Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Brain Stimul. Author manuscript; available in PMC 2017 May 1.
Published in final edited form as:
PMCID: PMC4875865

Downregulating aberrant motor evoked potential synergies of the lower extremity post stroke during TMS of the contralesional hemisphere

Andrew Q. Tan, MSE,1,3 Jon Shemmell, PhD,4 and Yasin Y. Dhaher, PhD1,2,3



Growing evidence demonstrates unique synergistic signatures in the lower limb (LL) post-stroke, with specific across-plane and across-joint representations. While the inhibitory role of the ipsilateral hemisphere in the upper limb (UL) has been widely reported, examination of the contralesional hemisphere (CON-H) in modulating LL expressions of synergies following stroke is lacking.


We hypothesize that stimulation of lesioned and contralesional motor cortices will differentially regulate paretic LL motor outflow. We propose a novel TMS paradigm to identify synergistic motor evoked potential (MEP) patterns across multiple muscles.


Amplitude and background activation matched adductor MEPs were elicited using single pulse TMS of L-H and CON-H (control ipsilateral) during an adductor torque matching task from 11 stroke and 10 control participants. Associated MEPs of key synergistic muscles were simultaneously observed.


By quantifying CON-H/L-H MEP ratios, we characterized a significant targeted inhibition of aberrant MEP coupling between ADD and VM (p=0.0078) and VL (p=0.047) exclusive to the stroke group (p=0.028) that was muscle dependent (p=0.039). We find TA inhibition in both groups following ipsilateral hemisphere stimulation (p=0.0014; p=0.015).


We argue that ipsilaterally mediated attenuation of abnormal synergistic activations post stroke may reflect an adaptive intracortical inhibition. The predominance of sub 3ms interhemispheric MEP latency differences implicates LL ipsilateral corticomotor projections. These findings provide insight into the association between CON-H reorganization and post-stroke LL recovery. While a prevailing view of driving L-H disinhibition for UL recovery seems expedient, presuming analogous LL neuromodulation may require further examination for rehabilitation. This study provides a step toward this goal.

Keywords: Transcranial Magnetic Stimulation, Muscle Synergies, Contralesional Hemipshere, Stroke, Lower Limb


During walking, abnormal kinematic deviations are apparent following stroke1 reflecting the loss of independent joint control2, 3. Subsequent studies suggest these deficits are not compensatory strategies4, but involve abnormal neuromechanical interactions5. Moreover, observations of such neural constraints include coupled cross planar kinetic outflow6, 7 and altered muscle synergy structure8,9. Recent evidence form our lab demonstrated that the lesioned hemisphere upregulates aberrant coupled LL corticospinal excitability10. There we reported that stroke subjects exhibit abnormal TMS induced corticospinal coupling of the quadriceps muscles while performing isometric hip adduction. These findings underscore the need to clarify neuromechanical constraints underlying hemiparetic gait for developing effective clinical interventions.

In this study we explore the potential role that acute modulation of motor cortices play in the regulation of lower limb (LL) neural coupling post stroke. Specifically, we examine the relative contribution of contralesional hemisphere (CON-H) activation to aberrant across joint coupling of ipsilateral lower-extremity muscles post stroke. While traditional TMS paradigms examine the cortical excitability of a single muscle, we propose a novel TMS paradigm that evaluates MEP patterns across multiple LL muscles. This differential MEP signature will be used to explore potential cortical regulation11 of abnormal synergistic muscle activations while characterizing the modulatory effect of CON-H post-stroke. To our knowledge, no study has systematically evaluated synergistic LL MEP patterns to clarify cortically mediated motor adaptation.

Extensive evidence has implicated CON-H in mediating functional recovery post stroke1217, motivating investigations of the underlying neurophysiology. The association of increased CON-H activity with paretic hand movements have been documented using fMRI12, 13 and EEG14, 15. However, post-stroke remapping of sensorimotor cortices results in simultaneous activation of various cortical regions and descending pathways16, 17. TMS has been used to discern descending inputs to upper limb (UL) motor function in primates18,19 and humans20. Previous inquiries of ipsilateral UL control21,22 using TMS of CON-H23,24 have presented mixed results. While some have indicated that CON-H modulates post-stroke motor plasticity transcallosaly25, 26, others observed intact cortico-motoneuronal circuitry27, 28. It is unclear how CON-H governs across joint coupling of the LL post stroke.

Few published studies elucidate contributions of CON-H to adaptive motor output in LL muscles following stroke. Accordingly, there is insufficient evidence supporting that CON-H activates analogous neural substrates for the UL and LL, given segmental modulation of walking29. We hypothesize that a biased stimulation of the lesioned and contralesional cortices will have differential effects on the MEP patterns in the paretic limb of stroke survivors, suggesting that each hemisphere plays different roles in regulating the expression of abnormal muscle couplings. We further characterize temporal features of the elicited MEPs to indirectly explore neural pathways proposed to mediate adaptive CON-H plasticity.



10 normal control participants (mean age 37, range 20–52) and 12 stroke participants (mean age 58, range 43–67) with single hemispheric stroke were recruited for this study (Table 1). All participants gave written informed consent according to the declaration of Helsinki. The study was approved by the Northwestern University Review Board. Stroke participants were included if 1) they had only one stroke at least one year prior to study participation 2) had the ability to perform hip adduction movements during body weight support standing on the non-paretic leg. All stroke participants presented with persistent hemiparesis due to chronic stroke. Further, all stroke participants were ambulatory and able to independently walk with or without the use of an orthotic or assistive device. Functional levels of the participants are tabulated when available. Exclusion criteria for both stroke and control groups include 1) contraindications for TMS 2) history of seizures and medications known to affect central nervous system excitability 3) history of orthopedic injury or surgery to their lower limbs.

