with available clinical reviews supporting its benefit in the treatment of TMD.
The aim of this critical review was to investigate the current evidence in order to assess the efficacy of BTX use in TMD, especially pertaining to myofascial pain and disc derangement disorders.
Materials and methods
This systematic review was performed in accordance with the PRISMA statement (Preferred Reporting Items for Systematic Reviews and Meta-Analyses). Studies aiming to assess the treatment efficacy of BTX for TMD, published from inception to October 2014, were sought using three electronic databases: PubMed, Cochrane CENTRAL, and EMBASE. Other sources were searched manually in the reference lists of the included studies and six journals highly likely to contain studies relevant to the review topic: Journal of Oral and Maxillofacial Surgery ; International Journal of Oral and Maxillofacial Surgery ; British Journal of Oral and Maxillofacial Surgery ; Journal of Orofacial Pain ; Oral Surgery , O ral Medicine, Oral Pathology, Oral Radiology, and Endodontology ; Journal of Cranio-Maxillo-Facial Surgery . The search was conducted by two of the reviewers (YWC and YWC).
Search strategy
The database was searched using the medical subject heading terms craniomandibular disorders, temporomandibular joint, botulinum toxins, and free text word terms temporomandibular$, craniomandibular$, tmj, Botox, BTX, and Botulin, with restrictions to language (English), study participants (humans), and adults (age ≥16 years). For example, the search terms used in PubMed were (‘Craniomandibular Disorders’[Mesh] OR ‘Temporomandibular Joint’[Mesh] OR temporomandibular[tw] OR temporo mandibular[tw] OR tmj[tw] OR craniomandibular[tw] OR cranio mandibular[tw]) AND (‘Botulinum Toxins’[Mesh] OR botulinum[tw] OR Botox[tw] OR BTX[tw] OR Botulin[tw]), in which ‘Mesh’ indicates a medical subject heading term and ‘tw’ represents a free text word. Three review authors (YWC, YWC, and CYC) reviewed the titles and abstracts of the articles identified in all of the searches in the initial screening. Irrelevant studies that did not meet the selection criteria were excluded. Potential articles were required to meet the inclusion/exclusion criteria to be eligible for critical appraisal.
Inclusion criteria for studies to be considered (PICO)
Participants (P) were adults of both genders, regardless of social or economic status, profession, or residential location, with a clinical diagnosis of TMD pertaining to myofascial pain and disc derangement disorders, as diagnosed using the researchers’ own criteria, or according to the American Association of Orofacial Pain (AAOP) guidelines, or the RDC/TMD (research diagnostic criteria for TMD).
The intervention (I) was BTX injection, irrespective of dose and type (e.g., A, B). Injections in different masticatory muscles were all considered.
The comparator/control (C) was any alternative intervention or placebo.
With regard to outcomes (O), the primary outcome was a subjectively assessed, recognized and validated pain scale reported by the patients themselves. The secondary outcome was objective jaw opening capacity (range of maximal mouth opening) measured by the assessors, the number of patients experiencing pain reduction, and adverse effects including any undesirable events associated with BTX injection.
The study types included were randomized and quasi-randomized controlled trials, including parallel studies or cross-over studies having a wash-out period of at least 3 months.
Exclusion criteria
Articles were excluded if the cases were diagnosed with entities other than TMDs, such as TMJ dislocation, congenital or developmental disorders, systemic inflammatory connective tissue diseases (e.g., rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis), fracture, neoplasia, etc.
Eligible articles were then reviewed independently in full text version by two examiners (YWC and YWC). Throughout the review process, disagreements were discussed with another author (CYC) until consensus was reached.
Quality of the studies included
All studies included in the present systematic review were prospective human controlled trials. The methodological quality of the studies included was assessed using the Cochrane risk of bias tool for randomized controlled trials. Each domain in the tool was judged as presenting a low or high risk of bias. However, if the information given in the articles was insufficient for a judgement to be made, the domain in question would be recorded as having an unclear risk of bias. The overall risk of bias of each study included was categorized as ‘low risk’ if all criteria were met, ‘unclear’ if there was plausible bias raising some doubt about the results, or ‘high risk’ if there was plausible bias that seriously weakened confidence in the results.
Data extraction
Two reviewers (YWC and YWC) independently extracted the study details and entered them into data collection forms. The characteristics recorded from the studies were the following: (1) type and setting of the study; (2) age, gender, and number of participants; (3) diagnosis of TMD, duration of symptoms, and previous treatment for TMD before the study; (4) type (brand) and dosage of BTX, route of administration, and accessory treatments; (5) interval and length of follow-up, outcome measures, and side effects.
