J Rehabil Res Dev. 2013 Jul;50(4):555-72.

Changes in passive ankle stiffness and its effects on gait function in people with chronic stroke.

A. Roya,b*, L.W. Forrestera,b,c, R.F. Mackoa,b,c,d, H.I. Krebsa,e

aDepartment of Neurology, University of Maryland School of Medicine, Baltimore, MD; bMaryland Exercise and Robotics Center of Excellence, Baltimore Department of Veterans Affairs Medical Center (VAMC), Baltimore, MD; cDepartment of Physical Therapy and Rehabilitation Science, University of Maryland School of Medicine, Baltimore, MD; dGeriatric Research Education and Clinical Center, Baltimore VAMC, Baltimore, MD; eDepartment of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, MA

*Corresponding author (email: ARoy@som.umaryland.edu)

 

Abstract

Mechanical impedance of the ankle is known to influence key aspects of ankle function. We recently investigated the effects of robot-assisted ankle training in chronic stroke survivors on the paretic ankle’s passive stiffness and over-ground (OG) gait function.[1] Eight participants with residual hemiparetic deficits engaged in a visually-evoked, visually-guided motor task (video game) while seated over 6 weeks that required them to dorsiflex (DF) or plantarflex (PF) their paretic ankle with an impedance-controlled ankle robot[2] (“anklebot”) assisting “as-needed”. Passive ankle stiffness (PAS) was measured in both the trained sagittal (plantar-dorsiflexion) and untrained frontal (inversion-eversion) planes. At 6 weeks, PAS decreased in both DF and PF directions, and reverted into the variability of age-matched controls in DF direction. Moreover, pre-post training changes in PAS in the PF direction correlated strongly to gains in paretic step (ρ = -0.88, P =.03) and paretic stride (ρ = -0.82, P =.05) lengths during independent floor walking. These findings suggest that training-induced changes in paretic ankle PAS strongly influence key temporal-distance (T-D) measures of over ground gait function.

PMID: 23934875

 

Supplemental Material

Layperson Summary. People who have had a stroke often have abnormally stiff joints in their affected limb. Recently, we reported the effects of a 6 week training program with a new ankle robot2 (“anklebot”) on the stiffness of the affected ankle in people with stroke greater than 6 months from onset.1 Eight individuals with lasting weakness from a stroke played a video game while seated for 1 hour (three times per week) by moving their affected ankle “up” or “down” with the robot assisting “as-needed.” The stiffness of the weak ankle was measured, both before and after participation in the study, by slowly stretching the ankle in different directions using the anklebot, while subjects were seated in a relaxed state. We found that at 6 weeks, the stiffness of the affected ankle decreased in the up and down directions and became similar to those of nondisabled people of similar age in the up direction. Importantly, decreased stiffness of the affected ankle led to improved quality of walking over ground. These findings suggest that measuring and monitoring ankle stiffness over the course of a therapy program can provide important insight into the process of ankle rehabilitation and assist in tracking recovery.

Background. On average, every 40 seconds someone in the United States suffers a stroke making it the number one cause of chronic disability. In the lower extremity, a common condition that occurs following a stroke is weakness in the dorsiflexor muscles that lift the foot during walking, commonly referred to as “drop foot”. The two major complications of drop foot—slapping of the foot after heel strike (“foot slap”) and dragging of the toe during swing (“toe drag”)—present a major challenge to efficient gait since clearing the ground during the swing phase and maintaining ankle stability during the stance phase are essential for safe and efficient walking. In particular, the ankle joint plays a fundamental role in bipedal locomotion several ways. First, it contributes to the maintenance of stable upright posture in the frontal and sagittal planes during quiet standing and upright gait. Second, it contributes to shock-absorption during locomotion by attenuating the impact force at floor contact. Third, the ankle muscles are the primary contributors to OG gait-–the soleus is the propulsion prime-mover, the gastrocnemius is the posture prime-mover, and the tibialis anterior is critical for toe-off. Common to these aspects of ankle function is that, they may all be characterized by its active and passive mechanical impedance, i.e., stiffness plus damping and any other dynamic factors. Humans adjust leg stiffness to accommodate surface changes and changes in gait speed primarily by modulating ankle stiffness. Adequate ankle impedance is also needed to control body momentum (forward and downward vector components of the body center of mass) during gait.