Table 1
Stroke and control participant information

Transcranial Magnetic Stimulation

Motor evoked potentials (MEP) were elicited in 8 LL muscles of the test limb using single pulse TMS. In stroke participants the test limb was the paretic leg whereas for control the test limb was the right leg. TMS was applied on separate trials to either the lesioned hemisphere (L-H) or contralesional hemisphere (CON-H). Surface electromyography (EMG) preamplifiers (Motion Analysis) were placed over the following muscles of the test limb: vastus medialis (VM), vastus lateralis (VL), rectus femoris (RF), adductor longus (ADD), tibialis anterior (TA), medial gastrocnemius (GAS), and biceps femoris (HAM). A common ground was placed over the dorsum of the hand. All electrodes were secured to the skin using adhesive tapes and self-adherent elastic Coban wrap (Dynarex Corp., Orangeburg, NY, USA). Single pulse TMS was delivered using a Magstim 200 stimulator (Magstim, Whitland, UK) via a 110 mm diameter double-cone coil. The coil was oriented to induce a posterior-to-anterior current flow over M1 region of each cortex. A tightly secured swimming cap placed over the participant’s head was used to map and demarcate the optimal coil position and orientation for placement consistency. The intersections of the inter-aural and nasion-inion lines were first marked on the cap as the vertex. The coil position was incrementally adjusted from an initial starting position of 1cm posterior and 2 cm lateral to vertex10 to locate the hotspot of the targeted ADD muscle in the test limb (paretic limb for stroke, right limb for control) over the lesioned motor cortex (L-H). Optimal coil position and intensity was mapped as the lowest intensity of magnetic stimulation required to evoke ADD MEPs of 50 μV in peak-to-peak amplitude in at least three of five consecutive trials. Coil positioning was performed in seated position to minimize subject fatigue. The lower motor hotspot for the test limb ADD was symmetrically mapped for stroke contralesional hemisphere (CON-H) or control ipsilateral hemisphere to the test limb. The initial coil position placement was approximately 1 cm posterior and 2 cm lateral to vertex on the hemisphere ipsilateral to the test limb30.

Experimental Procedure

Participants were secured in a motorized, instrumented exoskeleton (Lokomat; Hocoma, Zurich, Switzerland) and isometrically locked in the toeoff gait posture: 15° hip extension, 45° knee flexion, and 90° ankle dorsiflexion7. The dimensions of the orthosis were adjusted for each participant to align the orthosis joint centers with those of the participant. The participant’s lower extremities were secured to the orthosis via cuffs instrumented with 3 total six-degree-of freedom load cells (JR3, Woodland, CA) to measure the interaction forces and moments of the paretic test limb (Figure 1). The test limb was completely unloaded by the Lokomat such that participants did not have to actively support the limb during torque production.

Figure 1
Experimental Paradigm

Participants were first asked to produce maximum voluntary isometric contractions (MVIC) in hip adduction, abduction, flexion, and extension directions with their test limb. MVIC were performed to obtain maximum torque and EMG signals. Participants were next instructed to match voluntary isometric hip adduction torques at 40% of MVIC levels (Figure 1) with their test limb. Differences in background activity across muscles may influence the evoked MEP amplitude31. Therefore a minimum target torque of 30% MVIC was needed to enable consistent matching of the pre-stimulus background EMG activity in the non-targeted (non-ADD) muscles across trials and coil placements10. Furthermore, previous investigations reveal no significant variations in both the isometric torque signal and the TMS elicited action potentials in the post stroke lower limb for levels up to 60% MVIC32,33,34. Participants received instantaneous visual feedback of the hip joint torque magnitude and direction produced by moving a circular cursor on a LCD monitor to a circular target. A successful trial was considered if the participant was able to match the target adduction torque within ±5% of the torque magnitude hold for a minimum of 200ms. The torques generated at each joint were calculated from thigh and shank load cell signals using static equilibrium equations.

For each trial, the TMS coil was placed over the hotspot location for either hemisphere marked on the swim cap and secured to the participant’s head. CON-H and L-H were stimulated on separate trials. Stability of the coil placement was improved using a custom designed overhead pulley system and cable supporting stand. A research assistant manually stabilized and positioned the coil over each target hotspot for each hemisphere for all trials along with Velcro straps. When the participants successfully matched the target adduction torque, the TMS device was triggered to deliver a single pulse using a custom automated software trigger (MATLAB v7.01). The TMS device was not triggered if the participant produced any off axis torque during the hip adduction or if the target torque magnitude exceeded +−5%. MEPs for each trial were constantly monitored by oscilloscope to ensure that ADD amplitude cMEP and iMEP were matched for all trials (Figure 1, ,2).2). Stimulator output was accordingly adjusted to meet this condition and was initially set at 110% of the estimated seated stimulator output. The adjusted stimulator outputs for each subject for both L-H and CON-H coils placements while the participants were inside the Lokomat are listed in Table 1. 10 to 30 trials were performed for each hemisphere in blocks of 10, with two 5-minute rest periods between blocks.

Figure 2
Schematic representation of TMS stimulation paradigm

Given the low spatial resolution of TMS, we expect that we will activate both hemispheres even after CON-H placement. This is further amplified by the complex representation of the lower limb muscles in the cortex. MEP responses recorded at the muscles during CON-H stimulation will therefore be the summated responses from the activation of both ipsilateral and contralateral descending drive. We optimize the CON-H coil location and orientation for the test limb ADD by normalizing the stimulation intensity to the increased distance of a CON-H coil placement (Figure 2). This was accomplished by incrementally increasing the stimulator output in order to match the contralateral ADD MEP (cMEP) amplitude evoked by stimulation of CON-H (Figure 2). Successful matching of ADD MEP across coil placements ensures that the combined stimulation of both L-H and CON-H results in the same summated downstream motor output of the targeted ADD MEP from both hemispheres. Thus any non-uniform change in the pattern of MEPs for other non-targeted lower limb muscles must be a result of additional CON-H activation, facilitating meaningful comparisons between the patterns of MEP modulation across multiple muscles.

