Journal of Oral Rehabilitation
Journal of Oral Rehabilitation 2010 37; 430–451
Review Article
Management of TMD: evidence from systematic reviews and
meta-analyses
T. LIST* & S. AXELSSON† *Department of Stomatognathic Physiology, Faculty of Odontology, Malmö University, Malmö and
†
The Swedish Council on Technology Assessment in Health and Care, Stockholm, Sweden
SUMMARY This systematic review (SR) synthesises
recent evidence and assesses the methodological
quality of published SRs in the management of
temporomandibular disorders (TMD). A systematic
literature search was conducted in the PubMed,
Cochrane Library, and Bandolier databases for 1987
to September 2009. Two investigators evaluated the
methodological quality of each identified SR using
two measurement tools: the assessment of multiple
systematic reviews (AMSTAR) and level of research
design scoring. Thirty-eight SRs met inclusion crite-
ria and 30 were analysed: 23 qualitative SRs and
seven meta-analyses. Ten SRs were related to occlu-
sal appliances, occlusal adjustment or bruxism; eight
to physical therapy; seven to pharmacologic treat-
ment; four to TMJ and maxillofacial surgery; and six
to behavioural therapy and multimodal treatment.
The median AMSTAR score was 6 (range 2–11).
Eighteen of the SRs were based on randomised
clinical trials (RCTs), three were based on case–
Background
Population-based studies among adults report that
approximately 10–15% have temporomandibular dis-
orders (TMD) pain and 5% a perceived need for
treatment (1, 2). Several studies have pointed out that
persistent and recurrent pain has a potential impact on
daily life – mainly in the areas of psychological
discomfort, physical disability, and functional limita-
tions – which leads to limitations in quality of life (3–6).
Keeping current on new developments in a medical
or dental field is often an imposing task for researchers
and clinicians. One researcher estimated that to keep
abreast of developments in his field, he would need to
ª 2010 Blackwell Publishing Ltd
control studies, and nine were a mix of RCTs and
case series. Most SRs had pain and clinical measures
as primary outcome variables, while few SRs
reported psychological status, daily activities, or qual-
ity of life. There is some evidence that the following
can be effective in alleviating TMD pain: occlusal
appliances, acupuncture, behavioural therapy, jaw
exercises, postural training, and some pharmacolog-
ical treatments. Evidence for the effect of electro-
physical modalities and surgery is insufficient, and
occlusal adjustment seems to have no effect. One
limitation of most of the reviewed SRs was that the
considerable variation in methodology between the
primary studies made definitive conclusions impos-
sible.
KEYWORDS: systematic review, randomised clinical
trial, evidence-based medicine, treatment, assess-
ment
Accepted for publication 8 March 2010
read about 17–20 original articles – as a daily average
(7).
Systematic reviews (SRs) are a cornerstone in evi-
dence-based medicine. An SR is a compilation of all
published research that addresses a carefully formulated
question. Research results are collected based upon pre-
determined inclusion and exclusion criteria; data are
critically analysed and synthesised so that evidence-
based conclusions on the benefit or risks of different
methods can be drawn, with the final goal to improve
patient care. An SR may have a qualitative approach if
data from the primary studies are presented descriptively
or a quantitative approach if statistical analysis has
combined data in a meta-analysis. Of the over 24
doi: 10.1111/j.1365-2842.2010.02089.x

instruments that have been developed to assess SR
quality, one recent instrument deserves mention: assess-
ment of multiple systematic reviews (AMSTAR) (8). This
tool comprises 11 items and has good face and content
validity for measuring methodological quality (9).
This systematic review of published SRs that survey
TMD management efficacy adds to these existing
reviews by (i) synthesising recent evidence and (ii)
assessing SR methodological quality.
Methods
Procedure
Both authors selected the articles based on inclusion
and exclusion criteria and independently read all titles
and abstracts that were found in multiple searches to
identify potentially eligible articles for inclusion. All
potentially eligible SRs were then retrieved, and full-
text articles were reviewed to determine whether they
met inclusion criteria. Disagreement was resolved by
discussion among the investigators. Authors were not
contacted for missing information. The reviewers were
experienced oro-facial pain specialist clinicians or
methodologists in evidence-based medicine.
The investigators independently evaluated the
methodological quality of each identified SR using
AMSTAR and level of research design (LRD) scoring.
The following data were extracted: study design, diag-
nosis, number of patients, types of intervention, outcome
measures, results, quality score, and author’s conclusion.
Database search
The search encompassed all the articles that were (i)
indexed in PubMed, the Cochrane Library, and Bando-
lier, (ii) published in English, Swedish, or German, and
(iii) published between 1 January 1987 and 8 September
2009. The search strategy was designed to identify SRs
that focused on TMD management. The following search
terms were used for PubMed: Craniomandibular disor-
ders ⁄ drug therapy [MeSH] OR Craniomandibular disor-
ders ⁄ surgery [MeSH] OR Craniomandibular
disorders ⁄ surgery [MeSH] AND ‘‘Review’’ AND Meta-
analysis [MeSH] AND ‘‘Systematic review’’. For Coch-
rane Library, the search strategy included the terms
craniomandibular disorders and temporomandibular
joint (TMJ). References in original articles and SRs were
hand-searched to identify additional SRs.
ª 2010 Blackwell Publishing Ltd
MANAGEMENT OF TMD 431
Inclusion criteria
Besides the database search criteria listed previously,
the selected articles must:
1 Be an SR
2 Focus on the management of TMD treatment
Systematic reviews that covered oro-facial pain or
TMD prevention besides TMD treatment effect were
included.
Exclusion criteria
1 Dual publication
2 Update of the SR by the same author in a later
publication
3 SR of orthodontic treatment focusing on develop-
ment and not treatment of TMD
4 Systematic reviews of SRs
Quality assessment
Two instruments were used to assess the methodolog-
ical quality of the SRs: (i) AMSTAR and (ii) LRD.
Shea et al. developed AMSTAR by evaluating the
importance of 37 items commonly used in SR assess-
ment and reducing these items to an 11-item instru-
ment that addresses key domains in methodological
qualities (8). Assessment of multiple systematic
reviews is reported to have good face and content
validity (9).
Assessment of multiple systematic reviews appraises
these key items:
1 Was an á priori design provided?
2 Was there duplicate study selection and data extrac-
tion?
3 Was a comprehensive literature search provided?
4 Was the status of publication (i.e. grey literature)
used as an inclusion criterion?
5 Was a list of studies (included and excluded)
provided?
6 Were the characteristics of the included studies
provided?
7 Was the scientific quality of the included studies
assessed and documented?
8 Was the scientific quality of the included studies
used appropriately in formulating conclusions?
9 Were the methods used to combine the findings of
studies appropriate?
10 Was the likelihood of publication bias assessed?
432 T. LIST & S. AXELSSON

11 Was the conflict of interest stated? Computerised search of databases,

Each of these 11 items was scored as ‘Yes’, ‘No’, reference checking, abstracts:

‘Can’t answer’, or ‘Not applicable’. PubMed: 602 references

Cochrane library: 19 references
The LRD assesses internal validity of the primary Bandolier: 1 reference
studies based on design. The research design levels are Handsearch: 3 references

structured hierarchically according to evidence strength 625 references

(10):
1 Systematic review of randomised clinical trials (RCTs) Abstracts excluded:
2 Randomised clinical trial 587 references

3 Study without randomisation, such as a cohort study,

case–control study
Screening of articles applying
4 A non-controlled study, such as cross-sectional study, inclusion criteria:
case series, case reports 38 references

5 Narrative review or expert opinion

Reliability of quality assessment Abstracts excluded:

8 references
(11–18)
Both authors independently assessed SR quality to
determine inter-test reliability. The inter-reliability of
Articles included for quality
the two authors in assessing the quality of the SRs was assessment:
calculated with kappa-statistics. All articles were dis-
30 references
cussed to verify appraisal process until consensus was (19–47, 65)
(23) reported two SRs
reached. Disagreements on individual item scores were
resolved with discussion. Fig. 1. Flow diagram of references included and excluded in the
systematic review.

