Major Finding: Pooled data from two new systemic reviews showed improvements in pain (ES, 0.20) and physical function (ES, 0.23) after an average weight loss of 6.1 kg (approximately 13 pounds).
Data Source: An analysis of 65 systematic reviews, 266 randomized, controlled trials, and 21 economic evaluations.
Disclosures: Lead author Dr. Zhang had no financial conflicts to disclose. Several of the co-authors have received consulting fees, honoraria, and research support from multiple drug and device manufacturers. No conflict of interest was identified that would prevent any OARSI Ethics Committee members from participating in the review.
Some osteoarthritis treatments are less effective than previously thought, judging from findings from a review of research conducted since 2006.
The goal of the update of published evidence is to determine whether the current Osteoarthritis Research Society International's (OARSI) recommendations for the treatment of OA, published in 2008, need to be modified. After reviewing this update and collecting feedback, the OARSI Treatment Guidelines Committee will determine whether changes are needed in 2010.
Acetaminophen use and surgical lavage and debridement were among the therapies that may be falling out of favor to treat knee and hip OA, but evidence supporting weight reduction is on the upswing, said Dr. Weiya Zhang, a rheumatologist at the University of Nottingham (England), and colleagues.
The researchers identified 64 systematic reviews, 266 randomized controlled trials, and 21 economic evaluations related to hip and knee OA that were published between January 2006 and January 2009.
“Of the 51 modalities of treatment addressed in the OARSI recommendations, 35 have now been systematically reviewed with 16 new or updated systematic reviews in the last 3 years,” the researchers wrote (Osteoarthritis Cartilage 2010 Feb. [Epub 10.1016/j.joca.2010.01.013
The reviewers assessed the best available evidence for effect size (ES) with 95% confidence intervals for improving function and relieving pain and stiffness associated with OA.
The new evidence for nonpharmacological therapies included several studies supporting weight reduction. Pooled data from two new systemic reviews showed improvements in pain (ES, 0.20) and physical function (ES, 0.23) after an average weight loss of 6.1 kg (approximately 13 pounds). The ES for pain relief for hip and knee OA were not significantly changed for acupuncture, education, exercise, and self-management.
New research on electromagnetic therapy showed a relatively small improvement in function (ES, 0.33) and no significant effect on pain reduction (ES, 0.16), in contrast to data from a 2002 Cochrane review showing an ES of 0.77 that almost led to electromagnetic therapy's inclusion in the 2008 OARSI guidelines (it was not included).
The review also yielded changes in evidence for pharmacological treatments for OA, notably for acetaminophen.
A review of five new studies of acetaminophen for knee OA showed no significant reductions in effect size for pain relief (pooled ES, 0.14).
Other recent studies showed an increased risk of hospitalization due to perforation, peptic ulceration, and bleeding when acetaminophen doses of more than 3 grams per day were used to treat OA (hazard ratio 1.20). No data from recent studies identified significant changes in the risks and benefits of oral or topical nonsteroidal anti-inflammatory agents, diacerhein, or interarticular corticosteroid injections for treating OA.
For surgical treatments, pooled results showed no benefit for lavage, debridement, or a combination of the two for treating OA compared with placebo. Effect sizes for pain relief, improvement in function, and reduction in stiffness were 0.21, 0.11, and 0.05, respectively.
For alternative medicine treatments, recent studies showed a reduced effect size for pain relief of OA with treatments including glucosamine sulphate, chondroitin sulphate, intra-articular hyaluronic acid injections, and avocado soybean unsponifiables. Recent studies of these treatments also showed increased evidence of publication bias and heterogeneous outcomes.
The results of the review were limited by many factors, including the inability to make comparisons across treatments when meta-analyses and systematic reviews had different inclusion and exclusion criteria than randomized controlled trials.
The researchers noted that there is a need for “a continuously updated, comprehensive, and coherent database of well-characterized trials of all modalities of treatment of OA.”
But they emphasized that treatment guidelines must be based on the best evidence, not simply on updated cumulative evidence.