Dry eye disease

Most trials of acupuncture for dry eye disease compare it with artificial tears, the lubricating eye drops that are the usual first treatment, and most report better tear measurements with acupuncture. The few trials that compare it with sham acupuncture, a control procedure made to resemble it, report smaller differences or none, and where they find one it is on some measures and not others. Cochrane has not published a review of acupuncture for dry eye: a PubMed search of the Cochrane Database of Systematic Reviews on 10 October 2026 found that its dry eye reviews cover eye drops, punctal plugs, omega-3 supplements and devices, and none covers acupuncture. The evidence on this page comes from reviews in other journals, a handful of trials, and the reports of the Tear Film and Ocular Surface Society (TFOS) and the American Academy of Ophthalmology.

What the trials measure

Dry eye trials report two kinds of outcome, and acupuncture trials often move one without the other. Signs are measured on the eye. Tear break-up time (TBUT) is the number of seconds after a blink before dry spots appear in the tear film, and under 10 seconds suggests an unstable film, as the 2019 Pittsburgh trial below defines it. The Schirmer test measures tear production as the length of a paper strip, placed inside the lower eyelid for 5 minutes, that the tears wet; under 10 mm suggests at least moderate dry eye in the 2012 Korean trial. Corneal staining counts damaged spots on the eye surface. Symptoms are scored by questionnaire, most often the Ocular Surface Disease Index (OSDI), 12 questions scored from 0 to 100, higher meaning worse. The TFOS DEWS II management report notes that a proportion of patients have signs and symptoms that conflict.

Reviews

Every review below pooled trials against artificial tears, or acupuncture added to artificial tears against the drops alone. None pooled trials against sham, and none of those whose full text AcuiQ could read graded the certainty of its evidence with GRADE, the system Cochrane uses.

Lee and colleagues, at the Korea Institute of Oriental Medicine with Edzard Ernst, wrote the first of these reviews. Kim and colleagues at Kyung Hee University in Seoul found that 15 of their 19 studies were run in China and 4 blinded the assessor; one compared acupuncture with sham at nearby non-points and found no difference. The heterogeneity they reported means the trials disagreed with each other more than chance explains, and no subgroup by number of sessions, length of course or points used reduced it. A standardised mean difference (SMD), as Wang and colleagues at Hunan University of Chinese Medicine report, expresses the gap between groups in units of the spread of scores. In the network meta-analysis by Yan and colleagues in Hunan, which ranks several treatments against each other through shared comparators, only one of the 30 trials reported any blinding. A 2026 evidence map from Nanjing University of Chinese Medicine counted 319 randomised trials of acupuncture and moxibustion for dry eye to December 2025 and judged their methodological quality generally low.

A 2023 review by Park and colleagues at Daegu Haany University, which used the same 21 trials as Na 2021 to look for an optimal point prescription, noted that none of those trials had a sham arm. It identified two trials that did, and in both, break-up time and Schirmer scores did not improve more with acupuncture than with sham. Those two trials are the first and fourth below.

Trials

• Against sham, 2010. A trial by Shin and colleagues in Korea randomised 42 people with moderate to severe dry eye to nine sessions over three weeks of real acupuncture at 17 points, among them BL02 (Cuanzhu), GB14 (Yangbai), TB23 (Sizhukong), ST01 (Chengqi) and GB20 (Fengchi), or of shallow needling at 17 sham points. OSDI, discomfort, break-up time and Schirmer scores did not differ between the groups. Symptoms improved within both. No adverse events were reported.
• Against artificial tears, 2012. A multicentre trial by Kim and colleagues in three Korean hospitals randomised 150 people to 12 sessions over four weeks, at much the same points plus LI04 (Hegu) and LI11 (Quchi), or to carboxymethylcellulose drops. At the end of treatment the OSDI had fallen by about 16 points in both groups. Eight weeks later it was 16.15 points below baseline with acupuncture and 10.76 with drops (P = 0.030). Three people in the acupuncture group had a haematoma, a bruise from bleeding under the skin, and one stopped treatment because of it.
• Added to artificial tears, 2018. A trial by Louis Tong and colleagues at the Singapore Eye Research Institute randomised 150 people with mild to moderate dry eye to artificial tears alone, the tears plus eight acupuncture sessions over a month, or the tears plus a daily herbal formula. Assessors did not know the allocation. Symptoms responded in 88% with acupuncture and 72% with tears alone, a number needed to treat of 7 (95% CI 3 to 157): seven people treated for each extra responder. Conjunctival redness and two tear inflammatory markers also fell more with acupuncture; tear osmolarity, evaporation, staining and break-up time did not differ. The herbal arm did not differ from tears alone.
• Against sham, 2019. A double-blinded trial by Deepinder Dhaliwal and colleagues at the University of Pittsburgh gave 24 people acupuncture on two consecutive days at three ear points and on the index finger at LI01 (Shangyang) and LI02 (Erjian), and 25 people sham needling on the upper shoulders away from any acupuncture point. OSDI scores improved from baseline in the acupuncture group at every visit to 6 months, and at 6 months the improvement was greater than with sham. Tear flow, break-up time and staining did not change. No adverse effects were noted.
• Against sham, after cataract surgery, 2022. A trial by Kong and colleagues in Shanghai randomised 90 people with dry eye after cataract surgery to real or non-penetrating sham acupuncture twice weekly for 8 weeks. Non-invasive break-up time rose by 1.52 seconds with acupuncture and 0.77 with sham at week 8 (difference 0.75, 95% CI 0.12 to 1.39), and the gap held at week 12. OSDI fell by 20.13 against 13.44 points at week 8, a difference that was gone by week 12.
• Against artificial tears, one point, 2022. A trial by Zhang and colleagues in two Chinese hospitals randomised 120 people with moderate to severe dry eye to 8 weeks of acupuncture at BL01 (Jingming), the point at the inner corner of the eye, or artificial tears. Schirmer scores rose by 5.75 against 0.52 mm, and the OSDI fell 8.49 points more with acupuncture; symptom ratings, break-up time and staining did not differ. Five of the 60 people needled had an adverse event, four of them bleeding at the needle site.