MEP Analysis

A secondary analysis of onset latencies from TMS has been traditionally used to make inferences regarding putative neural pathways to the upper and lower limbs21,22,26. While the primary goal of this investigation was to characterize the contribution of CON-H to aberrant synergistic muscle activation post stroke, the differential onset latency across hemispheres may provide indirect evidence regarding select neural pathways. To identify the onset latency of the TMS-induced motor evoked potential (MEP) in the EMG of each muscle, the EMG signal during the target matching task was filtered with an 8th order Butterworth, low-pass, zero-phase digital filter with a 220 Hz cutoff frequency. The filtered EMG signals were rectified and smoothed using the same digital filter with a 20 Hz cutoff frequency. MEP onset latencies were computed for each primary muscle in each participant from the ensemble of successful hip adduction trials (Figure 5). A pre-TMS background level of EMG activity was defined by calculating the mean and standard deviation of activity within a 100 ms window prior to a recorded TMS trigger pulse35. An extended 80 ms window following the TMS trigger pulse was used to search for the onset of MEPs30,36 that were clear and distinguishable from background activation35,36. Consistent with published onset latencies for LL muscles greater than 30ms10,29,35, the initial 20 ms epoch following the TMS trigger was designated to account for any post stimulus artifact. Similar to a procedure outlined by Garvey et al37, MEP onset was identified when five consecutive points in the EMG trace after this 20ms window were above three standard deviations of the mean pre-TMS EMG activity37 on the ascending phase (Figure 5). MEP amplitudes were measured peak-to-peak and averaged off-line across successful trials38. The MEP amplitude was calculated as the difference between the background activation during the target matching task and the average of a 20 ms time window centered at the peak MEP rectified pulse.

Figure 5
Interhemispheric Onset Latency Identification

Pre-stimulus background EMG activation of ADD was matched across trials for across coil placements through the torque matching paradigm. A post-hoc analysis was performed to ensure that all successful trials included in the ensemble averages achieved statistically equivalent ADD background activations. Similarly, a post hoc analysis of the background activation of any coupled non-targeted muscles was performed. The ratio of the mean iMEP amplitude over the mean cMEP amplitude will be quantified for each muscle for each participant. A statistical value less than one indicates reduction in iMEP amplitude relative to the cMEP amplitude following TMS of the ipsilateral hemisphere.

Statistical Analysis

Paired t-tests were used to evaluate if the ensemble averaged cMEP and iMEP ADD amplitudes were statistically different. To assess the potential confounding effect of differences in background activations of the non-targeted muscles for the different stimulation conditions, a paired t-test was used to test if the mean ADD, VM, and VL background activation in the 100ms prestimulus window during contralateral and ipsilateral hemisphere trials were statistically different. A two-way ANOVA was performed to examine the effect of muscle on iMEP/cMEP ratios, with group (stroke vs control) as the between subjects factor. Post hoc analysis using one-way analysis of variance (ANOVA) was used to detect significant differences in MEP ratios within the stroke group. Further post hoc analysis using pairwise comparisons were used to determine significant differences in MEP ratios across muscles using Fishers Least Squares Difference. Finally, one sample t tests were used to evaluate if the iMEP/cMEP ratio for each muscle’s ratio is statistically different than 1 for each group. All statistical analysis was performed using SPSS for windows v 19.0 (NCSS 9 LLC, Kaysville, Utah USA). A significance level of α = 0.05 was set for all statistical analyses.


Consistency in Background EMG Activation

MEP traces across multiple muscles following coil shift from a typical stroke participant can be seen in Figure 3. For stroke participants, the average stimulator intensity on the CON-H was 71.25%, with a range of 50% to 85% of max stimulator output. For control subjects, the average stimulator intensity was 57.1 %, with a range of 40% to 66% of max stimulator output. For the 100 ms prestimulus window, no statistical difference was detected in the mean background activations for the targeted ADD muscle between L-H and CON-H trials for all participants (p>0.05, p=0.241). This indicates that after shifting the coil to the opposite hemisphere, all targeted ADD trials were successfully matched in background activation eliminating the confounding effect of background activation on cortico-motor excitability of the ADD. Similarly no statistical difference in background activation was detected for VM (p=0.213) and VL (p=0.269). Furthermore, the mean ADD MEP amplitude was not statistically different between hemispheres for all participants (p>0.05, p=0.618) such that cMEP = iMEP for all trials (Figure 3, ,4).4). Exemplar data from one stroke participant depicts the quantification of latency differences between iMEP and cMEP (Figure 3). The distribution of the MEP latencies for all trials is shown in Figure 6. A paired t-test of CON-H and L-H latencies for all muscles shows no significant difference in onset latencies (p<0.000).

