Hemifacial spasm is a chronic disorder in which the muscles supplied by the facial nerve on one side of the face contract involuntarily, and the 2020 Cochrane review of botulinum toxin for it notes that spontaneous recovery is very rare and that the two treatments routinely available are botulinum toxin injections and microvascular decompression, an operation that a 2025 review in the Journal of Neurology describes as moving the blood vessel that presses on the facial nerve away from it. Blepharospasm is a different disorder, which the Cochrane review of botulinum toxin for it describes as a focal dystonia: spasms of the muscle around the eye close the eyelids involuntarily. The Cochrane Library holds no review of acupuncture for either. For hemifacial spasm the reviews that exist were written in China from trials run in China, and a 2022 overview rated every one of them critically low in methodological quality; for blepharospasm there is a review protocol and a handful of small trials.
The Cochrane reviews
Cochrane’s two reviews on these conditions are of botulinum toxin type A injections. The hemifacial spasm review by Duarte and colleagues, updated in November 2020 from a 2005 original, searched to July 2020 for double-blind trials of the toxin against placebo and found none. Its authors could draw no conclusion from randomised evidence and point instead to observational data showing a strong association between the injections and improvement. The blepharospasm review, by the same group and updated the same month, found three placebo-controlled trials with 313 participants, all funded by industry. A single injection session reduced severity on the Jankovic Rating Scale, a clinical rating scale for the condition, by 0.93 points on its severity subscale at four to six weeks, which the authors rate with moderate certainty; it also raised the risk of a drooping eyelid and of visual complaints.
Systematic reviews of acupuncture for hemifacial spasm
The earliest review in English, a 2012 meta-analysis by Wang, Bai and Lei in Alternative Therapies in Health and Medicine, included 13 randomised trials with 1,262 participants, all conducted in China. The trials compared acupuncture with carbamazepine, mecobalamin (a form of vitamin B12) or massage, and counted two outcomes: the share of people with any response and the share with full recovery. Every trial reported acupuncture ahead of its comparison and the pooled result agreed, but the authors judged most of the trials poor in methodological quality and drew no affirmative conclusion, citing that quality and the single-country population.
The 2022 overview by Gong and colleagues at Changchun University of Chinese Medicine searched eight English and Chinese databases into 2022 and found eight systematic reviews of acupuncture for hemifacial spasm, published from 2012 to 2022. Seven were in Chinese; all were by authors in China and all reached a positive conclusion. They pooled 6 to 17 trials each, with 583 to 1,262 participants, and compared acupuncture, electroacupuncture, fire needling or moxibustion, alone or added to other care, mostly against drugs such as carbamazepine, methylcobalamin and vitamins, and in some reviews against sham acupuncture. Under AMSTAR 2, a checklist for the conduct of systematic reviews, all eight were rated critically low: none had registered a protocol and none listed the trials it excluded. The overview graded 27 outcomes with GRADE, which rates how far a pooled estimate can be trusted, and found none high, two moderate, 14 low and 11 critically low. Every outcome was downgraded for the risk of bias in the underlying trials, which mostly did not describe how they randomised, concealed allocation or blinded anyone.
| Review | Comparison | Outcome | Result | GRADE |
|---|---|---|---|---|
| Wu and colleagues 2020 | Acupuncture against sham acupuncture | Effective rate | Odds ratio 3.59 (95% CI 1.25 to 10.31) | Moderate |
| Wu and colleagues 2020 | Acupuncture against conventional drugs | Effective rate | Odds ratio 4.16 (95% CI 2.45 to 7.06) | Moderate |
| Wang and colleagues 2012 | Acupuncture against conventional drugs | Cure rate | Risk ratio 2.07 (95% CI 1.60 to 2.68) | Low |
| Chen and colleagues 2019 | Fire needling against acupuncture or drugs | Effective rate | Odds ratio 2.88 (95% CI 1.97 to 4.23) | Low |
| Huang and colleagues 2018 | Electroacupuncture against drugs | Effective rate | Odds ratio 5.48 (95% CI 2.28 to 13.13) | Critically low |
The “effective rate” and “cure rate” are categories each trial defined for itself rather than a common scale of spasm. On safety, five of the eight reviews mentioned adverse events; three had too few trials reporting them to analyse, and two recorded adverse events only in the control groups. The overview’s authors describe acupuncture as a promising complementary treatment and conclude that the reviews’ low methodological quality rules out wide recommendations until larger, better-designed trials report. A 2019 protocol by Yang and colleagues, registered with PROSPERO as CRD42019142473, set out a further review with the cure rate and the effective rate as its primary outcomes; this page found no published result in PubMed or Europe PMC.
