Viscosupplementation consists of injecting exagenous hyaluronan (HA) into the synovial joints to restore the normal rheological environment which deteriorates severely in osteoarthritic (OA) joints. Efficacy might be related to the rheological properties and molecular weight (MW) of the hyaluronan preparations. This prospective, controlled, double-blind, randomised clinical trial was aimed at comparing the elastoviscous properties of a high molecular weight viscosupplement, hylan G-F 20, with that of a lower molecular weight hyaluronan product in order to determine the relationship of elastoviscosity to efficacy, alongside placebo, in the treatment of patients with knee OA. The results were analysed as a "completers" analysis with 59 patients. Primary outcome measures included the Western Ontario and Mc Master Universities' Osteoarthritis Index (WOMAC) for pain, stiffness and function scores, and patient and physician global assessments (0-100 scale). For patient (PGA) and physician global assessments (PhGA), the 0-100 scale was used, with 100 being the worst. Follow-up assessments were made at intervals of 1, 3 and 6 months after the first injection. Local adverse events, such as transient pain at the injection site or warm knee lasting for one night, were recorded in two patients (3%). In all groups, the WOMAC pain score exhibited a significant difference from the baseline value; neither treatment group was significantly different from the placebo group, but total pain score was significantly better than baseline for both of the HA groups at the end of 6 months (p < 0.05). Improvement in WOMAC physical function score favoured both sodium hyaluronate and hylan G-F 20 after the first month, and remained significant until the end of 6 months (p < 0.01). In the placebo group, the physical function scores became worse after the end of the 1st month; the scores at the end of 6 months were no different from those at the beginning. The WOMAC stiffness scores of both of the hyaluronic acid groups improved with the first injection, and remained significantly better than the placebo group until the end of the survey (p < 0.001). All groups expressed improvement with PGA scores after the first injection. At the end of 6 months all three groups were similar, but the treatment groups were significantly better than the placebo group (p < 0.05), and all were significantly better than at the beginning (p < 0.05). The PhGA scores were similar in all groups until after the third injection. The second group was slightly better in the controls at 1 and 3 months, but all the groups were similar at the end of 6 months. Although the placebo group seemed worse, it was not statistically significant. Compared with lower molecular weight HA, the higher molecular weight HA might be more efficacious in treating knee OA, but heterogeneity of previous studies limited definitive conclusions. Patients treated by injection of either of two hyaluronan preparations showed clinical improvement for pain, though no different from the placebo group; WOMAC stiffness scores were better than placebo in the HA groups, whereas PGA scores showed improvement in all groups but HA groups were better than placebo. PhGA scores were worse in the placebo group, but not to a statistically-significant extent. The HA groups did not differ in terms of clinical efficacy.