Figure 3
Exemplar stroke participant
Figure 4
Targeted inhibition following CON-H stimulation
Figure 6
Interhemispheric Onset Latency Distribution

Effect of Stimulated Hemisphere on MEP Ratio

Average iMEP/cMEP ratios across participants can be seen in Figure 4 for stroke and control participants. MEP ratios for stroke VM and VL were significantly lower than all other muscles in the stroke group only. The MEP ratio for TA was also significantly lower for both groups (Figure 3,,4).4). Several statistical outcomes confirm these findings. Results from a second two-way factorial ANOVA between the main effects of group and muscle location revealed a significant effect of group on MEP ratio (p = 0.028) and a significant effect of muscle location (p=0.039). No significant interaction was detected between group and muscle, indicating that for both groups, responses in the same muscles were affected by changing the stimulated hemisphere. Specifically, a Tukey post hoc test revealed a significant reduction in TA MEP ratio in both groups (p<0.05). Furthermore, a one-way ANOVA of the stroke group revealed significant differences between muscle MEP ratios (p=0.0331). MEP ratio in the TA of stroke survivors was significantly lower from the same ratio in all other muscles in the same group except VM and VL (p< 0.05). Reductions in TA MEP amplitude for a typical stroke participant are plotted in Figure 3. Finally, one sample t tests of the MEP ratios for each muscle were not statistically different than 1 (p>0.05) with the exception of stroke VM (p=0.0078), stroke VL (p=0.047), stroke TA (p=0.0014), and control TA (p=0.015).


The present study characterizes differences in the contribution of the lesioned and contralesional motor cortex to abnormal LL muscle activations following stroke. By quantifying the ratio of the iMEP to cMEP amplitudes, we observe a significant attenuation of aberrant MEP coupling between ADD, VM and VL following stimulation of the contralesional primary motor cortex that is exclusive to the stroke group. This confirms our hypothesis of a differential effect of lesioned and contralesional hemisphere stimulation on LL muscles and supports the idea that motor signals from CON-H reduce the expression of abnormal synergistic corticomotor excitability that impair stroke gait. The predominance of sub 3ms iMEP latencies together with a significant reduction of TA ratios across groups suggests evidence of direct LL ipsilateral projections from CON-H. As such our observations seem incongruent with the notion of maladaptive motor output in CON-H post stroke suggested in the UL39 and LL30. Our observations indicate that fundamental differences exist between CON-H contributions to functional recovery in the upper and lower limbs. The primary finding reported here is the reduction of abnormally coupled corticomotor output across lower limb muscles post stroke resulting from a biased stimulation of each cortex. Within the context of TMS, it is not possible to further discern between putative descending pathways given the limited spatial resolution. By combining both the targeted attenuation and interhemispheric onset latency data, we make informed interpretations regarding specific neural pathways that may or may not have contributed to the observed modulation.

Targeted Inhibition of Abnormal Synergistic Coupling

Several TMS paradigms have been used to investigate UL cortico-motoneuronal connections from the CON-H post stroke40. However a systematic evaluation of coupled iMEP patterns in the post-stroke LL is lacking. While other groups have characterized changes in ipsilateral vs contralateral CST conductivity in the post stroke LL41, the iMEP/cMEP ratio employed here specifically highlights the reduction of coupled motor output across LL muscles resulting from biased activation of each motor cortex. We demonstrate a novel multi-muscle TMS paradigm that uniquely evaluates cortical modulation of abnormal synergistic neuromotor coupling. Our data elucidates a significant differential effect of hemisphere on the evoked MEP pattern across LL muscles. Two important findings argue for this interpretation: 1) consistent with a previous investigation10, stimulation of the lesioned hemisphere amplified abnormal adductor and knee extensor EMG activation coupling relative to controls 2) stimulation of CON-H significantly reduced this activation coupling only in stroke. Crucially, the observed inhibition was specific to muscles previously reported to be expressed in abnormal LL synergies2, 8. Our observations corroborate evidence of neurally constrained muscle synergies9 previously proposed to constrain paretic gait subtask performance42. These findings support the idea that neuromodulation from CON-H may yield a functionally relevant output of CON-H hyper excitability for hemiparetic LL gait despite mixed evidence for UL recovery43,44.

One could argue that ipsi MEPs from the lesioned hemisphere to the non-paretic muscle could potentially be a more appropriate control for changes in muscle activation coupling. It has been suggested that the unaffected side of stroke participants may not be completely unaffected given evidence of neural organization16,17. Applying the current methodology to healthy subjects arguably yields a more effective control case to investigate ipsilateral hemisphere contributions to non-paretic muscle. In healthy subjects we do not see reductions in muscle activation coupling following CON-H stimulation. However it must be acknowledged that the iMEP/cMEP ratio cannot directly apportion the relative effect the stimulation to ipsilateral vs contralateral conductivity. In the absence of bilateral assessments of TMS connectivity and imaging analysis of CST integrity, inferences regarding asymmetries in contra vs. ipsilateral connectivity post-stroke will prove difficult in the context of this study.

Intracortical Inhibition

The specificity of the observed attenuation further suggests that TMS potentially activated inhibitory pathways from ipsilateral cortex to LL muscles. In the absence of more invasive recordings20, we can only speculate regarding the mechanism of inhibition along the neuraxis. One possible neural substrate underlying our findings is the widely observed intracortical inhibition (ICI) mediated by GABA-ergic activity of cortical interneurons45 and directly demonstrated in humans20 Stimulation of CON-H may have activated intracortical inhibitory circuits that reduced the expression of abnormal muscle synergies. Investigations of UL cortico-cortical inhibitory circuits reported inhibitory modulation onsets of 1–3ms following a conditioning stimulus46, 23. The similar timescale of our iMEP latencies suggest that circuits pertaining to ICI may potentially account for the targeted attenuation via inhibitory circuits that decrease the excitability of cortical cells29. Although our paradigm does not directly test this hypothesis, short latency ICI has been demonstrated in LL muscles47 in which ICI was not strongly related to motor threshold or MEP recruitment47. There are limited studies investigating intracortical mechanisms in the LL post stroke and the relation to clinical recovery measures in the UL is equivocal28, 40,45. Nonetheless the observed multi-muscle iMEP signature may be reflective of an inhibitory connectivity between the cortical representations of abnormally synergistic muscles post stroke.