A sham-controlled trial
One trial has compared electroacupuncture with sham on a symptom scale: Li and colleagues 2026 in Movement Disorders, the journal of the International Parkinson and Movement Disorder Society. Run from 2016 to 2023 and described as randomised and double-blind, it allocated 84 people with hemifacial spasm to 30 sessions over 15 weeks of electroacupuncture on the hand, electroacupuncture on the face, or sham electroacupuncture. Low-frequency, low-intensity stimulation on the hand (2 Hz, 0.5 to 1 mA) reduced the Jankovic Rating Scale score by 3.32 points more than sham (95% CI 2.52 to 4.12), and the authors report that the change was still present six months after treatment. They tie it to reduced excitability in the brainstem and the facial area of the motor cortex, measured by electromyography and transcranial magnetic stimulation. The abstract does not report the facial-electroacupuncture group against sham.
The 2025 Journal of Neurology review, written before that trial was published, lists electroacupuncture among miscellaneous treatments reported mostly in single case reports. It suggests such treatments might offer options to people who do not respond to, cannot tolerate or are unsuitable for botulinum toxin and surgery, and states that larger studies are needed to assess their efficacy and safety.
Blepharospasm
No completed systematic review of acupuncture for blepharospasm was found. A 2021 protocol by Huang and colleagues, posted as a preprint and registered as CRD42020222489, planned to pool randomised trials using blepharospasm rating scales as its primary outcomes. The trials that exist are small. A 2014 trial in Zhongguo Zhen Jiu randomised 70 people to one of two needling methods at the same points, among them BL02 (Cuanzhu), ST02 (Sibai), XH03 (Yintang) and Yuwei threaded through to XH04 (Yuyao). It reported 77.1% of the stronger-stimulation group “cured or markedly effective” against 51.4% with conventional needling, and fewer relapses at six months, without a sham or untreated group. A 2019 trial in Current Eye Research randomised 30 people to botulinum toxin or acupuncture and measured lower-eyelid tension with a tensiometer rather than a symptom scale; tension fell in both groups.
Several of these points lie close to the eye. A 2023 review in the Korean Journal of Ophthalmology collected 14 case reports, published from 1982 to 2020, of serious eye injury after needling around the eye.
Guidelines
The American Academy of Neurology (AAN) has assessed botulinum toxin for both conditions, and neither document mentions acupuncture. Its 2008 assessment of botulinum toxin for movement disorders concluded the toxin may be considered for hemifacial spasm (Level C, from one Class II and one Class III study) and may be offered for blepharospasm (Level B). The 2016 update rated onabotulinumtoxinA and incobotulinumtoxinA probably effective for blepharospasm (Level B) and did not cover hemifacial spasm. Both are now marked retired in the journal record. Clinical guidelines sets out other bodies’ positions on acupuncture.
What the reviews could not settle
Every review of acupuncture for hemifacial spasm draws on trials from one country, most of them comparing acupuncture with drugs such as carbamazepine rather than with botulinum toxin or surgery, and scoring an outcome each trial defined for itself. The reviews did not register protocols, and the 2022 overview could rate no pooled result above moderate certainty. Since then one sham-controlled trial using the Jankovic Rating Scale has reported a difference; no review has yet pooled it with others, and none of the sources here reports a trial comparing acupuncture with botulinum toxin or surgery for hemifacial spasm. For blepharospasm the planned review has not appeared.
AcuiQ’s facial spasm, hemifacial spasm, hemifacial spasm (indexed as spasm, hemifacial) and blepharospasm pages list the protocols individual studies prescribed, each with its citation. Bell’s palsy covers weakness of the same facial nerve, where hemifacial spasm is its overactivity.