- Email: [email protected]
Perception of symmetry and asymmetry in individuals with anterior cruciate ligament reconstruction
Clinical Biomechanics 40 (2016) 52–57
Contents lists available at ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Perception of symmetry and asymmetry in individuals with anterior cruciate ligament reconstruction Jaimie A. Roper ⁎, Matthew J. Terza, Chris J. Hass University of Florida, College of Health and Human Performance, Department of Applied Physiology and Kinesiology, USA
a r t i c l e
i n f o
Article history: Received 4 January 2016 Accepted 29 October 2016 Keywords: ACL Reconstruction Symmetry Gait Walking Sensory
a b s t r a c t Background: Changes in the quantity, quality and integration of sensory information are thought to persist long after anterior cruciate ligament reconstruction and completion of physical therapy. Our purpose was to investigate the ability of individuals with anterior cruciate ligament reconstruction to perceive imposed asymmetry and symmetry while walking. Methods: Twenty participants with anterior cruciate ligament reconstruction and 20 controls walked on a splitbelt treadmill while we assessed the ability to detect symmetry and asymmetry at fast and slow speeds. Detection scores and spatiotemporal data during asymmetric and symmetric tasks in which the belts were coupled or decoupled over time were recorded. Findings: The ability to detect symmetry and asymmetry was not altered in individuals with anterior cruciate ligament reconstruction compared to healthy young adults. The belt-speed ratio at detection also correlated to asymmetry for step length, stride length, double support time, and stance time. However, the anterior cruciate ligament reconstruction group appeared to utilize unique information to determine detection. When asked to detect symmetry at a fast speed, no asymmetry scores significantly correlated with belt-speed ratio in the anterior cruciate ligament reconstruction group. Conversely, asymmetry in stride length, step length, and stance time all significantly correlated with belt-speed ratio at detection in the control group. Interpretation: Specific sensory cues arising from the speed of the leg may also augment perception of symmetry. This strategy may be necessary in order to successfully execute the motor task, and could develop due to altered sensory information from the reconstructed knee at faster walking speeds. Published by Elsevier Ltd.
1. Introduction Locomotor adaptation paradigms using split-belt treadmills (SBT) have provided important insight into the neural control of movement and the ability of an individual to alter motor behavior of gait. Adaptation is a short-term learning process, which provides information with respect to how much the body can adapt to a new behavior and is considered a surrogate for the amount of flexibility within a behavioral pattern (Bastian, 2008; Reisman et al., 2010). During SBT walking the two belts can be decoupled such that one leg walks faster than the other, resulting in both reactive (fast) and predictive (slow) adjustments adaptation to an asymmetric walking pattern (Reisman et al., 2010). Recently, we have documented that individuals with anterior cruciate ligament reconstruction (ACLR) following anterior cruciate ligament (ACL) injury demonstrate altered locomotor adaptation to decoupled belt speeds (Roper et al., 2016). Specifically, we observed impairments in both slow-adapting and fast-adapting derived gait parameters, suggesting that fundamental features of motor control remain altered ⁎ Corresponding author at: 100 Florida Gym, PO Box 118205, Gainesville, FL, USA 32611-8205. E-mail address: jaimier@ufl.edu (J.A. Roper).
http://dx.doi.org/10.1016/j.clinbiomech.2016.10.017 0268-0033/Published by Elsevier Ltd.
despite surgical reconstruction and completion of physical therapy. Because individuals with ACLR demonstrate alterations in locomotor adaptation and also in proprioception (Littmann et al., 2012; Reider et al., 2003), deficits in the ability to accurately perceive asymmetric and symmetric belt leg movements may explain the altered SBT adaptation pattern observed in this population. Diminished sensation brought on by ACL injury may therefore inaccurately render the representation of the positions of body segments relative to one another (Ivanenko et al., 2011). Sensory receptors are important for successful adaptation to the SBT, and provide critical afferent information concerning both the external and internal environmental conditions of the body and are important for the execution of proper movements. Indeed, a loss of sensory information could affect the ability of an individual with ACL-R to accurately perceive altered or asymmetric movements. This may at least in part explain why people with ACL reconstruction ACLR execute several functional activities with abnormal and asymmetric movements (Gao and Zheng, 2010; Roewer et al., 2011; White et al., 2013). These gait impairments have been shown to persist for up to 17 months after surgery despite surgical reconstruction and successful completion of rehabilitation programs (Bulgheroni et al., 1997; Lewek et al., 2002). Understanding to what extent persons with ACLR are able to detect asymmetric and symmetric
J.A. Roper et al. / Clinical Biomechanics 40 (2016) 52–57
movements would be important to shed light on locomotor adaptation; and provide therapists with knowledge to develop rehabilitation programs targeting locomotor control deficits. Previous studies attempting to evaluate changes related to perception following ACLR have presented an incomplete picture of sensory function of the ACLR knee. Both joint position sense (active reproduction of passive positioning) and kinesthesia (the threshold to detect passive motion) have been used in the literature to measure differences in sensory function of the ACL-R knee. When measuring the threshold to detect passive motion, persons with ACL-R display decreased levels of joint motion perception in the reconstructed knee compared to their intact side (Lephart et al., 1992). Yet, during joint position sense testing, persons with ACL-R were better able to actively reproduce knee angles with their reconstructed leg compared to their healthy side six months postoperatively (Reider et al., 2003). The joint position sense task and the threshold for detection of passive motion task can only provide researchers with a limited evaluation of the reconstructed knee and may not accurately detect or represent the functional role of proprioception during locomotion in individuals with ACLR. Additionally, because individuals with ACLR demonstrate alterations in locomotor control, it would seem important to investigate perception of limb movement during a functional locomotor task, which could be beneficial in determining the significance of these sensory changes. Unfortunately, no study has yet employed a method of measuring perception during a functional locomotor task, which could be beneficial in determining the clinical and neurophysiological significance of these changes in sensorimotor control. Therefore, the aim of this study was to understand the ability of individuals with ACLR to perceive imposed (1) asymmetry and (2) symmetry while walking on a split-belt treadmill (SBT). We hypothesized: 1) the ability to detect symmetry and asymmetry would be altered in individuals with ACLR when the speed of the belt under the reconstructed limb was manipulated compared to when the speed of the belt under the intact limb was manipulated; 2) the ability to detect symmetry and asymmetry would be altered when the reconstructed limb speed was manipulated in individuals with ACLR compared to healthy young adults; and 3) the ability to detect symmetry and asymmetry would be similar when the intact limb speed was manipulated in individuals with ACLR compared to healthy young adults. 2. Methods 2.1. Participants This study was approved by the University's Institutional Review Board, and informed consent was obtained from each individual prior to their participation. Twenty participants with ACLR (12 female, 8 male, age: 20(1) yrs, height 1.71(0.11) m; mass 69(14) kg, 8 autograft hamstring, 7 patellar tendon, 4 Achilles allograft, 1 hybrid allograft/autograft) and 20 healthy age and gender matched controls (HYA) (12 female, 8 male, age: 20(1) yrs, height 1.70(0.12) m; mass 67(14) kg) were recruited. Both groups reported 6(3) h of physical activity per week. None of the participants had walked on a SBT prior to participation in this study. The controls were free from any history of neurological impairment. The ACLR volunteers were able to walk unassisted for a total of 60 min (with optional rest periods) and had sustained at least one anterior cruciate ligament tear verified by examination of a doctor, followed with surgical reconstruction. Only one participant had undergone a second ACLR to the same limb as the first. The time since surgery, type of graft, duration of therapy and current activity level were recorded. 2.2. Experimental protocol Passive reflective markers were attached to the ankle of each limb. Kinematic data, time-synchronized to the kinetic data, were collected
53
using an 8-camera motion capture system (120 Hz; Vicon, Oxford, UK). Kinetic data were collected as the participants walked on an instrumented SBT (1200 Hz; Bertec Corporation, Columbus, OH). Participants wore noise-cancelling headphones and dribble glasses to limit visual and auditory feedback provided by the acceleration and deceleration of the belts, and were instructed to hold on to the handrails. Following a five-minute warm-up and acclimation to the slow (0.75 m/s) and fast speed (1.5 m/s), participants were asked to “indicate when they perceived a difference or similarity in leg movement” during walking conditions in which one belt speed changed incrementally every three strides (velocity = 0.03 m/s, acceleration = 0.1 m/s2). Participants indicated a perceived difference or similarity in leg speeds by raising their hand. For symmetry detection and asymmetry detection only one belt speed was manipulated at a time, so the change in speed occurred during the swing phase of every third stride of the leg on the belt being manipulated. Upon indication of perceived difference or similarity by the participant, both belts were adjusted such that they both moved at 1.0 m/s in order to undo the effects of the most-recent walking condition and establish a washout period lasting one minute. After completing the washout period, the next sub-condition was performed until all sub-conditions were completed. The three conditions performed are defined below.
