Acupuncture for knee arthritis fails one test but may still be worth a try

Acupuncture is a popular form of complementary and alternative therapy, but it has yet to win universal endorsement in the medical community—and usually isn't covered by health insurance. Many satisfied customers continue to pay for treatment out of pocket in spite of mixed findings on the effectiveness of this ancient healing art.

A report published this week in the Journal of the American Medical Association (JAMA) offers weak-to-no proof that acupuncture helps ease the pain of knee arthritis. On the other hand, it's just one moderately sized study in a long and continuing series, and there's still credible evidence to suggest that acupuncture helps some people with common pain conditions.

"I would be careful saying acupuncture doesn't work for all pain conditions and no one should do it; we simply do not know enough yet," says Peter Wayne, PhD, research director of the Osher Center for Integrative Medicine at Harvard-affiliated Brigham and Women's Hospital.

Testing acupuncture

In the JAMA study, Australian researchers recruited 282 people over age 50 with moderate to severe knee pain from osteoarthritis. They were assigned at random to one of four options:

• no treatment
• traditional needle acupuncture
• laser acupuncture, which uses laser beams to stimulate acupuncture points on the body
• sham laser acupuncture, which looks like the real thing to study participants but doesn't deliver any laser light

Including the sham treatment was important. When comparing acupuncture to no treatment at all, part of the pain relief is attributable to the placebo effect. That's when people report feeling better after undergoing a treatment that they expect to work. Comparing real to "fake" acupuncture could reveal the effect of the treatment itself, minus the placebo effect.

In reality, though, it's more complicated than that. Just talking to and touching an acupuncture client during the treatment can exert a subtle healing power.

Weak results

As in previous studies, people who had needle or laser acupuncture reported less pain and better physical function compared with the group that had no treatment at all. Keep in mind that the differences were quite small—like a 1-point reduction in pain on a scale of 0 to 10. That's unlikely to impress someone with significant arthritis pain.

As for the sham acupuncture, the benefit disappeared when researchers compared the real and simulated treatment groups. Does it mean the effect of acupuncture was all placebo in this study—essentially, all in the participants' heads?

Not necessarily. The study had only 70 people in each of the four groups. That would make it difficult to pick up differences in the effect of real and fake acupuncture. Wayne, a board member of the Society for Acupuncture Research, believes the totality of the evidence suggests acupuncture does actually relieve pain for many conditions—although exactly how remains a mystery.

"This is a small study that replicates what we already know," Wayne says. "When you compare acupuncture to no treatment, there seems to be clinically meaningful differences for many pain conditions, including back pain and knee pain. Based on this pragmatic comparison, if I were deciding whether to send a family member or friend for a pain-related acupuncture treatment, I would say 'yes'."

Give it a try?

It's common for people to use acupuncture to enhance their existing treatment with pain relievers and physical therapy—not to replace it. "Most of the studies have treated people once or twice a week for a couple of weeks, and then continued with monthly maintenance after that," Wayne says. If after a few months you see no change, consider stopping.

One caveat is that there are multiple forms of acupuncture, such as needles that deliver a weak electrical stimulation. Also, acupuncture styles vary quite a bit from one practitioner to the next. That means if acupuncture doesn't help, you can't always be sure if the treatment didn't work for you, or the acupuncturist.

The typical cost of acupuncture in the United States ranges from $65 to $125 per session. Private insurers usually don't pay for it, nor do Medicare or Medicaid. A handful of plans may reimburse for physician-acupuncturists.

Harms from acupuncture are rare. Make sure you find a skilled practitioner who follows best practices for sterilizing needles and hand washing.

Ask your doctor to recommend a trusted provider. You can search for a trained acupuncturist in your area on the website of the National Certification Commission for Acupuncture and Oriental Medicine (www.nccaom.org) or call the organization at 904-598-1005.