Trials of acupuncture for chronic constipation count bowel movements in a diary. Most use the complete spontaneous bowel movement (CSBM): one that leaves a sense of complete emptying and comes without a laxative, enema or other aid in the previous 24 hours, as a 2020 review by Lu Wang and colleagues at Chengdu University of Traditional Chinese Medicine defines it. Functional constipation is constipation with no structural or medical cause that tests can find. Cochrane has not published a review of acupuncture for chronic constipation in adults, so the evidence on this page comes from reviews in other journals and from a series of trials led by one group in Beijing.
Reviews against sham acupuncture
Sham acupuncture is a control procedure made to look like acupuncture, such as shallow needling away from the classical points; Sham acupuncture covers the kinds in use. Wang’s review searched English and Chinese databases to October 2019 and included 28 randomised trials with 3,525 people, comparing acupuncture or electroacupuncture with sham or with drugs. The authors rated with GRADE only the comparisons against sham.
| Outcome | Pooled difference with acupuncture | Certainty (GRADE) |
|---|---|---|
| CSBMs per week | 0.84 more (3 trials, 1,171 people) | Low |
| Stool form, Bristol Stool Form Scale | 0.24 points better (3 trials, 1,224 people) | Low |
| Responders | Risk ratio 2.16 (3 trials, 1,262 people) | Low |
| Constipation-related quality of life | Better, from a single trial of 1,075 people | Moderate |
| Adverse events | No difference (7 trials, 1,627 people) | Moderate |
The authors downgraded most outcomes for inconsistency between trials and for reporting bias, and concluded that the overall quality of the evidence was relatively low.
A 2020 meta-analysis by Zhang and colleagues in Lanzhou kept to electroacupuncture against sham electroacupuncture in chronic severe functional constipation, and pooled 6 trials with 1,457 people. CSBMs rose by 0.80 a week more with electroacupuncture at week 4 and by 1.25 at week 8. The difference was 1.38 a week in follow-up and narrowed to 0.87 by the twelfth week. The authors wrote that differences between trials in the points used, the depth of needling and the strength and frequency of stimulation may have influenced the results.
A 2019 network meta-analysis by Zheng and colleagues compared eight treatments that use neither drugs nor surgery, across 33 trials with 4,324 people published from 2000 to June 2016. Against placebo or sham, acupuncture was associated with more frequent stools and a higher responder rate (risk ratio 1.56), and against laxatives with more frequent stools and fewer adverse events. The authors ranked electrical nerve stimulation through the skin and acupuncture highest of the eight, and warned that small-study effects, the tendency of small trials to report larger benefits, called for caution.
The 2016 trial
The largest sham-controlled trial is a 2016 trial in Annals of Internal Medicine by Zhishun Liu and colleagues, run at 15 hospitals in China. It enrolled 1,075 people with chronic severe functional constipation: 536 to electroacupuncture and 539 to sham. Its published protocol defines severe as two or fewer CSBMs a week for more than 3 months, in adults aged 18 to 75 who met the Rome III criteria, the standard diagnostic definitions for functional bowel disorders. The protocol needled ST25 (Tianshu), SP14 (Fujie) and ST37 (Shangjuxu) on both sides, with electrical stimulation; the sham group was needled 1 to 1.5 cm deep at sites about 2 to 3 cm from ST25 and SP14 and beside ST37, with no twirling or thrusting, and connected to a stimulator whose wire had been cut inside so that no current flowed. Both groups had 28 sessions over 8 weeks.
- Over weeks 1 to 8, weekly CSBMs rose by 1.76 with electroacupuncture and 0.87 with sham, a difference of 0.90.
- Over weeks 9 to 20, after treatment had stopped, they were 1.96 and 0.89 above baseline, a difference of 1.09.
- 31.3% of the electroacupuncture group averaged 3 or more CSBMs a week during treatment and 37.7% during follow-up, against 12.1% and 14.1% with sham.
Adverse events linked to acupuncture were infrequent in both groups and all were mild or transient. The authors listed two limitations: follow-up stopped at week 20, and the acupuncturists could not be blinded. A 2024 secondary analysis of the 539 people given sham found response rates of about 9% to 11%, depending on the definition of a response, and response was more likely in people who had more CSBMs at the start. The authors, from the same research group, concluded that shallow sham needling produced no large placebo response in this condition.
