In recent decades, acupuncture has attracted extensive research spanning an astonishingly wide array of medical conditions, from chronic pain and neurological disorders to infectious diseases and psychiatric ailments. However, the proposed mechanisms of action—ranging from peripheral sensory stimulation to central nervous system modulation—fail to provide a coherent, biologically plausible explanation for efficacy across this disparate spectrum (Zhao et al., 2022; WHO, 2003).
The aim of this post is to examine the breadth of published acupuncture trials, delineate the leading scientific hypotheses for its mode of action, and outline the profound implausibility of these mechanisms universally applying to such varied pathologies, ultimately framing acupuncture as non-specific rather than a specific therapeutic modality (Meissner et al., 2019; Ernst, 2018).
Acupuncture has been subjected to thousands of randomized clinical trials (RCTs) and systematic reviews across virtually every medical specialty. A comprehensive 2022 evidence map published in BMJ Open synthesized 120 systematic reviews, encompassing 1,402 individual RCTs and addressing 77 distinct conditions within 12 broad therapeutic categories (Zhao et al., 2022). These categories include neurological disorders, musculoskeletal conditions, cardiovascular diseases, and beyond, reflecting a research enthusiasm that transcends conventional biomedical boundaries.
Neurological applications dominate, with trials targeting stroke sequelae such as hemiplegia and aphasia, vascular dementia symptoms, migraines, tension headaches, and facial nerve palsies like Bell’s palsy (Li et al., 2022; Zhao et al., 2022; WHO, 2003). Musculoskeletal trials are equally prolific, examining low back pain, knee osteoarthritis, fibromyalgia, tennis elbow (lateral epicondylitis), sciatica, shoulder periarthritis, rheumatoid arthritis, and even gouty arthritis (Li et al., 2022; Zhao et al., 2022; Choi et al., 2019; Lam et al., 2020; WHO, 2003). Cardiovascular research has probed essential hypertension, primary hypotension, and pain from thromboangiitis obliterans (Shanghai Medical Clinic, 2025; WHO, 2003). Gynecological and obstetric domains feature prominently, including dysmenorrhea, labor induction, breech presentation correction, pregnancy-related nausea and vomiting, and fertility enhancement (e.g., improved clinical pregnancy rates in IVF protocols) (Zhao et al., 2022; Shanghai Medical Clinic, 2025; Smith et al., 2021; Carr, 2022; WHO, 2003).
Acupuncture trials also extend to psychiatric conditions like generalized anxiety disorder (especially in perimenopause), depression, and other mental disturbances (Zhao et al., 2022; Zhang et al., 2025; WHO, 2003); respiratory issues such as allergic rhinitis and hay fever (Li et al., 2022; Shanghai Medical Clinic, 2025; WHO, 2003); gastrointestinal disorders including acute and chronic gastritis, biliary colic, and postoperative nausea/vomiting (Zhao et al., 2022; Shanghai Medical Clinic, 2025; WHO, 2003); urogenital and nephrological problems like renal colic and radiation-induced leucopenia (often in renal contexts) (Shanghai Medical Clinic, 2025; WHO, 2003); infectious diseases such as acute bacillary dysentery, pertussis (whooping cough), and epidemic hemorrhagic fever (WHO, 2003); pediatric applications, albeit more limited, for post-extubation pain relief and whooping cough (ClinicalTrials.gov, 2013; WHO, 2003); and oncology support for cancer-related fatigue and chemotherapy/radiation side effects (Zhao et al., 2022; Shanghai Medical Clinic, 2025). Additional niches include ear-nose-throat conditions (e.g., rhinitis), eye disorders, connective tissue diseases, metabolic/nutritional imbalances, and skin pathologies (Zhao et al., 2022; WHO, 2003).
This extraordinarily wide spectrum, drawn from seminal analyses like the World Health Organization’s (WHO) 2003 review of controlled clinical trials (WHO, 2003) and Cochrane overviews on pain (Choi et al., 2019; Lee et al., 2011), clearly demonstrates that acupuncture is considered by its proponents to be a ‘cure all’. This begs the question whether such an assumption can be reasonable. The effect sizes are typically modest, and true acupuncture is often no different from sham interventions (e.g., superficial needling at non-acupoints), suggesting limited specific efficacy (Lee et al., 2011).
The scientific literature proposes a constellation of mechanisms to explain how acupuncture might work, integrating peripheral, spinal, supraspinal, and systemic processes. These are often conceptualized through the “Neural Acupuncture Unit” (NAU) model, which posits low-threshold mechanosensitive afferents (Aδ and C fibers) at acupoints converging with brain networks to elicit bidirectional signaling (Zhang et al., 2012).