Results
tenderness ⁄ pain on palpation, and 10 of the SRs
Thirty-eight SRs were read in full text; 30 of these were included measures of psychological status, daily activ-
included in this SR: 23 were qualitative SRs and seven ities, or quality of life. The median AMSTAR score was 6
were meta-analyses. Eight SRs were excluded after (range 2–11). The LRD scores for the SRs are as follows:
reading because upon closer scrutiny, they did not fulfil
the inclusion criteria: three SRs had been updated in Level I–II 2 SRs
more recent SRs by the same author (11–13), one study Level I-III 1 SR
did not focus on TMD management (14), two SRs Level II 16 SRs
Level II–III 1 SR
focused on development of TMD following orthodontic
Level III 1 SR
intervention (15, 16), and two SRs were SRs of SRs (17, Level II–IV 9 SRs
18) (Fig. 1).
Tables 1–5 list characteristics and assess quality of the
included SRs. Although the most common diagnosis in The SRs were divided into five treatment groups:
the SRs was TMD, more specific TMD diagnoses such as 1 Occlusal appliances, occlusal adjustment, and bruxism
disc displacements and myofascial pain were sometimes 2 Physical treatment [acupuncture, transcutaneous
also reported. Two SRs focused on bruxism. Although it electrical nerve stimulation (TENS), exercise, and
is in the TMD domain, bruxism differs from other TMD mobilisation]
diagnoses: although it may be accompanied by pain, 3 Pharmacologic treatment
bruxism is not related to pain in many cases. The 4 TMJ and maxillofacial surgery
number of patients ranged from 0 to 7173 in the SRs. 5 Behavioural therapy and multimodal treatment
Twenty-nine SRs had pain intensity or pain reduction Ten SRs evaluated occlusal appliances, occlusal
as primary outcome measures, 25 of the SRs reported adjustment, or bruxism (Table 1). Eight of these con-
clinical outcome measures such as jaw movement and cerned TMD treatment (12, 19–26), one the effect of

ª 2010 Blackwell Publishing Ltd

 Table 1. Characteristics of systematic reviews (SRs) of occlusal appliances, occlusal adjustment, and orthodontic treatment

Study design, Intervention (I)

Authors, year, diagnosis, and and control (C) Outcome Quality Authors’ (A) conclusions
reference no. of patients groups measures Results score Reviewers’ (R) comments

Santacatterina A SR and I1: Occlusal Pain reduction I2 better than I1 AMSTAR 2 A: A comparison between the two kinds of
1998 (25) Meta-analysis of appliance TMJ click for pain reduction LRD II-IV treatment has demonstrated that the

ª 2010 Blackwell Publishing Ltd

6 RCTs I2: repositioning and TMJ click. repositioning splint is more effective both in the
Disc displacement splint resolution of the articular click and in the
with reduction resolution of the pain (P < 0Æ001).
212 patients R: Methodological weaknesses of primary studies
such as heterogeneous patient material, outcome
measures not clearly defined, and three of the
studies were not RCTs.
Al-Ani MZ Qualitative SR of I1: Occlusal Pain reduction I1 no better than C1 AMSTAR 6 A: There is insufficient evidence either for or
2003 (19) 12 RCTs appliance Jaw motion I1 somewhat better LRD II against the use of stabilisation splint therapy over
Myofascial pain C1: Other treatment than C2 other active interventions for the treatment of
496 patients (biofeedback, jaw I1 no better than C3 temporomandibular myofascial pain. However,
exercises, it appears that stabilisation splint therapy may be
acupuncture) beneficial for reducing pain severity at rest and on
C2: No treatment palpation and depression when compared to no
C3: Placebo treatment.
R: Included studies were small. Patient material
was heterogeneous. Outcome measures varied
between studies. Short-term follow-up.
Exclusion, inclusion, and diagnostic criteria
not clearly defined.
Forssell H Qualitative I1: Occlusal Pain reduction Contradictory AMSTAR 7 A: Occlusal splints yielded equivocal results. Even
2004 (20) SR of 20 RCTs appliance Global results for I1 LRD II for the most studied area, stabilisation splints for
TMD I2: Occlusal improvement compared with myofascial pain, the results do not justify definite
1138 patients adjustment Clinical C1–C3 conclusions about the efficacy of splint therapy.
C1: Other treatment examination I2 no better than Their clinical effectiveness to relieve pain also
(biofeedback, jaw Depression scale C1 or C2 seems modest when compared with treatment
exercises, methods in general. None of the occlusal
acupuncture) adjustment studies provided evidence supporting
C2: No treatment the use of this treatment method.
C3: Placebo R: Included studies small, often not blinded,
heterogeneity concerning outcome measures and
control treatment.
MANAGEMENT OF TMD
433
 434

Table 1. (Continued)

Study design, Intervention (I)

Authors, year, diagnosis, and and control (C) Outcome Quality Authors’ (A) conclusions
reference no. of patients groups measures Results score Reviewers’ (R) comments

Türp JC Qualitative I1: Intra-oral Pain reduction I1 better than C2 AMSTAR 6 A: Based on the currently best available evidence,
2004 (21) SR of 9 RCTs appliance Clinical I1 no better than C1 LRD II it appears that most patients with masticatory
T. LIST & S. AXELSSON

Myofascial pain C1: Other treatment examination muscle pain are helped by incorporation of a
482 patients including placebo Depression scale stabilisation splint. A stabilisation splint does not
C2: No treatment appear to yield a better clinical outcome than a
soft splint, a non-occluding palatal splint, physical
therapy, or acupuncture.
R: Well-conducted SR. Limitations: small patient
studies, outcome measures vary between studies,
no long-term results.
Fricton J Qualitative I1: Stabilisation Pain reduction No difference AMSTAR 4 A: Stabilisation splints can reduce TMD pain
2006 (22) SR of 39 RCTs splint between I1 and I2. LRD II compared to placebo splints. Stabilisation splints
TMD I2: Anterior I1 and C2 have are equally effective in reducing pain compared to
patients* positioning and similar effects physical therapy, acupuncture and behavioural
soft splints I2, I1, and C1 have therapy in the short term. The long-term effects of
C1: Placebo similar effects behavioural therapy may be better than splints in
C2: Other treatment reducing symptoms in more severe patients with
psychosocial problems.
R: The article is an overview and separate articles
are under publishing with details regarding
methods and results.
SBU Qualitative I1: Stabilisation Pain reduction I1 better than C3 AMSTAR 6 A: Occlusal appliances gave better pain reduction
2006 (23) SR of 3 SRs and splint Clinical I1 and C2 have LRD I–II than no treatment. Treatment with occlusal
3 RCTs I2: Occlusal examination similar effect appliance had similar effect as other therapies
TMD adjustment Depression scale Results of I1 whereas the effect compared with placebo was
2299 patients C1: Placebo compared with C1 contradictory. No study found occlusal
C2: Other are contradictory adjustment to be effective compared to a control.
treatments I2 and C1 have R: Reviewers and authors are identical persons
C3: No treatment similar effect

ª 2010 Blackwell Publishing Ltd

 Table 1. (Continued)

Study design, Intervention (I)

Authors, year, diagnosis, and and control (C) Outcome Quality Authors’ (A) conclusions
reference no. of patients groups measures Results score Reviewers’ (R) comments

Stapelman H Qualitative I1: NTI splint EMG activity Reducing EMG AMSTAR 7 A: NTI-tss devices may be successfully used to
2008 (24) SR of 5 RCTs C1: Flat occlusal Polysomnographic activity: I1 more LRD II manage bruxism and TMDs. To avoid potential

ª 2010 Blackwell Publishing Ltd

TMD splint evaluation effective than C1 unwanted effects, it should be chosen only if a
Bruxism C2: Bleaching tray Pain intensity Improvement in patient will be compliant with follow-up. Two of
Tension-type Jaw opening pain reduction: five studies focused on treatment of TMD.
headache (TTH) Comfort results for I1 Overall, small groups, median of 14 participants
Migraine Analgesic compared with C1 in each group. As adverse events were reported
190 patients consumption are contradictory for I1, and I1 was not more effective than C1,
Treating TTH and only limited indication is warranted.
migraine: I1 more R: Well-conducted SR. Limitations are small
effective than C2 studies. Methodological weaknesses in primary
Five reports of studies such as heterogeneity concerning
complications or outcome measures and diagnostic criteria.
side-effects
reported for I1
Koh H SR and I1: Occlusal Global symptoms No difference AMSTAR 10 A: There is an absence of evidence, from RCTs, that
2009 (26) meta-analysis adjustment Relief of headache between I1 and C1 LRD II occlusal adjustment treats or prevents TMD.
of 6 RCTs C1: Placebo, no Quality of life Occlusal adjustment cannot be recommended for
TMD treatment or management or prevention of TMD.
392 patients reassurance R: Small studies, diagnostic criteria inaccurate.
Outcome measures poorly defined.
Marcedo CR SR and I1: Occlusal splint Sleep variables: e.g. No difference AMSTAR 9 A: There is not sufficient evidence to state that the
2009 (27) meta-analysis I2: Other appliances EMG activity, between I1 and I2 LRD II occlusal splint is effective for treating sleep
of 5 RCTs I3: Other therapies arousal index in the bruxism. Indication of its use is questionable
Bruxism C1: No treatment Report of bruxism meta-analysis of concerning sleep outcomes, but it may be that
63 patients by partner arousal index. there are some benefits concerning tooth wear.
Tooth wear No difference R: Small studies, methodological weaknesses in
between I1 and C1 primary studies
for tooth wear
facets
No difference
between I1 and I3
for TMD pain
MANAGEMENT OF TMD
435
436 T. LIST & S. AXELSSON

occlusal appliances on sleep bruxism (27), and one the

devices. In a case series of patients with severe

A: One study was prospective and reported less
attrition in young children wearing bite plates

