INTRODUCTION — The word "acupuncture" is derived from the Latin words "acus" (needle) and "punctura" (penetration). Acupuncture originated in China approximately 2000 years ago and is one of the oldest medical procedures in the world.
Over its long history and dissemination, acupuncture has diversified and encompasses a large array of styles and techniques. Common styles include Traditional Chinese, Japanese, Korean, Vietnamese, and French acupuncture, as well as specialized forms such as hand, auricular, and scalp acupuncture.
Acupuncture also refers to a family of procedures used to stimulate anatomical points. Aside from needles, acupuncturists can incorporate manual pressure, electrical stimulation, magnets, low-power lasers, heat, and ultrasound.
Despite this diversity, the techniques most frequently used and studied are manual manipulation and/or electrical stimulation of thin, solid, metallic needles inserted into skin. Except where specifically stated, "acupuncture" in this topic refers to these two most common procedures.
A general discussion of acupuncture is presented here. Additional discussions of acupuncture for rheumatic conditions, cancer, and other conditions are presented separately. (See "Overview of the management of overuse (persistent) tendinopathy", section on 'Acupuncture' and "Management of knee osteoarthritis", section on 'Acupuncture' and "Complementary and alternative therapies for cancer" and "Overview of fatigue, weakness, and asthenia in palliative care", section on 'Acupuncture' and "Nonpharmacologic approaches to management of labor pain", section on 'Acupuncture' and "Postoperative nausea and vomiting", section on 'Acupuncture'.)
HISTORY AND USAGE PATTERNS — The precise origin of acupuncture is a source of debate. There is no single archaeological finding that points to a momentary emergence of acupuncture. Rather, evidence exists for a variety of potential antecedent practices like bloodletting, tattoos for religious purposes, and the use of bones to extract abscess [1].
China — The first written document to record the use of acupuncture is the Nei Jing (Inner Classic of the Yellow Emperor) dated approximately 100 BC. It is a collection of 81 treatises divided into two parts [2]. By the time of its compilation, acupuncture was already a signature therapy of Chinese medicine.
The importance of acupuncture as medical therapy emerged around the same time that Confucianism and Taoism gained prominence in China. These philosophies are imprinted in the fundamental principles of acupuncture theory, and their influence is patently evident throughout the ancient texts [1,3]. Acupuncture underwent significant development and expansion within the ensuing 1500 years and arguably climaxed in the Ming era (1368 to 1644) when The Great Compendium of Acupuncture and Moxibustion was published in 1601 [4]. Afterwards, it experienced waxing and waning popularity due to political and social pressures arising from Western influences, but it gained a modern resurgence after Mao Zedong encouraged its use among "barefoot doctors" [1].
Historically, there are around 10,000 treatises on acupuncture from the centuries preceding the modern era [5]. Past acupuncture scholars freely edited prior texts and added personal interpretations, commentaries, and clinical experiences [3]. As a result, present copies of ancient texts often represent the work of multiple acupuncture scholars and demonstrate a medley of teachings, each susceptible to variable interpretations. This has contributed to the marked heterogeneity seen in acupuncture practice.
Asia and Europe — Acupuncture was disseminated to Korea and Japan in the sixth century, to Southeast Asia around the ninth century through commercial trade routes from China, and to Europe as early as the 16th century when Asian texts and translations were brought back by traders and missionaries [6]. Acupuncture became relatively established in some parts of Europe, such as France, around the 18th century and persisted due to perpetual colonial influences (eg, Indochine) [4].
United States — In the United States, traces of acupuncture appeared as early as 18th century and appeared in the early editions of William Osler's Principles and Practice of Medicine [7]. However, acupuncture did not enter the mainstream until 1971, when a New York Times journalist, James Reston, visited China and reported his experiences with acupuncture for postoperative pain relief [8].
Several surveys suggest that acupuncture is the complementary and alternative medicine (CAM) therapy most likely to be recommended by conventional medical professionals in resource-rich settings [9]. Based on a 2012 survey in the United States, an estimated 3.8 million adults, or 1.5 percent of the adult population, had used acupuncture in the previous year [10]. The five most commonly treated conditions were back pain, neck pain, joint pain, headache, and "head/chest cold." Other commonly treated conditions include fatigue, anxiety, insomnia, and depression. Acupuncture use is probably more prevalent among patients who have immigrated from certain Asian countries (eg, Chinese and Vietnamese Americans) [11].