2.2.1. Symmetry detection This condition allowed a representative measure of the participant's acuity in detecting symmetry when a walking trial began asymmetric and rapidly became more symmetric. Participants began walking with one leg moving at the fast speed, and the other at the slow speed. The following sub-conditions occurred in a random order: 1-a) the left leg starts moving at the fast speed and was decreased. 1-b) the left leg starts moving at the slow speed and was increased. 1-c) the right leg starts moving at the fast speed and was decreased. 1-d) the right leg starts moving at the slow speed and was increased.
2.2.2. Asymmetry detection The purpose of this condition was to observe the participant's sensitivity in detecting when symmetry was no longer present between their leg movements when a walking trial began symmetric and rapidly became more asymmetric. Participants walked on the treadmill with both belts moving together at either the slow speed or fast speed. This condition was repeated three times so that the participant completed the following sub-conditions: 2-a) right leg continues moving at fast speed, left leg gradually decreases. 2-b) left leg continues moving at fast speed, right leg gradually decreases. 2-c) right leg continues moving at slow speed, left leg gradually increases. 2-d) left leg continues moving at slow speed, right leg gradually increases.
2.3. Data processing All variables calculated for the leg on the fast belt are henceforth referred to as the “fast” leg, and the leg on the slow belt will be referred to as the “slow” leg. Initial contact (heel-strikes) and toe-offs were labeled in Vicon software and were determined using marker velocity profiles and a 50-N force-plate threshold (Neckel et al., 2008). Marker data were filtered using a 4th-order low-pass Butterworth filter with a cutoff frequency of 10 Hz. The speed of each belt was also recorded throughout each condition.
54
J.A. Roper et al. / Clinical Biomechanics 40 (2016) 52–57
2.4. Primary outcome measures 2.4.1. Belt-speed ratio For symmetry detection and asymmetry detection, perception performance was measured by the belt-speed ratio, which was calculated as the ratio of the slow belt speed to the fast belt speed, relative to the “best” score a person could achieve for the specific detection task. The greater the belt-speed ratio, the more symmetric the belt speeds when the participant raised their hand. For example, a ratio of 1.0 would indicate a tied-belt condition (right belt = 0.50 m/s and left belt = 0.50 m/s) during symmetry detection, but would indicate a near tied-belt situation (right belt = 0.53 m/s and left belt = 0.50 m/s) during asymmetry detection. For the symmetry detection condition in cases where participants did not detect symmetry after reaching the tied belt condition after 9 strides, a ratio of 0 was assigned. We defined the belt-speed ratio at detection using the following index:
6) Increasing speed of reconstructed limb of ACLR vs. decreasing speed of un-injured limb of ACLR (ACLR reconstructed vs ACLR uninjured) 2.7. Gait cycle parameters at detection In order to understand the type of information individuals were utilizing to detect symmetry/asymmetry, we sought to understand the relationship between the ratio and asymmetry value of each spatiotemporal gait parameter. Relationships between belt-speed ratios and asymmetry scores of each gait parameter were assessed by Pearson correlation coefficients. All statistics were performed using SPSS version 22 software and level of significance for t-tests was set at α b 0.05. For the correlation assessment, significance was set at α ≤ 0.01 to adjust for multiple correlation analyses. 3. Results
Belt−speed Ratio ¼ ðslow belt speed at perception indicationÞ=ðfast belt speed at indicationÞ =best possible ratio for the given detection taskÞ:
2.5. Secondary outcome measures 2.5.1. Spatiotemporal gait parameters Stride length was defined as the anterior posterior distance traveled by the ankle marker from heel-strike to toe-off (Reisman et al., 2005) Stance time was defined as the percent of the gait cycle between heelstrike and subsequent toe-off of the same limb. Step length was defined as the anterior posterior distance between the ankle markers at heelstrike. Fast step length refers to the step length calculated at the heel strike of the fast leg and slow step length to that calculated at the heel strike of the slow leg. Slow double limb support refers to the time from fast leg heel-strike to slow leg toe-off and fast double limb support refers to the time from slow leg heel-strike to fast leg toe-off. We defined symmetry in each spatiotemporal gait parameter using the following asymmetry index: Asymmetry ¼ ðfast leg parameter−slow leg parameterÞ =ðfast leg parameter þ slow leg parameterÞ: Asymmetry scores for spatiotemporal data were averaged over the three strides prior to the participant indicating perception in asymmetry or symmetry. An asymmetry score of 0 would indicate spatiotemporal values were symmetric between both legs at the time of detecting belt symmetry or asymmetry. 2.6. Statistical analyses 2.6.1. Belt-speed ratio T-tests were performed to compare belt-speed ratios between the following pairs of conditions during asymmetry and symmetry detection: 1) Decreasing speed of reconstructed limb of ACLR vs. decreasing speed of non-dominant of HYA (ACLR reconstructed vs HYA nondominant) 2) Decreasing speed of un-injured limb of ACLR vs. decreasing speed of dominant of HYA (ACLR un-injured vs HYA dominant) 3) Decreasing speed of reconstructed limb of ACLR vs. decreasing speed of un-injured limb of ACLR (ACLR reconstructed vs ACLR uninjured) 4) Increasing speed of reconstructed limb of ACLR vs. decreasing speed of non-dominant of HYA (ACLR reconstructed vs HYA nondominant) 5) Increasing speed of un-injured limb of ACLR vs. decreasing speed of dominant of HYA (ACLR un-injured vs HYA dominant)
3.1. Belt speed ratio T-tests did not reveal a difference between belt-speed ratio at detection of symmetry or asymmetry (Fig. 1). 3.2. Gait cycle parameters at detection Correlations and p-values of symmetry and asymmetry scores for stride length, step length, stance time, and double support time during detection for each leg are displayed in Tables 1–4. 4. Discussion Contrary to our hypotheses, individuals with ACLR were able to detect asymmetric and symmetric movements of both limbs comparable to HYA. Also, participants with ACLR detected symmetry and asymmetry equally well regardless of which leg (reconstructed or un-injured) was manipulated. It can be argued that a wide repertoire of feedback information from multiple joints is likely available during walking, (Duysens et al., 1998, 2000; MacKinnon and Winter, 1993) and if enough accurate sensory information is available to the individual, successful identification of asymmetric and symmetric leg movements can result. The findings of this study further suggest that lingering asymmetric behaviors observed during gait in persons with ACLR reported in the literature (Gao and Zheng, 2010; Roewer et al., 2011) are not likely a result of an inability to detect symmetric and asymmetric leg movements. Both acuity (symmetry detection) and sensitivity (asymmetry detection) measures provide a proxy for sensory feedback processing. During active movement it is hypothesized that distinct contributions to feedback exist (Bhanpuri et al., 2013; Elangovan et al., 2014). First, feedback from peripheral transmission of proprioceptive signals is thought to provide information regarding neuromuscular control of gait. Additionally, active movement is influenced by peripheral sensory feedback and underlying sensorimotor integration processes influenced by primary motor and sensory cortices, as well as premotor cortical and subcortical (cerebellum and putamen) regions (Bhanpuri et al., 2013). This contribution is thought to be formed by predicted movement outcomes computed by the cerebellum (Bhanpuri et al., 2013; Ciccarelli et al., 2005). Our locomotor detection tasks were primarily modified by active movements rather than passive movements, and as a result likely used an integration of the two sensory estimates described above. Therefore it is possible that the peripheral sensory dysfunction of the ACLR limb did not largely influence the contributing feedback necessary for active movement. Moreover, we do not suspect feedback formed by cerebellar predictions was largely influenced either (Bhanpuri et al., 2013). As a result individuals with ACLR were able to detect differences
J.A. Roper et al. / Clinical Biomechanics 40 (2016) 52–57
55
Fig. 1. Mean and standard error for belt-speed ratios at detection of symmetry and asymmetry.
