OBJECTIVE: To analyze the associations between body composition, notably low lean mass, and clinical symptoms [pain, physical function, quality of life (QoL)] in patients with symptomatic hip and/or knee OA. METHODS: Cross-sectional study using data from the 3-year follow-up visit of the Knee and Hip OsteoArthritis Long-term assessment (KHOALA) cohort. Skeletal muscle and fat mass were measured by dual X-ray absorptiometry (DXA). Fat mass index (FMI) was defined as total fat mass/height2. Appendicular lean mass was adjusted on body mass index (ALM/BMI), and low lean mass was defined according to the definition of FNIH Sarcopenia Project recommendations. Pain and function were measured by the WOMAC index and QoL by the SF-36. RESULTS: In total, 358 patients underwent DXA (67% women, mean [SD] age 63.4 [8.4] years, mean BMI 29.5 [5.6]kg/m2). The visual analog scale (0-100) pain score was 38.0 [24.7] and 25.4% had hip and 74.6% knee OA. Low lean mass and ALM/BMI were associated with impaired QoL and WOMAC scores on bivariate analysis (all p ≤ 0.001) but not on multivariate analysis after adjustment for FMI. For patients with normal BMI, mean [SD] WOMAC scores were higher (greater impairment) with low lean mass than normal body composition (WOMAC function 33.4 [23.3] and 24.0 [17.4], p = 0.02), and mean SF-36 physical component score was lower (greater impairment) 40.3 [10.2] and (44.3 [8.4], p = 0.04). Among patients with obesity, low lean mass had no additional effect. CONCLUSION: For patients with OA and normal BMI, QoL and function were more impaired for those with than without low lean mass. Conserving muscle mass in people with OA could have functional and antalgic benefits especially for those with normal BMI. Copyright © 2017 Elsevier Inc. All rights reserved.