- Peripheral and Local Mechanisms. Needle manipulation is claimed to induce immediate tissue responses: adenosine triphosphate (ATP) breakdown to adenosine activates A1 receptors, dampening nociceptor firing (Kelly & Suckley, 2016); axonal reflexes release neuropeptides like substance P and calcitonin gene-related peptide (CGRP), modulating local inflammation; and stromal cells exhibit cytoskeleton remodeling, with collagen fibers “wrapping” around needles to propagate mechanical signals (Kelly & Suckley, 2016; Zhang et al., 2012; Li et al., 2025). The characteristic deqi sensation (aching, soreness) correlates with these events, potentially amplifying sensory input (Staud & Price, 2014).
- Spinal Cord Level. Ascending afferents are said to activate the gate control system, presynaptic inhibition, and diffuse noxious inhibitory controls (DNIC), releasing endogenous opioids (β-endorphin, enkephalins, dynorphins), serotonin, norepinephrine, and acetylcholine to suppress nociceptive transmission in the dorsal horn (Kelly & Suckley, 2016; Zhang et al., 2012; Staud & Price, 2014). This underpins analgesia and autonomic regulation, such as reduced sympathetic outflow (Kelly & Suckley, 2016).
- Central Nervous System Modulation. Functional neuroimaging (fMRI, PET) reveals deactivated limbic hyperactivity (amygdala, anterior cingulate), normalized hypothalamic-pituitary-adrenal (HPA) axis activity, and enhanced prefrontal connectivity, particularly in pain, stress, and mood disorders (Kelly & Suckley, 2016; Zhang et al., 2012; Wang et al., 2025). Top-down expectancy modulates descending inhibitory pathways, integrating with reward and mirror neuron systems (Zhang et al., 2012).
- Systemic and Humoral Effects. Acupuncture is also thought to influence immune homeostasis by shifting cytokine profiles (e.g., ↑IL-10, ↓TNF-α, ↓IL-6), autonomic balance (vagal enhancement), and endocrine axes, providing a basis for visceral, metabolic, and inflammatory conditions (Kelly & Suckley, 2016; Li et al., 2025). Recent integrative studies emphasize network pharmacology, where multi-point stimulation perturbs interconnected pathways (Li et al., 2025).
These potential mechanisms have been empirically observed in animal models and/or human imaging studies. They might offer a partial rationale, primarily for analgesia and stress-related syndromes (Kelly & Suckley, 2016; Zhang et al., 2012). The question, however, is whethr they can provide a full explanation for acupuncture’s efficacy in all the above-named conditions.
No synthesis of these mechanisms plausibly accounts for acupuncture’s claimed benefits across unrelated conditions, exposing a core scientific paradox. Musculoskeletal pain might align with local adenosine/opioid effects and spinal gating (Kelly & Suckley, 2016), but how do these explain microbial clearance in bacillary dysentery, hypertensive vascular remodeling, or synaptic imbalances in major depression? (Meissner et al., 2019; Ernst, 2018). Gynecological infertility involves ovarian endocrinology, distant from needle-evoked sensory cues; infectious pertussis implicates Bordetella immunity, not HPA modulation (WHO, 2003; Meissner et al., 2019). This biological implausibility echoes homeopathy critiques: a single intervention cannot verifiably target such heterogeneous pathophysiologies without invoking non-specific forces (Fabrizio et al., 2010).
Trial data reinforce these doubts: meta-analyses consistently show that verum acupuncture is hardly different from sham acupuncture, and sham elicit up to 80% of verum’s effects (Kelly & Suckley, 2016; Meissner et al., 2019; Fabrizio et al., 2010; Kaptchuk et al., 2013). Such considerations implicate patient and therapist expectations, therapeutic ritual, and patient-practitioner alliance as the true mechanism behing the observed outcomes (Meissner et al., 2019; Kaptchuk et al., 2013). Neuroimaging effects often mirror expectancy manipulations in non-needling studies, suggesting top-down confounds (Fabrizio et al., 2010). Lab phenomena (e.g., adenosine release) occur but yield trivial clinical effects, dwarfed by psychosocial amplification (Fabrizio et al., 2010).
Acupuncture’s elaborate ritual maximizes contextual healing, outperforming inert pills but lacking disease-modifying specificity (Meissner et al., 2019; Ernst, 2018). Paradoxes abound—positive preclinical signals evaporate in blinded RCTs; cultural bias inflates Asian trial positives; poor sham penetration and blinding failures perpetuate illusions (Fabrizio et al., 2010; Ernst, 2018). For non-pain conditions, evidence thins further, with publication bias and flexible outcome reporting inflating apparent successes (Fabrizio et al., 2010).
