Complementary and alternative medicine

Incidence of adverse events

Much of the information about adverse events associated with complementary therapies is anecdotal, and assessment and classification of causality is often not possible. Likewise, there have been few attempts to determine systematically the incidence of adverse events associated with non-orthodox therapies.

Two prospective interview surveys have generated incidence figures for adverse reactions associated with chiropractic treatment. In a Norwegian investigation [ ], 102 chiropractors were asked to monitor 1058 new patients. At least one adverse event was reported by 55% of the patients at some time during the course of a maximum of six treatments. The most common adverse events were local discomfort (53% of total), headache (12%), tiredness (11%), or radiating discomfort (10%). A Swedish survey used similar methods to monitor 1858 chiropractic consultations and demonstrated an incidence rate for adverse events of 44% [ ]. In both studies, all the adverse events were mild and transient and no serious events were recorded.

In a prospective Japanese survey of adverse effects of acupuncture, 55 291 treatments were monitored between 1992 and 1997 [ ]. There were 64 adverse events in all; none was serious or led to permanent damage.

General adverse events

Even a perfectly safe remedy (mainstream or unorthodox) can become unsafe when used incompetently. Medical competence can be defined as doing everything in the best interest of the patient according to the best available evidence. There are numerous circumstances, both in orthodox and complementary medicine, when competence is jeopardized:

• Missed diagnosis;

• Misdiagnosis;

• Disregarding contraindications;

• Preventing/delaying more effective treatments (for example misinformation about effective therapies, loss of herd immunity through negative attitude towards immunization);

• Clinical deterioration not diagnosed;

• Adverse reaction not diagnosed;

• Discontinuation of prescribed drugs;

• Self-medication.

The most obvious danger is that some patients (or their parents) may elect to abandon conventional therapies for serious diseases in favor of alternative approaches. This can have fatal consequences [ ]. Survey data suggest that a sizeable proportion of complementary practitioners advise their clients to reduce their prescribed medication [ ].

Diagnostic techniques can also cause serious risks; an example is the reported over-use of X-rays by chiropractors [ ]. Although this practice seems to be in decline, it can put patients at risk through the mutagenic effects of X-rays.

The attitude of consumers towards complementary medicine may also constitute a risk. When 515 users of herbal remedies were interviewed about their behavior vis-a-vis adverse reactions to herbal versus synthetic over-the-counter drugs, a clear difference emerged. While 26% would consult their doctor for a serious adverse reaction to a synthetic medication, only 0.8% would do the same in relation to an herbal remedy [ ]. A further risk might lie in the plethora of lay books and other forms of information on complementary medicine now available in many bookstores and online. The authors of a pilot project attempted to evaluate the value of a random selection of such books and concluded that the lay literature on complementary medicine was far from adequate and had the potential to put the health of the reader at risk [ ].

The fear has been expressed that “with the increased interest in alternative medicine, we [shall] see a reversion to irrational approaches to medical practice” [ ]. The only way to minimize incompetence is proper education and training, combined with responsible regulatory control. While training and control are self-evident features of mainstream medicine they are often not fully incorporated in complementary medicine. Thus, the issue of indirect health risk is particularly pertinent to complementary medicine. Whenever complementary practitioners take full responsibility for a patient, it is desirable that this should be matched with full medical competence; if on the other hand, competence is not demonstrably complete, the practitioner in question should not assume full responsibility [ ].

Frequency of adverse reactions

Acupuncture keeps being associated with serious adverse events [ ]. Some reports of complications after acupuncture are summarized in Table 2 . Several attempts to define the size of this problem more closely have been published.

In a prospective investigation of the adverse effects of Japanese acupuncture, all adverse events experienced by patients in an acupuncture clinic seen between November 1992 and October 1997 were recorded [ ]. There were 64 adverse events. Failure to remove the needles, dizziness, discomfort, and sweating were the most frequent adverse reactions. There were no serious adverse events. These data suggest that superficial needling as used in Japanese acupuncture is relatively safe.

A Japanese survey of 391 patients who received acupuncture in 1441 treatment sessions involving a total of 30 338 needle insertions showed the following systemic adverse events: tiredness (8.2%), drowsiness (2.8%), aggravation of the presenting condition (2.8%), itching in the punctured region (1.0%), dizziness or vertigo (0.8%), faintness or nausea during treatment (0.8%), headache (0.5%), and chest pain (0.3%) [ ]. The incidences of local reactions were: minor bleeding after withdrawal of the needle (2.6%), pain on insertion of the needle (0.7%), petechiae or ecchymoses (0.3%), local pain after treatment (0.1%), subcutaneous hematomas (0.1%).