and similarities in leg movements at similar belt-speed ratios in comparison to the HYA group. This would further suggest that the two groups have similar acuity and sensitivity abilities between legs. To better understand what criterion was used in identifying symmetry and asymmetry, we also correlated the belt-speed ratio at detection to the amount of asymmetry for step length, stride length, double support time, and stance time for both groups. A recent study investigating the perception threshold in a healthy elderly population suggested that stance time is the criterion used to identify gait asymmetry/symmetry (Lauzière et al., 2014). These results are substantiated by other observations that indicate a strong link between stance time and belt speed during SBT walking (Reisman et al., 2005, 2007, 2013). We also observed a high correlation between stance time asymmetry and belt-speed ratio for certain symmetry detection conditions. Specifically, both ACLR and HYA groups displayed a significant relationship between stance time and detection of symmetry at the slow speed, regardless of what limb was being manipulated. Our study provides evidence that criteria
other than stance time were also highly correlated to belt-speed ratio. Stride length asymmetry, step length asymmetry, and double support time asymmetry all were moderately and significantly correlated (rvalues ranging from −0.801 to 0.783) with belt-speed ratios at detection of symmetry and asymmetry. These findings are in contrast to those reported by Lauzière et al. (2014) who did not observe a significant correlation between double support time and belt-speed ratio. The differences reported between our study and by Lauzière et al. (2014) may be explained by methodological differences. Perhaps the largest difference between the two studies was the strategy utilized in changing belt speeds. Lauzière et al. (2014) manipulated belt speeds by 0.01 m/s every five seconds, while the current study incrementally changed the speed every three strides by 0.03 m/s which began at ipsilateral toe off. We selected a strategy such that the speed of the belt would change while the leg was in swing to prevent the leg from receiving input from a closed chain position regarding the acceleration. Additionally, we chose to apply changes every three strides rather than over
Table 1 Correlations and p-values of Asymmetry Scores for Gait Parameters when Detecting Symmetry at the Fast Speed. ACL
Stride Length Asymmetry
Step Length Asymmetry
Stance Time Asymmetry
Double Support Time Asymmetry
Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value
HYA
Reconstructed
Noninjured
Non-dominant
Dominant
0.054(0.034) −0.305 0.190 −0.015(0.034) 0.107 0.654 −0.056(0.029) 0.106 0.657 −0.027(0.061) 0.384 0.095
0.046(0.028) 0.048 0.841 −0.008(0.016) −0.030 0.901 −0.061(0.024) 0.162 0.495 0.009(0.063) 0.151 0.525
0.060(0.037) −0.519 0.019 −0.001(0.019) 0.457 0.043 −0.068(0.039) 0.747 b 0.001 0.003(0.056) 0.360 0.119
0.046(0.041) −0.660 0.002 −0.016(0.016) 0.469 0.037 −0.077(0.043) 0.671 0.001 −0.005(0.037) 0.466 0.038
Values listed for each limb represents the condition in which the limbs began asymmetric with the specified limb at the slow speed which was then increased until the participant detected symmetry between legs at the fast speed. Bold text indicates significance set at α b 0.05.
56
J.A. Roper et al. / Clinical Biomechanics 40 (2016) 52–57
Table 2 Correlations and p-values of Asymmetry Scores for Gait Parameters when Detecting Symmetry at the Slow Speed. ACL
Stride Length Asymmetry
Step Length Asymmetry
Stance Time Asymmetry
Double Support Time Asymmetry
Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value
HYA
Reconstructed
Noninjured
Non-dominant
Dominant
0.087(0.045) −0.801 b0.001 −0.015(0.028) 0.606 0.005 −0.076(0.039) 0.783 b0.001 0.010(0.078) 0.562 0.010
0.095(0.034) −0.774 b0.001 −0.016(0.028) 0.454 0.044 −0.063(0.038) 0.639 0.002 −0.011(0.063) 0.249 0.289
0.087(0.048) −0.580 0.007 −0.015(0.025) 0.053 0.824 0.079(0.028) 0.551 0.012 0.023(0.038) −0.092 0.700
0.084(0.039) −0.791 b0.001 0.001(0.024) 0.533 0.015 −0.071(0.054) 0.742 b0.001 0.002(0.072) 0.371 0.107
Values listed for each limb represents the condition in which the limbs began asymmetric with the specified limb at the fast speed which was then decreased until the participant detected symmetry between legs at the slow speed. Bold text indicates significance set at α b 0.05.
a set period of time, as recent locomotor adaptation studies have suggested one stride is equal to one practice trial (Roemmich et al., 2014) Thus, we exposed the legs to equal amounts of practice regardless of what speed the belts began or ended at. Stride frequency increases with gait speed (Nilsson and Thorstensson, 1987) and consequently faster walking speeds could have a greater amount of practice trials than slower walking speeds for a given amount of time. Furthermore, if the incremental change is minimal the difference between the legs is not only small, but participants may experience more practice trials depending upon the speed of the leg. We believe the methodological differences listed here may influence sensorimotor integration response, which could impact the spatiotemporal parameters related to detection of asymmetric or symmetric gait. Although the ACLR group performed similarly to the HYA group by measure of belt-speed ratios, they may have utilized unique information to determine detection. None of the spatiotemporal parameters in the ACLR group correlated with belt-speed ratio when asked to detect symmetry at a fast walking speed. In the case of detecting symmetry at a slow walking speed, stride length asymmetry, step length asymmetry and stance time asymmetry correlated with belt-speed ratio in both groups. In sum, it appears that the ACLR group presented the largest differences in spatiotemporal strategy during symmetry detection at the fast walking speed, regardless of what limb was being manipulated. In gait, sensory information may be at least partially dependent on changes in leg speed and may generate specific responses in locomotor control (Vasudevan and Bastian, 2010). These findings in the ACLR group strongly support the implication that specific sensory cues arising from the speed of the leg may also augment perception of symmetry. This alternative strategy may be necessary to successfully execute the
motor task, and could develop due to faulty sensory information from the reconstructed knee at faster walking speeds. In sum, our results convey the importance of investigating perception during locomotor tasks. Specifically, stride length asymmetry, step length asymmetry, stance time asymmetry and double support time asymmetry were significantly related to the belt speed ratio at detection of symmetry at the slow speed, yet none of the asymmetry parameters were related to the belt speed ratio at detection of symmetry at the fast speed in the ACLR group. Strategies for detecting symmetry between individuals with ACLR and healthy individuals are most similar at detecting symmetry at slow speeds compared to fast speeds. Because people with ACLR are at risk for re-injury, it is important to consider the meaning of these alternate strategies. Although individuals with ACLR are still successful at detecting symmetry at faster speed, it is possible that these alternate strategies are masking perceptual deficits that could underlie mechanisms of injury. Our findings suggest that clinicians and researchers should examine sensory function across a range of walking speeds in persons with ACLR, as sensory differences may be most evident at faster walking speeds. The current investigation was not without limitations. First, it is necessary to acknowledge other potential factors related to sensory feedback from the limbs. Parameters such as swing speed asymmetry, (Bruijn et al., 2012) position and stretch of muscles at the hip flexors, as well as kinetic sense could have important roles in perception of interlimb symmetry and asymmetry (Hoogkamer et al., 2014). Second, it is important to note that participants were asked to detect when they felt their limbs were moving different or moving similarly. While individuals should have been attending to how their legs were moving rather than the symmetry of the belts, it is possible that participants
Table 3 Correlations and p-values of Asymmetry Scores for Gait Parameters when Detecting Asymmetry at the Fast Speed. ACL
Stride Length Asymmetry
Step Length Asymmetry
Stance Time Asymmetry
Double Support Time Asymmetry
Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value
HYA
Reconstructed
Noninjured
Non-dominant
Dominant
0.035(0.015) −0.688 0.001 −0.024(0.026) −0.207 0.380 −0.043(0.031) 0.428 0.059 −0.021(0.059) −0.040 0.866
0.043(0.026) −0.120 0.615 −0.032(0.036) −0.051 0.832 −0.036(0.036) 0.158 0.507 −0.052(0.056) −0.162 0.495
0.017(0.047) 0.275 0.240 −0.007(0.036) −0.275 0.241 −0.008(0.060) −0.184 0.438 −0.003(0.058) 0.182 0.443
0.024(0.044) −0.104 0.663 −0.008(0.027) 0.339 0.144 −0.025(0.050) 0.046 0.846 −0.008(0.056) 0.157 0.508
Values listed for each limb represents the condition in which both limbs began at the slow speed and the specified limb speed was increased until the participant detected asymmetry. Bold text indicates significance set at α b 0.05.
