Non-penetrating Sham Acupuncture

Most randomised-controlled trials of acupuncture use penetrating sham acupuncture as control intervention. Sham acupuncture is described as acupuncture at points which are not known as acupuncture points. Because of physiological changes due to skin penetration sham acupuncture should not be defined as placebo control. In 1998 a new placebo needle was introduced by Streitberger. This placebo needle allows patient-blinding without penetration of the skin. Since then, this placebo needle and similar devices were used frequently as non-penetrating sham acupuncture in validation studies, experimental studies and randomised controlled trials (RCTs). This brief review will introduce the Streitberger placebo needle, the Park sham device and a simple blunt needle technique. Since 1999 more than 40 RCTs used one of these non penetrating devices. Significant better improvement in acupuncture compared to control was shown mainly in some pain conditions. However, many studies could not show a statistical significant difference between the two groups. Reasons might include a different power of placebo effects according to the condition treated, minimal effects by touching the skin with the sham needle and the fact that many studies were pilot studies with small sample sizes. Due to the heterogeneity of these studies a conclusive statement about the clinical effects of acupuncture is not possible yet. Especially in pain conditions further studies of acupuncture compared to non-penetrating sham acupuncture are necessary to lead to a better understanding of the importance of needle insertion. A critical discussion might allow to detect problems in existing studies and to improve protocols for further studies. 75 Konrad Streitberger, Japanese Acupuncture and Moxibustion, 2010; Vol.6(1): 74-79 ing into their body. For fixation of the needle a plastic ring is covered by a plaster (fig.1). In real acupuncture the same procedure is used and after puncturing the plaster the sharp tip of the needle is inserted into deeper tissue layers. Before using the placebo needle in a clinical trial it was tested in a cross-over experiment with 60 healthy volunteers whether a needling with the placebo-needle is as credible as with a real acupuncture needle. After randomisation the volunteers were needled in a cross-over design with a real acupuncture nee dle and with the placebo-needle at acupoint "Large Intestine 4". The volunteers were asked if they felt the needle pene76 Konrad Streitberger, Japanese Acupuncture and Moxibustion, 2010; Vol.6(1): 74-79 tration through the skin, how painful the penetration of the needle was (on a Visual Analogue Scale = VAS), and if they felt a dull pain (DEQI-feeling). Baseline data and results are shown in table 1. Concerning the comparative feeling of needle penetration 43 of 60 volunteers (72%, 95%-confidence interval: 59%; 83%) felt no difference. 12 volunteers felt penetration only with acupuncture (20%), 5 only with placebo (8%). As most important none of the volunteers suspected that the skin had not been punctured in one of both procedures. Therefore in this experiment the placeboneedle proved to be sufficiently credible to be used as a placebo in single blind conditions. Since then, this placebo needle and similar devices were used frequently as non-penetrating sham acupuncture in validation studies, experimental studies and RCTs. The Park sham device and other techniques Shortly after the description of the Streitberger needle a similar device with the same principle of a blunt telescopic needle but using a different applicator was introduced and evaluated by Park (fig 2). Also studies with more simple concepts like sticking a blunt needle into a foam applicator were described. Randomized controlled trials Acupuncture studies using non-penetrating sham acupuncture as control were identified by a systematic search covering the period from 1998 onwards in MEDLINE. Database searches were supplemented by screening of reference lists of systematic reviews and eligible primary studies. Participant blinded RCTs were described qualitatively in tables. Yet, the methodological quality of the studies was not analysed in detail. Since 1999 at least 40 RCTs used one of those non penetrating devices (table 2). Significant better improvement in acupuncture compared to control was shown 15 trials, mainly in some pain conditions. However, 25 studies could not show a statistical significant difference between both groups in the main outcome criterion. Using multiple testing in 11 of these studies at least in one secondary criterion a significant better result for acupuncture could be detected. From 20 studies including more than 50 patients only 6 showed a clearly significant effect in the main outcome criteria (table 3). Only 2 of 8 studies which included more than 100 patients proved a significant effect. 77 Konrad Streitberger, Japanese Acupuncture and Moxibustion, 2010; Vol.6(1): 74-79