Norwegian researchers sent questionnaires about acupuncture to a random sample of the Norwegian general population [ ]. Of the 653 respondents, 7% claimed to have had adverse reactions. The most common were dizziness, fatigue, and pain from the needles. No serious adverse events were reported.

In a prospective UK survey of members of the medical and physiotherapy acupuncture organizations in Britain the preliminary data included 25 500 treatments given by 77 acupuncturists [ ]. There were 29 major events, including four episodes of loss of consciousness and one tonic-clonic seizure. The most common minor events were bleeding or hematoma (3%), aggravation of symptoms (1%), and pain during needling (0.9%).

In a survey of 1100 Australian providers of traditional Chinese medicine the adverse events of acupuncture were also monitored [ ]. There were 3222 events, including 64 cases of pneumothorax and 80 convulsions. No deaths were recorded.

A German survey of 29 acupuncturists included 409 patients who received 3535 acupuncture sessions [ ]. In 11.4% of these there were adverse events, which were usually mild and transient: slight bleeding (2.9%), hematoma (2.2%), and dizziness (1%). There were no serious events and the authors concluded that acupuncture is safe. Others have pointed out after reviewing the relevant literature that serious complications are infrequent but “responsible clinicians practicing acupuncture and seeing patients who use acupuncture should be aware of the adverse events associated with it” [ ].

In a prospective study of the use of acupuncture in the framework of statutory health insurance in Germany, 454 920 patients with at least one of three chronic pain conditions (headache, low back pain, and osteoarthritis) were treated by 8727 medical acupuncturists [ ]. At least one of the predefined adverse reactions or complications was seen in 7.9% of the patients. The most frequent reactions were “needling pain” (4.0%), “hematoma” (3.3%), and “bleeding” at the point of insertion (1.6%). In 0.4% of the patients, the acupuncturists reported “orthostatic problems”, in 0.3% “forgotten needles”, and in 0.7% “other events” (including “local skin irritation”, “worsening of symptoms”, “symptom aggravation”, “fatigue”, and a “sensation of warmth”). During the 2-year reporting period, serious adverse reactions due to acupuncture were reported in 13 patients: pneumothorax (n = 3), acute hypertensive or hypotensive crises (n = 6), erysipelas (n = 2), acute asthma (n = 1), and aggravation of suicidal thoughts (n = 1). This represents a ratio of 1 in 34 994 of the total patient population and 1 in 295 000 of the total number of acupuncture sessions. The authors concluded that acupuncture provided by qualified therapists is safe and that patients benefited. They also found that more acupuncture training did not correspond to better therapeutic effects.

Two prospective UK studies with a total of about 70 000 consultations have confirmed that serious adverse events of acupuncture are true rarities in Britain [ , ]. Bleeding and needle pain were the most frequent adverse events, with a prevalence of about 1:1000.

In a Korean retrospective cross-sectional survey of 1095 subjects who used acupuncture, 75 (6.8%) described negative short-term acupuncture reactions, including feelings of pain in 37 (3.4%), tiredness in 24 (2.2%), and dizziness in 9 (0.8%). The only adverse event reported was bleeding in 92 (8.4%) of the participants [ ].

In a German observational study of 503 397 treatments documented between July 2001 and June 2003, physicians recorded at least one adverse event in 7.8% of all patients, the most frequent being needling pain in 3.9% [ ]. Serious adverse events were reported in 17 cases, the most frequent event being pneumothorax (five cases).

Of 6140 patients who received acupuncture, 9.3% reported adverse reactions, and a quarter of these were considered to be troublesome. The most frequent adverse reactions were pain, fatigue, and circulatory disturbances. In 20 patients who received acupuncture for acute non-penetrating limb injuries, only minor complications were reported, including local pain at the acupuncture site, light-headedness, sweating and pruritus, erythema, and minor bleeding at the acupuncture site [ ].

Systematic reviews

A systematic review of case reports from the Japanese literature yielded 105 cases of suspected acupuncture adverse events not previously reported in Western publications [ ]. These included 21 spinal lesions, 21 cases of pneumothorax, 19 infections, 15 cases of foreign bodies in organs, 10 instances of argyria, 10 neural injuries, and 11 other adverse events, including two cases of cardiac tamponade.

A systematic review of all prospective studies of adverse events associated with acupuncture included nine primary investigations [ ]. The most commonly reported adverse events were needle pain (1–45%), tiredness (2–41%), and bleeding (0.03–38%). Pneumothorax was the only serious complication in these studies; it was reported twice in about 250 000 patients.

A systematic review of all adverse events associated with acupuncture in the Japanese medical literature located 124 cases [ ]. These included 25 cases of pneumothorax, 18 cases of spinal cord injury, 11 cases of hepatitis B, and 10 cases of localized argyria.

