Intended for healthcare professionals

Clinical Review

Temporomandibular disorders

BMJ 2015; 350 doi: (Published 12 March 2015) Cite this as: BMJ 2015;350:h1154
  1. Justin Durham, NIHR clinician scientist, senior lecturer1,
  2. Toby R O Newton-John, senior lecturer, clinical psychology2,
  3. Joanna M Zakrzewska, professor3
  1. 1Centre for Oral Health Research & Institute of Health and Society, Newcastle University, Newcastle, UK
  2. 2Graduate School of Health, University of Technology, Sydney, NSW, Australia
  3. 3Facial Pain Unit, Eastman Dental Hospital, UCLH NHS Foundation Trust, London, UK
  1. Correspondence to: J Durham, Level 5—School of Dental Sciences, Newcastle University, Newcastle NE2 4BW, UK justin.durham{at}

The bottom line

  • TMDs represent a range of conditions with multifactorial complex pathophysiology, which are common up to the age of 40 years

  • TMDs can present as a continuous or episodic pain of varying intensity around and in the ears radiating up to the temple or down the mandible to the neck

  • Persistent (chronic) TMDs can be associated with other chronic pain conditions

  • Early management with education and counselling is highly effective

  • More complex patients or those with persistent TMDs require a biopsychosocial approach delivered by a multidisciplinary pain team

After odontogenic pain, temporomandibular disorders (TMDs) are one of the most common causes of pain in the mouth and face and also have the potential to produce persisting (chronic) pain.1 2 Chronic or persistent (myogenous) TMDs can be associated with other chronic pain conditions,3 including migraine, fibromyalgia, and widespread pain.4 5 6 7 8 They are also known to be comorbid with bruxism, depression, irritable bowel syndrome, and chronic fatigue.6 9 10 With or without these comorbidities, TMDs are recognised to have a considerable impact on quality of life.11 12 13 14 15 16 17

Sources and selection criteria

We conducted a systematic search of Medline (1966 to 5 August 2014), PROSPERO, and Cochrane libraries for systematic reviews relating to interventions for TMDs. For systematic reviews in Medline we used the search strategy suggested by BMJ Clinical Evidence search strategy. We searched PROSPERO and Cochrane libraries using their own search engines with the terms “temporomandibular”, “temporomandibular disorder”, “temporomandibular joint”, and “pain dysfunction syndrome”. The supplementary table summarises the most recent and relevant systematic reviews. It excludes older systematic reviews that duplicate more recent ones unless they are within a five year period and offer contradictory evidence. All Cochrane reviews identified are contained within the summary table.

Early diagnosis and explanation followed by management is likely to be key to improving prognosis and reducing the impact of this group of conditions on quality of life.11 18 The purpose of this review is to give non-specialists an overview of the diagnosis and management of TMDs.

What are TMDs?

Some of the signs and symptoms suggestive of TMDs were identified back in the 1930s by Costen, who incorrectly attributed them to changes in occlusion (bite of the teeth).19 Over the next 80 years the terminology has changed and been revised several times20 21 22 until the collective term temporomandibular disorders has been accepted.23 24 25 26 The most widely accepted definition is “a group of musculoskeletal . . . conditions that involve the temporomandibular joints (TMJs), the masticatory muscles and all associated tissues.”27 A new expanded taxonomy is now available defining the various subtypes of disorder or disease incorporated in this group.28

How common are TMDs and who gets them?

Data from a US large and ongoing prospective cohort study (OPPERA)29 suggest that the incidence of first onset painful TMDs is 3-4% per annum. Among the population studied (adults aged 18-44) for first onset of painful TMDs there seemed to be only a slight preponderance for women to be affected more than men (imputed hazard ratio 1.3, P>0.05) and an increasing incidence with age: 18-24 age band 2.5%, 25-34 age band 3.7%, and 35-44 age band 4.5%.30 The probability of first onset TMDs is also strongly associated with concurrent headaches and body pain.31 32 The prevalence of painful TMDs in adolescents is still a matter of debate,33 with some estimates reaching 25%, but in developed countries it is estimated to be between 2% and 6%.34 35 36 In adults aged more than 45 the prevalence of painful TMDs is likely to lie in the range of 2-7%,37 although estimates vary and those aged over 45 are more likely to display objective signs of TMDs but be asymptomatic.38

Among all incident cases in the OPPERA study, pain occurred as a singular episode in 12%, recurrent episode in 65%, and persistent episode in 19%. Pain was most commonly reported in both the joint and the muscles of mastication (arthralgia with myalgia: 73% of the incident cases); the next most common presentation was myalgia alone (23% of incident cases).30 OPPERA and other US data suggest there is a disproportionate increase in healthcare utilisation with TMDs.32 39 The reasons behind this are not fully explained, but ongoing research40 in the United Kingdom seeking to determine the effectiveness and efficacy of care pathways for orofacial pain ( may help to explain this from a UK context in due course.