What is passive joint stiffness and its origins? In general, the mechanical impedance of a joint is a function of both passive (e.g., mechanical stiffness of ligaments, tendons, and connective tissue) and active (e.g., muscle activation, contraction mediated by stretch reflex) mechanisms. The passive component of impedance—passive stiffness—may be defined as the resistance to elongation or shortening of a muscle when it is quiescent, thus generating passive tension. The series elastic and parallel elastic elements of muscle (e.g., the tendon, structural proteins within the myofibril, connective tissue around the muscle fibers and fascicles) play a role in generating this passive tension. Structures such as perimysium that contain collagen within the muscle tendonunit, also contribute to passive stiffness, mostly at end-range (long sarcomere lengths). Withinthe physiological range of muscle length change, passive stiffnesshas been attributed to protein structures within the myofibril, such as titin.

How does stroke impact passive stiffness? Following a stroke, hypertonus and/or reflex hyper-excitability (spasticity) often disrupt the functional use of already-weakened muscles that include manifestation of increased resistance to passive movement for example, at the paretic ankle.[3] Surprisingly, very few studies have quantifiably measured and monitored changes in PAS over the course of some lower limb intervention; especially those that engage and attempt to increase the contribution of the paretic ankle into walking and balance tasks. Hence, a gap in knowledge remains as to whether and how changes in PAS affect locomotor function. In a recent undertaking,1 we sought to address these questions by evaluating the effects of visually-evoked, visually-guided, performance-based, progressive anklebot training on paretic PAS in chronic stroke and its influence on selected aspects of unassisted OG gait. We hypothesized that following 6 weeks of anklebot training the paretic PAS will change in the trained sagittal plane (DF-PF), but not in the untrained frontal plane (INV-EV). Moreover, we expected that, to be functionally meaningful, a robotic treatment protocol must emphasize a sequence and timing of sensorimotor stimuli similar to those naturally occurring during gait. Hence a secondary hypothesis was that changes in PAS resulting from training in the seated position would not carryover and confer improvements in gait function.

Visuomotor ankle training. The anklebot was used for training as well as PAS measurement.1,3,[4] During training sessions (6 weeks, 3xweekly = 18 sessions), subjects sat in a chair wearing the anklebot on their paretic leg with the knee flexed at 45° (Figure 1A). In this configuration, subjects played a video game with their paretic ankle by making alternate movements in DF and PF, which moved a robot-controlled cursor “up”/“down” on a display screen in order to pass through targets that approached across the display screen at different vertical levels. Task difficulty was adjusted to deficit severity in that, target locations were set at +/-80% and +/-40% of active range of motion (AROM) in each direction. In each session, subjects made 560 targeted ankle movements that comprised of 8 blocked trials with the first/last being “record-only” blocks consisting of 40 targets without any robotic assistance, while the 6 intermediate blocks each consisted of 80 targets. It is important to note that ankle movement was uncontrolled and unactuated in the INV-EV directions —in other words, the ankle was free to move in the untrained frontal plane. A performance-based progression algorithm was adopted that included increasing the target ROM by 10% in weeks 3-4 and frequency of target presentation by 0.06 Hz in weeks 5-6 in the assisted trials but only if tolerated and based on past performance; in this context, achievement of at least 64 out of 80 targets in at least one assisted block with the new settings. Otherwise, we used the prior settings. The target presentations for the “record-only” trials were held constant throughout training period.

Anindo Roy-1Figure 1: (A) Photograph of experimental setup showing a stroke subject training with the prototype MIT ankle robot (anklebot) in a seated position while playing an ankle targeting video game. Arrows denote motion of vertical gates that serve as targets for the anklebot-and-foot controlled cursor. The knee brace (partly seen) is mounted to a fixed plate that supports the anklebot and restricts knee and hip motions, effectively isolating the ankle to move freely in either PF–DF or INV–EV planes. The heel maintains contact with a platform so as to provide a pivot for the foot; (B) Measurement of passive ankle stiffness using the anklebot. (Top, left): commanded ramp-and-hold displacement perturbation (qcommand) of 15° in DF with constant velocity (v) of 5°/s and hold time (thold) of 1 s. Raw traces of (Top, right) ankle angle and (Bottom, right) torque resulting from each commanded positional perturbation taken from a single representative stroke subject, shown with initial (q0, t0) and final conditions (q, t) for a single trace. (Bottom, left): The torque-angle data is fitted with least-squares linear regression line in each direction, the slope of which is the estimate of PAS in that direction.

