Modulation of Spastic Anke Stiffness Dynamics with Voluntary Contraction in Spinal Cord Injury

A parallel-cascade system identification technique was used to examine the intrinsic and reflex contributions to overall ankle stiffness in normal (control) and spastic spinal cord injured (SCI) subjects as a function of voluntary contraction level. Intrinsic dynamics were modeled as a linear, 2nd-order system relating intrinsic torque to joint position. Reflex dynamics were described by a linear, 3rdorder system relating half-wave rectified velocity and reflextorque. Intrinsic stiffness was similar in magnitude in both groups and increased with voluntary contraction at similar rates. In contrast, reflex stiffness dynamics behaved differently in the two groups: (1) reflex stiffness gain was significantly greater in SCI than control subjects at all contraction levels, (2) the modulation of reflex gain with voluntary contraction was abnormal, and (3) the reflex frequency parameter was lower in SCIs and decreased with contraction level while it increased in controls. These differences were significant across a wide range of contraction levels with the gain difference being largest at low levels of contraction and the frequency difference being largest at high levels of contraction. INTRODUCTIOIN Spasticity is a motor disorder associated with lesions at different levels of the nervous system due to spinal cord or brain injury, multiple sclerosis, cerebral palsy, or stroke [1]. Common clinical symptoms in SCIs include hypertonia, autonomic hyperreflexia, flexor or adductor spasms, clonus, and weakness of voluntary contraction [2]. Nevertheless, hypertonia, an abnormal increase in muscle tone, is regarded as the defining feature of spasticity [3] that has both diagnostic and therapeutic significance [2]. In our previous studies, we used a parallel-cascaded system identification technique to quantify muscle tone in terms of ankle dynamic stiffness [4, 5]. Our findings demonstrated that reflex stiffness gain was significantly increased in SCIs [5]. We argued that this abnormality could be due, at least partially, to the inappropriate recruitment of larger motor units at low levels of contraction (0-10% MVC) where only small motor units are recruited in normal subjects. This argument was supported by our finding that reflex latency was shorter in SCIs than controls [6], consistent with faster conduction by larger motor axons. Based on these findings we hypothesized that inappropriate recruitment of motoneurons could result in abnormal modulation of reflex stiffness with voluntary contraction in spastic. We designed this study to test this hypothesis by examining dynamic stiffness and reflex function of the ankle joint as a function of voluntary activation level in SCIs and controls. EXPERIMENTAL PROTOCOL Eight control subjects (4 females, 4 males) and nine SCI subjects (3 females, 6 males) with different degrees of spasticity were examined. Subjects lay supine with their foot attached to the pedal of a stiff, position controlled, electrohydraulic actuator by a custom fitted fiber-glass boot. Joint position and torque were measured by transducers in the actuator. Electromyograms from the tibialis anterior and gastrocnemius muscles were recorded using bipolar surface electrodes. A series of pseudorandom binary sequences with an amplitude of 0.03 rad and a switching-interval of 150 ms were used to perturb the ankle at neutral position (90). PRBS trials were then recorded at tonic contractions ranging from 0 to 50% of the PF MVC, with the ankle at the neutral position. Trials were done at 3 Nm intervals up to –24 Nm. The different levels were examined in a different random order for each subject. Subjects had difficulty maintaining stable contractions at high torque levels. Consequently, only contractions less than 50% MVC and -24 Nm were included in the analysis. Plantarflexion is considered negative by convention. ANALYSIS METHODS Parallel-cascade Identification Technique Intrinsic and reflex contributions to the ankle stiffness dynamics were separated using a parallel-cascade identification method [7], shown in Figure 1. Intrinsic stiffness dynamics were estimated by determining the impulse response function ( PTQIRF ) between position and torque. Reflex dynamics were modeled as a differentiator, in series with a delay, a static nonlinear element (a half-wave rectifier) and then a dynamic linear element. Reflex stiffness dynamics were estimated by determining the impulse response function (VTQIRF ), between half-wave rectified velocity as the input and the reflex-torque as the output, using Hammerstein identification methods [7]. The non-linearity was found to be a half-wave rectifier, and thus the overall reflex gain was measured from the linear dynamic element. Report Documentation Page Report Date 25OCT2001 Report Type N/A Dates Covered (from... to) Title and Subtitle Modulation of Spastic Anke Stiffness Dynamics with Voluntary Contraction in Spinal Cord Injury Contract Number Grant Number Program Element Number Author(s) Project Number Task Number Work Unit Number Performing Organization Name(s) and Address(es) Dept of Biomedical Engineering Therapy, McGill University, Canada Performing Organization Report Number Sponsoring/Monitoring Agency Name(s) and Address(es) US Army Research, Development & Standardization Group (UK) PSC 802 Box 15 FPO AE 09499-1500 Sponsor/Monitor’s Acronym(s) Sponsor/Monitor’s Report Number(s) Distribution/Availability Statement Approved for public release, distribution unlimited Supplementary Notes Papers from the 23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, October 25-28, 2001, held in Istanbul, Turkey. See also ADM001351 for entire conference on cd-rom.