Functional dyspepsia

Functional dyspepsia is recurring discomfort in the upper abdomen that a routine clinical evaluation cannot explain. The Rome V criteria, published in 2026 by Törnblom and colleagues, divide it into two subtypes that can overlap: postprandial distress syndrome, in which meals bring on fullness or early satiation (feeling full after a few mouthfuls), and epigastric pain syndrome, in which pain or burning in the upper abdomen does not come only after eating. The trials below diagnosed people under the earlier editions, Rome II to Rome IV, and the largest recent trials studied postprandial distress. The 2014 Cochrane review found the evidence too weak to judge; a 2026 meta-analysis with more than three times as many trials rated a benefit over sham acupuncture as high certainty, from trials all run in Asia.

The Cochrane review

The 2014 Cochrane review by Lan and colleagues included 7 randomised trials with 542 people diagnosed by Rome II or Rome III, comparing manual acupuncture or electroacupuncture with drugs, sham acupuncture or no treatment. Most trials gave too little detail about how patients were allocated, and most were not blinded.

• Against drugs. In 4 trials against the prokinetics cisapride, domperidone and itopride, which speed stomach emptying, acupuncture made no statistically significant difference to symptom scores or to how often symptoms flared.
• Against sham. In 3 trials, acupuncture improved symptom scores and quality-of-life and anxiety and depression scales as much as or more than sham acupuncture.
• Adverse events. One trial recorded fewer adverse events with acupuncture than with cisapride, all minor; the sham trials found no difference.

The authors graded all the evidence low or very low quality and concluded that it remains unknown whether acupuncture is more effective or safer than other treatments for functional dyspepsia. No later version has appeared in the Cochrane Library.

Later reviews

A 2026 meta-analysis by Li and colleagues, with Lan among the authors, gives the Cochrane review’s record as its registration. It searched six databases and two trial registries to March 2025 and included 23 trials with 2,454 people: 13 against sham acupuncture, 4 against no treatment or usual care, and 8 against drugs, with two three-arm trials counted in two groups. All 23 were run in Asia, 21 in China and 2 in South Korea. Symptoms were scored on the Nepean Dyspepsia Symptom Index, which runs from 0 to 195, and quality of life on the Nepean life-quality index, from 0 to 100.

The sham result for symptoms comes from the 5 trials at low risk of bias only, because trials at high risk of bias gave a different answer and the authors reported the two groups separately. Effects on anxiety and depression were rated very low certainty. Manual acupuncture and electroacupuncture did not differ. The trials most often used ST36 (Zusanli), CV12 (Zhongwan) and PC06 (Neiguan), typically 3 to 5 sessions a week for about 4 weeks.

A 2017 overview and network meta-analysis by Ho and colleagues took 22 trials from 5 earlier systematic reviews and compared acupuncture-based treatments, used alone or added to prokinetics, with prokinetics alone. Manual acupuncture eased overall symptoms marginally more than domperidone or itopride, and manual acupuncture combined with the prokinetic clebopride ranked most likely to be the most effective option. No serious adverse events were reported.

A 2022 meta-analysis by Liu and colleagues measured the sham arms themselves. Across 13 trials, 8 of them pooled, symptom scores improved in the groups given sham manual acupuncture (standardised mean difference −0.42) and sham electroacupuncture (−0.54), and quality of life improved with sham manual acupuncture. Two objective measures of the stomach’s electrical activity, dominant frequency and dominant power, did not change. Sham acupuncture explains what a sham control can and cannot rule out.

Landmark trials

• Acupoints against sham and a drug, 2012. A trial by Ma and colleagues randomised 712 people to six groups: four sets of acupuncture points, sham needling at non-acupoints, and itopride. Treatment ran five sessions a week for 4 weeks, with 12 weeks of follow-up. All groups improved. Response was 70.7% in the group needled at selected points on the stomach meridian and 34.8% with sham, and the stomach-meridian group also did better than itopride and the other acupuncture groups.
• Refractory dyspepsia against sham, 2018. A trial by Zheng and colleagues at three Chinese hospitals gave 200 people whose symptoms had not responded to drugs 20 sessions of electroacupuncture or sham electroacupuncture over 4 weeks. At week 16, 17% against 6% had no dyspeptic symptoms at all, and 62% against 22% reported adequate relief. Results were similar at week 24, and no serious adverse event was reported.
• Refractory dyspepsia added to a drug, 2019. A single-centre trial by Chung and colleagues randomised 132 people with postprandial distress that had not responded to acid-suppressing or prokinetic drugs. Both groups took gastrocaine, a local anaesthetic combined with antacids, as needed for 12 weeks, and one group also had 20 electroacupuncture sessions over 10 weeks. The assessors were blinded but the patients were not. At week 12 electroacupuncture gave adequate relief to one extra person for every 2.4 treated.
• Postprandial distress against sham, 2020. A trial in 5 Chinese hospitals by Yang and colleagues randomised 278 people aged 18 to 65 who met Rome IV criteria for postprandial distress syndrome to 12 sessions of acupuncture or sham acupuncture over 4 weeks. At week 4, 83.0% against 51.6% responded, and 27.8% against 17.3% were free of all three main symptoms: fullness after meals, upper abdominal bloating and early satiation. The difference held through 12 weeks of follow-up, and no serious adverse event occurred. The authors listed as limitations the lack of objective outcomes, the 18% of participants missing at week 16, and that the acupuncturists could not be blinded.

What guidelines say

The American College of Gastroenterology and Canadian Association of Gastroenterology guideline of 2017, by Paul Moayyedi and colleagues, states in its statement 12 that it does not recommend routine use of complementary and alternative medicines for functional dyspepsia (conditional recommendation, very low quality evidence). Its supporting text cites the 2014 Cochrane review and adds that such treatments may suit individual patients who know the evidence cannot yet show their benefit or risk. The 2017 document is still the dyspepsia guideline that the ACG lists in October 2026.

The British Society of Gastroenterology’s 2022 guideline by Black and colleagues makes no recommendation on acupuncture. In its section on drugs in development and other therapies it notes that acupuncture beat a sham procedure for postprandial distress in one Chinese trial of more than 200 patients, the Yang trial above, and that the result needs confirming in other regions. It also cites the Chung trial of electroacupuncture with gastrocaine as giving symptom relief in refractory dyspepsia. Clinical guidelines sets out other bodies’ positions.

What the reviews could not settle

Every trial in the 2026 meta-analysis came from China or South Korea, and the British guideline asks for the sham result to be confirmed in other regions. Where the trials come from discusses what the country of a trial tells a reader. The same review found little evidence for epigastric pain syndrome, for older adults or for people with psychiatric conditions alongside their dyspepsia. Acupuncturists cannot be blinded, and the sham arms of these trials improve on symptom questionnaires but not on objective measures, so how much of the difference over sham comes from needling at particular points, which the Ma trial was designed to test, remains open. The Cochrane review has not been updated since 2014, and the American guideline rests on it. Irritable bowel syndrome reports the trials in a related disorder of the gut.

AcuiQ’s functional dyspepsia and epigastric pain pages list the protocols individual studies prescribed, each with its citation.