- Email: [email protected]
Stimulation site and phase modulation of the withdrawal reflex during gait initiation
Clinical Neurophysiology xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Stimulation site and phase modulation of the withdrawal reflex during gait initiation Miguel A. Richard a, Erika G. Spaich a,⇑, Mariano Serrao b, Ole K. Andersen a a b
Integrative Neuroscience group, Center for Sensory-Motor Interaction (SMI), Aalborg University, Denmark Department of Medico-Surgical Sciences and Biotechnologies, Sapienza University of Rome, Italy
a r t i c l e
i n f o
Article history: Accepted 18 January 2015 Available online xxxx Keywords: Gait initiation Nociceptive withdrawal reflex Site modulation Phase modulation
h i g h l i g h t s During gait initiation the nociceptive withdrawal reflex is modulated by posture and stimulation site. The nociceptive withdrawal reflex responses are largest during heel off and after stimulation of the
arch of the foot. The nociceptive withdrawal reflex modulation followed a functional principle, which may be
exploited in rehabilitation of the gait initiation process.
a b s t r a c t Objective: To investigate how the nociceptive withdrawal reflex (NWR) is modulated during gait initiation. Methods: The NWR was elicited in ten subjects using electrical stimulation at four sites in the right foot during symmetrical stance (50% of body weight on each foot) or while performing the first step during gait initiation: either during heel off (HO, 20% of body load on the starting leg) or heel contact (HC, 80% of body load on the starting leg in the first step). Kinematics and EMG responses from major muscles of the ipsilateral leg were recorded. Results: The NWR was modulated by stimulation site in all muscles except Soleus. The NWR responses elicited after stimulation of the arch were significantly larger than those evoked at all other sites in Tibialis Anterior, Rectus Femoris, and Vastus Lateralis. At the hip joint, the largest flexion was obtained during HO, whereas the smallest flexion was observed during HC, both following stimulation on the arch of the foot. Conclusions: The NWR responses were modulated to maintain balance and continue the development of the gait initiation process. Significance: The NWR modulation followed a functional principle, which might allow a functional use in rehabilitation strategies. Ó 2015 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction Muscular activity, neural mechanisms, and biomechanical forces are highly integrated in the gait initiation process (GIP) (Mann et al., 1979). In this complex process, forces generated by the legs serve to limit postural perturbations and provide forward propulsion. However, in patients with Parkinson’s disease (PD) or stroke, this process is sometimes disrupted (Hesse et al., 1997; Hass et al., 2005). In particular, people who suffered a stroke often ⇑ Corresponding author. Tel.: +45 9940 7462. E-mail address: [email protected] (E.G. Spaich).
have problems generating force in the starting leg and difficulties to load the limbs symmetrically during gait initiation (Brunt et al., 1995). As a consequence, gait initiation forces and momentums must be generated by the stance leg (Brunt et al., 1995) causing balance problems and difficulties when transitioning from quiet stance to steady-state dynamic gait. Functional electrical stimulation can be used to obtain muscle contractions (see review (Lyons et al., 2002)). Electrical pulses are used to activate motor nerves that generate an action potential leading to direct muscle activation, allowing the possibility of controlling paralyzed muscles (Fuhr et al., 2008). Electrical stimulation can also be used to evoke reflexes, like the nociceptive withdrawal
http://dx.doi.org/10.1016/j.clinph.2015.01.019 1388-2457/Ó 2015 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
2
M.A. Richard et al. / Clinical Neurophysiology xxx (2015) xxx–xxx
reflex (NWR) (Spaich et al., 2006). The NWR can be used to activate a synergistic set of muscles in a more functional manner than by electrical stimulation of individual motor nerves and to activate muscles that are difficult to access by surface electrodes like the hip flexors. The NWR has been successfully used to evoke ankle dorsiflexion during the swing phase by afferent stimulation in stroke individuals (Quintern et al., 2003). The reflex produces also flexion of the hip and knee during gait in hemiparetic patients when elicited appropriately (Quintern et al., 2003; Spaich et al., 2006). Swinging of the leg can be achieved even in subjects with spinal cord injury using the NWR (Nicol et al., 1995). In symmetrical standing, larger reflexes are obtained at higher degrees of unloading (Serrao et al., 2012b). In the GIP, as soon as the starting leg is unloaded and prior to any movement, the excitability of the NWR in hip and knee flexor muscles is also increased (Serrao et al., 2012a). Previous studies have been carried out during rhythmic walking (Spaich et al., 2004) and during symmetrical standing (Andersen et al., 2003) to characterize stimulation site dependencies for the NWR during these two stable, steady-state conditions. However, an understanding of the stimulation site and phase dependencies during the GIP is lacking. The GIP is a biomechanically relatively unstable transition from standing to walking and hence the influence on the spinal reflex organization in this phase is not known, e.g. is forward propulsion, balance control or effective limb withdrawal governing the net reaction to a stimulus perturbation applied at different skin sites? A better understanding of the behavior of the NWR during the GIP is needed to be able to administrate an electrical stimulation strategy to help patients with neuromuscular pathologies during gait initiation (Berardelli et al., 2001). The purpose of the present work was to explore the behavior of the NWR during the transition from standing to walking in healthy subjects. The kinematics and electromyographic activity (EMG) of five muscles in the starting leg were examined following stimulation of the sole of the foot during gait initiation. 2. Materials and methods 2.1. Participants Ten healthy volunteers (8 males, 2 females, age range: 19–31 years). Exclusion criteria: left-leg dominance or history of neuromuscular or osteoarticular disorders. The protocol of the study was approved by the local ethical committee (case number VN2007-0026) and was in accordance with The Declaration of Helsinki. All volunteers gave their written informed consent before participating in the study.
painful or painful. When the stimulus was non-painful the intensity was increased until it was painful, then, the stimulus intensity was decreased until it was non-painful. This procedure was repeated three times and the pain threshold was determined as the average of the three painful intensities. Subsequently, the stimulus intensities were obtained as a multiple of the pain thresholds detected at each electrode site. The multiplication factor was the same for all sites and was determined based on the evoked NWR with the subjects in symmetrical standing posture to ensure a clear and stable reflex. 2.3. Outcome measures 2.3.1. EMG responses EMG was acquired from tibialis anterior (TA), soleus (SOL), vastus lateralis (VL), rectus femoris (RF), and biceps femoris (BF) of the ipsilateral leg. For VL, RF, and BF a single differential configuration was used to record the EMG signals. To minimize crosstalk contamination, EMG from TA and SOL was acquired using a double differential configuration (Frahm et al., 2012). The recordings were obtained using surface electrodes (Neuroline 720, AMBU, Denmark), amplified, band-pass filtered (0.5–500 Hz, 2nd order), sampled at 2 kHz, displayed, and stored for further analysis. 2.3.2. Kinematic responses Three goniometers (Biometrics Ltd., typeSG150 and SG110/A, Gwent, UK) were mounted on the lateral side of the ankle, knee, and hip joints of the right leg to record movements in the sagittal plane. The goniograms were sampled at 2 kHz, displayed, and stored together with the EMG recordings. A force plate (AMTI, type OR6-7, USA) was used to measure the body load of the starting leg and to detect displacement of the center of pressure (CoP) during gait initiation. 2.4. Postures The subjects were stimulated on the right foot during symmetrical standing or while performing the first step during gait initiation. The specific postures were: Symmetric standing, with both feet inside the force plate; 50% of body load on each foot (ST). Heel off (HO), with 20% of the body load on the starting leg, starting to walk with the right leg. Heel contact (HC), with 80% of the body load on the starting leg during the first contact with the ground; starting to walk with the right leg. 2.5. Experimental procedure
2.2. Evoking the nociceptive withdrawal reflex The nociceptive withdrawal reflex was elicited by transcutaneous electric stimulation delivered in random order to three sites in the sole of the right foot: the third metatarsophalangeal joint (forefoot), the medial arch of the foot (arch), and the plantar side of calcaneus (heel), and one site on the posterior side of calcaneus (post-heel). The stimulation was delivered through selfadhesive electrodes (AMBU Neuroline 700, Denmark). A reference electrode (7 10-cm electrode, Pals, Axelgaard Ltd., USA) was placed on the dorsal aspect of the foot (Emborg et al., 2009). Each stimulus consisted of a constant current pulse burst of five individual 1-ms pulses delivered at 200 Hz. Pain thresholds for the four stimulation sites were determined with the volunteers in seated position, using a staircase method consisting of a series of increasing and decreasing stimuli. After each stimulus the volunteers rated the evoked sensation as non-
Before starting the actual reflex recordings, the subjects were instructed on how to maintain the different postures helped by visual feedback. Based on the force plate measures, the visual feedback consisted of a small square on a computer screen depicting the CoP. Additionally, the square changed color according to the weight load from red (0% of body weight) to blue (100% of body weight), but at 50% of the body weight of the subject, the square’s color changed to black in order to be easily distinguishable. The screen to visualize the feedback was placed at a comfortable height in front of the subject (Fig. 1). At the beginning of the experiment, subjects stood upright and barefoot at the center of the force plate, with their feet parallel and the inner edges of their heels approximately 20 cm apart in order to measure body weight and to set the moving square to zero position and blue color (100% of body load). To obtain the ST posture, the subjects were requested to keep the moving square within an area on the screen that corresponded
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
3
M.A. Richard et al. / Clinical Neurophysiology xxx (2015) xxx–xxx
Table 1 Positions of the three main joints of the starting leg at the moment when the stimulation was delivered. For hip and knee, positive values are flexion and for ankle, values larger than 90 degrees are dorsiflexion. Mean ± standard error of the mean [degrees] Symmetrical standing
Hip Knee Ankle
1.04 ± 2.07 6.05 ± 3.20 88.02 ± 2.03
Heel off
Hip Knee Ankle
6.52 ± 2.47 17.33 ± 3.57 90.21 ± 2.68
Heel contact
Hip Knee Ankle
9.78 ± 2.16 13.92 ± 2.81 90.50 ± 1.77
Fig. 1. Experimental setup.
