Randhawa 2016

REVIEW ARTICLE
The Effectiveness of Noninvasive Interventions for

Temporomandibular Disorders
A Systematic Review by the Ontario Protocol for Traffic Injury
Management (OPTIMa) Collaboration
Kristi Randhawa, BHSc, MPH,*wz Richard Bohay, DMD, MSc, MRCD(C),y
Pierre Côte´, DC, PhD,*8z Gabrielle van der Velde, DC, PhD,#**ww
Deborah Sutton, BScOT, Med, MSc,*w
Jessica J. Wong, BSc, DC, FCCS(C),*z Hainan Yu, MBBS, MSc,*w
Danielle Southerst, BScH, DC, FCCS(C),*z
Sharanya Varatharajan, BSc, MSc,*wz Silvano Mior, DC, PhD,wz
Maja Stupar, DC, PhD,*wz Heather M. Shearer, DC, MSc, FCCS(C),*
Craig Jacobs, BFA, DC, MSc, FCCS(C),zzyy and
Anne Taylor-Vaisey, MLS*
Results: Our search for effectiveness studies yielded 16,995 citations;

Objective: To determine the effectiveness and cost-effectiveness of 31 were relevant and 7 randomized controlled trials (published in 8
noninvasive interventions for temporomandibular disorders articles) had a low risk of bias. We found no relevant cost-effec-
(TMD). tiveness studies. The evidence suggests that for persistent TMD: (1)
cognitive-behavioral therapy and self-care management lead to sim-
Methods: We systematically searched MEDLINE, EMBASE, ilar improvements in pain and disability but cognitive-behavioral
CINAHL, PsycINFO, and Cochrane Central register from 1990 to therapy is more effective for activity interference and depressive
2014 for effectiveness studies and the Cochrane Health Technology symptoms; (2) cognitive-behavioral therapy combined with usual
Assessment Database, EconLit, NHS Economic Evaluation treatment provides short-term benefits in pain and ability to control
Database, and Tufts Medical Center Cost-Effectiveness Analysis pain compared with usual treatment alone; (3) intraoral myofascial
Register from 1990 to 2014 for cost-effectiveness studies. Random therapy may reduce pain and improve jaw opening; and (4) struc-
pairs of independent reviewers critically appraised eligible studies tured self-care management may be more effective than usual treat-
using the Scottish Intercollegiate Guidelines Network criteria. ment. The evidence suggests that occlusal devices may not be effective
Evidence from eligible studies was synthesized using best-evidence in reducing pain and improving motion for TMD of variable dura-
synthesis methodology. tion. Evidence on the effectiveness of biofeedback is inconclusive.
Discussion: The available evidence suggests that cognitive-behav-
ioral therapy, intraoral myofascial therapy, and self-care manage-
ment are therapeutic options for persistent TMD.
Received for publication April 24, 2014; revised May 13, 2015; accepted
April 7, 2015. Key Words: temporomandibular disorders, craniofacial disorders
From the *UOIT-CMCC Centre for the Study of Disability Prevention
and Rehabilitation, University of Ontario Institute of Technology conservative, noninvasive, nonsurgical, systematic review, cost-
(UOIT)-Canadian Memorial Chiropractic College (CMCC); effectiveness
wDivisions of Graduate Education and Research; zUndergraduate
Education; yyClinical Education, Canadian Memorial Chiropractic (Clin J Pain 2016;32:260–278)
College (CMCC); #Toronto Health Economics and Technology
Assessment (THETA) Collaborative; **Leslie Dan Faculty of
Pharmacy, University of Toronto; wwInstitute for Work and
Health; zzUOIT-CMCC Centre for the Study of Disability Pre-
vention and Rehabilitation, Toronto; 8Canada Research Chair in
Disability Prevention and Rehabilitation; zFaculty of Health Sci-
T emporomandibular disorders (TMD) include conditions
of the masticatory muscles, temporomandibular joint
(TMJ), and surrounding structures that present as pain,
ences, University of Ontario Institute of Technology (UOIT),
Oshawa; and ySchulich School of Medicine and Dentistry, Western abnormal joint sounds, limited jaw movement, and joint and
University, London, ON, Canada. muscle tenderness.1–3 The point prevalence of TMD in adults
This research was undertaken, in part, thanks to funding from the ranges from 38% in Finland4 to 44.4% in The Netherlands5
Canada Research Chairs program to P.C.
Supported by the Ontario Ministry of Finance and the Financial and 49.9% in Germany,6 and from 58% to 68% in Turkish
Services Commission of Ontario, Toronto, Canada (RFP No.: children and adolescents.7 TMD most commonly affect
OSS_00267175). The authors declare no conflict of interest. females and prevalence peaks around 50 years of age.8–11
Reprints: Kristi Randhawa, MPH, UOIT-CMCC Centre for the Study TMD are associated with worse oral health-related quality of
of Disability Prevention and Rehabilitation, 6100 Leslie Street,
Toronto, ON, Canada M2H 3J1 (e-mail: kristi.randhawa@uoit.ca). life.12 Individuals with TMD also report more depression and
Supplemental Digital Content is available for this article. Direct URL higher interference with daily activities.13 In the United Sates,
citations appear in the printed text and are provided in the HTML the cost of managing TMD during a 6- to 12-month period is
and PDF versions of this article on the journal’s Website, estimated to be $2 billion.14
www.clinicalpain.com.
Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved. Little is known about the etiology of TMD. It is
DOI: 10.1097/AJP.0000000000000247 hypothesized that TMD are commonly associated with
260 | www.clinicalpain.com Clin J Pain Volume 32, Number 3, March 2016

Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Clin J Pain Volume 32, Number 3, March 2016 Effectiveness of Noninvasive Interventions for TMD
whiplash injuries.15 One systematic review suggests that the more of the following signs or symptoms: pain, abnormal
incidence of TMD is higher in individuals with whiplash joint sounds, limited jaw movement, joint tenderness, and
injuries (range, 4% to 35%) compared with individuals muscle tenderness.3
without a history of such injuries (range, 4.7% to 7%).16 In
Saskatchewan, Canada, the incidence of self-reported Interventions
reduced or painful jaw movement was 15.8% in individuals We included any noninvasive intervention. These
who reported whiplash-associated disorders (WAD) versus include passive physical modalities, exercise, occlusal devi-
4.7% in those without WAD.9 In the Saskatchewan cohort, ces, manual therapy, soft-tissue therapy, acupuncture,
reduced or painful jaw movement was associated with psychological interventions, and education. We excluded
female sex, age less than 50 years, having hit one’s head in pharmacological interventions and invasive interventions
the collision, and postinjury symptoms of difficulty swal- (eg, injections, surgical procedures).
lowing, ringing in the ears, dizziness or unsteadiness, and
more intense neck pain.9 The management of TMD com- Comparison Groups
monly involves noninvasive interventions such as acu- We included studies that compared noninvasive
puncture, physical therapy, passive modalities, orthodontic interventions to other interventions, placebo/sham inter-
therapy, and occlusal appliances.1 ventions, or no intervention.
We identified 9 systematic reviews that evaluated the
effectiveness of noninvasive interventions for TMD.17–25 Outcomes
However, the effectiveness and cost-effectiveness of most Eligible studies included at least one of the following
interventions remains unclear19,22,24; this may be partially outcomes: (1) pain intensity; (2) functional recovery (eg,
attributed to the methodology used in these systematic disability, return to work); (3) self-rated recovery: (4)
reviews. Most reviews included results from studies with health-related quality of life; (5) psychological outcomes
significant biases (eg, no randomization, no intention-to- such as depression; or (6) adverse effects. Eligible economic
treat analysis, no allocation concealment, differences in evaluations reported health outcomes expressed as quality-
baseline characteristics)17–19,21–25 and small sample sizes.17–25 adjusted life-years, the health economic outcome measure.
Moreover, several systematic reviews only considered stat-
istical significance, and did not assess the clinical significance Study Characteristics
of results.18,20,23,25 Finally, we did not identify any system- Eligible studies met the following criteria: (1) English
atic reviews on the cost-effectiveness of noninvasive inter- language; (2) published in a peer-reviewed journal; (3) study
ventions for TMD. designs including: randomized controlled trials (RCTs),
The objective of our review was to investigate the cohort studies, and case-control studies; and (4) an incep-
effectiveness and cost-effectiveness of noninvasive inter- tion cohort of 30 individuals per treatment arm for RCTs
ventions compared with other interventions, placebo/sham or 100 patients per exposed group for cohort studies.
interventions, or no intervention in improving self-rated Studies were excluded if the following criteria were met: (1)
recovery, functional recovery, or clinical outcomes in adults letters, editorials, commentaries, unpublished manuscripts,
and children with TMD. dissertations, government reports, books and book chap-
ters, conference proceedings, meeting abstracts, lectures
METHODS and addresses, consensus development statements, guide-
line statements; (2) study designs including: clinical practice
Registration guidelines, cross-sectional studies, case reports, case series,
This review protocol was registered with the Interna- guidelines, qualitative studies, nonsystematic and system-
tional Prospective Register of Systematic Reviews (PROS- atic reviews (with or without meta-analyses), biomechanical
PERO) on October 7, 2013 (CRD42013005910) for studies, laboratory studies, studies not reporting on
effectiveness and March 14, 2014 (CRD42014009041) for methodology; or (3) cadaveric or animal studies.
cost-effectiveness. Eligible cost-effectiveness studies were full economic
evaluations of any noninvasive intervention for treating
Eligibility Criteria TMD in children and adults. Full economic evaluations
were defined as comparisons that jointly analyzed costs
Population (resource use) and consequences (health outcomes), and
Our review included studies of adults and/or children expressed cost-effectiveness with the incremental cost-
diagnosed with TMD. We excluded studies of patients with effectiveness ratio or incremental net-benefit statistic.
severe injuries, which include but are not limited to grade
III sprains/strains, dislocations, fractures, ruptures, and Information Sources
osteonecrosis. We developed search strategies in consultation with a
We defined TMD as a group of conditions that affect health sciences librarian (Appendix 1 and 2, Supplemental
the masticatory muscles, the TMJ, and surrounding struc- Digital Content 1, http://links.lww.com/CJP/A178). The
tures, and include sprain and strain injuries to this ana- search strategies were reviewed by a second librarian to
tomic region.3 Sprains may involve supporting ligaments of ensure completeness and accuracy using the Peer Review of
the TMJ, including the collateral ligament, capsular liga- Electronic Search Strategies checklist.26 We systematically
ment, and temporomandibular ligament. Sprains may also searched the following electronic databases for effectiveness
involve ligaments adjacent to the TMJ, including the evidence from January 1, 1990 to November 4, 2014:
sphenomandibular ligament and stylomandibular ligament. MEDLINE, EMBASE, CINAHL, PsychINFO, and
Strains may involve supporting muscles of the TMJ, Cochrane Central Register of Controlled Trials. We sys-
including the masseter, medial and lateral pterygoid, tem- tematically searched for cost-effectiveness evidence in the
poralis, and digastrics. TMD presents clinically with one or following databases from January 1, 1990 to January 7,
Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved. www.clinicalpain.com | 261
Randhawa et al Clin J Pain Volume 32, Number 3, March 2016
2014: CINAHL, Cochrane Health Technology Assessment Statistical Analyses