Acupuncture carries risks including minor issues like bleeding, needle site pain, vegetative reactions (e.g., dizziness or nausea), and symptom aggravation, alongside rarer serious events such as pneumothorax, infections, or organ injury. Overall, at least one adverse event in 9.31% of patients undergoing a treatment series or 7.57% of treatments, with half of these being mild local reactions. Serious adverse events seem to be uncommon. Reliable prevalence figures do not exist because there is no adequate surveillance system in place (Ernst 2006).
Acupuncture’s trial proliferation signals cultural and patient-driven demand rather than mechanistic or evidential triumph. Its broad therapeutic claims by far overreach evidence (Staud & Price, 2014). Rigorous advancement would require objective biomarkers (e.g., cytokine assays, EEG), dose-response optimization, adaptive sham designs, and large pragmatic trials stratifying contextual from specific effects (Zhang et al., 2012; Fabrizio et al., 2010). Until compelling evidence exists, acupuncture remains a testament to human suggestibility’s power, but not a biomedical panacea.
References
- Carr, D. (2022). Acupuncture as Treatment for Female Infertility. Medical Acupuncture, 34(1), 12-21.
- Choi, D., et al. (2019). Cochrane reviews on acupuncture therapy for pain: a snapshot of the current evidence. Systematic Reviews, 8, 231.
- ClinicalTrials.gov. (2013). Pediatric Laser Acupuncture and Renal Biopsy (NCT01879826).
- Ernst, E. (2006). Acupuncture–a critical analysis. J Intern Med, 259(2):125-37.
- Ernst, E. (2018). Acupuncture Research: The Problem. Pain Medicine, 19(6), 1287-1288.
- Fabrizio, P., et al. (2010). Paradoxes in Acupuncture Research: Strategies for Moving Forward. Explore (NY), 6(4), 231-239.
- Kaptchuk, T. J., et al. (2013). Are All Placebo Effects Equal? Placebo Pills, Sham Acupuncture, or Placebo Needle in Irritable Bowel Syndrome. PLoS ONE, 8(7), e67485.
- Kelly, R., & Suckley, S. (2016). Mechanisms of acupuncture. European Journal of Integrative Medicine, 20, 1-11.
- Lam, M., et al. (2020). Acupuncture and Chronic Musculoskeletal Pain. Medical Acupuncture, 32(6), 357-366.
- Lee, M. S., et al. (2011). Acupuncture for pain: an overview of Cochrane reviews. Chinese Journal of Integrative Medicine, 17(3), 187-189.
- Li, T., et al. (2022). Evidence on acupuncture therapies is underused in clinical practice. Frontiers in Medicine.
- Li, Y., et al. (2025). Integrative research on the mechanisms of acupuncture. Neural Regeneration Research.
- Meissner, K., et al. (2019). Acupuncture for the Treatment of Pain – A Mega-Placebo? Frontiers in Neuroscience, 13, 1119.
- Shanghai Medical Clinic. (2025). WHO Approved Acupuncture List of Conditions.
- Smith, C. A., et al. (2021). An Overview of Systematic Reviews of Acupuncture for Respiratory Diseases. Frontiers in Public Health.
- Staud, R., & Price, D. D. (2014). Acupuncture therapy: mechanism of action, efficacy, and safety. International Review of Neurobiology, 111, 171-189.
- Wang, L., et al. (2025). Possible antidepressant mechanism of acupuncture. Frontiers in Neuroscience, 19, 1512073.
- WHO. (2003). Acupuncture: Review and Analysis of Reports on Controlled Clinical Trials.
- Zhang, R., et al. (2012). Neural Acupuncture Unit: A New Concept for Interpreting Effects and Mechanisms of Acupuncture. Evidence-Based Complementary and Alternative Medicine, 2012, 429412.
- Zhang, Y., et al. (2025). Patient-reported outcome tools of acupuncture clinical trials. Journal of Pain Research.
- Zhao, C., et al. (2022). Evidence mapping and overview of systematic reviews of the effects of acupuncture therapies. BMJ Open, 12(6), e056803.
6 Responses to Acupuncture: too good to be true?
Leave a Reply
This site uses Akismet to reduce spam. Learn how your comment data is processed.
On Acupuncture: Some aspects which are usually forgotten
Professor Ernst is essentially correct. While acupuncture is currently the most frequently used method in so-called alternative or complementary medicine, and while there are countless supposedly scientific studies on it, we still know virtually nothing about its fundamental questions.
1) It is certain that the “meridians” of acupuncture do not exist, and that outside of arteries, veins, capillaries, and lymphatic vessels, there is no “flow of blood and Qi.”