to a maximum displacement of the CoP of ±4 cm. In order to get the HO posture, the subjects were asked to keep only one foot on the force plate, load the starting leg with 50% of their body weight (square in black color) and then initiate the first step off the force plate. For the HC posture, the subjects were instructed to stand upright outside of the force plate, load, as far as possible, each leg equally and take the first step onto the force plate. In these two last conditions, the subjects were asked to walk forward at a comfortable speed beginning with their right foot, and to stop after one step with the left leg. The position of the starting leg at the moment when the stimulus was delivered is presented in the Table 1. In the ST posture, the subjects were stimulated while keeping the CoP displacement within the limit of ±4 cm after a random waiting time ranging from 1 to 5 s. In the HC and HO postures, the stimulation was triggered by the body load on the force plate during the execution of the first step by the starting leg. There were 24 possible combinations for the different stimulation intensities (noxious and control), stimulation sites (forefoot, arch, heel and post-heel), and postures (ST, HO and HC). Each combination was presented once, in random order, within a block of 24 trials. After each block of 24 trials, the subjects rested for 5 min. At least 120 trials per subject (5 blocks) were recorded for all participants. Before the actual recordings were performed, subjects were thoroughly familiarized with the stimulations to avoid eliciting a startle response in conjunction with the NWR. This comprised delivering several stimulations randomly to the different sites and postures, until the subjects felt confident with the sensations produced by the stimulation. 2.6. Data analysis The EMG recordings were band-pass filtered (Butterworth, 10– 500 Hz, fourth order, no phase lag). In the trials with stimulation, the root mean square (RMS) of the EMG amplitude in the window 60–220 ms after the stimulus onset was calculated. In the trials without stimulation, control trials, the RMS of the EMG amplitude in the corresponding window was calculated to determine the background activity. The reflex response was calculated as the difference between the RMS of the EMG in stimulated steps and the corresponding control steps. To decrease the inter-subject variability (Yang and Winter, 1984), the background activity and the EMG reflex responses in each muscle were normalized to the grand mean of the RMS values calculated previously in all control trials of the respective muscle for all three conditions. Finally, they were averaged across blocks for every site and posture. The signals from the goniometers were low-pass filtered (Butterworth, 25 Hz, sixth order, no phase lag). The kinematic
reflex response for the hip and knee was calculated as the peak angle change between the stimulated condition and the control condition. The analysis was assessed in the interval 110–470 ms post-stimulation. The time window was selected by visual inspection of all the data from the goniometers and was long enough to contain the reflex response. Unlike the hip and knee joints, the ankle joint changed the direction of its movement in the window of analysis. The movement changed from plantarflexion to dorsiflexion and thus, the kinematic response of the ankle was assessed in two windows in the interval 110–470 ms poststimulation. The split-point was set as the time point of peak plantarflexion in the 110–470 ms reflex analysis window (Spaich et al., 2006). 2.7. Statistics Repeated-measures ANOVA was applied to the data with factors ‘‘stimulation site’’ and ‘‘posture’’. Student Newman–Keuls statistics (SNK) were used for post hoc pair-wise comparisons when a main effect or an interaction was detected. P < 0.05 was regarded as statistically significant and data are presented as mean and standard error of the mean. 3. Results All subjects were able to complete the experiment and described the evoked sensation as sharp, pinpricking, and localized at the different stimulation sites. The stimulation did not evoke any sensation originating from the dorsum of the foot. The multiplication factor for the stimulation intensities was 1.4 for nine subjects and 1.5 for one subject resulting in stimulation intensities of 24.8 ± 1.3 mA for the forefoot, 18.2 ± 1.2 mA for the arch of the foot, 39.4 ± 3.1 mA for the heel, and 24.1 ± 1.7 mA for the post-heel stimulation site. Fig. 2 shows the NWR responses in the TA muscle following electrical stimulation. 3.1. EMG reflex responses The evoked reflex response was significantly modulated by stimulation site (RM ANOVA main effect) for all muscles, except for SOL (TA, F(3, 27) = 17.51, P < 0.001; BF, F(3, 27) = 4.36, P < 0.05; RF, F(3, 27) = 6.48, P < 0.01, and VL (F(3, 27) = 7.28, P < 0.001). The post hoc analysis showed that the largest TA reflex responses (Fig. 3A) were elicited after stimulating the arch of the foot, compared to the forefoot, heel, and post-heel stimulation sites (SNK, P values <0.05). Similar results were found for the RF (SNK, P < 0.05) and the VL muscles (SNK, P < 0.05) (Fig. 3D–E). In the BF muscle, the post hoc analysis showed that the largest reflex response was obtained by stimulation of the arch of the foot
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
4
M.A. Richard et al. / Clinical Neurophysiology xxx (2015) xxx–xxx
Fig. 2. TA EMG response to electrical stimulation at specific sites and postures. In the background, in thin line, the control signal is plotted. The plots are the mean of the 5 repetitions from subject 8. The vertical full line indicates the stimulation onset. The stippled lines delimit the analysis window.
(SNK, P < 0.05) compared to the heel and post-heel sites (Fig. 3C). The analysis of the EMG responses revealed no significant interactions between site and posture in any of the muscles. Posture significantly modulated the evoked reflex responses in TA (F(2, 18) = 6.05, P < 0.01), SOL (F(2, 18) = 6.83, P < 0.01) and VL (F(2, 18) = 8.42, P < 0.01) (Fig. 3), while no significant modulation was observed in BF and RF. In SOL muscle, results showed that the size of the evoked reflex response at heel contact (mean 0.15 ± 0.19) was not different from zero; post hoc pairwise comparisons showed that responses evoked during symmetrical standing (mean 0.34 ± 0.14) and heel off (mean 0.32 ± 0.09) were not different from each other (SNK, P = 0.89) (Fig. 3B). A tendency for a suppressive response was observed when the stimulation was applied at the forefoot site at heel contact (Fig. 3B) but this response was not different from zero. In the TA muscle, the post hoc analysis showed that the size of the evoked reflex response was smallest during symmetrical standing (mean 1.55 ± 0.40), and that there were no significant differences between the heel off (mean 2.30 ± 0.46) and heel contact (mean 2.24 ± 0.37) postures (Fig. 3A). Regarding the VL muscle, the heel contact posture showed the largest evoked reflex response (mean 4.74 ± 1.30), while no significant differences were observed between stimulation at heel off (mean 1.85 ± 0.54) and symmetrical standing (mean 2.37 ± 0.71) (Fig. 3C). 3.2. Kinematics At the hip joint, flexion was the primary reflex response with posture and site as main effects (F(2, 18) = 7.24, P < 0.01; F(3, 27) = 6.24, P < 0.01). Moreover a significant posture and site interaction (F(6, 54) = 2.99, P < 0.05) was found. The post hoc analysis revealed that NWR responses elicited after stimulation of the arch during heel off were significantly larger than those evoked at all other posture and site combinations (SNK, P values <0.01). Additionally, at heel contact the stimulation at the forefoot, heel, and post-heel sites did not produce a response different from zero (95% CI included zero, Fig. 4B). The NWR responses after stimulation of the arch during heel contact were significantly smaller than those evoked at all other posture and site combinations (SNK, P < 0.01). During symmetric standing, the NWR responses after stimulation in the arch and post-heel were larger (SNK, P values <0.01) than after stimulation on the forefoot and heel. At the knee joint, the main reflex response was flexion modulated by stimulation site (F(3, 27) = 5.62, P < 0.01). Hence, the largest knee response was evoked by stimulation on the arch of the foot (SNK P < 0.05) (Fig. 4A). There were no significant differences between responses evoked by stimulation on the forefoot, the heel, and post-heel sites. No interaction between factors was found.
