RSS Feeds
-
Acupuncture for OA knee is unlikely to be principally placebo
Submit responseTo the editor: We congratulate the authors of this paper for performing a rigorous review of acupuncture for osteoarthrosis of the knee (1). We also performed a review with meta-analysis, and our results were very similar (2). Our interpretation of the findings differed however, and we would like to draw this to the attention of your readers.
The methodological difficulties of devising a suitable placebo in order to study the specific effects of acupuncture have been debated for some time (3). As Manheimer et al themselves recognise, sham acupuncture usually involves inserting needles and is likely to have a direct physiological effect. Therefore it is unfair to demand that acupuncture must show a clinical superiority over sham acupuncture; statistical superiority is enough to demonstrate that acupuncture has a biologic effect. When it comes to choosing treatments for patients, we should look to the comparisons with usual care, or standard care, since this is the choice faced by patients and their physicians. The large effect identified by Manheimer et al (standardised mean difference (SMD) 0.62 for pain compared with usual care) is in line with our own calculations (4) and is clinically highly relevant. The largest trial included in these reviews (n=1007), published in this journal, demonstrated that both acupuncture and sham acupuncture were superior (SMD = 0.67 for acupuncture) to standard care involving physiotherapy and as-needed anti-inflammatory drugs (5). We suggest that the overall effect of acupuncture is due to the physiological effects of needle insertion at any point (shown in sham acupuncture) together with the specific effect of correct stimulation (as in real acupuncture). Both these effects are facilitated by expectation, as has been demonstrated for analgesic drugs (6). It seems inherently unlikely that these large effects of acupuncture are due principally to placebo, as suggested by Manheimer et al in their conclusion.
In our review, we applied ‘adequacy of acupuncture’ as an inclusion criterion. This was an important methodological innovation. We also applied criteria for the suitability of sham. Only one trial used what we would consider to be optimal acupuncture from a physiological perspective, and compared this with a suitable (non-penetrating) sham (7). The effect size in this trial was much larger than the rest, and while this was a relatively small trial (n=97), we hope that it will guide future trial designs and reviews in this field.
Reference List
(1) Manheimer E, Linde K, Lao L, Bouter LM, Berman BM. Meta- analysis: acupuncture for osteoarthritis of the knee. Ann Intern Med. 2007;146:868-77.
(2) White A, Foster NE, Cummings M, Barlas P. Acupuncture treatment for chronic knee pain: a systematic review. Rheumatology (Oxford). 2007;46:384-90.
(3) White AR, Filshie J, Cummings TM. Clinical trials of acupuncture: consensus recommendations for optimal treatment, sham controls and blinding. Complement Ther Med. 2001;9:237-45.
(4) White A, Kawakita K. The evidence on acupuncture for knee osteoarthritis - editorial summary on the implications for health policy. Acupunct Med. 2006;24 Suppl:S71-S76.
(5) Scharf HP, Mansmann U, Streitberger K, Witte S, Kramer J, Maier C et al. Acupuncture and knee osteoarthritis: a three-armed randomized trial. Ann Intern Med. 2006;145:12-20.
(6) Colloca L, Lopiano L, Lanotte M, Benedetti F. Overt versus covert treatment for pain, anxiety, and Parkinson's disease. Lancet Neurol. 2004;3:679-84.
(7) Vas J, Mendez C, Perea-Milla E, Vega E, Panadero MD, Leon JM et al. Acupuncture as a complementary therapy to the pharmacological treatment of osteoarthritis of the knee: randomised controlled trial. BMJ. 2004;329:1216.
Conflict of Interest:
Lead author is employed by the British Medical Acupuncture Society, which is a registered charity in the UK, set up to encourage the use and scientific evaluation of acupuncture for the public benefit.
-
Needling With Acupuncture May Still Have A Benefit
Submit responseIt is with great interest into both the article and subject matter that we reviewed the article "Meta-analysis: Acupuncture for Osteoarthritis of the Knee" by Manheimer et al. As a medical student doing a rheumatology summer scholar program we have much interest into acupuncture and other complementary medicine modalities. However, before dismissing the role of acupuncture in the treatment of osteoarthritis, we have several concerns about the conclusions from this meta-analysis.
We are impressed with the result of acupuncture showing short term improvement in pain (-0.96) and function (-0.93) compare to wait list control and short and long term improvement compare to the usual care control group. However, when acupuncture is compared to the sham control group there is no difference in improvement. The author mentioned patient preference, expectations, and drop-out rates as being possible confounding variables preventing accurate interpretation. These nonrandomized patient selection bias can be avoided if the experiments are done using cross over design studies, in which each patients have a period of treatment and then nontreatment and can act as their own control, thereby minimizing variability between patients and between acupuncturists. And wash out periods may be put in between to avoid carry over effects. With cross over studies everyone will get treated, thus avoiding the high drop out rate. Also, it will not be necessary to have multiple (shame, usual care, wait list) control groups. In addition, the lack of reliable long term studies on the effect of acupuncture seems to be a limitation in the study.
Hopefully, further studies will further clarify the role of acupuncture and other CAM's in the treatment of OA and other rheumatic disorders.
Conflict of Interest:
None declared
-
If acupuncture is "clinicallly irrelevant" so is fluoxetine
Submit responseDear Sirs, Manheimer et al. (1) show that the effect size of acupuncture for osteoarthritis when compared to sham acupuncture is 0.35 and label it as "clinically irrelevant". They therefore suggest that the much larger effect sizes observed when acupuncture is compared to a waiting list or usual care control (0.45 to 0.96) can be attributed to a placebo effect.
Before jumping to this conclusion, it is important to bear in mind that the effect size of fluoxetine in the short-term treatment (at least 6 weeks) of major depressive disorder is at best 0.37 when it, too, is compared to a placebo. (2)
Yet, few psychiatrists or internists would be comfortable relinquishing their reliance on fluoxetine or equivalent antidepressants (none of which have ever been found to be more effective than fluoxetine) in the treatment of major depressive disorder. Certainly, few among us would consider fluoxetine to be "clinically irrelevant".
1. Manheimer E, Linde K, Lao L, Bouter LM, Berman BM. Meta-analysis: Acupuncture for Osteoarthritis of the Knee. Archives of Internal Medicine 2007;146(12):868-77.
2. Bech P, Cialdella P, Haugh MC, et al. Meta-analysis of randomised controlled trials of fluoxetine v. placebo and tricyclic antidepressants in the short-term treatment of major depression. British Journal of Psychiatry 2000;176:421-8.
Conflict of Interest:
None declared