|| Checking for direct PDF access through Ovid
NSAIDs and acetaminophen are the most commonly used treatments for knee osteoarthritis; however, these agents have limited effectiveness, and NSAIDS in particular can have serious adverse effects. Safer and more effective therapies are needed.To evaluate acupuncture as an alternative therapy for knee osteoarthritis.This was a meta-analysis of trials examining acupuncture as a treatment for knee osteoarthritis. The MEDLINE, EMBASE and Cochrane databases were searched for randomized controlled trials that compared patients treated by needle acupuncture with sham treatment, usual care, or waiting list control groups (waiting list patients received no care while waiting for acupuncture). The analysis only included trials in which traditional meridian points were used; needles could also be inserted into tender points, and could be electrically stimulated. Trials of dry needling, trigger point therapy, those that compared only two types of active acupuncture, and those with <6 weeks' treatment duration were excluded. Standardized mean differences were used as the principal measure of effect size.The main outcome measure was a clinically relevant improvement in osteoarthritis symptoms, defined as standardized mean differences of 0.39 for pain and 0.37 for function.The systematic review included 11 trials, two of which were excluded from the meta-analysis because of insufficient outcome data. The total number of patients was 2,821. Of the seven trials that used a sham control, most used either penetrating or nonpenetrating needles placed away from acupuncture points. One trial used needles placed superficially at acupuncture points with mock electrical stimulation, and one used patch electrodes attached at the knee with mock electrical stimulation. Compared with sham controls, acupuncture produced short-term improvements in pain (standardized mean difference −0.35 [95% CI −0.55 to −0.15]) and function (standardized mean difference −0.35 [95% CI −0.56 to −0.14]), but these improvements were not clinically relevant. Six-month improvements (−0.13 [95% CI −0.24 to −0.01] and −0.14 [95% CI −0.26 to −0.03] for pain and function, respectively) were also not clinically relevant. Compared with usual care, acupuncture did produce clinically relevant short-term improvements in both pain and function (−0.62 [95% CI −0.75 to −0.49] and −0.56 [95% CI −0.69 to −0.43] respectively); these improvements remained clinically relevant for 6 months after baseline (−0.52 [95% CI −0.66 to −0.39] and −0.45 [95% CI −0.59 to −0.32], respectively). Patients receiving acupuncture also had clinically relevant improvements compared with those on a waiting list (−0.96 [95% CI −1.21 to −0.70] for pain and −0.93 [95% CI −1.16 to −0.43] for function).Acupuncture has no clinical benefit when compared with sham-control trials, but is clinically beneficial when compared with usual care or waiting list control groups. These improvements might be due to placebo or expectation effects.