Adiposity and clinical outcomes of knee osteoarthritis: Sex-specific analysis from the FEAST trial

Greater body weight and body mass index (BMI) are associated with worse clinical outcomes in knee osteoarthritis, but these anthropometric measures do not assess fat mass. We evaluated sex-specific relationships between total and visceral adiposity and clinical outcomes in knee osteoarthritis. Baseline data from all 144 participants with knee osteoarthritis (age 65±8 years, BMI 30.3±6.2 kg/m, 64% female) enrolled in the FEAST randomised controlled trial were analysed. Adiposity measures (total fat mass, percent fat mass, visceral fat mass, percent visceral fat mass) were assessed using dual-energy X-ray absorptiometry. Clinical outcomes included knee symptoms (five subscales of Knee Osteoarthritis Outcome Score (KOOS)), health-related quality-of-life (QoL) (EQ-5D) and functional performance (40 m walk). Linear regression models with a sex*adiposity interaction term evaluated associations between adiposity measures and clinical outcomes. Significant sex*adiposity interactions were observed for KOOS-QoL and 40 m walk, with higher percent body fat and percent visceral fat, respectively, associated with worse outcomes in females but not males (e.g. for KOOS-QoL 1-unit increase in percent body fat in females -1.2 (95%CI -1.8 to -0.6); males 0.1 (95%CI -0.9-1.1). While no significant interaction effect was observed for other comparisons, in females greater total and visceral fat mass were largely associated with worse clinical outcomes, which was generally not the case in males. Adiposity was associated with poorer outcomes, with significant sex-specific effects observed for knee-related QoL and functional performance for females but not males. While most outcomes did not demonstrate significant sex-interactions, effect estimates were typically larger and more consistent in females.