At the ankle joint, in the first analysis window (110 ms to maximum plantarflexion), site significantly modulated the degree of plantar flexion (RM ANOVA, main effect, F(3, 27) = 10.00, P < 0.001). The post hoc analysis showed that stimulation of the arch of the foot and the post-heel sites resulted in the largest plantarflexion reduction (SNK P < 0.01). The posture also significantly modulated the early ankle response (RM ANOVA, main effect, F(2, 18) = 9.27, P < 0.01). At heel contact, stimulation on the arch, heel, and post-heel sites did not produce a response that was significantly different from the unperturbed condition (95% CI included zero degrees). Increased plantarflexion was observed at heel contact (SNK P < 0.01) (Fig. 4C) for the forefoot site. The largest reduction in plantarflexion was observed at heel off (SNK P < 0.05). No interaction between factors was found. In the second analysis window (maximum plantarflexion to 670 ms) (Fig. 4D), there was a significant posture and site interaction (F(6, 54) = 2.96, P < 0.05). Stimulation during symmetrical standing evoked the largest increase of dorsiflexion, no matter which site was stimulated (SNK P values <0.001). Stimulation at heel off resulted in reduced dorsiflexion when stimulating at the heel (SNK P < 0.05) and post-heel sites (SNK P < 0.01) compared to all other sites and postures; at the same time, stimulation at the forefoot and arch sites evoked a response that was not different from zero. Stimulation during heel contact evoked an increase of dorsiflexion no matter which site was stimulated (SNK P values <0.001). 4. Discussion A single experimental session was carried out where NWR reflex responses were assessed in order to determine how the NWR reflex is modulated during gait initiation. Electrical stimulation was applied to ten healthy subjects on four sites of the foot and in three different postures. Both, stimulation site and posture modulated the evoked reflexes. Regarding stimulation site, stimuli delivered on the arch of the foot resulted in larger reflex responses compared to heel and post-heel stimulation sites for all muscles, except for SOL. Similarly, the largest flexion of the knee and hip was obtained by stimulation of the arch of the foot. In relation to postures, the largest reflexes were obtained at HC and HO in the TA muscle, while the largest responses were obtained at ST and HO in the SOL muscle. 4.1. Methodological considerations The gait initiation process is composed of four sub-phases (Breniere et al., 1981). The first two sub-phases form the preparatory phase (PP) that goes from initial movement of the COP of the
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
M.A. Richard et al. / Clinical Neurophysiology xxx (2015) xxx–xxx
5
Fig. 3. EMG responses to electrical stimulation at the specified stimulation postures and sites. The normalized background EMG activity is shown as empty circles and solid lines. Values are mean and standard error of the mean. Significant differences between groups are indicated (⁄P < 0.05; ⁄⁄P < 0.01). Acronyms: TA, Tibialis anterior; SOL, Soleus; BF, Biceps Femoris; RF, Rectus Femoris; VL, Vastus Lateralis; ST, Symmetrical standing; HO, Heel off; HC, Heel contact.
subject (Winter, 1995) until toe-off of the starting foot. The next two sub-phases form the stepping phase (SP). This phase of the gait initiation process is divided in the single support sub-phase that finalizes with the heel contact of the starting foot (Serrao et al., 2012a) and is followed by the double support sub-phase. In the present study, three different postures that represent events of the gait initiation process were selected. The first posture, standing, is the starting point of the gait initiation process, where the loading of the starting leg was maintained at 50% of the total body load. This posture was included because it requires the
coordination of all postural muscles to maintain equilibrium. The second posture, heel off corresponds to an instant before the end of the PP. A threshold of 20% of the body weight was selected as trigger for the stimulation, since at that moment most of the body weight is supported by the standing leg, allowing lifting of the starting foot. The third posture was heel contact which corresponds with the end of the single support sub-phase. In order to ensure a clear HC, the stimulation was triggered when 80% of the body weight was supported by the starting leg. Thus, this experimental setup allowed to investigate the loading/unloading of the
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
6
M.A. Richard et al. / Clinical Neurophysiology xxx (2015) xxx–xxx
Fig. 4. Kinematic responses at the knee, hip, and ankle joints induced by electrical stimulation at specified postures and sites. The error bars indicate standard error of the mean. Significant results regarding posture modulation are shown in panel C, and posture and site interactions in panels B and D (⁄P < 0.05; ⁄⁄P < 0.01). Acronyms: ST, symmetrical standing; HC, Heel contact; HO, Heel off.
leg during gait initiation in a natural motor context. Moreover, the motor behavior of the joints is expected to be the same at each posture: hip and knee flexion and ankle dorsiflexion at HO (20% BW) and hip and knee flexion (with ankle in neutral position) at HC (80% BW). With the three different postures, a large variation in the pressure on the stimulation electrodes must be expected, which could explain, in principle, some of the observed differences in responses. However, recent evidence indicates that when stimulating the sole of the foot, Ad fibers were activated within the most superficial 100 lm of the dermis while Ab fibers were activated throughout the dermis (Frahm et al., 2013a). Furthermore, the current flow between the stimulating and reference electrodes happened primarily through the dermis, with much smaller current densities in the fat, muscle, and bone layers (Frahm et al., 2013b) suggesting that even in both extreme conditions (fully loading/unloading of the foot) the current flow will most likely cause depolarization of the same fibers. In this study, the stimulation intensity was 1.4–1.5 times the pain threshold, which was clearly perceived as painful. Strong, painful electrical stimulation will inherently activate Ad as well as Ab afferents eliciting a withdrawal reflex response. Andersen (2007) has shown that from 60 ms after the stimulus onset, the EMG can contain components mediated by Ad afferents, constituting therefore a nociceptive withdrawal reflex response. However, this response could also be mediated by non-nociceptive inputs from proprioceptors of the foot, muscle and cutaneous afferents which are known to converge in the spinal withdrawal reflex arc.
4.2. Stimulation site modulation Previous studies investigating the relationship between gait and the NWR found a modulation of the reflex amplitude depending on where the electrical stimulation was delivered (Granat et al., 1993; Spaich et al., 2004). This modulation was primarily observed in the TA muscle during symmetrical stance and during steady gait (Spaich et al., 2004; Emborg et al., 2009), and also in BF, and VL muscles (Serrao et al., 2012b). However, it has not previously been investigated how the NWR is site modulated during gait initiation. The results of this study showed that the stimulation site modulated the amplitude of the NWR in BF, RF, TA, and VL muscles during gait initiation. Specifically, stimulations of the arch of the foot evoked the largest reflexes in all muscles, except SOL, and in the knee joint, independent of the posture. Stimulation delivered at the proximal part of the sole of the foot evoked the smallest NWR response in the BF, RF, TA, and VL muscles. This is in accordance with previous studies for the BF, TA, and VL muscles (Spaich et al., 2004; Serrao et al., 2012b). Similarly, the smallest flexion of the knee joint was evoked by stimulation at the proximal part of the sole of the foot. These two patterns of stimulation-site modulation reflect that electrical stimulation was applied within the reflex receptive field (RRF) of the different muscles, defined as the area of the skin from which a reflex can be elicited (Schouenborg, 2002; Andersen et al., 2005). Hence, stimulation in the medial, distal part of the sole of the foot mainly activated BF, RF, TA, and VL (Sonnenborg et al., 2000; Spaich et al., 2004) associated with hip and knee flexion and reduced plantarflexion in the initial part of the ankle response,
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
M.A. Richard et al. / Clinical Neurophysiology xxx (2015) xxx–xxx
while stimulation of the proximal part of the sole of the foot evoked a NWR mainly in BF, RF, and VL. Moreover, there were no significant differences in NWR amplitude in the SOL muscle across stimulation sites and the angle changes in the second part of the ankle response did not show modulation related to the site of stimulation. The reflex receptive fields for antagonistic muscles overlap, e.g. both VL/RF and BF were activated simultaneously when the stimulus was applied to the proximal part of the sole of the foot. Obviously, the net kinematic response reflects the balance between activation of the antagonists and most likely the function of the cocontraction is to provide additional stability to the joint during the withdrawal movement. These results support the idea that the NWR response to stimulation during gait initiation was highly related to the location of the stimulus in accordance with the RRF of the muscle, and consequently, activated a neuronal pool that triggered a coordinated movement to withdraw the limb away from the stimulation. In the present study, this stimulation site dependency at muscle level was independent of the degree of load on the leading limb (no ANOVA interaction). However, the results showed an interaction at the hip joint, meaning that the hip flexion was largest when the stimulation was delivered in the arch of the foot at HO, but the response was suppressed at HC when the stimulation was delivered in the other sites. This lack of NWR response has likely a functional meaning since it avoids postural instability and collapsing during heel-contact during the gait initiation process. 