BASIC THEORY — Acupuncture's early development coincided with the rise and prominence of two major Chinese philosophies, Confucianism and Taoism. As a result, acupuncture theory is largely grounded in these philosophies [1].
One notable, early influence of these philosophies was the recognition that one's observation and experience were sufficient to explain the human condition [12]. This was a significant departure from primordial Chinese healing arts which usually ascribed illness to some superstitious force or moral punishment [12].
The two philosophies, particularly Taoism, emphasized the importance of understanding the laws of nature and for humans to integrate and abide by these laws rather than resist them. The human body was regarded as a microcosmic reflection of the macrocosm of the universe. For this reason, concepts used to explain nature, such as yin/yang and Five Elements (described below), became central to acupuncture theory [3]. The goal of the clinician was to maintain the body's harmonious balance both internally and in relation to the external environment.
Eastern medicine values the clinician's initial assessment and encourages the practitioner to hone his/her own intuition to extract additional subtleties. Eastern thought perceives the world as dynamic and interconnected [13]. To the acupuncturist, it makes little sense to isolate a symptom such as back pain. Symptoms necessarily arise from a particular context. Acupuncture treatments are therefore usually individualized, and two patients with the same symptoms often do not get the same treatment. The same patient also may not receive the same treatment on subsequent visits.
Three important concepts in acupuncture are qi, yin/yang, and Five Elements.
●Qi (pronounced "chee") is frequently translated as "vital energy" [14]. It is felt to permeate all things, may assume different forms, and travels through meridians located on the body. It can be described as stagnant, depleted, collapsed, or rebellious. Whether qi is a quantitative force or a metaphor for the way people experience and depict connections and interconnections is not clear. It likely provides a rationale for explaining change and linking phenomena [12].
●Yin and yang are felt to be complementary opposites and are used to describe all things in nature. Yin is used to represent more material, dense states of matter, while yang represents more immaterial, rarefied states of matter [15]. The interplay between the two opposites is dynamic and cyclical. To the acupuncturist, health is a constant state of dynamic balance, and one must employ a series of qualitative assessments to establish a patient's present disposition (table 1). The evaluation is more complex than merely designating a patient as "more yin" or "more yang." An intricate set of qualitative measures, examination tools, and symptom evaluations are used [15].
●Five Elements along with yin/yang theory form the basis of Chinese medical theory. The Five Elements are wood, water, fire, earth, and metal. These elements are not basic constituents of nature but represent different basic processes, qualities, or phases of a cycle [15]. Each element can generate or counteract another element. Most vital organs, acupuncture meridians, emotions, and other health-related variable are assigned an element (table 2), thus providing a global description of the balancing dynamics seen in each person.
The Eastern medical practitioner relies on these principles for diagnosis and treatment selection. Once the nature of imbalance is determined, the practitioner aims to shift the constitution towards balance with the use of various interventions. Acupuncture is one important option.
ACUPUNCTURE ENCOUNTER — The typical acupuncture treatment begins with identification of the patient's constitutional pattern. To accomplish this, acupuncturists use the "Four Pillars of Evaluation": inspection, auscultation, inquiring, and palpation [14].
According to traditional Chinese medical theory, practically everything, such as skin, complexion, bones, channels, smells, sounds, mental state, preferences, emotions, demeanor, and body build, reflects the state of the internal organs and can be used in diagnosis [15]. The diagnostic evaluation may therefore be extensive, often incorporating seemingly unrelated symptoms (as an example, discerning one's incapacity to make decisions or dislike of speaking for complaints of abdominal pain) [15]. In traditional Chinese acupuncture, the tongue and radial pulse are often evaluated. In the Japanese style, strategic "reflex points" may be identified [14].
