Keele Research Repository

BACKGROUND AND OBJECTIVES: The PRevention of knee Osteoarthritis in Overweight Females (PROOF) study (ISRCTN 42823086) described a trend for a decrease in the incidence of knee osteoarthritis (OA) by a tailored diet and exercise program (DEP) or by oral glucosamine sulfate in women at risk for the disease, using a composite clinical and/or radiological outcome. The aim of this updated post-hoc analysis was to re-assess the results according to more precise techniques and take advantage of the 2×2 factorial design. METHODS: A total of 407 overweight (BMI ≥ 27kg/m(2)) women of 50-60 years of age with no diagnosis of knee OA were randomized to: (1) no DEP + placebo (Control, N = 102), (2) DEP + placebo (DEP, N = 101), (3) glucosamine sulfate + no DEP (GS, N = 102), and (4) DEP + glucosamine sulfate (DEP + GS, N =102) and followed for 2.5 years, with standardized postero-anterior, semiflexed (MTP) view knee radiographs at baseline and end of the study. DEP consisted of a tailored low fat and/or low caloric diet and easy to implement physical activities. Glucosamine was given as oral crystalline glucosamine sulfate 1500mg once daily, double-blinded vs. placebo. Incident knee OA was defined as radiographic progression of ≥1mm minimum joint space narrowing (mJSN) in the medial tibiofemoral compartment, as previously assessed by the visual (manual) technique and by a new semi-automated method. Logistic regression analysis was used to calculate the odds ratio for the effect of the interventions. RESULTS: After 2.5 years, 11.8% of control subjects developed knee OA. This incidence was decreased with glucosamine sulfate, either alone or in combination with the DEP, but not by the DEP alone. Since there was no statistical interaction between treatments, the 2×2 factorial design allowed analysis of patients receiving glucosamine sulfate (N = 204) vs. those not receiving it (N = 203), similarly for those on the DEP (N = 203) or not (N = 204). Glucosamine sulfate significantly decreased the risk of developing knee OA: odds ratio (OR) = 0.41 (95% CI: 0.20-0.85, P = 0.02) by the manual JSN assessment method and OR = 0.42 (95% CI: 0.20-0.92, P = 0.03) by the semi-automated technique. Conversely, there was no decrease in risk with the DEP. CONCLUSIONS: Glucosamine sulfate decreased the risk of developing radiographic knee OA over 2.5 years in overweight, middle-aged women at risk, as determined by medial mJSN progression. Conversely a tailored diet and exercise program exerted no preventive effect, possibly because of the lower than expected effect on weight loss.