4.3. Posture modulation The modulation of the NWR response related to different postures has been studied in animals (Schotland et al., 1989) and in humans during standing, rhythmic walking (Spaich et al., 2004), and in the transition from gait initiation to steady-state walking (Park et al., 2009). In general, previous studies dealing with the relationship between gait and the NWR showed a modulation of the reflex amplitude depending on which moment of the gait process the stimulation was delivered (Spaich et al., 2004, 2006; Emborg et al., 2009; Serrao et al., 2012a). However, the present study also revealed interactions between postures and stimulation sites in the GIP for the hip and ankle measurements but not for the EMG responses. At HO, where an increase of the response was expected for BF and RF (Serrao et al., 2012a), our results showed that neither BF nor RF reflex responses seem to be posture dependent. The modulation on TA can be related to supporting the development of the displacement of the CoP towards the stance leg and the forward movement of the body. The largest TA EMG reflex responses were detected during the two dynamic conditions (HC and HO) compared to the symmetrical stance posture. This suggests that the large TA reflex response at HO might be related to the preactivation of this muscle normally associated with the production of the swing phase. This pre-activation might facilitate the foot withdrawal (Mann et al., 1979). At HC, the VL reflex response was largest contrasting with previous results (Serrao et al., 2012a) where no differences in VL responses during the sub-phases of gait initiation were found; other studies reported however a facilitation of cutaneous reflexes in VL in the transition from swing to stance (Crenna and Frigo, 1984; Zehr et al., 1998) as a mechanism to decrease the possibility of leg collapse (Zehr et al., 1998). When the subject was standing, an excitatory response of the NWR was also found in the quadriceps muscle (Decchi et al., 1997), but in the present study the VL reflex response was almost the same at HO compared to symmetrical standing. As expected, the main response at the hip joint was flexion (Emborg et al., 2009). The largest reflex was obtained during HO
7
when the stimulation was delivered in the arch of the foot. This response is related to the degree of unloading of the starting leg in line with prior studies in symmetrical stance (Serrao et al., 2012b). Additionally, NWR responses were smallest during HC in order to prevent loss of balance and destabilization in agreement with observations during walking (Spaich et al., 2004). Therefore, the results suggest that the flexion of the hip and knee are the main movements involved in the withdrawal of the lower limb during gait initiation. Unlike the knee where the NWR response is only modulated by stimulation site, at the hip joint the response is mainly modulated by the current posture. In this regard, the interaction found at the hip joint suggests that the central nervous system modulates the hip kinematics according to the motor context (HO, HC). Thus, the central nervous system seems to attribute an important role to the hip joint in the first step of the gait cycle. 4.4. Functional modulation The present study indicates that the NWR response elicited by localized stimulation of the foot during gait initiation depends on the stimulation site and on the posture during gait initiation. Our findings suggest that stimulation on the arch of the foot elicited the largest reflexes regardless of the posture, and in general the reflex responses were larger during heel off. Larger reflexes during heel off suggest a functional modulation of the reflex gain as the evoked response supported the GIP by facilitating the flexion of the joints. This is in agreement with prior observations for the knee and hip joints during steady state gait (Spaich et al., 2004) and indicate a priority to maintain balance and ensure continued forward propulsion (Rietdyk and Patla, 1998). Moreover, at heel off, the center of pressure has been shifted to the stance limb and thereby the initial inhibition of the NWR as part of the preparatory balance adjustments during steady state standing (McIlroy et al., 1999) was reduced. To what degree the reflex gain modulation is a result of spinal motor programmes (Zehr and Duysens, 2004) or is a result of descending motor commands during the relatively unstable gait initiation process remains unclear. The functional, modular organization of the NWR might be used in rehabilitation strategies involving functional electrical stimulation. This might, in particular, be of relevance for patients with hypokinetic gait disorders that could benefit from reflexes elicited with proper timing and at the optimal stimulation site resulting in functional support of the gait initiation process. Acknowledgement This work was supported by the Svend Andersen foundation. Conflict of interest: Authors declare that there are no financial or other relationships that might lead to a conflict of interests. References Andersen OK. Studies of the organization of the human nociceptive withdrawal reflex. Focus on sensory convergence and stimulation site dependency. Acta Physiol 2007;189:1–35. Andersen OK, Sonnenborg F, Matjacˇic´ Z, Arendt-Nielsen L. Foot-sole reflex receptive fields for human withdrawal reflexes in symmetrical standing position. Exp Brain Res 2003;152:434–43. Andersen OK, Spaich EG, Madeleine P, Arendt-Nielsen L. Gradual enlargement of human withdrawal reflex receptive fields following repetitive painful stimulation. Brain Res 2005;1042:194–204. Berardelli A, Rothwell JC, Thompson PD, Hallett M. Pathophysiology of bradykinesia in parkinson’s disease. Brain 2001;124:2131–46. Breniere Y, Do MC, Sanchez J. A biomechanical study of the gait initiation process. J Biophys Med Nucl 1981;5:197–205. Brunt D, Vander Linden DW, Behrman AL. The relation between limb loading and control parameters of gait initiation in persons with stroke. Arch Phys Med Rehabil 1995;76:627–34. Crenna P, Frigo C. Evidence of phase-dependent nociceptive reflexes during locomotion in man. Exp Neurol 1984;85:336–45.
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
8
M.A. Richard et al. / Clinical Neurophysiology xxx (2015) xxx–xxx
Decchi B, Zalaffi A, Spidalieri R, Arrigucci U, Di Troia AM, Rossi A. Spinal reflex pattern to foot nociceptive stimulation in standing humans. Electroencephalogr Clin Neurophysiol Electromyogr Mot Control 1997;105:484–9. Emborg J, Spaich EG, Andersen OK. Withdrawal reflexes examined during human gait by ground reaction forces: site and gait phase dependency. Med Biol Eng Comput 2009;47:29–39. Frahm KS, Jensen MB, Farina D, Andersen OK. Surface EMG crosstalk during phasic involuntary muscle activation in the nociceptive withdrawal reflex. Muscle Nerve 2012;46:228–36. Frahm KS, Mørch CD, Grill WM, Lubock NB, Hennings K, Andersen OK. Activation of peripheral nerve fibers by electrical stimulation in the sole of the foot. BMC Neurosci 2013a;8:114–6. Frahm KS, Mørch CD, Grill WM, Andersen OK. Experimental and model-based analysis of differences in perception of cutaneous electrical stimulation across the sole of the foot. Med Biol Eng Comput 2013b;51:999–1009. Fuhr T, Quintern J, Riener R, Schmidt G. Walking with WALK! IEEE Eng Med Biol Mag 2008;27:38–48. Granat MH, Heller BW, Nicol DJ, Baxendale RH, Andrews BJ. Improving limb flexion in FES gait using the flexion withdrawal response for the spinal cord injured person. J Biomed Eng 1993;15:51–6. Hass CJ, Waddell DE, Fleming RP, Juncos JL, Gregor RJ. Gait initiation and dynamic balance control in Parkinson’s disease. Arch Phys Med Rehabil 2005;86:2172–6. Hesse S, Reiter F, Jahnke M, Dawson M, Sarkodie-Gyan T, Mauritz KH. Asymmetry of gait initiation in hemiparetic stroke subjects. Arch Phys Med Rehabil 1997;78:719–24. Lyons GM, Sinkjær T, Burridge JH, Wilcox DJ. A review of portable FES-based neural orthoses for the correction of drop foot. IEEE Trans Neural Syst Rehabil Eng 2002;10:260–79. Mann RA, Hagy JL, White V, Liddell D. The initiation of gait. J Bone JT Surg Ser A 1979;61:232–9. McIlroy WE, Bent LR, Potvin JR, Brooke JD, Maki BE. Preparatory balance adjustments precede withdrawal response to noxious stimulation in standing humans. Neurosci Lett 1999;267:197–200. Nicol DJ, Granat MH, Baxendale RH, Tuson SJM. Evidence for a human spinal stepping generator. Brain Res 1995;684:230–2. Park S, Choi H, Ryu K, Kim S, Kim Y. Kinematics, kinetics and muscle activities of the lower extremity during the first four steps from gait initiation to the steadystate walking. J Mech Sci Technol 2009;23:204–11.
Quintern J, Krewer C, Husemann B, Hartmann E, Heller S, Müller F, et al. Controlled clinical study of stimulation of flexor reflex afferents for gait rehabilitation in patients with hemiplegia. In: Gantchev N, editor. From basic motor control to functional recovery III. University Press; 2003. p. 240. Rietdyk S, Patla AE. Context-dependent reflex control: some insights into the role of balance. Exp Brain Res 1998;119:251–9. Schotland JL, Lee WA, Rymer WZ. Wipe and flexion withdrawal reflexes display different EMG patterns prior to movement onset in the spinalized frog. Exp Brain Res 1989;78:649–53. Schouenborg J. Modular organisation and spinal somatosensory imprinting. Brain Res Rev 2002;40:80–91. Serrao M, Ranavolo A, Andersen OK, Conte C, Don R, Cortese F, et al. Adaptive behaviour of the spinal cord in the transition from quiet stance to walking. BMC Neurosci 2012a;13:80. Serrao M, Spaich EG, Andersen OK. Modulating effects of bodyweight unloading on the lower limb nociceptive withdrawal reflex during symmetrical stance. Clin Neurophysiol 2012b;123:1035–43. Sonnenborg FA, Andersen OK, Arendt-Nielsen L. Modular organization of excitatory and inhibitory reflex receptive fields elicited by electrical stimulation of the foot sole in man. Clin Neurophysiol 2000;111:2160–9. Spaich EG, Arendt-Nielsen L, Andersen OK. Modulation of lower limb withdrawal reflexes during gait: a topographical study. J Neurophysiol 2004;91:258–66. Spaich EG, Hinge HH, Arendt-Nielsen L, Andersen OK. Modulation of the withdrawal reflex during hemiplegic gait: effect of stimulation site and gait phase. Clin Neurophysiol 2006;117:2482–95. Winter DA. Human balance and posture control during standing and walking. Gait Posture 1995;3:193–214. Yang JF, Winter DA. Electromyographic amplitude normalization methods: improving their sensitivity as diagnostic tools in gait analysis. Arch Phys Med Rehabil 1984;65:517–21. Zehr EP, Duysens J. Regulation of arm and leg movement during human locomotion. Neuroscientist 2004;10:347–61. Zehr EP, Stein RB, Komiyama T. Function of sural nerve reflexes during human walking. J Physiol 1998;507:305–14.