Data with regard to the outcome measures were also extracted independently. As well as the primary and secondary outcomes, other subjective and objective evaluation outcomes of TMD were reviewed, including questionnaire scores for global or functional improvement, tenderness of the masticatory muscles, the frequency of analgesic use, any adverse events, and emotional evaluations. However, only the number of patients experiencing a reduction in pain, pain scores (primary outcome), and maximal jaw opening range (secondary outcome) were entered into Review Manager Software (RevMan 5.2; The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark) for the analysis of between-group differences for the effects of BTX therapy.
Statistical analysis
It was intended to perform a meta-analysis for this systematic review, however due to the considerable variations in study methods and evaluation of results, with only two papers potentially eligible for meta-analysis, this could not be done.
Results
Search outcome
Figure 1 illustrates the process of evaluating articles for inclusion in the review and meta-analysis. The search strategy yielded a total of 124 non-duplicate reports from all databases. One hundred and eleven articles were excluded after reading the titles and abstracts, and the full text articles of the remaining 13 studies were reviewed by two authors for eligibility. At this stage of the analysis, eight studies were excluded because of a duplicate database (two studies), prospective cohort (three studies), case series (two studies), and ineligible diagnosis (containing TMJ arthralgia) with a previous history of TMJ surgery (one study). Finally a total of five randomized controlled trials met the inclusion criteria and were processed for critical review.
Of the five studies included, three were parallel studies and two were cross-over studies. The cross-over study design involves a within-individual comparison, which usually provides more precise comparisons than studies with between-individual designs. To avoid the carry-over effects of BTX, a wash-out period of at least 3 months was required; both of the cross-over controlled trials included in this review had waited for a sufficient period of time (4 months) to allow the residual effects of BTX to dissipate.
Study description and population epidemiology
The study characteristics were evaluated, catalogued, and tabulated, and a descriptive synthesis is summarized in Table 1 . These selected studies were published between 2002 and 2012. They were performed in different countries, with a total of 123 patients receiving BTX injection. However, a total six participants in two of the studies dropped out before evaluation of the effect. Nearly all of the studies had more female participants than males, and two studies focused only on females.
| Study | Study design | Participants | Intervention/control | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Trial type | Country | Diagnosis (criteria) | Mean age, years | Symptoms | Symptom duration, months | Sample size | Gender | Intervention | BTX brand | |
| Ernberg et al. (2011) | Cross-over | Sweden and Denmark | MFP (RDC/TMD) | 38 | Persistent myofascial TMD pain (average intensity 3/10 VAS) | ≥6 | G1, 21 G2, 21 | 21 F | G1, BTX-A injection G2, NS injection |
Botox |
| Guarda-Nardini et al. (2012) | Parallel | Italy | MFP ± DD (RDC/TMD) | 45.45 | Bilateral masticatory muscle pain with or without limited mouth opening range | ≥6 | G1, 15 G2, 15 |
G1, 4 M, 11 F G2, 4 M, 11 F |
G1, BTX-A injection G2, Fascial manipulation |
Dysport |
| Kurtoglu et al. (2008) | Parallel | Turkey | MFP ± DD (RDC/TMD) | 26.5 | Pain with or without functional disturbance | N/A | G1, 12 G2, 12 |
G1, 2 M, 10 F G2, 2 M, 10 F |
G1, BTX-A injection G2, NS injection |
Botox |
| von Lindern et al. (2003) | Parallel | Germany | MFP with bruxism (own criteria) | N/A | Chronic facial pain, with parafunctional movement | ≥3 | G1, 60 G2, 30 |
G1, N/A G2, N/A |
G1, BTX-A injection G2, NS injection |
Botox |
| Nixdorf et al. (2002) | Cross-over | Canada | MFP ± DD (RDC/TMD) | 33 | Moderate to severe muscle pain (VAS >5/10) with or without limited opening | ≥6 | G1, 15 G2, 15 |
15 F | G1, BTX-A injection G2, NS injection |
Botox |
| Study | Intervention/control | |||||||
|---|---|---|---|---|---|---|---|---|
| Methods to identify muscle | Intervention frequency | Injection muscle, dose (U) × points each side | Unilateral/bilateral | Total dose of BTX | Follow-up, months | Drop-outs | Adverse effects | |
| Ernberg et al. (2011) | Palpation with EMG confirmation | G1, Single session G2, Single session |
G1, Masseter, 10 U × 1 and 20 U × 2 G2, Same volume as G1 |
