Acupuncture as Placebo

Sticking needles in someone’s body is not a placebo therapy, or at least not an effective one, argue two letters in today’s issue of Archives of Internal Medicine. From personal experience I’d argue that the critics got this call right.

Last year, a paper in Archives described findings from a randomized, multicenter program to treat chronic lower back pain. In it, Michael Haake of the University of Regensburg, in Bad Abbach, and his colleagues administered conventional therapy (drugs, physical therapy, and exercise) to one-third of their nearly 1,200 volunteers. The remaining patients got 30-minute acupuncture sessions, generally twice a week for five weeks. Of these, half of the patients were punctured at sites suspected of being useful, and the rest received needles at other spots on their bodies.

Interestingly, comparable shares of each acupuncture group – roughly 45 percent – reported decreased back pain for at least six months after their sessions ended. That success rate was almost twice as high as for the group getting conventional therapy. The authors concluded: “The underlying mechanism may be a kind of superplacebo effect produced by placebo and all nonspecific factors working together.”

Critics from the University of Notre Dame Australia make some interesting points in their Archives letter today. “It is highly likely,” they say, “that conventional therapy had already failed in most of this population.” Otherwise, why would they have taken part? It’s therefore also unlikely they would have responded to conventional treatment in the German trial. By contrast, the Australian team says, acupuncture’s novelty for most Europeans would be expected “to lead to a strong placebo response.”

Indeed, because both the ostensibly therapeutic and sham acupuncture sites proved equally beneficial, these researchers conclude that “a reasonable interpretation of this result is that acupuncture is not effective beyond a placebo effect.” For the German team to have made the opposite interpretation the focus of its paper “is both misleading and unscientific,” the UNDA researchers contend. Writing in a second letter, researchers from the Federal University of Santa Catarina, Brazil, offer a different spin. Because the sham acupuncture sites were as beneficial as the test sites, they constituted “neither an inert nor innocuous procedure” – which is what a placebo should be.

The latter is something I learned the hard way.

I’m a long-time migraine sufferer and four years ago my neurologist offered me a chance to enroll in an acupuncture trial at GeorgeWashingtonUniversity. Three times a week for eight weeks I showed up to be stuck with longish needles in my feet and hands by a traditional Chinese acupuncturist. (During the trial, I was not allowed to speak with him; then again, I’m not sure whether he spoke English very fluently.) Neither the physicians nor the volunteers knew until after the trial who was getting needles inserted in the test versus the sham sites.

As it turned out, I got the sham sites.

I also got pain. Lots of it. For whatever reason, about 10 minutes into my 30- or 40-minute treatment, I’d develop throbbing pain deep within my right upper arm. The discomfort eventually became intense by the time each session finished. And that throbbing didn’t disappear for another 90 minutes or so. Which made riding home on a crowded subway car challenging, since I couldn’t lift that arm to hold onto the grab bar (or hold my bags in that arm if I tried to grab the stabilizing bar with my other hand).

My doctor offered to let me drop out of the trial. But I was doing this for science. And if it reduced the frequency of my headaches, it would have been a small price to pay. In fact, I still got roughly 80 headaches a year after the trial, the same rate I had before it – and that I experience still.

As the Brazilian team points out, such a test procedure is hardly inert. It’s more like comparing effects of a cancer drug to a heart drug in a cholesterol trial. So I, like the Brazilian clinicians, would argue sham acupuncture is no placebo.

And Haake’s team, too, now concedes “sham acupuncture is not a placebo” in a reply they offer to their critics. It’s possible, they write, that whatever nerves were stimulated by inserting needles at the test sites were also inadvertently stimulated by the presence of needles at the sham sites. Moveover, the German team adds, “even a placebo effect is associated with real biochemical process in the brain, which blur the boundaries between a ‘real’ treatment with specific action mechanisms and a placebo treatment without such ‘specific’ mechanisms.”

Hmmm… What the German study and the one I participated in would both seem to point out is that there may be some treatments for which a true placebo is unavailable.