J.A. Roper et al. / Clinical Biomechanics 40 (2016) 52–57
57
Table 4 Correlations and p-values of Asymmetry Scores for Gait Parameters when Detecting Asymmetry at the Slow Speed. ACL
Stride Length Asymmetry
Step Length Asymmetry
Stance Time Asymmetry
Double Support Time Asymmetry
Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value
HYA
Reconstructed
Noninjured
Non-dominant
Dominant
0.005(0.047) −0.256 0.275 0.002(0.022) 0.262 0.264 −0.018(0.022) 0.077 0.746 −0.006(0.060) 0.374 0.104
−0.005(0.042) −0.081 0.733 −0.005(0.068) −0.098 0.680 −0.015(0.031) 0.132 0.579 0.050(0.068) 0.124 0.602
0.017(0.011) −0.438 0.054 −0.009(0.012) 0.029 0.904 −0.021(0.028) 0.657 0.002 −0.033(0.047) 0.092 0.701
0.005(0.017) −0.052 0.826 −0.011(0.022) −0.009 0.971 −0.018(0.027) 0.145 0.542 −0.027(0.054) 0.120 0.614
Values listed for each limb represents the condition in which both limbs began at the fast speed and the specified limb speed was decreased until the participant detected asymmetry.
used information derived from the belts upon detection. Finally, participants were asked to hold on to the treadmill rails which could have also influenced the outcomes of this study. Upper extremity movement and limb coordination was likely limited by holding on to the rails, and it is possible that sensory information from upper extremities may be used to detect asymmetric or symmetric movements of the legs. However, because the vision of participants was also being manipulated via dribble goggles, we felt it was necessary for participants to hold on for safety purposes. 5. Conclusion The current study is the first to document the ability of individuals with ACLR to perceive imposed spatiotemporal asymmetry and symmetry. Individuals with ACLR are capable of detecting symmetry and asymmetry similar to a healthy control group, but likely utilize different information to determine symmetry at faster walking speeds. Indeed, sensory feedback from each leg may be amplified by increasing or decreasing speeds and thereby influence the relationship of certain factors related to symmetry and asymmetry detection. Overall, these findings have important implications for developing interventions targeting ACL injury. Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10. 1016/j.clinbiomech.2016.10.017. References Bastian, A.J., 2008. Understanding sensorimotor adaptation and learning for rehabilitation. Curr. Opin. Neurol. 21, 628–633. Bhanpuri, N.H., Okamura, A.M., Bastian, A.J., 2013. Predictive modeling by the cerebellum improves proprioception. J. Neurosci. 33, 14301–14306. Bruijn, S.M., Van Impe, A., Duysens, J., Swinnen, S.P., 2012. Split-belt walking: adaptation differences between young and older adults. J. Neurophysiol. 108, 1149–1157. Bulgheroni, P., Bulgheroni, M.V., Andrini, L., Guffanti, P., Giughello, A., 1997. Gait patterns after anterior cruciate ligament reconstruction. Knee Surg. Sports Traumatol. Arthrosc. 5, 14–21. Ciccarelli, O., Toosy, A.T., Marsden, J.F., Wheeler-Kingshott, C.M., Sahyoun, C., Matthews, P.M., Miller, D.H., Thompson, A.J., 2005. Identifying brain regions for integrative sensorimotor processing with ankle movements. Exp. Brain Res. 166, 31–42. Duysens, J., van Wezel, B.M., van de Crommert, H.W., Faist, M., Kooloos, J.G., 1998. The role of afferent feedback in the control of hamstrings activity during human gait. Eur. J. Morphol. 36, 293–299. Duysens, J., Clarac, F., Cruse, H., 2000. Load-regulating mechanisms in gait and posture: comparative aspects. Physiol. Rev. 80, 83–133. Elangovan, N., Herrmann, A., Konczak, J., 2014. Assessing proprioceptive function: evaluating joint position matching methods against psychophysical thresholds. Phys. Ther. 94, 553–561.
Gao, B., Zheng, N.N., 2010. Alterations in three-dimensional joint kinematics of anterior cruciate ligament-deficient and -reconstructed knees during walking. Clin. Biomech. (Bristol, Avon) 25, 222–229. Hoogkamer, W., Bruijn, S.M., Duysens, J., 2014. Gait parameters affecting the perception threshold of locomotor symmetry: comment on Lauzière, et al. (2014). Percept. Mot. Skills 119, 474–477. Ivanenko, Y.P., Dominici, N., Daprati, E., Nico, D., Cappellini, G., Lacquaniti, F., 2011. Locomotor body scheme. Hum. Mov. Sci. 30, 341–351. Lauzière, S., Miéville, C., Duclos, C., Aissaoui, R., Nadeau, S., 2014. Perception threshold of locomotor symmetry while walking on a split-belt treadmill in healthy elderly individuals. Percept. Mot. Skills 118, 475–490. Lephart, S.M., Perrin, D.H., Fu, F.H., Gieck, J.H., McCue, F.C., Irrgang, J.J., 1992. Relationship between selected physical characteristics and functional capacity in the anterior cruciate ligament-insufficient athlete. J. Orthop. Sports Phys. Ther. 16, 174–181. Lewek, M., Rudolph, K., Axe, M., Snyder-Mackler, L., 2002. The effect of insufficient quadriceps strength on gait after anterior cruciate ligament reconstruction. Clin. Biomech. 17, 56–63. Littmann, A.E., Iguchi, M., Madhavan, S., Kolarik, J.L., Shields, R.K., 2012. Dynamic-positionsense impairment's independence of perceived knee function in women with ACL reconstruction. J. Sport Rehabil. 21, 44–53. MacKinnon, C.D., Winter, D.A., 1993. Control of whole body balance in the frontal plane during human walking. J. Biomech. 26, 633–644. Neckel, N.D., Blonien, N., Nichols, D., Hidler, J., 2008. Abnormal joint torque patterns exhibited by chronic stroke subjects while walking with a prescribed physiological gait pattern. J. Neuroeng. Rehabil. 5 (19-0003-5-19). Nilsson, J., Thorstensson, A., 1987. Adaptability in frequency and amplitude of leg movements during human locomotion at different speeds. Acta Physiol. Scand. 129, 107–114. Reider, B., Arcand, M.A., Diehl, L.H., Mroczek, K., Abulencia, A., Stroud, C.C., Palm, M., Gilbertson, J., Staszak, P., 2003. Proprioception of the knee before and after anterior cruciate ligament reconstruction. Arthroscopy 19, 2–12. Reisman, D.S., Block, H.J., Bastian, A.J., 2005. Interlimb coordination during locomotion: what can be adapted and stored? J. Neurophysiol. 94, 2403–2415. Reisman, D.S., Wityk, R., Silver, K., Bastian, A.J., 2007. Locomotor adaptation on a split-belt treadmill can improve walking symmetry post-stroke. Brain 130, 1861–1872. Reisman, D.S., Bastian, A.J., Morton, S.M., 2010. Neurophysiologic and rehabilitation insights from the split-belt and other locomotor adaptation paradigms. Phys. Ther. 90, 187–195. Reisman, D.S., McLean, H., Keller, J., Danks, K.A., Bastian, A.J., 2013. Repeated split-belt treadmill training improves poststroke step length asymmetry. Neurorehabil. Neural Repair 27, 460–468. Roemmich, R.T., Nocera, J.R., Stegemöller, E.L., Hassan, A., Okun, M.S., Hass, C.J., 2014. Locomotor adaptation and locomotor adaptive learning in Parkinson's disease and normal aging. Clin. Neurophysiol. 125, 313–319. Roewer, B.D., Di Stasi, S.L., Snyder-Mackler, L., 2011. Quadriceps strength and weight acceptance strategies continue to improve two years after anterior cruciate ligament reconstruction. J. Biomech. 44, 1948–1953. Roper, J.A., Terza, M.J., Tillman, M.D., Hass, C.J., 2016. Adaptation strategies of individuals with anterior cruciate ligament reconstruction. Orthop. J. Sports Med. 4 (2). http://dx. doi.org/10.1177/2325967115627611. Vasudevan, E.V., Bastian, A.J., 2010. Split-belt treadmill adaptation shows different functional networks for fast and slow human walking. J. Neurophysiol. 103, 183–191. White, K., Di Stasi, S.L., Smith, A.H., Snyder-Mackler, L., 2013. Anterior cruciate ligamentspecialized post-operative return-to-sports (ACL-SPORTS) training: a randomized control trial. BMC Musculoskelet. Disord. 14, 108.