Database, EconLit, EMBASE, Medline, National Health We computed interrater agreement for the screening of
Services Economic Evaluation Database, PsychINFO, and articles and reported the kappa coefficient (k) with 95%
Tufts Medical Center Cost-effectiveness Analysis Register. confidence interval (CI).44 We also calculated the percentage
The subject headings used in MEDLINE (eg, MeSH, agreement for the critical appraisal of articles for admis-
or Medical Subject Headings) were adapted to conform to sible/inadmissible results. For effectiveness evidence, we
other bibliographic databases. In addition, we included calculated the difference in mean change between groups
free-text words relevant to noninvasive interventions, and and the 95% CI to quantify the effectiveness of inter-
TMD. ventions. The computation of the 95% CI for the difference
in mean change was based on the assumption that pre-
Study Selection intervention and postintervention outcomes were highly
We used a 2-phase screening process to identify eligible correlated (r = 0.8).45,46 A difference in mean change was
studies. In phase 1, random pairs of independent reviewers classified as clinically significant if change was greater than
screened citation titles and abstracts to determine the eli- or equal to minimal clinically important differences
gibility of studies. Phase 1 screening resulted in studies (MCIDs): (1) 10/100 mm on the Visual Analog Scale47; (2)
being classified as relevant, possibly relevant, or irrelevant. 2/10 points on the Numeric Rating Scale48; (3) 2/10 points
In phase 2, the same paired reviewers independently on the Graded Chronic Pain Scale49; and (4) 5/63 on the
reviewed the manuscripts of possibly relevant studies to Beck Depression Inventory.50 The MCIDs for some out-
make a final determination of eligibility. Reviewers met to come measures used in the admissible studies have not been
resolve disagreements and reach consensus on the eligibility reported in the published literature (Mandibular Function
of studies. We used an independent third reviewer if con- Impairment Questionnaire, Survey of Pain Attitudes).
sensus could not be reached.
Reporting
The systematic review was structured and reported
Assessment of Risk of Bias according to the Preferred Reporting Items for Systematic
Random pairs of independent reviewers critically Reviews and Meta-Analyses (PRISMA) statement.51 The
appraised the internal validity of eligible studies using the aim of the PRISMA statement is to improve and stand-
Scottish Intercollegiate Guidelines Network (SIGN) ardize the reporting of systematic reviews and meta-
Methodology Checklist for randomized controlled trials, analyses.51
cohort, case-control studies, and economic evaluations.27
The SIGN criteria were used to qualitatively evaluate the
presence and impact of selection bias, information bias, and RESULTS
confounding on the results of a study. We did not use a Study Selection
quantitative score or a cutoff point to determine the inter-
Our search of studies on effectiveness retrieved 16,995
nal validity of studies.28 Rather, the SIGN criteria were
citations. We removed 2046 duplicates and screened 14,949
used to assist reviewers in making an informed overall
citations (Fig. 1). Of those, 30 RCTs (published in 31
judgment on the internal validity of studies. This method-
articles) were eligible for critical appraisal and 7 RCTs
ology has been previously described.29–34
(published in 8 articles) had a low risk of bias. The inter-
We critically appraised the following methodological
rater agreement for the screening of articles was k = 0.874
aspects of studies: (1) clarity of the research question;
(95% CI, 0.814-0.933). The percentage agreement for the
(2) randomization method; (3) concealment of treatment
critical appraisal of articles was 83.3% (25/30 RCTs) based
allocation; (4) blinding of treatment and outcomes; (5)
on admissible/inadmissible results. For the 6 studies where
similarity of baseline characteristics between/among treat-
reviewers disagreed, consensus was reached through
ment arms; (6) cointervention contamination; (7) validity
discussion.
and reliability of outcome measures; (8) follow-up rates;
Our search of studies on cost-effectiveness retrieved
(9) analysis according to intention-to-treat principles; and
4764 nonduplicate citations (Fig. 2). Of these, none were
(10) comparability of results across study sites (where
eligible economic evaluations. The interrater agreement for
applicable).
screening of articles for economic evaluations was 100%.
Reviewers reached consensus through discussion. An
independent third reviewer was used to resolve disagree- Study Characteristics
ments if consensus could not be reached. We contacted All RCTs with a low risk of bias included adults. Four
authors when additional information was needed to com- studies addressed persistent TMD,35,36,38,41–42 and 3
plete the critical appraisal. Studies with adequate internal addressed TMD of variable duration.37,39,40 The following
validity had a low risk of bias and were included in our interventions were investigated in the 7 RCTs: cognitive-
evidence synthesis (Table 1).43 behavioral therapy (CBT), intraoral myofascial therapy,
education, self-care management, occlusal device with
Data Extraction and Synthesis of Results exercises, intraoral appliance, biofeedback with stress
The lead author extracted data from studies with a low management, and biofeedback.
risk of bias (Table 2). A second reviewer independently
checked the extracted data. The evidence was synthesized Risk of Bias Within Studies
according to the principles of best-evidence synthesis.43 All RCTs with a low risk of bias clearly stated their
Finally, a senior epidemiologist (P.C.) reviewed the accu- research questions, reported adequate blinding methods,
racy of the extracted data at the manuscript preparation and had adequate follow-up rates (Table 1). Nevertheless,
stage by cross-checking the extracted data with the original these studies had minor limitations. Specifically, the
studies. method of treatment allocation concealment was not
262 | www.clinicalpain.com Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
TABLE 1. Summary of Assessment of Risk of Bias for Accepted Studies Based on SIGN Criteria
Results
Similarity Similarity Outcome Comparable
Research Randomiz- Conceal- Blind- at Between Measure- Drop- Intention- Between
References Question ation ment ing Baseline Arms ment Out (%)* to-Treat Sites
Dworkin et al35 Y CS N Y Y CS Y Posttreatment: Y CS
CC: 12%
UT:16%
12 mo:
CC: 5%
UT: 12%
Dworkin et al36 Y CS N CS N N Y SC: 21.3% Y NA
UT: 19%
Gatchel et al37 Y CS N N N N Y Overall: 3/ Y NA
101 = 2.97%
Drop- out rate
for each group
unknown
Kalamir et al38 Y Y Y Y CS CS Y 6 wk: Y NA
IMT: none
IMTESC: 0%
Control: 0%
6 mo:
IMT: 0%
IMTESC: 0%
Control: 0%
1 y:
Control: 3%
IMT: 0%
IMTESC: 0%
Niemela et al39 Y Y N Y Y CS Y 1 mo: N NA
Splint: 0%
Control: 10%
Turk et al40 Y CS N Y N CS Y 7 wk: N NA
IA: 6.67%; BF/
SM: 0%
6 mo:
IA: 10.7% BF/
SM: 13.3%
Turner Y Y Y Y Y N Y Postintervention: Y NA
and The mean
colleagues41,42 percent of
electronic
interviews
completed was
90% in both
the CBT and
SCM groups
(for total
sample,
SD = 9%)
8 wk:
CBT: 11%
SCM: 9%
6 mo:
CBT: 14%
SCM: 9%
12 mo:
CBT: 14%
SCM: 11%
*Percent drop-out includes drop-outs and loss to follow-up.
BF indicates biofeedback; CBT, cognitive-behavioral therapy; CC, comprehensive care; CS, can’t say; IA, intraoral appliance; IMT, intraoral myofascial
therapy; IMTESC, intraoral myofascial therapy, education, and stretching; N, no; NA, not applicable; SC, self-care; SCM, self-care management; SM, stress
management; UT, usual treatment; Y, yes.
described in 5 studies35–37,39–40 and randomization was not usual treatment included multimodal care prescribed by the
adequately described in 4 RCTs.35–37,40 Many studies (4/7) attending dentists that included a combination of: (1) phys-
did not describe cointerventions35,38–40 or reported differ- iotherapy; (2) patient education; (3) medication; and (4)
ences in cointervention between groups (3/7).36,37,41–42 Two intraoral flat-plane occlusal appliances. The authors reported
studies did not report an intention-to-treat analysis.39,40 statistically significant differences between groups immedi-