R: Small studies, methodological weaknesses in

effect of occlusal appliances on attrition (28). Six SRs

primary studies. The results seem to confirm

attrition, occlusal splints were found to slow
compared with subjects who did not wear
reported that occlusal appliances and other therapies
had similar effect, three that occlusal appliances were
better than no treatment, and four that there was no

AMSTAR, assessment of multiple systematic reviews; LRD, level of research design; NTI, nociceptive trigeminal inhibition; RCT, randomised clinical trial; TMD,
difference between occlusal appliances and non-occlud-

down the rate of tooth wear.

ing splints (placebo). One SR reported stabilisation
Reviewers’ (R) comments
Authors’ (A) conclusions

splints and anterior positioning and soft splints to yield

similar results, while one SR reported anterior posi-

clinical experience.
tioning splints to be more effective than stabilisation
splints. One SR reported contradictory results in a
comparison of nociceptive trigeminal inhibition (NTI)
splints and occlusal appliances. Several complications
and adverse events were documented for the NTI-tss
device. Three SRs reported no difference in outcome
between occlusal adjustments and control treatment.
AMSTAR 6
LRD II-IV

Eight SRs evaluated various physical treatments in

Quality

patients with TMD (Table 2). Of these SRs, six assessed

score

acupuncture (23, 29–32), three jaw exercises (23, 33,

34), two manual therapy (33, 34), and three various
forms of electrical therapy (30, 34). Three SRs reported
compared with C1.
I1 reduces attrition

that acupuncture was better than no treatment, four

SRs that acupuncture and other therapies yielded
similar results, and one SR that acupuncture and
Results

placebo treatment had similar results. Two SRs found

no support for acupuncture treatment.
As for the other types of physical treatment, three
SRs reported that jaw exercises were better than
measures
Outcome

Attrition

controls, and two SRs found postural training to be

better than no treatment. No SR indicated that electro-
physical modalities such as TENS, laser, and ultrasound
were effective in reducing pain.
C1: no treatment
Intervention (I)
and control (C)

Seven SRs evaluated various pharmacological treat-

I1: Occlusal

ments in patients with TMD, atypical facial pain,

appliance

trigeminal neuralgia, and burning mouth syndrome

groups

(23, 35–40) (Table 3). One SR reported anti-epileptic

drugs to be more effective than placebo in patients
with trigeminal neuralgia. In patients with TMD,
2 studies related to

and 1 case series.

pharmacological treatment that was found to be more

SR of 33 studies

bruxism. 1 RCT
no. of patients
diagnosis, and

effective than placebo was analgesics (two SRs),

Intervention

*Number of patients not reported.

Study design,

because of

27 patients
Qualitative

clonazepam or diazepam (three SRs), antidepressants

temporomandibular disorders.

(four SRs), and hyaluronate (one SR). This last SR also

reported hyaluronate and glycocorticoid to produce a
Table 1. (Continued)

similar effect. Three SRs reported amitriptyline to have

a moderate effect in patients with atypical facial pain.
van¢t Spijker A
Authors, year,

Two SRs reported the use of botulinum toxin in

2007 (28)
reference

patients with pain or masseter hypertrophy. One SR

reported positive outcome based upon one RCT,
whereas the other SR found no studies that met its

ª 2010 Blackwell Publishing Ltd

 Table 2. Characteristics of systematic reviews (SRs) of physical therapy: acupuncture, TENS, exercise, and mobilisation

Study design, Intervention (I)

Authors, year, diagnosis, and and control (C) Outcome Quality Authors’ (A) conclusions
reference no. of patients groups measures Results score Reviewers’ (R) comments

Ernst E Qualitative I1: Acupuncture Pain Intensity I1 better than C2 AMSTAR 5 A: Although all studies agree with the notion that
1999 (29) SR of 6 RCTs C1: Occlusal Daily activity No difference LRD II acupuncture is effective for TMD, this hypothesis
TMD appliance Global between I1 and C1 requires confirmation through more rigorous

ª 2010 Blackwell Publishing Ltd

205 patients C2: No treatment improvement investigations.
Clinical R: Short follow-up time in two studies. One study
examination reports a 1-year follow-up. No reported side-effects.
SR based on three studies with low quality.
Jedel E Qualitative I1: Biofeedback Self-reported No evidence of an AMSTAR 3 A: The studies were heterogeneous with low quality,
2003 (30) SR of 7 RCTs I2: Acupuncture symptoms effect for any LRD II and therefore it is not possible to draw any
TMD I3: TENS Pain intensity treatment mode conclusions.
379 patients C1: No treatment Clinical R: Limitations in the SR methodology and therefore
C2: Other treatment examination difficult to draw any conclusions
Rosted P Qualitative I1: Acupuncture Pain intensity No difference AMSTRAR 4 A: Acupuncture and conventional treatment were
1998 (31) SR of 15 RCTs C1: Other treatment between I1 and C1 LRD II-IV found to have similar effect. The effect in treating
(7 RCTs on TMD) TMD and facial pain seems real, and acupuncture
Acute toothache could be an alternative to conventional treatment.
TMD R: Non-RCTs were included in the evaluation.
patients* Methodology and results not clearly described.
SBU Qualitative I1: Acupuncture Pain intensity I1 better than C3 AMSTAR 6 A: Acupuncture was found to show better pain
2006 (23) SR of 3 C1: Placebo Daily activities I1 and C2 have LRD I–II reduction than no treatment and similar effect
SRs and 1 RCT acupuncture Global similar effect compared to other treatments. Compared to placebo,
TMD C2: Other treatment improvement I1 and C1 have the results are contradictory.
575 patients C3: No treatment Clinical contradictory R: Reviewers and authors are identical persons
examination results
SBU Qualitative I1: Jaw exercises Pain intensity I1 better than C1 AMSTAR 6 A: The evidence to draw any conclusions regarding
2006 (23) SR of 6 RCTs I2: Pulsed radio Daily activities I1–I4 no different LRD I–II physical treatment for TMD is insufficient because
TMD frequency therapy Clinical from C2 the studies are heterogeneous regarding diagnosis
279 patients I3: Laser examination I5 better than C3 and treatment method.
I4: Pulsed R: Some of the studies showed a difference compared
electromagnetic with controls. Because these findings were not
fields supported by more studies, there is lack of evidence.
I5: Postural Reviewers and authors are identical persons
correction
C1: Minimal
information
C2: Placebo
C3: No treatment
MANAGEMENT OF TMD
437
 438

Table 2. (Continued)

Study design, Intervention (I)

Authors, year, diagnosis, and and control (C) Outcome Quality Authors’ (A) conclusions
reference no. of patients groups measures Results score Reviewers’ (R) comments

Fink M Qualitative I1: Acupuncture Pain intensity I1 and C2 have AMSTAR 7 A: The analysed studies on acupuncture in the
2006 (32) SR of 6 RCTs C1: Sham Global similar effects LRD II treatment of TMD confirm acupuncture to be as
TMD acupuncture improvement No difference effective as conservative treatment.
223 patients C2: Other treatment Daily activities between I1 and C1 R: Short follow-up time in three studies. One study
C3: No treatment Clinical reports a 1-year follow-up. No reported side-effects.
T. LIST & S. AXELSSON

examination SR based on three studies with low and one with

Pain thresholds high quality.
McNeely M Qualitative I1: Exercise and Pain reduction No synthesis of AMSTAR 7 A: The results support use of active and passive oral
2006 (33) SR of 12 RCTs manual therapy Jaw mobility results LRD II exercises and exercises to improve posture as
TMD I2: Acupuncture effective interventions to reduce symptoms
480 patients I3: Electrophysical associated with TMD. There is inadequate
modalities (PRFE, information to either support or refute use of
TENS, biofeedback, acupuncture in TMD treatment. There is no evidence
laser) to support use of electrophysical modalities to reduce
C1: Other therapies TMD pain.
C2: Placebo R: Methodological weaknesses of primary studies such
as diagnostic criteria of TMD, outcome measures and
chosen controls limit conclusions in the study.
Medlicott MS Qualitative I1: Exercise Pain intensity No synthesis of AMSTAR 5 A: Active exercise and manual mobilisation may be
2006 (34) SR of 24 RCTs I2: Manual therapy Global results LRD II-IV effective. Postural training may be used in
and 6 uncontrolled I3: Electrotherapy improvement combination with other interventions, as
studies (ultrasound, TENS, Clinical independent effects of postural training are
TMD laser, PRFE) examination unknown. Mid-laser therapy may be more effective
1071 patients I4: Relaxation Jaw mobility than other electrotherapy modalities. Programmes
training and Pressure pain involving relaxation techniques and biofeedback,
education threshold electromyographic training, and proprioceptive
C1: Occlusal splint re-education may be more effective than placebo
C2: Placebo treatment or occlusal splints. Combinations of active
C3: Waiting-list exercise, manual therapy, postural correction, and
relaxation techniques may be effective.
R: Because of the heterogeneous population and to
differences in diagnosis and outcome measures, it not
possible to draw any clear conclusions from this SR.

