Journal of
Oral Rehabilitation
Journal of Oral Rehabilitation 2015 42; 847–861
Review
Manual therapy for the management of pain and limited range of motion in subjects with signs and symptoms of temporomandibular disorder: a systematic review of randomised controlled trials L. B. CALIXTRE*, R. F. C. MOREIRA*, G. H. FRANCHINI*, F. ALBURQUERQUES E N D IN † & A . B . O L I V E I R A * *Department of Physiotherapy, Federal University of S~ao Carlos (UFSCar), S~ao Carlos, Brazil and †Department of Nursing and Physiotherapy, Salamanca University, Salamanca, Spain
SUMMARY There is a lack of knowledge about the effectiveness of manual therapy (MT) on subjects with temporomandibular disorders (TMD). The aim of this systematic review is to synthetise evidence regarding the isolated effect of MT in improving maximum mouth opening (MMO) and pain in subjects with signs and symptoms of TMD. MEDLINEâ, Cochrane, Web of Science, SciELO and EMBASETM electronic databases were consulted, searching for randomised controlled trials applying MT for TMD compared to other intervention, no intervention or placebo. Two authors independently extracted data, PEDro scale was used to assess risk of bias, and GRADE (Grading of Recommendations Assessment, Development and Evaluation) was applied to synthetise overall quality of the body of evidence. Treatment effect size was calculated for pain, MMO and pressure pain threshold (PPT). Eight trials were included, seven of high methodological quality. Myofascial release and massage techniques applied on the masticatory muscles are more effective than control
Background Temporomandibular disorder (TMD) is a general term that refers to disorders associated with the temporomandibular joint (TMJ) and masticatory muscles (1). The main signs in patients with TMD are noise and limitations or deviations during mouth opening, and © 2015 John Wiley & Sons Ltd
(low to moderate evidence) but as effective as toxin botulinum injections (moderate evidence). Upper cervical spine thrust manipulation or mobilisation techniques are more effective than control (low to high evidence), while thoracic manipulations are not. There is moderate-to-high evidence that MT techniques protocols are effective. The methodological heterogeneity across trials protocols frequently contributed to decrease quality of evidence. In conclusion, there is widely varying evidence that MT improves pain, MMO and PPT in subjects with TMD signs and symptoms, depending on the technique. Further studies should consider using standardised evaluations and better study designs to strengthen clinical relevance. KEYWORDS: musculoskeletal manipulations, physical therapy modalities, temporomandibular joint disorders, craniomandibular disorders, facial pain, articular range of motion Accepted for publication 17 May 2015
mandibular dysfunction. The main symptoms are pain in the region of the TMJ, the masticatory muscles or both (2). Thirty-nine per cent of the general population presents at least one sign or symptom of TMD (3). There is evidence that occlusal appliances, acupuncture, behavioural therapy, jaw exercises, postural doi: 10.1111/joor.12321
848
L . B . C A L I X T R E et al. training and some pharmacological treatments can be effective in alleviating those signs and symptoms (4). Physiotherapy usually applies different therapy modalities, such as electrical stimulation, low-level laser therapy, ultrasound, acupuncture and relaxation exercises using biofeedback, as well as active exercises and MT (5). These modalities are often combined in rehabilitation protocols. According to the American Academy of Orthopaedic Manual Physical Therapists, MT, also called orthopaedic manual physical therapy, comprises ‘any hands-on treatment provided by the physical therapist’. Treatment may include moving joints in specific directions and at different speeds to regain movement (joint mobilisation and manipulation), muscle stretching, passive movements of the affected body part or having the patient move the body part against the therapist’s resistance to improve muscle activation and timing. Selected specific soft tissue techniques may also be used to improve the mobility and function of tissue and muscles (6). Generally, MT improves circulation, decreases muscle spasm, relaxes muscles around the joint, realigns soft tissue, breaks adhesions, increases range of motion and decreases pain (7). In patients with signs and symptoms of TMD, MT (by itself or in combination with other techniques) has been applied directly to the TMJ and masticatory musculature (8–10), to the cervical spine and neck muscles (11–14) or to both regions or structures (15–18). Two systematic reviews (5, 19) on the effectiveness of physical therapy treatment of patients with TMD were identified in the literature. Both were published in 2006 and synthetised limited evidence regarding the effectiveness of MT techniques to treat those subjects. McNeely et al. (5) found only one primary study using MT in association with exercises, and they highlight the need for better designed studies giving more attention to methodological issues and clinical relevance. Medlicot et al. (19) included nine studies applying MT for TMD. However, most of them were case series, and only three studies were randomised controlled trials (RCT). Furthermore, the authors investigated the effects of MT associated with other therapeutic modalities, leading to a biased evidence synthesis regarding the effectiveness of MT itself. Although physical therapy treatment, in general, presents itself as a multidisciplinary and multimodal
approach, the knowledge about the effectiveness of each technique to improve symptoms is essential for planning an efficient treatment protocol. The use of non-effective techniques may compromise a patient’s adherence to treatment. Therefore, it is important to emphasise that the evaluation of combined techniques jeopardises the identification of which modality is actually effective in the improvement of symptoms. Therefore, the aim of this study is to synthetise evidence regarding the isolated effect of MT in improving TMJ function, considering maximum mouth opening (MMO) and pain as main outcomes. The review followed the Cochrane Collaboration recommendations for systematic reviews (20) and applied the GRADE (Grading of Recommendations Assessment, Development and Evaluation) approach (21).
