Publications Archive

Making sense of the science—
Acupuncture for treating HIV neuropathy

Acupuncture therapy for HIV-related neuropathic pain was recently evaluated in a large, randomized placebo-controlled clinical trial. This study, which failed to support this use for acupuncture, has been criticized for the choice of acupuncture methods that it evaluated. Although the conclusions that can be drawn from this study may be limited to the specific acupuncture intervention, some of the methodological considerations for which this study has been criticized can be generalized to any acupuncture study.

Given the lack of an effective therapy for HIV-related neuropathic pain, and the reluctance of many infected individuals to add to an already heavy pill burden, many who suffer from this condition have sought alternative therapeutic approaches such as acupuncture.

Although acupuncture is apparently effective in treating chronic pain from many causes, and anecdotal information supports the use of acupuncture for treating HIV-related neuropathic pain, scientific evidence for this has been lacking. Since neuropathy affects as many as a third of all HIV-infected individuals, such data would be expected to substantially impact the quality of life of a great number of people.

This interest led to a randomized, controlled clinical trial to test whether acupuncture could reduce HIV-related neuropathic pain. In this study, a standard acupuncture regimen was compared against a "sham" acupuncture, which consisted of inserting needles at points not expected to produce any effects. (In pain research, the placebo effect can be quite substantial, making it necessary to devise a sham treatment against which the real treatment is compared). No difference was seen in the average reduction in reported pain between the acupuncture and sham groups.

How can these results be interpreted? Clearly, the standardized acupuncture treatment that was assessed in the study performed no better than placebo, but whether this can be generalized to other acupuncture treatments has been questioned on several grounds. Three of the main ones are:

• Choice of acupuncture treatment. A number of schools of acupuncture exist, each of which have a different philosophical approach to treatment. Thus, reaching a consensus on what the best acupuncture regimen to study is problematic. It is quite possible that acupuncture systems other than the one examined in the present study would be of benefit to HIV-infected people with neuropathic pain.

• Standardizing the acupuncture. Those patients that were randomly assigned to receive real (as opposed to sham acupuncture) were given the same treatment. On this basis, it has been argued that the present study did not address the effectiveness of acupuncture as it is routinely practiced, because in practice acupuncture is tailored to each individual. However, without incorporating standardized regimens into a study design, the study itself may not be replicable (that is, it may not be possible to either verify or study the results in further detail), and if the treatment proved effective, it would not be possible to know how to apply the results to routine practice.

Many clinical trials of acupuncture incorporate a standardized regimen, and in some cases have shown benefits for the procedure. A literature review of randomized clinical trials of acupuncture for the treatment of back pain was recently published by researchers from the University of Exeter in England. Several studies were clearly able to document significant benefits for back pain using a standardized regimen. On the other hand, studies that utilized individualized acupuncture did tend to show stronger beneficial effects.

• Quality of the sham. Some studies of acupuncture have shown that needle points chosen for the "sham" treatment can produce physiological effects. Since these physiological changes might produce the same benefits as the real acupuncture, this would tend to make acupuncture appear less effective. In the present study, the "sham" patients did no better than those who received a placebo consisting of an inactive pill, which reduces the likelihood that the benefits of the real acupuncture were masked in this way.

Controlled clinical trials are performed to assess the relative effectiveness of a treatment on average in a select group of people. Such trials are not designed to provide information about how a given individual is likely to respond to the treatment. The gap between trial results and the specific needs of each patient in routine care must be bridged by the expertise of the healthcare provider with the individual on a case-by-case basis.

Acupuncture is an example of a treatment option that magnifies this gap, since it is intended to be an individualized treatment. As such, although it can be studied in a clinical trial setting, the interpretations of trial results may be more problematic than trials of Western-style therapies.

These issues notwithstanding, one can surely conclude that the acupuncture intervention that was evaluated in the present trial was ineffective. Other acupuncture treatments might well deserve further study, but to date all that can be said is that the clinical evidence is lacking to support the use of acupuncture in treating HIV-related neuropathic pain.

The study discussed above was conducted at multiple sites of the Terry Biern Community Programs for Clinical Research on AIDS. Findings were published in the Journal of the American Medical Association (JAMA, Nov. 11, 1998, Vol. 280, page 1590). For a different perspective of the acupuncture study discussed above, and for more information on neuropathy, see the article "Numb and number" in the Jan. 1999 issue of Poz.