T.E. Towheed M.D., M.Sc., F.R.C.P.C., F.A.C.P., , Associate Professor of Medicine and of Epidemiology
Department of Medicine, Queen's University, Kingston, Ontario, Canada
Objective To systematically review published randomized controlled trials (RCTs) evaluating pharmacological and non-pharmacological therapies in patients with hand osteoarthritis (OA), with an emphasis on trial methodology.
Methods RCTs published between 1966 and August 2004 were identified by searching several electronic data sources as well as by searching reference lists. Details of study demographics, methodology, quality and outcomes were analyzed. A meta-analysis was planned, if feasible.
Results Thirty-one RCTs evaluating various pharmacological and non-pharmacological therapies in hand OA were analyzed in this systematic review. When compared with hip and knee OA, there are surprisingly few published RCTs in hand OA. Generally, these RCTs are of low quality. RCTs are weakened by a lack of consistent case definition and by a lack of standardized outcome assessments. The methods used for randomization, blinding, and allocation concealment were rarely described. The number and location of symptomatic hand joints per treatment group at baseline was usually not stated. The number and location of evaluated hand joints at the end of the study was also usually not stated. A meta-analysis could not be performed since most of the treatments studied did not have more than one identical comparison to allow pooling of the data.
Conclusion It is apparent that hand OA is a more complex area in which to study the efficacy of therapies when compared to hip and knee OA. Consensus guidelines are urgently needed to help improve the design and conduct of RCTs in hand OA. Additional RCTs of high quality that follow consensus recommendations are needed to evaluate the wide range of possible therapeutic options available for patients with hand OA.
Osteoarthritis and Cartilage Volume 13, Issue 6 , June 2005, Pages 455-462.