Methods Data sources and searches Electronic searches were carried out in MEDLINE, Cochrane, Web of Science, SciELO and EMBASE databases, using the following combination of keywords: (Musculoskeletal Manipulations[MeSH Terms] OR massage[MeSH Terms] OR manipulation, chiropractic [MeSH Terms] OR physical therapy modalities[MeSH Terms]) AND (craniomandibular disorders[MeSH Terms] OR temporomandibular joint disorders [MeSH Terms] OR facial pain[MeSH Terms]). The search strategy focused on publications from the last 21 years (1993–2014). There was no restriction regarding the publication language. Grey literature was not accessed; therefore, such publications as reports, conference proceedings, and theses or dissertations were not considered for analysis (22). The reference lists of the selected papers were also checked to identify potentially relevant studies not retrieved in the electronic search. Trial selection Two independent reviewers screened titles and abstracts of the retrieved publications to exclude those not related to the topic of the review. Full texts of potentially relevant articles were retrieved for final evaluation. The whole selection process was conducted by consensus. When consensus was not achieved, a third reviewer was consulted for final © 2015 John Wiley & Sons Ltd
MANUAL THERAPY FOR TEMPOROMANDIBULAR JOINT TREATMENT judgment. START (State of the Art through Systematic Review*), a reference manager software, was used during the selection of papers. Eligibility criteria for initial study selection Type of study. Studies were considered eligible for inclusion if they were RCTs comparing an MT physical therapy intervention to a reference group (placebo intervention, controlled comparison intervention, standard treatment or other treatment). Participants. Trials with adult participants presenting signs and symptoms of temporomandibular disorders were included. Studies that applied manual therapy in non-symptomatic subjects were excluded. Type of intervention. Primary studies addressing the treatment of TMJ impairments through MT protocols with emphasis on joint, ligaments, fascia, muscles or a combination of these techniques were included. At least one of the intervention groups must have exclusively received some modality of MT. Temporomandibular joint treatments after surgery, fracture, dislocation or trauma were not included in the review, and hyperplasia, ankylosis, synovitis, Eagle’s syndrome and treatments for rheumatic diseases were excluded. Outcome measures. Studies reporting outcomes related to pain, MMO or pressure pain threshold (PPT) of the masticatory muscles were included. These outcomes were selected because they represent the main complaints of subjects with TMD. The methods applied to evaluate outcome measures had to be valid and reliable. Quality assessment and clinical relevance The PEDro (Physiotherapy Evidence-Based Database) scale was used to assess the methodological quality of the studies included in this review. The reliability of this tool is fair to good (23). Although this scale has 11 items, specification of eligibility criteria was not accounted in the final score, so the total score ranges from zero to ten. Each satisfied item scored one point, and the final score was obtained by the sum of all positive answers. The stud*Laboratory of Research on Software Engineering from Federal University of Sc´o Carlos, Sc´o Carlos, Brazil. © 2015 John Wiley & Sons Ltd
ies indexed in the PEDro database already had a rating, which was maintained. The non-indexed studies were independently evaluated by two reviewers (L.B.C., G.H.F.). In case of disagreement, a third reviewer (R.F.C.M.) was consulted. The literature suggests that high-quality studies should achieve a total score higher than 50% of the possible maximum (24, 25). Thus, for this review, RCTs with a PEDro score higher than or equal to 50 were classified as high methodological quality studies. Recommended criteria proposed by Higgins and Green (20) were used to evaluate clinical relevance. Data extraction After the final consensus and selection of the primary studies, the reviewers (L.B.C., G.H.F.) worked independently. A standardised form, adapted to the model proposed by the Cochrane Collaboration, was used for data extraction including information regarding study design, follow-up, sample characteristics, intervention characteristics (type, duration and number of sessions), comparison groups and outcomes. Data synthesis and analysis Because of the heterogeneity of the primary studies, it was not possible to perform a meta-analysis. To compare the effect size (ES) of each MT technique, standardised mean difference (26) was calculated for each comparison group separately, considering the values before and after intervention. These were further classified as small (080), according to Cohen’s criteria (27). The quality of the body of evidence was determined using the GRADE approach, which analyses the following domains: trial design limitations due to risk of bias, inconsistency of results, indirectness, imprecision of results and publication bias. The details of this method have been reported previously (21, 28, 29); a detailed description is presented in Data S1. The studies were primarily divided according to the body structure that received MT: masticatory musculature, cervical or thoracic spine, or a combination of several techniques in an MT protocol. Subsequently, distinctions were made according to the outcomes (pain, MMO and PPT) and comparison groups.
849
850
L . B . C A L I X T R E et al.
Results Trial selection The electronic search returned 6581 published references. The final selection process resulted in eight included studies for evidence synthesis (13, 16, 17, 30–34). The main reasons for exclusion are specified in Data S2. Most of them were excluded because of the trial design (non-randomised and controlled studies), and the combination of therapies. Details of the selection process are presented in Fig. 1.
The most critical criteria to be satisfied were blinding subjects (25%), intention-to-treat analysis (375%) and concealed allocation (625%). The studies presented a mean score of 29 on the clinical relevance scale. Only one study (32) achieved the highest score (five points). The least satisfied criterion was related to the clinical importance of the effect size, achieved by two primary studies (17, 32). Only three studies (13, 32, 34) provided confidence intervals. Sample size calculation was reported by five studies (30, 32–35) and the ES by three studies (13, 33, 34). Finally, none of the studies tested the reliability of the assessment tool.
Quality assessment and clinical relevance Seven studies presented high methodological quality. The mean PEDro total score obtained for the studies was 662 (Table 1). Only one study (13) was classified as low methodological quality. Only one study (30) was not indexed by the PEDro database. Therefore, the reviewers assessed its methodological quality.
Trial characteristics The characteristics of the eight included papers are listed in Table 2. Although all included studies were RCT, there was a variation in the applied protocols regarding the number of sessions and the frequency of therapy application.
