ABSTRACT
Medical acupuncture combines the understanding of neuroanatomy and pain physiology with the basic tenets of Eastern thought and the concept of Qi. More than 100 scientific papers suggest that pain-relieving aspects of acupuncture are in part mediated by endogenous monoamines and neuropeptides that are activated by needling. In the United States, acupuncture has its greatest success and acceptance in the treatment of musculoskeletal pain. When acupuncturists are appropriately trained, significant adverse events are rare. Scientific data confirm that medical acupuncture is effective for some medical conditions, while findings are inconclusive for others.
KEY WORDS
Acupuncture, Medical Acupuncture, Pain, RCTs, Efficacy, Safety
INTRODUCTION
Medical acupuncture is acupuncture that has been successfully incorporated into medical or allied health practices in Western countries. It is the complementary discipline most commonly integrated into Western medical practice. Derived from Asian and European sources of acupuncture, it is practiced in both pure and hybrid forms.1 The biomedical viewpoint that relies on quantitative measurement and objective technology is very different from the Chinese Medicine assertion that "contemplation and reflection on sensory perception and ordinary appearances are sufficient to understand the human condition, including health and illness."2
The basic idea of medical acupuncture is to combine the understand- ing of neuroanatomy and pain physiology with the basic tenets of Eastern thought and the concept of a subtle circulation network of a vivifying source known as Qi. Medical acupuncture is the therapeutic insertion of solid needles in various combinations and patterns based on traditional concepts such as Qi and Yin/Yang on neuroanatomical and segmental distribution, or a combination of the two.
Variation in acupuncture technique and concepts is not confined to the West. China, Japan, and Korea each developed a distinct version of acupuncture with multiple approaches. Japanese acupuncture is often more superficial than that practiced as part of Traditional Chinese Medicine (TCM). European interpretation of the original Chinese writings gave way to the energetic concept, with further variation by the British and French. Somatotopic representations of the body are utilized in 3 specialized systems of acupuncture developed in France (auriculotherapy, or ear acupuncture), Korea (Korean hand acupuncture), and Japan (scalp acupuncture). Medical acupuncture allows for all of these systems to be integrated into the knowledge gained from traditional biomedical training. The approach also emphasizes choosing acupuncture points based on Western understanding of trigger points, scientific discoveries on mechanism of action, and more formulaic approaches to point location.3,4
History
The first codification of acupuncture as part of East Asian medicine was sometime in the 1st or 2nd century BC. The Huang Di Nei Jing (Inner Classic of the Yellow Emperor) discussed acupuncture in terms of keeping the body in harmonious balance internally and with regard to the external environment. The Nan Jing (Classic of Difficult Issues) advanced the theories of points, channels, etiology, and treatment of illness in the 2nd century AD. The Zhen Jiu Da Cheng (Great Compendium of Acupuncture and Moxibustion), published in 1601, synthesized many earlier classical texts. It was the primary source of acupuncture information in Europe in the 17th to 20th centuries after its translation into Latin by various missionaries and scholar-diplomats.1 The continual codevelopment of acupuncture and biomedical sciences in Europe during the later 20th century influenced the discipline of medical acupuncture today.
Recent US interest in acupuncture dates back to James Restons 1971 New York Times article discussing his acupuncture-assisted postsurgical pain management during an international visit by US President Nixon. As doors opened to China, so did more information and positive reports from Westerners. Work by Pomeranz5 and others in the late 1970s elucidated some of the physiologic mechanisms of acupuncture and fueled acceptance by Western practitioners and the public. Two landmark events in the 1990s further legitimized acupuncture as a viable medical system. In 1996, the US Food and Drug Administration removed the experimental tag and approved the acupuncture needle for routine use,6 and in 1997, the National Institutes of Health Consensus Conference on Acupuncture determined that there was scientific evidence of acupunctures efficacy in some clinical situations.7
Today, acupuncture is the most widely accepted form of complementary and alternative medicine (CAM) recommended by physicians.8 It is also popular with the US public: Eisenberg et al showed that US spending on CAM treatments increased from $13 billion in 1990 to $27 billion in 1997.9,10 Acupuncture and other CAM treatments are big business. But is acupuncture safe, and does it work?