Once the diagnosis is established, fine metal needles are inserted into precisely defined points to correct disruption in harmony. Classic theory recognizes about 365 points, said to be located on 14 main channels (or meridians) connecting the body. The 14 main channels are associated with specific organs, although theoretically not in the anatomic sense to which biomedical clinicians are accustomed.
Half are yin and other half are yang channels. Additional acupuncture points (both on- and off-channel) have been added with time and the total number of points has increased to at least 2000 [16]. In practice, however, the repertoire of a typical acupuncturist may be only 150 points. In a typical session, 5 to 20 needles are used [12]. Each session usually lasts up to one hour, although sessions can be as short as 15 minutes. Once needles are inserted, they are often left for 10 to 15 minutes while the patient lies relaxed. Needles are removed at the end of the session. Treatments occur one to two times a week and the total number of sessions is variable, depending on the condition, disease severity, and chronicity.
In traditional Chinese acupuncture, needle effectiveness is frequently measured by the elicitation of de qi [17]. De qi is obtained by manipulation of the acupuncture needle and is perceived as an "aching" or "throbbing" sensation by the patient and a "grasp" by the acupuncturist [4,18-20]. For the patient, a treatment session may be considered painful, although there is clear cultural and interpersonal variability. Other styles, such as Japanese acupuncture, tend to be more subtle and utilize more superficial needling with little or no manipulation [14,21].
Heat stimulation, a technique known as moxibustion, which burns the herb Artemisia vulgaris near the acupuncture point, is sometimes used. Hand pressure is also sometimes applied. Numerous other techniques can also be used including the addition of low-level electric current (electroacupuncture), low-power laser, magnets, and ultrasound. The type of intervention and level of stimulation varies with acupuncture style and between acupuncturists. Some styles, such as auricular, hand, and scalp acupuncture, limit their stimulation to a particular body part.
Acupuncture treatments are usually individualized, catered to the individual and not to the condition [22]. Two patients with identical problems will frequently get different treatments. Point combinations can also vary between sessions.
Acupuncture is often used in conjunction with other modalities. Chinese herbal interventions have historically been the mainstay of East Asian therapy. Acupuncturists may also use massage, cupping (using vacuum suction over particular areas of the skin), and scarification [12].
Lifestyle counseling, around issues such as diet, exercise, and mental health, is a component of acupuncture care. In addition, the acupuncture experience itself is purported to be therapeutic. Patients are frequently required to lay relaxed while the needles are left embedded in the skin. Consequently, the experience is frequently described as relaxing and soothing. Furthermore, acupuncturists historically have considered the patient-clinician relationship and therapeutic encounter itself to be inherently "potent" and sufficient to promote healing [3].
PROPOSED MECHANISMS OF ACTION — Multiple physiologic models have been proposed to explain the effects of acupuncture; however, for many proposed models, the data have been either too inconsistent or inadequate to draw significant conclusions. Various models have implicated cytokines, hormones (eg, cortisol and oxytocin), biomechanical effects, electromagnetic effects, the immune system, and the autonomic and somatic nervous systems. Questions remain as to what is the optimal form of acupoint stimulation and whether acupuncture points have any physiological specificity as espoused by traditional Chinese theory [23].
Endorphins — The most thoroughly studied application of acupuncture is for pain relief. Studies performed in the 1970s and 1980s have contributed tremendously to our present understanding of acupuncture's analgesic effects [24-42]. According to this theory, acupuncture stimulation is associated with neurotransmitter effects such as endorphin release at both the spinal and supraspinal levels [43,44].
Different electrical frequencies appear to stimulate different activation pathways [45]: low-frequency (2 to 4 Hz) electroacupuncture mobilize enkephalin, beta endorphin, and endomorphin at all three centers (spinal cord, midbrain, and pituitary/hypothalamus), while high-frequency (100 Hz) electroacupuncture induces only dynorphin (kappa) at the spinal cord level. In addition, the effects of low-frequency stimulation tend to last longer and become cumulative with each subsequent stimulation, while the effects of high-frequency stimulation are shorter in duration and noncumulative.
