Background/Purpose: Obesity and metabolic syndrome (MetS) are the strongest modifiable risk factors of knee OA, and frequently affect the same individual. The reported associations in hip OA are less consistent. Weight reduction alone has not been shown to prevent the progression of hip OA, which led us to consider the metabolic effects of obesity. This study tests the interaction between obesity and MetS on hip OA in older men and women, to determine whether odds of hip OA were higher among individuals with both MetS and obesity as compared to those with only one or neither of these conditions.

Methods: Cross-sectional analysis was performed on two groups: 4871 women from the Study of Osteoporotic Fractures (SOF) and 3567 men from the Osteoporotic Fractures in Men Study (MrOS) cohorts. Participants were included if they had available hip radiographs, documented hip exam, and measurements of body mass index (BMI), systolic blood pressure, triglycerides, high density lipoprotein, and fasting glucose. Clinical hip OA was defined as a modified Croft Score ³2 or THR by radiographic review, and either hip pain or limited hip range of motion. The odds ratio for hip OA in obese (BMI³30) as compared to non-obese and in MetS as compared to non-MetS participants was estimated using multivariable logistic regression, adjusting for covariates (Table). An interaction term was then included to determine whether MetS modified associations between obesity and hip OA. In SOF, serum measurements required to evaluate MetS were only available for 401 participants. Compared to the 4871 SOF participants, these 401 women were 0.6 years older, less educated, had 2.6% more DM; and were similar in prevalent hip OA and anthropometry.

Results: The prevalence of obesity was 21.4% in men and 19.3% in women. Fifty percent of women and 60.2% of men met the definition for MetS. In men, there was a 70% higher odds of hip OA in obese as compared to non-obese individuals in fully adjusted analysis (95% CI: 1.03-2.84). Among women, no significant association between obesity and hip OA was found in fully adjusted analysis (Table 1). There was no significant association between MetS and hip OA (Table 1) or interactive effect (data not shown) between MetS and obesity on hip OA in either women(p=0.996) or men (p=0.394).

Conclusion: Obesity, but not MetS, conferred a higher odds of hip OA in a population of community-dwelling older men, suggesting that mechanical load has a more significant role in hip OA pathogenesis than the metabolic effects of increased adiposity. Neither obesity nor MetS were associated with hip OA in the women studied. Given the small sample size, future studies are needed to evaluate MetS and hip OA in women.

Table 1. Odds ratios (and 95% Confidence Intervals) for associations of obesity (BMI³30 kg/m²) and metabolic syndrome (2 of 4 NCEP-ATPIII criteria, excluding waist circumference) with clinical hip osteoarthritis in SOF and MrOS.