The NICE guideline NG206 for England describes myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) by four symptoms that must all be present: debilitating fatigue that activity worsens and rest does not significantly relieve; post-exertional malaise, a worsening of symptoms after activity that is often delayed by hours or days, out of proportion to the activity and slow to recover from; unrefreshing or disturbed sleep; and cognitive difficulties, sometimes called brain fog. It diagnoses the condition when these have lasted 3 months and no other condition explains them. Most trials of acupuncture used older case definitions, which matters for reading them, and the trials measure fatigue on questionnaires such as the Chalder Fatigue Scale, the 14-item Fatigue Scale (FS-14) and the Fatigue Severity Scale (FSS).
No Cochrane review
Cochrane has published no review of acupuncture or moxibustion for ME/CFS, and a search of PubMed in October 2026 found no protocol for one. The nearest Cochrane work in Chinese medicine is a 2009 review of Chinese herbal medicine for chronic fatigue by Adams and colleagues, which located more than 2,400 references and included none, because no study met its criteria; it noted that many trials labelled randomised in China had not used rigorous randomisation. Cochrane withdrew that review in 2018.
Later reviews
A 2008 review by Wang and colleagues of the Chinese acupuncture and moxibustion literature found that every study reported the treatment effective, with response rates of 79% to 100%, and that none was a randomised controlled trial. The reviews below pool the randomised trials that followed. Most of those trials report a response rate, also called an effective rate: the share of patients judged improved by a set amount. A risk ratio (RR) or odds ratio (OR) above 1 favours the treatment arm. A standardised mean difference (SMD) states the gap between groups in units of the spread of scores, so that different fatigue scales can be pooled, and I² is the share of the variation between trials that chance does not explain.
| Review | Trials and people | Comparisons | Main pooled result | Certainty |
|---|---|---|---|---|
| Wang 2017, searched to May 2016 | 31 trials | Herbal medicine, Western medicine and sham, analysed separately | Response rate RR 5.90 against sham, 1.51 against Western medicine, 1.22 against herbal medicine, for a single acupuncture or moxibustion method | Not graded; trials of relatively poor quality |
| Zhang 2019, searched to March 2018 | 16 trials, 1,346 people | Sham, herbal medicine and other controls, analysed separately | Response rate RR 2.08 against sham (4 trials, 281 people, I² 64%); RR 1.17 against herbal medicine (3 trials, 290 people) | Low (GRADE) |
| You 2021, moxibustion, searched to January 2021 | 15 trials, 1,030 people | Acupuncture and drugs; no sham | Response rate OR 4.58 against acupuncture and 6.36 against drugs; FS-14 2.2 points lower | Moderate against acupuncture; low or very low for the other outcomes (GRADE) |
| Fang 2022, network meta-analysis, searched to August 2021 | 51 trials, 3,473 people | Acupuncture, moxibustion, herbal medicine, Western medicine and no treatment, compared in one network | Moxibustion and its combinations ranked highest on response rate and FS-14 | Low (GRADE) |
| Li 2026, searched to April 2026 | 8 trials, 715 people | Usual care, drugs, lifestyle advice, herbal medicine, non-acupoint and sham needling, pooled | Overall fatigue SMD −1.31 (95% CI −2.16 to −0.47; I² 91.7%); no difference in adverse events | Risk of bias some concerns in 7 trials, high in 1 |
| Zhang 2026, network meta-analysis, searched to January 2026 | 29 trials, 2,234 people | Moxibustion, acupuncture, massage, cupping and other methods against conventional care | Overall fatigue lower with moxibustion than conventional care (SMD −1.84) | Assessed with CINeMA; authors advise caution |
Wang’s review, from the Third Military Medical University in Chongqing, ranked combined acupuncture and moxibustion first and sham acupuncture last on response rate. Zhang’s review, from Huazhong University of Science and Technology in Wuhan, kept the sham comparison apart and graded it: the authors judged most trials of low methodological quality and wrote that no firm conclusion could be reached, owing to limited data, poor quality and potentially exaggerated effect sizes. In Li’s review, from Guangdong Medical University, 7 of the 8 trials were run in China and 1 in Korea, treatment lasted 14 days to 4 weeks, and only one trial followed patients after treatment ended. The authors noted that the estimate against sham (SMD −0.92) and the estimate against usual care (−0.85) carry different meanings, the second including the effects of attention and expectation, and that pooling every control together may obscure which is which.
An overview of 10 systematic reviews by Yin and colleagues at Chengdu University of Traditional Chinese Medicine, searched to September 2019, rated the methods of every one as critically low on AMSTAR 2, a checklist for appraising reviews. Of the 17 outcomes it graded on GRADE, the scale reviewers use from high to very low certainty, none was high, 3 were moderate, 10 low and 4 very low. A 2020 review of all randomised trials for ME/CFS by Kim and colleagues at Daejeon University in Korea, drugs and non-drug treatments alike, found 55 trials with 6,316 people. Acupuncture was one of 8 interventions with a statistically significant result, and the authors found no intervention effective with coherence and reproducibility.