In support of this theory, there is evidence that opioid antagonists block the analgesic effects of acupuncture [46]. In contrast to this theory, however, the endorphin effects appear to be short-term, only lasting 10 to 20 minutes and possibly up to several days [47], while many acupuncture clinical trials have documented longer effects [47-49]. Additionally, endorphin release can be induced by strongly stimulating any free nerve ending or muscle afferents. The specificity of acupuncture point location and the rationale for needling certain points in various conditions remain unexplained.
For these and other reasons, researchers have acknowledged the limitations of the endorphin-related mechanism [50].
Functional MRI — Functional magnetic resonance imaging (MRI) studies have demonstrated physiologic effects with acupuncture. Compared with “sham” acupuncture, verum acupuncture (ie, manual needling or electrical stimulation of acupuncture points) is generally associated with more widespread and sustained changes in the blood oxygenation level-dependent (BOLD) signals in various regions of the central nervous system (cerebral, limbic, and brainstem) [23].
Connective tissue — Another theory is that acupuncture points are associated with anatomic locations of loose connective tissue. As an example, a study that evaluated points and meridians in the arm concluded that such an association with intermuscular fascia was present [51]. It is possible that such an association might relate to the concept of "grasp" noted by practitioners, attributable to collagen twisting around the acupuncture needle, as demonstrated by light and electron microscopy [52,53]. The associated mechanical forces are hypothesized to change local purinergic (adenosine) signaling and inflammatory pathways contributing to analgesic effects [54].
CLINICAL APPLICATION
Proposed indications — There have been hundreds of controlled trials of acupuncture for various conditions. The best trials are discussed below. (See 'High-quality trials' below.)
Conditions for which acupuncture has been studied and appears to have possible efficacy include:
●Chronic pain [55-59]
●Postoperative nausea and vomiting [60]
●Chemotherapy-induced nausea [61,62]
●Acute pain including dental pain [63-65]
●Headache [48,66,67]
●Chronic obstructive pulmonary disease (COPD) [68]
●Seasonal allergic rhinitis [69]
●Menopausal hot flashes [70]
●Acute hordeolum [71]
●Cancer-related pain [72]
Acupuncture has been studied for many other conditions, including stroke [73,74], depression [75], fibromyalgia [76,77], functional dyspepsia [78], schizophrenia [79], and tobacco use [80,81], but the evidence is insufficient to recommend the use of acupuncture for these conditions.
Adverse events — Acupuncture is generally safe but can lead to the complications seen with any type of needle use. These include transmission of infectious diseases, retained needle fragments, nerve damage, and very rarely, pneumothorax, pneumoperitoneum, organ puncture, cardiac tamponade, and osteomyelitis [82,83]. Local complications may include bleeding, contact dermatitis, infection, pain, and paresthesias [82].
Despite the variety of listed complications and the occasional case reports in major journals [84-89], major adverse events are exceedingly rare and are usually associated with poorly trained, unlicensed acupuncturists [90].
●A prospective study in Japan of 65,482 acupuncture treatments reported no major adverse events [91].
●A prospective investigation in Germany of 97,733 patients constituting 760,000 treatment sessions reported that the two most frequently reported adverse events were needling pain (3.3 percent) and hematoma (3.2 percent) [92]. Potentially serious adverse events included two cases of pneumothorax. An asthma attack, a vasovagal reaction, an acute hypertensive crisis, and an exacerbation of depression were considered to be possibly related to treatment.
●Another two surveys performed in the United Kingdom totaling 66,000 treatments reported no serious adverse events [93,94].
In summary, acupuncture is considered very safe if rates of adverse effects are compared with those seen in many pharmacologic treatments. Practitioners should use sterile needles to prevent transmission of disease. In the United States, acupuncture practitioners are required to use disposable sterile needles.
Precautions — There are few absolute contraindications to acupuncture treatment, as it is generally safe and well tolerated.
●Acupuncture should be avoided in patients with severe neutropenia as seen after myelosuppressive chemotherapy [95].
●The insertion of acupuncture needles at sites of active infection or malignancy is contraindicated. In the case of malignancy, there is a theoretical risk of causing metastatic dispersal of tumor cells [96].
