Content area
Abstract
It could be argued that the importance of a good biological explanation is reduced by the demonstration of therapeutic efficacy in a robust clinical trial. Certainly, the study by Jean-Yves Reginster and colleagues shows many hallmarks of a well-conducted trial, including rigorous descriptions of randomisation, blinding, and allocation concealment, as well as use of state-of-the-art approaches to measurement of outcome. Even so, some elements of the methods merit attention. As recommended by expert consensus, the study used change in width of medial tibiofemoral joint space as its primary measure of the biological effect.7 Precise measurement of this variable is contingent on standardised radiographic techniques, and could be biased when the presence of pain impairs the ability of participants to extend the knee fully! How much this theoretical limitation translates into real bias is unclear, but there are also fundamental questions about the relevance of width of radiographic joint space as a single measure of biological severity in knee OA. The use of this measure derives from the conceptualisation of OA as a disorder of progressive hyaline-cartilage loss, and the belief that hyaline-cartilage thickness correlates with width of radiographic joint space. In fact, recent magneticresonance-imaging studies have highlighted the panarticular nature of knee OA and have shown that mild to moderate joint-space loss reflects meniscal extrusion rather than hyaline-cartilage erosion.9 Even more troublesome is the common clinical observation that radiographic severity is one of the least important of the predictors of outcomes that matter, such as pain and physical function.10