Basic Science
There is a substantial body of data supported by more than 100 scientific papers that the pain-relieving aspects of acupuncture are in part mediated by endogenous monoamines and neuropeptides that are activated by the needle. Opioids are released into the system as confirmed by reversal with naloxone and transfer of effect to another animal via cerebrospinal fluid transfusions.5,11 Dale12 demonstrated that acupuncture points are enhanced conductors of electromagnetic signals. Stimulating acupuncture points increases the relay capability of the electromagnetic signals that may direct flow of immune cells and other neurohormones to the injured area. Recently, Joos et al reported that acupuncture as an adjunct to conventional asthma therapy had significant immunomodulating effects.
Safety
When acupuncturists are appropriately trained, significant adverse events are rare. It is difficult to introduce long-lasting adverse physiologic/energetic effects with 1 or 2 acupuncture treatments. However, fear of litigation resulting from a coincidental adverse medical outcome may be a concern when performing acupuncture in certain situations. Addressing this issue, one recent study evaluated the pregnancy outcome in patients who had received acupuncture for nausea and vomiting in the 1st trimester. Smith et al14 studied 583 women as part of a blinded, controlled trial of acupuncture to treat nausea in early pregnancy. There were no differences in pregnancy complications, spontaneous abortion, stillbirth, neonatal death, or congenital anomalies among groups of acupuncture treatment, sham treatment, and no treatment.
Most adverse events from acupuncture are due to the introduction of a needle into the body. The most severe adverse reactions include pneumothorax, spinal cord injury, and hepatitis B infection. Most of these serious events were retrospectively reported in Japan and China and were due to broken or purposefully embedded needles. Yamashita et al15 attributes this to poor training programs in Japan.
In the United States and Britain, prospective studies identified 2 cases of pneumothorax out of 250,000 treatments16 and no incidence of spinal cord injury. When carefully performed, acupuncture involving the chest wall for asthma and chronic lung disease is safe without significant risk of pneumothorax.17 In separate prospective reviews of medical acupuncturists and non-physician providers, White et al18 and MacPherson et al19 reported that minor adverse events occurred in 1.3-1.4 per 1,000 treatments. These included nausea and fainting, prolonged aggravation of the presenting complaint, significant pain and bruising, and psychoemotional reactions. In a few instances, clinician error contributed to problems such as needles left in patients and burns from using moxibustion. Fifteen percent of patients reported transient reactions such as feeling relaxed or energized. None of the adverse events required hospitalization or unscheduled office visits.
One case of a possible wound infection to a total knee arthroplasty (TKA) following acupuncture was reported.20 However, a follow-up survey to medical acupuncturists in the American Academy of Medical Acupuncture (AAMA) identified no other incidence of wound infection in total joint surgery following nearly 8,000 acupuncture treatments. Because the acupuncture infection rate was less than the expected TKA postsurgical infection rate, the authors concluded that the infection was most likely coincidental and not due to the acupuncture. Hepatitis B infection has not been reported with the use of sterile disposable acupuncture needles.
The combined data from prospective and retrospective studies and case reports confirm that acupuncture is safe in the hands of trained providers.
Indications and Efficacy
In the United States, acupuncture has its greatest success and acceptance in the treatment of musculoskeletal pain. Problems such as acute sprains and strains are most easily treated, while chronic pain patients make up the largest numbers of patients seeking acupuncture from US physicians.21 Table 1 lists general indications for acupuncture as adapted from the AAMA Web site22 and the World Health Organization.23 Most of these indications are supported by textbooks or at least 1 journal article. However, definitive conclusions based on research findings are rare because the state of acupuncture research is poor but improving.