●Electroacupuncture should be avoided in patients with an automatic implantable cardioverter-defibrillator (AICD) or pacemaker because of risk of electrical interference with the device [97]. Traditional acupuncture is safe in these patients.
There are other conditions that are not contraindications but warrant special consideration:
●Pregnancy is not an absolute contraindication, since acupuncture has been used and studied for gestational conditions such as breech presentation and pregnancy-associated nausea [98-103]. According to acupuncture theory, however, some points can induce labor, and the acupuncturist should be informed of the pregnancy [104,105].
●Bleeding disorders and use of anticoagulants are also not contraindications to acupuncture treatment, with no increased incidence of bleeding complications among patients taking antiplatelet agents, direct oral anticoagulants (DOACs) or warfarin [106-109]. The acupuncturist should, however, be notified of any bleeding risks.
Referral — There is wide variability in skill level among acupuncture practitioners, including those licensed to perform the procedure. In the United States, identifying a good acupuncturist is typically by referral or word of mouth.
In the United States, acupuncturists should be certified by the National Certification Commission for Acupuncture and Oriental Medicine (NCCAOM) or the American Board of Medical Acupuncture (ABMA); acupuncturists should be licensed if they are in one of the states that have such licensure.
Clinicians should try to identify acupuncturists who will work with "traditional" medical providers and treatments and who will not encourage patients to discontinue standard medical therapies. In the United States, as long as the referring clinician appropriately diagnoses and manages a condition, referral of patients to an independent licensed practitioner for whom it is clear they have no supervisory role will not typically create a significant risk of legal liability [110].
In the United States, approximately 70 percent of acupuncturists practice alone or in groups of acupuncturists; 30 percent practice in multidisciplinary environments [12].
United States insurance coverage — Referring clinicians should familiarize themselves with insurance coverage, as some insurance providers will cover acupuncture for certain conditions.
Medicare does not cover acupuncture; however, in some states, Medicaid has begun to cover acupuncture for the treatment of chronic pain [111]. In addition, many other insurance carriers have some form of acupuncture coverage [112]. The amount of coverage varies widely, ranging from a small discount to total coverage. Some plans require the services be performed by clinicians or chiropractors; some limit coverage to certain conditions.
According to the 2012 National Health Interview Survey (NHIS), only one-quarter of adults seeing an acupuncture practitioner in the previous year had some form of health insurance coverage for it: one-third of the patients with coverage had complete coverage (8.5 percent of all acupuncture users), and the other two-thirds (16.5 percent of all users) had only partial coverage [113].
If cost is a major concern, patients should check with their insurance carrier before using acupuncture. Given the number of sessions frequently required for treatment, the cost can accumulate and become substantial. This should be considered when referring a patient to an acupuncturist.
CLINICAL EVIDENCE
Difficulties in research — Some of the problems encountered with acupuncture randomized trials are shared by trials in many domains: inadequate sample size, lack of follow up, imprecise outcomes, improper statistical analysis, and others. Some problems, however, are particular to acupuncture research. Issues include:
●Identifying an acupuncture treatment for a biomedically defined disease can be difficult. One disease in biomedicine can be many "patterns" within the Eastern medicine classification schema [12,114]. As an example, diabetes can have Eastern medical diagnoses of "stomach fire," "kidney fire," or "lung fire" [115].
●Individualized treatments seen in acupuncture run counter to the standardized treatments used in randomized trials. Researchers have tried to deal with this by performing pragmatic trials (where acupuncturists are given full freedom) or trials using semi-standardized treatment (where acupuncturists are assigned mandatory points but given additional individualized options). Whether this latter approach approximates real acupuncture treatments is uncertain, as few studies have reported on the acupuncturists' perceptions of whether their treatments were constrained.
●Acupuncture entails many different styles and techniques. In the United States alone, at least eight different styles of acupuncture are taught in the various accredited schools [116]. Differences exist on what points are to be needled, how the needle should be manipulated, how long the needle should be kept in, and what is the appropriate response elicited from the patient [21]. Thus, it is difficult to know whether the results of a trial of single type of acupuncture can be generalized to other types.
●Due to the heterogeneity of acupuncture, an optimal control for one style may not be ideal for another.