Fig. 1. Trial flow of the selection process. © 2015 John Wiley & Sons Ltd
MANUAL THERAPY FOR TEMPOROMANDIBULAR JOINT TREATMENT Table 1. Methodological quality evaluated according to PEDro scale and clinical relevance criteria Risk of bias (PEDro scale)
Study
1
2
3
4
Kalamir et al. 2012 (31) Guarda Nardini et al. 2012 (30) Gomes et al. 2014 (33) Mansilla-Ferragut et al. 2008 (13) La Touche et al. 2013 (32) Packer et al. 2014 (34) Cuccia et al. 2010 (16) Piekartz et al. 2011 (17) Percentage of studies which satisfied the criteria
+
+
+
+
+
+
+
+ +
+ +
+
+
+
+
+
+ + +
+ + +
+
100
100
625
5
Clinical relevance
6
+
7
8
9
10
11
Total score
12
13
14
+
+
+
+
+
8
+
+
+
+
+
5
+
+ +
+ +
7 4
+
+
+
8
+ + +
+ + +
9 6 6
100
100
+ +
+
+
+
+
+ + +
+
+ + +
+ + +
+
75
25
875
75
375
0
16
Total Score
+
+
4
+
+
3
+ +
+
+ +
2 4
+
+
+
+
+
5
+ + +
+
+ + +
+ +
+ +
3 4 4
875
625
975
375
975
15
1. Eligibility criteria; 2. Random allocation; 3. Concealed allocation; 4. Baseline comparability; 5. Blind subjects; 6. Blind therapists; 7. Blind assessors; 8. Adequate follow-up; 9. Intention-to-treat analysis; 10. Between-group comparisons; 11. Point estimates and variability; 12. Are the patients described in detail so that you can decide whether they are comparable to those that you see in your practice? 13. Are the interventions and treatment settings described well enough so that you can provide the same for your patients? 14. Were all clinically relevant outcomes measured and reported? 15. Is the size of the effect clinically important? 16. Are the likely treatment benefits worth the potential harms? Eligibility criteria item is not accounted in the total score.
Two of these studies (13, 36) analysed the immediate effect of MT. The six remaining studies performed a longer follow-up evaluation after the treatment. Two of them (30, 32) applied a protocol of three sessions, while the other three studies (17, 31, 33) used longer protocols (more than three sessions). One study (16) did not specify the number of sessions applied. Outcome measurement tools Pain intensity at rest was evaluated through the visual analogue scale in five studies (16, 30–32, 34). The coloured analogue scale was also applied (17) both at rest and at MMO. MMO measurements were taken in all studies except for 2 (32, 34), while PPT was assessed in three of eight included studies (13, 32, 34). Characteristics of subjects included in the primary studies The total number of participants in the studies ranged from 30 to 93, and the mean number of subjects was 456 182. All evaluated subjects were adults, based © 2015 John Wiley & Sons Ltd
on a minimum age of 18 years and a maximum age of 50 to 65 years. Females predominated (626% of the sample) in most of the studies, and in one study (34), the sample was composed exclusively of women. Four studies (30–32, 34) used the standardised evaluation protocol Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD) (37): All of them included patients with myogenic dysfunction. The remaining studies adopted non-standard evaluation protocols. In general, the presence of signs and symptoms of TMD (13, 16, 17, 33) was required for subject inclusion, and some studies (13, 17, 32, 34) also included subjects with TMD and associated neck pain symptoms. Most of these studies have clearly reported their exclusion criteria (16, 30–34). The most frequent criteria adopted were subjects who suffer from malignancy, fibromyalgia, TMJ inflammatory arthritis, or from metabolic, connective tissue, rheumatic, or haematological diseases. Subjects with previous oral or cervical surgery, history of neck trauma or whiplash were also excluded. Studies involving manipulative
851
Risk of Bias: 4 Clin. Relevance: 4
Mansilla-Ferragut et al. (13)
Risk of Bias:7 Clin. Relevance: 2
Risk of Bias: 5 Clin. Relevance: 3 Gomes et al. (33)
Guarda Nardini et al. (30)
Subjects with 1 month, not received any osteopathic treatment 3 months in duration and minimum baseline graded chronic pain score of 3 Age: 18–50 years N: 93 (52 W and 41 M)
Kalamir et al. (31)
Risk of Bias: 8 Clin. Relevance: 4
Subjects’ characteristics
Trial
(b) Asymptomatic comparison group was not submitted to any form of intervention (not considered in analysis) (a) 1 session of atlanto-occipital joint thrust manipulation
(a) Massage therapy involving sliding and kneading manoeuvres on the masseters and temporalis muscles
(a) 3(1) sessions based on fascial manipulation guidelines
(b) control group
(b) 1 session of multiple botulin toxin injections in temporalis and masseters muscles (c) treatment with occlusal splint for 4 weeks
PPT of temporalis muscles (algometer)
MMO (calliper)
Pain (VAS) MMO (calliper)
Jaw pain at rest, maximal active opening and clenching (VAS) MMO (NE tool)
MMO (calliper)
(c) control group
(a) 10 sessions of intra-oral temporalis release, intra-oral medial and lateral pterygoid (origin) technique, intra-oral sphenopalatine ganglion technique (b) group A +instructions + home exercise (not considered in analysis)
Outcomes and tools
Comparison Group
Intervention group(s)
Table 2. Characteristics of primary studies considering study design, subjects, intervention and outcomes
Immediately after intervention
After 4 weeks of intervention
Immediately after intervention, and 2 months follow-up
After five weeks of intervention, six months and a year follow-up
Follow-up
852
L . B . C A L I X T R E et al.