The 23 studies with a high risk of bias had important ately postintervention for pain intensity, ability to control
methodological limitations. The most common limitations pain, and satisfaction and helpfulness of treatment favouring
related to: (1) randomization method (16/23); (2) conceal- the CBT group. The clinical importance of these outcomes is
ment of treatment allocation (18/23); (3) differences in unknown. There were no differences between groups at the
baseline characteristics (19/23); and (4) report of coin- 6- and 12-month follow-ups.
terventions (19/23). Furthermore, several studies with a
high risk of bias had >20% attrition (7/23) or did not Self-Care Management
conduct or report having performed an intention-to-treat Evidence from 1 RCT suggests that structured self-
analysis (18/23). We contacted the authors of 7 studies to care management may be more beneficial than usual
obtain clarification about their methodology37,39,40,42,52–54; treatment for the management of persistent TMD.36
6 authors responded to our queries.35,39–41,53–54 Dworkin and colleagues randomized participants to: (1)
structured self-care management; or (2) usual treatment
Summary of Evidence (Table 2).48 Structured self-care management was provided
We did not find studies with low risk of bias that by a dental hygienist and involved 3 sessions over 2 and a
informed the management of recent onset (< 3 mo) of half months. The structured self-care management empha-
TMD. sized education, guided reading with structured feedback,
relaxation and stress management training, self-monitoring
Persistent TMD (Z3 mo)
of signs and symptoms, development of a “Personal TMD
Cognitive-Behavioral Therapy Self-care Plan,” supervised practice and reinforcement of
Evidence from 2 RCTs suggests that CBT may benefit dentist-prescribed self-care treatments, and maintenance
patients with persistent TMD.35,41,42 In her trial, Turner and relapse prevention. The usual care included multimodal
and colleagues randomized participants to 4 sessions (over care prescribed by a dentist and included: (1) physi-
8 wk) of CBT by a clinical psychologist or self-management otherapy; (2) education; (3) medication; and (4) intraoral
by a patient educator41,42 (Table 2). CBT included pro- flat-plane occlusal appliances. There were statistically sig-
gressive relaxation and abdominal/diaphragmatic breathing nificant between-group differences at 12 months for pain
techniques, a relaxation audiotape, discussion regarding intensity, activity interference, depression and somatisation,
fear-avoidance, the identification and challenging of neg- helpfulness of treatment, coping with pain, level of TMD
ative thoughts in response to pain, relapse prevention, ways knowledge, and number of sites painful to palpation
to maintain gains, and methods to deal with setbacks. favouring structured self-care management.
Participants randomized to CBT also received a weekly
personal TMD health care plan. The self-care management Myofascial Therapy
group received information about TMD and a manual on Evidence from 1 RCT suggests that intraoral myo-
general health care. Both groups received usual treatment fascial therapy may be more effective than wait listing in
from a dentist (jaw posture monitoring and correction; reducing pain and improving jaw opening in patients with
advice heat/cold to facial areas; diet modification; medi- persistent TMD.38 However, the addition of structured
cation; jaw-stretching exercises, and occlusal device if education and TMD stretching to intraoral myofascial
indicated). Compared with those who received the self-care therapy does not provide additional benefits. In their trial,
management intervention, participants randomized to CBT Kalamir et al38 randomized participants to: (1) intraoral
were 4 times more likely to report no interference with their myofascial therapy; (2) intraoral myofascial therapy with
activities at the 1-year follow-up (odds ratio [OR] = 4.2; structured education and TMD stretching; or (3) wait list
95% CI, 1.7-10.2). Similarly, those in the CBT group were (Table 2). The intraoral myofascial therapy included: (1)
more likely to report low to moderate depressive symptoms intraoral temporalis release; (2) intraoral medial and lateral
(OR = 3.8; 95% CI, 1.2-12.0). The differences in mean pterygoid (origin) technique; and (3) intraoral sphenopa-
change between groups for pain intensity (Graded Chronic latine ganglion technique twice a week for 5 weeks. Par-
Pain Scale) were statistically significant but were not clin- ticipants who received intraoral myofascial therapy with
ically important. Finally, the CBT group demonstrated structured education and TMD stretching were instructed
statistically significant improvement in masticatory and on chewing technique and relaxation stretching. They also
nonmasticatory disability, but the clinical importance of attended short lectures on basic TMJ anatomy, bio-
this result is not known. mechanics, disk displacement and dysfunction, and the role
In the second RCT, Dworkin and colleagues random- of psychoemotional factors in TMD. At the 6-month fol-
ized participants to: (1) CBT combined with usual treatment; low-up, participants in the intraoral myofascial therapy
or (2) usual treatment alone (Table 2). CBT was delivered by group reported greater improvement in jaw pain at rest
clinical psychologists and involved 6 visits and 3 phone calls (mean difference: 2.59/10 [95% CI, 0.69-4.49]); jaw pain
over 6 weeks. The CBT emphasized patient education and upon maximal active opening (mean difference 2.53/10
self-care, the identification and modification of maladaptive [95% CI, 2.23-2.84]); and jaw pain upon clenching (mean
thought patterns related to pain, training in relaxation and difference: 3.66/10 [95% CI, 3.33-3.99]). Moreover, the
other pain management coping skills, and discussion of intraoral myofascial therapy group had greater improve-
relapse prevention and long-term maintenance of gains ment in jaw opening (Table 2). Similarly, participants who
achieved in treatment. Furthermore, modules were included received the myofascial therapy with structured education
to address somatization and depression when present. The and TMD stretching intervention had statistically and
TABLE 2. Evidence Table for Accepted Randomized Controlled Trials Assessing the Effectiveness of Noninvasive Interventions for TMD
Patients and
Setting;
Number (n) Interventions; Comparisons;
References Enrolled No. Patients No. Patients Follow-up Outcomes Key Findings
Dworkin et al35 Patients (18- Comprehensive UT by attending Postintervention Characteristic Difference in
70 y) referred care (CC): dentist: (1) and 6, 12 mo pain intensity mean change
to the CBT by physiotherapy (RDC/TMD from baseline
Orofacial Pain clinical (passive and axis II (CC UT):
Clinics in the psychologist (6 active jaw measures 0- Characteristic
Department of visits + 3 range-of- 10); pain- pain intensity
Oral Medicine, phone calls/ motion and related activity (0-10)*w
University of 16 wk) and stretching interference Postintervention:
Washington UT. CBT exercises and score (RDC/ 1.4
School of included application of TMD axis II 12 mo: 0.4
Dentistry. education, heat or cold measures 0- Statistically
Case definition: identification, packs); (2) 10); depression significant
(1) self-report and patient and differences
of facial ache modification education somatization postinterven-
or pain in the of maladaptive (parafunc- scales (SCL- tion favouring
muscles of thought tional oral 90); ability to CC (P = 0.02).
mastication, patterns behaviors, control pain Pain-related
the TMJ, or related to diet, nature of (0-6); process activity
the region in pain, the condition, of care ratings interference
front of the relaxation, and rationale (0-10); vertical (0-10)*w
ear or inside and other pain for treatment); range of jaw Postintervention:
the ear; (2) coping skills, (3) motion (RDC/ 1.2
interference of and discussion medications TMD axis I No statistically
daily activities of relapse (analgesics, measures); significant
(RDC/TMD prevention NSAIDs, number of differences.
axis II GCP and long-term muscle sites of painful Depression and
score of II maintenance relaxants, palpation somatisation
high, III, or of gains antidepres- (RDC/TMD (SCL-90)