In a review of case reports and prospective surveys of adverse events focusing on “Japanese acupuncture”, almost all of the adverse reactions commonly seen in acupuncture practice, such as fatigue, drowsiness, aggravation of existing symptoms, minor bleeding, pain on insertion, and subcutaneous hemorrhage, were mild and transient [ ].

Cardiovascular

Acupuncture has been associated with hemopericardium due to ventricular puncture [ ].

• An 83-year-old Austrian woman developed syncope and cardiogenic shock shortly after acupuncture over the sternum. Echocardiography showed cardiac tamponade and pericardiocentesis revealed hemopericardium. At operation a small bleeding perforation of the right ventricle was found and closed. The acupuncture at the point “Ren 17” was above a sternal foramen, which allowed the needle to penetrate the heart.

A case in which acupuncture was apparently responsible for cardiac dysrhythmias has been reported [ ].

Vasovagal syncope has been reported after routine use of acupuncture; the authors suggested that this is an uncommon complication [ ].

• A 72-year-old man visited an acupuncture clinic in Taipei, Taiwan because of pain over the right forearm for 5 months. He had no previous history of neurological deficits. He received six sessions of manual acupuncture with limited improvement. During the seventh session acupuncture and electro-acupuncture stimulation, 2 Hz, was applied at points of Li-11 and TP-5. After 5 minutes he complained of dizziness. He then experienced cold sweating and suddenly lost consciousness. Irregular tonic–clonic movements of the right arm, upward rolling of the eyes, and batting of both eyelids lasted 20–30 seconds after the attack of syncope. All needles were immediately removed. Strong needling stimulation at Du-26 (Renzhong) was performed. He remained confused for 2–3 minutes. About 10 minutes after this episode, his supine blood pressure was 144/82 mmHg with a pulse rate of 62/minute. He felt well, except for fatigue, after 40 minutes of rest. His laboratory tests, including blood chemistry, blood glucose, thyroid function, a complete blood count, and urinalysis, were normal. An electrocardiogram showed mild ST-T wave depression, implying myocardial ischemia. Further evaluation was suggested but he refused.

• A 63-year-old woman received acupuncture for pain over the left ankle for 5 days. She had been well in the past and denied systemic diseases and use of medications. She received manual needling at GB-34 and GB-40 points. After a 3-minute session she complained of dizziness and nausea, and then experienced cold sweating and suddenly lost consciousness. She had irregular tonic–clonic movements of the left arm with batting of both eyelids lasting for 10–15 seconds and lip cyanosis after the episode of syncope. All needles were immediately removed and she received strong needling stimulation at Du-26. The event lasted for 1 minute and she remained confused for 2–3 minutes. About 15 minutes after the episode, her supine blood pressure was 126/74 mmHg with a pulse rate of 56/minute. After 60 minutes of rest, she felt chest tightness and fatigue. She was referred to a cardiologist for further evaluation. Laboratory tests, including blood chemistry, blood glucose, thyroid function, a complete blood count, and urinalysis, were normal. Electroencephalography was also normal. However, electrocardiography showed mild ST-T depression in V3–5. A treadmill test was positive and 24-hour Holter monitoring showed intermittent ischemic ST segment depression.

Respiratory

Pneumothorax has been reported in association with acupuncture [ ].

• A 50-year-old Caucasian woman suddenly developed severe right-sided chest pain and breathlessness 8 hours after receiving acupuncture to her scapulothoracic region [ ]. She was hypoxic and tachycardic (oxygen saturation 91% breathing room air). Her trachea was central, but there were reduced breath sounds, poor chest wall excursion, and hyper-resonance of the right hemithorax. A chest radiograph confirmed a moderate sized right-sided pneumothorax. An intercostal chest drain was inserted, with good effect. She required opiates for pain but otherwise made an uneventful recovery.

• A 28-year-old Chinese woman developed bilateral pneumothoraces after receiving acupuncture in the upper thoracic and paraspinal regions [ ]. She was treated conservatively and was discharged after 2 days.

• A 30-year-old woman developed bilateral chest pain and dyspnea after paraspinal acupuncture resulted in bilateral pneumothorax; she recovered fully within 2 days [ ].

A fatal case of acupuncture-induced pneumothorax has been reported [ ]. The patient developed dyspnea and chest pain directly after the treatment and died before adequate medical help was available. At autopsy ecchymoses were noted on the parietal pleura, suggesting that needles had been inserted into the thoracic cavity and had perforated the lungs. Microscopically many black spots were seen on the parietal pleura along the vertebral column. The authors interpreted them as traces from previous acupuncture treatments, which had been dangerously close to causing pneumothorax.