What causes TMDs?

The causes and underlying pathophysiology of TMDs have been the subject of much debate and some controversy over the years.26 41 It is suggested that the cause is likely to be both multifactorial and biopsychosocial, consisting of initiating, predisposing, and perpetuating factors, for example, macro-trauma or micro-trauma, linked to a failure to heal appropriately because of the underlying psychosocial profile of the patient, failed treatment, or genotype.41 42 43 Evidence is now emerging for factors involved in the complex multifactorial pathophysiology of TMDs relating to both genotype and phenotype.44 45

In recent years major advances have occurred in the knowledge of the underlying pathophysiology of TMDs. It is suggested that women with an oestrogen receptor polymorphism have twice and three times the odds of developing a non-painful or painful TMD, respectively, compared with women without the polymorphism.46 Several studies suggest the pain of TMDs to be potentially mediated by several different mechanisms: central sensitisation, through neuroplasticity and calcitonin gene related peptide,47 48 49 50 51 and differing haplotypes of the gene encoding catecholamine-O-methyltransferase, which produce less activity of this enzyme.52 53 Some of the neural abnormalities detected in chronic TMDs are in the trigeminal and limbic systems,54 and cortical activity may be changed such that cognitive and emotional tasks are harder to perform.55

Further to this genetic polymorphisms in the folate and oxidative metabolic pathways may be associated with oxidative stress and be implicated in degenerative joint disease.56 57 58 Such disease has also been associated with a metalloproteinase gene polymorphism.59

Clearly there may be further advances in the coming years, but what is truly exciting are the potential advances in identifying targets for treatment, such as calcitonin gene related peptide and β adrenergic receptors. β Adrenergic receptors have already been shown in a proof of concept trial to be of benefit if targeted by treatment.60

What are the clinical features?

TMDs can present with any or all of the following signs and symptoms: pain in the temporomandibular joint or muscles of mastication (figures 1 and 2), which may radiate or refer to local or distant structures; clicking, popping, or crepitus of the temporomandibular joint on any of its movements with or without locking of the joint; headache limited to the temporal region61; and otalgia or tinnitus, or both in the absence of aural disease. The correlation between self reported severity of pain and pathological changes in the joints or muscles is poor.26


Fig 1 Surface anatomy of temporalis and masseter muscles. Adapted from Standring 2008113


Fig 2 Surface anatomy of temporomandibular joint. Mouth closed (left) and mouth open (right). Adapted from Drake et al 2005114

The table summarises the 12 most common TMDs and the most salient points of their history and examination findings. Some patients present principally with a self reported painless clicking (disc displacement with reduction) and are mistakenly concerned that this indicates damage. Others report episodic painful (closed) locking, an inability to open the mouth wide (disc displacement with reduction with intermittent limited opening), persistent closed lock (disc displacement without reduction with limited opening), or repeated dislocation of the temporomandibular joint, which are all clearly of functional importance.

Summary of salient history and examination findings for 12 most common (numbers in parentheses) temporomandibular disorders62

View this table:

How are TMDs diagnosed?

Over the past two decades progress in revising and validating the diagnostic criteria for TMDs has been considerable.62 63 64 65 66 67 68 69 70 71 72 73 74 In their validated and revised form these criteria, originally known as the RDC/TMD (Research Diagnostic Criteria for TMD),62 are now known as the DC/TMD (Diagnostic Criteria for TMD).75 They provide a standard and operationalised manner in which to examine the temporomandibular joint and its associated structures (axis 1) physically and to screen for psychosocial comorbidity (axis 2).

A simple method to screen for painful TMDs (as a physical entity) is to use a six item self complete questionnaire developed in tandem with the Diagnostic Criteria for TMD.76 When used in a facial pain clinic this method has 99% sensitivity and 98% specificity for painful TMDs and may be of utility in busy general medical practices. The questionnaire is freely available on the RDC/TMD consortium’s website ( A score of 3 or more is indicative of a positive screening for TMDs. Alternatively the questions posed could be incorporated into a standard clinical history for orofacial pain.

