Atopic dermatitis, also called atopic eczema, is a chronic, relapsing inflammatory skin disease marked by dry skin and itch, as a 2024 review describes it. Chronic spontaneous urticaria, in the international urticaria guideline’s terms, is urticaria lasting more than six weeks with no definite eliciting factor: wheals (hives), swelling under the skin, or both. Itch is the symptom both share, and the trials below measure it. The two literatures differ in size and design: the eczema evidence is a handful of small trials, several of them single-session laboratory experiments, while the urticaria evidence includes one 330-person trial with a sham arm and four reviews published in 2025.
No Cochrane review
Cochrane has published no review of acupuncture for either condition. Two broader Cochrane protocols that would have covered acupuncture, complementary and alternative medicine treatments for atopic eczema and interventions for chronic idiopathic urticaria excluding antihistamines, were withdrawn in 2017 and 2018 before either reached a review. The reviews below were all published outside Cochrane.
Atopic dermatitis: the reviews
A 2020 review by Jiao and colleagues at Guang’anmen Hospital in Beijing searched 13 databases to July 2018 and found eight randomised trials with 434 participants. Seven of the eight enrolled people diagnosed with chronic eczema, the term Chinese practice uses for the condition, and seven used fire needling, in which a heated needle is inserted briefly. Seven compared acupuncture with conventional medicine and one, with 10 people, with no treatment; none used a sham. Pooling six trials, the authors found a lower Eczema Area and Severity Index (EASI, a clinician’s score of extent and severity) with acupuncture than with medication, a mean difference of 1.89 points. They rated every outcome low certainty on GRADE, the scale that grades confidence in a result from high to very low, and wrote that without sham-controlled trials the specific effects of acupuncture could not be evaluated.
A 2024 review in BMJ Open by Liang and colleagues in Ningbo searched to October 2024 and included eight trials with 463 participants: two in Germany, two in Korea and four in China. Two compared acupuncture with no treatment, two with a non-penetrating sham at non-acupoints and four with conventional medicine, usually loratadine with or without a steroid cream. It did not grade certainty. The pooled result for SCORAD, a severity score that combines a clinician’s assessment with the patient’s itch and sleep loss, favoured acupuncture by 10.6 points, but the trials disagreed so much (I² 97%) that the subgroups say more:
| Comparison | Trials | SCORAD | EASI |
|---|---|---|---|
| Acupuncture vs no treatment | 2 | +2.7, no clear difference | −0.9 (one trial) |
| Acupuncture vs sham | 2 | −12.6 | −1.8, no clear difference |
| Acupuncture plus medication vs medication | 2 (SCORAD), 1 (EASI) | −17.9 | −11.5 |
Across three trials, itch on a visual analogue scale was 14.7 points lower with acupuncture, above the 13.4-point threshold the authors took as the smallest change patients notice. No serious adverse events were reported; the minor ones were bruising, numbness, diarrhoea, indigestion and heartburn. The points used most often were LI11 (Quchi), ST36 (Zusanli) and SP10 (Xuehai). The authors call for large multicentre trials.
Atopic dermatitis: the trials
Florian Pfab’s group at the Technical University of Munich tested acupuncture against itch provoked in the laboratory. In a 2010 crossover trial, 30 adults with atopic eczema had an allergen (house dust mite or grass pollen) pricked into the skin of the forearm under three conditions in turn: needling at Quchi and Xuehai, needling at “placebo points” and no needling. Mean itch intensity was 35.7 with the real points, 40.4 with the placebo points and 45.9 with none, on a 0 to 100 scale; when needling came before the allergen, the wheal and the surrounding redness were smaller after the real points. A 2012 trial of 20 patients, seven sessions each in random order, added the antihistamine cetirizine and a placebo pill. Electroacupuncture at LI-11 and HT03 (Shaohai) during the allergen challenge gave a mean itch of 31.9, against 36.5 for a non-penetrating placebo needle at non-points, 36.8 for cetirizine and 45.7 for no intervention. Cetirizine, and none of the acupuncture arms, lowered scores on an attention test. Both are single-session experiments on provoked itch, not courses of treatment. A 2011 pilot from the same group gave five patients ten treatments and five none, and reported lower itch and less allergen-induced activation of basophils, a type of white blood cell, in the treated group.
At Kyung Hee University in Seoul, a 2018 three-arm pilot of 30 people was followed by a 2021 trial that randomised 36 adults aged 19 to 38 with mild to moderate atopic dermatitis to real or sham acupuncture twice a week for four weeks. The real group was needled at LI11 (Quchi), ST36 (Zusanli) and PC06 (Neiguan) on both sides, with up to ten further points chosen by symptoms, and then had press tack needles, tiny needles taped in place, at LI11 and the ear point Shenmen. The sham group had blunt needles that did not pierce the skin, about 1 cm from the real points. Participants and assessors did not know the allocation; the acupuncturist did. SCORAD fell by a mean of 11.8 points with real acupuncture and rose by 0.5 with sham at four weeks, and the authors report the gap persisting four weeks after treatment ended. No serious adverse events occurred. Press needles describes the taped needles.
A trial from the Charité in Berlin, published in 2022, randomised 121 adults with atopic dermatitis to acupuncture, osteopathic medicine or routine care alone, without blinding. After 12 weeks, adjusted SCORAD was 22.3 with acupuncture, 26.4 with osteopathy and 23.7 with routine care, with no difference between groups. Itch scored 27.9, 35.0 and 42.3 out of 100, a difference that was statistically significant. Neither treatment was cost-effective compared with routine care.