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Stimulation site and phase modulation of the withdrawal reflex during gait initiation Miguel A. Richard a, Erika G. Spaich a,⇑, Mariano Serrao b, Ole K. Andersen a a b
Integrative Neuroscience group, Center for Sensory-Motor Interaction (SMI), Aalborg University, Denmark Department of Medico-Surgical Sciences and Biotechnologies, Sapienza University of Rome, Italy
a r t i c l e
i n f o
Article history: Accepted 18 January 2015 Available online xxxx Keywords: Gait initiation Nociceptive withdrawal reflex Site modulation Phase modulation
h i g h l i g h t s During gait initiation the nociceptive withdrawal reflex is modulated by posture and stimulation site. The nociceptive withdrawal reflex responses are largest during heel off and after stimulation of the
arch of the foot. The nociceptive withdrawal reflex modulation followed a functional principle, which may be
exploited in rehabilitation of the gait initiation process.
a b s t r a c t Objective: To investigate how the nociceptive withdrawal reflex (NWR) is modulated during gait initiation. Methods: The NWR was elicited in ten subjects using electrical stimulation at four sites in the right foot during symmetrical stance (50% of body weight on each foot) or while performing the first step during gait initiation: either during heel off (HO, 20% of body load on the starting leg) or heel contact (HC, 80% of body load on the starting leg in the first step). Kinematics and EMG responses from major muscles of the ipsilateral leg were recorded. Results: The NWR was modulated by stimulation site in all muscles except Soleus. The NWR responses elicited after stimulation of the arch were significantly larger than those evoked at all other sites in Tibialis Anterior, Rectus Femoris, and Vastus Lateralis. At the hip joint, the largest flexion was obtained during HO, whereas the smallest flexion was observed during HC, both following stimulation on the arch of the foot. Conclusions: The NWR responses were modulated to maintain balance and continue the development of the gait initiation process. Significance: The NWR modulation followed a functional principle, which might allow a functional use in rehabilitation strategies. Ó 2015 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction Muscular activity, neural mechanisms, and biomechanical forces are highly integrated in the gait initiation process (GIP) (Mann et al., 1979). In this complex process, forces generated by the legs serve to limit postural perturbations and provide forward propulsion. However, in patients with Parkinson’s disease (PD) or stroke, this process is sometimes disrupted (Hesse et al., 1997; Hass et al., 2005). In particular, people who suffered a stroke often ⇑ Corresponding author. Tel.: +45 9940 7462. E-mail address: [email protected] (E.G. Spaich).
have problems generating force in the starting leg and difficulties to load the limbs symmetrically during gait initiation (Brunt et al., 1995). As a consequence, gait initiation forces and momentums must be generated by the stance leg (Brunt et al., 1995) causing balance problems and difficulties when transitioning from quiet stance to steady-state dynamic gait. Functional electrical stimulation can be used to obtain muscle contractions (see review (Lyons et al., 2002)). Electrical pulses are used to activate motor nerves that generate an action potential leading to direct muscle activation, allowing the possibility of controlling paralyzed muscles (Fuhr et al., 2008). Electrical stimulation can also be used to evoke reflexes, like the nociceptive withdrawal
http://dx.doi.org/10.1016/j.clinph.2015.01.019 1388-2457/Ó 2015 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
2
M.A. Richard et al. / Clinical Neurophysiology xxx (2015) xxx–xxx
reflex (NWR) (Spaich et al., 2006). The NWR can be used to activate a synergistic set of muscles in a more functional manner than by electrical stimulation of individual motor nerves and to activate muscles that are difficult to access by surface electrodes like the hip flexors. The NWR has been successfully used to evoke ankle dorsiflexion during the swing phase by afferent stimulation in stroke individuals (Quintern et al., 2003). The reflex produces also flexion of the hip and knee during gait in hemiparetic patients when elicited appropriately (Quintern et al., 2003; Spaich et al., 2006). Swinging of the leg can be achieved even in subjects with spinal cord injury using the NWR (Nicol et al., 1995). In symmetrical standing, larger reflexes are obtained at higher degrees of unloading (Serrao et al., 2012b). In the GIP, as soon as the starting leg is unloaded and prior to any movement, the excitability of the NWR in hip and knee flexor muscles is also increased (Serrao et al., 2012a). Previous studies have been carried out during rhythmic walking (Spaich et al., 2004) and during symmetrical standing (Andersen et al., 2003) to characterize stimulation site dependencies for the NWR during these two stable, steady-state conditions. However, an understanding of the stimulation site and phase dependencies during the GIP is lacking. The GIP is a biomechanically relatively unstable transition from standing to walking and hence the influence on the spinal reflex organization in this phase is not known, e.g. is forward propulsion, balance control or effective limb withdrawal governing the net reaction to a stimulus perturbation applied at different skin sites? A better understanding of the behavior of the NWR during the GIP is needed to be able to administrate an electrical stimulation strategy to help patients with neuromuscular pathologies during gait initiation (Berardelli et al., 2001). The purpose of the present work was to explore the behavior of the NWR during the transition from standing to walking in healthy subjects. The kinematics and electromyographic activity (EMG) of five muscles in the starting leg were examined following stimulation of the sole of the foot during gait initiation. 2. Materials and methods 2.1. Participants Ten healthy volunteers (8 males, 2 females, age range: 19–31 years). Exclusion criteria: left-leg dominance or history of neuromuscular or osteoarticular disorders. The protocol of the study was approved by the local ethical committee (case number VN2007-0026) and was in accordance with The Declaration of Helsinki. All volunteers gave their written informed consent before participating in the study.
painful or painful. When the stimulus was non-painful the intensity was increased until it was painful, then, the stimulus intensity was decreased until it was non-painful. This procedure was repeated three times and the pain threshold was determined as the average of the three painful intensities. Subsequently, the stimulus intensities were obtained as a multiple of the pain thresholds detected at each electrode site. The multiplication factor was the same for all sites and was determined based on the evoked NWR with the subjects in symmetrical standing posture to ensure a clear and stable reflex. 2.3. Outcome measures 2.3.1. EMG responses EMG was acquired from tibialis anterior (TA), soleus (SOL), vastus lateralis (VL), rectus femoris (RF), and biceps femoris (BF) of the ipsilateral leg. For VL, RF, and BF a single differential configuration was used to record the EMG signals. To minimize crosstalk contamination, EMG from TA and SOL was acquired using a double differential configuration (Frahm et al., 2012). The recordings were obtained using surface electrodes (Neuroline 720, AMBU, Denmark), amplified, band-pass filtered (0.5–500 Hz, 2nd order), sampled at 2 kHz, displayed, and stored for further analysis. 2.3.2. Kinematic responses Three goniometers (Biometrics Ltd., typeSG150 and SG110/A, Gwent, UK) were mounted on the lateral side of the ankle, knee, and hip joints of the right leg to record movements in the sagittal plane. The goniograms were sampled at 2 kHz, displayed, and stored together with the EMG recordings. A force plate (AMTI, type OR6-7, USA) was used to measure the body load of the starting leg and to detect displacement of the center of pressure (CoP) during gait initiation. 2.4. Postures The subjects were stimulated on the right foot during symmetrical standing or while performing the first step during gait initiation. The specific postures were: Symmetric standing, with both feet inside the force plate; 50% of body load on each foot (ST). Heel off (HO), with 20% of the body load on the starting leg, starting to walk with the right leg. Heel contact (HC), with 80% of the body load on the starting leg during the first contact with the ground; starting to walk with the right leg. 2.5. Experimental procedure
2.2. Evoking the nociceptive withdrawal reflex The nociceptive withdrawal reflex was elicited by transcutaneous electric stimulation delivered in random order to three sites in the sole of the right foot: the third metatarsophalangeal joint (forefoot), the medial arch of the foot (arch), and the plantar side of calcaneus (heel), and one site on the posterior side of calcaneus (post-heel). The stimulation was delivered through selfadhesive electrodes (AMBU Neuroline 700, Denmark). A reference electrode (7 10-cm electrode, Pals, Axelgaard Ltd., USA) was placed on the dorsal aspect of the foot (Emborg et al., 2009). Each stimulus consisted of a constant current pulse burst of five individual 1-ms pulses delivered at 200 Hz. Pain thresholds for the four stimulation sites were determined with the volunteers in seated position, using a staircase method consisting of a series of increasing and decreasing stimuli. After each stimulus the volunteers rated the evoked sensation as non-
Before starting the actual reflex recordings, the subjects were instructed on how to maintain the different postures helped by visual feedback. Based on the force plate measures, the visual feedback consisted of a small square on a computer screen depicting the CoP. Additionally, the square changed color according to the weight load from red (0% of body weight) to blue (100% of body weight), but at 50% of the body weight of the subject, the square’s color changed to black in order to be easily distinguishable. The screen to visualize the feedback was placed at a comfortable height in front of the subject (Fig. 1). At the beginning of the experiment, subjects stood upright and barefoot at the center of the force plate, with their feet parallel and the inner edges of their heels approximately 20 cm apart in order to measure body weight and to set the moving square to zero position and blue color (100% of body load). To obtain the ST posture, the subjects were requested to keep the moving square within an area on the screen that corresponded
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
3
M.A. Richard et al. / Clinical Neurophysiology xxx (2015) xxx–xxx
Table 1 Positions of the three main joints of the starting leg at the moment when the stimulation was delivered. For hip and knee, positive values are flexion and for ankle, values larger than 90 degrees are dorsiflexion. Mean ± standard error of the mean [degrees] Symmetrical standing
Hip Knee Ankle
1.04 ± 2.07 6.05 ± 3.20 88.02 ± 2.03
Heel off
Hip Knee Ankle
6.52 ± 2.47 17.33 ± 3.57 90.21 ± 2.68
Heel contact
Hip Knee Ankle
9.78 ± 2.16 13.92 ± 2.81 90.50 ± 1.77
Fig. 1. Experimental setup.