Axis 1

The physical axis of the DC/TMD, axis 1, defines 38 different types of TMDs of which 12 types present most commonly (table).75 As this is a summary table the list is not exhaustive. We believe that the easiest way to recall these 12 separate TMDs is to subdivide them into four main categories or groups: myalgia, arthralgia, intra-articular disorders, and headache attributable to TMDs. Full details on DC/TMD diagnostic algorithms and their sensitivity and specificity (where determined) are available in an open access publication75 and on the RDC/TMD consortium’s website (table).

The concept of eliciting pain that is “familiar” by provocation underpins the new revised and validated diagnostic criteria for TMD.75 Familiar pain is the process of recreating the patients’ symptoms during an examination and confirming with them that this is what they experience. The following features on examination should raise suspicion of, but do not formally confirm or exclude, a diagnosis of painful TMD: provocation of familiar pain through palpation of masseter and or temporalis (fig 1), through palpation of the temporomandibular joint (fig 2), or by putting the joint through its normal range of movements.

Of note in the diagnostic algorithm of the DC/TMD is the rationalised, limited, use of either plain or computed imaging (employing ionising radiation), which is outlined on the consortium’s website (table). Generally, imaging is only performed if clinical examination indicates a functional problem. Great care should be taken when explaining the results of any imaging to patients as the use of words such as “degenerative” may cause great concern and consequently affect patients’ behaviour.

Axis 2

Axis 2 of the DC/TMD can be used in a shortened form (a brief screening axis for psychosocial comorbidity) or in its full form (a comprehensive axis for psychosocial comorbidity). The instruments comprising axis 2 are available in full on the consortium’s website. Axis 2’s focus on psychosocial comorbidity is because this is known to impact on prognosis and treatment outcome.5 45 77

What are the red flags?

Box 1 outlines some of the red flags that may mimic a TMD,78 79 80 81 although this is not an exhaustive list. These red flags can act as signposts for conditions to consider in differential diagnoses.

Box 1 Red flags that may mimic TMDs

  • History of malignancy—potential for a new primary, recurrence, or metastases

  • Presence of lymphadenopathy or neck masses—consider a neoplastic, infective, or autoimmune cause

  • Sensory or motor function changes (specifically focusing on cranial nerves V, VII, and VIII)—consider intracranial causes, or malignancy affecting the nerve’s peripheral branches

  • Recurrent epistaxis, purulent nasal drainage, or anosmia—consider nasopharyngeal carcinoma or chronic sinusitis

  • Trismus especially if history of using paan or betel nut—consider oral malignancy as part of differential diagnoses and think about employing the “Trismus checklist”82

  • Unexplained pyrexia or weight loss—consider malignant tumours, immunosuppression, and infective causes as part of differential diagnoses

  • First episode in over 50s, unilateral headache accompanied by jaw claudication, and general malaise—consider temporal arteritis

  • Facial asymmetry or masses (uncommon in TMD unless there is masseteric hypertrophy)—consider neoplastic, infective, or inflammatory causes

  • Occlusal changes (bite of teeth changes) as determined by dentist that do not predate the start of the TMD—consider growth disturbance of condyle, neoplasia, rheumatoid arthritis, and traumatic causes as part of differential diagnoses

  • Ipsilateral objective change in hearing—consider acoustic neuroma, or other ear disease as part of differential diagnoses

  • Persisting or worsening symptoms despite treatment—consider a misdiagnosis or more complex case

  • History of recent head and neck trauma

  • Paroxysmal unilateral lancinating pain with or without autonomic features—more likely to be associated with trigeminal neuralgia or one of the trigeminal autonomic cephalagias

Specialties that will help in the case of relevant oral and maxillofacial red flag symptoms include oral medicine, oral and maxillofacial surgery, and oral surgery. Referral to ear, nose, and throat surgeons should be considered for worrisome signs in these areas and referral to a headache neurologist should be considered for upper face pain.

What makes TMDs more likely to become a chronic musculoskeletal pain?