TENS, transcutaneous electric nerve stimulation; PRFE, pulsed radio frequency energy; AMSTAR, assessment of multiple systematic reviews; LRD, level of research design; RCT,
randomised clinical trials; TMD, temporomandibular disorders.

ª 2010 Blackwell Publishing Ltd

 Table 3. Characteristics of systematic reviews (SRs) of pharmacologic treatment

Study design, Intervention (I)

Authors, year, diagnosis, and and control (C) Outcome Quality Authors’ (A) conclusions
reference no. of patients groups measures Results score Reviewers’ (R) comments

ª 2010 Blackwell Publishing Ltd

Sommer C Qualitative I1: Carbamazepin, Pain reduction Trigeminal neuralgia: AMSTAR 4 A: Apart from studies on trigeminal neuralgia,
2002 (35) SR of 27 RCTs baclofen, >50% I1 significantly better LRD II there is little evidence for pharmacotherapy in
TMD, trigeminal lamotrigine than C1 oro-facial pain.
neuralgia, atypical I2: Clonazepam, TMD: moderate R: Some of the studies showed a difference
facial pain Diazepam evidence that I2 and compared with controls. Limitations in the
931 patients I3: Amitriptyline I3 are better than C1 primary studies were small groups,
C1: Placebo Atypical facial pain: a heterogeneous outcome measures, and no
moderate effect of I3 long-term follow-up.
compared with C1
List T Qualitative I1: Analgesics Pain reduction TMD and atypical AMSTAR 7 A: The common use of analgesics in TMD, AFP,
2003 (36) SR of 11 RCTs I2: Antidepressants Global facial pain: few LRD II and BMS is not supported by scientific
TMD, atypical I3: Benzodiazepines improvement studies found better evidence.
facial pain, I4: Miscellaneous Depression scale effect of I1-I4 R: The studies are too heterogeneous to draw
burning mouth (corticosteroids, compared with C1 any conclusions from.
syndrome sodium Burning mouth:
368 patients hyaluronate, I2 = C1
sumatriptan,
cocaine)
C1: Placebo
SBU Qualitative I1: Analgesics Pain reduction TMD and atypical AMSTAR 6 A: Studies on pharmacological treatment of
2006 (23) SR of 1 SR and I2: Antidepressants facial pain: few LRD I–II TMD, atypical facial pain, and burning mouth
13 RCTs I3: Benzodiazepines studies found better syndrome report contradictory results.
TMD, atypical I4: Miscellaneous effect of I1–I4 No conclusions can be drawn as the studies are
facial pain, (corticosteroids, compared with C1 heterogeneous regarding diagnosis and
burning mouth sodium For burning mouth, treatment method.
syndrome hyaluronate, I = C1 R: Reviewers and authors are identical persons
968 patients sumatriptan,
capsaicin,
botulinum toxin)
C1: Placebo
MANAGEMENT OF TMD
439
 440

Table 3. (Continued)

Study design, Intervention (I)

Authors, year, diagnosis, and and control (C) Outcome Quality Authors’ (A) conclusions
reference no. of patients groups measures Results score Reviewers’ (R) comments

Shi ZC SR and I1: Hyaluronate Symptoms Long-term effects AMSTAR 11 A: There is insufficient consistent evidence to
2009 (37) meta-analysis I2: Hyaluronate + (e.g. pain, favour I1 compared LRD II support or refute the use of hyaluronate for
of 7 RCTs Arthroscopy ⁄ lavage Clinical to C1 treating patients with TMD.
TMD, rheumatoid C1: Placebo examination I1 had the same R: Methodological weaknesses of primary
arthritis C2: Glycocorticoid Adverse events long-term effects on studies such as diagnostic criteria of TMD and
364 patients C3: Arthroscopy ⁄ lavage symptoms and clinical outcome measures in the study.
T. LIST & S. AXELSSON

signs compared to C2
Comparing I1 to C3,
results were
inconsistent
Al-Muharraqi MA SR and I1: Botulinum toxin Self-reported 167 references were AMSTAR 7 A: No randomised trial on the efficacy of
2009 (38) meta-analysis of C1: Placebo facial retrieved, but none LRD II intra-muscular injections of botulinum toxin
RCTs appearance matched the with bilateral benign masseter hypertrophy
No studies Pain and inclusion criteria. was identified.
included discomfort R: No trend of the effect can be drawn because
Masseter all studies were excluded.
hypertrophy
0 patients
Cascos-Romero J Qualitative I1: Antidepressants Pain I1 better than C1 AMSTAR 4 A: The use of tricyclic antidepressants for the
2009 (39) SR of 1 SR, C1: Placebo LRD I–III treatment of TMD is recommended.
1 RCT and 1 R: Synthesis of results from primary studies are
case–control study missing, and therefore, because of limitations
TMD in the SR, it is difficult to draw any
patients* conclusions.
Ihde S 2007 (40) Qualitative I1: Botulinum toxin Pain reduction I1 better than C1 for AMSTAR 3 A: Botulinum toxin appears relative safe and
SR of 1 RCT and C1 Placebo Jaw opening reducing pain based LRD II-IV effective in treating chronic facial pain
10 case series Functional on one RCT. associated with masticatory hyperactivity.
TMD improvement No synthesis of R: Methodological limitations in the SR.
Bruxism Aesthetic result results. Results only relate to one RCT study.
Masseter Synthesis of results missing so a conclusion of
hypertrophy the effect of Botulinum toxin is difficult to
Oro-mandibular determine
dystonia
402 patients

AMSTAR, assessment of multiple systematic reviews; LRD, level of research design; RCT, randomised clinical trials; TMD, temporomandibular disorders.
*Number of patients not reported.

ª 2010 Blackwell Publishing Ltd

 Table 4. Characteristics of systematic reviews (SRs) of the TMJ and maxillofacial surgery

Study design,
Authors, year, diagnosis, and Intervention (I) and Outcome Quality Authors’ (A) conclusions
reference no. of patients control (C) groups measures Results score Reviewers’ (R) comments

ª 2010 Blackwell Publishing Ltd

Reston JT SR and I1: Arthrocentesis Pain reduction Disc displacement AMSTAR 4 A: Surgical treatment appears to provide some
2003 (41) meta-analysis of I2: Arthroscopy Global with reduction: LRD II–IV benefit to patients refractory to non-surgical
30 studies (3 RCTs I3: Disc improvement I2 and I3 (most studies, therapies. The most reliable evidence supports the
and 27 repair ⁄ repositioning Jaw mobility comparable results level IV) effectiveness of arthrocentesis and arthroscopy for
uncontrolled I4: Discectomy Disc displacement patients with disc displacement with reduction.
studies) without reduction: R: Methodological weaknesses of primary studies
Disc displacement Similar results for such as heterogeneous patient material, outcome
with reduction, I1, I2, and I3 measures not clearly defined, and majority of the
Disc displacement studies were not RCTs.
without reduction
1463 patients
Kropmans TJ Qualitative SR of 24 I1: Arthroscopy Pain intensity No synthesis of AMSTAR 2 A: No distinguishing effects on jaw mobility, jaw
1999 (42) studies (6 RCTs I2: Arthrocentesis Jaw function results presented LRD II–IV function, or pain intensity was seen between
and 6 case–control I3: Physical therapy Jaw mobility arthroscopic surgery, arthrocentesis, and physical
and 12 (e.g. exercise, therapy in patients with permanent
uncontrolled massage, TENS) temporomandibular joint disc displacement.
studies) C1: Placebo R: Methodological weaknesses of primary studies
Disc displacement C2: No treatment such as diagnostic criteria of TMD and outcome
without reduction measures in the study. A majority of the studies
4916 patients are LRD level IV, and therefore, it not possible to
draw any clear conclusions from this SR.
Abrahamsson C Qualitative I1: Bilateral sagittal Self-report of Contradictory AMSTAR 8 A: Because of few studies with unambiguous
2007 (44) SR of 3 split and ⁄ or Le Fort symptoms results in signs LRD III results and heterogeneity in study design, the
case–control I osteotomy Clinical and symptoms scientific evidence was insufficient to evaluate the
studies C1: No treatment examination effects that orthognathic surgery had on TMD.
280 patients R: Well-designed SR. Methodological weaknesses
of primary studies such as diagnostic criteria of
TMD and outcome measures in the study.
MANAGEMENT OF TMD
441
 442

Table 4. (Continued)

Study design,
T. LIST & S. AXELSSON

Authors, year, diagnosis, and Intervention (I) and Outcome Quality Authors’ (A) conclusions
reference no. of patients control (C) groups measures Results score Reviewers’ (R) comments

Al-Belasy FA Qualitative I1: Arthrocentesis Pain intensity Overall success AMSTAR 2 A: The majority of the reviewed publications were
2007 (43) SR of 19 Studies C1: Not specified Jaw mobility varied between LRD II-IV prospective case series with flawed methodology
(2 RCTs and 6 Clinical 60% -100%. and, despite the impression that arthrocentesis
case–control and examination No comparison may be beneficial for patients with TMJ closed
11 uncontrolled between I1 lock, there have been no good prospective
studies) and C1. randomised clinical trial confirm the efficacy
Anchored disc of the procedure.
phenomenon, R: The overall success rate was high from the
Disc displacement primary studies. The results are difficult to
with or without interpret because of methodological weaknesses
reduction, of primary studies such as diagnostic criteria of
capsulitis ⁄ TMD, outcome measures, missing analysis
synovitis. between Intervention and control treatment in
571 patients the studies. In addition, a majority of the studies
patients received complementary pharmacologic
or conservative treatment besides lavage. The
majority of the studies are LRD level IV, and
therefore, it not possible to draw any clear
conclusions from this SR.

