Asthma

The Global Initiative for Asthma (GINA) describes asthma in its 2026 strategy report as a varied disease, usually with chronic inflammation of the airways, defined by wheeze, shortness of breath, chest tightness and cough that come and go, together with airflow out of the lungs that varies over time. Trials of acupuncture for asthma test two kinds of treatment. One is needling, with or without an electric current or heat. The other is point application, in which a paste of powdered herbs is taped over acupuncture points on the back and left for hours; in its best-known form, sanfu application, the plasters go on during sanfu, the hottest days of summer, to prevent attacks in winter. Cochrane last searched the literature on needling in 2008. The trials since then come mostly from China, and the largest placebo-controlled trial of either treatment, published in 2026, tested the summer plasters.

The Cochrane review

The Cochrane review by McCarney, Brinkhaus, Lasserson and Linde was published in 2004, updated with a search to August 2008 and edited in 2009 without any change to its conclusions; no later version has been published. It took in randomised trials of needle acupuncture or other ways of stimulating acupuncture points, against sham, no added treatment or an active treatment, in people with asthma or asthma-like symptoms. Twelve studies with 350 participants met its criteria. The reviewers judged trial reporting poor and the quality of the trials too low to generalise from, and noted that some trials used as their sham points that traditional Chinese medicine itself uses for asthma. They found no statistically significant or clinically relevant effect of acupuncture against sham. Two small studies could be pooled for lung function after treatment, measured as FEV1, the volume of air a person can blow out in 1 second, and showed no difference (standardised mean difference 0.12, 95% CI −0.31 to 0.55). The reviewers concluded that there was not enough evidence to make recommendations about acupuncture for asthma, and asked future trials to compare acupuncture with no treatment as well as with sham.

Later reviews

Reviews since the Cochrane search draw on trials published after it, nearly all run in China. Their comparisons differ, so the table keeps them apart. Sham acupuncture is a control procedure made to look like the real one, and a placebo plaster is one made to look and feel like the herbal plaster with little or no active herb in it; Sham acupuncture covers the kinds in use. GRADE is the system reviewers use to rate how certain a pooled result is, from high to very low.

Against sham, Pang’s review, by a team from Guangzhou University of Chinese Medicine and RMIT University in Melbourne, found FEV1 as a share of the predicted value 6.11 points higher with acupuncture (95% CI 0.54 to 11.68; 603 people), but the trials disagreed widely with each other (I² 93%). Scores on Cai’s Asthma Quality of Life Questionnaire, a Chinese instrument, were 7.26 points better (358 people), and symptom scores lower (120 people). One trial, of 111 people, reported a 2-point better score on the Asthma Control Test and one fewer attack a year. Other measures of lung function, and medication use, did not differ significantly. The authors called for larger trials with appropriate controls.

Wang’s review in children, from Tianjin University of Traditional Chinese Medicine, pooled trials of needling, catgut embedding (a short length of suture left under the skin at a point), electrical stimulation through skin pads, laser and press needles. Added to standard treatment, these were associated with FEV1 as a share of predicted 6.02 points higher, and no significant difference in the ratio of FEV1 to total exhaled volume or on the Pediatric Asthma Quality of Life Questionnaire (difference 0.01 points). The authors downgraded most outcomes to low certainty for risk of bias and inconsistency. Only 1 of the 16 trials reported adverse events. The trials used BL13 (Feishu) most often, in 15, followed by ST36 (Zusanli) in 13, Dingchuan (XB01B, written EX-B1 in the WHO standard) in 12, and CV17 (Danzhong) in 11. Acupuncture for children covers what the safety literature records for younger patients.

A pragmatic trial in Germany

Benno Brinkhaus, one of the Cochrane authors, led a 2017 pragmatic trial from the Charité in Berlin. A pragmatic trial tests a treatment as it would be given in practice, here against routine care alone and with no sham. It randomised 357 adults with allergic asthma: 184 to up to 15 acupuncture sessions over 3 months on top of routine care, and 173 to routine care alone. The primary outcome was the Asthma Quality of Life Questionnaire (AQLQ), scored from 1 to 7. At 3 months the acupuncture group scored 0.7 points higher (95% CI 0.5 to 1.0), with better physical and mental scores on the SF-36, a general health questionnaire, and the difference held at 6 months. A further 1,088 people who declined randomisation had acupuncture and were followed as a cohort; they improved by a similar amount. With no sham arm, the trial cannot separate the needling from the attention and expectation that come with it.