Box 2 describes the factors that are more likely to be associated with chronic pain.83 84 85 86 87

Box 2 Factors associated with pain from TMDs becoming chronic

  • Females with myogenous TMD

  • Increasing age at presentation

  • Graded chronic pain scale88 score of 3 or 4

    • Higher pain intensity

    • Higher pain interference

  • More widespread non-specific symptoms (or higher number of pain locations)

  • Concurrent psychiatric diagnosis or mood disturbance (depression, anxiety, anger)

For patients with a noted affective problem in addition to the pain or where there is a preoccupation with somatic symptoms and experiences, referral to a clinical psychologist or psychiatrist is an important adjunctive pain management option.89

Convincing evidence shows that psychosocial factors are influential in the development of general chronic pain states90 and of TMD specifically.5 45 77

Importantly, research has shown that a chronic pain trajectory can be modified. People with acute TMDs identified as being at high risk of chronic TMDs have been shown to benefit considerably from early psychological intervention.91 This suggests that although mood disorder, pain catastrophising, somatic preoccupation, and poor coping skills may predispose people with acute TMDs to develop long term problems, intervening with appropriate psychological input before these problems become entrenched can be of great value.

What is the management?

The recurring theme in the systematic reviews examined for this review (see supplementary table) is that well designed pragmatic randomised controlled trials using standardised and patient based outcome measures in TMD are urgently needed.

An international consensus reached in 1995 and revised slightly in 201027 92 was that reversible, conservative treatments should generally form the first line intervention for TMDs. This is based on data that suggest between 75% and 90% of patients with TMDs will respond to these simpler, potentially less costly, and much less invasive techniques.93 After an initial diagnosis of acute TMD the first approach to management should be a careful explanation of the (provisional) diagnosis alongside education about the condition and its largely benign course. This can be supplemented by advice on appropriate, time limited, simple analgesic use and altered dietary consistency for acute exacerbations plus information on self management techniques.

Reversible and conservative treatments

Education—education about the condition and reassurance of its benign and usually non-progressive nature can be helpful. Lack of, or uncertainty in, diagnosis for patients with TMD has a major impact on their quality of life and can lead to care pathways that are not efficacious and possibly compound or worsen symptoms.11 12 94 Health professionals should seek to explain a clear (provisional) diagnosis to patients at the earliest opportunity and then signpost helpful information and supportive organisations. Providing diagrams to increase knowledge of the anatomy are useful. Education would be the starting point for management of any patient with a suspected TMD. Useful resources to help with this are signposted in the additional educational resources box.

Self management (or self care) techniques—specific suggestions for self management and for physiotherapy are given in the recently published primary care guidance for TMDs ( Simple generic advice around sleep hygiene, caffeine consumption, and avoidance of parafunctional activities such as daytime clenching, nail biting, or excessive chewing of gum can be helpful. This advice along with suggestions on relaxation techniques such as diaphragmatic breathing, and local measures for pain relief such as the use of covered ice or warm flannels to painful areas can be of benefit for some patients.

Physiotherapy and acupuncture—there is limited evidence to suggest that physiotherapeutic modalities such as massage or targeted exercises may help provide short term relief for acute TMDs, although the results of a Cochrane systematic review (see supplementary table) are awaited. A systematic review on acupuncture suggests that it may reduce pain intensity and specifically help with masseteric tenderness (see supplementary table).

Cognitive behavioural therapy—such treatment is considered the ideal psychological treatment for chronic pain.95 Cognitive behavioural treatment for pain is an active, structured, collaborative therapy, which aims to teach patients the most adaptive ways of coping with their pain. Techniques might include dealing with pain related anxieties, behavioural activation plans, management of pain flare-ups, sleep hygiene, and optimal communication strategies with family members. The evidence for cognitive behavioural treatment in relation to TMDs indicates important long term improvements in pain intensity and mood, and reductions in activity limitation due to pain.96 97 Cognitive behavioural treatment does not seem to reduce pain on muscle palpation.

Splints—“mouth guards” in either a soft polyethylene or a hard acrylic that provide full coverage of all the upper or lower teeth worn, can be worn at night in those who grind or clench their teeth. Mouth guards help to protect tooth substance and may also provide an element of biofeedback.

Simple analgesia—non-steroidal anti-inflammatory drugs can be used, being mindful of gastrointestinal side effects from continual use and drug overuse headache.

