The effect of sham acupuncture can differ depending on the points needled in knee osteoarthritis: A systematic review and network meta-analysis

Abstract

Objective: In sham acupuncture-controlled acupuncture clinical trials, although sham acupuncture techniques are different from those of verum acupuncture, the same acupuncture points are often used for verum and sham acupuncture, raising the question of whether sham acupuncture is an appropriate placebo. We aimed to examine the effects of sham and verum acupuncture according to the points needled (same or different between verum and sham acupuncture) in knee osteoarthritis.

Methods: Ten databases were searched to find randomized controlled clinical trials (RCTs) assessing the effects of verum acupuncture with sham acupuncture or waiting lists on knee osteoarthritis. Sham acupuncture was classified as using the same acupuncture points as those in verum acupuncture (SATV) or using sham points (SATS). A frequentist network meta-analysis (NMA) was conducted, and the certainty of evidence was evaluated.

Results: A total of 10 RCTs involving 1628 participants were included. Verum acupuncture was significantly superior to SATS but not different from SATV in terms of pain reduction. Additionally, SATV was significantly superior to the waiting list. For physical function, no difference were found between verum acupuncture, SATV, and SATS. The certainty of evidence was low to moderate.

Conclusion: For knee osteoarthritis, the pain reduction effect of acupuncture may differ according to the needling points of sham acupuncture, and the control group should be established according to the specific aim of the study design and treatment mechanism.

Keywords: Acupuncture therapy; Knee osteoarthritis; Network meta-analysis; Placebo; Sham acupuncture.

Fig. 1

Flow diagram of the literature screening and selection processes.

Fig. 2

Network map of (A) pain and (B) physical function. AT, acupuncture therapy; SATS, sham acupuncture needling at points different from those in the verum acupuncture group; SATV, sham acupuncture needling at the same acupuncture points as those in the verum acupuncture group; WL, waiting list.

Fig. 3

Risk of bias summary for all included studies. Low risk of bias, some concerns, and high risk of bias, respectively, are represented with the following symbols: “+”, “?”, and “-”.

Fig. 4

Interval plots of (A) pain and (B) physical function. AT, acupuncture therapy; CI, confidence interval; SATS, sham acupuncture needling at points different from those in the verum acupuncture group; SATV, sham acupuncture needling at the same acupuncture points as those in the verum acupuncture group; SMD, standardized mean difference; WL, waiting list.

Fig. 5

Funnel plot: Pain. AT, acupuncture therapy; SATS, sham acupuncture needling at points different from those in the verum acupuncture group; SATV, sham acupuncture needling at the same acupuncture points as those in the verum acupuncture group; WL, waiting list.