Stimulated by Deng et al 2026.[1]
MwoA – migraine without aura
VAS – visual analogue scale
RCT – randomised controlled trial
IF – impact factor
ICHD-3 – International Classification for Headache Disorders 3rd edition
MA – manual acupuncture
PSD – Park sham device
CIs – 95% confidence intervals
SR – systematic review
GRADE – Grading of Recommendations, Assessment, Development and Evaluation (a method of grading the quality or ‘certainty’ and strength or ‘effect size’ of evidence in healthcare)– key to acronyms
There are lots of big studies of acupuncture in headache, but most are concerned with the prophylaxis of chronic headache. We don’t see so many studies in acute headache, so that is why this one caught my eye.
The first one I remember seeing in acute migraine pitted acupuncture head-to-head with an injection of sumatriptan (n=179).[2] There was a 3rd group with a placebo injection. A full migraine attack within 48 hours was prevented in 21 of 60 patients (35%) receiving acupuncture, 21 of 58 patients (36%) receiving a sumatriptan injection, and 11 of 61 patients (18%) receiving a placebo injection.
Subsequently, there have been a couple of moderately large sham controlled trials from China. One from Chengdu (n=175) in 2009,[3] and one from Beijing (n=150) in 2012.[4] Both showed significant benefits for real acupuncture over sham, but the reductions in VAS pain were far from impressive. Interestingly, both trials utilised more than 1 sham group, and all the sham interventions involved needle penetration.
Ear acupuncture was tried in 2011 in 90 women with MwoA in Turin.[5] Patients were randomised to having ASP studs (see Ears and the battlefield from 2019) inserted at sensitive points in an area (just inside the antitragus) that is supposed to be specific for migraine or non-sensitive points on a different part of the ear (the inferior antihelix crus). The results were dramatically positive for needling in one part of the ear compared with the other. The lack of effect in the control group that received 4 ASP studs is somewhat dubious to my mind.
There is one further RCT of ear acupuncture in acute migraine (n=60) from Iran.[6] Both groups had needles inserted in their ears and it looks as though everybody was better after a couple of hours. Rescue medication was applied after 10 minutes if any patient had not had at least a 30mm reduction in their VAS pain. The rescue medication was an injection of dexamethasone. This paper did not convince me either.
Finally, I found a paper that used EA between TE5 and GB20 for 30 minutes compared with EA at non-meridian points (n=110).[7] The paper is in Chinese but has an English abstract. The immediate analgesic effect was 87% versus 53%, although it is not clear exactly how that was determined and to what time point immediate refers.
The current trial is not quite as big as the first 3 discussed above (n=80), but it is the first to employ a non-penetrating sham. It was published in the journal Headache (IF 4.0) last week.
The trial seems to have been based at a single centre in Kunming, Yunnan province, which is in the southwest of China, just above Laos and Vietnam. I have contacts in Kunming because the university there has a museum dedicated to the spread of Chinese Medicine in the West. This includes some artifacts curated from the BMAS including books by Felix Mann, but the main focus is really on the French connection and Soulie de Morant.
80 patients with unilateral headache and meeting the ICHD-3 criteria for MowA were recruited if their headache VAS was between 4 and 9. They were randomised to either real or non-penetrating sham MA at the following points on the ‘healthy’ side of the body – TE3, TE5, GB41, GB34, GB20, GB8. The PSD was used in both groups, and a fancy hair grip was used for each of the points on the head (GB20 and GB8).
Healthy side presumably refers to the opposite side from the headache, and it looks as though the needling was performed starting from distal points and ending with those on the head.
The primary outcome was the analgesic response at 10 minutes. The former was defined as a 50% reduction in headache VAS from baseline.
Over ¾ of patients were female and the average age was 38. They had suffered with their current headache for an average of over 9 hours, and the baseline headache VAS was over ~6.5.
37 out of 40 patients in the real MA group achieved an analgesic response at 10 minutes, but only 5 of 40 in the sham MA group. This difference was of course very highly statistically significant.
This seems like a very dramatically positive study, but it looks to me as if all the patients still had headaches at the 10-minute time point, since the mean VAS scores were 2.0 and 4.6 respectively, with very tight CIs (1.8 to 2.3 and 4.4 to 4.9).
I was hoping that the sham group might get treated after the measurement of the primary outcome, but there is no mention of this.
Having spent a couple of days looking at these trials it seems that the one of the key problems in trying to make comparisons between them is the differences in populations included. For example, the first one I mentioned recruited patients who thought they were about to have a migraine,[2] whereas in the second and third the patients appeared to have already been suffering with a migraine attack for 24 hours.[3,4] By contrast, one of the ear acupuncture studies recruited patients who had been suffering for less than 4 hours,[5] and the other did not give any information concerning the duration of the migraine attack.[6]
So, it is a tricky to summarise all this neatly. An SR from 2023 found most of the papers I have mentioned here,[8] apart from the one that is just out,[1] of course. It also included a bunch more in the Chinese literature. The SR had a couple of outcomes with moderate certainty (GRADE). Acupuncture was associated with a modest improvement in headache intensity at 2 hours compared with sham and did not have a significantly greater rate of AEs.
References
1 Deng Y, Zhang S, Yao F, et al. Acupuncture for acute treatment of migraine without aura: A randomized clinical trial. Headache. Published Online First: 9 June 2026. doi: 10.1111/head.70144
2 Melchart D, Thormaehlen J, Hager S, et al. Acupuncture versus placebo versus sumatriptan for early treatment of migraine attacks: a randomized controlled trial. J Intern Med. 2003;253:181–8. doi: 10.1046/j.1365-2796.2003.01081.x
3 Li Y, Liang F, Yang X, et al. Acupuncture for treating acute attacks of migraine: a randomized controlled trial. Headache. 2009;49:805–16. doi: 10.1111/j.1526-4610.2009.01424.x
4 Wang L-P, Zhang X-Z, Guo J, et al. Efficacy of acupuncture for acute migraine attack: a multicenter single blinded, randomized controlled trial. Pain Med. 2012;13:623–30. doi: 10.1111/j.1526-4637.2012.01376.x
5 Allais G, Romoli M, Rolando S, et al. Ear acupuncture in the treatment of migraine attacks: a randomized trial on the efficacy of appropriate versus inappropriate acupoints. Neurol Sci. 2011;32 Suppl 1:S173-175. doi: 10.1007/s10072-011-0525-4
6 Farahmand S, Shafazand S, Alinia E, et al. Pain Management Using Acupuncture Method in Migraine Headache Patients; A Single Blinded Randomized Clinical Trial. Anesthesiol Pain Med. 2018;8:e81688. doi: 10.5812/aapm.81688
7 Zhang H, Hu Y, Wu J, et al. [Timeliness law on the immediate analgesia on acute migraine treated with electroacupuncture at shaoyang meridian points]. Zhongguo Zhen Jiu. 2015;35:127–31.
8 Wang Y, Du R, Cui H, et al. Acupuncture for acute migraine attacks in adults: a systematic review and meta-analysis. BMJ Evid-Based Med. 2023;28:228–40. doi: 10.1136/bmjebm-2022-112135
The BMAS Blog 
You must be logged in to post a comment.