Other drug interventions—more “advanced” approaches include the use of neuromodulatory agents for persistent pain, which would all come under off-label prescribing. A small number of published trials support the use of gabapentin, benzodiazepines, amitriptyline, and propranolol for the management of pain of a myogenous origin.60 98 99 100 Botulinum toxin applied to the muscles of mastication has been the subject of a small number of randomised controlled trials of varying quality and with the potential for type II errors.101 102 103 104 Considering the results of these various studies as a whole the evidence at present is at best equivocal for the use of botulinum toxin in myogenous TMDs, but there is still a need for a definitive and appropriately powered randomised controlled trial.105 Some concerns are emerging from a small scale study about the potential osteopenic effects on the temporomandibular joint of repeatedly administered botulinum toxin to the muscles of mastication over a long period.106

Irreversible approaches

The American Association for Dental Research has also issued a statement to recommend strongly that unless there are “specific and justifiable indications,” irreversible approaches should not be used as primary management for TMDs.27

Orthodontics and occlusal equilibration—systematic reviews of irreversible approaches such as orthodontics and occlusal equilibration (careful grinding of the natural teeth to achieve a more “harmonious” bite) show these approaches lack evidence to support their use in the management of TMDs.107 108

Surgical interventions—these are only indicated in a small number of cases and only if there is an intra-articular disorder, degenerative joint disease, or other evident disease affecting the temporomandibular joint.96

Why is psychological management of TMDs important?

TMDs, as with other causes of pain, are best conceptualised in biopsychosocial terms.109 The patient’s beliefs, expectations, emotional state, family environment, and cultural background can all play a role in the onset, development, and maintenance of persistent pain. The axis 2 classification of the DC/TMD explicitly recognises the importance of psychological and social aspects of pain by assessing not just pain intensity but also depressed mood, anxiety, pain related disability, parafunctional behaviours, and a variety of associated comorbidities. The DC/TMD also helps to perform a basic (“screening”) or a “comprehensive” assessment of the patient’s psychosocial status and can be used to inform referrals for specialised psychological input.

In addition to cognitive behavioural therapy, there is limited evidence for a range of other psychologically based interventions for TMDs, including electromyographic biofeedback, stress management, and hypnosis.96 97 In the past, referral for psychological therapy has tended to be the last resort, once all biological treatments have been exhausted and present treatment “failed.” As a result of this type of approach patients may interpret the referral to a psychologist as an indication that their medical or dental practitioner has given up on them, or that their pain is considered to be “psychological” and somehow not genuine.110 Making time for adequate explanation of the multidisciplinary nature of complex pain treatment as early as possible in the diagnosis and management process is generally time well spent.

Having laid the groundwork that chronic pain is best managed using a team approach (which includes the patient as an active member of the team), the literature relating to facilitating the engagement of patients in biopsychosocial intervention offers several useful suggestions.111 As always, the quality of patient-provider communication is important, with empathy for the patient’s pain experience particularly relevant.112 The use of patient centered communication skills, promoting patient self efficacy, and providing positive feedback on patient’s coping efforts can all enhance patient motivation to engage in self management approaches.111 A further communication problem for clinicians relates to patients wanting to pursue treatment options that are risky, ill advised, or dangerous, such as an irreversible intervention where there is no objective clinical evidence to support its use. These requests are best understood as reflecting the patient’s desperation in living with intractable pain. In response, clinicians should firstly empathise with the difficulty of the patient’s situation, but provide a clear, evidence based explanation as to the inadvisability of the treatment. Encouraging patients to refrain from falling into the trap of seeking “quick fixes” for complex problems and assisting and supporting them to develop and expand their own coping resources, offers a better long term solution.

Tips for non-specialists

  • Problems associated with TMDs are common and associated with pain and sometimes clicking and locking

  • Consider either liaision with the patient’s dental practitioner, or use of a simple screening questionnaire if confirmation of clinical diagnosis is required

  • It is important to exclude temporal arteritis, especially if the patient is older than 50 years

  • Reassurance, explanation, and non-steroidal anti-inflammatory drugs used early in the history of the disorder are highly effective

  • Management of both comorbid migraine and TMDs may help improve outcome

  • In patients with other chronic pains and depression, consider referral to specialist centres for biopsychosocial management

Additional educational resources

Resources for healthcare professionals
Resources for patients


Cite this as: BMJ 2015;350:h1154


  • Contributors: JD and JMZ conceived and designed the article. JD drafted the manuscript. TNJ contributed the psychological aspects. All authors approved the final version of the manuscript and are the guarantors.

  • Competing interests: We have read and understood the BMJ policy on declaration of interests and declare the following: JMZ is part funded by UCL/UCLHT, which received a proportion of funding from the Department of Health’s NIHR Biomedical Research Centre funding scheme; JD is funded by a NIHR clinician scientist award (NIHR-CS-011-003). The views expressed in this publication are those of the authors and not necessarily those of the National Health Service, the National Institute for Health Research, or the Department of Health in the United Kingdom.

  • Provenance and peer review: Commissioned; externally peer reviewed.


View Abstract