2) Even in Traditional Chinese Medicine (TCM) theory, there is no answer to the question of a) what force (since the heart is not at the center of these “meridians”) is supposed to cause the alleged “flow through the meridians,” and b) that, according to acupuncture theory, there should be not ONE, but TWO of these alleged “meridian circuits”—one on the right and one on the left.
3) There is no answer to the logical question that arises from the actual non-existence of the “meridians”: If the “meridians” do not exist, what are the so-called “acupuncture points”? Do at least the frequently needled points actually have a specific effect? Is there really a connection between the points of a “meridian”? Is there truly a demonstrable connection between the “acupuncture points” and the internal organs?
4) All the cited studies also overlook the fact that acupuncture was largely clinically insignificant in ancient China and that, for example, it was probably never used to treat the approximately 400 emperors in China’s history.
5) Strangely, one of the few aspects of needle acupuncture that is truly established is almost always forgotten: namely, that needling constitutes a micro-injury, and that the physiological reactions to this injury are among the established effects of acupuncture.
6) Equally forgotten is usually the SETTING of modern acupuncture, both in China and in the West: 30 minutes of quiet lying down, a haven of peace and relaxation amidst the hectic 3-minute routine of conventional medicine. This setting alone should be sufficient to assume with some certainty that it has a psychosomatic healing effect. Explaining the effectiveness of acupuncture in this way is still not really science – but at least it’s a bit of applied common sense.
@Dr. med. Hanjo Lehmann
Not to mention the fact that traditionally, acupuncture was, erm, slightly different from its modern-day incarnation, as also described by a Scottish doctor and missionary working in China around 1900:
“The only mode of treatment in vogue which might be called surgical is acu-puncture, practised for all kinds of ailments. The needles are of nine forms, and are frequently used red-hot, and occasionally left in the body for days. Having no practical knowledge of anatomy, the practitioners often pass needles into large blood-vessels and important organs, and immediate death has some- times resulted. A little child was carried to the dispensary presenting a pitiable spectacle. The doctor had told the parents that there was an excess of fire in its body, to let out which he must use cold needles, so he had pierced the abdomen deeply in several places. The poor little sufferer died shortly afterwards. For cholera the needling is in the arms. For some children’s diseases, especially convulsions, the needles are inserted under the nails. For eye diseases they are often driven into the back between the shoulders to a depth of several inches. Patients have come to us with large surfaces on their backs sloughing by reason of excessive treatment of this kind with instruments none too clean.”
Richard Rasker is being ironic. However, he is actually pointing out an important issue confirmed by all reputable sources: After acupuncture was banned for use at China’s imperial court in 1822, needle therapy became a rarity throughout the country—at least if it wasn’t already. This was justified because ancient acupuncture—without anatomical knowledge, clean needles, and without disinfection and sterilization—always posed a real danger to patients. Since in ancient China the physician was held responsible if a patient died, acupuncture was also dangerous for the practitioner (for more on this, see Lehmann: Acupuncture in Ancient China – How important was it really? https://www.sciencedirect.com/science/article/abs/pii/S2095496414600984?via%3Dihub ).
The acupuncture revival that began in China after 1954 (not, as one might assume, as early as 1949) was therefore by no means a continuation of a long-term development, but largely a new development. The internal debates surrounding this revival are evident in the first acupuncture textbooks published in China for foreigners. For example, the first of these books—the “Outline of Chinese Acupuncture” from 1975—makes no mention of either the Yin-Yang theory or the Five Elements theory.
The hype that began in the West after James Reston’s acupuncture article in the New York Times on July 26, 1971, coincided with the height of the murderous Cultural Revolution in China. Since no one could travel to China during this period and learn acupuncture firsthand, it was inevitable that the development of acupuncture in the West would be dominated by charlatans. The first and most influential of these charlatans had been George Soulié de Morant in France after 1929 (for more information, see Lehmann, Akupunktur im Westen: Am Anfang war ein Scharlatan https://www.aerzteblatt.de/archiv/akupunktur-im-westen-am-anfang-war-ein-scharlatan-ad4c01bf-880e-4db9-af2f-424757a937bc ). He was followed in France by Roger de la Fuye. Germans and Austrians such as Gerhard Bachmann and Johannes Bischko learned from him, as well as Felix Mann from Great Britain and the charismatic swindler Worsley. Worsley’s “5-Element Acupuncture” was just as much a product of fantasy as the “energetic acupuncture” of figures like Helms in the USA or Thambirajah in GB. But that is another story entirely, one that truly deserves a separate article.
yes, the recent history of acupuncture is littered with charlatans – not a surprise really, if we consider acupuncture to be little more than charlatanery!
And even so, still NICE endorses it!
Why do you think NICE endorses it?