To the Editor:

Nonsteroidal antiinflammatory drugs (NSAIDs) are widely used for pain relief in subjects with osteoarthritis (OA). What effects these agents have on the progression of OA, remains controversial (1). Recently, Reijman et al (2) reported that in an observational study, subjects who were receiving diclofenac for >180 days had a 2.4-fold increased risk of progression of hip radiographic OA and a 3.2-fold increased risk of knee radiographic OA, compared with subjects who were considered short-term users (diclofenac for 1–30 days). These results are consistent with previous observations in animal studies (1) and indicate that diclofenac may induce accelerated progression of hip and knee OA.

While the associations between NSAID use and the progression of radiographic OA were adjusted for age, sex, body mass index, baseline radiographic OA, followup time, and defined daily dosage, they were not adjusted for baseline hip or knee pain, which is a very important confounder. Instead, the authors analyzed the associations only in subjects with prevalent hip and knee pain at baseline, which were also not adjusted for the severity of pain (no information about pain severity was obtained in this study). This could lead to false-positive results because baseline joint pain is a predictor of joint space narrowing (3) and cartilage loss (4), and it is possible that joint pain is more severe and more prevalent in subjects with long-time NSAID use; thus, the progression of radiographic OA is increased in this group. The findings presented by Reijman and colleagues would be more convincing if those authors reported the results in subjects without baseline joint pain, rather than in all subjects or only in subjects with baseline joint pain.

It is of great importance and interest to determine the effects of NSAIDs on the progression of OA; however, radiographic assessment of OA progression is insensitive and provides only a broad-brush view of joint pathology. In contrast, magnetic resonance imaging (MRI) can be used to visualize joint structure directly and is a valid, accurate, and reproducible tool for measuring articular cartilage lesions and loss (5). MRI has great potential to reveal the associations between NSAID use and progression of OA. In addition, randomized controlled trials, rather than observational studies, should be used to determine these effects of NSAIDs.


  1. Top of page

Changhai Ding MD*, * University of Tasmania, Hobart, Tasmania, Australia.