Acupuncture delivered either traditionally through needles or by laser does not significantly improve pain or function at 12 weeks compared with sham acupuncture in older patients with chronic knee pain, new research shows.
The study was published in JAMA.
Rana Hinman, PhD, from the University of Melbourne, Victoria, Australia, and multicenter colleagues found that neither needle nor laser acupuncture significantly improved pain or function compared with sham controls at 12 weeks in adults 50 years of age and older who had chronic knee pain.
The researchers saw modest improvement in pain in both needle and laser acupuncture groups at 12 weeks compared with in control patients, but they found no differences between the groups in pain levels at 1 year.
Similarly, needle acupuncture resulted in modest improvements in function compared with control patients at 12 weeks but was not significantly different from improvements seen with sham acupuncture.
Significant functional differences between the needle acupuncture group and the control patients were lost at 1 year as well.
“Clinical guidelines recommend conservative nondrug treatments for knee osteoarthritis, but many patients seek complementary and alternative treatments, the most common being acupuncture, and use of such treatments is increasing,” coinvestigator Kim Bennell, PhD, from Centre for Health, Exercise and Sports Medicine, University of Melbourne, Australia said.
“But we found that neither needle nor laser acupuncture conferred benefit over sham or pain or function in patients 50 years of age and older with moderate or severe chronic knee pain. So our findings do not support acupuncture for these patients.”
The investigators recruited 282 patients aged 50 years or older to the study. All had knee pain lasting more than 3 months and on most days, with an average severity of 4 or more out of 10 on a numeric rating scale. Patients also had morning stiffness lasting less than 30 minutes.
Roughly equal numbers of patients were randomly assigned to 1 of the 4 treatment groups: needle acupuncture (n = 70), laser acupuncture (n = 71), sham laser (n = 70), or no acupuncture control (n = 71). At 12 weeks, 9% of the group was lost to follow-up, as was 18% of the group at 1 year.
“We used a combined Western and traditional Chinese medicine style of acupuncture,” the investigators write. A total of 8 family physicians who were members of the Australian Medical Acupuncture College delivered the acupuncture. Practitioners had a mean of 19.6 years of acupuncture experience.
Participants completed questionnaires at baseline, 12 weeks, and 1 year after receiving their respective treatment.
The primary outcomes of the study were valid and reliable self-reported pain and function measures for osteoarthritis. Secondary outcomes included average knee pain on walking and standing, average daily activity restriction during the previous week, pain on the Western Ontario and McMaster Universities Osteoarthritis Index pain subscale, health-related quality of life, and physical and mental component summary scores.
As Dr. Bennell said patients typically volunteer for studies because they have an interest, and a belief, in the treatment under investigation.
“Hence, for acupuncture studies, patients volunteer to participate often because they have an expectation that acupuncture will be effective,” she observed. It is known that expectation of benefit from a given intervention positively influences treatment outcomes.
To control for this, investigators used what is called a “Zelen” design. As Dr. Bennell explained, in this study design, participants initially consented to a natural history study that involved completion of questionnaires about knee pain and symptoms over the course of 1 year.
Once enrolled, the patients were covertly randomly assigned to 1 of the 4 treatment groups in the trial. “Patients allocated to 1 of the 3 treatment groups were invited to take part in the treatment, while those in the control group continued to be unaware they were part of a clinical trial,” Dr. Bennell added.
This design ensured trialists did not recruit patients with positive acupuncture attitudes, but it also ensured that participants in the no-treatment group were not disappointed in being allocated to the control group, both of which can adversely affect outcomes, Dr. Bennell pointed out.
Other acupuncture trials, including a meta-analysis suggested acupuncture was superior to both sham acupuncture and standard care for different types of chronic pain, indicating the effects of acupuncture are more than just a placebo effect.
Asked why she thought her study did not support these findings, Dr. Bennell suggested minimization of sources of bias across the trial design might explain why their results differed from those of other studies.
“Clinical guidelines recommend exercise and weight loss if patients are overweight or obese for people with knee osteoarthritis,” Dr. Bennell said. “So treatment should involve patient education, strengthening or aerobic exercise or both, and weight loss, if appropriate.”
Andrew Vickers, PhD, lead author of the meta-analysis and attending research methodologist, Memorial Sloan-Kettering Cancer Center, New York City said that there are 2 common myths about medical studies.
The first is that all studies should come up roughly with the same result, he noted. And the second is that whatever the latest result shows is the truth.
“In fact, studies come up with different answers for reasons that are very well understood, and there is no reason to give prominence to the most recent research,” Dr. Vickers observed.
Rather, most clinical trialists agree that data from different studies need to be combined into a single meta-analysis to give a bigger picture about a specific area of research.
“Results of the current study are very, very similar to that of the Acupuncture Trialists’ Collaboration meta-analysis,” Dr. Vickers elaborated. “In both studies, the difference between acupuncture and no treatment is about half a standard deviation, and between acupuncture and sham is about one fifth of a standard deviation.”
The reason why the current study did not find a significant difference between acupuncture and sham is because it was a small study involving a total of 280 patients compared with nearly 18,000 patients who were included in the meta-analysis, he added.
“The story here is not ‘acupuncture doesn’t work’ but why authors continue to overinterpret their own studies as the ‘final answer’ when, in fact, they are a small part of the overall evidence base,” Dr. Vickers concluded.
The results of the meta-analysis in acupuncture, run by a large international group called the Acupuncture Trialists’ Collaboration, under Dr. Vickers and colleagues were published in the Archives of Internal Medicine in 2012 (2012;172:1444-1453).