●It is difficult to perform a double-blind acupuncture study. Acupuncturists are typically able to distinguish real treatment from sham treatment.
●Delivering acupuncture is not as simple as administering pills and, much like psychotherapy and surgery, experience may play a critical role in determining outcome.
High-quality trials — Despite the difficulties discussed above, a number of trials have compared active acupuncture with a sham control procedure that allows evaluation of the efficacy of acupuncture compared with placebo.
Low back pain — Well-designed clinical trials have found that both acupuncture and sham acupuncture are more effective than control interventions for low back pain.
In a systematic review of six randomized trials for chronic non-specific low back pain, acupuncture was found to have a small beneficial effect in reducing pain and improving functional status compared with sham, placebo, or other passive modalities at short-term (one month) and intermediate-term follow-up (three and/or six months) [117]. Similarly, in five randomized trials, the combination of acupuncture plus an intervention was found to have a small beneficial effect in reducing pain and improving functional status compared with the intervention alone (physiotherapy, exercise, or standard medical care) at short- and intermediate-term follow-up.
A subsequent randomized trial in 638 adults with chronic low back pain compared acupuncture treatment individualized to the patient, acupuncture treatment standardized for low back pain, sham acupuncture, and usual care [118]. Patients received 10 treatments over seven weeks. Patients were lying prone wearing a mask, and the sham therapy consisted of tapping and twisting a toothpick contained in an acupuncture needle guide tube against the skin for a few seconds at the points used in the standardized acupuncture intervention. At 10 and 20 minutes into the procedure, a toothpick was touched to the skin and twisted at the same points. Patients apparently rated individualized, standardized, and sham acupuncture very similarly in terms of credibility. At eight weeks, back dysfunction scores improved by similar amounts in the individualized, standardized, and sham acupuncture groups and more than in the usual care group (4.4, 4.5, and 4.4 points versus 2.1 points, respectively).
Knee osteoarthritis — A multicenter randomized trial compared 10 sessions of acupuncture, sham acupuncture, or clinician visits in 1007 patients with chronic knee osteoarthritis who were also being treated with physical therapy and antiinflammatory medications as needed [57]. The primary outcome measure was the rate of success at 26 weeks, defined as a 36 percent improvement in a standardized osteoarthritis index.
Acupuncture treatments were semi-standardized: practitioners were instructed in certain points that were to be needled and could then choose individually to needle additional points. Sham acupuncture involved the use of points not deemed to be useful in the treatment of knee osteoarthritis administered at minimal needling depths.
The researcher assessing endpoints was blinded to treatment assignment. Patient blinding between acupuncture and sham acupuncture was successful, with about half of patients who thought they knew which treatment they were receiving guessing incorrectly.
Rates of success were similar for acupuncture and sham acupuncture and greater than with conservative therapy (53 and 51 versus 29 percent).
Two other high-quality randomized trials that compared acupuncture with sham acupuncture found some added benefit with acupuncture; however, some blinding breakdown appears to have occurred in these trials [119,120].
A meta-analysis of randomized trials of acupuncture for knee osteoarthritis concluded that acupuncture may have had some additional measurable benefits compared with sham acupuncture but that the differences were too small to be clinically relevant [121].
A randomized trial published after the above meta-analysis compared six sessions of acupuncture, sham acupuncture (performed with a needle that retracted such that it did not penetrate the skin), and no additional therapy in 352 adults, all of whom were treated with advice and exercise [122]. Patients found both acupuncture and sham acupuncture to be credible.
The primary outcome was change in pain score at six months, and there were no significant differences between the three groups. At six weeks, patients treated with sham acupuncture had a small but statistically significant improvement in pain compared with those receiving advice and exercise alone; true acupuncture showed no significant benefit compared with advice and exercise alone.
Although like many other trials, this study found similar effects with acupuncture and sham acupuncture, the results were unusual in that the overall benefits of acupuncture and sham acupuncture were very small. This trial had fewer sessions of acupuncture or sham acupuncture than were used in most other studies.