• A 25-year-old woman had a bilateral pneumothorax with pericardial and peritoneal effusions during acupuncture treatment [ ]. She was successfully resuscitated and made a full recovery.

The Japanese press reported a similar case [ ], but in this instance the life of the patient was not saved.

All patients should be counseled about the possibility of pneumothorax when needling is performed in the thoracic region. The acupoints on the inner bladder lines should be needled obliquely towards the spine, in order to avoid needle-induced pneumothorax.

Nervous system

Several reports have illustrated that acupuncture needles can cause nerve damage.

• A 62-year-old woman developed left drop foot, anterior leg pain, and numbness [ ]. The symptoms occurred after acupuncture for sciatica. Radiographs showed a needle-like object near the fibular head; it was removed surgically.

• A 68-year-old woman developed symptoms consistent with spinal stenosis [ ]. Microscopic examination showed a chronic inflammatory epidural granuloma compressing the lumbar fourth nerve and dural sac. Her history suggested that it had been caused by acupuncture. Surgical excision of the granuloma led to a full recovery.

• When a 67-year-old woman was investigated for pancytopenia, multiple needle fragments were noted on an X-ray along the anterior and posterior thoracic and abdominal walls [ ]. They had probably originated from acupuncture 17 years before, when she had had “Hari” acupuncture, which involves leaving small gold needles permanently at acupuncture points. She did not seem to have had any symptoms from these remnants.

The authors of the last report cautioned that such fragments may be mistaken for calcified organs at X-ray.

Acupuncture has been associated with subarachnoid hemorrhage due to arterial puncture [ ].

• A 44-year-old Chinese man had severe occipital headache, nausea, and vomiting during acupuncture in the posterior neck. A CT scan showed hemorrhage in the third, fourth, and lateral ventricles, and blood was found in the lumbar fluid. The problem was due to puncture of a branch of the vertebral artery at the “feng fu” point, which coincides with the site for performing cisternal puncture. The patient made a spontaneous full recovery within 28 days.

• A 43-year-old woman suddenly developed a headache, dizziness, and heavy sweating with vomiting during the second acupuncture treatment for low back and leg pain [ ]. The needles were removed immediately and she was helped to lie down. Most of her symptoms resolved but the headache persisted for 2 days. A subarachnoid hemorrhage was later confirmed by CT scan.

Convulsions, inadvertent anesthesia, loss of co-ordination, and tinnitus have been associated with acupuncture [ ].

Xiaozhendao (which means “small-needle-knife”) is a modified form of acupuncture, in which needles with flat edges on the needle tips are used to perform minimal soft-tissue dissection. In recent decades this form of treatment has become increasingly popular for a variety of conditions, such as acute and chronic pain, bursitis, tenosynovitis, herniated intervertebral disks, arthritis, spinal stenosis, and congenital bony abnormalities. Reports of adverse events associated with this modified form of acupuncture are rare. However, a broken needle was found in the cervical spine 3 years after Xiaozhendao treatment [ ].

• A 29-year-old man gradually developed neck pain, bilateral shoulder pain, and sensory disturbance of the left lateral arm. A cervical spine radiograph showed a 4-cm metallic needle that tilted toward the tracheal wall at the level of C2. A cervical spine CT scan confirmed that the needle had penetrated the epidural space from the left side into the lateral mass of the C2 vertebra. He had received Xiaozhendao treatment in his posterior neck region for chronic dizziness and stiffness of the neck from a non-medical practitioner about 3 years before. A 4.8 cm Xiaozhendao needle was removed surgically and he made an uneventful recovery.

Endocrine

Galactorrhea has been associated with acupuncture.

• A 41-year-old woman with breast cancer was treated with acupuncture for pain control [ ]. She had an episode of galactorrhea 6 days after the first treatment and also during the second acupuncture treatment. No reason for this unusual phenomenon other than acupuncture could be found.

The authors pointed out that in Chinese medicine acupuncture at the points used in this patient promotes lactation.

Liver

• A 52-year-old woman presented with a 3-day history of anorexia and jaundice after receiving acupuncture twice a day for 7 weeks, performed bilaterally at the Zusanli (ST36) acupoint to a depth of 22 mm [ ]. Electrical stimulation was performed, with the stimulation frequency fixed at 5 Hz for 20 minutes. Her aspartate aminotransferase and alanine aminotransferase activities and total bilirubin concentrations were 84 U/l, 109 U/l, and 216 mmol/l respectively She developed pale stools, dark urine, pruritus, pedal edema, and diarrhea during the next 12 days. Her laboratory results continued to worsen. No specific therapy was provided for the severe cholestatic jaundice. Over the next 12 weeks, her symptoms and the laboratory results gradually improved.