BACKGROUND: Acupuncture reportedly relieves chronic knee pain and improves physical function in patients diagnosed with osteoarthritis, but the duration of these effects is controversial. The aim of this study was to evaluate the temporal effects of acupuncture on chronic knee pain due to knee osteoarthritis by means of a meta-analysis.
METHODS: The PubMed, Embase, and Cochrane Central Register of Controlled Trials databases were searched for studies published through March 2015. Ten randomized controlled trials of acupuncture compared with sham acupuncture, usual care, or no intervention for chronic knee pain in patients with clinically diagnosed or radiographically confirmed knee osteoarthritis were included. All of the studies were available in English. Weighted mean differences (WMDs), 95% confidence intervals (CIs), publication bias, and heterogeneity were calculated.
RESULTS: The acupuncture groups showed superior pain improvement (p < 0.001; WMD = -1.24 [95% CI, -1.92 to -0.56]; I(2) > 50%) and physical function (p < 0.001; WMD = 4.61 [95% CI, 2.24 to 6.97]; I(2) > 50%) in the short term (up to 13 weeks). The acupuncture groups showed superior physical function (p = 0.016; WMD = 2.73 [95% CI, 0.51 to 4.94]; I(2) > 50%) but not superior pain improvement (p = 0.199; WMD = -0.55 [95% CI, -1.39 to 0.29]; I(2) > 50%) in the long term (up to 26 weeks). Subgroup analysis revealed that the acupuncture groups tended to have better outcomes compared with the controls. Significant publication bias was not detected (p > 0.05), but the heterogeneity of the studies was substantial.
CONCLUSIONS: This meta-analysis demonstrates that acupuncture can improve short and long-term physical function, but it appears to provide only short-term pain relief in patients with chronic knee pain due to osteoarthritis.
LEVEL OF EVIDENCE: Therapeutic Level I. See Instructions for Authors for a complete description of levels of evidence.
I found it difficult to read and the statistics confusing. I am not convinced the control groups were comparable. I was also unsure if patients getting sham acupuncture are "blind": is sham a valid placebo control if the real acupuncture produces a sensation and the sham does not. I doubt it is useful information for most general practitioners.
As the authors point out, "the lack of standardization of acupuncture treatment protocols also limits our findings". Moreover, the "control group" was highly heterogeneous including “sham acupuncture”, “standard, conservative pharmacological therapy" and no intervention”.