Latency of Ipsilateral Responses to TMS

Given the multiplicity of pathways through which CON-H can access the LL, the skewed distribution of the observed iMEP latencies (Figure 6) favor certain neurophysiological interpretations. Short interhemispheric iMEP onset latencies suggest that TMS of the ipsilateral cortex may have excited a fast conducting corticospinal pathway to LL muscles. We observe short latencies from all LL muscles, whereas previously reported UL iMEPs were rarely found in distal muscles but reported in proximal muscles in normal participants48, 49. Direct cortical control of ipsilateral limb muscles has been previously proposed from TMS studies of UL CON-H projections17,28. Moreover anatomical evidence from primates has established direct monosynaptic corticospinal projections to hindlimb motoneurons50 along with an uncrossed corticospinal tract that descends ipsilaterally51. Therefore, it seems plausible that unlike the UL52, ipsilateral pathways from CON-H motor cortex play an important role in LL functional recovery post-stroke. Nevertheless, the dominance of short onset latency differences provides indirect evidence that ipsilateral corticospinal inputs to the human LL drive the observed inhibition, a notion debated in the healthy UL53. Improving the sampling resolution may yield additional insight regarding pathways that may or may not contribute to the observed responses.

Differential Contributions of Each Hemisphere

Given that both cortices are activated even after a biased placement of the coil, the extent to which each hemisphere contributes to the lower limb expression of MEP coupling cannot be discriminated by the data. Nevertheless, biasing the stimulation of one hemisphere via coil placement has previously been employed in TMS studies investigating the underlying electrophysiology of both healthy and pathological cohorts17,21,22,26. To account for the increased distance of the coil to the targeted ADD representation, amplifying the stimulation output to match ADD cMEP and iMEP ensures that the aggregate effect of stimulating both L-H and CON-H results in the same motor output for the target ADD across coil placements. However, if distance from the lesioned hemisphere was the only parameter that changed following coil shift, then an increase in stimulator output should modulate iMEPs for all non-targeted muscles. To this end, we do not observe a modulatory effect on the iMEP amplitudes across non-targeted muscles. If CON-H has no role to play in the targeted attenuation, we would alternatively expect an amplification of abnormal knee extensor/hip adductor coupling consistent with previously reported results10. Our findings in Figure 4 presents the opposite case in which the inhibitory effect of CON-H stimulation was specific to the knee extensor muscles. One could also argue that the observed targeted attenuation could potentially be attributed to the relative spatial location of each muscle’s cortical neurons. The representations of VM and VL in motor cortex may be located further away from the other lower limb muscles and experience a diminished stimulation. This seems unlikely given anatomical studies in primates where the LL representations are clustered together54. Together, these controls undermine the possibility that current spread merely activated contralateral cortex.

CON-H Influence of Lower Limb

A significant feature of our findings was that we robustly evoked amplitude matched adductor iMEPs in all study participants by stimulating ipsilateral motor cortex. This finding seems incompatible with previous UL studies where some authors elicit iMEPS in the healthy UL53, 22 while others have failed to evoke ipsilateral responses55, 56, 27 in stroke participants. These equivocal results may be explained by predominant plastic reorganization in LH57 or differential inputs to distal muscles21. The lack of correlation between CON-H activity and UL motor recovery58, 26 undermine a role of CON-H in shaping functional recovery. However, the neurophysiological substrates underpinning LL control are different given interactions with segmental modulation of neural walking circuits59. The ability to preferentially modulate one lower limb cortex has been demonstrated in a limited number of studies using tDCS60, rTMS61, and paired associative stimulation62. Reliable ADD iMEPs were elicited in all participants at 40% maximum torque, suggesting that the proximal LL joints may be more readily excitable post stroke than the distal UL.

In our study, we remarkably observe a targeted inhibition of TA across groups following coil shift. This corroborates previous research supporting privileged cortical control for TA63,64,65 where direct monosynaptic connections have been implicated in the healthy TA63, 29. This reduction in TA across groups may thus represent a preserved feature of healthy motor control where post-stroke plasticity has limited effect. These findings imply that direct inhibitory pathways exist in healthy participants and implicate the ipsilateral motor cortex in the regulation of the lower limb.

Transcallosal Pathways

While transcallosal pathways and their inhibitory effect58, 66 have been widely documented during paretic hand movements20, 67, their influence in LL pathways remains unclear. A majority of the observed iMEP latency differences were below previously estimated minimum UL interhemispheric conduction times ranging from 5–8ms68. The minority of long latency differences in our data undermines the idea that the inhibition was mediated transcallosally. However, this does not conclusively preclude contributions of IHI as our data represents an aggregate effect of both excitatory and inhibitory activity. While reducing IHI from CON-H has been leveraged to enhance paretic hand function69, 39, the inhibition from CON-H seems advantageous. Contrary to the UL, our data suggests that downregulating abnormally high IHI drive from CON-H70, 71 may be beneficial for LL gait.

Subthreshold Responses

One could argue that increasing the stimulator output could have differentially biased the electrophysiology of the underlying intracortical circuitry. Interestingly for control participants, while the stimulation needed to evoke an MEP in the contralateral hemisphere is lower, a similar increase in stimulator output was needed to match ADD MEPS for ipsilateral coil placement (~6% stimulator output increase, TABLE 1). It has been reported in healthy subjects that short interval intracortical inhibition can be achieved in the upper limb using conditioning pulse intensities ranging from 0.7 to 1.0 of active motor threshold (AMT)45,46 within a paired pulse paradigm. However, there is currently no consensus regarding a shifted balance in intracortical inhibition/facilitation post stroke, especially in the lower limb at these stimulation ranges. We are aware of only one lower limb study that reports a shift toward lower intracortical facilitation in TA post-stroke using a 1.2 AMT conditioning pulse72. The maximum stimulation intensity used in our study was 1.12 AMT. Nevertheless, Chen and colleagues reported that the absolute conditioning stimulus intensities required to elicit intracortical inhibition may not be related to active motor threshold or MEP recruitment in both lower and upper limb muscles47. Given that our CONH stimulation intensities never exceeded 1.2 AMT, there is no available evidence to suggest that stimulus levels used shifted the balance in intracortical circuitry toward facilitation or inhibition.