Trials
The one sham-controlled trial that reached NICE’s review is a 2013 trial in Hong Kong by Ng and Yiu at the University of Hong Kong. It randomised 127 people with chronic fatigue syndrome to 8 sessions of acupuncture over 4 weeks or a sham using blunt needles that pricked the skin without entering it, at the same points; 99 completed treatment. Both groups improved. The authors reported effect sizes of 0.92 for physical fatigue and 0.78 for mental fatigue within the sham group, and net effect sizes of moderate magnitude in favour of acupuncture, at 0.52 and 0.63. They suggested that pressure from the sham needles on the points may have contributed to the sham group’s improvement. Sham acupuncture covers why a sham is hard to make inert.
A 2015 multicentre trial by Kim and colleagues ran at four Korean hospitals. It randomised 150 people with chronic fatigue syndrome or idiopathic chronic fatigue, fatigue of 6 months or more that does not meet the full case definition, to body acupuncture, Sa-am acupuncture (a Korean method that selects points by five-phase theory), or usual care alone. Both acupuncture arms had 10 sessions over 4 weeks, and no one was blinded. Body acupuncture used GV20 (Baihui), GB20 (Fengchi) and six points on the back: BL11 (Dazhu), BL13 (Feishu), BL15 (Xinshu), BL18 (Ganshu), BL20 (Pishu) and BL23 (Shenshu). One week after treatment, FSS scores were lower with body acupuncture than with usual care, a difference of 0.43 points on the 7-point scale (95% CI 0.05 to 0.81); by 13 weeks the difference was no longer statistically significant, and Sa-am acupuncture did not differ from usual care on the FSS at either point. Two of the 10 adverse events recorded were linked or possibly linked to acupuncture, both mild.
For moxibustion, the burning of mugwort near the skin (Moxibustion describes the methods), a 2013 double-blind trial by Kim and colleagues at Daejeon University randomised 44 people to indirect moxibustion or a placebo at CV04 (Guanyuan) and CV08 (Shenque), three times a week for 4 weeks, and reported lower fatigue scores with moxibustion. Its participants had idiopathic chronic fatigue rather than chronic fatigue syndrome.
What guidelines say
NICE NG206, published on 29 October 2021, replaced NICE’s 2007 guideline. NICE last reviewed it on 24 January 2025, found no new evidence that affects the recommendations, and will review it again only if new evidence is likely to change them. Its recommendations do not mention acupuncture. They state that there is currently no cure for ME/CFS, pharmacological or non-pharmacological; tell clinicians not to offer any medicines or supplements to cure it, not to offer graded exercise therapy, defined as fixed incremental increases in activity, and not to offer the Lightning Process; and centre management on energy management, a self-management strategy for staying within a person’s energy limits. In the guideline’s evidence review G, the committee assessed the Hong Kong trial and found no clinically important difference between acupuncture and sham on quality of life, fatigue or psychological status, at very low quality on GRADE; it noted that the trial’s randomisation may in fact have been alternation, and that the 1994 CDC case definition the trial used does not require post-exertional malaise. Across all complementary therapies, the committee wrote that there was “not enough robust evidence to recommend any type of complementary therapy for ME/CFS”.
The US Centers for Disease Control and Prevention (CDC) publishes clinical care advice for health professionals, last updated in May 2024, which is not a graded guideline. Under pain, it states that acupuncture by a licensed practitioner might help with pain for some patients. It recommends activity management, or pacing, for post-exertional malaise, and warns that standard exercise advice for healthy people can substantially harm people with ME/CFS. Clinical guidelines lists what other guideline bodies say about acupuncture.
What the reviews could not settle
The reviews could not separate the needling from the attention around it. Most trials compared acupuncture with drugs, herbal medicine or usual care, with patients aware of their treatment and rating their own fatigue, and the comparison against sham rests on a few small trials, among them the Hong Kong trial whose sham group improved substantially and which NICE read as showing no clinically important difference. They could not say how long any change lasts, since almost no trial followed patients after a course of 2 to 4 weeks. Nor could they say whether the trials studied the condition NICE now defines: most used the 1994 CDC definition, under which post-exertional malaise is optional, and some enrolled people with idiopathic chronic fatigue alongside ME/CFS. Reviews of acupuncture for related conditions are in Fibromyalgia and Cancer-related fatigue.
AcuiQ’s chronic fatigue syndrome and fatigue pages list the protocols individual studies prescribed, each with its citation.