The objective of this study was to investigate mechanisms of action of intra-articular hyaluronic acid in osteoarthritis (OA) of the knee. Twelve patients with bilateral knee OA and synovial effusions entered a randomized, single-blind, blind observer study. Hyaluronic acid ("Hyalgan", Fidia SpA, Italy) or placebo were given by intra-articular injection weekly for 5 weeks. Assessments included clinical indices and imaging (magnetic resonance imaging (MRI) and 99m Tc bone scanning) before and after the course of injections. In addition, synovial fluid keratan sulfate (KS), chondroitin sulfate (CS) and C-propeptide of type II collagen (CPII) were measured. MRI and 99m Tc scanning showed no change in either treated or placebo knees over the 6-week study period. A fall in KS levels occurred in treated knees compared with placebo (Wilcoxon paired test, P = 0.1), although this did not reach significance perhaps due to small sample numbers). Ten out of 12 treated knees showed a fall in KS, compared with four out of 12 placebo knees. CS and CPII levels did not change significantly. Intra-articular injection of hyaluronic acid did not result in any improvement in the clinical indices compared to the placebo. In conclusion, assessment of cartilage markers may be of value when studying novel therapies in OA. MRI appearances remain remarkably stable over a 6-week period.