Future Directions

Other limitations may potentially confound the generalizability of the observed MEP patterns. Given that the current results focus on a hip adduction/knee extension synergy, it remains to be seen if CON-H also modulates other abnormal across joint kinetic coupling. Additionally, the stroke participants tested here are relatively high functioning and well recovered given their ability to independently ambulate in the community. Furthermore, individuals with only brain stem and cerebellar stroke were not tested in this study. Experiments with stroke subsets may clarify inferences on brain stem pathways as generalizations to the broader stroke population must be further qualified. Correlations with clinical recovery level and other gait related performance metrics are an important topic of future research.


While a prevailing view of driving LH disinhibition for UL recovery seems expedient, we argue that presuming analogous neuromodulation for the LL may adversely affect rehabilitation outcomes. Our results demonstrate that neural substrates underlying LL post-stroke muscle recovery may activate alternative pathways than the UL following CON-H stimulation. Therefore, developing appropriate gait therapies requires further clarity of post-stroke neural plasticity specific to LL motor pathways. This study takes an important step in exploring the relationship between CON-H cortical reorganization and the modulation of abnormal LL synergies post-stroke.


  • Novel multi-muscle magnetic stimulation paradigm used to evaluate abnormal muscle synergies
  • Targeted inhibition of stroke specific synergy during contralesional hemisphere stimulation
  • Short latency motor evoked potentials across muscles imply direct lower limb ipsilateral pathways
  • Clarify differences between contralesional hemisphere contributions to upper and lower limbs
  • Insight into contralesional hemisphere reorganization and lower limb recovery post stroke


We are grateful to Franz Nigl, MSc, for technical support, Heidi Roth, DPT, and Dr. Gilles Hoffman for experimental support. The study was supported by NIH-NINDS Grant R01NS064084-02 and NIH T32 HD057845.