G1, Unilateral/bilateral G2, Unilateral/bilateral |
50 or 100 U | 1 (short-term) and 3 (long-term) | 1 | Headache (7 after BTX and 9 after NS), muscle weakness (2 after BTX and 4 after NS), increase in pain (3 after BTX and 1 after NS), influenza-like symptoms (2 after BTX and 1 after NS) |
| Guarda-Nardini et al. (2012) | G1, Palpation only G2, Not applicable |
G1, Single session G2, Weekly |
G1, Masseter and temporalis (dose unclear) G2, Not applicable |
G1, Bilateral G2, Bilateral |
300 U | 1 (short-term) and 3 (long-term) | G1, 0 G2, 0 |
G1, 1/15 minor discomfort during chewing for 2–3 weeks G2, Some pain with digital pressure of manipulation points |
| Kurtoglu et al. (2008) | Palpation with EMG confirmation | G1, Single session G2, Single session |
G1, Masseter, 10 U × 3 and temporalis, 10 U × 2 G2, Same volume as G1 |
G1, Bilateral G2, Bilateral |
100 U | 0.5 and 1 | G1, 0 G2, 0 |
G1, Nil G2, Nil |
| von Lindern et al. (2003) | Palpation in accordance with topography | G1, Single session G2, Single session |
G1, Masseter, temporalis, or MTM, areas with pain, dose unknown G2, Same volume as G1 |
G1, Bilateral G2, Bilateral |
Mean 70 U | 1 | G1, 0 G2, 0 |
G1, Temporary (<4 weeks) swallowing difficulty and paralysis of facial expression (1 patient) G2, Nil |
| Nixdorf et al. (2002) | Palpation with EMG confirmation | G1, Single session G2, Single session |
G1, masseter, 50 U × 2 and temporalis, 25 U × 2 G2, Same volume as G1 |
G1, Bilateral G2, Bilateral |
150 U | 2 | 5 | G1, Increasing pain and paralysis of zygomaticus major resulting in asymmetric smile G2, Increasing pain |
All study participants were diagnosed with myofascial TMJ disorders, with or without concurrent disc displacement symptoms. The participants’ symptoms were consistent with chronic muscle pain, and they had either suffered with this for more than 3 months or had experienced no symptom improvement after conservative treatment.
Quality assessment
Figure 2 a presents the individual domain risk of bias for each controlled trial. Of the five controlled trials, two provided little information on how subjects were randomized and one failed to conceal allocation to the patients because the control group was treated by physical manipulation. These studies were deemed unclear and to have a high risk of selection bias. Four studies achieved blinding of the participants, and they were regarded as having a low risk of bias for performance and detection. In the study by Guarda-Nardini et al. using fascial manipulation therapy in the control group, it was not possible to achieve blinding of the patients or clinical practitioners. Regarding attrition bias, one third of subjects (5/15) in the cross-over trial by Nixdorf et al. dropped out of the study, thus this study was considered as having a high risk of bias. The trial by von Lindern et al., which failed to report demographic and comprehensive data results, was regarded as having a moderate risk of bias.
With regard to the required sample size in controlled trials, 80% power and a two-sided confidence level of 95% were needed to give a low-risk bias score. We also needed to postulate the expected risk reduction from the intervention and the ratio of the experimental to the control group in the sample, based on the previous literature and our own judgement. In the cross-over controlled trials, because each patient served as their own control, it was shown that 18 patients would be sufficient to obtain even 90% power. Nevertheless, the study by Nixdorf et al., from which one third of subjects dropped out, did not achieve a sufficient sample size ( n = 10) and it was considered that there would be a high risk of bias in their results.
Finally, regarding reporting bias and the follow-up time, short-term (1 to 2 months of follow-up) outcome measures were available for all of the controlled trials. However, only the cross-over trial by Ernberg et al. and the parallel controlled trial by Guarda-Nardini et al. evaluated the long-term (3-month) effect of BTX. The summary assessment for the overall risk of bias is shown in Figure 2 b.
Dosage and sites of BTX injection
All five studies were targeted primarily on the masseter and temporalis muscles, and most of them administered injections at bilateral muscle sites. In the study by von Lindern et al., in which the participants all had problems of bruxism, injections were given in the masticatory muscles with tenderness, including the medial pterygoid muscles. The method they used to find the muscles to target were all based on physical examination (palpation in accordance with topography), with three of five studies using electromyography (EMG) as guidance.
