Quadriceps neuromuscular function and self-reported functional ability in knee osteoarthritis

The purposes of this study were to determine 1) the relationships of self-reported function scores in patients with knee osteoarthritis (OA) to both maximal isometric torque and to isotonic power at a variety of loads, and 2) the degree to which muscle volume (MV) or voluntary activation (VA) are associated with torque and power measures in this population. Isometric maximal voluntary contraction (MVC) torque and isotonic power [performed at loads corresponding to 10, 20, 30, 40, and 50% MVC, and a minimal load ("Zero Load")] were measured in 40 participants with knee OA. Functional ability was measured with the Western Ontario and McMaster Osteoarthritis Index (WOMAC) function subscale. MV was determined with magnetic resonance imaging, and VA was measured with the interpolated twitch technique. In general, power measured at lower loads (Zero Load and 10-30% MVC, r(2) = 0.21-0.28, P < 0.05) predicted a greater proportion of the variance in function than MVC torque (r(2) = 0.18, P < 0.05), with power measured at Zero Load showing the strongest association (r(2) = 0. 28, P < 0.05). MV was the strongest predictor of MVC torque and power measures in multiple regression models (r(2) = 0.42-0.72). VA explained only 6% of the variance in MVC torque and was not significantly associated with power at any load (P > 0.05). Quadriceps MVC torque and power are associated with self-reported function in knee OA, but muscle power at lower loads is more predictive of function than MVC torque. The variance in MVC torque and power between participants is due predominantly to differences in MV and has little to do with deficits in VA.