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


1. Kerrigan DC, Frates EP, Rogan S, Riley PO. Hip hiking and circumduction: quantitative definitions. Am J Phys Med Rehabil. 2000;79:247–52. [PubMed]
2. Cruz TH, Dhaher YY. Evidence of lower limb torque synergies after stroke: an isometric study. Stroke. 2008;39:139–47. [PMC free article] [PubMed]
3. Cruz TH, Lewek MD, Dhaher YY. Biomechanical impairments and gait adaptations post-stroke: multi-factorial associations. J Biomech. 2009;42:1673–1677. [PMC free article] [PubMed]
4. Sulzer JS, Gordon KE, Dhaher YY, Peshkin MA, Patton JL. Preswing knee flexion assistance is coupled with hip abduction in people with stiff-knee gait after stroke. Stroke. 2009;41:1709–1714. [PMC free article] [PubMed]
5. Rogers LM, Brown DA, Gruben KG. Foot force direction control during leg pushes against fixed and moving pedals in persons post-stroke. Gait Posture. 2004;19(1):58–68. [PubMed]
6. Hunter BV, Thelen DG, Dhaher YY. A three-dimensional biomechanical evaluation of quadriceps and hamstrings function using electrical stimulation. IEEE Trans on Neural Sys and Rehab Engin. 2009;17:167–175. [PubMed]
7. Tan AQ, Dhaher YY. Evaluation of lower limb cross planar kinetic connectivity signatures post-stroke. J Biomech. 2014;47:949–956. [PMC free article] [PubMed]
8. Cruz TH, Dhaher YY. Impaired lower limb muscle synergies post-stroke. Conf Proc IEEE Eng Med Biol Soc. 2009;2009:3956–9. [PubMed]
9. Clark DJ, Ting LH, Zajac FE, Neputne RR, Kautz SA. Merging of healthy motor modules predicts reduced locomotor performance and muscle coordination complexity post-stroke. J Neurophys. 2010;103:844–857. [PubMed]
10. Krishnan C, Dhaher YY. Corticospinal responses of quadriceps are abnormally coupled with hip adductors in chronic stroke survivors. Exp Neuro. 2010;233:400–407. [PMC free article] [PubMed]
11. Gerachschenko T, Rymer WZ, Stinear JW. Abnormal corticomotor excitability assessed in biceps brachii preceding pronator contraction post-stroke. Clin Neurophy. 2008;119:683–692. [PMC free article] [PubMed]
12. Cao Y, D’Olhaberriague L, Vikingstad EM, Levin SR, Welch KM. Pilot study of functional MRI to assess cerebral activation of motor function after poststroke hemiparesis. Stroke. 1998;29:112–22. [PubMed]
13. Cramer SC, Nelles G, Benson RR, et al. A functional MRI study of subjects recovered from hemiparetic stroke. Stroke. 1997;28:2518–2527. [PubMed]
14. Verleger R, Adam S, Rose M, Vollmer C, Wauschkuhn B, Kompf D. Control of hand movements after striatocapsular stroke: high-resolution temporal analysis of the function of ipsilateral activation. Clin Neurophysiol. 2003;114:1468–76. [PubMed]
15. Serrien DJ, Strens LH, Cassidy MJ, Thompson AJ, Brown P. Functional significance of the ipsilateral hemisphere during movement of the affected hand after stroke. Exp Neurol. 2004:425–432. [PubMed]
16. Johansen-Berg H, Rushworth MFS, Bogdanovic MD, Kischka U, Wimalaratna S, Mathews PM. The role of ipsilateral premotor cortex in hand movement after stroke. PNAS. 2002;99:14518–4523. [PubMed]
17. Gerloff C, Bushara K, Wassermann EM, et al. Multimodal imaging of brain reorganization in motor areas of the contralesional hemisphere of well recovered patients after capsular stroke. Brain. 2006;129:791–808. [PubMed]
18. Edgley SA, Eyre JA, Lemon RN, Miller S. Comparison of activation of corticospinal neurons and spinal motor neurons by magnetic and electrical transcranial stimulation in the lumbosacral cord for the anaesthetized monkey. Brain. 1991;120:839–53. [PubMed]
19. Fisher KM, Zaaimi B, Baker SN. Reticular formation responses to magnetic brain stimulation of pirmary motor cortex. J Physiol. 2012;590(16):4045–4060. [PMC free article] [PubMed]
20. Di Lazzaro V, Oliviero A, Profice P, et al. Direct demonstration of interhemispheric inhibition of the human motor cortex produced by transcranial magnetic stimulation. Exp Brain Res. 1999;124:520–524. [PubMed]
21. Turton A, Wroe S, Trepte N, Fraser C, Lemon RN. Contralateral and ipsilateral EMG responses to transcranial magnetic stimulation during recovery of arm and hand function after stroke. Electroenceph and Clin Neurophys. 1996;101:316–328. [PubMed]
22. Ziemann U, Ishii K, Borgheresi A, Yaseen Z, Battaglia F, Hallet M, Cincotta M, Wassermann EM. Dissociation of the pathways mediating ipsilateral and contralateral motor-evoked potentials in human hand and arm muscles. J Physio. 1999;518(3):895–906. [PubMed]
23. Liepert J, Hamzei F, Weiler C. Motor cortex disinhibition of the unaffected hemisphere after acute stroke. Muscle Nerve. 2000;23:1761–1763. [PubMed]
24. Cicinelli P, Pasqualetti P, Zaccagnini M, Traversa R, Oliveri M, Rossini PM. Interhemispheric asymmetries of motor cortex excitability in the postacute stroke stage. Stroke. 2003;34:2653–2658. [PubMed]
25. Boroojerdi B, Diefenbach K, Ferbert A. Transcallosal inhibition in cortical and subcortical cerebral vascular lesions. J Neuro Sci. 1996;144:160–170. [PubMed]
26. Netz J, Lammers T, Homberg V. Reorganization of motor output in the non-affected hemisphere after stroke. Brain. 1997;120:1579–1586. [PubMed]
27. Shimizu T, Hosaki A, Hino T, Sato M, Komori T, Hirai S, Rossini PM. Motor cortical disinhibition in the unaffected hemisphere after unilateral cortical stroke. Brain. 2002;125:1896–1907. [PubMed]
28. Butefisch CM, Netz J, Webling M, Seita RJ, Homberg V. Remote changes in cortical excitability after stroke. Brain. 2003;126:470–481. [PubMed]
29. Petersen NT, Butler JE, Marchand-Pauvert V, et al. Suppression of EMG activity by transcranial magnetic stimulation in human subjects during walking. J Physiol. 2001;537(2):651–656. [PubMed]
30. Madhavan S, Rogers LM, Stinear JW. A paradox: after stroke the non-lesioned lower limb motor cortex may be maladaptive. Eur J Neruosci. 2010;32:1032–1039. [PMC free article] [PubMed]
31. Hess CW, Mills KR, Murray NM. Magnetic stimulation of the human brain: facilitation of motor responses by voluntary contraction of ipsilateral and contralateral muscles with additional observations on an amputee. Neurosci Lett. 1986;71:235–40. [PubMed]
32. Kischka U, Fajfr R, Fellenberg T, Hess CW. Facilitation of motor evoked potentials from magnetic brain stimulation in man: a comparative study of different target muscles. J Clin Neurophys. 1993;10(4):505–512. [PubMed]
