Trials of acupuncture for irritable bowel syndrome (IBS) split according to their control group. Trials against sham acupuncture, in which neither patient nor assessor is meant to know who received the real treatment, have mostly found small differences or none. Trials against usual care or drugs, which cannot be blinded, have mostly found acupuncture ahead. The 2012 Cochrane review by Eric Manheimer and colleagues at the University of Maryland set out that split, and the trials since have added to both sides of it.
The Cochrane review
The review included 17 randomised trials with 1,806 adults, searched to November 2011 in English and Chinese databases. Five compared acupuncture with sham, and the authors rated those five at low risk of bias. The others had a high risk of bias, from lack of blinding or weak randomisation.
| Comparison | Finding | Quality (GRADE) |
|---|---|---|
| Against sham acupuncture | No difference in symptom severity (standardised mean difference −0.11, 95% CI −0.35 to 0.13; 4 trials, 281 people) or quality of life (−0.03; 3 trials, 253 people) | Moderate, because the data were sparse |
| Against antispasmodic drugs | Symptoms improved in 84% with acupuncture and 63% with drugs (risk ratio 1.28, 95% CI 1.12 to 1.45; 5 trials, 449 people) | Low, because the trials were unblinded and small |
| Against no specific treatment | 63% against 34% improved (risk ratio 2.11; 2 trials, 181 people) | Not rated in the abstract |
| Against a probiotic (Bifidobacterium) or psychotherapy | No difference (1 to 2 trials each) | Not rated in the abstract |
| Added to Chinese herbal medicine, against the herbs alone | 93% against 79% improved (risk ratio 1.17; 4 trials, 466 people) | Not rated in the abstract |
The drug comparisons came from Chinese trials against pinaverium bromide and trimebutine maleate, antispasmodics that the authors note have modest benefit in IBS. They wrote that future trials might show whether acupuncture’s larger reported benefit over these drugs comes from patients’ preference for acupuncture or their greater expectation of it. Nine trials reported adverse events, and they recorded one, a faint during treatment. Sham acupuncture explains what a sham control can and cannot rule out.
A network meta-analysis
A 2019 overview and network meta-analysis by Wu and colleagues in Hong Kong drew 27 trials with 2,141 people from 15 systematic reviews, which it rated of mediocre quality. None of the 27 trials blinded properly. Needle acupuncture and electroacupuncture each improved overall symptoms more than pinaverium bromide, and acupuncture combined with a Chinese herbal formula ranked highest of 14 options. The authors wrote that the risk of bias and inconsistency in the network limited how far the ranking could be trusted.
Trials since the review
Four trials show the split by control group.
- Against sham, 2009. A trial by Lembo and colleagues randomised 230 adults to 6 sessions of real or sham acupuncture over 3 weeks, after a 3-week run-in on sham, with a waiting-list group alongside. Global improvement was 41% with real and 32% with sham acupuncture, a difference the trial could not distinguish from chance. Both acupuncture groups together improved more than the waiting list (37% against 4%).
- Added to usual care, 2012. A UK primary care trial by Hugh MacPherson and colleagues offered 116 people 10 weekly individualised sessions on top of usual care and kept 117 on usual care alone. At three months the acupuncture group’s symptom score had fallen by 27 points more, and 49% against 31% met the trial’s definition of success, a number needed to treat of 6: six people offered acupuncture for each extra success. The difference largely persisted at 6, 9 and 12 months.
- Against drugs, 2020. A trial in 7 Chinese hospitals by Pei and colleagues randomised 531 people, two to one, to 18 acupuncture sessions over 6 weeks or to polyethylene glycol for constipation-predominant IBS or pinaverium bromide for diarrhoea-predominant IBS. The symptom severity score fell by 124 points with acupuncture and 95 with drugs, a difference of 29 points, and the gap lasted to 12 weeks of follow-up.
- Against sham, 2025. The ACTION trial by Yang and colleagues, in 6 Chinese hospitals, randomised 280 people with diarrhoea-predominant IBS to 15 sessions of acupuncture or of sham acupuncture with blunt needles at non-acupuncture points over 6 weeks. At week 6, 57.9% against 41.4% met the combined target of at least 30% less abdominal pain and at least 50% fewer days with diarrhoea (risk ratio 1.40). The difference appeared from week 3 and held to week 18 at every point except week 16. No serious adverse event was reported. The journal has since published two comments on the trial, one of them on its methods.
What guidelines say
NICE’s guideline on IBS in adults, CG61, states in recommendation 1.2.4.1 that “the use of acupuncture should not be encouraged for the treatment of IBS”. The recommendation dates from the guideline’s publication in 2008, before the Cochrane review. NICE updated other parts in 2017 and, on review in April 2025, decided not to update the section on clinical management. The British Society of Gastroenterology’s 2021 guideline on managing IBS does not mention acupuncture. Clinical guidelines sets out other bodies’ positions.
What the reviews could not settle
The Cochrane review has not been updated since 2012, so no review with Cochrane methods has weighed the ACTION trial against the earlier sham trials that found no difference. Whether the advantage over drugs and usual care reflects the needling, the attention of a practitioner or patients’ expectations is the question the Cochrane authors left open, and unblinded trials cannot answer it. ACTION enrolled only people with diarrhoea-predominant IBS, and the two most recent trials above, both favouring acupuncture, were run in China. Where the trials come from discusses what the country of a trial tells a reader. Constipation reports the trials in chronic constipation outside IBS. Functional dyspepsia reports them in a disorder of the upper gut.
AcuiQ’s irritable bowel syndrome, diarrhoea-predominant IBS and constipation-predominant IBS pages list the protocols individual studies prescribed, each with its citation.