For acupressure, the only trial is a 2012 pilot in which 15 adults were randomised to press a 1.2 mm pellet at LI11 for three minutes three times a week for four weeks, alongside their usual care, or usual care alone; 12 finished. The authors report a greater fall in itch with acupressure and no change in the overall EASI.
Chronic urticaria: the sham-controlled trial
A 2023 trial in Annals of Internal Medicine by Zheng and colleagues randomised 330 people with chronic spontaneous urticaria at three teaching hospitals in China to acupuncture, sham acupuncture or a waiting list. According to the published protocol, the real group was needled at LI11 (Quchi), SP10 (Xuehai), ST36 (Zusanli), ST25 (Tianshu), SP06 (Sanyinjiao), HT07 (Shenmen) and CV12 (Zhongwan), 16 sessions over four weeks; the sham used blunt retractable needles at non-acupoints. Antihistamines were not allowed. The primary outcome was the weekly urticaria activity score (UAS7), which runs from 0 to 42, higher meaning more active disease. At four weeks UAS7 fell by 8.2 points with acupuncture, 4.1 with sham and 2.2 on the waiting list. Acupuncture’s advantage was 4.1 points over sham and 6.1 over the waiting list, and the authors state that neither reached the minimal clinically important difference. Fifteen people in the acupuncture group (13.6%) and none in the other groups reported adverse events, all mild or transient. The authors list incomplete blinding, self-reported outcomes and short follow-up as limitations. Sham acupuncture sets out the kinds of sham in use.
Chronic urticaria: the reviews
| Review | Trials (people) | Compared with | GRADE certainty for the main outcome |
|---|---|---|---|
| Wei | 22 (1,867) | Loratadine, cetirizine or sham | Low for UAS7 |
| Yu | 6 (615) | Sham or no acupuncture | Very low for UAS7 at the end of treatment |
| Tong | 8 (564) | Standard-dose antihistamines | Low for response rate |
| Wu | 18 (1,829) | Antihistamines, sham or waiting list | Moderate for UAS7 |
Wu and colleagues searched to September 2025; all 18 trials came from China. Against sham in six trials, UAS7 was 5.9 points lower with acupuncture; against a waiting list in four, 8.6 points lower; the one trial against an antihistamine found no difference. Their last author, Liang FR, is also an author of the 2023 trial. They note that the gains did not reach the minimal clinically important difference, and that a test for small-study effects suggested publication bias. Yu and colleagues pooled six trials published from 2021 to 2024 and found UAS7 3.4 points lower with four weeks of manual acupuncture, below the 8-point threshold they cite, and more bruising under the skin with acupuncture (risk ratio 4.5). Wei and colleagues rated UAS7 and itch severity low certainty, and reported moderate to high risk of bias across their trials. Tong and colleagues compared acupuncture with antihistamines on the “clinical effective rate”, the share of patients rated improved, and found a risk ratio of 1.19 at low certainty, with possible publication bias.
The guidelines
The American Academy of Dermatology’s 2014 guideline on flare prevention and adjunctive therapies for atopic dermatitis wrote that the evidence on acupuncture was confined to small studies of limited quality, and made no recommendation on it; it listed acupressure among therapies with insufficient evidence to recommend. The Academy’s 2023 update covers topical therapies.
The European guideline on atopic eczema, EuroGuiDerm, published its second part, on non-systemic treatment and complementary medicine, in 2022. Its complementary medicine chapter recommends against acupuncture as standard therapy for atopic eczema, by expert consensus with 13 of 13 votes. Its evidence summary describes the Jiao review, rates its certainty low partly because the trials enrolled people with chronic rather than explicitly atopic eczema, and states that the effects may have been exaggerated. That chapter is still dated June 2022 on the European Dermatology Forum’s site.
The international urticaria guideline, last issued in 2026 after a consensus conference in December 2024 and replacing the 2022 edition, does not mention acupuncture or complementary therapies in either edition. It recommends a second-generation antihistamine first, the dose raised up to fourfold if needed, then adding omalizumab, with dupilumab, remibrutinib and ciclosporin as further options.
What the reviews could not settle
For atopic dermatitis, the two sham-controlled treatment trials enrolled 66 people between them, both run by the same Korean group, and of the two reviews one did not grade certainty and the other rated every outcome low. The Berlin trial was unblinded, and its itch result came without a difference in disease severity. The laboratory trials measured itch provoked in a single session, not the itch of daily disease. None of the trials the reviews included enrolled children.
For chronic urticaria, the reviews pool results in acupuncture’s favour against sham and waiting lists, and the 2023 trial and two of the reviews state that the gain over sham falls short of the minimal clinically important difference. Every trial in the Wu and Yu reviews came from China, three of the four reviews flag possible publication bias or high risk of bias, and the 2023 trial followed patients for four weeks after treatment. None of the reviews reports a comparison with omalizumab, the add-on the guideline recommends after antihistamines. Reading an acupuncture trial covers how sham and waiting-list comparisons differ.
AcuiQ lists the protocols individual studies prescribed for atopic dermatitis, eczema, chronic urticaria, hives and itching, each with its citation. Allergic rhinitis covers the trials in the other common allergic condition.