The US otolaryngology guideline defines tinnitus as the perception of sound without an external source. Tinnitus trials measure what a person reports: loudness or annoyance marked on a line (a visual analogue scale), or questionnaires such as the Tinnitus Handicap Inventory (THI), which scores how much tinnitus interferes with daily life. The acupuncture reviews divide along one line. Trials that compared acupuncture with sham acupuncture, a control procedure built to resemble it, have mostly found no significant difference; reviews that also pool comparisons with drugs, other treatments or none report better questionnaire scores, with low-quality evidence. Sham acupuncture covers the kinds of control in use.
No Cochrane review
Cochrane registered a review of acupuncture for tinnitus and published its protocol in 2009. In November 2016 it withdrew the protocol because the review had not been completed within the expected time. The evidence therefore comes from reviews published elsewhere.
The sham-controlled trials
The first review, a 2000 analysis by Park, White and Ernst, found six randomised trials, four of them crossover trials in which each participant had both treatments in turn. The two unblinded trials reported a positive result; the four blinded trials found no significant effect. The authors concluded that rigorous trials had not shown acupuncture to be efficacious for tinnitus.
A 2012 review by Kim, Ernst and colleagues searched 14 databases to July 2012 and included nine randomised trials that used acupuncture as the only treatment: three from Denmark, two each from China and Brazil, and one each from England and Korea. Their methodological quality was mostly poor.
- Five trials compared acupuncture or electroacupuncture with sham, and none showed a statistically significant improvement.
- Two compared one short session of scalp acupuncture with needling at non-acupuncture points, and both reported relief on a visual analogue scale.
- Two compared acupuncture with drugs: one found a significant difference in the share of people who responded, the other did not.
The authors concluded that the trials were too few, too small and too weak to support a definitive conclusion. Six of the nine trials needled points around the ear: GB02 (Tinghui) in five, TB17 (Yifeng) in four and SI19 (Tinggong) in three. The review’s tables write the triple burner points with the WHO prefix TE, which AcuiQ writes as TB; Reading a point code explains the difference. Scalp acupuncture describes the method the two single-session trials used.
Later reviews
A 2016 meta-analysis by Liu and colleagues searched English and Chinese databases and found a split by language: nearly all the Chinese trials reported positive results and most of the English-language trials negative ones. The Chinese and English-language trials used different points and numbers of sessions, and the authors found methodological flaws in many trials, especially the Chinese ones. The authors concluded acupuncture may offer subjective benefit to some patients but that the flaws prevented a definitive conclusion. Where the trials come from sets out what the share of positive results by country shows across acupuncture research.
| Review | Trials (people) | Controls pooled | Finding |
|---|---|---|---|
| Huang 2021 | 8 (504) | Conventional treatment, sham, no treatment, or the same treatment without acupuncture | No significant difference on its primary outcome, a visual analogue scale; THI 10.1 points better and Tinnitus Severity Index 8.4 points better with acupuncture |
| Wu 2023 | 34 (3,086) | Drugs, oxygen or physical therapies, or no treatment; no sham | Lower THI scores and a higher response rate with acupuncture and moxibustion; GRADE evidence low, with considerable inconsistency between trials |
Huang and colleagues judged the trials too few and too low in quality for a definitive conclusion. Wu and colleagues, whose review added moxibustion, the burning of mugwort on or above points, reported a good safety profile and called for larger trials with longer follow-up.
A 2022 overview by Xu and colleagues assessed 14 published systematic reviews of acupuncture for tinnitus. Eleven reported acupuncture as effective and three said no firm conclusion was possible. On AMSTAR-2, a checklist for the quality of a review, one was rated moderate and the other 13 very low. Of the 54 results the overview graded, none reached high-quality evidence: 17 were moderate, 25 low and 12 very low.
What guidelines say
NICE’s tinnitus guideline, NG155, published in March 2020, does not mention acupuncture in its recommendations. Those on management cover psychological therapies, amplification devices such as hearing aids, and sound therapy, and advise against offering the drug betahistine.
The American Academy of Otolaryngology–Head and Neck Surgery Foundation’s 2014 clinical practice guideline, cited above for its definition and written for adults with tinnitus that is persistent and bothersome, recommended cognitive behavioural therapy and a hearing aid evaluation where hearing loss is documented. It recommended against routine antidepressants, anticonvulsants, anxiolytics, dietary supplements including Ginkgo biloba, and transcranial magnetic stimulation, and made no recommendation either way on acupuncture. Clinical guidelines compares other guideline bodies.
What the reviews could not settle
The sham-controlled trials are few and small, and the reviews that report benefit lean on comparisons with drugs, other treatments or no treatment, where participants know what they received and the outcome is self-reported. None of the reviews could say whether results differ by the kind or cause of tinnitus. A Cochrane review that would apply one standard method across the trials was registered and never completed. Dizziness and vertigo reports the trials in Ménière disease, where tinnitus comes with attacks of vertigo.
AcuiQ’s tinnitus page and its subjective tinnitus page list the protocols individual studies prescribed, each with its citation.