To explore the potential contributions of joint innervation towards joint stability, we propose to study the role of joint afferents in promoting medio-lateral stability of the human knee joint and seek to characterize the input-output properties of joint afferents-based reflexes in a well-controlled environment. The choice of knee varus/valgus (v/v) as a model is advantageous because there is less likelihood that muscle proprioceptors will contribute to reflex activation in these directions, allowing more straight-forward examination of input/output properties of ligament/capsular mechanoreceptor effects. Based on our preliminary studies, mechanical perturbations that stretch joint ligaments produced reflex response in muscles traversing the knee joint, and that these reflexes were organized so as to compensate selectively for varus-valgus loading. Accordingly, we hypothesize that joint afferents mediated muscle activation can be elicited by the application of mechanical v/v loading to the human knee, and that the reflex intensity will be an increasing function of the angular perturbation size. We also propose that medial muscles will show a larger reflex response to lateral perturbation, and lateral muscles will show a larger response to medial perturbations. This reflex response will significantly increase knee joint stiffness in the v/v direction. Finally, we hypothesize that spatial distribution of these reflex-based activation patterns maximize resistance to the valgus (or varus) loading at the knee. Quantitative measures of periarticular mediated reflex activation of knee muscles will be determined in 72 subjects. A ramp valgus positional perturbation will be applied at the subject knee via a servomotor. [Load bearing conditions will be simulated by the application of an axial force to the sole of the foot during positional perturbation using a linear servomotor system. The axial force will be used to increase joint compression, and will be represented as a percent of the subject s body weight.] Torque and EMG activity will be recorded and analyzed in major knee muscles. To examine the mechanics of reflex action quantitatively, a three-dimensional subject-specific patello-femoral and tibio-femoral joint model will be constructed based on complete knee MRIs of a subset of 10 subjects. In the model, each muscle activation will be represented as a distribution function with a mean and standard deviation bounded by the reflex activation patterns obtained from the same subject in Aim 1. We will use Monte Carlo simulations with this model to obtain estimates of the likely distribution of random quadriceps activation patterns that can generate varus moment in response to an applied valgus load. These model-based varus moments are then compared to the varus moment computed by the model when experimental muscle activations are used (data from Aim 1). A quasistatic and dynamic reflex stiffness will be quantified using ramp and pseudo-random binary v/v positional perturbation applied to the knee in 9 subjects, respectively. Joint stiffness and damping coefficients will be estimated using delayed linear models. Identification of the input/output properties of the mechanoreceptor-based knee joint reflexes obtained under static neurological and mechanical states will serve as solid basis for future studies, which will investigate joint afferents contributions during a functional task. We believe that stretch receptors in periarticular tissues of the knee joint play a major role in promoting joint medial-lateral stability, which will potentially have a major impact on how training may improve overall joint stability.
