Responsiveness, effect size, and smallest detectable difference of Magnetic Resonance Imaging in knee osteoarthritis

SummaryObjectiveThe aim of this study was to determine the responsiveness, effect size (ES) and smallest detectable difference (SDD) of two Magnetic Resonance Imaging (MRI) measures for osteoarthritis (OA) of the knee: a whole-organ semiquantitative evaluation and cartilage volume.DesignThis analysis was performed on a dataset from a randomized, double-blind trial (Roche NI-15713) conducted in 1998 of a novel therapy in subjects with mild-moderate knee OA, with MRI at baseline and 6-month follow-up. The trial measurements included (1) cartilage volume measured using a proprietary software method; and (2) semiquantitative scoring of other parameters important for “whole organ” evaluation of OA knee joint pathology, using the Whole-Organ MRI Score (WORMS). The analysis initially examined the distributional characteristics of WORMS items, such as cartilage morphology. Standardized response mean (SRM), ES, and SDD between baseline and 6-month follow-up were then calculated in the whole group and the placebo group alone.ResultsIn general, the differences were small and this was reflected in the small ESs and SRMs. There was also a suggestion of a treatment effect with reduction in differences between baseline and follow-up in the treatment group.ConclusionOf the MRI semiquantitative measures, cartilage morphology, synovitis and osteophytes appeared to be responsive to change and the focus of repeat measures should highlight these articular features. In general, the ESs and SRMs were small.