Regarding the dosage, the total dose of BTX for bilateral muscles ranged from 70 U to 300 U, although the majority used 100–150 U. All of the studies gave a single session of BTX injection and re-evaluated participants at least 1 month following the injection. Only in the studies by Ernberg et al. and Guarda-Nardini et al. were the participants re-evaluated at a long-term follow-up time (3 months) after injection (see Table 1 ).
Effects of BTX on pain improvement in TMD patients
Data examining BTX efficacy and between-group differences from baseline to follow-up are tabulated in Table 2 . Two parallel studies, one by Kurtoglu et al. and the other by von Lindern et al., evaluated the short-term (1 month) effect of BTX compared with a control (normal saline, NS) group. In the study by Kurtoglu et al., pain status was evaluated by RDC/TMD Axis II Biobehavioral Questionnaire (characteristic pain intensity, CPI) filled out by the patients themselves, and the score changes between baseline and days 14 and 28 were analyzed statistically. The BTX group showed a great improvement in pain status, but no statistical difference over time was found in either within-groups or between-groups comparisons. von Lindern et al. used a modified visual analogue pain scale (VAS, 10-point scale) to evaluate pain reduction. A greater number of patients had pain reduction in the BTX group (33/60) than in the NS group (1/30). The study also demonstrated a statistically significant mean reduction in pain in the BTX groups (3.2 points) compared with the NS group (0.4 points). However, no baseline, follow-up, or standard deviation (SD) data were available. The cross-over study by Ernberg et al. used the CPI to evaluate pain status at 1 and 3 months after BTX injection. At the 1-month follow-up, there was no difference in pain relief in the between-groups analysis, and the number of patients who experienced a significant or 30% reduction in pain was not significantly larger in the BTX group than in the NS group.
| Study | Diagnosis | Number of participants | Intervention | Follow-up (short-term), months | Primary outcome | No. with pain improvement a | Pre-treat pain intensity, mean (SD) | Post-treat pain intensity, mean (SD) | Pre-treat MMO, mm, mean (SD) | Post-treat MMO, mm, mean (SD) | Overall risk of bias |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Ernberg et al. (2011) | MFP | 21 (cross-over) | G1, BTX-A injection G2, NS injection |
1 | RDC/TMD questionnaire (CPI) | G1, 9/21 G2, 7/21 |
G1, 69 (11) G2, 67 (14) |
G1, 61 (15) G2, 65 (15) |
G1, 42.7 (11.3) G2, 43.4 (7.3) |
G1, 44.3 (7.2) G2, 44.3 (7.3) |
Low |
| Guarda-Nardini et al. (2012) | MFP ± DD | G1, 15 G2, 15 |
G1, BTX-A injection G2, Fascial manipulation |
1 | Pain VAS (10-point scale) | G1, N/A G2, N/A |
G1, 7.3 (1.1) G2, 6.0 (2.0) |
G1, 5.2 (2.1) G2, 2.1 (1.4) |
G1, 48.7 (8.3) G2, 52.0 (9.5) |
G1, N/A G2, N/A |
High |
| Kurtoglu et al. (2008) | MFP ± DD | G1, 12 G2,12 |
G1, BTX-A injection G2, NS injection |
1 | RDC/TMD questionnaire (CPI) | G1, N/A G2, N/A |
G1, 56.1 (17.1) G2, 58.9 (14.7) |
G1, 43.9 (25.2) G2, 51.4 (23.0) |
G1, N/A G2, N/A |
G1, N/A G2, N/A |
Moderate |
| von Lindern et al. (2003) | MFP with bruxism | G1, 60 G2, 30 |
G1, BTX-A injection G2, NS injection |
1 | Pain VAS (10-point scale) | G1, 33/60 G2, 1/30 |
G1, N/A G2, N/A |
G1, −3.2 (N/A) G2, −0.4 (N/A) |
G1, N/A G2, N/A |
G1, N/A G2, N/A |
High |
| Nixdorf et al. (2002) | MFP ± DD | 10 (cross-over) | G1, BTX-A injection G2, NS injection |
2 | Pain VAS (100-point scale) | G1, N/A G2, N/A |
56 (N/A) | G1, −19 (31) G2, −1 (16) |
43 (N/A) | G1, 0 G2, +10 (9) |
High |
Stay updated, free dental videos. Join our Telegram channel
VIDEdental - Online dental courses
