Sensorimotor changes in knee osteoarthritis: from muscle spindle function to brain organisation and activity

Abstract

Knee osteoarthritis (OA) is a major cause of disability and it is a growing problem due to the aging population and rising obesity rates. Better understanding of the disease process is required to inform the most efficient and effective methods of treatment for knee OA.

Knee OA is associated with changes in motor and sensory function including impaired proprioception and motor output. The underlying causes of these impairments in knee OA remain unclear. Little previous research has investigated whether the proprioceptive impairments associated with knee OA are localised to the knee joint or are generalised across multiple joints in both the upper and lower limb joints. Muscle spindle function and the central processing of muscle spindle related afferent input in the brain are integral to optimal proprioception, however muscle spindle function in people with knee OA has not previously been investigated. Many of the documented motor and sensory function changes associated with knee OA alter input from the periphery. These changes along with the functional or structural reorganisation of the brain associated with other conditions that affect control of movement indicate likely reorganisation of at least the motor cortex and possibly other brain regions. No previous study has investigated reorganisation of the motor cortex or changes in activation across the whole brain associated with knee OA.

To investigate whether joint repositioning proprioceptive deficits are localised to the diseased joint or generalised across other joints in people with knee OA, 30 individuals with right knee OA and 30 healthy asymptomatic controls performed active joint repositioning tests of the knee, ankle and elbow. Participants with knee OA had a larger relative error for joint repositioning of the knee than the controls. Relative error did not differ between groups for the ankle or elbow. These results are consistent with a mechanism for proprioceptive change that is localised to the knee joint.

Given the important contribution muscle spindles make to proprioception differences, the function of quadriceps, triceps surae and tibialis anterior muscle spindles between people with and without knee OA were investigated. Thirty individuals with knee OA and 30 healthy asymptomatic controls stood comfortably and blindfolded on a force plate, with mechanical vibration applied over the quadriceps, triceps surae or tibialis anterior muscles. Anterior-posterior displacement of centre of pressure was analysed. Although there were no differences between groups for trials with vibration applied to the quadriceps or tibialis anterior, participants with knee OA were initially perturbed more by triceps surae vibration and accommodated less to repeated exposure than controls. This indicates that people with knee OA have less potential to detect or compensate for disturbed input to triceps surae, possibly due to an inability to compensate using muscles spindles in the quadriceps muscle.

Differences in the organisation of the motor cortex between people with and without knee OA and possible associations between cortical organisation and accuracy of performance of a motor task were investigated. Functional magnetic resonance imaging (fMRI) data were collected while 11 participants with right knee OA and seven asymptomatic controls performed three force-matching motor tasks involving: 1) quadriceps, 2) tibialis anterior, and 3) finger/thumb flexor muscles. fMRI data were used to map the loci of peak activation in the motor cortex during the three tasks and to assess whether there were differences in the organisation of the motor cortex between the groups for the three motor tasks. Task accuracy was also quantified. A more anterior representation of the knee, and an opposite relative position of the knee and ankle representations in the motor cortex were found in people with knee OA. Poorer performance of the knee task was associated with more anterior placement of motor cortex loci in both groups.

To investigate differences in activation of brain regions involved in sensorimotor processing between people with and without knee OA, brain activation during force matching tasks of the knee, ankle, or hand was assessed. fMRI data were collected with the participants, protocol and methods described above for investigations of motor cortex organisation. Areas of activation that differed between groups for each of the three motor tasks across the whole brain were identified. Task accuracy was also quantified. The combination of findings of changes in brain activation largely localised to the knee and of similar levels of performance accuracy across the knee, ankle and hand in the OA group indicate that a combination of factors, including those specific to sensorimotor control, may underlie changes in brain activation during knee, and to a lesser extent, ankle movements in people with knee OA. Brain activation was either not different or minimally different during the hand and ankle tasks, respectively. This indicates differences in sensorimotor processing are largely specific to the knee and not a generalised phenomenon.

Overall this thesis provides evidence that in knee OA: 1) proprioceptive impairments are localised to the knee joint and not generalised across the ankle and elbow joints, 2) there are changes to muscle spindle function associated with knee OA, 3) organisation of the motor cortex differs between people with and without knee OA and differing organisation of the motor cortex is related to motor task performance, and 4) there are differences in brain activation between people with and without knee OA across a broad sensorimotor network associated with knee movement and a small number of areas of differing brain activation associated with ankle movement.