to a maximum displacement of the CoP of ±4 cm. In order to get the HO posture, the subjects were asked to keep only one foot on the force plate, load the starting leg with 50% of their body weight (square in black color) and then initiate the first step off the force plate. For the HC posture, the subjects were instructed to stand upright outside of the force plate, load, as far as possible, each leg equally and take the first step onto the force plate. In these two last conditions, the subjects were asked to walk forward at a comfortable speed beginning with their right foot, and to stop after one step with the left leg. The position of the starting leg at the moment when the stimulus was delivered is presented in the Table 1. In the ST posture, the subjects were stimulated while keeping the CoP displacement within the limit of ±4 cm after a random waiting time ranging from 1 to 5 s. In the HC and HO postures, the stimulation was triggered by the body load on the force plate during the execution of the first step by the starting leg. There were 24 possible combinations for the different stimulation intensities (noxious and control), stimulation sites (forefoot, arch, heel and post-heel), and postures (ST, HO and HC). Each combination was presented once, in random order, within a block of 24 trials. After each block of 24 trials, the subjects rested for 5 min. At least 120 trials per subject (5 blocks) were recorded for all participants. Before the actual recordings were performed, subjects were thoroughly familiarized with the stimulations to avoid eliciting a startle response in conjunction with the NWR. This comprised delivering several stimulations randomly to the different sites and postures, until the subjects felt confident with the sensations produced by the stimulation. 2.6. Data analysis The EMG recordings were band-pass filtered (Butterworth, 10– 500 Hz, fourth order, no phase lag). In the trials with stimulation, the root mean square (RMS) of the EMG amplitude in the window 60–220 ms after the stimulus onset was calculated. In the trials without stimulation, control trials, the RMS of the EMG amplitude in the corresponding window was calculated to determine the background activity. The reflex response was calculated as the difference between the RMS of the EMG in stimulated steps and the corresponding control steps. To decrease the inter-subject variability (Yang and Winter, 1984), the background activity and the EMG reflex responses in each muscle were normalized to the grand mean of the RMS values calculated previously in all control trials of the respective muscle for all three conditions. Finally, they were averaged across blocks for every site and posture. The signals from the goniometers were low-pass filtered (Butterworth, 25 Hz, sixth order, no phase lag). The kinematic
reflex response for the hip and knee was calculated as the peak angle change between the stimulated condition and the control condition. The analysis was assessed in the interval 110–470 ms post-stimulation. The time window was selected by visual inspection of all the data from the goniometers and was long enough to contain the reflex response. Unlike the hip and knee joints, the ankle joint changed the direction of its movement in the window of analysis. The movement changed from plantarflexion to dorsiflexion and thus, the kinematic response of the ankle was assessed in two windows in the interval 110–470 ms poststimulation. The split-point was set as the time point of peak plantarflexion in the 110–470 ms reflex analysis window (Spaich et al., 2006). 2.7. Statistics Repeated-measures ANOVA was applied to the data with factors ‘‘stimulation site’’ and ‘‘posture’’. Student Newman–Keuls statistics (SNK) were used for post hoc pair-wise comparisons when a main effect or an interaction was detected. P < 0.05 was regarded as statistically significant and data are presented as mean and standard error of the mean. 3. Results All subjects were able to complete the experiment and described the evoked sensation as sharp, pinpricking, and localized at the different stimulation sites. The stimulation did not evoke any sensation originating from the dorsum of the foot. The multiplication factor for the stimulation intensities was 1.4 for nine subjects and 1.5 for one subject resulting in stimulation intensities of 24.8 ± 1.3 mA for the forefoot, 18.2 ± 1.2 mA for the arch of the foot, 39.4 ± 3.1 mA for the heel, and 24.1 ± 1.7 mA for the post-heel stimulation site. Fig. 2 shows the NWR responses in the TA muscle following electrical stimulation. 3.1. EMG reflex responses The evoked reflex response was significantly modulated by stimulation site (RM ANOVA main effect) for all muscles, except for SOL (TA, F(3, 27) = 17.51, P < 0.001; BF, F(3, 27) = 4.36, P < 0.05; RF, F(3, 27) = 6.48, P < 0.01, and VL (F(3, 27) = 7.28, P < 0.001). The post hoc analysis showed that the largest TA reflex responses (Fig. 3A) were elicited after stimulating the arch of the foot, compared to the forefoot, heel, and post-heel stimulation sites (SNK, P values <0.05). Similar results were found for the RF (SNK, P < 0.05) and the VL muscles (SNK, P < 0.05) (Fig. 3D–E). In the BF muscle, the post hoc analysis showed that the largest reflex response was obtained by stimulation of the arch of the foot
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
4
M.A. Richard et al. / Clinical Neurophysiology xxx (2015) xxx–xxx
Fig. 2. TA EMG response to electrical stimulation at specific sites and postures. In the background, in thin line, the control signal is plotted. The plots are the mean of the 5 repetitions from subject 8. The vertical full line indicates the stimulation onset. The stippled lines delimit the analysis window.
(SNK, P < 0.05) compared to the heel and post-heel sites (Fig. 3C). The analysis of the EMG responses revealed no significant interactions between site and posture in any of the muscles. Posture significantly modulated the evoked reflex responses in TA (F(2, 18) = 6.05, P < 0.01), SOL (F(2, 18) = 6.83, P < 0.01) and VL (F(2, 18) = 8.42, P < 0.01) (Fig. 3), while no significant modulation was observed in BF and RF. In SOL muscle, results showed that the size of the evoked reflex response at heel contact (mean 0.15 ± 0.19) was not different from zero; post hoc pairwise comparisons showed that responses evoked during symmetrical standing (mean 0.34 ± 0.14) and heel off (mean 0.32 ± 0.09) were not different from each other (SNK, P = 0.89) (Fig. 3B). A tendency for a suppressive response was observed when the stimulation was applied at the forefoot site at heel contact (Fig. 3B) but this response was not different from zero. In the TA muscle, the post hoc analysis showed that the size of the evoked reflex response was smallest during symmetrical standing (mean 1.55 ± 0.40), and that there were no significant differences between the heel off (mean 2.30 ± 0.46) and heel contact (mean 2.24 ± 0.37) postures (Fig. 3A). Regarding the VL muscle, the heel contact posture showed the largest evoked reflex response (mean 4.74 ± 1.30), while no significant differences were observed between stimulation at heel off (mean 1.85 ± 0.54) and symmetrical standing (mean 2.37 ± 0.71) (Fig. 3C). 3.2. Kinematics At the hip joint, flexion was the primary reflex response with posture and site as main effects (F(2, 18) = 7.24, P < 0.01; F(3, 27) = 6.24, P < 0.01). Moreover a significant posture and site interaction (F(6, 54) = 2.99, P < 0.05) was found. The post hoc analysis revealed that NWR responses elicited after stimulation of the arch during heel off were significantly larger than those evoked at all other posture and site combinations (SNK, P values <0.01). Additionally, at heel contact the stimulation at the forefoot, heel, and post-heel sites did not produce a response different from zero (95% CI included zero, Fig. 4B). The NWR responses after stimulation of the arch during heel contact were significantly smaller than those evoked at all other posture and site combinations (SNK, P < 0.01). During symmetric standing, the NWR responses after stimulation in the arch and post-heel were larger (SNK, P values <0.01) than after stimulation on the forefoot and heel. At the knee joint, the main reflex response was flexion modulated by stimulation site (F(3, 27) = 5.62, P < 0.01). Hence, the largest knee response was evoked by stimulation on the arch of the foot (SNK P < 0.05) (Fig. 4A). There were no significant differences between responses evoked by stimulation on the forefoot, the heel, and post-heel sites. No interaction between factors was found.