The 1997 National Institutes of Health Consensus Conference on Acupuncture7 concluded that clear evidence existed to support acupuncture efficacy in postoperative and chemotherapy-induced nausea and vomiting and in postoperative dental pain. The panel members cited other situations in which acupuncture may be useful as an adjunct treatment or an acceptable alternative: addiction, stroke rehabilitation, headache, menstrual cramps, tennis elbow, fibromyalgia, myofascial pain, osteoarthritis, low back pain, carpal tunnel syndrome, and asthma.
The main problem in evaluating scientific efficacy data is that many studies provide equivocal results due to flaws in design, small sample size, and the inherent difficulty in the use of controls such as placebo and sham acupuncture. Other problems with randomized controlled trials (RCTs) include testing of poorly defined illness with imprecise outcomes, heterogeneous study groups (e.g., all types of back or neck pain), and inadequate follow-up. Acupuncture RCTs are also subject to bias due to the difficulty in blinding both the acupuncturists and the patients. Sham techniques that include needle insertion in non-acupuncture points do not remove all the potential positive physiologic effects of needle insertion itself. Finally, the training and experience of the researcher, or lack thereof, may affect the appropriateness of the treatment arm of a particular study.
Acupuncture RCTs range from single-point treatment protocols24-27 to set programmatic treatments based on Western diagnoses,28,29 and to individualized treatments determined by the acupuncturist based on an Eastern diagnosis (e.g., Kidney Yin deficiency).30,31 Each of
the different types of RCTs represent variations in the practice of acupuncture. However, that technique variation makes it difficult to consolidate results into systematic reviews to determine whether acupuncture is effective or ineffective for a particular pathological condition. Rather, the studies show that various treatment protocols within acupuncture are effective or ineffective against a particular (sometimes not well-defined) disease process. From a Western perspective, it would be similar to studying the effect of penicillin therapy on a group of patients with viral and/or bacterial pneumonia and concluding that antibiotics are ineffective in the treatment of pneumonia. Another problem is the lack of agreement on the adequacy of an acupuncture treatment in terms of frequency and duration of treatments. To use the pneumonia analogy, insufficient dosages of the antibiotic would likely result in poor outcome results.
The problems above help explain why many of the systematic reviews and meta-analyses on acupuncture result in contradictory and inconclusive results.32 Nonetheless, they are used by the medical community and insurance industry to determine whether acupuncture is effective and should be covered as a benefit by 3rd-party insurers. A summary of some of the more recent RCTs and systematic reviews follows.
The study often quoted as supporting acupuncture for the treatment of depression is an example of small sample size leading to possibly faulty conclusions that permeate the literature. Allen33 concluded that acupuncture was helpful in treating unipolar depression when more than half of the 11 patients showed significant improvement in symptoms. No larger study is available for comparison.
Two areas where recent RCTs and systematic reviews consistently show a lack of efficacy for acupuncture are stroke and tinnitus. Two RCTs showed negative results for acupuncture efficacy in stroke when compared with transcutaneous electrical nerve stimulation34 and placebo.35 Additionally, a 2001 systematic review by Park et al36 found that the best-constructed RCT demonstrated the least acupuncture efficacy. These findings overcame early optimism generated by reported positive results from less-rigorous RCTs. Likewise, a systematic review of acupuncture for tinnitus found that none of the blinded studies demonstrated improvement.37 Wong et al demonstrated greater clinical improvement in neurological function of patients with spinal cord injury who underwent programmatic acupuncture treatment in conjunction with rehabilitation in Taiwan.38 However, as with the early stroke studies, there is not yet a large enough sample size from multiple trials to make a definitive conclusion regarding the impact of acupuncture on spinal cord injury.
The 2 areas best studied with consistent positive results for acupuncture are emesis and dental pain. Vickers systematic review on emesis39 found that every study that used the acupuncture point Pericardium 6 had a positive outcome on emesis of all causes. Recent RCTs demonstrate specific efficacy in the treatment of chemotherapy-induced emesis,40,41 hyperemesis gravidarum,24 and postoperative nausea and vomiting following pediatric tonsillectomy.27 Dental pain studies are summarized in a systematic review by Ernst and Pittler,42 who found that 12 of 16 studies demonstrated a positive effect of acupuncture.