© 2015 John Wiley & Sons Ltd
© 2015 John Wiley & Sons Ltd
Risk of Bias: 6 Clin. Relevance: 4
Risk of Bias: 9 Clin. Relevance: 3 Cuccia et al. (16)
Packer et al. (34)
Myofascial pain or mixed TMD according to the RDC/TMD, complaints of pain or fatigue in the masticatory muscles during functional activities >6 months, diagnosis of neck pain (NDI) and body mass index 3 mm on VAS scale; neck and/or shoulder pain; NDI ≥ 15 Age: 3319 949 and 3456 784 years 32 (21 W and 11 M)
La Touche et al. (32)
Risk of Bias: 8 Clin. Relevance: 5
Subjects’ characteristics
Trial
Table 2. (continued)
(a) OMT: myofascial release, balanced membranous tension, muscle energy, myofascial release, joint articulation, high-velocity, low-amplitude thrust and cranialsacral therapy
(a) Upper thoracic manipulation technique applied to the T1 vertebral segment. Therapist hand in pistol grip.
(b) conventional treatment: oral appliance, muscle stretching and relaxing exercises, hot and/or cold packs, TENS
(b) placebo technique: same position of the subject; however, therapist hand was open and not in pistol grip.
Pain (VAS)
PPT in masseters and temporalis MMO (calliper)
After intervention and 2 months follow-up
Immediately, 48 and 72 h after the procedure
After 3 sessions of intervention
Pain (VAS)
(b) Placebo technique: same hand position of the therapist. However, mobilisation was not applied to the cervical spine
(a) Anterior–posterior upper cervical mobilisation on the 3 upper cervical segments (C0-C3). The mobilisation was applied at a rate of 1 oscillation per 2 s (05 Hz)
PPT in masseters and temporalis (2 points each muscle) Pain (VAS)
Follow-up
Outcomes and tools
Comparison Group
Intervention group(s)
MANUAL THERAPY FOR TEMPOROMANDIBULAR JOINT TREATMENT 853
Risk of Bias: 6 Clin. Relevance: 4
N, Number of subjects; Clin, Clinical, MMO; Maximum mouth opening; NE, Not specified; W, Women; M, Men; TrPs, Trigger points; TMJ, Temporomandibular joint; TMD, Temporomandibular disorder; VAS, Visual analogue scale; CAS, Coloured analogue scale; PPT, Pressure pain threshold; TMI, Temporomandibular index; RDC/TMD, Research Diagnostic Criteria for Temporomandibular Disorders; OMT, Osteopathic manual therapy; NDI, Neck Disability Index; ICDCH, International Classification of Diagnostic Criteria of Headaches; C1, First cervical vertebrae; FAI, Fonseca Anamnestic Index.
Pain (CAS)
After 6 weeks of intervention and 6 months follow-up MMO (ruler) (b) usual care group: continued their treatment of the craniocervical region Subjects with 4 of following signals: joint sounds, deviation during MMO, extra oral muscle pain muscle pain at a minimum of two tender points in the masseters or temporalis muscles and pain during passive MO + cervicogenic headache according to the ICDCH lasting more than 3 months and NDI > 15 Age: 18–65 years N: 43 (27 W and 16 M) Piekartz et al. (17)
(a) translator movements of the TMJ and/or masticatory muscle techniques (TrPs treatment and muscle stretching), active and passive movements facilitating optimal function of cranial nerve tissue, coordination and home exercises, depending on the therapists’ clinical decisions. The therapist could also, when necessary, opt for additional treatment of the neck
Follow-up Comparison Group Subjects’ characteristics Trial
Intervention group(s)
Outcomes and tools
L . B . C A L I X T R E et al.
Table 2. (continued)
854
techniques to the cervical spine (13, 16) have applied other specific exclusion criteria. Manual therapy techniques Some studies applied intra-oral (31) myofascial release techniques as well as massage therapy (33) on masticatory muscles. One study (13) applied an atlanto-occipital joint thrust manipulation, another (34) applied a thoracic manipulation, and another applied a mobilisation for the upper cervical spine (32). Finally, two studies (16, 17) applied many MT techniques to treat patients with signs and symptoms of TMD. Effect of manual therapy techniques The quality of evidence for each MT technique according to GRADE approach is presented in Tables 3–5. Reasons for downgrading the quality of the body of evidence are cited in the legends below each table. The evidence synthesis showed moderate and low evidence that myofascial release and massage techniques are more effective than placebo or no intervention for MMO and pain outcomes respectively. There is also moderate evidence that no significant difference exists between myofascial release and toxin botulinum for improvement on the same outcomes. There is moderate evidence that atlanto-occipital joint thrust manipulation is more effective than placebo in improving MMO in individuals with TMD. However, there is low evidence that C7-T1 thrust manipulation does not improve PPT in subjects with TMD when compared to placebo manipulation. In addition, an upper cervical mobilisation proposed by La Touche et al. (32) showed high evidence on reducing pain and increasing PPT comparing to placebo. Finally, MT osteopathic technique protocols showed greater MMO, pain and PPT improvement when compared to a usual care group (revealing high evidence for MMO and moderate evidence for pain and PPT). Only two studies (30, 31) reported absence of adverse effects as a consequence of the applied techniques; the remaining did not report.