 

PAS measurement. To estimate PAS,1,3 the paretic ankle was stretched by the anklebot at a constant (5°/sec) velocity according to ramp up-hold-ramp down positional reference trajectory (Figure 1B). The rationale to stretch the ankle at 5°/sec (both ramp-up and ramp-down) was to avoid evoking stretch reflex. Subjects under relaxed condition experienced a series of perturbations during which the ankle was stretched to a commanded position, held at steady-state for 1 second, and returned back to neutral. The range of stretch amplitudes depended on the plane and direction of stretch: in the sagittal plane, displacements ranged from 20° in PF to the subject’s passive range of motion (PROM) in DF. In the frontal plane, stretch amplitudes ranged from 25° in INV to 20° in EV. Stretches were made in 5° increments. Ankle angles and torques were recorded by the anklebot during each stretch. Stiffness estimates were obtained in each direction of movement by fitting the pair-wise steady-state torque and angle data using least-squares linear regression (Figure 1B).

Main findings. This study1 revealed two important findings: #1 Interactive, robotic training in a seated position exercising the paretic ankle in DF and PF positively affected sagittal, but not frontal-plane PAS in chronic stroke survivors, with the DF PAS reverting into the PAS ranges of healthy and age-matched subjects (Figure 2A). Specifically, at baseline the sagittal-plane PAS was anisotropic,[5] with significantly greater stiffness in DF (53.4±8.2 Nm/rad) than in PF (13.2±0.85 Nm/rad, P = .001); but not in the frontal plane i.e., the stiffness did not significantly differ between EV (51.6±7.5 Nm/rad) and INV (44.6±3.6 Nm/rad, P = .72) directions. After 6 weeks of training, the PAS decreased in all four directions (DF, PF, INV and EV) but statistically significant changes were observed only in the sagittal plane PAS–i.e., DF and PF (Figure 2A). In one of those directions–i.e., DF, the PAS (24.6±4.1 Nm/rad) reverted into the ranges of young (DF: 12-48.2 Nm/rad) as well as age-matched healthy subjects (DF: 22.4-53 Nm/rad).3 In PF, however, the paretic PAS at termination (discharge) was outside the variability band of both young (10.7-25.5 Nm/rad) and age-matched healthy controls (12.2-13.8 Nm/rad).3 Similar to baseline, the PAS at discharge was anisotropic in the sagittal plane i.e., significantly higher in DF (24.6±4.1 Nm/rad) than in PF (10.0±0.47 Nm/rad, P = .03), but not in the frontal plane (Figure 2B)—PAS did not significantly differ between EV (40.8±8.6 Nm/rad) and INV (35.7±6.8 Nm/rad, P = .72) directions (Figure 2B); #2 Changes in sagittal-plane PAS in the PF direction, had a very strong and significant relationship with gains in selected T-D parameters of unassisted OG gait namely, stride and paretic step lengths (Figure 2C, D). Following training, subjects significantly increased their self-selected OG walking speed through a combination of longer stride lengths, faster cadence, and longer duration spent on paretic single support stance (SST) with concomitant decreases in double support stance.4 Correlation analyses[6] between changes in PAS and gait outcomes revealed that changes in passive PF stiffness was significantly correlated with changes in two key temporal-temporal parameters of gait function namely, (a) paretic step (STP) length (ρ = -0.88, P = .03), and (b) stride (STR) length (ρ = -0.82, P =.05), suggesting that improvements in STR and paretic STP lengths occurred in part due to changes in the PF PAS that in turn, contributed to improvements in OG gait. In both cases the correlation was negative; indicating that subjects whose ankles became more compliant in PF with training, took longer steps and strides on their paretic leg during OG walking.

Anindo Roy-2

Figure 2: (A) Changes in sagittal plane PAS i.e., DF (filled gray) and PF (unfilled) across time (PRE vs. POST). In both directions, the PAS decreased post-training (*P < .05). The PAS was anisotropic i.e., higher in one direction versus another, at both time points and this property was preserved across training with a more pronounced difference between the two directions at baseline (**P < .01); (B) Changes in frontal plane PAS i.e., EV (filled) and INV (unfilled) across time (PRE vs. POST). In both directions, the PAS decreased post-training but failed to achieve statistical significance. Unlike the sagittal plane PAS, the frontal plane PAS was not anisotropic at either time points; (C, D) Linear regression between changes in PAS in PF directions vs. changes in STR and paretic STP lengths.