One of the first systematic reviews of acupuncture trials of chronic pain was a meta-analysis done by Patel et al in 1989.43 They reported that acupuncture was efficacious in the treatment of chronic pain.
However, the majority of the studies they included were unblinded trials with small sample sizes (n=720 in 14 studies). The few placebo-controlled trials in their summary had negative results. More recent systematic reviews have had inconclusive or mostly negative findings.44,45 Systematic reviews46-48 are also inconclusive about acupuncture for chronic neck and back pain. Two recent RCTs show conflicting results when considered together. Leibing et al49 studied programmatic acupuncture treatments (all patients got the same treatment) for chronic low back pain vs sham acupuncture and physical therapy; they found that acupuncture was more favorable than the other modalities. On the other hand, Cherkin et al30 published a well-designed study comparing acupuncture with therapeutic massage and self-care education for chronic low back pain. That study was relatively unique in that the acupuncture arm allowed the TCM acupuncturist to select an individualized treatment plan for each patient. The authors concluded that therapeutic massage was better than acupuncture.
The last set of examples comes from studies on hip and knee osteoarthritis (OA). Most studies demonstrate positive efficacy for acupuncture.28,29,32,50-52 What sets these studies apart is that they follow a sequence that first identifies a successful treatment protocol and then refines it. Bermans initial studies28,52 confirmed that acupuncture works in knee OA. The next group of studies identified the timing of treatment in regard to the disease course50 (the earlier the better) and the amount of treatment necessary for effect; unilateral treatment was as good as bilateral treatment.51 The success of the research methodology in knee OA serves as a model of how rigorous Western scientific methodology may yet provide answers regarding the efficacy of acupuncture.
TREATMENT
The initial evaluation and course of treatment differ based on whether the acupuncturist combines treatment with Western biomedical paradigms or Eastern acupuncture paradigms. The medical acupuncturist will often use a combination of both. Therefore, the initial evaluation will include a history and physical examination that identifies signs and symptoms related to both paradigms. For example, the patient with an S1 radiculopathy would not only have the classic symptoms and signs of dropped reflexes and sciatica but also the signs and symptoms associated with kidney and bladder Yin/Yang imbalance or Qi blockage. The acupuncture treatment may vary based on a neuroanatomic vs TCM approach, but nonsteroidal anti-inflammatory drugs or other medications may be used as well. Other referral models have a physician taking care of the patient from a Western perspective and referring him/her to a non-physician acupuncturist outside of or connected with the practice.
Acupuncture treatments are usually weekly but sometimes more frequent for acute conditions. Once the effectiveness of a treatment lasts a full week, the treatments are scheduled further apart. As the response stabilizes for longer periods, the treatments are spread out further until a determination is made whether to stop or schedule periodic maintenance visits. Chronic conditions usually require maintenance visits every 2-3 months. The number of needles in a treatment ranges from 1-20, depending on whether the treatment is bilateral and which acupuncture method is used. Electrical stimulation is added to the needles in most medical acupuncture treatments. Heated moxibustion is also used in some treatments. A typical initial visit will last more than an hour with follow-up treatment visits lasting 30-60 minutes.
Approximately 70% of non-physician acupuncturists practice alone or in acupuncture groups; 30% work in multidisciplinary settings, usually in association with other alternative providers.2 Most physician acupuncturists are primary care providers in private practice and treat about 25% of their patients with acupuncture. Few do medical acupuncture exclusively. The 2 most common non-primary care specialties are anesthesiology and physiatry. Endorsement or use of other complementary methods such as manual medicine, herbal medicine, and supplements are common.21 Third-party reimbursement varies by locale with different fees and rules between physician and non-physician acupuncturists. Insurance coverage is more common in the Western states, though most patients still pay out-of-pocket for acupuncture.9,10
The AAMA lists medical acupuncturists who accept referrals.22 Non-physician acupuncturists maintain referral lists on a site maintained by TCM providers.53
According to medicalacupuncture