33. Chow JW, Stokic DS. Impaired force steadiness in associated with changes in force frequency composition in subacute stroke. Neuroscience. 2013;242:69–77. [PubMed]
34. Chow JW, Stokic DS. Variability, frequency composition, and complexity of submaximal isometric knee extension force from subacute to chronic stroke. Neuroscience. 2014;273:18–198. [PubMed]
35. Roy FD, Gorassini MA. Peripheral sensory activation of cortical circuits in the leg motor cortex of man. J Physio. 2008;586:4091–4105. [PubMed]
36. Barthelemy D, Nielsen JB. Corticospinal contribution to arm muscle activity during human walking. J Physiol. 2010;588:967–979. [PubMed]
37. Garvey MA, Ziemann U, Becker DA, Barker CA, Bartko JJ. New graphical method to measure silent periods evoked by transcranial magnetic stimulation. Clin Neurophys. 2001;112(8):1451–1460. [PubMed]
38. Perez MA, Lungholt BK, Nyborg K, Nielsen JB. Motor skill training induces changes in the excitability of the leg cortical area in healthy humans. Exp Brain Res. 2004;159:197–205. [PubMed]
39. Kobayashi M, Hutchinson S, Theoret H, Schlaug G, Pascual-Leone A. Repetitive TMS of motor cortex improves ipsilateral sequential simple finger movements. Neurology. 2004;62:91–98. [PubMed]
40. Talelli P, Greenwood RJ, Rothwell JC. [Review] Arm function after stroke: Neurophysiological correlates and recovery mechanism assessed by transcranial magnetic stimulation. Clin Neurophys. 2006;117:1641–1659. [PubMed]
41. Madhavan S, Krishnan C, Jayaraman A, Rymer WZ, Stinear JW. Corticospinal tract integrity correlates with knee extensor weakness in chronic stroke survivors. Clin Neurophys. 2011;122:1588–1594. [PMC free article] [PubMed]
42. Allen JL, Kautz SA, Neptune RR. The influence of merged muscle excitation modules on post-stroke hemiparetic walking performance. Clin Biomech. 2013;28(6):697–704. [PMC free article] [PubMed]
43. Alagona G, Delvaux V, Gerard P, et al. Ipsilateral motor reponses to focal transcranial magnetic stimulation in healthy subjects and acute-stroke patients. Stroke. 2001;32:1304–1309. [PubMed]
44. Werhann KJ, Conforto AB, Kadom N, Hallett M, Cohen LG. Contribution of the ipsilateral motor cortex to recovery after chronic stroke. Ann Neurol. 2003;54:464–472. [PubMed]
45. Chen R. Interactions between inhibitory and excitatory circuits in the human motor cortex [Review] Exp Brain Res. 2004;154:1–10. [PubMed]
46. Kujirai T, Caramia MD, Rothwell JC, Day BL, Thompson PD, Ferbert A, et al. Corticocortical inhibition in human motor cortex. J Physiol. 1993;471:501–519. [PubMed]
47. Chen R, Tam A, Butefisch C, Corwell B, Ziemann U, Rothwell JC. Intracortical inhibition and facilitation in different representations of the human motor cortex. J Neurophysiol. 1998;80(6):2870–81. [PubMed]
48. Carr LJ, Harrison LM, Evans AL, Stephens JA. Patterns of central motor reorganization in hemiplegic cerbral palsy. Brain. 1993;116:1223–1247. [PubMed]
49. Basu AP, Turton A, Lemon RN. Activation of ipsilateral upper limb muscles by transcranial magnetic stimulation. J Physiol. 1994;479:144–145.
50. Jankowska E, Tanaka R. Projections of pyramidal tract cells to alpha-motoneruons innervating hind-limb muscles in the monkey. J Physiol. 1975;249:637–667. [PubMed]
51. Kuypers HG. Handbook of physiology, section 1: the nervous system. Vol II. Motor control. In: Brooks VB, editor. Anatomy of descending pathways. Bethesda: American Physiological Society; 1981. pp. 596–666.
52. Soteropoulos D, Edgley SA, Baker SN. Lack of evidence for direct corticospinal contributions to control of the ipsilateral forelimb in monkey. J Neurosci. 2011;31:11208–11219. [PMC free article] [PubMed]
53. Wasserman EM, Fuhr P, Cohen LG, Hallet M. Effects of transcranial magnetic stimulation on ipsilateral muscles. Neurology. 1991;41:1795–1799. [PubMed]
54. Hudson HM, Griffin DM, Belhaj-Saif A, Cheney PD. Properties of primary motor cortex output to hindlimb muscles in the macaque monkey. J Neurophys. 2014 doi: 10.1152/jn.00099.2014. [PubMed] [Cross Ref]
55. Bastings EP, Greenberg JP, Good DC. Hand motor recovery after stroke: a transcranial magnetic stimulation mapping study of motor output areas and their relation to functional status. Neurorehabil Neural Rep. 2002;16:275–282. [PubMed]
56. Caramia MD, Palmieri MG, Giacomini P, Iani C, Dally L, Silvestrini M. Ipsilateral activation of the unaffected motor cortex in patients with hemiparetic stroke. Clin Neurophys. 2000;111:1990–1996. [PubMed]
57. Nudo RJ, Wise BM, SiFuentes F, Milliken GW. Neural substrates for the effects of rehabilitative training on motor recovery after ischemic infarct. Science. 1996;272:1791–4. [PubMed]
58. Traversa R, Cicinelli P, Pasqualetti P, Filippi M, Rossini PM. Follow-up of interhemispheric differences of motor evoked potentials from the ‘affected’ and ‘unaffected’ hemispheres in human stroke. Brain Res. 1998;803:1–8. [PubMed]
59. Hultborn H, Nielsen JB. Spinal control of locomotion-from cat to man. Acta Physiol. 2007;189:111–121. [PubMed]
60. Jayaram G, Stinear JW. The effects of transcranial stimulation o paretic lower limb motor excitability during walking. J Clin Neurophysiol. 2009;26(4):272–9. [PubMed]
61. Khedr EM, Etraby AE, Hemeda M, Nasef AM, Razek AE. Long-term effect of repetitive transcranial magnetic stimulation on motor function recovery after acute ischemic stroke. Acta Neurol Scand. 2010;121:30–37. [PubMed]
62. Jayaram G, Stinear JW. Contralesinal paired associative stimulation increases paretic lower limb motor excitability post-stroke. Exp Brain Res. 2008;185(4):563–70. [PubMed]
63. Brouwer B, Ashby P. Corticopsinal projections to upper and lower limb spinal motoneurons in man. Electroencephalogr Clin Neurophysiol. 1990;76:509–19. [PubMed]
64. Drew T, Jiang W. Contributions of the motor cortex to the control of the hindlimbs during locomotion in the cat. Brain Research Reviews. 2002;40(1–3):178–191. [PubMed]
65. Schubert M, Curt A, Jensen L, Dietz V. Corticospinal input in human gait: modulation of magnetically evoked motor responses. Exp Brain Res. 1997;115:234–246. [PubMed]
66. Poon DE, Roy FD, Gorassini MA, Stein RB. Interaction of paired cortical and peripheral nerve stimulation on human motor neurons. Exp Brain Res. 2008;188:13–21. [PubMed]
67. Hanajima R, Ugawa Y, Machii K, Mochizuki H, Terao Y, Enomoto H. Interhemispheric facilitation of the hand motor area in humans. J Physiol. 2001;531:849–59. [PubMed]
68. