To evaluate the association between muscle mass and knee pain in relation to radiographic severity of knee osteoarthritis. We consulted nationwide health examination and survey records collected from 2010 to 2011 and extracted data regarding female patients aged > 50 years and diagnosed with knee osteoarthritis. Radiographic severity was assessed on plain radiographs using the Kellgren-Lawrence system, whereas appendicular skeletal mass was obtained from dual-energy X-ray absorptiometry data. We performed multivariate logistic regression to evaluate the association between knee pain and muscle mass index (appendicular skeletal muscle mass divided by body weight in percentile) in patient groups stratified by radiographic severity of knee osteoarthritis. Among 17,476 participants of the national survey, 2013 female knee osteoarthritis patients were identified and stratified by radiographic severity (grade ≤ 1, n = 1136; grade 2, n = 240; grade 3, n = 379; and grade 4, n = 258). For mild osteoarthritis (Kellgren-Lawrence grade 2), muscle mass index was significantly lower in patients with knee pain than in those without knee pain (24.9 ± 3.9 vs 26.5 ± 6.3%, P = 0.023), whereas no such difference was noted for severe osteoarthritis (Kellgren-Lawrence grade > 2). After adjusting for clinical variables by multivariate logistic regression, decreased muscle mass index remained significantly associated with knee pain in patients with mild osteoarthritis but not in those with severe osteoarthritis (regression coefficient 0.915, 95% confidence interval 0.854-0.981, P = 0.012). Lower muscle mass may be a risk factor for knee pain in patients with radiographically mild knee osteoarthritis but not in those with radiographically severe osteoarthritis.
Keywords: Knee osteoarthritis; Knee pain; Muscle mass; Radiographic severity; Sarcopenia.