AMSTAR, assessment of multiple systematic reviews; LRD, level of research design; RCT, randomised clinical trials; TMD, temporomandibular disorders.

ª 2010 Blackwell Publishing Ltd

 Table 5. Characteristics of systematic reviews (SRs) of behavioural therapy and multimodal treatments

Study design,
Authors, year, diagnosis, and no. Intervention (I) and Outcome Quality Authors’ (A) conclusions
reference of patients control (C) groups measures Results score Reviewers’ (R) comments

Crider AB SR and I1: Electromyographic Pain reduction Pain reduction and AMSTAR 4 A: Although limited in extent, the available

ª 2010 Blackwell Publishing Ltd

1999 (45) meta-analysis of biofeedback Clinical signs of clinical signs: I1 LRD II-IV data support the efficacy of EMG biofeedback
13 RCTs and C1: Active control dysfunction better than treatments for TMD.
un-controlled C2: No treatment or Global assessment C1 and C2 R: Methodological weaknesses of primary
studies placebo studies such as heterogeneous patient
TMD material, outcome measures not clearly
patients* defined, and several studies were not RCTs.
Jedel E Qualitative I1: Biofeedback Self-reported No evidence of an AMSTAR 3 A: The studies were heterogeneous with low
2003 (30) SR of 7 RCTs I2: Acupuncture symptoms effect for any LRD II quality, and therefore, it is not possible to
TMD I3: TENS Pain intensity treatment mode draw any conclusions.
379 patients C1: No treatment Clinical R: Limitations in the SR methodology and
C2: Other treatment examination therefore difficult to draw any conclusions
Crider AB Qualitative I1: Biofeedback Pain intensity I1 was superior to AMSTAR 3 A: Biofeedback training with adjunctive CBT
2005 (46) SR of 6 RCTs training Global C1 in one of two LRD II–III was reported to be an efficacious treatment for
TMD I2: Biofeedback improvement RCTs TMD; both biofeedback training as the sole
449 patients training + CBT Limitation in jaw I2 was significantly intervention and biofeedback-assisted
I3: Biofeedback- function better than relaxation training are probably efficacious
assisted relaxation Depression C2 and I4 treatments.
training Clinical I3 was better R: Small studies, no long-term follow-up.
I4: Alternative examination than C2 Methodological limitations in SR.
treatment I3 was better than
C1: Sham treatment I4 in one of two
C2: No treatment RCTs
SBU (23) Qualitative I1: CBT Pan intensity I1 was better than AMSTAR 6 A: Behavioural treatment such as biofeedback
SR of 2 SRs I2: Biofeedback Daily activities C1 and C3 LRD I–II and CBT have better effect than no treatment.
and 6 RCTs I3: Education Depression I2 was better R: Reviewers and authors are identical persons
TMD I4: Education + home Sleep quality than C2
XX patients instruction I3 was similar to I4
C1: Brief information
C2. No treatment
C3: Conventional
treatment
MANAGEMENT OF TMD
443
 444

Table 5. (Continued)

Study design,
Authors, year, diagnosis, and no. Intervention (I) and Outcome Quality Authors’ (A) conclusions
reference of patients control (C) groups measures Results score Reviewers’ (R) comments

McNeely M Qualitative I1: CBT Pain reduction Pain reduction: I1 AMSTAR 7 A: Programmes involving relaxation techniques
T. LIST & S. AXELSSON

2006 (33) SR of 4 RCTs I2: Biofeedback Jaw mobility better than C2 LRD II and biofeedback, electromyographic training,
TMD I3: Relaxation I2 similar effect and proprioceptive re-education may be more
207 patients C1: Occlusal splint to C1 effective than placebo treatment or occlusal
C2: No treatment I2 and I3 similar splints.
effect R: Studies had small numbers of participants
and outcome measures were poorly defined,
so it is difficult to draw any conclusions.
Türp J Qualitative I1: Simple treatment Pain intensity Disc displacement AMSTAR 4 A: Current research suggests that individuals
2007 (47) SR of 11 RCTs I2: Multimodal Graded Chronic without reduction LRD II without major psychological symptoms do not
TMD: treatment Pain Scale with pain: I1 = I2. require more than simple therapy. In contrast,
Disc displacement Analgesic TMD pain, without patients with major psychological
without reduction, consumption major involvement need multimodal,
with pain Psychologic status psychological interdisciplinary therapeutic strategies.
TMD pain, without Pain threshold symptoms: I1 = I2 R: Methodological weaknesses of primary
major TMD pain, with studies such as diagnostic criteria of TMD,
psychological major poor description of how the treatment was
symptoms psychological conducted, and outcome measures in the
TMD pain, with symptoms: study
major I2 better than I1.
psychological
symptoms
895 patients

AMSTAR, assessment of multiple systematic reviews; CBT, cognitive behavioural therapy; LRD, level of research design; RCT, randomised clinical trials; TMD, temporomandibular
disorders.
*Number of patients not reported.

ª 2010 Blackwell Publishing Ltd

 MANAGEMENT OF TMD 445

inclusion criteria because of limitations in the quality
of the studies.
Three SRs evaluated surgical treatment of the TMJ in
patients with disc displacements (41–43) and one SR
orthognathic surgery in patients with TMD (44)
(Table 4). In patients with disc displacements with
reduction, one SR reported similar treatment effects for
arthrocentesis, arthroscopy, and discectomy. In patients
with disc displacement without reduction, one SR
reported similar effect for arthrocentesis, arthroscopy,
and physical therapy. One SR reported overall high
success rate for arthrocentesis but made no comparison
of arthrocentesis with other interventions. In patients
with TMD pain, one SR reported contradictory results
following orthognathic surgery. But all SRs included in
these SRs had low levels of evidence.
Five SRs evaluated the treatment of various behavio-
ural therapies in patients with TMD (23, 30, 33, 45, 46),
and one SR analysed the effect of multimodal and
simple treatment in TMD (47) (Table 5). Biofeedback
was reported to be (i) better than active control or no
treatment (two SRs), (ii) better than no treatment
when administered in combination with cognitive
behavioural therapy (CBT) or relaxation (two SRs),
and (iii) similar in effect compared with relaxation
training (one SR). Cognitive behavioural therapy was
reported to be (i) better than conventional treatment
(two SRs) and (ii) better than no treatment (one SR).
Education at the clinic and education at the clinic
combined with home exercises produced similar results.
In patients with disc displacement without reduction
and pain and in patients with TMD pain without major
psychological symptoms, no difference in outcome was
found between simple treatment and multimodal
treatment (one SR). But in patients with TMD pain
and major psychological symptoms, multimodal treat-
ment was found to be better than simple treatment.
The overall inter-reliability agreement of the two
authors in assessing the quality of the SRs was 0Æ70 and
free-marginal kappa 0Æ67. Figure 2 presents percentage
of primary studies cited in one or more of the different
SRs for each treatment area.

Occlusal appliance Occlusal adjustment Bruxism

6 SR, 55 studies 4 SR, 12 studies 2 SR, 6 studies
5 4
4 8%
5% 3 2
9%
17% 33%
1
3
43%
18% 1
2 58% 1
17% 67%
2
25%

Acupuncture Physical treatment Pharmacologic treatment

6 SR, 10 studies 4 SR, 26 studies 7 SR, 60 studies
4 4
3 8%
8% 3 13%
5 1 5%
40% 40% 2
12% 2
20% 1
1 62%
4 3 2 72%
10% 0% 10%

TMJ and maxillofacial surgery Behavioral therapy and

4 SR, 54 studies Multimodadal treatment
6 SR, 30 studies
5
3 4
2 7%
4% 7%
15% 1 1 Reference cited number of times
3 36% 2
Fig. 2. Percentage of same primary 17% 3
4
study cited in one or more of the
5
different systematic reviews, for each 1 2
81%
treatment area. 33%

