Commentary on the Cochrane Review of acupuncture for peripheral joint osteoarthritis.

BSTRACT OF THE OCHRANE REVIEW ackground: Peripheral joint osteoarthriis is a major cause of pain and functional imitation. Few treatments are safe and efective. bjectives: To assess the effects of acuuncture for treating peripheral joint oseoarthritis. earch Strategy: We searched the Cohrane Central Register of Controlled Trils (The Cochrane Library 2008, Issue 1), EDLINE, and EMBASE (both through ecember 2007), and scanned reference ists of articles. election Criteria: Randomized controlled rials (RCTs) comparing needle acupuncure with a sham, another active treatent, or a waiting list control group in eople with osteoarthritis of the knee, hip, r hand. ata Collection and Analysis: Two auhors independently assessed trial quality nd extracted data. We contacted study uthors for additional information. We alculated standardized mean differences sing the differences in improvements beween groups. ain Results: Sixteen trials involving 498 people were included. Twelve of the CTs included only people with OA of he knee, three only OA of the hip, and 1 a ix of people with OA of the hip and/or nee. In comparison with a sham control, cupuncture showed statistically signifihritis pain (standardized mean difference 0.28, 95% confidence interval 0.45 to 0.11; 0.9 point greater improvement han sham on 20 point scale; absolute% hange 4.59%; relative% change 10.32%; ine trials; 1835 participants) and funcion ( 0.28, 0.46 to 0.09; 2.7 point reater improvement on 68 point scale; abolute% change 3.97%; relative% change .63%); however, these pooled short-term enefits did not meet our predefined threshlds for clinical relevance (ie, 1.3 points for ain; 3.57 points for function) and there was ubstantial statistical heterogeneity. Addiionally, restriction to sham-controlled trils using shams judged most likely to adeuately blind participants to treatment ssignment (which were also the same hams judged most likely to have physioogical activity), reduced heterogeneity nd resulted in pooled short-term benefits f acupuncture that were smaller and nonignificant. In comparison with sham acuuncture at the six-month follow-up, acuuncture showed borderline statistically ignificant, clinically irrelevant improveents in osteoarthritis pain ( 0.10, 0.21 to 0.01; 0.4 point greater improveent than sham on 20 point scale; absoute% change 1.81%; relative% change .06%; four trials;1399 participants) and unction ( 0.11, 0.22 to 0.00; 1.2 point reater improvement than sham on 68 oint scale; absolute% change 1.79%; reltive% change 3.89%). In a secondary uncture was associated with statistically ignificant, clinically relevant short-term mprovements in osteoarthritis pain ( 0.96, 1.19 to 0.72; 14.5 point greater improveent than sham on 100 point scale; absoute% change 14.5%; relative% change 9.14%; four trials; 884 participants) and unction ( 0.89, 1.18 to 0.60; 13.0 oint greater improvement than sham on 00 point scale; absolute% change 13.0%; elative% change 25.21%). In the head-on omparisons of acupuncture with the ‘suervised osteoarthritis education’ and the physician consultation’ control groups, cupuncture was associated with clinically elevant shortand long-term improveents in pain and function. In the head n comparisons of acupuncture with home exercises/advice leaflet’ and ‘superised exercise,’ acupuncture was associted with similar treatment effects as the ontrols. Acupuncture as an adjuvant to n exercise based physiotherapy program id not result in any greater improvements han the exercise program alone. Informaion on safety was reported in only eight rials and even in these trials there was limted reporting and heterogeneous methds. uthors’ Conclusions: Sham-controlled rials show statistically significant benefits; owever, these benefits are small, do not eet our pre-defined thresholds for clinial relevance, and are probably due at least artially to placebo effects from incomlete blinding. Waiting list-controlled tri-