Against drugs
A 2021 trial led from the same department, at Guang’anmen Hospital in Beijing, randomised 560 people with severe chronic constipation to 28 sessions of electroacupuncture over 8 weeks or to prucalopride, a drug that speeds the gut, at 2 mg a day for 32 weeks. The trial was designed to test whether electroacupuncture was no worse than the drug, within a margin of 10 percentage points. Over weeks 3 to 8, 36.2% of the electroacupuncture group and 37.8% of the prucalopride group averaged 3 or more CSBMs a week, which met that test. Adverse events occurred in 17.7% of the electroacupuncture group and 44.2% of the prucalopride group.
A 2024 meta-analysis by Xu and colleagues pooled 12 trials with 1,473 people comparing electroacupuncture with prucalopride or mosapride, drugs of the same class. It found no significant difference in weekly CSBMs, spontaneous bowel movements or stool consistency, and better scores with electroacupuncture on a constipation quality-of-life questionnaire and on anxiety and depression scales. The authors rated the certainty of every outcome low, for risk of bias, small samples and inconsistency between trials. Wang’s 2020 review, against a wider set of drugs including polyethylene glycol and lactulose, found more CSBMs with acupuncture but no significant difference in responders, with high inconsistency between trials, and did not grade those comparisons.
Opioid-induced constipation
Constipation is common in people taking opioids for cancer pain, as the authors of a 2023 trial in JAMA Network Open note. Their trial, also led from Guang’anmen Hospital, randomised 100 adults with cancer and opioid-induced constipation at 6 Chinese hospitals to 24 sessions of electroacupuncture or sham over 8 weeks, at ST25, SP14 and ST37. At week 8, 40.1% of the electroacupuncture group and 9.0% of the sham group were responders, defined as at least 3 spontaneous bowel movements a week, with an increase of at least 1 over baseline, in at least 6 of the 8 treatment weeks. Electroacupuncture did not change cancer pain or opioid dose, and the adverse events linked to it were rare and mild.
A 2026 network meta-analysis by Liu and colleagues, two of them authors of that trial, gathered 36 trials with 3,351 cancer patients testing 15 acupuncture-related methods, from needles to acupressure and ear acupuncture, alone or with laxatives. Three combinations with laxatives reduced a bowel function score more than laxatives alone, on low-certainty evidence; no significant difference appeared in weekly bowel movements, symptom scores or stool form. The authors rated most estimates low or very low in certainty, and called their rankings exploratory because several methods rested on a single trial.
What guidelines say
The Oncology Nursing Society’s 2020 guideline on constipation in people with cancer recommends acupuncture and electroacupuncture for constipation not caused by opioids in the context of a clinical trial. ONS still lists that guideline on its constipation resource page, last reviewed in August 2020. The American Gastroenterological Association and American College of Gastroenterology’s 2023 guideline on chronic idiopathic constipation covers drug treatment only, so it does not weigh acupuncture. The Japan Gastroenterological Association’s guidelines for chronic constipation 2023, published in English in 2025, make no recommendation on acupuncture. Clinical guidelines compares other bodies’ positions.
What the reviews could not settle
The 2016 sham trial, the prucalopride trial and the opioid trial come from one research group, so the trials that carry most of this page are not independent of one another. All three were run in China, and the 2016 and opioid trials needled the same three points. Where the trials come from discusses what the country of a trial tells a reader. None of the three followed people beyond 32 weeks from the start of treatment. The reviews against drugs carry low certainty or no grade, and no Cochrane review has weighed the trials together. A 2022 protocol set out an individual-patient-data meta-analysis, which pools each trial’s raw data to find which kinds of patient respond, and PubMed listed no results from it by October 2026. Electroacupuncture describes the method most of these trials tested, and Irritable bowel syndrome covers the trials in constipation-predominant IBS, and Parkinson’s disease the sham-controlled trial in constipation that comes with that disease.
AcuiQ’s constipation, functional constipation, chronic constipation and opioid-induced constipation pages list the protocols individual studies prescribed, each with its citation.