Temporomandibular disorders (TMD) are a group of conditions that cause pain in the jaw joint, the chewing muscles and the structures around them. The 2023 BMJ guideline by Busse and colleagues puts them second only to low back pain among chronic musculoskeletal pain disorders, affecting 6 to 9% of adults, and lists headache, earache, clicking or grating sounds in the joint and difficulty moving the jaw among the ways they present. The UK guideline separates myogenous TMD, pain from the chewing muscles, from arthrogenous TMD, pain from the joint. Reviews that pool acupuncture against any comparison report less pain; reviews restricted to sham-controlled trials disagree with one another, and the guideline built on the largest analysis recommends acupuncture conditionally.

No Cochrane review

Cochrane has not published a review of acupuncture for TMD. Its reviews of the condition cover occlusal interventions (2024) and psychological therapies (2022). The largest review that includes acupuncture is a network meta-analysis, a method that compares many treatments at once by linking trials through shared comparators.

The 2023 network meta-analysis

The network meta-analysis by Yao and colleagues, published in the BMJ in December 2023, searched five databases to January 2023 for trials in people with chronic TMD pain. It found 233 eligible trials and pooled 153 of them, with 8,713 participants and 59 treatments or combinations, each compared with placebo or sham. Eight treatments had moderate or high certainty evidence of relieving pain, led by cognitive behavioural therapy with biofeedback or relaxation, therapist-assisted jaw mobilisation and manual trigger point therapy; acupuncture was not among the eight. For physical function, moderate certainty evidence showed that acupuncture probably raised by 42 percentage points (95% CI 33 to 50) the share of people who improved by at least 5 points on the physical summary score of the SF-36 questionnaire, alongside supervised jaw exercise, manipulation and jaw exercise with mobilisation. Evidence on adverse events was low or very low certainty for every treatment.

Reviews against sham

  • La Touche and colleagues, 2010. This meta-analysis found 8 randomised trials and judged 4 of acceptable quality. Pooling the trials that compared acupuncture with placebo or sham, 96 participants in all, it found acupuncture reduced pain intensity more (standardised mean difference 0.83, 95% CI 0.41 to 1.25). The authors called acupuncture a reasonable adjunct for short-term pain relief and said substantial bias limited the result.
  • Jung and colleagues, 2011. This review searched 13 databases, including seven Korean and one Chinese, to July 2010 and found 7 trials against sham. It concluded that the evidence for acupuncture as a treatment for TMD symptoms is limited.
  • Al-Moraissi and colleagues, 2023. This network meta-analysis of trials to September 2019 in masticatory muscle pain compared acupuncture and dry needling with active and inactive placebo. Neither differed from placebo on pain intensity or on pressure pain threshold, the pressure on a muscle at which it starts to hurt. Dry needling increased maximum mouth opening, on very low quality evidence. The authors concluded that placebo accounts for most of the effect.
  • Di Francesco and colleagues, 2024. This meta-analysis searched three databases to July 2023 and included 11 trials of needle or laser acupuncture. Both had a higher response rate than placebo, and the authors still described the evidence for needle acupuncture as limited.

Sham acupuncture explains what these controls can and cannot rule out.

Reviews against other treatments

A 2023 meta-analysis by Park and colleagues included 32 trials and pooled 22, with 471 participants. Acupuncture improved outcomes against active treatments, and when added to another treatment it raised the response rate (risk ratio 1.36, 95% CI 1.04 to 1.77) and lowered pain on a visual analogue scale by 1.23 points (95% CI 0.67 to 1.79). The authors rated the evidence low to very low.

A 2026 meta-analysis by Ning and colleagues searched to January 2026 and pooled 19 trials with 708 people against a mix of control treatments. Acupuncture reduced pain (standardised mean difference −0.65, 95% CI −0.83 to −0.48) and increased mouth opening (0.57, 95% CI 0.38 to 0.77). Needle and laser acupuncture did about as well as each other, and two or more sessions a week may have done better than one.

A 2026 network meta-analysis by Zhang and colleagues compared non-drug treatments for muscular TMD across 41 trials with 2,021 adults. Photobiomodulation, a group of treatments that includes low-level laser, relieved pain more than acupuncture (standardised mean difference −1.07, 95% CI −1.93 to −0.21), and manual therapy improved mouth opening more than acupuncture (2.01, 95% CI 0.12 to 3.90).