ª 2010 Blackwell Publishing Ltd

446 T. LIST & S. AXELSSON
Discussion
Systematic reviews are a synthesis and critical assess-
ment of primary studies, or even other SRs, and they
play an important role in evidence-based decision
making. The SR has the benefit that it provides a
systematic overview of what has been published on a
specific issue and what current trends are, such as the
effect of various treatment modes in the management
of TMD. Because the conclusions in our study are based
on several SRs published by independent researchers,
findings are well supported.
Because many researchers and practitioners may not
have time to read SRs – much less all the primary
studies on which these SRs are based – a systematic
review of these SRs may (i) give a valuable, time-saving
overview of a specific issue, (ii) provide a pool of
references of SRs and of primary studies that have been
quality assessed, and (iii) serve as a useful tool for
giving undergraduate students an understanding of
how to conduct systematic literature searches and
quality assessments of a topic.
Studies have indicated that it is difficult to change
competences and knowledge that a dentist learned in
undergraduate dental education (48). It is therefore
important to implement SRs in undergraduate teaching
so that students understand evidence-based medicine
(49).
The results of the SRs are impacted by the quality of
the primary studies included in the review. However, the
methodological quality of the SR should not be affected
by flaws in the primary studies, as long as these flaws are
reported, commented on and sufficiently regarded in the
conclusions. One SR resulted in no included studies but
it was included because it reflected our second aim, to
assess the methodological quality of SRs (38).
In AMSTAR, one key item is whether the scientific
quality of the included SRs was assessed and docu-
mented The most commonly used quality assessment
tool used in the SRs was the Jadad score (50). Grades of
Recommendation Assessment, Development and Eval-
uation (GRADE) is a recently developed tool for
assessing the synthesised evidence for specified out-
come measures. As one of the aims of this study was to
detect trends in the evidence base regarding the overall
effect of different methods for treating TMD, we
decided that introducing GRADE in this context would
give too detailed information for this purpose (51).
A concern in this SR was that 30% of the SRs included
case series, that is, no control group, which limits the
conclusions of the SR. Several leading scientific journals
have made a united effort to provide guidelines for how
a clinical trial should be conducted in the CONSORT
statement (52). These guidelines can be used as a
checklist for avoiding pitfalls in conducting clinical
trials.
Our SR found inconsistencies in several SRs con-
cerning diagnosis. Temporomandibular disorders were
inconsistently defined in the primary studies but, more
importantly, also in the SRs reviewed here. Similar
results were reported for radiographic findings in TMD
(53). The classification that has been found to provide
acceptable reliability and validity and is commonly used
in TMD research is the research diagnostic criteria for
temporomandibular disorders (RDC ⁄ TMD) (54, 55).
The future challenge is to convince researchers to use
the diagnostic system with the best evidence – accept-
able sensitivity and specificity (56).
In this SR, a large variety of outcome measures were
used, and instrument reliability and validity were often
not described. It is essential that evidence be based on
measures that are important indicators of patient health
and that are not surrogate endpoints. For example, pain
reduction and daily functioning are more important
aspects in an evaluation than clicking sounds, deviation
in mm, or tenderness on palpation. Clinical trials and
SRs must begin meeting international standards, and
clinical scientists must begin designing and reporting
such trials. To create a core of common outcome
measures in the field of pain, Dworkin et al. and a
consensus group undertook to put together the
Initiative on Methods, Measurements, and Pain Assess-
ment in Clinical Trials (IMMPACT) (57). IMMPACT
recommendations include the following domains:
pain, physical functioning, emotional functioning,
participant rating of improvement and satisfaction
with treatment, symptoms and adverse events, and
participant disposition.
It is also important that (i) these measures are
expressed in clinically relevant terms such as numbers
of responders and non-responders to treatment, for
example, number of patients with >30% pain reduction
and (ii) data are presented in a transparent manner, so
that mathematical syntheses or meta-analyses of the
results may be conducted. In this review, only a few of
the SRs had conducted a meta-analysis because of
heterogeneity of the original data or methodological
differences such as variability of measures. In future,
ª 2010 Blackwell Publishing Ltd

the ability to synthesise the results of several primary
studies would allow more accurate assessment of
treatment efficacy and treatment effectiveness. This
approach would allow the continuous update of RCTs
in meta-analyses, which would (i) limit the number of
qualitative SRs and (ii) allow more accurate, overall
assessment of treatment result.
Quality assessment
In this SR, AMSTAR scores ranged from 2 to 11. But it is
important to point out that item scores are not equal in
weight; for example, characteristics of included SRs and
conflict of interest statement have different weights.
With its focus on study design, the LRD was used to
supplement AMSTAR in SR quality assessment. Use of
both instruments was essential to gain perspective on
SR quality.
The number of SRs evaluated in the various treatment
groups ranged from 10 for occlusal splint, occlusal
adjustment, and bruxism to 4 for TMJ and maxillofacial
surgery. In each treatment area, a small number of well-
designed primary studies overlapped and were cited in
several of the SRs that covered that area. But 40–80% of
the primary studies did not overlap between different
SRs and were only cited once (Fig. 2). Variations
between the SRs in aims, inclusion criteria, and time of
data collection may explain this lack of overlap. Despite
these differences, conclusions drawn in several of the
SRs for a specific treatment form had similar evidence.
Thus, synthesising evidence from several SRs can also be
a tool for validation of this kind of meta-research.
A general impression from this study was that
strength of evidence of an SR was weak if the SR (i)
had a low AMSTAR score (e.g. <5) and (ii) was based
upon non-randomised studies. In SRs where AMSTAR
scores ranged between 5 and 10, the results were
similar, regardless of quality. It has been emphasised
that a clinical trial should follow strict rules; likewise an
SR should be conducted in a standardised manner.
Occlusal appliances, occlusal adjustment, and bruxism
Several of the SRs concluded that management of TMD
with a stabilisation splint worn at night is likely to lead
to short-term improvement when compared with no
treatment, but is inconclusive compared with placebo
(non-occluding palatinal splint). In the short term,
stabilisation splints were equally effective in reducing
ª 2010 Blackwell Publishing Ltd
MANAGEMENT OF TMD 447
TMD pain compared to other treatment modalities such
as physical medicine, behavioural medicine, and acu-
puncture treatment. Overall, documentation on the
long-term pain-relieving effect of occlusal appliances is
limited, as it is for patient compliance in occlusal
appliance treatment. Few SRs reported data on adverse
events related to the use of occlusal appliances. The
major concern with adverse events has been related to
partial non-occluding splints such as the NTI, where the
design of the splint may contribute to tooth pain and
occlusal changes (24).
One SR evaluated the use of splints in bruxism,
which was assessed as number of bruxism episodes per
hour [electromyographic (EMG) activity] and episodes
with grinding noises. The SR found no significant
differences between occlusal splints, no treatment, and
palatinal splints. Small sample size was one explanation
for the lack of significance between outcomes which
the authors of the SR emphasised (27). It should also be
emphasised that some primary studies, particularly
those that use polysomnographic registration, are
technically very difficult to conduct on large patient
samples. One SR examined tooth attrition related to
bruxism, and based on two small studies, found that
occlusal appliances retarded wear.
No SR found evidence that occlusal adjustments are
more or less effective than placebo in the treatment of
TMD pain. All SRs were restrictive in recommending
the use of occlusal adjustments for treatment for TMD
pain, especially because this therapy is non-reversible.
Physical therapy (acupuncture, TENS, exercise, and
mobilisation)
Most SRs found evidence that acupuncture is better
than no treatment and comparable to other forms of
conservative treatment. But because of methodological
shortcomings, MacPherson et al. recommended specific
guidelines to improve the quality of intervention
primary studies; before effectiveness of acupuncture
can be determined more primary studies are needed
(58). Few SRs reported any adverse events or side-
effects from acupuncture treatment. In treatment of
patients with TMD, side-effects seem to be rare or
complications only minor (59). MacPherson et al. came
to similar conclusions concerning use of acupuncture in
general for pain treatment in a large population group
(60). Major adverse events are very rare, but because
some have occurred following acupuncture treatment,
448 T. LIST & S. AXELSSON
it is strongly recommended that those practicing acu-
puncture have good theoretical knowledge and training
in the therapy (61).
Next to information, patient education, and occlusal
appliance, jaw exercises are a common form of TMD
treatment (62). One SR found active exercise and
postural training to be effective in treatment of TMD
pain but no evidence for the effectiveness of various
electrical modalities. Overall, few primary studies have
been published and more research is needed to estab-
lish the efficacy of the various physical treatment
modalities, including acupuncture.
Pharmacologic treatment
Several SRs indicated that analgesics, antidepressants,
diazepam, hyaluronate, and glycocorticoid may be
effective in TMD pain. Few primary studies were well
designed with a relevant follow-up time, so the main
conclusion in the SR was that results were heteroge-
neous, and no conclusions could be drawn. But it is
important to differentiate between lack of evidence and
evidence for lack of effect. Because of current limitations in
knowledge of pharmacologic effects on TMD pain, only
comparisons between similar pain conditions such as
backache or tension-type headache can be made. In
several chronic pain conditions, drugs such as analge-
sics, opioids, antidepressants, and anti-epileptics have
been found to be effective in relieving pain (63); these
drugs would probably be effective in TMD pain.
Important endpoints such as numbers needed to treat
(NNT) and numbers needed to harm (NNH) were rare
in these primary studies, despite being recommended
for use in pharmacologic treatment studies because
they are easy to understand and provide a clinically
relevant measure of the success rate and rate of harm of
an intervention (63).
The SR on pharmacologic treatment reported minor
adverse events. Because there is currently no criterion
standard in the pharmacological treatment of chronic
oro-facial pain, the positive effects of drugs must be
weighed against possible adverse and toxic effects, and
risk of dependency.
TMJ and maxillofacial surgery
The SRs of surgical treatment of TMD concluded that
arthroscopic surgery, arthrocentesis, and physical ther-
apy affected mandibular movement, reduction in pain
intensity, and mandibular functioning to the same
degree. Success rates were often high, independent of
treatment mode. The effect of maxillofacial surgery on
TMD pain is unclear. Many of the RCTs included in the
SRs had low quality scores, and outcome measures were
often coarse, which made data difficult to interpret.
One important aspect in the evaluation of invasive
TMJ interventions versus conservative treatment is the
patient groups. In most TMJ surgery primary studies,
the inclusion criterion for entering the study was
that the patient had been refractory to conserva-
tive treatment for 6 months, while the inclusion
criterion for conservative treatment was pain duration
of 6 months. So a comparison between the groups is
inconclusive.
More well-designed primary studies with proper
evidence-based standards are needed to identify the
patients who are most suited for surgical interventions.
Behavioural therapy and multimodal treatment
All SRs of behavioural therapy concluded that this type
of treatment was effective in treating TMD pain. The
treatment modalities included education, biofeedback,
relaxation training, stress management, and CBT.
Treatment modes were often combined, for example,
biofeedback and relaxation could be compared with
biofeedback and CBT, making it difficult to determine
which part of the treatment was most important.
Several primary studies indicated that behavioural
therapy was as effective as other forms of conservative
TMD treatment.
One SR reported that most patients with TMD
without psychological involvement benefited from
simple treatments. Patients with TMD pain and major
psychological disturbances were in need of a combined
therapeutic approach. This emphasises the need for
combining the clinical examination with a behavioural
assessment to be able to direct the patient to the proper
mode of treatment. The RDC ⁄ TMD is a classification
system that uses a clinical (axis I) and a psychosocial
assessment (axis II) to gain a more complete picture of
the patient; Garofalo and Wesley recommended this
approach – use of a dual axis – in chronic pain
assessment (64).
Evidence-based medicine is defined as the integration
of best research evidence with clinical expertise and
patient values. In future, we need to expand our
understanding of how to:
ª 2010 Blackwell Publishing Ltd