 

Interpretation of findings. The 1st finding of this study was that robotic training of the paretic ankle joint decreased its PAS in the sagittal plane; in one of those directions (DF), the PAS reverted into the ranges of young and older non-disabled adults. Because of the long elapsed time since the stroke, we assume that the ankle condition was stable and that the obtained improvements were not due to natural recovery. Without direct evidence of muscle morphological data we can at best, conjecture that robot-assisted, repetitive massed practice may have induced intrinsic changes in ankle muscle physiology. There is indirect evidence to link PAS to the summed physiologic cross-sectional area (SPCA) and to the square of the mean moment arm of the antagonist group of muscles undergoing passive stretch.3 According to this hypothesis, training-induced reduction in the SPCA of plantar- flexors as a whole would explain the reduction in PAS in the DF direction. Reasoning along these lines may also explain unexpected changes in PAS in the untrained frontal plane given a synergistic functional role played by the plantar flexor muscles that are also evertors of the ankle. As an illustration, the peroneus brevis and peroneus longus muscles are the primary evertors of the ankle but are also (weak) plantar-flexors. A reduction in the plantar flexor SPCA could, therefore, potentially contribute to a reduction in the overall SPCA of the evertors taken as a muscle group. If true, this in turn would lead to a decrease in inversion PAS, a prediction consistent with the findings in this study. Similarly, the TA is the primary dorsiflexor but also acts to invert the ankle, so a reduction in the dorsiflexor SPCA could contribute to a reduction in the inverters as a group. If so, one would expect to see a reduction in the plantar flexion and eversion PAS that is consistent with our experimental findings. But at a mechanistic level, what could have lead to changes in the SPCA of a muscle group (e.g., plantar flexors)? We believe that there could be two plausible origins i.e., changes in either: a) the cellular structure; or b) the fiber-type distribution. Either (or both) of these mechanisms may have been induced due to the large volume of robotics-driven exercise of the paretic ankle. It is known that exercise or training promotes a chronic increase in the so-called “collagen turnover” process in which collagen is broken down or degraded by as much as 50%. Changes induced by collagen turnover have been shown to modify the biomechanical (e.g., viscoelastic) or structural (e.g., cross-sectional area) properties of soft tissue leading to altered resistance to loading. A decrease in the collagen level has been linked to a reduction in the ratio of collagen-to-muscle fiber tissue, thereby increasing muscle compliance. An equally plausible explanation is that repetitive exercise promoted changes in fiber-type distribution, i.e., an increase in the proportion of slow-to-fast twitch fibers; since the former type has a smaller diameter, it may have led to a decrease in the volume of the muscle undergoing passive stretching.

The 2nd finding was that changes in PAS were strongly correlated to key T-D OG gait parameters. The latter—i.e., expected benefits of seated ankle training extended to OG gait,4 is contrary to task-specificity of training since subjects were not trained on a walking task. Could changes in paretic PAS have potentially contributed to the increases in gait speed by means of increased STR and paretic STP lengths? Because PAS contributes to the total mechanical impedance of the joint, a conjecture is that changes in PAS may have enabled subjects to utilize their paretic ankle to orient their paretic foot more efficiently, thereby increasing their stride and paretic STP lengths. For example, dorsiflexor control of the foot is essential to clear the ground during the swing phase of gait and for ecological foot strike. A reduction in the DF PAS would contribute to a reduction in the total mechanical impedance of the ankle (in DF) that may lead to better dorsiflexor control of the foot for greater swing clearance, as well as controlled landing. Similarly, the plantar flexors play a critical role in stabilizing the forefoot rocker action during terminal stance and we know that plantar flexor muscle–tendons generate the largest power burst during trailing leg push-off in a single stride to enable forward propulsion (~35% total lower limb positive mechanical work and ~66% total ankle muscle-tendon positive work). Therefore, a reduction in the total mechanical impedance in the PF direction could in fact, lead to increased anterior-posterior (A/P) positive propulsion during paretic SST. Indeed, some of our participants increased their A/P positive propulsion by as much as 18% during paretic SST.4
  

[1] Roy et al. Changes in passive ankle stiffness and its effects on gait function in chronic stroke survivors. J Rehabilitation Research and Development; 50(4):555-572, 2013.

[2] Roy et al. Robot-aided neurorehabilitation: a novel robot for ankle rehabilitation. IEEE Transactions on Robotics; 25(3):569-582, 2009.

[3] Roy et al. Measurement of passive ankle stiffness in subjects with chronic hemiparesis using a novel ankle robot. J Neurophysiology; 105:2132-2149, 2011.

[4] Forrester et al. Ankle training with a robotic device improves hemiparetic gait after stroke. Neurorehabilitation Neural Repair; 25(4):369-377, 2011.

[5] Anisotropy is the property of being directionally dependent.

[6] Correlations assessed using Spearman’s correlation coefficient (ρ). Statistical significance was set at P<.05.

Multiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier SchönmannMultiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier Schönmann