At the ankle joint, in the first analysis window (110 ms to maximum plantarflexion), site significantly modulated the degree of plantar flexion (RM ANOVA, main effect, F(3, 27) = 10.00, P < 0.001). The post hoc analysis showed that stimulation of the arch of the foot and the post-heel sites resulted in the largest plantarflexion reduction (SNK P < 0.01). The posture also significantly modulated the early ankle response (RM ANOVA, main effect, F(2, 18) = 9.27, P < 0.01). At heel contact, stimulation on the arch, heel, and post-heel sites did not produce a response that was significantly different from the unperturbed condition (95% CI included zero degrees). Increased plantarflexion was observed at heel contact (SNK P < 0.01) (Fig. 4C) for the forefoot site. The largest reduction in plantarflexion was observed at heel off (SNK P < 0.05). No interaction between factors was found. In the second analysis window (maximum plantarflexion to 670 ms) (Fig. 4D), there was a significant posture and site interaction (F(6, 54) = 2.96, P < 0.05). Stimulation during symmetrical standing evoked the largest increase of dorsiflexion, no matter which site was stimulated (SNK P values <0.001). Stimulation at heel off resulted in reduced dorsiflexion when stimulating at the heel (SNK P < 0.05) and post-heel sites (SNK P < 0.01) compared to all other sites and postures; at the same time, stimulation at the forefoot and arch sites evoked a response that was not different from zero. Stimulation during heel contact evoked an increase of dorsiflexion no matter which site was stimulated (SNK P values <0.001). 4. Discussion A single experimental session was carried out where NWR reflex responses were assessed in order to determine how the NWR reflex is modulated during gait initiation. Electrical stimulation was applied to ten healthy subjects on four sites of the foot and in three different postures. Both, stimulation site and posture modulated the evoked reflexes. Regarding stimulation site, stimuli delivered on the arch of the foot resulted in larger reflex responses compared to heel and post-heel stimulation sites for all muscles, except for SOL. Similarly, the largest flexion of the knee and hip was obtained by stimulation of the arch of the foot. In relation to postures, the largest reflexes were obtained at HC and HO in the TA muscle, while the largest responses were obtained at ST and HO in the SOL muscle. 4.1. Methodological considerations The gait initiation process is composed of four sub-phases (Breniere et al., 1981). The first two sub-phases form the preparatory phase (PP) that goes from initial movement of the COP of the
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
M.A. Richard et al. / Clinical Neurophysiology xxx (2015) xxx–xxx
5
Fig. 3. EMG responses to electrical stimulation at the specified stimulation postures and sites. The normalized background EMG activity is shown as empty circles and solid lines. Values are mean and standard error of the mean. Significant differences between groups are indicated (⁄P < 0.05; ⁄⁄P < 0.01). Acronyms: TA, Tibialis anterior; SOL, Soleus; BF, Biceps Femoris; RF, Rectus Femoris; VL, Vastus Lateralis; ST, Symmetrical standing; HO, Heel off; HC, Heel contact.
subject (Winter, 1995) until toe-off of the starting foot. The next two sub-phases form the stepping phase (SP). This phase of the gait initiation process is divided in the single support sub-phase that finalizes with the heel contact of the starting foot (Serrao et al., 2012a) and is followed by the double support sub-phase. In the present study, three different postures that represent events of the gait initiation process were selected. The first posture, standing, is the starting point of the gait initiation process, where the loading of the starting leg was maintained at 50% of the total body load. This posture was included because it requires the
coordination of all postural muscles to maintain equilibrium. The second posture, heel off corresponds to an instant before the end of the PP. A threshold of 20% of the body weight was selected as trigger for the stimulation, since at that moment most of the body weight is supported by the standing leg, allowing lifting of the starting foot. The third posture was heel contact which corresponds with the end of the single support sub-phase. In order to ensure a clear HC, the stimulation was triggered when 80% of the body weight was supported by the starting leg. Thus, this experimental setup allowed to investigate the loading/unloading of the
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
6
M.A. Richard et al. / Clinical Neurophysiology xxx (2015) xxx–xxx
Fig. 4. Kinematic responses at the knee, hip, and ankle joints induced by electrical stimulation at specified postures and sites. The error bars indicate standard error of the mean. Significant results regarding posture modulation are shown in panel C, and posture and site interactions in panels B and D (⁄P < 0.05; ⁄⁄P < 0.01). Acronyms: ST, symmetrical standing; HC, Heel contact; HO, Heel off.
leg during gait initiation in a natural motor context. Moreover, the motor behavior of the joints is expected to be the same at each posture: hip and knee flexion and ankle dorsiflexion at HO (20% BW) and hip and knee flexion (with ankle in neutral position) at HC (80% BW). With the three different postures, a large variation in the pressure on the stimulation electrodes must be expected, which could explain, in principle, some of the observed differences in responses. However, recent evidence indicates that when stimulating the sole of the foot, Ad fibers were activated within the most superficial 100 lm of the dermis while Ab fibers were activated throughout the dermis (Frahm et al., 2013a). Furthermore, the current flow between the stimulating and reference electrodes happened primarily through the dermis, with much smaller current densities in the fat, muscle, and bone layers (Frahm et al., 2013b) suggesting that even in both extreme conditions (fully loading/unloading of the foot) the current flow will most likely cause depolarization of the same fibers. In this study, the stimulation intensity was 1.4–1.5 times the pain threshold, which was clearly perceived as painful. Strong, painful electrical stimulation will inherently activate Ad as well as Ab afferents eliciting a withdrawal reflex response. Andersen (2007) has shown that from 60 ms after the stimulus onset, the EMG can contain components mediated by Ad afferents, constituting therefore a nociceptive withdrawal reflex response. However, this response could also be mediated by non-nociceptive inputs from proprioceptors of the foot, muscle and cutaneous afferents which are known to converge in the spinal withdrawal reflex arc.
4.2. Stimulation site modulation Previous studies investigating the relationship between gait and the NWR found a modulation of the reflex amplitude depending on where the electrical stimulation was delivered (Granat et al., 1993; Spaich et al., 2004). This modulation was primarily observed in the TA muscle during symmetrical stance and during steady gait (Spaich et al., 2004; Emborg et al., 2009), and also in BF, and VL muscles (Serrao et al., 2012b). However, it has not previously been investigated how the NWR is site modulated during gait initiation. The results of this study showed that the stimulation site modulated the amplitude of the NWR in BF, RF, TA, and VL muscles during gait initiation. Specifically, stimulations of the arch of the foot evoked the largest reflexes in all muscles, except SOL, and in the knee joint, independent of the posture. Stimulation delivered at the proximal part of the sole of the foot evoked the smallest NWR response in the BF, RF, TA, and VL muscles. This is in accordance with previous studies for the BF, TA, and VL muscles (Spaich et al., 2004; Serrao et al., 2012b). Similarly, the smallest flexion of the knee joint was evoked by stimulation at the proximal part of the sole of the foot. These two patterns of stimulation-site modulation reflect that electrical stimulation was applied within the reflex receptive field (RRF) of the different muscles, defined as the area of the skin from which a reflex can be elicited (Schouenborg, 2002; Andersen et al., 2005). Hence, stimulation in the medial, distal part of the sole of the foot mainly activated BF, RF, TA, and VL (Sonnenborg et al., 2000; Spaich et al., 2004) associated with hip and knee flexion and reduced plantarflexion in the initial part of the ankle response,
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
M.A. Richard et al. / Clinical Neurophysiology xxx (2015) xxx–xxx
while stimulation of the proximal part of the sole of the foot evoked a NWR mainly in BF, RF, and VL. Moreover, there were no significant differences in NWR amplitude in the SOL muscle across stimulation sites and the angle changes in the second part of the ankle response did not show modulation related to the site of stimulation. The reflex receptive fields for antagonistic muscles overlap, e.g. both VL/RF and BF were activated simultaneously when the stimulus was applied to the proximal part of the sole of the foot. Obviously, the net kinematic response reflects the balance between activation of the antagonists and most likely the function of the cocontraction is to provide additional stability to the joint during the withdrawal movement. These results support the idea that the NWR response to stimulation during gait initiation was highly related to the location of the stimulus in accordance with the RRF of the muscle, and consequently, activated a neuronal pool that triggered a coordinated movement to withdraw the limb away from the stimulation. In the present study, this stimulation site dependency at muscle level was independent of the degree of load on the leading limb (no ANOVA interaction). However, the results showed an interaction at the hip joint, meaning that the hip flexion was largest when the stimulation was delivered in the arch of the foot at HO, but the response was suppressed at HC when the stimulation was delivered in the other sites. This lack of NWR response has likely a functional meaning since it avoids postural instability and collapsing during heel-contact during the gait initiation process. 