Skip navigation

Acupuncture for Temporomandibular Joint Dysfunction

Clinical bottom line: There is no evidence to show that acupuncture is an effective treatment for Temporomandibular joint dysfunction (TMJD). The few existing randomised controlled trials (RCTs) are of poor methodological quality.

TMJD is a poorly defined complex pain condition. Current treatment options are limited and none have proven efficacy.

Systematic review

Ernst E and White AR. Acupuncture as a Treatment for Temporomandibular Joint Dysfunction: A Systematic Review of Randomised Trials. Arch Otolaryngol Head Neck Surg; 125: 269-272

Date review completed: 1997

Number of trials included: 3

Number of patients: A total of 205 patients in the active and control groups

Control group: standard treatment (occlusal splints), no treatment (waiting list)

Main outcomes: clinical dysfunction index, mouth opening, and subjective patient ratings

Inclusion criteria were RCTs of acupuncture compared with sham acupuncture, standard treatment or no treatment

Reviewers conducted extensive searches of the main databases and reference lists of retrieved reports. Experts in the field were also contacted. There was no restriction to language of publication. A descriptive summary of trial findings was presented in the review.


Included patients in all studies were hospital specialist referrals from family dentists or doctors for TMJD treatment. Overall, the included trials are of poor methodological quality. Although trials were described as randomised, details of randomisation were insufficient to determine the method of treatment allocation and if it remained concealed throughout the study. None of the trials were observer or patient blind.

Two trials used formula acupuncture (needling of a pre-defined set of points according to conventional diagnosis); one trial gave traditional Chinese acupuncture (needling of individualised sets of points according to traditional Chinese diagnosis). Acupuncture needles were manually stimulated, except in one study where electrical stimulation at 2Hz to produce muscle contraction was given to two needles for the majority of the sessions. The number of treatment sessions ranged from three to eight.

The first study with a total of 50 patients showed standard stomatognathic treatment significantly better than acupuncture one week after treatment (p=0.04) for the dysfunction index. At three months there was no significant difference between treatments (p=0.12) on any outcome. The authors claim a marked improvement in both groups.

The second study with 45 patients (approx. 15 per group) compared six sessions of acupuncture to maxillary occlusal splints and an untreated control group. Ninety percent of the acupuncture group and 86% of the occlusal splint group improved by two month follow up. The reviewers do not state how many improved in the untreated control group but both treatment groups were significantly better than the untreated group for all outcomes (p<0.01).

The third study with 110 patients compared six to eight acupuncture sessions (40 patients) with occlusal splints (40 patients) and an untreated control group (30 patients). Evaluation of 96 patients at three months showed a statistically significant benefit of acupuncture compared with both other groups on three out of seven outcomes. Acupuncture and occlusal splints were significantly better than the untreated control group for two out of seven outcomes. It is unclear from the review whether corrections for multiple testing were applied. Follow up after six months and one year showed no significant difference between the acupuncture and occlusal splint group for any outcome.

The authors of the review state that the evidence from these three trials suggest that acupuncture is as effective as standard therapy and more effective than no treatment and that the effect lasts for as long as one year post-treatment.

All three trials have serious methodological flaws; none of the trials have an adequate placebo control group. The fact that acupuncture was better than no treatment does not rule out a placebo effect. In these trials, acupuncture was shown to be as effective as standard treatment. However, the trials were not sensitive for the long term outcomes and in the absence of a placebo group or an additional active control group of proven efficacy it is unclear if both treatments were indistinguishable from each other because they were equally effective or equally ineffective. The purported long-term effects of acupuncture and standard therapy may just be due to the natural history of the disease. These trials are inadequately designed to answer these questions.

Adverse effects were not reported by the individual reports of included trials

Further reading

Related topics