Discussion Most of the RCTs included in this review were high methodological quality studies and presented positive © 2015 John Wiley & Sons Ltd
© 2015 John Wiley & Sons Ltd
Serious 3 Serious 3
Serious 2,3 Serious 2
Imprecision
N/A N/A
N/A Undetected
Publication Bias
Guarda-Nardini (30) Guarda-Nardini (30)
Kalamir (31) Kalamir (31) Gomes (33)
Trial
Serious 1 – >75% of the sample show conflicting results. Serious 2 –
Oral Rehabilitation
Journal of Oral Rehabilitation 2015 42; 847–861
Review
Manual therapy for the management of pain and limited range of motion in subjects with signs and symptoms of temporomandibular disorder: a systematic review of randomised controlled trials L. B. CALIXTRE*, R. F. C. MOREIRA*, G. H. FRANCHINI*, F. ALBURQUERQUES E N D IN † & A . B . O L I V E I R A * *Department of Physiotherapy, Federal University of S~ao Carlos (UFSCar), S~ao Carlos, Brazil and †Department of Nursing and Physiotherapy, Salamanca University, Salamanca, Spain
SUMMARY There is a lack of knowledge about the effectiveness of manual therapy (MT) on subjects with temporomandibular disorders (TMD). The aim of this systematic review is to synthetise evidence regarding the isolated effect of MT in improving maximum mouth opening (MMO) and pain in subjects with signs and symptoms of TMD. MEDLINEâ, Cochrane, Web of Science, SciELO and EMBASETM electronic databases were consulted, searching for randomised controlled trials applying MT for TMD compared to other intervention, no intervention or placebo. Two authors independently extracted data, PEDro scale was used to assess risk of bias, and GRADE (Grading of Recommendations Assessment, Development and Evaluation) was applied to synthetise overall quality of the body of evidence. Treatment effect size was calculated for pain, MMO and pressure pain threshold (PPT). Eight trials were included, seven of high methodological quality. Myofascial release and massage techniques applied on the masticatory muscles are more effective than control
Background Temporomandibular disorder (TMD) is a general term that refers to disorders associated with the temporomandibular joint (TMJ) and masticatory muscles (1). The main signs in patients with TMD are noise and limitations or deviations during mouth opening, and © 2015 John Wiley & Sons Ltd
(low to moderate evidence) but as effective as toxin botulinum injections (moderate evidence). Upper cervical spine thrust manipulation or mobilisation techniques are more effective than control (low to high evidence), while thoracic manipulations are not. There is moderate-to-high evidence that MT techniques protocols are effective. The methodological heterogeneity across trials protocols frequently contributed to decrease quality of evidence. In conclusion, there is widely varying evidence that MT improves pain, MMO and PPT in subjects with TMD signs and symptoms, depending on the technique. Further studies should consider using standardised evaluations and better study designs to strengthen clinical relevance. KEYWORDS: musculoskeletal manipulations, physical therapy modalities, temporomandibular joint disorders, craniomandibular disorders, facial pain, articular range of motion Accepted for publication 17 May 2015
mandibular dysfunction. The main symptoms are pain in the region of the TMJ, the masticatory muscles or both (2). Thirty-nine per cent of the general population presents at least one sign or symptom of TMD (3). There is evidence that occlusal appliances, acupuncture, behavioural therapy, jaw exercises, postural doi: 10.1111/joor.12321
848
L . B . C A L I X T R E et al. training and some pharmacological treatments can be effective in alleviating those signs and symptoms (4). Physiotherapy usually applies different therapy modalities, such as electrical stimulation, low-level laser therapy, ultrasound, acupuncture and relaxation exercises using biofeedback, as well as active exercises and MT (5). These modalities are often combined in rehabilitation protocols. According to the American Academy of Orthopaedic Manual Physical Therapists, MT, also called orthopaedic manual physical therapy, comprises ‘any hands-on treatment provided by the physical therapist’. Treatment may include moving joints in specific directions and at different speeds to regain movement (joint mobilisation and manipulation), muscle stretching, passive movements of the affected body part or having the patient move the body part against the therapist’s resistance to improve muscle activation and timing. Selected specific soft tissue techniques may also be used to improve the mobility and function of tissue and muscles (6). Generally, MT improves circulation, decreases muscle spasm, relaxes muscles around the joint, realigns soft tissue, breaks adhesions, increases range of motion and decreases pain (7). In patients with signs and symptoms of TMD, MT (by itself or in combination with other techniques) has been applied directly to the TMJ and masticatory musculature (8–10), to the cervical spine and neck muscles (11–14) or to both regions or structures (15–18). Two systematic reviews (5, 19) on the effectiveness of physical therapy treatment of patients with TMD were identified in the literature. Both were published in 2006 and synthetised limited evidence regarding the effectiveness of MT techniques to treat those subjects. McNeely et al. (5) found only one primary study using MT in association with exercises, and they highlight the need for better designed studies giving more attention to methodological issues and clinical relevance. Medlicot et al. (19) included nine studies applying MT for TMD. However, most of them were case series, and only three studies were randomised controlled trials (RCT). Furthermore, the authors investigated the effects of MT associated with other therapeutic modalities, leading to a biased evidence synthesis regarding the effectiveness of MT itself. Although physical therapy treatment, in general, presents itself as a multidisciplinary and multimodal
approach, the knowledge about the effectiveness of each technique to improve symptoms is essential for planning an efficient treatment protocol. The use of non-effective techniques may compromise a patient’s adherence to treatment. Therefore, it is important to emphasise that the evaluation of combined techniques jeopardises the identification of which modality is actually effective in the improvement of symptoms. Therefore, the aim of this study is to synthetise evidence regarding the isolated effect of MT in improving TMJ function, considering maximum mouth opening (MMO) and pain as main outcomes. The review followed the Cochrane Collaboration recommendations for systematic reviews (20) and applied the GRADE (Grading of Recommendations Assessment, Development and Evaluation) approach (21).
Methods Data sources and searches Electronic searches were carried out in MEDLINE, Cochrane, Web of Science, SciELO and EMBASE databases, using the following combination of keywords: (Musculoskeletal Manipulations[MeSH Terms] OR massage[MeSH Terms] OR manipulation, chiropractic [MeSH Terms] OR physical therapy modalities[MeSH Terms]) AND (craniomandibular disorders[MeSH Terms] OR temporomandibular joint disorders [MeSH Terms] OR facial pain[MeSH Terms]). The search strategy focused on publications from the last 21 years (1993–2014). There was no restriction regarding the publication language. Grey literature was not accessed; therefore, such publications as reports, conference proceedings, and theses or dissertations were not considered for analysis (22). The reference lists of the selected papers were also checked to identify potentially relevant studies not retrieved in the electronic search. Trial selection Two independent reviewers screened titles and abstracts of the retrieved publications to exclude those not related to the topic of the review. Full texts of potentially relevant articles were retrieved for final evaluation. The whole selection process was conducted by consensus. When consensus was not achieved, a third reviewer was consulted for final © 2015 John Wiley & Sons Ltd
MANUAL THERAPY FOR TEMPOROMANDIBULAR JOINT TREATMENT judgment. START (State of the Art through Systematic Review*), a reference manager software, was used during the selection of papers. Eligibility criteria for initial study selection Type of study. Studies were considered eligible for inclusion if they were RCTs comparing an MT physical therapy intervention to a reference group (placebo intervention, controlled comparison intervention, standard treatment or other treatment). Participants. Trials with adult participants presenting signs and symptoms of temporomandibular disorders were included. Studies that applied manual therapy in non-symptomatic subjects were excluded. Type of intervention. Primary studies addressing the treatment of TMJ impairments through MT protocols with emphasis on joint, ligaments, fascia, muscles or a combination of these techniques were included. At least one of the intervention groups must have exclusively received some modality of MT. Temporomandibular joint treatments after surgery, fracture, dislocation or trauma were not included in the review, and hyperplasia, ankylosis, synovitis, Eagle’s syndrome and treatments for rheumatic diseases were excluded. Outcome measures. Studies reporting outcomes related to pain, MMO or pressure pain threshold (PPT) of the masticatory muscles were included. These outcomes were selected because they represent the main complaints of subjects with TMD. The methods applied to evaluate outcome measures had to be valid and reliable. Quality assessment and clinical relevance The PEDro (Physiotherapy Evidence-Based Database) scale was used to assess the methodological quality of the studies included in this review. The reliability of this tool is fair to good (23). Although this scale has 11 items, specification of eligibility criteria was not accounted in the final score, so the total score ranges from zero to ten. Each satisfied item scored one point, and the final score was obtained by the sum of all positive answers. The stud*Laboratory of Research on Software Engineering from Federal University of Sc´o Carlos, Sc´o Carlos, Brazil. © 2015 John Wiley & Sons Ltd
ies indexed in the PEDro database already had a rating, which was maintained. The non-indexed studies were independently evaluated by two reviewers (L.B.C., G.H.F.). In case of disagreement, a third reviewer (R.F.C.M.) was consulted. The literature suggests that high-quality studies should achieve a total score higher than 50% of the possible maximum (24, 25). Thus, for this review, RCTs with a PEDro score higher than or equal to 50 were classified as high methodological quality studies. Recommended criteria proposed by Higgins and Green (20) were used to evaluate clinical relevance. Data extraction After the final consensus and selection of the primary studies, the reviewers (L.B.C., G.H.F.) worked independently. A standardised form, adapted to the model proposed by the Cochrane Collaboration, was used for data extraction including information regarding study design, follow-up, sample characteristics, intervention characteristics (type, duration and number of sessions), comparison groups and outcomes. Data synthesis and analysis Because of the heterogeneity of the primary studies, it was not possible to perform a meta-analysis. To compare the effect size (ES) of each MT technique, standardised mean difference (26) was calculated for each comparison group separately, considering the values before and after intervention. These were further classified as small (080), according to Cohen’s criteria (27). The quality of the body of evidence was determined using the GRADE approach, which analyses the following domains: trial design limitations due to risk of bias, inconsistency of results, indirectness, imprecision of results and publication bias. The details of this method have been reported previously (21, 28, 29); a detailed description is presented in Data S1. The studies were primarily divided according to the body structure that received MT: masticatory musculature, cervical or thoracic spine, or a combination of several techniques in an MT protocol. Subsequently, distinctions were made according to the outcomes (pain, MMO and PPT) and comparison groups.
849
850
L . B . C A L I X T R E et al.
Results Trial selection The electronic search returned 6581 published references. The final selection process resulted in eight included studies for evidence synthesis (13, 16, 17, 30–34). The main reasons for exclusion are specified in Data S2. Most of them were excluded because of the trial design (non-randomised and controlled studies), and the combination of therapies. Details of the selection process are presented in Fig. 1.
The most critical criteria to be satisfied were blinding subjects (25%), intention-to-treat analysis (375%) and concealed allocation (625%). The studies presented a mean score of 29 on the clinical relevance scale. Only one study (32) achieved the highest score (five points). The least satisfied criterion was related to the clinical importance of the effect size, achieved by two primary studies (17, 32). Only three studies (13, 32, 34) provided confidence intervals. Sample size calculation was reported by five studies (30, 32–35) and the ES by three studies (13, 33, 34). Finally, none of the studies tested the reliability of the assessment tool.
Quality assessment and clinical relevance Seven studies presented high methodological quality. The mean PEDro total score obtained for the studies was 662 (Table 1). Only one study (13) was classified as low methodological quality. Only one study (30) was not indexed by the PEDro database. Therefore, the reviewers assessed its methodological quality.
Trial characteristics The characteristics of the eight included papers are listed in Table 2. Although all included studies were RCT, there was a variation in the applied protocols regarding the number of sessions and the frequency of therapy application.
Fig. 1. Trial flow of the selection process. © 2015 John Wiley & Sons Ltd
MANUAL THERAPY FOR TEMPOROMANDIBULAR JOINT TREATMENT Table 1. Methodological quality evaluated according to PEDro scale and clinical relevance criteria Risk of bias (PEDro scale)
Study
1
2
3
4
Kalamir et al. 2012 (31) Guarda Nardini et al. 2012 (30) Gomes et al. 2014 (33) Mansilla-Ferragut et al. 2008 (13) La Touche et al. 2013 (32) Packer et al. 2014 (34) Cuccia et al. 2010 (16) Piekartz et al. 2011 (17) Percentage of studies which satisfied the criteria
+
+
+
+
+
+
+
+ +
+ +
+
+
+
+
+
+ + +
+ + +
+
100
100
625
5
Clinical relevance
6
+
7
8
9
10
11
Total score
12
13
14
+
+
+
+
+
8
+
+
+
+
+
5
+
+ +
+ +
7 4
+
+
+
8
+ + +
+ + +
9 6 6
100
100
+ +
+
+
+
+
+ + +
+
+ + +
+ + +
+
75
25
875
75
375
0
16
Total Score
+
+
4
+
+
3
+ +
+
+ +
2 4
+
+
+
+
+
5
+ + +
+
+ + +
+ +
+ +
3 4 4
875
625
975
375
975
15
1. Eligibility criteria; 2. Random allocation; 3. Concealed allocation; 4. Baseline comparability; 5. Blind subjects; 6. Blind therapists; 7. Blind assessors; 8. Adequate follow-up; 9. Intention-to-treat analysis; 10. Between-group comparisons; 11. Point estimates and variability; 12. Are the patients described in detail so that you can decide whether they are comparable to those that you see in your practice? 13. Are the interventions and treatment settings described well enough so that you can provide the same for your patients? 14. Were all clinically relevant outcomes measured and reported? 15. Is the size of the effect clinically important? 16. Are the likely treatment benefits worth the potential harms? Eligibility criteria item is not accounted in the total score.
Two of these studies (13, 36) analysed the immediate effect of MT. The six remaining studies performed a longer follow-up evaluation after the treatment. Two of them (30, 32) applied a protocol of three sessions, while the other three studies (17, 31, 33) used longer protocols (more than three sessions). One study (16) did not specify the number of sessions applied. Outcome measurement tools Pain intensity at rest was evaluated through the visual analogue scale in five studies (16, 30–32, 34). The coloured analogue scale was also applied (17) both at rest and at MMO. MMO measurements were taken in all studies except for 2 (32, 34), while PPT was assessed in three of eight included studies (13, 32, 34). Characteristics of subjects included in the primary studies The total number of participants in the studies ranged from 30 to 93, and the mean number of subjects was 456 182. All evaluated subjects were adults, based © 2015 John Wiley & Sons Ltd
on a minimum age of 18 years and a maximum age of 50 to 65 years. Females predominated (626% of the sample) in most of the studies, and in one study (34), the sample was composed exclusively of women. Four studies (30–32, 34) used the standardised evaluation protocol Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD) (37): All of them included patients with myogenic dysfunction. The remaining studies adopted non-standard evaluation protocols. In general, the presence of signs and symptoms of TMD (13, 16, 17, 33) was required for subject inclusion, and some studies (13, 17, 32, 34) also included subjects with TMD and associated neck pain symptoms. Most of these studies have clearly reported their exclusion criteria (16, 30–34). The most frequent criteria adopted were subjects who suffer from malignancy, fibromyalgia, TMJ inflammatory arthritis, or from metabolic, connective tissue, rheumatic, or haematological diseases. Subjects with previous oral or cervical surgery, history of neck trauma or whiplash were also excluded. Studies involving manipulative
851
Risk of Bias: 4 Clin. Relevance: 4
Mansilla-Ferragut et al. (13)
Risk of Bias:7 Clin. Relevance: 2
Risk of Bias: 5 Clin. Relevance: 3 Gomes et al. (33)
Guarda Nardini et al. (30)
Subjects with 1 month, not received any osteopathic treatment 3 months in duration and minimum baseline graded chronic pain score of 3 Age: 18–50 years N: 93 (52 W and 41 M)
Kalamir et al. (31)
Risk of Bias: 8 Clin. Relevance: 4
Subjects’ characteristics
Trial
(b) Asymptomatic comparison group was not submitted to any form of intervention (not considered in analysis) (a) 1 session of atlanto-occipital joint thrust manipulation
(a) Massage therapy involving sliding and kneading manoeuvres on the masseters and temporalis muscles
(a) 3(1) sessions based on fascial manipulation guidelines
(b) control group
(b) 1 session of multiple botulin toxin injections in temporalis and masseters muscles (c) treatment with occlusal splint for 4 weeks
PPT of temporalis muscles (algometer)
MMO (calliper)
Pain (VAS) MMO (calliper)
Jaw pain at rest, maximal active opening and clenching (VAS) MMO (NE tool)
MMO (calliper)
(c) control group
(a) 10 sessions of intra-oral temporalis release, intra-oral medial and lateral pterygoid (origin) technique, intra-oral sphenopalatine ganglion technique (b) group A +instructions + home exercise (not considered in analysis)
Outcomes and tools
Comparison Group
Intervention group(s)
Table 2. Characteristics of primary studies considering study design, subjects, intervention and outcomes
Immediately after intervention
After 4 weeks of intervention
Immediately after intervention, and 2 months follow-up
After five weeks of intervention, six months and a year follow-up
Follow-up
852
L . B . C A L I X T R E et al.
© 2015 John Wiley & Sons Ltd
© 2015 John Wiley & Sons Ltd
Risk of Bias: 6 Clin. Relevance: 4
Risk of Bias: 9 Clin. Relevance: 3 Cuccia et al. (16)
Packer et al. (34)
Myofascial pain or mixed TMD according to the RDC/TMD, complaints of pain or fatigue in the masticatory muscles during functional activities >6 months, diagnosis of neck pain (NDI) and body mass index 3 mm on VAS scale; neck and/or shoulder pain; NDI ≥ 15 Age: 3319 949 and 3456 784 years 32 (21 W and 11 M)
La Touche et al. (32)
Risk of Bias: 8 Clin. Relevance: 5
Subjects’ characteristics
Trial
Table 2. (continued)
(a) OMT: myofascial release, balanced membranous tension, muscle energy, myofascial release, joint articulation, high-velocity, low-amplitude thrust and cranialsacral therapy
(a) Upper thoracic manipulation technique applied to the T1 vertebral segment. Therapist hand in pistol grip.
(b) conventional treatment: oral appliance, muscle stretching and relaxing exercises, hot and/or cold packs, TENS
(b) placebo technique: same position of the subject; however, therapist hand was open and not in pistol grip.
Pain (VAS)
PPT in masseters and temporalis MMO (calliper)
After intervention and 2 months follow-up
Immediately, 48 and 72 h after the procedure
After 3 sessions of intervention
Pain (VAS)
(b) Placebo technique: same hand position of the therapist. However, mobilisation was not applied to the cervical spine
(a) Anterior–posterior upper cervical mobilisation on the 3 upper cervical segments (C0-C3). The mobilisation was applied at a rate of 1 oscillation per 2 s (05 Hz)
PPT in masseters and temporalis (2 points each muscle) Pain (VAS)
Follow-up
Outcomes and tools
Comparison Group
Intervention group(s)
MANUAL THERAPY FOR TEMPOROMANDIBULAR JOINT TREATMENT 853
Risk of Bias: 6 Clin. Relevance: 4
N, Number of subjects; Clin, Clinical, MMO; Maximum mouth opening; NE, Not specified; W, Women; M, Men; TrPs, Trigger points; TMJ, Temporomandibular joint; TMD, Temporomandibular disorder; VAS, Visual analogue scale; CAS, Coloured analogue scale; PPT, Pressure pain threshold; TMI, Temporomandibular index; RDC/TMD, Research Diagnostic Criteria for Temporomandibular Disorders; OMT, Osteopathic manual therapy; NDI, Neck Disability Index; ICDCH, International Classification of Diagnostic Criteria of Headaches; C1, First cervical vertebrae; FAI, Fonseca Anamnestic Index.
Pain (CAS)
After 6 weeks of intervention and 6 months follow-up MMO (ruler) (b) usual care group: continued their treatment of the craniocervical region Subjects with 4 of following signals: joint sounds, deviation during MMO, extra oral muscle pain muscle pain at a minimum of two tender points in the masseters or temporalis muscles and pain during passive MO + cervicogenic headache according to the ICDCH lasting more than 3 months and NDI > 15 Age: 18–65 years N: 43 (27 W and 16 M) Piekartz et al. (17)
(a) translator movements of the TMJ and/or masticatory muscle techniques (TrPs treatment and muscle stretching), active and passive movements facilitating optimal function of cranial nerve tissue, coordination and home exercises, depending on the therapists’ clinical decisions. The therapist could also, when necessary, opt for additional treatment of the neck
Follow-up Comparison Group Subjects’ characteristics Trial
Intervention group(s)
Outcomes and tools
L . B . C A L I X T R E et al.
Table 2. (continued)
854
techniques to the cervical spine (13, 16) have applied other specific exclusion criteria. Manual therapy techniques Some studies applied intra-oral (31) myofascial release techniques as well as massage therapy (33) on masticatory muscles. One study (13) applied an atlanto-occipital joint thrust manipulation, another (34) applied a thoracic manipulation, and another applied a mobilisation for the upper cervical spine (32). Finally, two studies (16, 17) applied many MT techniques to treat patients with signs and symptoms of TMD. Effect of manual therapy techniques The quality of evidence for each MT technique according to GRADE approach is presented in Tables 3–5. Reasons for downgrading the quality of the body of evidence are cited in the legends below each table. The evidence synthesis showed moderate and low evidence that myofascial release and massage techniques are more effective than placebo or no intervention for MMO and pain outcomes respectively. There is also moderate evidence that no significant difference exists between myofascial release and toxin botulinum for improvement on the same outcomes. There is moderate evidence that atlanto-occipital joint thrust manipulation is more effective than placebo in improving MMO in individuals with TMD. However, there is low evidence that C7-T1 thrust manipulation does not improve PPT in subjects with TMD when compared to placebo manipulation. In addition, an upper cervical mobilisation proposed by La Touche et al. (32) showed high evidence on reducing pain and increasing PPT comparing to placebo. Finally, MT osteopathic technique protocols showed greater MMO, pain and PPT improvement when compared to a usual care group (revealing high evidence for MMO and moderate evidence for pain and PPT). Only two studies (30, 31) reported absence of adverse effects as a consequence of the applied techniques; the remaining did not report.
Discussion Most of the RCTs included in this review were high methodological quality studies and presented positive © 2015 John Wiley & Sons Ltd
© 2015 John Wiley & Sons Ltd
Serious 3 Serious 3
Serious 2,3 Serious 2
Imprecision
N/A N/A
N/A Undetected
Publication Bias
Guarda-Nardini (30) Guarda-Nardini (30)
Kalamir (31) Kalamir (31) Gomes (33)
Trial
Serious 1 – >75% of the sample show conflicting results. Serious 2 –