Contents lists available at ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Perception of symmetry and asymmetry in individuals with anterior cruciate ligament reconstruction Jaimie A. Roper ⁎, Matthew J. Terza, Chris J. Hass University of Florida, College of Health and Human Performance, Department of Applied Physiology and Kinesiology, USA
a r t i c l e
i n f o
Article history: Received 4 January 2016 Accepted 29 October 2016 Keywords: ACL Reconstruction Symmetry Gait Walking Sensory
a b s t r a c t Background: Changes in the quantity, quality and integration of sensory information are thought to persist long after anterior cruciate ligament reconstruction and completion of physical therapy. Our purpose was to investigate the ability of individuals with anterior cruciate ligament reconstruction to perceive imposed asymmetry and symmetry while walking. Methods: Twenty participants with anterior cruciate ligament reconstruction and 20 controls walked on a splitbelt treadmill while we assessed the ability to detect symmetry and asymmetry at fast and slow speeds. Detection scores and spatiotemporal data during asymmetric and symmetric tasks in which the belts were coupled or decoupled over time were recorded. Findings: The ability to detect symmetry and asymmetry was not altered in individuals with anterior cruciate ligament reconstruction compared to healthy young adults. The belt-speed ratio at detection also correlated to asymmetry for step length, stride length, double support time, and stance time. However, the anterior cruciate ligament reconstruction group appeared to utilize unique information to determine detection. When asked to detect symmetry at a fast speed, no asymmetry scores significantly correlated with belt-speed ratio in the anterior cruciate ligament reconstruction group. Conversely, asymmetry in stride length, step length, and stance time all significantly correlated with belt-speed ratio at detection in the control group. Interpretation: Specific sensory cues arising from the speed of the leg may also augment perception of symmetry. This strategy may be necessary in order to successfully execute the motor task, and could develop due to altered sensory information from the reconstructed knee at faster walking speeds. Published by Elsevier Ltd.
1. Introduction Locomotor adaptation paradigms using split-belt treadmills (SBT) have provided important insight into the neural control of movement and the ability of an individual to alter motor behavior of gait. Adaptation is a short-term learning process, which provides information with respect to how much the body can adapt to a new behavior and is considered a surrogate for the amount of flexibility within a behavioral pattern (Bastian, 2008; Reisman et al., 2010). During SBT walking the two belts can be decoupled such that one leg walks faster than the other, resulting in both reactive (fast) and predictive (slow) adjustments adaptation to an asymmetric walking pattern (Reisman et al., 2010). Recently, we have documented that individuals with anterior cruciate ligament reconstruction (ACLR) following anterior cruciate ligament (ACL) injury demonstrate altered locomotor adaptation to decoupled belt speeds (Roper et al., 2016). Specifically, we observed impairments in both slow-adapting and fast-adapting derived gait parameters, suggesting that fundamental features of motor control remain altered ⁎ Corresponding author at: 100 Florida Gym, PO Box 118205, Gainesville, FL, USA 32611-8205. E-mail address: jaimier@ufl.edu (J.A. Roper).
http://dx.doi.org/10.1016/j.clinbiomech.2016.10.017 0268-0033/Published by Elsevier Ltd.
despite surgical reconstruction and completion of physical therapy. Because individuals with ACLR demonstrate alterations in locomotor adaptation and also in proprioception (Littmann et al., 2012; Reider et al., 2003), deficits in the ability to accurately perceive asymmetric and symmetric belt leg movements may explain the altered SBT adaptation pattern observed in this population. Diminished sensation brought on by ACL injury may therefore inaccurately render the representation of the positions of body segments relative to one another (Ivanenko et al., 2011). Sensory receptors are important for successful adaptation to the SBT, and provide critical afferent information concerning both the external and internal environmental conditions of the body and are important for the execution of proper movements. Indeed, a loss of sensory information could affect the ability of an individual with ACL-R to accurately perceive altered or asymmetric movements. This may at least in part explain why people with ACL reconstruction ACLR execute several functional activities with abnormal and asymmetric movements (Gao and Zheng, 2010; Roewer et al., 2011; White et al., 2013). These gait impairments have been shown to persist for up to 17 months after surgery despite surgical reconstruction and successful completion of rehabilitation programs (Bulgheroni et al., 1997; Lewek et al., 2002). Understanding to what extent persons with ACLR are able to detect asymmetric and symmetric
J.A. Roper et al. / Clinical Biomechanics 40 (2016) 52–57
movements would be important to shed light on locomotor adaptation; and provide therapists with knowledge to develop rehabilitation programs targeting locomotor control deficits. Previous studies attempting to evaluate changes related to perception following ACLR have presented an incomplete picture of sensory function of the ACLR knee. Both joint position sense (active reproduction of passive positioning) and kinesthesia (the threshold to detect passive motion) have been used in the literature to measure differences in sensory function of the ACL-R knee. When measuring the threshold to detect passive motion, persons with ACL-R display decreased levels of joint motion perception in the reconstructed knee compared to their intact side (Lephart et al., 1992). Yet, during joint position sense testing, persons with ACL-R were better able to actively reproduce knee angles with their reconstructed leg compared to their healthy side six months postoperatively (Reider et al., 2003). The joint position sense task and the threshold for detection of passive motion task can only provide researchers with a limited evaluation of the reconstructed knee and may not accurately detect or represent the functional role of proprioception during locomotion in individuals with ACLR. Additionally, because individuals with ACLR demonstrate alterations in locomotor control, it would seem important to investigate perception of limb movement during a functional locomotor task, which could be beneficial in determining the significance of these sensory changes. Unfortunately, no study has yet employed a method of measuring perception during a functional locomotor task, which could be beneficial in determining the clinical and neurophysiological significance of these changes in sensorimotor control. Therefore, the aim of this study was to understand the ability of individuals with ACLR to perceive imposed (1) asymmetry and (2) symmetry while walking on a split-belt treadmill (SBT). We hypothesized: 1) the ability to detect symmetry and asymmetry would be altered in individuals with ACLR when the speed of the belt under the reconstructed limb was manipulated compared to when the speed of the belt under the intact limb was manipulated; 2) the ability to detect symmetry and asymmetry would be altered when the reconstructed limb speed was manipulated in individuals with ACLR compared to healthy young adults; and 3) the ability to detect symmetry and asymmetry would be similar when the intact limb speed was manipulated in individuals with ACLR compared to healthy young adults. 2. Methods 2.1. Participants This study was approved by the University's Institutional Review Board, and informed consent was obtained from each individual prior to their participation. Twenty participants with ACLR (12 female, 8 male, age: 20(1) yrs, height 1.71(0.11) m; mass 69(14) kg, 8 autograft hamstring, 7 patellar tendon, 4 Achilles allograft, 1 hybrid allograft/autograft) and 20 healthy age and gender matched controls (HYA) (12 female, 8 male, age: 20(1) yrs, height 1.70(0.12) m; mass 67(14) kg) were recruited. Both groups reported 6(3) h of physical activity per week. None of the participants had walked on a SBT prior to participation in this study. The controls were free from any history of neurological impairment. The ACLR volunteers were able to walk unassisted for a total of 60 min (with optional rest periods) and had sustained at least one anterior cruciate ligament tear verified by examination of a doctor, followed with surgical reconstruction. Only one participant had undergone a second ACLR to the same limb as the first. The time since surgery, type of graft, duration of therapy and current activity level were recorded. 2.2. Experimental protocol Passive reflective markers were attached to the ankle of each limb. Kinematic data, time-synchronized to the kinetic data, were collected
53
using an 8-camera motion capture system (120 Hz; Vicon, Oxford, UK). Kinetic data were collected as the participants walked on an instrumented SBT (1200 Hz; Bertec Corporation, Columbus, OH). Participants wore noise-cancelling headphones and dribble glasses to limit visual and auditory feedback provided by the acceleration and deceleration of the belts, and were instructed to hold on to the handrails. Following a five-minute warm-up and acclimation to the slow (0.75 m/s) and fast speed (1.5 m/s), participants were asked to “indicate when they perceived a difference or similarity in leg movement” during walking conditions in which one belt speed changed incrementally every three strides (velocity = 0.03 m/s, acceleration = 0.1 m/s2). Participants indicated a perceived difference or similarity in leg speeds by raising their hand. For symmetry detection and asymmetry detection only one belt speed was manipulated at a time, so the change in speed occurred during the swing phase of every third stride of the leg on the belt being manipulated. Upon indication of perceived difference or similarity by the participant, both belts were adjusted such that they both moved at 1.0 m/s in order to undo the effects of the most-recent walking condition and establish a washout period lasting one minute. After completing the washout period, the next sub-condition was performed until all sub-conditions were completed. The three conditions performed are defined below.