Relatively small improvements in pain with acupuncture, laser acupuncture, and sham laser acupuncture were also seen in a randomized trial in 282 patients with chronic knee pain [123]. A concern with trials of acupuncture is that the large improvements seen may, in part, reflect the recruitment of patients who expect acupuncture to work [124]. This trial used an unusual design to try to minimize this effect by randomizing patients before they were aware of the intervention arms of the trial. This may explain the smaller benefits seen in both the active and sham arms of the trial; however, the design may also have diluted measured benefits by including outcomes of patients who ultimately refused to be assigned to an intervention arm of the trial.
Migraine — A meta-analysis of 22 randomized trials (including 4985 participants) evaluated the efficacy of acupuncture in preventing episodic migraines compared with no acupuncture, sham acupuncture, or prophylactic drug treatments [125]. The primary outcome measure was migraine frequency, defined as the number of migraine days (or attacks) per month, at completion of the treatment and at six-month follow-up. The secondary outcome was treatment response, a dichotomized outcome of either greater than 50 percent reduction in headache frequency or not. The analysis found the following:
●For comparisons with no acupuncture (acute abortive treatment only or “routine” care), acupuncture demonstrated a “moderate” reduction of headache frequency at study completion (-0.56, 95% CI -0.65 to -0.48) and at follow-up (-0.36, 95% CI -0.59 to -0.12). Approximately 40 percent of participants in the acupuncture groups noted ≥50 percent reduction in headache frequency compared with 17 percent in the no-acupuncture groups (pooled risk ratio 2.40, 95% CI 2.08 to 2.76) at study completion.
●Compared with the sham acupuncture, acupuncture was associated with a small reduction in headache frequency at both study completion (-0.18, 95% CI -0.28 to -0.08) and at follow-up (-0.19. 95% CI -0.30 to -0.09). This diminished difference in response, compared with no acupuncture, is consistent with findings from other studies that sham acupuncture is associated with clinical improvements [126].
●Acupuncture reduced migraine frequency more than prophylactic drug treatment at study completion (-0.25, 95% CI -0.39 to -0.10) but not at follow-up (-0.13. 95% CI -0.28 to 0.01). A similar pattern was noted for the secondary outcome of dichotomized treatment response. Trial participants receiving acupuncture were less likely to report adverse effects (odds ratio [OR] 0.25, 95% CI 0.10 to 0.62) than participants receiving the prophylactic drug treatment.
Based on this meta-analysis, acupuncture is associated with moderate reduction in migraine episodes at study completion (and possibly sustained till a six-month post-randomization follow-up) when compared with no-acupuncture groups. This difference is smaller yet statistically significant when compared with both sham acupuncture and prophylactic drug treatment at completion of treatment. Acupuncture is associated with fewer adverse outcomes relative to prophylactic drug treatment.
Failure to effectively blind the subjects for no-acupuncture and drug treatment comparisons and failure to blind the acupuncturists for all intervention comparisons were noted as limitations for these included studies. Furthermore, the substantial heterogeneity in acupuncture treatments (five trials with standardized approach, seven with semi-standardized treatments, and 10 with individualized-treatments) makes the generalization of these results difficult for all acupuncture approaches.
Hot flashes — A randomized trial compared acupuncture with sham acupuncture (device that did not puncture the skin applied in locations not normally used for acupuncture) in 327 women with menopausal hot flashes [127]. Ten treatments were administered over eight weeks, and the primary outcome was the hot flash score (a validated measure) at the end of eight weeks. Both the acupuncture and sham acupuncture groups showed an approximately 40 percent reduction in symptoms after treatment. The mean hot flash score (where higher values reflect more severe or more frequent symptoms, and where a difference of four points between the groups was felt to be clinically meaningful a priori) was similar between the two groups (15.36 versus 15.04; mean difference 0.33, 95% CI -1.87 to +2.52). Acupuncture was not superior to sham acupuncture for treating menopausal hot flashes.
Summary — These studies suggest that there is little difference in the effects on pain between acupuncture and sham acupuncture. A meta-analysis of randomized controlled trials of acupuncture for pain that included both sham acupuncture and no treatment arms (three-armed trials) found that the superiority of acupuncture over sham acupuncture, if real, appeared to be too small to be clinically important [128].