IV). (n = 117) achieved in sants); and (4) axis I No statistically
treatment. intraoral flat- measures); significant
Somatization plan occlusal axis I differences.
and depression appliances. No diagnoses. Ability to
were limitations on Adverse effects. control pain
addressed number of (0-6)*w
when present visits (n = 58). Postintervention:
(n = 59). 1.0
Statistically
significant
between-group
differences
postinterven-
tion favouring
CC
(P < 0.001).
Process of care
ratings
(satisfaction
and
helpfulness of
treatment; 0-
10)
Statistically
significant
between-group
differences
postinterven-
tion for
helpfulness of
treatment
favouring CC
(P = 0.03).
Vertical range of
jaw motion
(Continued )
TABLE 2. (continued)
Patients and
Setting;
(mm)
No statistically
significant
differences.
Number of sites
painful to
palpation
Extraoral
muscles: 0.3
(95% CI,
1.5 to 0.9)
Intraoral
muscles: 0.2
(95% CI,
0.1 to 0.5)
TMJ: 0.4
(95% CI,
0.7 to 0.1)
Axis I diagnosis
No statistically
significant
differences.
Adverse effects
None.
Dworkin et al36 Patients (18- SCM by dental UT (no Postintervention Characteristic Difference in
70 y) referred hygienists (3 limitation on and 6, 12 mo pain intensity mean change
to the sessions/ number of (RDC/TMD from baseline
Orofacial Pain 2.5 mo): visits/2.5 mo): axis II (SCMUT):
Clinics in the education and noninvasive measures 0- Characteristic
Department of self-care for treatments: (1) 10); pain- pain intensity
Oral Medicine, TMD, physiotherapy related activity (0-10)
University of incorporating (passive and interference Statistically
Washington CBT methods. active jaw score (RDC/ significant
School of SC range-of- TMD axis II between-group
Dentistry. emphasized motion and measures 0- differences
Case definition: education, stretching 10); depression reported at
(1) self-report guided reading exercises and and 12 mo
of facial ache with application of somatization favouring
or pain in the structured heat or cold scales (SCL- SCM
muscles of feedback, packs); (2) 90); process of (P = 0.036).
mastication, relaxation and patient care ratings (0- Pain-related
the TMJ, or SM training, education; (3) 10); vertical activity
the region in self- medications range of jaw interference
front of the monitoring of (analgesics, motion (RDC/ (0-10)
ear or inside signs and muscle TMD axis I Statistically
the ear; (2) symptoms, relaxants, measures); significant
interference of development antidepres- number of between-group
daily activities of a “personal sants); and (4) sites of painful differences
(RDC/TMD TMD self-care intraoral flat- palpation reported at
axis II GCP plan,” plan occlusal (RDC/TMD 12 mo
score of 0, I, supervised appliances axis I favouring
or II-low practice, and (n = 63). measures). SCM
(n = 124). reinforcement Adverse events. (P = 0.01).
of dentist- Depression and
prescribed SC somatisation
treatments, (SCL-90)
maintenance, Statistically
and relapse significant
prevention between-group
(n = 61). differences for
somatization
at 12 mo
favouring
(Continued )
Patients and
Setting;
SCM
(P = 0.002).
Process of care
ratings
(satisfaction
and
helpfulness of
treatment; 0-
10)
Statistically
significant
between-group
differences at
12 mo for
helpfulness of
treatment
(P = 0.0002);
helpfulness in
overall coping
with TMD
pain
(P < 0.0001);
level of
knowledge
about TMD
(P < 0.0001);
and treatment
satisfaction
(P = 0.280)
favouring SC.
Vertical range of
jaw motion
(mm)
No statistically
significant
differences.
Number of sites
painful to
palpation*w
Extraoral
muscles: 1.0
Statistically
significant
differences
between
groups at
12 mo
favouring SC
(P = 0.007)
Adverse events
None.
Gatchel at al37 Adults (18-70 y) EI: CBT and Nonintervention 1y Primary: Difference in
recruited from biofeedback group (NI) characteristic mean change
referrals to a (BF) provided (n = 45). pain intensity from baseline
university by clinical (CPI). (EI–NI):
TMD clinical psychology Secondary: ways CPI: 12.39 (95%
research research of coping CI, 8.08-16.70)
program, personnel (WOC): blame Ways of Coping
fliers, and (6 1 h self, wishful Blame Self: 2.13
newspaper sessions). thinking, (95% CI, 0.95-
advertise- CBT: avoidance. 3.31)
ments, in education Depressive Wishful
Texas. mind-body symptom thinking: 0.16
(Continued )
Patients and
Setting;
Case definition: relationship (BDI), (95% CI,
acute jaw or emphasizing affective 0.82 to 1.14)
facial pain stress, disorder, Avoidance: 0.82
<6 mo; relaxation anxiety (95% CI, 0.03-
myofascial training, disorder, 1.61)
pain based on distraction somatoform Problem focused:
axis I-group and pleasant disorder, 1.19 (95%
1a of RDC activity substance CI, 2.27 to
form, and scheduling, abuse, pain 0.11)
response to cognitive disorder. Seeks social
RDC history restructuring, support:
question: self- 1.83 (95%
“Have you instructional CI, 3.13 to
had pain in the training. 0.53)
face, jaw, BF: EMG and BDI (Mann-
temple, in temperature Whitney):
front of the biofeedback P = 0.03
ear, or in the units. Odds ratio (95%
ear in the last Work books CI)( NI vs.
month” with reading EI):
(n = 101). assignments Affective
and disorder: 2.70
homework to (0.91-7.97)
complete Anxiety
between disorder: 7.18
sessions (1.90-27.13)
(n = 56). Somatoform
disorder: 12.50
(4.44-35.22)
Pain disorder:
9.00 (3.46-
23.44)
Kalamir et al38 Patients (18- IMT (2 times/wk Wait list (WL) Postinterven- Primary: jaw Difference in
50 y) recruited for 5 wk): (1) (n = 31). tion, 6 and pain at rest mean change
from referrals intraoral 12 mo (11-point from
from dental temporalis scale), jaw baseline(IMT-
clinics in release; (2) pain upon WL)*:
Australia in intraoral maximal Pain at rest
2006-2007. medial and active opening Postintervention:
Case definition: lateral (11-point 2.65 (95% CI,
myogenous pterygoid scale), jaw 0.74-4.56)
TMD: daily technique; and pain upon 6 mo: 2.59 (95%
periauricular (3) intraoral clenching (11- CI, 0.69-4.49)
pain with or sphenopala- point scale). 1 y: 1.66 (95%
without joint tine ganglion Secondary: CI, 0.03 to
sounds technique opening range 3.35)
(Z3 mo); jaw (n = 31). of motion Pain upon
pain Z3/10 IMTES (12 (vernier maximal
for pain at visits/6 wk): callipers; mm). opening
rest, on education: Adverse events. Postintervention:
maximal short lectures 2.23 (95% CI,
opening, and on basic TMJ 1.93-2.54)
pain on anatomy, 6 mo: 2.53 (95%
clenching biomechanics, CI, 2.23-2.84)
(n = 93). disk 1 y: 1.47 (95%
displacement CI, 1.17-1.78)
and Pain upon
dysfunction, clenching
the role of Postintervention:
psychoemo- 3.55 (95% CI,
tional factors 3.21-3.88)
in TMD. 6 mo: 3.66 (95%
TMD stretching: CI, 3.33-3.99)
(Continued )
Patients and
Setting;
mandibular 1 y: 2.86 (95%
home CI, 2.53-3.19)
stretching Opening range of
including motion:
mandibular Postintervention:
body-condylar 5.16 (95% CI,
cross-pressure 3.49-6.83)
chewing 6 mo: 4.64 (95%
technique and CI, 2.97-6.32)
postisometric 1 y: 4.93 (95%
relaxation CI, 3.23-6.63)
stretches Difference in
(twice a day) mean change
(n = 31). from baseline
(IMTES W-
L)*:
Jaw pain at rest
Postintervention:
2.0 (95% CI,
0.03 to 4.03)
6 mo: 2.30 (95%
CI, 0.18-4.41)
1 y: 3.54 (95%
CI, 1.07-6.02)
Pain upon
maximal
opening
Postintervention:
2.15 (95% CI,
1.85-2.46)
6 mo: 2.31 (95%
CI, 2.01-2.61)
1 y: 3.75 (95%
CI, 3.45-4.05)
Pain upon
clenching
Postintervention:
2.42 (95% CI,
2.09-2.75)
6 mo: 2.84 (95%
CI, 2.51-3.16)
1 y: 3.73 (95%
CI, 3.40-4.05)
Opening range of
motion
Postintervention:
4.52 (95% CI,
2.72-6.31)
6 mo: 6.52 (95%
CI, 4.72-8.31)
1 y: 8.61 (95%
CI, 6.78-10.44)
Difference in
mean change
from baseline
(IMTES-
IMT)*:
Pain at rest
Postintervention:
0.65 (95%
CI, 3.16 to
1.86)
(Continued )
Patients and
Setting;
6 mo: 0.30
(95% CI,
2.87 to 2.27)
1 y: 1.88 (95%
CI, 0.85 to
4.61)
Jaw pain upon
maximal
opening
Post: 0.65
(95% CI,
0.95 to
0.35)
6 mo: 0.22
(95% CI,
0.52 to 0.08)
1 y: 2.27 (95%
CI, 1.97 to
2.57)
Jaw pain upon
clenching
Postintervention:
1.12 (95%
CI, 1.45 to
0.80)
6 mo: 0.83
(95% CI,
1.15 to
0.50)
1 y: 0.86 (95%
CI, 0.54 to
1.19)
Opening range of
motion:
Post: 0.65
(95% CI,
2.26 to 0.97)
6 mo: 1.87 (95%
CI, 0.27-3.48)
1 y: 3.68 (95%
CI, 2.08-5.28)
No adverse
events
reported.
Niemela et al39 Patients (Z20 y) Occlusal device Structured 1 mo Pain intensity Difference in
referred to the (heat-cured masticatory (VAS 0-10), mean change