2.2.1. Symmetry detection This condition allowed a representative measure of the participant's acuity in detecting symmetry when a walking trial began asymmetric and rapidly became more symmetric. Participants began walking with one leg moving at the fast speed, and the other at the slow speed. The following sub-conditions occurred in a random order: 1-a) the left leg starts moving at the fast speed and was decreased. 1-b) the left leg starts moving at the slow speed and was increased. 1-c) the right leg starts moving at the fast speed and was decreased. 1-d) the right leg starts moving at the slow speed and was increased.
2.2.2. Asymmetry detection The purpose of this condition was to observe the participant's sensitivity in detecting when symmetry was no longer present between their leg movements when a walking trial began symmetric and rapidly became more asymmetric. Participants walked on the treadmill with both belts moving together at either the slow speed or fast speed. This condition was repeated three times so that the participant completed the following sub-conditions: 2-a) right leg continues moving at fast speed, left leg gradually decreases. 2-b) left leg continues moving at fast speed, right leg gradually decreases. 2-c) right leg continues moving at slow speed, left leg gradually increases. 2-d) left leg continues moving at slow speed, right leg gradually increases.
2.3. Data processing All variables calculated for the leg on the fast belt are henceforth referred to as the “fast” leg, and the leg on the slow belt will be referred to as the “slow” leg. Initial contact (heel-strikes) and toe-offs were labeled in Vicon software and were determined using marker velocity profiles and a 50-N force-plate threshold (Neckel et al., 2008). Marker data were filtered using a 4th-order low-pass Butterworth filter with a cutoff frequency of 10 Hz. The speed of each belt was also recorded throughout each condition.
54
J.A. Roper et al. / Clinical Biomechanics 40 (2016) 52–57
2.4. Primary outcome measures 2.4.1. Belt-speed ratio For symmetry detection and asymmetry detection, perception performance was measured by the belt-speed ratio, which was calculated as the ratio of the slow belt speed to the fast belt speed, relative to the “best” score a person could achieve for the specific detection task. The greater the belt-speed ratio, the more symmetric the belt speeds when the participant raised their hand. For example, a ratio of 1.0 would indicate a tied-belt condition (right belt = 0.50 m/s and left belt = 0.50 m/s) during symmetry detection, but would indicate a near tied-belt situation (right belt = 0.53 m/s and left belt = 0.50 m/s) during asymmetry detection. For the symmetry detection condition in cases where participants did not detect symmetry after reaching the tied belt condition after 9 strides, a ratio of 0 was assigned. We defined the belt-speed ratio at detection using the following index:
6) Increasing speed of reconstructed limb of ACLR vs. decreasing speed of un-injured limb of ACLR (ACLR reconstructed vs ACLR uninjured) 2.7. Gait cycle parameters at detection In order to understand the type of information individuals were utilizing to detect symmetry/asymmetry, we sought to understand the relationship between the ratio and asymmetry value of each spatiotemporal gait parameter. Relationships between belt-speed ratios and asymmetry scores of each gait parameter were assessed by Pearson correlation coefficients. All statistics were performed using SPSS version 22 software and level of significance for t-tests was set at α b 0.05. For the correlation assessment, significance was set at α ≤ 0.01 to adjust for multiple correlation analyses. 3. Results
Belt−speed Ratio ¼ ðslow belt speed at perception indicationÞ=ðfast belt speed at indicationÞ =best possible ratio for the given detection taskÞ:
2.5. Secondary outcome measures 2.5.1. Spatiotemporal gait parameters Stride length was defined as the anterior posterior distance traveled by the ankle marker from heel-strike to toe-off (Reisman et al., 2005) Stance time was defined as the percent of the gait cycle between heelstrike and subsequent toe-off of the same limb. Step length was defined as the anterior posterior distance between the ankle markers at heelstrike. Fast step length refers to the step length calculated at the heel strike of the fast leg and slow step length to that calculated at the heel strike of the slow leg. Slow double limb support refers to the time from fast leg heel-strike to slow leg toe-off and fast double limb support refers to the time from slow leg heel-strike to fast leg toe-off. We defined symmetry in each spatiotemporal gait parameter using the following asymmetry index: Asymmetry ¼ ðfast leg parameter−slow leg parameterÞ =ðfast leg parameter þ slow leg parameterÞ: Asymmetry scores for spatiotemporal data were averaged over the three strides prior to the participant indicating perception in asymmetry or symmetry. An asymmetry score of 0 would indicate spatiotemporal values were symmetric between both legs at the time of detecting belt symmetry or asymmetry. 2.6. Statistical analyses 2.6.1. Belt-speed ratio T-tests were performed to compare belt-speed ratios between the following pairs of conditions during asymmetry and symmetry detection: 1) Decreasing speed of reconstructed limb of ACLR vs. decreasing speed of non-dominant of HYA (ACLR reconstructed vs HYA nondominant) 2) Decreasing speed of un-injured limb of ACLR vs. decreasing speed of dominant of HYA (ACLR un-injured vs HYA dominant) 3) Decreasing speed of reconstructed limb of ACLR vs. decreasing speed of un-injured limb of ACLR (ACLR reconstructed vs ACLR uninjured) 4) Increasing speed of reconstructed limb of ACLR vs. decreasing speed of non-dominant of HYA (ACLR reconstructed vs HYA nondominant) 5) Increasing speed of un-injured limb of ACLR vs. decreasing speed of dominant of HYA (ACLR un-injured vs HYA dominant)
3.1. Belt speed ratio T-tests did not reveal a difference between belt-speed ratio at detection of symmetry or asymmetry (Fig. 1). 3.2. Gait cycle parameters at detection Correlations and p-values of symmetry and asymmetry scores for stride length, step length, stance time, and double support time during detection for each leg are displayed in Tables 1–4. 4. Discussion Contrary to our hypotheses, individuals with ACLR were able to detect asymmetric and symmetric movements of both limbs comparable to HYA. Also, participants with ACLR detected symmetry and asymmetry equally well regardless of which leg (reconstructed or un-injured) was manipulated. It can be argued that a wide repertoire of feedback information from multiple joints is likely available during walking, (Duysens et al., 1998, 2000; MacKinnon and Winter, 1993) and if enough accurate sensory information is available to the individual, successful identification of asymmetric and symmetric leg movements can result. The findings of this study further suggest that lingering asymmetric behaviors observed during gait in persons with ACLR reported in the literature (Gao and Zheng, 2010; Roewer et al., 2011) are not likely a result of an inability to detect symmetric and asymmetric leg movements. Both acuity (symmetry detection) and sensitivity (asymmetry detection) measures provide a proxy for sensory feedback processing. During active movement it is hypothesized that distinct contributions to feedback exist (Bhanpuri et al., 2013; Elangovan et al., 2014). First, feedback from peripheral transmission of proprioceptive signals is thought to provide information regarding neuromuscular control of gait. Additionally, active movement is influenced by peripheral sensory feedback and underlying sensorimotor integration processes influenced by primary motor and sensory cortices, as well as premotor cortical and subcortical (cerebellum and putamen) regions (Bhanpuri et al., 2013). This contribution is thought to be formed by predicted movement outcomes computed by the cerebellum (Bhanpuri et al., 2013; Ciccarelli et al., 2005). Our locomotor detection tasks were primarily modified by active movements rather than passive movements, and as a result likely used an integration of the two sensory estimates described above. Therefore it is possible that the peripheral sensory dysfunction of the ACLR limb did not largely influence the contributing feedback necessary for active movement. Moreover, we do not suspect feedback formed by cerebellar predictions was largely influenced either (Bhanpuri et al., 2013). As a result individuals with ACLR were able to detect differences
J.A. Roper et al. / Clinical Biomechanics 40 (2016) 52–57
55
Fig. 1. Mean and standard error for belt-speed ratios at detection of symmetry and asymmetry.
and similarities in leg movements at similar belt-speed ratios in comparison to the HYA group. This would further suggest that the two groups have similar acuity and sensitivity abilities between legs. To better understand what criterion was used in identifying symmetry and asymmetry, we also correlated the belt-speed ratio at detection to the amount of asymmetry for step length, stride length, double support time, and stance time for both groups. A recent study investigating the perception threshold in a healthy elderly population suggested that stance time is the criterion used to identify gait asymmetry/symmetry (Lauzière et al., 2014). These results are substantiated by other observations that indicate a strong link between stance time and belt speed during SBT walking (Reisman et al., 2005, 2007, 2013). We also observed a high correlation between stance time asymmetry and belt-speed ratio for certain symmetry detection conditions. Specifically, both ACLR and HYA groups displayed a significant relationship between stance time and detection of symmetry at the slow speed, regardless of what limb was being manipulated. Our study provides evidence that criteria
other than stance time were also highly correlated to belt-speed ratio. Stride length asymmetry, step length asymmetry, and double support time asymmetry all were moderately and significantly correlated (rvalues ranging from −0.801 to 0.783) with belt-speed ratios at detection of symmetry and asymmetry. These findings are in contrast to those reported by Lauzière et al. (2014) who did not observe a significant correlation between double support time and belt-speed ratio. The differences reported between our study and by Lauzière et al. (2014) may be explained by methodological differences. Perhaps the largest difference between the two studies was the strategy utilized in changing belt speeds. Lauzière et al. (2014) manipulated belt speeds by 0.01 m/s every five seconds, while the current study incrementally changed the speed every three strides by 0.03 m/s which began at ipsilateral toe off. We selected a strategy such that the speed of the belt would change while the leg was in swing to prevent the leg from receiving input from a closed chain position regarding the acceleration. Additionally, we chose to apply changes every three strides rather than over
Table 1 Correlations and p-values of Asymmetry Scores for Gait Parameters when Detecting Symmetry at the Fast Speed. ACL
Stride Length Asymmetry
Step Length Asymmetry
Stance Time Asymmetry
Double Support Time Asymmetry
Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value
HYA
Reconstructed
Noninjured
Non-dominant
Dominant
0.054(0.034) −0.305 0.190 −0.015(0.034) 0.107 0.654 −0.056(0.029) 0.106 0.657 −0.027(0.061) 0.384 0.095
0.046(0.028) 0.048 0.841 −0.008(0.016) −0.030 0.901 −0.061(0.024) 0.162 0.495 0.009(0.063) 0.151 0.525
0.060(0.037) −0.519 0.019 −0.001(0.019) 0.457 0.043 −0.068(0.039) 0.747 b 0.001 0.003(0.056) 0.360 0.119
0.046(0.041) −0.660 0.002 −0.016(0.016) 0.469 0.037 −0.077(0.043) 0.671 0.001 −0.005(0.037) 0.466 0.038
Values listed for each limb represents the condition in which the limbs began asymmetric with the specified limb at the slow speed which was then increased until the participant detected symmetry between legs at the fast speed. Bold text indicates significance set at α b 0.05.
56
J.A. Roper et al. / Clinical Biomechanics 40 (2016) 52–57
Table 2 Correlations and p-values of Asymmetry Scores for Gait Parameters when Detecting Symmetry at the Slow Speed. ACL
Stride Length Asymmetry
Step Length Asymmetry
Stance Time Asymmetry
Double Support Time Asymmetry
Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value
HYA
Reconstructed
Noninjured
Non-dominant
Dominant
0.087(0.045) −0.801 b0.001 −0.015(0.028) 0.606 0.005 −0.076(0.039) 0.783 b0.001 0.010(0.078) 0.562 0.010
0.095(0.034) −0.774 b0.001 −0.016(0.028) 0.454 0.044 −0.063(0.038) 0.639 0.002 −0.011(0.063) 0.249 0.289
0.087(0.048) −0.580 0.007 −0.015(0.025) 0.053 0.824 0.079(0.028) 0.551 0.012 0.023(0.038) −0.092 0.700
0.084(0.039) −0.791 b0.001 0.001(0.024) 0.533 0.015 −0.071(0.054) 0.742 b0.001 0.002(0.072) 0.371 0.107
Values listed for each limb represents the condition in which the limbs began asymmetric with the specified limb at the fast speed which was then decreased until the participant detected symmetry between legs at the slow speed. Bold text indicates significance set at α b 0.05.
a set period of time, as recent locomotor adaptation studies have suggested one stride is equal to one practice trial (Roemmich et al., 2014) Thus, we exposed the legs to equal amounts of practice regardless of what speed the belts began or ended at. Stride frequency increases with gait speed (Nilsson and Thorstensson, 1987) and consequently faster walking speeds could have a greater amount of practice trials than slower walking speeds for a given amount of time. Furthermore, if the incremental change is minimal the difference between the legs is not only small, but participants may experience more practice trials depending upon the speed of the leg. We believe the methodological differences listed here may influence sensorimotor integration response, which could impact the spatiotemporal parameters related to detection of asymmetric or symmetric gait. Although the ACLR group performed similarly to the HYA group by measure of belt-speed ratios, they may have utilized unique information to determine detection. None of the spatiotemporal parameters in the ACLR group correlated with belt-speed ratio when asked to detect symmetry at a fast walking speed. In the case of detecting symmetry at a slow walking speed, stride length asymmetry, step length asymmetry and stance time asymmetry correlated with belt-speed ratio in both groups. In sum, it appears that the ACLR group presented the largest differences in spatiotemporal strategy during symmetry detection at the fast walking speed, regardless of what limb was being manipulated. In gait, sensory information may be at least partially dependent on changes in leg speed and may generate specific responses in locomotor control (Vasudevan and Bastian, 2010). These findings in the ACLR group strongly support the implication that specific sensory cues arising from the speed of the leg may also augment perception of symmetry. This alternative strategy may be necessary to successfully execute the
motor task, and could develop due to faulty sensory information from the reconstructed knee at faster walking speeds. In sum, our results convey the importance of investigating perception during locomotor tasks. Specifically, stride length asymmetry, step length asymmetry, stance time asymmetry and double support time asymmetry were significantly related to the belt speed ratio at detection of symmetry at the slow speed, yet none of the asymmetry parameters were related to the belt speed ratio at detection of symmetry at the fast speed in the ACLR group. Strategies for detecting symmetry between individuals with ACLR and healthy individuals are most similar at detecting symmetry at slow speeds compared to fast speeds. Because people with ACLR are at risk for re-injury, it is important to consider the meaning of these alternate strategies. Although individuals with ACLR are still successful at detecting symmetry at faster speed, it is possible that these alternate strategies are masking perceptual deficits that could underlie mechanisms of injury. Our findings suggest that clinicians and researchers should examine sensory function across a range of walking speeds in persons with ACLR, as sensory differences may be most evident at faster walking speeds. The current investigation was not without limitations. First, it is necessary to acknowledge other potential factors related to sensory feedback from the limbs. Parameters such as swing speed asymmetry, (Bruijn et al., 2012) position and stretch of muscles at the hip flexors, as well as kinetic sense could have important roles in perception of interlimb symmetry and asymmetry (Hoogkamer et al., 2014). Second, it is important to note that participants were asked to detect when they felt their limbs were moving different or moving similarly. While individuals should have been attending to how their legs were moving rather than the symmetry of the belts, it is possible that participants
Table 3 Correlations and p-values of Asymmetry Scores for Gait Parameters when Detecting Asymmetry at the Fast Speed. ACL
Stride Length Asymmetry
Step Length Asymmetry
Stance Time Asymmetry
Double Support Time Asymmetry
Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value
HYA
Reconstructed
Noninjured
Non-dominant
Dominant
0.035(0.015) −0.688 0.001 −0.024(0.026) −0.207 0.380 −0.043(0.031) 0.428 0.059 −0.021(0.059) −0.040 0.866
0.043(0.026) −0.120 0.615 −0.032(0.036) −0.051 0.832 −0.036(0.036) 0.158 0.507 −0.052(0.056) −0.162 0.495
0.017(0.047) 0.275 0.240 −0.007(0.036) −0.275 0.241 −0.008(0.060) −0.184 0.438 −0.003(0.058) 0.182 0.443
0.024(0.044) −0.104 0.663 −0.008(0.027) 0.339 0.144 −0.025(0.050) 0.046 0.846 −0.008(0.056) 0.157 0.508
Values listed for each limb represents the condition in which both limbs began at the slow speed and the specified limb speed was increased until the participant detected asymmetry. Bold text indicates significance set at α b 0.05.
