Acupuncture is a complementary approach that originated from traditional Chinese medicine, from a theory that the flow of vital energy, or "Qi" (pronounced CHEE, I believe, unless someone who speaks Chinese tells me otherwise) can be regulated by stimulation of key body areas with needles, heat, or pressure. It is generally felt to be quite safe when performed by properly qualified individuals, although caution is appropriate for patients with a tendency toward increased bleeding or known low platelets. Acupuncture has been the subject of a reasonable amount of study, and it appears that there are significant physiologic effects of acupuncture, mediated through effects on the nervous system. Specifically, studies show effects in terms of changes during acupuncture in release of proteins in the brain, called neurotransmitters, as well as neuro-imaging like functional MRI. While it has historically been utilized for a very wide range of medical problems, the value of acupuncture in evidence-based medicine appears to be considerably more limited, but definitely relevant for a few common issues facing patients with cancer.

It is currently strongly recommended in the recent ACCP guidelines (abstract here) as an additional therapy for pain control among patients who don’t have good pain control, who have side effects from other treatments (such as the well-known constipation, nausea, somnolence, and mental status changes that can accompany otherwise very helpful narcotics), or who are trying to cut down on pain medications. Pain is the most common use and the setting in which acupuncture has been best studied. It’s been shown to be beneficial in treating both acute pain such as following surgery, as well as chronic pain such as headaches (abstracts here and here). One trial (abstract here) that randomized 570 patients with arthritic knee pain significantly demonstrated significant improvement among those who underwent a 26-week course of acupuncture compared with those who underwent sham acupuncture (conducted in a way outside of the proper principles but otherwise technically similar). All of these patients received other standard care for osteoarthritis, but eight weeks into the program, both pain and knee function were significantly better among those who received acupuncture according to proper techniques. A similar study (actually a companion article in the same issue of the same journal; abstract here) in chronic neck pain showed that acupuncture also improved survival in this setting. However, a recent meta-analysis indicated that acupuncture hasn’t been significantly beneficial for low back pain beyond other standard therapies (abstract here).

Fortunately, the available evidence indicates that acupuncture can be effective against cancer-related pain. One study of acupuncture against a placebo, sham acupuncture procedure tested acupuncture around the ear for patients who continued to have pain despite being on a stable pain regimen (abstract here). This wasn’t a large trial, including just 90 patients, but those who had needles placed at the supposedly effective acupuncture points had a decrease in pain intensity by 36% by two months into treatment; in the patients who did not receive acupuncture to presumed appropriate acupuncture points, there was little to no change in pain levels. What was also particularly impressive in this small study was that most of the patients had a type of pain called neuropathic pain that is generally regarded as particularly challenging to treat effectively. Studies of brain imaging show changes in certain key areas of brain activity that are seen with the "proper" form of acupuncture but not with control/placebo stimulations that are done at presumed ineffective points of stimulation (abstract here).

Currently, acupuncture is also being studied for "post-thoracotomy syndrome", the chronic pain that can result from cut nerves after open lung surgery; we don’t have results yet on its value in this setting.

Next, I’ll turn to the use of acupuncture for settings other than pain control.