However, in a subsequent 2018 individual patient data meta-analysis including 39 trials and almost 21,000 patients, acupuncture treatment of various pain conditions (including osteoarthritis, chronic headache, shoulder, or musculoskeletal pain) was superior to both sham acupuncture and no acupuncture control for improvement in pain (-0.2 standard deviations [SD] and -0.5 SD, respectively) [129]; treatment effects persisted at one year.
One likely explanation for the results seen in high-quality randomized trials is that both acupuncture and sham acupuncture moderate pain through a strong placebo effect. An alternate possibility is that sham needling at non-acupuncture points to minimal depths has physiologic effects on pain. Against this latter possibility is the result of a randomized trial that examined the effects of acupuncture and sham acupuncture on postoperative nausea and vomiting [60]. This trial used a sham device that did not penetrate the skin and still found similar effects with acupuncture and sham acupuncture.
As discussed above, it is difficult to know whether acupuncture constrained by the requirements of a clinical trial has the same efficacy as when it is performed according to the practitioner's preferences. However, the marked superiority of acupuncture and sham acupuncture over untreated controls demonstrates the strong effects of treatment seen even under study conditions.
Trials for conditions other than pain — Most of the trials that have used the best sham controls have involved treatment of pain conditions.
A trial in patients with seasonal allergic rhinitis, all of whom had access to “rescue medication” treatment with cetirizine (and oral glucocorticoids if response was inadequate), compared acupuncture with sham acupuncture or no additional therapy beyond rescue medication [69]. Symptom reduction was greater with acupuncture than with sham or rescue medication alone, but the improvements were small and may not have been clinically important. Although this trial included a sham arm, it appears that acupuncture and sham acupuncture were performed somewhat differently: patients treated with acupuncture had more needles placed on average (16 versus 10), and this may be a marker for other interactions between therapist and patient also being different during acupuncture and sham. Also, a larger fraction of patients in the acupuncture arm had high expectations for acupuncture efficacy at baseline (85 versus 73 percent). As such, this trial does not provide convincing evidence for a benefit of acupuncture beyond that of sham acupuncture.
CREDENTIALING — In the United States, the American Board of Medical Acupuncture (ABMA) certifies clinician acupuncturists while the National Certification Commission for Acupuncture and Oriental Medicine (NCCAOM) certifies non-clinician acupuncturists.
Certifications require passing a standardized exam and demonstration of adequate training. The standard for an acupuncturist is usually between 2000 and 3000 hours of training in a three- or four-year program that is independently accredited [12].
Although some states allow clinicians to practice acupuncture without additional education, most states require between 200 and 300 hours of special training.
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topic (see "Patient education: Complementary and alternative medicine (The Basics)")
SUMMARY AND RECOMMENDATIONS
●The word "acupuncture" is derived from the Latin words "acus" (needle) and "punctura" (penetration) and can refer to a family of procedures used to stimulate anatomical points. (See 'Introduction' above.)
●The traditional theory of acupuncture involves qi, yin and yang, and the Five Elements. (See 'Basic theory' above.)
●There are a number of physiologic models that have been proposed to explain the effects of acupuncture. (See 'Proposed mechanisms of action' above.)
●Acupuncture has been studied for many conditions including chronic and acute pain and postoperative and chemotherapy-associated nausea. (See 'Proposed indications' above.)
●Although there are difficulties in studying acupuncture, randomized trials suggest that acupuncture and sham acupuncture may have similar efficacy. Given this, much or all of the effect of acupuncture may be related to the placebo effect. (See 'High-quality trials' above.)
●Acupuncture is generally very safe as long as appropriate sterile techniques are followed. (See 'Adverse events' above.)
●In patients with chronic pain, both acupuncture and sham acupuncture appear to have much greater efficacy than when patients are left untreated. We suggest that patients with chronic pain who are interested or open to acupuncture be referred for a trial of acupuncture when the availability of safe alternatives is limited (Grade 2B). Patients with other conditions may also benefit from a trial of acupuncture. (See 'Clinical evidence' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Andrew Ahn, MD, MPH, who contributed to earlier versions of this topic review.