What a sham can change

A 2011 crossover trial in the New England Journal of Medicine by Michael Wechsler, Ted Kaptchuk and colleagues at Brigham and Women’s Hospital in Boston was not a test of acupuncture, but it bears on how asthma trials are read. Each of 46 people with asthma received, in random order and three times over, an albuterol inhaler (a bronchodilator), a placebo inhaler, sham acupuncture with a validated sham needle, and no intervention. Among the 39 who finished, FEV1 rose 20.1% with albuterol and about 7% with each of the other three, which did not differ from one another. Asked how much better they felt, people reported 50% improvement with albuterol, 45% with the placebo inhaler, 46% with sham acupuncture and 21% with no intervention. The authors concluded that self-reported outcomes in asthma can be unreliable and that trials need a no-treatment arm to show what happens untreated. GINA’s 2026 report cites the trial for the point that sham treatments can control symptoms while leaving airway inflammation untreated.

Summer plasters

Point application has its own literature, set out in Point application. In Wei’s review of 11 trials in children, the herbs used most were white mustard seed, corydalis and kansui, and the points BL13, CV17, BL17 (Geshu) and CV22 (Tiantu). Added to Western medicine, the plasters were associated with 1.62 fewer acute attacks a year and better peak flow and FEV1; the authors asked for caution because of the trials’ methodological flaws. Three trials reported adverse events, mainly redness, itching, stinging and blisters, in 16.1% of children given the plaster. Against placebo, Wu’s review found 1.84 fewer acute attacks with short-term use and AQLQ scores 6.57 points higher, both on moderate-certainty evidence.

The two largest placebo-controlled trials disagree. In a 2010 trial at the University of Hong Kong, published in 2016, Zhu and colleagues randomised 323 people over 13 with asthma in remission to herbal or placebo plasters on three days that summer, added to their usual treatment. Lung function, the primary outcome, did not differ, and neither did the questionnaire scores; a sub-analysis of follow-up questionnaires reported less need for medicine in the herbal group.

A 2026 multicentre trial by Xu and colleagues at nine hospitals in China gave everyone placebo plasters in the summer of 2015 to set a baseline, then randomised 835 people with asthma to real (418) or placebo (417) plasters in 2016 and 2017. The real plaster held 3 g of white mustard seed, asarum, corydalis and kansui in ginger juice, on BL13, BL15 (Xinshu) and BL17 on both sides, four times a summer for 2 to 6 hours each. The placebo held 0.2 g of herb powder with corn cob, starch and pigment, also in ginger juice so that it stung a little, at sites further from the spine. Over 24 months attacks fell by 4.08 with the real plaster and 2.66 with placebo, a difference of 1.42 (95% CI 0.69 to 2.14). FEV1 as a share of predicted was higher at 24 months but not at 12; peak flow and the Asthma Control Test did not differ. Treatment-related adverse events occurred in 19.4% of the real-plaster group and 5% of the placebo group, mostly skin allergy (17.7% against 3.4%), with no serious events. The authors noted that 22.9% dropped out and that the placebo may not have been inert.

What guidelines say

The British guideline on the management of asthma (SIGN 158) from the British Thoracic Society and the Scottish Intercollegiate Guidelines Network was published in July 2019 and revised in November 2024. Its section on acupuncture, unchanged in the revision, summarises the 2000 version of the Cochrane review, which found no clinically valuable benefit and no effect on lung function, and a later meta-analysis that found no effect on asthma severity and raised concern about publication bias. It makes no graded recommendation. SIGN lists its sections on non-pharmacological management as under review.

The joint BTS, NICE and SIGN guideline NG245, published on 27 November 2024, replaces NICE’s 2017 guideline and the parts of SIGN 158 on diagnosis, monitoring and chronic asthma management. Its recommendations do not mention acupuncture or complementary therapies. GINA’s 2026 report lists the non-drug strategies it advises, including stopping smoking, physical activity, breathing exercises and weight reduction, and neither acupuncture nor point application is among them. Allergic rhinitis reports the trials of acupuncture and summer plasters for hay fever.

What the reviews could not settle

No Cochrane review has weighed the trials published since 2008. The sham-controlled needling trials pooled in 2023 disagreed with each other on lung function, and the quality-of-life gain rests on a Chinese questionnaire. In children, the trials compared acupuncture added to medicine with medicine alone, and almost none reported adverse events. The Wechsler trial shows that in asthma a sham can match an active drug on how people feel while leaving lung function where it was, so trials scored on symptoms or quality of life without a sham, such as the German pragmatic trial, cannot isolate the needling. For summer plasters the 2026 trial reports fewer attacks than placebo while the Hong Kong trial found no difference in lung function, the plasters cause skin reactions, and the formula and points vary from trial to trial. COPD reports the trials in the other common airway disease.

AcuiQ’s asthma, bronchial asthma and cough-variant asthma pages list the protocols individual studies prescribed, each with its citation.