A PubMed search found no trial of acupuncture for dry eye in Ophthalmology, JAMA or its network journals, the American Journal of Ophthalmology, the British Journal of Ophthalmology or BMJ. Sham acupuncture sets out what penetrating and non-penetrating shams can and cannot rule out.

Needling around the eye

Dry eye prescriptions often use points on the rim of the orbit, the bony socket of the eye. A 2023 review of case reports by Lee and colleagues in Korea collected 14 reports, covering 15 eyes, of serious eye injury after periocular acupuncture between 1982 and 2020; 12 eyes had a penetrating injury, and only three reports named the point: ST01 (Chengqi), XH04 (Yuyao) and GB01 (Tongziliao). The authors concluded that acupuncture around the eye is “a risky procedure, with no robust evidence of its benefits”. A 2025 report from Singapore described two women, aged 61 and 86, with severe bruising around one eye after acupuncture for dry eye; the younger was taking 100 mg of aspirin daily. Both recovered without loss of vision. Bleeding and anticoagulants covers needling in people on blood thinners.

What guidelines say

The TFOS DEWS II Management and Therapy Report, published in 2017 by Lyndon Jones and colleagues for the Tear Film and Ocular Surface Society, discussed acupuncture under complementary therapies. It reported that trials found greater improvements than with artificial tears, named as a shortcoming that many used artificial tears rather than sham needling as the control, and cited the 2010 sham trial above. It concluded that acupuncture, with herbal and other natural products, needed more Level 1 studies, its term for well-conducted randomised trials, before wider acceptance.

The TFOS DEWS III Management and Therapy report, published in the American Journal of Ophthalmology in November 2025, succeeds it. It mentions acupuncture in one section, under natural products and interventions. It describes the 2019 sham-controlled trial, says the absence of change in tear flow, break-up time and staining suggests acupuncture may act on pain rather than on tear production, and cites the Wang 2024 review and the Singapore trial.

The American Academy of Ophthalmology’s Dry Eye Syndrome Preferred Practice Pattern, approved in September 2023 and current, mentions acupuncture once: alongside cognitive behavioural therapy and hypnosis, as a complementary therapy to consider for neuropathic ocular pain, in which symptoms outweigh the visible signs of disease. It does not discuss acupuncture for dry eye itself. Clinical guidelines sets out other bodies’ positions on acupuncture.

What the reviews could not settle

The pooled results rest on trials against artificial tears, mostly unblinded and, in the 2018 review, mostly run in China, and no review has pooled the sham-controlled trials or graded the evidence. The sham trials do not agree on which outcome moves. The 2010 trial found no difference on any measure. The 2019 trial found a difference on symptoms and none on tear measurements. The 2022 cataract trial found one on break-up time that lasted and one on symptoms that did not. Against artificial tears, the BL01 trial moved tear production and not symptoms, and the Singapore trial moved symptoms and not tear stability. Follow-up beyond the end of treatment was a matter of weeks in most trials, and the longest, 6 months, was in a trial of 49 people. Reading an acupuncture trial covers outcome choice and blinding, and Where the trials come from discusses what a trial’s country tells a reader.

AcuiQ’s dry eye, dry eye disease, dry eye syndrome and Sjögren’s syndrome pages list the protocols individual studies prescribed, each with its citation.