1 Tailor treatment for the individual patient, so they
benefit from the best treatment for them.
2 Transfer research findings to the practicing dentist,
because they are the main caregiver of patients with
TMD.
3 Balance our judgment of best research evidence with
clinical expertise in the choice of treatment.
4 Gain more information about patient preferences
and values and what impact this has on treatment
outcome.
Conclusions
There is some evidence that occlusal appliances,
acupuncture, behavioural therapy, jaw exercises, pos-
tural training, and some pharmacological treatments
can be effective in alleviating pain in patients with
TMD. Evidence is insufficient for the effect of electro-
physical modalities and surgery. Occlusal adjustment
seems to have no effect according to the available
evidence. One limitation of most of the SRs reviewed
was that the considerable variation in methodology
between the primary studies made definitive conclu-
sions impossible.
Acknowledgments
To the participants of the 2009 Colloquium on Oral
Rehabilitation in Sienna, who provided valuable feed-
back on the manuscript.
Declaration of interests
Dr Axelsson is a staff member and a project director at
the Swedish Council on Technology Assessment in
Health Care (SBU).
Authors’ contributions
Dr List conceived the project, developed the protocol,
conducted searches, and prepared the manuscript. Both
authors undertook data collection and extraction. Dr
Axelsson contributed to manuscript preparation.
References
1. Drangsholt M. Temporomandibular pain. In: Crombie IK,
Croft PR, Linton SJ, LeResche L, Von Korff M, eds. Epidemi-
ology of pain. Seattle (WA): IASP Press; 1999:203–233.
ª 2010 Blackwell Publishing Ltd
MANAGEMENT OF TMD 449
2. Macfarlane TV, Glenny AM, Worthington HV. Systematic
review of population-based epidemiological studies of oro-
facial pain. J Dent. 2001;29:451–467.
3. Dworkin SF, Von Korff M, LeResche L. Multiple pains and
psychiatric disturbance. An epidemiologic investigation. Arch
Gen Psychiatry. 1990;47:239–244.
4. List T, Dworkin SF. Comparing TMD diagnoses and clinical
findings at Swedish and US TMD centers using research
diagnostic criteria for temporomandibular disorders. J Orofac
Pain. 1996;10:240–253.
5. John MT, Reissmann DR, Schierz O, Wassell RW. Oral health-
related quality of life in patients with temporomandibular
disorders. J Orofac Pain. 2007;21:46–54.
6. Turp JC, Motschall E, Schindler HJ, Heydecke G. In patients
with temporomandibular disorders, do particular interven-
tions influence oral health-related quality of life? A qualita-
tive systematic review of the literature. Clin Oral Implants
Res. 2007;18(Suppl 3):127–137.
7. Davidoff F, Haynes B, Sackett D, Smith R. Evidence based
medicine. BMJ. 1995;310:1085–1086.
8. Shea BJ, Grimshaw JM, Wells GA, Boers M, Andersson N,
Hamel C et al. Development of AMSTAR: a measurement tool
to assess the methodological quality of systematic reviews.
BMC Med Res Methodol. 2007;7:10.
9. Shea BJ, Hamel C, Wells GA, Bouter LM, Kristjansson E,
Grimshaw J et al. AMSTAR is a reliable and valid measure-
ment tool to assess the methodological quality of systematic
reviews. J Clin Epidemiol. 2009;62:1013–1020.
10. Antes G. [Evidence-based medicine]. Internist (Berl).
1998;39:899–908.
11. Forssell H, Kalso E, Koskela P, Vehmanen R, Puukka P,
Alanen P. Occlusal treatments in temporomandibular disor-
ders: a qualitative systematic review of randomized controlled
trials. Pain. 1999;83:549–560.
12. Koh H, Robinson PG. Occlusal adjustment for treating and
preventing temporomandibular joint disorders. J Oral Reha-
bil. 2004;31:287–292.
13. Al-Ani Z, Gray RJ, Davies SJ, Sloan P, Glenny AM. Stabiliza-
tion splint therapy for the treatment of temporomandibular
myofascial pain: a systematic review. J Dent Educ.
2005;69:1242–1250.
14. Ismail AI, Bader JD. Evidence-based dentistry in clinical
practice. J Am Dent Assoc. 2004;135:78–83.
15. Mohlin B, Axelsson S, Paulin G, Pietila T, Bondemark L,
Brattstrom V et al. TMD in relation to malocclusion and
orthodontic treatment. Angle Orthod. 2007;77:542–
548.
16. Kim MR, Graber TM, Viana MA. Orthodontics and temporo-
mandibular disorder: a meta-analysis. Am J Orthod Dentofa-
cial Orthop. 2002;121:438–446.
17. Rinchuse DJ, McMinn JT. Summary of evidence-based
systematic reviews of temporomandibular disorders. Am J
Orthod Dentofacial Orthop. 2006;130:715–720.
18. Bessa-Nogueira RV, Vasconcelos BC, Niederman R. The
methodological quality of systematic reviews comparing
temporomandibular joint disorder surgical and non-surgical
treatment. BMC Oral Health. 2008;8:27.
450 T. LIST & S. AXELSSON
19. Al-Ani MZ, Davies SJ, Gray RJ, Sloan P, Glenny AM.
Stabilisation splint therapy for temporomandibular pain
dysfunction syndrome. Cochrane Database Syst Rev. 2004;1:
CD002778.
20. Forssell H, Kalso E. Application of principles of evidence-based
medicine to occlusal treatment for temporomandibular disor-
ders: are there lessons to be learned? J Orofac Pain.
2004;18:9–22; discussion 23–32.
21. Turp JC, Komine F, Hugger A. Efficacy of stabilization splints
for the management of patients with masticatory muscle pain:
a qualitative systematic review. Clin Oral Investig. 2004;8:
179–195.
22. Fricton J. Current evidence providing clarity in management
of temporomandibular disorders: summary of a systematic
review of randomized clinical trials for intra-oral appliances
and occlusal therapies. J Evid Based Dent Pract. 2006;6:48–52.
23. SBU. Methods of treating chronic pain. In: Axelsson S, Boivie J,
Eckerlund I, Gerdle B, Johansson E, Kristiansson M, List T,
Lundberg B, Mannheimer C, Marke¢ L-Å, Olsson G, Segerdahl
M, Sjöström B, Söderlund A, Willman A, eds. SBU-report no
177:1+2. Stockholm: The Swedish Council on Technology
Assessment in Health and Care (SBU); 2006:299–410, 437–443.
24. Stapelmann H, Turp JC. The NTI-tss device for the therapy of
bruxism, temporomandibular disorders, and headache – where
do we stand? A qualitative systematic review of the literature.
BMC Oral Health. 2008;8:22.
25. Santacatterina A, Paoli M, Peretta R, Bambace A, Beltrame A.
A comparison between horizontal splint and repositioning
splint in the treatment of ‘disc dislocation with reduction’.
Literature meta-analysis. J Oral Rehabil. 1998;25:81–88.
26. Koh H, Robinson PG. Occlusal adjustment for treating and
preventing temporomandibular joint disorders. Cochrane
Database Syst Rev. 2003;1:CD003812.
27. Macedo CR, Silva AB, Machado MA, Saconato H, Prado GF.
Occlusal splints for treating sleep bruxism (tooth grinding).
Cochrane Database Syst Rev. 2007;4:CD005514.
28. van ‘tSpijker A, Kreulen CM, Creugers NH. Attrition, occlu-
sion, (dys)function, and intervention: a systematic review.
Clin Oral Implants Res. 2007;18(Suppl 3):117–126.
29. Ernst E, White AR. Acupuncture as a treatment for tempo-
romandibular joint dysfunction: a systematic review of ran-
domized trials. Arch Otolaryngol Head Neck Surg.
1999;125:269–272.
30. Jedel E, Carlsson J. Biofeedback, acupuncture and transcuta-
neous electric nerve stimulation in the management of
temporomandibular disorders: a systematic review. Phys Ther
Rev. 2003;8:217–223.
31. Rosted P. The use of acupuncture in dentistry: a review of the
scientific validity of published papers. Oral Dis. 1998;4:
100–104.
32. Fink M, Rosted P, Bernateck M, Stiesch-Scholz M, Karst M.
Acupuncture in the treatment of painful dysfunction of the
temporomandibular joint – a review of the literature. Forsch
Komplementmed. 2006;13:109–115.
33. McNeely ML, Armijo Olivo S, Magee DJ. A systematic review
of the effectiveness of physical therapy interventions for
temporomandibular disorders. Phys Ther. 2006;86:710–725.
34. Medlicott MS, Harris SR. A systematic review of the effec-
tiveness of exercise, manual therapy, electrotherapy, relaxa-
tion training, and biofeedback in the management of
temporomandibular disorder. Phys Ther. 2006;86:955–973.
35. Sommer C. [Pharmacotherapy of orofacial pain]. Schmerz.
2002;16:381–388.
36. List T, Axelsson S, Leijon G. Pharmacologic interventions in
the treatment of temporomandibular disorders, atypical facial
pain, and burning mouth syndrome. A qualitative systematic
review. J Orofac Pain. 2003;17:301–310.
37. Shi Z, Guo C, Awad M. Hyaluronate for temporomandibular
joint disorders. Cochrane Database Syst Rev. 2003;1:
CD002970.
38. Al-Muharraqi MA, Fedorowicz Z, Al Bareeq J, Al Bareeq R,
Nasser M. Botulinum toxin for masseter hypertrophy. Coch-
rane Database Syst Rev. 2009;1:CD007510.
39. Cascos-Romero J, Vazquez-Delgado E, Vazquez-Rodriguez E,
Gay-Escoda C. The use of tricyclic antidepressants in the
treatment of temporomandibular joint disorders: systematic
review of the literature of the last 20 years. Med Oral Patol
Oral Cir Bucal. 2009;14:E3–E7.
40. Ihde SK, Konstantinovic VS. The therapeutic use of botu-
linum toxin in cervical and maxillofacial conditions: an
evidence-based review. Oral Surg Oral Med Oral Pathol Oral
Radiol Endod. 2007;104:e1–e11.
41. Reston JT, Turkelson CM. Meta-analysis of surgical treatments
for temporomandibular articular disorders: a reply to the
discussants. J Oral Maxillofac Surg. 2003;61:737–738.
42. Kropmans TJ, Dijkstra PU, Stegenga B, de Bont LG. Ther-
apeutic outcome assessment in permanent temporomandib-
ular joint disc displacement. J Oral Rehabil. 1999;26:
357–363.
43. Al-Belasy FA, Dolwick MF. Arthrocentesis for the treatment of
temporomandibular joint closed lock: a review article. Int J
Oral Maxillofac Surg. 2007;36:773–782.
44. Abrahamsson C, Ekberg E, Henrikson T, Bondemark L.
Alterations of temporomandibular disorders before and after
orthognathic surgery: a systematic review. Angle Orthod.
2007;77:729–734.
45. Crider AB, Glaros AG. A meta-analysis of EMG biofeedback
treatment of temporomandibular disorders. J Orofac Pain.
1999;13:29–37.
46. Crider A, Glaros AG, Gevirtz RN. Efficacy of biofeedback-
based treatments for temporomandibular disorders. Appl
Psychophysiol Biofeedback. 2005;30:333–345.
47. Turp JC, Jokstad A, Motschall E, Schindler HJ, Windecker-
Getaz I, Ettlin DA. Is there a superiority of multimodal as
opposed to simple therapy in patients with temporomandib-
ular disorders? A qualitative systematic review of the litera-
ture. Clin Oral Implants Res. 2007;18(Suppl 3):138–150.
48. Greenhalgh T, Robert G, Macfarlane F, Bate P, Kyriakidou
O. Diffusion of innovations in service organizations: system-
atic review and recommendations. Milbank Q. 2004;82:
581–629.
49. Sackett D, Straus S, Richardson WS, Rosenberg W, Haynes R.
Evidence-based medicine. How to practice and teach EBM.
Edinburgh: Churchill Livingstone; 2000.
ª 2010 Blackwell Publishing Ltd

50. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ,
Gavaghan DJ et al. Assessing the quality of reports of
randomized clinical trials: is blinding necessary? Control Clin
Trials. 1996;17:1–12.
51. Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S
et al. Grading quality of evidence and strength of recommen-
dations. BMJ. 2004;328:1490.
52. Moher D, Schulz KF, Altman D. The CONSORT statement:
revised recommendations for improving the quality of reports
of parallel-group randomized trials. JAMA. 2001;285:
1987–1991.
53. Koh KJ, List T, Petersson A, Rohlin M. Relationship between
clinical and magnetic resonance imaging diagnoses and
findings in degenerative and inflammatory temporomandib-
ular joint diseases: a systematic literature review. J Orofac
Pain. 2009;23:123–139.
54. Dworkin SF, LeResche L. Research diagnostic criteria for
temporomandibular disorders: review, criteria, examinations
and specifications, critique. J Craniomandib Disord. 1992;
6:301–355.
55. Schiffman E, Truelove E, Ohrbach R, Anderson GC, John MT,
List T, Look J. The research diagnostic criteria for tempor-
omandibular disorders. I: overview and methodology for
assessment of validity. J Orofacial Pain. 2010;24:7–24.
56. Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM, Kleijnen J.
The development of QUADAS: a tool for the quality assess-
ment of studies of diagnostic accuracy included in systematic
reviews. BMC Med Res Methodol. 2003;3:25.
57. Dworkin RH, Turk DC, Farrar JT, Haythornthwaite JA, Jensen
MP, Katz NP et al. Core outcome measures for chronic pain
clinical trials: IMMPACT recommendations. Pain. 2005;113:
9–19.
ª 2010 Blackwell Publishing Ltd
MANAGEMENT OF TMD 451
58. MacPherson H, White A, Cummings M, Jobst KA, Rose K,
Niemtzow RC. Standards for Reporting Interventions in
Controlled Trials of Acupuncture: the STRICTA recommen-
dations. J Altern Complement Med. 2002;8:85–89.
59. List T, Helkimo M. Adverse events of acupuncture and
occlusal splint therapy in the treatment of craniomandibular
disorders. Cranio. 1992;10:318–324; discussion 324–6.
60. MacPherson H, Thomas K, Walters S, Fitter M. A prospective
survey of adverse events and treatment reactions following
34,000 consultations with professional acupuncturists. Acu-
punct Med. 2001;19:93–102.
61. Ernst E, White A. Life-threatening adverse reactions
after acupuncture? A systematic review. Pain. 1997;71:123–
126.
62. Tegelberg A, List T, Wahlund K, Wenneberg B. Temporo-
mandibular disorders in children and adolescents: a survey of
dentists’ attitudes, routine and experience. Swed Dent J.
2001;25:119–127.
63. McQuay H, Moore A. An evidence-based resource for pain
relief. Oxford: Oxford University Press; 1998.
64. Garofalo J, Wesley A. Research diagnostic criteria for tempo-
romandibular disorders: reflection pf the physical-psycholog-
ical interface. ASP Bull. 1997;3:4–16.
65. Jedel E, Carlsson J. Biofeedback, acupuncture and transcuta-
neous electric nerve stimulation in the management of
temporomandibular disorders. A systematic review. J Orofac
Pain. 2003;8:217–223.
Correspondence: Thomas List, DDS, Odont. dr, Department of
Stomatognathic Physiology, Faculty of Odontology, Malmö Univer-
sity, SE-205 06 Malmö, Sweden. E-mail: Thomas.list@mah.se