J.A. Roper et al. / Clinical Biomechanics 40 (2016) 52–57
57
Table 4 Correlations and p-values of Asymmetry Scores for Gait Parameters when Detecting Asymmetry at the Slow Speed. ACL
Stride Length Asymmetry
Step Length Asymmetry
Stance Time Asymmetry
Double Support Time Asymmetry
Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value Mean(SD) Pearson Correlation Coefficient p-value
HYA
Reconstructed
Noninjured
Non-dominant
Dominant
0.005(0.047) −0.256 0.275 0.002(0.022) 0.262 0.264 −0.018(0.022) 0.077 0.746 −0.006(0.060) 0.374 0.104
−0.005(0.042) −0.081 0.733 −0.005(0.068) −0.098 0.680 −0.015(0.031) 0.132 0.579 0.050(0.068) 0.124 0.602
0.017(0.011) −0.438 0.054 −0.009(0.012) 0.029 0.904 −0.021(0.028) 0.657 0.002 −0.033(0.047) 0.092 0.701
0.005(0.017) −0.052 0.826 −0.011(0.022) −0.009 0.971 −0.018(0.027) 0.145 0.542 −0.027(0.054) 0.120 0.614
Values listed for each limb represents the condition in which both limbs began at the fast speed and the specified limb speed was decreased until the participant detected asymmetry.
used information derived from the belts upon detection. Finally, participants were asked to hold on to the treadmill rails which could have also influenced the outcomes of this study. Upper extremity movement and limb coordination was likely limited by holding on to the rails, and it is possible that sensory information from upper extremities may be used to detect asymmetric or symmetric movements of the legs. However, because the vision of participants was also being manipulated via dribble goggles, we felt it was necessary for participants to hold on for safety purposes. 5. Conclusion The current study is the first to document the ability of individuals with ACLR to perceive imposed spatiotemporal asymmetry and symmetry. Individuals with ACLR are capable of detecting symmetry and asymmetry similar to a healthy control group, but likely utilize different information to determine symmetry at faster walking speeds. Indeed, sensory feedback from each leg may be amplified by increasing or decreasing speeds and thereby influence the relationship of certain factors related to symmetry and asymmetry detection. Overall, these findings have important implications for developing interventions targeting ACL injury. Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10. 1016/j.clinbiomech.2016.10.017. References Bastian, A.J., 2008. Understanding sensorimotor adaptation and learning for rehabilitation. Curr. Opin. Neurol. 21, 628–633. Bhanpuri, N.H., Okamura, A.M., Bastian, A.J., 2013. Predictive modeling by the cerebellum improves proprioception. J. Neurosci. 33, 14301–14306. Bruijn, S.M., Van Impe, A., Duysens, J., Swinnen, S.P., 2012. Split-belt walking: adaptation differences between young and older adults. J. Neurophysiol. 108, 1149–1157. Bulgheroni, P., Bulgheroni, M.V., Andrini, L., Guffanti, P., Giughello, A., 1997. Gait patterns after anterior cruciate ligament reconstruction. Knee Surg. Sports Traumatol. Arthrosc. 5, 14–21. Ciccarelli, O., Toosy, A.T., Marsden, J.F., Wheeler-Kingshott, C.M., Sahyoun, C., Matthews, P.M., Miller, D.H., Thompson, A.J., 2005. Identifying brain regions for integrative sensorimotor processing with ankle movements. Exp. Brain Res. 166, 31–42. Duysens, J., van Wezel, B.M., van de Crommert, H.W., Faist, M., Kooloos, J.G., 1998. The role of afferent feedback in the control of hamstrings activity during human gait. Eur. J. Morphol. 36, 293–299. Duysens, J., Clarac, F., Cruse, H., 2000. Load-regulating mechanisms in gait and posture: comparative aspects. Physiol. Rev. 80, 83–133. Elangovan, N., Herrmann, A., Konczak, J., 2014. Assessing proprioceptive function: evaluating joint position matching methods against psychophysical thresholds. Phys. Ther. 94, 553–561.
Gao, B., Zheng, N.N., 2010. Alterations in three-dimensional joint kinematics of anterior cruciate ligament-deficient and -reconstructed knees during walking. Clin. Biomech. (Bristol, Avon) 25, 222–229. Hoogkamer, W., Bruijn, S.M., Duysens, J., 2014. Gait parameters affecting the perception threshold of locomotor symmetry: comment on Lauzière, et al. (2014). Percept. Mot. Skills 119, 474–477. Ivanenko, Y.P., Dominici, N., Daprati, E., Nico, D., Cappellini, G., Lacquaniti, F., 2011. Locomotor body scheme. Hum. Mov. Sci. 30, 341–351. Lauzière, S., Miéville, C., Duclos, C., Aissaoui, R., Nadeau, S., 2014. Perception threshold of locomotor symmetry while walking on a split-belt treadmill in healthy elderly individuals. Percept. Mot. Skills 118, 475–490. Lephart, S.M., Perrin, D.H., Fu, F.H., Gieck, J.H., McCue, F.C., Irrgang, J.J., 1992. Relationship between selected physical characteristics and functional capacity in the anterior cruciate ligament-insufficient athlete. J. Orthop. Sports Phys. Ther. 16, 174–181. Lewek, M., Rudolph, K., Axe, M., Snyder-Mackler, L., 2002. The effect of insufficient quadriceps strength on gait after anterior cruciate ligament reconstruction. Clin. Biomech. 17, 56–63. Littmann, A.E., Iguchi, M., Madhavan, S., Kolarik, J.L., Shields, R.K., 2012. Dynamic-positionsense impairment's independence of perceived knee function in women with ACL reconstruction. J. Sport Rehabil. 21, 44–53. MacKinnon, C.D., Winter, D.A., 1993. Control of whole body balance in the frontal plane during human walking. J. Biomech. 26, 633–644. Neckel, N.D., Blonien, N., Nichols, D., Hidler, J., 2008. Abnormal joint torque patterns exhibited by chronic stroke subjects while walking with a prescribed physiological gait pattern. J. Neuroeng. Rehabil. 5 (19-0003-5-19). Nilsson, J., Thorstensson, A., 1987. Adaptability in frequency and amplitude of leg movements during human locomotion at different speeds. Acta Physiol. Scand. 129, 107–114. Reider, B., Arcand, M.A., Diehl, L.H., Mroczek, K., Abulencia, A., Stroud, C.C., Palm, M., Gilbertson, J., Staszak, P., 2003. Proprioception of the knee before and after anterior cruciate ligament reconstruction. Arthroscopy 19, 2–12. Reisman, D.S., Block, H.J., Bastian, A.J., 2005. Interlimb coordination during locomotion: what can be adapted and stored? J. Neurophysiol. 94, 2403–2415. Reisman, D.S., Wityk, R., Silver, K., Bastian, A.J., 2007. Locomotor adaptation on a split-belt treadmill can improve walking symmetry post-stroke. Brain 130, 1861–1872. Reisman, D.S., Bastian, A.J., Morton, S.M., 2010. Neurophysiologic and rehabilitation insights from the split-belt and other locomotor adaptation paradigms. Phys. Ther. 90, 187–195. Reisman, D.S., McLean, H., Keller, J., Danks, K.A., Bastian, A.J., 2013. Repeated split-belt treadmill training improves poststroke step length asymmetry. Neurorehabil. Neural Repair 27, 460–468. Roemmich, R.T., Nocera, J.R., Stegemöller, E.L., Hassan, A., Okun, M.S., Hass, C.J., 2014. Locomotor adaptation and locomotor adaptive learning in Parkinson's disease and normal aging. Clin. Neurophysiol. 125, 313–319. Roewer, B.D., Di Stasi, S.L., Snyder-Mackler, L., 2011. Quadriceps strength and weight acceptance strategies continue to improve two years after anterior cruciate ligament reconstruction. J. Biomech. 44, 1948–1953. Roper, J.A., Terza, M.J., Tillman, M.D., Hass, C.J., 2016. Adaptation strategies of individuals with anterior cruciate ligament reconstruction. Orthop. J. Sports Med. 4 (2). http://dx. doi.org/10.1177/2325967115627611. Vasudevan, E.V., Bastian, A.J., 2010. Split-belt treadmill adaptation shows different functional networks for fast and slow human walking. J. Neurophysiol. 103, 183–191. White, K., Di Stasi, S.L., Smith, A.H., Snyder-Mackler, L., 2013. Anterior cruciate ligamentspecialized post-operative return-to-sports (ACL-SPORTS) training: a randomized control trial. BMC Musculoskelet. Disord. 14, 108.