Landmark trials

  • Points against non-points, 2002. In a trial by Goddard and colleagues, 18 people with myofascial jaw pain were needled either at acupuncture points or at skin away from them. Pain provoked by pressing on the masseter, the main chewing muscle, fell in both groups by a similar amount: 7 of 10 and 4 of 8 improved by at least 10 mm on a visual analogue scale.
  • A sham device on the face, 2007. A double-blind trial at the University of Manchester dental school, by Smith and colleagues, gave 27 people with myofascial pain of the jaw either acupuncture or a non-penetrating sham through the Park sham device. Real acupuncture did better on most outcomes, and the authors judged the device a credible control for points on the face.
  • One point, one session, 2009. A trial by Shen and colleagues gave 28 people with chronic myofascial jaw pain 15 minutes of needling at LI04 (Hegu) or a blunt needle that pricked the skin elsewhere. Jaw pain, jaw tightness and neck pain fell, and tolerance of pressure on the masseter rose, with real needling only. Participants could not tell which treatment they had received.
  • Specific against non-specific points, 2020. A double-blinded trial in Germany by Şen and colleagues randomised 49 people with painful TMD of less than chronic severity, and 41 completed it. Needling at classical points and at non-specific points both reduced pain, and the authors concluded that the effect could not be attributed to which points were chosen.
  • Chronic TMD against sham, 2024. A single-centre trial at Beijing Hospital of Traditional Chinese Medicine, by Liu and colleagues, randomised 60 people with pain-related TMD of at least 3 months, diagnosed under the Diagnostic Criteria for TMD and excluding disorders inside the joint. Both groups had 12 sessions over 4 weeks: needling at LI04 and GB34 (Yanglingquan) on both sides and at SI19 (Tinggong), ST06 (Jiache) and ST07 (Xiaguan) on the painful side, or Park sham needles at the same points. Pain on a visual analogue scale fell by 1.49 points more with acupuncture at week 4 (95% CI 0.65 to 2.32) and by 1.23 points more at week 8. Jaw opening, jaw function, sleep and mood scores also improved more; pressure pain threshold and muscle electrical activity did not differ. Two participants given acupuncture had an adverse event, a bruise under the skin and pain after needling; none was serious. Only the participants were blinded, not the acupuncturists.

What guidelines say

The BMJ Rapid Recommendation by Busse and colleagues, published in December 2023 and drawn from the Yao network meta-analysis, covers chronic TMD pain lasting 3 months or more and not acute pain. Against placebo or sham, an international panel that included patients made a conditional recommendation in favour of acupuncture, alongside manipulation and jaw exercise with mobilisation. It made strong recommendations for cognitive behavioural therapy, therapist-assisted mobilisation, manual trigger point therapy, supervised postural and jaw exercise, and usual care such as home exercises and reassurance. It recommended conditionally against reversible occlusal splints, low-level laser therapy, TENS and several drugs and injections; TENS is electrical nerve stimulation through the skin. No later version has been published.

The UK guideline on painful TMD in adults, from NHS England’s Getting It Right First Time programme and the Royal College of Surgeons’ Faculty of Dental Surgery, was first issued in March 2024; version 1.1 is dated 27 August 2025. It grades as a strong recommendation on moderate evidence that acupuncture added to other care for muscular TMD is likely to help pain, drawing on reviews of needle, laser and electroacupuncture and of dry needling. It places acupuncture as an adjunct to supported self-management, encourages referral to an acupuncturist with experience in facial pain on expert opinion, and lists acupuncture as an option to discuss for joint pain that has improved too little at review.

NICE’s Clinical Knowledge Summary on TMD, last revised in May 2026, suggests physiotherapy referral for jaw stretching, posture training, and massage or acupuncture to relax muscle spasm, where available. Its supporting evidence notes that acupuncture may have limited long-term benefit and cites the UK guideline above.

What the reviews could not settle

Whether the needle’s location matters is unresolved. Goddard and Şen found points and non-points relieved pain equally, Smith, Shen and Liu found real needling beat a non-penetrating sham, and the Al-Moraissi analysis found no difference from placebo on pain. The trials above enrolled between 18 and 60 people each, the reviews pooled short-term results, and Ning and colleagues call for trials with long-term follow-up. The Yao analysis found no moderate or high certainty evidence that acupuncture relieves pain, and the BMJ panel says further research may change its recommendations. Chronic pain reports the broader trials of acupuncture for persistent musculoskeletal pain.

AcuiQ’s temporomandibular disorder, jaw pain and masticatory muscle pain pages list the protocols individual studies prescribed, each with its citation.