4.3. Posture modulation The modulation of the NWR response related to different postures has been studied in animals (Schotland et al., 1989) and in humans during standing, rhythmic walking (Spaich et al., 2004), and in the transition from gait initiation to steady-state walking (Park et al., 2009). In general, previous studies dealing with the relationship between gait and the NWR showed a modulation of the reflex amplitude depending on which moment of the gait process the stimulation was delivered (Spaich et al., 2004, 2006; Emborg et al., 2009; Serrao et al., 2012a). However, the present study also revealed interactions between postures and stimulation sites in the GIP for the hip and ankle measurements but not for the EMG responses. At HO, where an increase of the response was expected for BF and RF (Serrao et al., 2012a), our results showed that neither BF nor RF reflex responses seem to be posture dependent. The modulation on TA can be related to supporting the development of the displacement of the CoP towards the stance leg and the forward movement of the body. The largest TA EMG reflex responses were detected during the two dynamic conditions (HC and HO) compared to the symmetrical stance posture. This suggests that the large TA reflex response at HO might be related to the preactivation of this muscle normally associated with the production of the swing phase. This pre-activation might facilitate the foot withdrawal (Mann et al., 1979). At HC, the VL reflex response was largest contrasting with previous results (Serrao et al., 2012a) where no differences in VL responses during the sub-phases of gait initiation were found; other studies reported however a facilitation of cutaneous reflexes in VL in the transition from swing to stance (Crenna and Frigo, 1984; Zehr et al., 1998) as a mechanism to decrease the possibility of leg collapse (Zehr et al., 1998). When the subject was standing, an excitatory response of the NWR was also found in the quadriceps muscle (Decchi et al., 1997), but in the present study the VL reflex response was almost the same at HO compared to symmetrical standing. As expected, the main response at the hip joint was flexion (Emborg et al., 2009). The largest reflex was obtained during HO
7
when the stimulation was delivered in the arch of the foot. This response is related to the degree of unloading of the starting leg in line with prior studies in symmetrical stance (Serrao et al., 2012b). Additionally, NWR responses were smallest during HC in order to prevent loss of balance and destabilization in agreement with observations during walking (Spaich et al., 2004). Therefore, the results suggest that the flexion of the hip and knee are the main movements involved in the withdrawal of the lower limb during gait initiation. Unlike the knee where the NWR response is only modulated by stimulation site, at the hip joint the response is mainly modulated by the current posture. In this regard, the interaction found at the hip joint suggests that the central nervous system modulates the hip kinematics according to the motor context (HO, HC). Thus, the central nervous system seems to attribute an important role to the hip joint in the first step of the gait cycle. 4.4. Functional modulation The present study indicates that the NWR response elicited by localized stimulation of the foot during gait initiation depends on the stimulation site and on the posture during gait initiation. Our findings suggest that stimulation on the arch of the foot elicited the largest reflexes regardless of the posture, and in general the reflex responses were larger during heel off. Larger reflexes during heel off suggest a functional modulation of the reflex gain as the evoked response supported the GIP by facilitating the flexion of the joints. This is in agreement with prior observations for the knee and hip joints during steady state gait (Spaich et al., 2004) and indicate a priority to maintain balance and ensure continued forward propulsion (Rietdyk and Patla, 1998). Moreover, at heel off, the center of pressure has been shifted to the stance limb and thereby the initial inhibition of the NWR as part of the preparatory balance adjustments during steady state standing (McIlroy et al., 1999) was reduced. To what degree the reflex gain modulation is a result of spinal motor programmes (Zehr and Duysens, 2004) or is a result of descending motor commands during the relatively unstable gait initiation process remains unclear. The functional, modular organization of the NWR might be used in rehabilitation strategies involving functional electrical stimulation. This might, in particular, be of relevance for patients with hypokinetic gait disorders that could benefit from reflexes elicited with proper timing and at the optimal stimulation site resulting in functional support of the gait initiation process. Acknowledgement This work was supported by the Svend Andersen foundation. Conflict of interest: Authors declare that there are no financial or other relationships that might lead to a conflict of interests. References Andersen OK. Studies of the organization of the human nociceptive withdrawal reflex. Focus on sensory convergence and stimulation site dependency. Acta Physiol 2007;189:1–35. Andersen OK, Sonnenborg F, Matjacˇic´ Z, Arendt-Nielsen L. Foot-sole reflex receptive fields for human withdrawal reflexes in symmetrical standing position. Exp Brain Res 2003;152:434–43. Andersen OK, Spaich EG, Madeleine P, Arendt-Nielsen L. Gradual enlargement of human withdrawal reflex receptive fields following repetitive painful stimulation. Brain Res 2005;1042:194–204. Berardelli A, Rothwell JC, Thompson PD, Hallett M. Pathophysiology of bradykinesia in parkinson’s disease. Brain 2001;124:2131–46. Breniere Y, Do MC, Sanchez J. A biomechanical study of the gait initiation process. J Biophys Med Nucl 1981;5:197–205. Brunt D, Vander Linden DW, Behrman AL. The relation between limb loading and control parameters of gait initiation in persons with stroke. Arch Phys Med Rehabil 1995;76:627–34. Crenna P, Frigo C. Evidence of phase-dependent nociceptive reflexes during locomotion in man. Exp Neurol 1984;85:336–45.
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019
8
M.A. Richard et al. / Clinical Neurophysiology xxx (2015) xxx–xxx
Decchi B, Zalaffi A, Spidalieri R, Arrigucci U, Di Troia AM, Rossi A. Spinal reflex pattern to foot nociceptive stimulation in standing humans. Electroencephalogr Clin Neurophysiol Electromyogr Mot Control 1997;105:484–9. Emborg J, Spaich EG, Andersen OK. Withdrawal reflexes examined during human gait by ground reaction forces: site and gait phase dependency. Med Biol Eng Comput 2009;47:29–39. Frahm KS, Jensen MB, Farina D, Andersen OK. Surface EMG crosstalk during phasic involuntary muscle activation in the nociceptive withdrawal reflex. Muscle Nerve 2012;46:228–36. Frahm KS, Mørch CD, Grill WM, Lubock NB, Hennings K, Andersen OK. Activation of peripheral nerve fibers by electrical stimulation in the sole of the foot. BMC Neurosci 2013a;8:114–6. Frahm KS, Mørch CD, Grill WM, Andersen OK. Experimental and model-based analysis of differences in perception of cutaneous electrical stimulation across the sole of the foot. Med Biol Eng Comput 2013b;51:999–1009. Fuhr T, Quintern J, Riener R, Schmidt G. Walking with WALK! IEEE Eng Med Biol Mag 2008;27:38–48. Granat MH, Heller BW, Nicol DJ, Baxendale RH, Andrews BJ. Improving limb flexion in FES gait using the flexion withdrawal response for the spinal cord injured person. J Biomed Eng 1993;15:51–6. Hass CJ, Waddell DE, Fleming RP, Juncos JL, Gregor RJ. Gait initiation and dynamic balance control in Parkinson’s disease. Arch Phys Med Rehabil 2005;86:2172–6. Hesse S, Reiter F, Jahnke M, Dawson M, Sarkodie-Gyan T, Mauritz KH. Asymmetry of gait initiation in hemiparetic stroke subjects. Arch Phys Med Rehabil 1997;78:719–24. Lyons GM, Sinkjær T, Burridge JH, Wilcox DJ. A review of portable FES-based neural orthoses for the correction of drop foot. IEEE Trans Neural Syst Rehabil Eng 2002;10:260–79. Mann RA, Hagy JL, White V, Liddell D. The initiation of gait. J Bone JT Surg Ser A 1979;61:232–9. McIlroy WE, Bent LR, Potvin JR, Brooke JD, Maki BE. Preparatory balance adjustments precede withdrawal response to noxious stimulation in standing humans. Neurosci Lett 1999;267:197–200. Nicol DJ, Granat MH, Baxendale RH, Tuson SJM. Evidence for a human spinal stepping generator. Brain Res 1995;684:230–2. Park S, Choi H, Ryu K, Kim S, Kim Y. Kinematics, kinetics and muscle activities of the lower extremity during the first four steps from gait initiation to the steadystate walking. J Mech Sci Technol 2009;23:204–11.
Quintern J, Krewer C, Husemann B, Hartmann E, Heller S, Müller F, et al. Controlled clinical study of stimulation of flexor reflex afferents for gait rehabilitation in patients with hemiplegia. In: Gantchev N, editor. From basic motor control to functional recovery III. University Press; 2003. p. 240. Rietdyk S, Patla AE. Context-dependent reflex control: some insights into the role of balance. Exp Brain Res 1998;119:251–9. Schotland JL, Lee WA, Rymer WZ. Wipe and flexion withdrawal reflexes display different EMG patterns prior to movement onset in the spinalized frog. Exp Brain Res 1989;78:649–53. Schouenborg J. Modular organisation and spinal somatosensory imprinting. Brain Res Rev 2002;40:80–91. Serrao M, Ranavolo A, Andersen OK, Conte C, Don R, Cortese F, et al. Adaptive behaviour of the spinal cord in the transition from quiet stance to walking. BMC Neurosci 2012a;13:80. Serrao M, Spaich EG, Andersen OK. Modulating effects of bodyweight unloading on the lower limb nociceptive withdrawal reflex during symmetrical stance. Clin Neurophysiol 2012b;123:1035–43. Sonnenborg FA, Andersen OK, Arendt-Nielsen L. Modular organization of excitatory and inhibitory reflex receptive fields elicited by electrical stimulation of the foot sole in man. Clin Neurophysiol 2000;111:2160–9. Spaich EG, Arendt-Nielsen L, Andersen OK. Modulation of lower limb withdrawal reflexes during gait: a topographical study. J Neurophysiol 2004;91:258–66. Spaich EG, Hinge HH, Arendt-Nielsen L, Andersen OK. Modulation of the withdrawal reflex during hemiplegic gait: effect of stimulation site and gait phase. Clin Neurophysiol 2006;117:2482–95. Winter DA. Human balance and posture control during standing and walking. Gait Posture 1995;3:193–214. Yang JF, Winter DA. Electromyographic amplitude normalization methods: improving their sensitivity as diagnostic tools in gait analysis. Arch Phys Med Rehabil 1984;65:517–21. Zehr EP, Duysens J. Regulation of arm and leg movement during human locomotion. Neuroscientist 2004;10:347–61. Zehr EP, Stein RB, Komiyama T. Function of sural nerve reflexes during human walking. J Physiol 1998;507:305–14.
Please cite this article in press as: Richard MA et al. Stimulation site and phase modulation of the withdrawal reflex during gait initiation. Clin Neurophysiol (2015), http://dx.doi.org/10.1016/j.clinph.2015.01.019