ASSOCIATION BETWEEN METABOLIC FACTORS AND KNEE OSTEOARTHRITIS IN WOMEN - THE CHINGFORD STUDY

Objective. Several studies have shown an association with obesity and knee osteoarthritis (OA), however the role of other metabolic factors is unclear, with conflicting data in the literature. We studied the as sociation between metabolic risk factors and knee OA in women in the g eneral population. Methods. One thousand three women aged 45-64 from t he Chingford population study completed risk factor questionnaires. Cu rrent blood pressure and ever hypertension were noted and fasting bloo d glucose, serum cholesterol, triglycerides, high density lipoprotein (HDL), and uric acid levels were measured. AP weight bearing radiograp hs were available in 979 women and scored using the Kellgren and Lawre nce system. Grade 2+ (definite osteophytes) was used as a definition o f knee OA. Odds ratios (OR) and 95 % confidence intervals were calcula ted for risk of knee OA in highest tertile versus lowest for each risk factor. All OR were adjusted for age and body mass index as potential confounders for OA. Results. Radiological evidence of knee OA was fou nd in 118 women (12 %). For knee OA in either knee the variables signi ficantly associated were raised blood glucose OR = 1.95 (1.08-3.59), a nd moderately raised serum cholesterol OR = 2.06 (1.06-3.98). For symp tomatic women (n = 58) raised blood glucose OR = 2.77 (1.13-6.76), and use of diuretics OR = 2.27 (1.11-4.65) were significantly associated. For bilateral knee disease (n = 55) significant associations were fou nd for ever hypertension OR = 3.02 (1.51-6.06), subjects taking diuret ics OR = 2.84 (1.37-5.89), and both high and moderately raised serum c holesterol OR = 3.91 (1.07-14.25), and OR = 3.63 (1.00-13.88), respect ively. In all categories of knee OA serum uric acid was nonsignificant ly increased. No association was found with raised triglyceride or HDL levels or with current systolic blood pressure. Further adjustment fo r physical activity and social class did not affect the results. Concl usion. These data suggest that hypertension, hypercholesterolemia, and blood glucose are associated with both unilateral and bilateral knee OA independent of obesity, and support the concept that OA has an impo rtant systemic and metabolic component in its etiology.