Oral and acrylic) and exercises active from baseline
Maxillofacial structured (same as maximal (Occlusal
Department, masticatory intervention opening (mm), deviceexer-
Oulu exercises group) and laterotrusion cise)
University (active mouth jaw stretching movement Pain intensity:
Hospital, openings, (n = 41). (left and right) 1 mo: 0.99
Finland in laterotrusive (mm), (95% CI,
2008-2009. movements, protrusion 2.21 to 0.22)
Case definition: and protrusive movement No statistically
diagnosis of movements) (mm), number significant
TMD and jaw of painful differences in
according to stretching muscle sites, laterotrustion
Research (n = 39). TMJ pain on movement,
Diagnostic palpation (% protrusion,
Criteria RDC/ of patients). active
TMD Adverse effects. maximal
(n = 80). opening, and
(Continued )
Patients and
Setting;
TMJ pain on
palpation.
Adverse events:
Pain on TMJ
palpation
increased
significantly in
the occlusal
device group,
which may
indicate a
negative effect
of the occlusal
device on
distinct TMJ
findings.
Turk et al40 Patients (18- BF/SM: 1 h/wk/ Wait list Postintervention Pain (PSS 0-10); Difference in
55 y) referred 6 wk by (n = 20). and 6 mo muscle mean change
to an psychologist palpation pain from baseline
outpatient BF: (1) index (PPI); (BF/SMIA):
TMD clinic at electrodes over depression Pain (PSS 0-10)
the University masseter (CES-D); Posttreatment:
of Pittsburgh. muscles; (2) depression 0.9 (95%
Case definition: computer- (POMS); CI, 1.3 to
pain and controlled credibility of 0.5)
tenderness of auditory tone treatment (0- 6 mo: 0.2 (95%
the muscles of and pulsating 10) CI, 0.3 to
mastication feedback; (3) 0.7)
and TMJ 5 min of no- Muscle
region and feedback palpation pain
limited resting (PPI):
mandibular baseline Posttreatment:
movements period with 1.5 (95%
Z2 mo jaw relaxed, CI, 2.2 to
(n = 80). 20 min of BF, 0.8)
and 5 min of 6 mo: 0.5 (95%
no-feedback. CI, 0.1 to
SM: (1) 1.1)
education on Depression
stress, (CES-D)
increased Posttreatment:
muscle 2.7 (95%
tension, and CI, 5.9 to
pain; (2) 0.5)
coping skills 6 mo: 2.7 (95%
training; (3) CI, 0.4 to
homework to 5.8)
practice Depression
relaxations (POMS):
skills and Posttreatment:
techniques. 2.3 (95%
(n = 30) CI, 5.4 to
IA: full-arch, flat 0.8)
heat-cured 6 mo: 4.4 (95%
acrylic resin CI, 1.3-7.5)
splint, worn at Credibility of
all times for treatment (0-
first 6 wk of 10)
treatment Posttreatment:
except during 0.5 (95% CI,
eating and oral 0.1-0.9)
hygiene.
Weekly
(Continued )
Patients and
Setting;
meeting with
dentist for
instruction on
oral habits,
review and
adjustments of
IA use,
inspection for
irritation
(n = 30).
Turner and Patients (Z18 y) CBT by SCM Immediately Primary: pain Absence of
colleagues41,42 recruited from psychologists (education/ postinterven- intensity activity
the University (4 times/8 wk) attention tion, 3, 6, (GCPS); interference
of Washington and a manual control 12 mo activity (CBT vs.
Orofacial Pain (education condition) interference SCM):
Clinic between about TMD; provided by (GCPS). 12 mo: OR 4.2
June 2001 and psychological patient Jaw disability: (95% CI, 1.7-
February aspects of educators (4 masticatory 10.2)
2004. pain, times/8 wk). and Difference in
Case definition: relaxation, Manual: nonmastica- mean change
axis I TMD behavioural general health tory disability from baseline
diagnosis techniques for care (MFIQ). (CBT–SCM):
(RDC/TMD); pain information Depressive Pain intensity
facial pain management, (pain symptom 3 mo: 0.00 (95%
>3 mo; facial coping). medications; (BDI). CI, 0.38 to
pain-related Weekly personal communicat- Secondary: pain- 0.38)
disability rated TMD health ing with health related jaw use 6 mo: 0.80 (95%
as II high, III, care plan. care providers; limitations CI, 0.36-1.24)
or IV on the Progressive making (NRS 0-10); 12 mo: 0.80 (95%
GCPS relaxation and treatment pain beliefs CI, 0.331.27)
(n = 158). abdominal/ decisions); (SOPA); self- MFIQ
diaphragmatic information efficacy (TMD (nonmastica-
breathing on TMD, SES); pain tory):
techniques. TMD castrophizing 3 mo: 0.05 (95%
Relaxation treatments, (CSQ); CI, 0.01-0.09)
audiotape for TMD self- rumination 6 mo: 0.06 (95%
home practice. care, sessions (PCS); rest, CI, 0.02-0.10)
Discussion of reviewed main task 12 mo: 0.08 (95%
fear- points of persistence, CI, 0.05-0.12)
avoidance, manual coping self- MFIQ
identify and articles statements, (masticatory):
challenge (n = 79). relaxation 3 mo: 0.10 (95%
negative (CPCI). CI, 0.05-0.15)
thoughts in 6 mo: 0.10 (95%
response to CI, 0.05-0.15)
pain; relapse 12 mo: 0.14 (95%
prevention, CI, 0.09-0.19)
ways to Depression:
maintain gains 3 mo: 2.2 (95%
and deal with CI, 0.30-4.10)
setbacks 6 mo: 1.9 (95%
(n = 79). CI, 0.10-3.70)
Both groups 12 mo: 3.1 (95%
received usual CI, 1.26-4.94)
treatment: jaw Jaw use
posture limitation:
monitoring Postintervention:
and 0.06w
correction; Pain beliefs:
advice heat/ Disability:
cold to facial 3 mo: 0.40 (95%
areas; diet CI, 0.23-0.57)
modification; 6 mo: 0.60 (95%
(Continued )
Patients and
Setting;
medication; CI, 0.44-0.76)
jaw-stretching 12 mo: 0.40 (95%
exercises; CI, 0.23-0.57)
occlusal device Harm:
if indicated. 3 mo: 0.50 (95%
CI, 0.37-0.63)
6 mo: 0.50 (95%
CI, 0.37-0.63)
12 mo: 0.50 (95%
CI, 0.26-0.54)
Control:
3 mo: 0.60
(95% CI,
0.75 to 0.45)
6 mo: 0.70
(95% CI,
0.85 to
0.55)
12 mo: 0.30 (95%
CI, 0.14-0.46)
Self-efficacy:
3 mo: 1.30
(95% CI,
1.70 to
0.90)
6 mo: 1.30
(95% CI,
1.73 to
0.87)
12 mo: 1.50
(95% CI,
1.93 to
1.07)
Catastrophizing:
3 mo: 2.00 (95%
CI, 1.20-2.80)
6 mo: 1.20 (95%
CI, 0.40-2.00)
12 mo: 1.40 (95%
CI, 0.59-2.21)
Coping:
3 mo: 0.30
(95% CI,
0.65 to 0.50)
6 mo: 0.00 (95%
CI, 0.35 to
0.35)
12 mo: 0.10 (95%
CI, 0.26 to
0.46)
Relaxation:
3 mo: 1.30
(95% CI,
1.59 to
1.01)
6 mo: 0.90
(95% CI,
1.20 to
0.60)
12 mo: 0.80
(95% CI,
1.09 to
0.51)
(Continued )
Patients and
Setting;
Rest:
3 mo: 0.10
(95% CI,
0.45 to 0.25)
6 mo: 0.60
(95% CI,
0.96 to
0.24)
12 mo: 0.10
(95% CI,
0.46 to 0.26)
Depression:
BDI score <21
at 12 mo (CBT
vs. SCM):
OR = 3.8
(95% CI, 1.2-
12.0)
Proportions of
groups with
>50%
improvement
immediately
postinterven-
tion:
Significantly
more CBT
than
SCM
participants
reported
>50%
improvement
in daily
activity
interference
(34 vs. 13%
[P = 0.006])
and jaw use
limitations (29
vs. 10%
[P = 0.005]).
*Recalculated data from study.
wConfidence intervals could not be calculated.
BDI indicates Beck Depression Inventory; BF, biofeedback; CBT, cognitive behavioural therapy; CC, comprehensive care; EI, early intervention; EMG,
electromyography; GCPS, Graded Chronic Pain Scale; IA, intraoral appliance; IMT, intraoral myofascial therapy; IMTESC, intraoral myofascial therapy and
self-care exercises; MFIQ, Mandibular Function Impairment Questionnaire; MRI, magnetic resonance imaging; NAD, neck pain and associated disorders;
RCT, randomized controlled trial; RDC, research diagnostic criteria; ROM, range of motion; SC, self-care; SCL, Symptom Checklist; SCM, self-care
management; SM, stress management; TMD, temporomandibular disorder; TMJ, temporomandibular joint; UT, usual treatment; VAS, Visual Analog Scale.
clinically important improvement in pain and jaw opening occlusal device to a structured masticatory exercise pro-
compared with the wait list group. However, there were no gram does not provide added benefits compared with a
differences between intraoral myofascial therapy with structured masticatory muscle exercise program alone.39
structured education and TMD stretching compared with The trial by Niemela et al39 compared a multimodal inter-
intraoral myofascial therapy alone (Table 2). vention that combined an occlusal device and structured
exercises for the masticatory muscles to structured exercises
alone (Table 2). The participants were asked to wear the
TMD of Variable Duration
occlusal device regularly at night and were instructed to
Occlusal Device perform the exercises 2 to 3 times daily. There were no
Evidence from 2 RCTs suggests that occlusal devices statistically or clinically important differences between groups
are ineffective for the management of TMD of variable in terms of pain, active maximal opening, laterotrusion
duration.39,40 Evidence from 1 RCT suggests that adding an movements, protrusion movement, number of painful muscle
FIGURE 2. Flow diagram of the number (n) of selected studies

