Association of metabolic syndrome with knee and hand osteoarthritis: a community-based study of women

Objective: It is unclear whether the association between osteoarthritis (OA) and metabolic syndrome (MetS) varies with the site of the affected joint and the presence of pain. Our aim was to describe the association between MetS and radiographic OA (ROA) affecting the knee or the hand in the presence or absence of concurrent joint pain. Methods: Cross-sectional data of 952 women, aged 45–65 years from the Chingford study, a population-based longitudinal cohort of middle-aged women initiated in 1988–1989 in London (UK), was analysed. MetS was defined using the National Cholesterol Education Program Treatment Panel III criteria. Data was collected on components of MetS: waist circumference, triglycerides, high-density lipoprotein (HDL), blood pressure and blood glucose. The outcome was four knee and hand OA groups: painful ROA, ROA only, pain only and neither ROA nor pain (reference category). Multinomial logistic regression models adjusted for age and body mass index (BMI) were used to evaluate the effect of presence of MetS and its individual components on OA subgroups for knee and hand separately. Results: 952 eligible women, aged 45–65 years was analysed. A significant association was observed between the presence and the number of MetS with painful knee ROA when adjusted for age; however, this association disappeared when BMI was included in the model. In contrast, the presence and the number of MetS were associated with painful interphalangeal (IPJ) OA after adjusting for both age and BMI. Four out of the five MetS components, including triglycerides, HDL-c, hypertension and glucose, were associated with painful IPJ OA. Conclusions: MetS is associated with painful IPJ OA but not with knee OA once BMI is taking into consideration. Further attention to MetS and OA at different sites is needed to understand the metabolic phenotype in OA.