for cost-effectiveness.
hour sessions of CBT combined with biofeedback by clinical

psychology research personnel. There were statistically sig-
nificant improvements in pain intensity (mean difference:
12.39/100 [95% CI, 8.08-16.70]), self-blaming (mean difference:
2.13 [95% CI, 0.95-3.31]), avoidance (mean difference: 0.82
FIGURE 1. Flow diagram of the number (n) of selected studies [95% CI, 0.03-1.61]), and depressive symptoms (P = 0.03)
for effectiveness. RCT indicates randomized controlled trial.
favoring CBT combined with biofeedback. The clinical sig-
nificance of these differences is not known. The diagnoses of
sites, or TMJ pain on palpation at the 1-month follow-up. anxiety disorder (OR = 7.18 [95% CI, 1.90-27.13]), somato-
Evidence from a second RCT comparing biofeedback/stress form disorder (OR = 12.50 [95% CI, 4.44-35.22]), and pain
management and an intraoral appliance provides conflicting disorder (OR = 9.00 [95% CI, 3.46-23.44]) were more likely in
results (Table 2).40 In their trial, Turk and colleagues com- the group receiving no intervention than in those treated with
pared biofeedback/stress management, an intraoral appliance, CBT and biofeedback. It is important to note that important
and wait list. The biofeedback/stress management inter- differences in baseline characteristics (race, marital status,
vention was delivered by a psychologist in 6 weekly working status, health insurance) and cointerventions may
1-hour sessions. The intervention included: (1) 20 minutes of have biased these results.
biofeedback-assisted relaxation; and (2) stress management.
The intraoral appliance was worn for 6 weeks. There were Adverse Events
small, short-term statistically significant reductions in pain One RCT reported increased pain with palpation over
favoring intraoral appliance over biofeedback/stress manage- the TMJ in individuals randomized to the occlusal device
ment postintervention, however, the intraoral group relapsed group.39 Three studies reported no adverse events.35,36,38
at the 6-month follow-up. Greater reduction in depressive Three studies did not report on adverse events.37,40–42
symptomatology favored biofeedback/stress management
over intraoral appliance at 6 months. The clinical importance
of these results is unknown. It is important to note that
DISCUSSION
important differences in baseline characteristics (sex) and Summary of Evidence
cointerventions may have biased these results. Our systematic review provides a synthesis of the best
evidence for the conservative management of TMD. The
CBT results suggest that CBT may benefit patients with persis-
Evidence from 1 RCT suggests that CBT may be more tent TMD and that self-care management may be an option
effective than no intervention37 for TMD disorders of variable for the management of pain and jaw disability. We also
duration. Gatchel et al37 randomized participants, with a mean found that compared with wait listing, intraoral myofascial
duration of symptoms of 3 months, to 2 groups: (1) CBT; or therapy may reduce pain and improve jaw opening in
(2) no intervention (Table 2). The CBT group received 6 one- patients with persistent TMD. We did not find evidence
that occlusal devices are effective in providing persistent are treated for TMD. Finally, we used condition-specific
pain and disability reduction in patients with TMD. MCIDs that were available in the literature. It is possible that
these MCIDs were computed from populations that differ
Comparison of Results to Previous Systematic from those reported in the RCTs included in our review and
Reviews may lack generalizability.
Liu et al25 previously reviewed the literature on the
effectiveness of CBT for TMD. However, they were unable
to determine the effectiveness of CBT. The difference in CONCLUSIONS
conclusions by Liu and colleagues and our review may be Our review suggests that patients with persistent TMD
due to the inclusion of different studies. Liu and colleagues may benefit from CBT, self-care management, or intraoral
included 2 small studies with a high risk of bias that were myofascial therapy. The current evidence does not support the
not designed to detect statistically signficant differences.55,56 use of occlusal device to reduce pain and disability in patients
Our results on the effectiveness of an occlusive device do with TMD. Other conservative interventions for the manage-
not support those of a recent meta-analysis.23 In their ment of TMD have not been supported by studies with a low
review, Ebrahim et al23 found “promising” evidence for the risk of bias. Moreover, economic evaluations to support
use of occlusal splints in reducing pain (but not in noninvasive interventions for TMD are needed.
improving quality of life and depression). These con-
clusions were based on a meta-analysis of 11 RCTs with ACKNOWLEDGMENTS
varying levels of methodological quality, heterogeneous
The authors acknowledge the invaluable contributions to
clinical conditions, and different control interventions.
this review from: Angela Verven, BA—Research Analyst,
State of the TMD Literature and Future Research UOIT-CMCC Centre for the Study of Disability Prevention
and Rehabilitation, University of Ontario Institute of
Directions
Technology (UOIT) and Canadian Memorial Chiropractic
The quality of evidence to guide the management of College (CMCC), Toronto, Ontario, Canada; Carlo
TMD is poor. Many studies identified in our search are Ammendolia, DC, PhD—University of Toronto, Toronto,
small and include important methodological limitations Ontario, Canada; David Cassidy, PhD, DrMedSc—
that limit their internal validity. Future studies need to University of South Denmark, University of Toronto, Tor-
follow accepted standards for trial designs and reporting of onto, Ontario, Canada; Doug Gross, BScPT, PhD—
results such as those found in CONSORT.57 Methodolog- University of Alberta, Edmonton, Alberta, Canada; Gail
ically robust studies are needed to evaluate the effectiveness Lindsay, RN, PhD—University of Ontario Institute of
of passive physical modalities, acupuncture, and manual Technology, Toronto, Ontario, Canada; John Stapleton,
therapies. Moreover, there is a need for more studies that MA—Open Policy Ontario, Toronto, Ontario, Canada;
evaluate the effectiveness of noninvasive interventions for Linda Carroll, PhD—University of Alberta, Edmonton,
the management of recent onset TMD (r3 mo). Finally, we Alberta, Canada; Margareta Nordin, Dr Med Sci, PT, CIE—
found no health economic evidence to inform the man- New York University, New York, New York, United States;
agement of TMD. Health economic evidence on the effi- Michel Lacerte, MDCM, MSc, FRCPC—University of
ciency (cost-effectiveness) of noninvasive interventions for Toronto, Toronto, Ontario, Canada; Mike Paulden, MA,
TMD in children and adults is a vital research priority. MSc—University of Alberta, Edmonton, Alberta, Canada;
Strengths and Limitations Murray Krahn, MD, MSc, FRCPC—University of Toronto,
Toronto, Ontario, Canada; Patrick Loisel, MD—University
Our study has strengths. First, we collaborated with a
of Toronto, University of Ontario Institute of Technology,
librarian to develop a sensitive search strategy that was
Toronto, Ontario, Canada; Poonam Cardoso, BHSc—UOIT-
methodologically rigorous. The search strategy was checked
CMCC Centre for the Study of Disability Prevention and
through peer-review to minimize errors. Second, we outlined
Rehabilitation, University of Ontario Institute of Technology
exhaustive inclusion and exclusion criteria to identify all rele-
(UOIT) and Canadian Memorial Chiropractic College
vant citations from the searched literature. Third, 2 inde-
(CMCC), Toronto, Ontario, Canada; Robert Brison, MD,
pendent reviewers conducted screening and critical appraisal to
MPH, FRCPC, CCFPC—Queen’s University, Kingston,
minimize error and bias. Fourth, we employed a well-accepted
Ontario, Canada; Shawn Marshall, MD, MSc, FRCPC—
and valid set of criteria (SIGN) for critical appraisal. Fifth, we
University of Ottawa, Ottawa, Ontario, Canada; HON.
used best-evidence synthesis to exclude studies with low
Roger Salhany, QC, BA, LLB—Retired Judge from the
internal validity to minimize bias in the reported results.
Ontario Superior Court of Justice, Canada. The authors also
Furthermore, we used a standardized methodology and all
thank Trish Johns-Wilson at the University of Ontario
reviewers undertook critical appraisal training. Sixth, we
Institute of Technology, Oshawa, ON, Canada for her review
searched the literature from 1990 onward in an effort to cap-
of the search strategy.
ture as many relevant studies as possible.
Our review also has limitations. First, we only included
REFERENCES
studies published in the English language and may have
excluded relevant non-English studies. However, previous 1. Bessa-Nogueira R, Vasconcelos B, Niederman R. The method-
systematic reviews have reported that the exclusion of non- ological quality of systematic reviews comparing temporoman-
English studies does not appear to bias results.58–61 Second, dibular joint disorder surgical and non-surgical treatment.
BMC Oral Health. 2008;8:27.
our search may have missed potentially relevant studies. 2. Di Fabio RP. Physical therapy for patients with TMD: a
However, the likelihood of this was small as we created a descriptive study of treatment, disability, and health status.
comprehensive and methodologically rigorous search strategy. J Orofac Pain. 1998;12:124–135.
Third, we limited our review to quantitative studies and 3. Canadian Dental Association. TMD (Temporomandibular
therefore did not capture the lived experience of patients who Joint Disorder). Available at http://www.cda-adc.ca/en/oral_
health/complications/temporomandibular_disorder/. Accessed 24. La Touche R, Goddard G, De-la-Hoz JL, et al. Acupuncture

May 3, 2013. in the treatment of pain in temporomandibular disorders: a
4. Rutkiewicz T, Kononen M, Suominen-Taipale L, et al. systematic review and meta-analysis of randomized
Occurrence of clinical signs of temporomandibular disorders controlled trials. Clin J Pain. 2010;26:541–550.
in adult Finns. J Orofac Pain. 2006;20:208–217. 25. Liu HX, Liang QJ, Xiao P, et al. The effectiveness of cognitive-
5. De Kanter RJ, Truin GJ, Burgersdijk RC, et al. Prevalence in behavioural therapy for temporomandibular disorders:
the Dutch adult population and a meta-analysis of signs and systematic review. J Oral Rehabil. 2012;39:55–62.
symptoms of temporomandibular disorder. J Dent Res. 26. McGowan J, Sampson M, Lefebvre C. An evidence based
1993;72:1509–1518. checklist for the Peer Review of Electronic Search
6. Gesch D, Bernhardt O, Alte D, et al. Prevalence of signs and Strategies (PRESS EBC). Evid Based Libr Inf Pract. 2010;5:
symptoms of temporomandibular disorders in an urban and 149–154.
rural German population: results of a population-based Study 27. Harbour R, Miller J. A new system for grading recommenda-
of Health in Pomerania. Quintessence Int (Berlin, Germany: tions in evidence based guidelines. BMJ. 2001;323:334–336.
1985). 2004;35:143–150. 28. van der Velde G, van Tulder M, Côté P, et al. The sensitivity of
7. Sonmez H, Sari S, Oksak Oray G, et al. Prevalence of review results to methods used to appraise and incorporate
temporomandibular dysfunction in Turkish children with mixed trial quality into data synthesis. Spine (Phila Pa 1976).
and permanent dentition. J Oral Rehab. 2001;28:280–285. 2007;32:796–806.
8. Nilsson IM, List T, Drangsholt M. Incidence and temporal 29. Spitzer WO, Skovron ML, Salmi LR, et al. Scientific
patterns of temporomandibular disorder pain among Swedish monograph of the Quebec Task Force on Whiplash-Associated
adolescents. J Orofac Pain. 2007;21:127–132. Disorders: redefining “whiplash” and its management. Spine
9. Carroll LJ, Ferrari R, Cassidy JD. Reduced or painful jaw (Phila Pa 1976). 1995;20:1S–73S.
movement after collision-related injuries: a population- 30. Carroll LJ, Cassidy JD, Peloso PM, et al. Methods for the best
based study. J Am Dent Assoc (1939). 2007;138:86–93. evidence synthesis on neck pain and its associated disorders:
10. Mobilio N, Casetta I, Cesnik E, et al. Prevalence of self- the Bone and Joint Decade 2000-2010 Task Force on Neck
reported symptoms related to temporomandibular disorders in Pain and Its Associated Disorders. J Manipulative Physiol
an Italian population. J Oral Rehab. 2011;38:884–890. Ther. 2009;32:S39–S45.
11. Yekkalam N, Wanman A. Prevalence of signs and symptoms 31. Carroll LJ, Cassidy JD, Peloso PM, et al. Systematic search
indicative of temporomandibular disorders and headaches in and review procedures: results of the WHO Collaborating
35-, 50-, 65- and 75-year-olds living in Vasterbotten, Sweden. Centre Task Force on mild traumatic brain injury. J Rehabil
Acta Odontol Scand. 2014;72:458–465. Med. 2004;(43 Suppl):11–14.
12. Dahlstrom L, Carlsson GE. Temporomandibular disorders 32. Cote P, Cassidy JD, Carroll L, et al. A systematic review of the
and oral health-related quality of life. A systematic review. prognosis of acute whiplash and a new conceptual framework
Acta Odontol Scand. 2010;68:80–85. to synthesize the literature. Spine (Phila Pa 1976). 2001;26:
13. Morris S, Benjamin S, Gray R, et al. Physical, psychiatric and E445–E458.
social characteristics of the temporomandibular disorder pain 33. Hayden JA, Cote P, Bombardier C. Evaluation of the quality
dysfunction syndrome: the relationship of mental disorders to of prognosis studies in systematic reviews. Ann Intern Med.
presentation. Br Dent J. 1997;182:255–260. 2006;144:427–437.
14. Drangsholt ML, LeResche L. Temporomandibular disorder 34. Hayden JA, van der Windt DA, Cartwright JL, et al. Assessing
pain. In: Crombie I, ed. Epidemiology of pain: A Report of the bias in studies of prognostic factors. Ann Intern Med.
Task Force on Epidemiology of the International Association for 2013;158:280–286.
the Study of Pain. Seattle: IASP Press; 1999:203–233. 35. Dworkin SF, Turner JA, Mancl L, et al. A randomized clinical
15. Häggman-Henrikson B, Rezvani M, List T. Prevalence of trial of a tailored comprehensive care treatment program for
whiplash trauma in TMD patients: a systematic review. J Oral temporomandibular disorders. J Orofac Pain. 2002;16:
Rehab. 2014;41:59–68. 259–276.
16. Haggman-Henrikson B, List T, Westergren HT, et al. Tem- 36. Dworkin SF, Huggins KH, Wilson L, et al. A randomized
poromandibular disorder pain after whiplash trauma: a clinical trial using research diagnostic criteria for temporo-
systematic review. J Orofac Pain. 2013;27:217–226. mandibular disorders-axis II to target clinic cases for a tailored
17. Brantingham JWC, Cassa TK, Bonnefin D, et al. Manipulative self-care TMD treatment program. J Orofac Pain. 2002;16:
and multimodal therapy for upper extremity and temporo- 48–63.
mandibular disorders: a systematic review. J Manipulative 37. Gatchel RJ, Stowell AW, Wildenstein L, et al. Efficacy of an
Physiol Ther. 2013;36:143–201. early intervention for patients with acute temporomandibular
18. Petrucci A, Sgolastra F, Gatto R, et al. Effectiveness of low- disorder-related pain: a one-year outcome study. J Am Dent
level laser therapy in temporomandibular disorders: a system- Assoc (1939). 2006;147:339–347.
atic review and meta-analysis. J Orofac Pain. 2011;25:298–307. 38. Kalamir A, Bonello R, Graham P, et al. Intraoral myofascial
19. La Touche R, Angulo-Dı́az-Parreño S, de-la-Hoz JL, et al. therapy for chronic myogenous temporomandibular disorder:
Effectiveness of acupuncture in the treatment of temporoman- a randomized controlled trial. J Manipulative Physiol Ther.
dibular disorders of muscular origin: a systematic review of the 2012;35:26–37.
last decade. J Altern Complement Med. 2010;16:107–112. 39. Niemela K, Korpela M, Raustia A, et al. Efficacy of
20. Maia ML, Bonjardim LR, Quintans JS, et al. Effect of low- stabilisation splint treatment on temporomandibular disorders.
level laser therapy on pain levels in patients with tempor- J Oral Rehab. 2012;39:799–804.
omandibular disorders: a systematic review. J Appl Oral Sci. 40. Turk DC, Zaki HS, Rudy TE. Effects of intraoral appliance
2012;20:594–602. and biofeedback/stress management alone and in combination
21. Jung A, Shin BC, Lee MS, et al. Acupuncture for treating in treating pain and depression in patients with temporoman-
temporomandibular joint disorders: a systematic review and dibular disorders. J Prosthet Dent. 1993;70:158–164.
meta-analysis of randomized, sham-controlled trials. J Dent. 41. Turner JA, Mancl L, Aaron LA. Short- and long-term efficacy
2011;39:341–350. of brief cognitive-behavioral therapy for patients with chronic
22. Cho SH, Whang WW. Acupuncture for temporomandibular temporomandibular disorder pain: a randomized, controlled
disorders: a systematic review. J Orofac Pain. 2010;24:152–162. trial. Pain. 2006;121:181–194.
23. Ebrahim S, Montoya L, Busse JW, et al. The effectiveness of 42. Turner JA, Mancl L, Aaron LA. Brief cognitive-behavioral
splint therapy in patients with temporomandibular disorders: a therapy for temporomandibular disorder pain: effects on daily
systematic review and meta-analysis. J Am Dent Assoc. electronic outcome and process measures. Pain. 2005;117:
2012;143:847–857. 377–387.
43. Slavin RE. Best evidence synthesis: an intelligent alternative to 53. Yuasa H, Kurita K. Treatment Group on Temporomandibular,
meta-analysis. J Clin Epidemiol. 1995;48:9–18. Disorders. Randomized clinical trial of primary treatment for
44. Viera AJ, Garrett JM. Understanding interobserver agreement: temporomandibular joint disk displacement without reduction
the kappa statistic. Fam Med. 2005;37:360–363. and without osseous changes: a combination of NSAIDs and
45. Abrams KR, Gillies CL, Lambert PC. Meta-analysis of mouth-opening exercise versus no treatment. Oral Surg Oral
heterogeneously reported trials assessing change from baseline. Med Oral Pathol Oral Radiol Endod. 2001;91:671–675.
Stat Med. 2005;24:3823–3844. 54. Oliveira-Campelo NM, Rubens-Rebelatto J, Marti N-Vallejo
46. Follmann D, Elliott P, Suh I, et al. Variance imputation for FJ, et al. The immediate effects of atlanto-occipital joint
overviews of clinical trials with continuous response. J Clin manipulation and suboccipital muscle inhibition technique on
Epidemiol. 1992;45:769–773. active mouth opening and pressure pain sensitivity over latent
47. Stauffer ME, Taylor SD, Watson DJ, et al. Definition of myofascial trigger points in the masticatory muscles. J Orthop
nonresponse to analgesic treatment of arthritic pain: an Sports Phys Ther. 2010;40:310–317.
analytical literature review of the smallest detectable difference, 55. Mishra KD, Gatchel RJ, Gardea MA. The relative efficacy of
the minimal detectable change, and the minimal clinically three cognitive-behavioral treatment approaches to tempor-
important difference on the pain visual analog scale. Int J omandibular disorders. J Behav Med. 2000;23:293–309.
Inflam. 2011;2011:231926. 56. Gardea MA, Gatchel RJ, Mishra KD. Long-term efficacy
48. Carroll LJ, Jones DC, Ozegovic D, et al. How well are you of biobehavioral treatment of temporomandibular disorders.
recovering? The association between a simple question about J Behav Med. 2001;24:341–359.
recovery and patient reports of pain intensity and pain 57. Schulz KF, Altman DG, Moher D. CONSORT 2010 State-
disability in whiplash-associated disorders. Disabil Rehabil. ment: updated guidelines for reporting parallel group
2012;34:45–52. randomised trials. BMJ. 2010;340:340.
49. Yuan J, Purepong N, Kerr DP, et al. Effectiveness of 58. Moher D, P B, Lawson ML, et al. The inclusion of reports of
acupuncture for low back pain: a systematic review. Spine. randomised trials published in languages other than English
2008;33:E887–E900. in systematic reviews. Health Technol Asses (Winchester,
50. Hiroe T, Kojima M, Yamamoto I, et al. Gradations of clinical England). 2003;7:1–90.
severity and sensitivity to change assessed with the Beck 59. Moher D, Fortin P, Jadad AR, et al. Completeness of
Depression Inventory-II in Japanese patients with depression. reporting of trials published in languages other than English:
Psychiatry Res. 2005;135:229–235. implications for conduct and reporting of systematic reviews.
51. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting Lancet. 1996;347:363–366.
items for systematic reviews and meta-analyses: the PRISMA 60. Morrison A, Polisena J, Husereau D, et al. The effect of
statement. BMJ. 2009;339:b2535. English-language restriction on systematic review-based meta-
52. Dworkin SF, Huggins KH, Wilson L, et al. A randomized analyses: a systematic review of empirical studies. Int J Technol
clinical trial using research diagnostic criteria for temporo- Asses Health Care. 2012;28:138–144.
mandibular disorders-axis II to target clinic cases for a tailored 61. Sutton AJ, Duval SJ, Tweedie RL, et al. Empirical assessment
self-care TMD treatment program. J Orofac Pain. 2002;16: of effect of publication bias on meta-analyses. BMJ. 2000;320:
48–63. 1574–1577.

Randhawa 2016

Uploaded by

Copyright:

Available Formats

Randhawa 2016

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Randhawa 2016

Uploaded by

Copyright:

Available Formats

REVIEW ARTICLE

The Effectiveness of Noninvasive Interventions for

Results: Our search for eﬀectiveness studies yielded 16,995 citations;

260 | www.clinicalpain.com Clin J Pain Volume 32, Number 3, March 2016

2014: CINAHL, Cochrane Health Technology Assessment Statistical Analyses

FIGURE 2. Flow diagram of the number (n) of selected studies

hour sessions of CBT combined with biofeedback by clinical

health/complications/temporomandibular_disorder/. Accessed 24. La Touche R, Goddard G, De-la-Hoz JL, et al. Acupuncture

You might also like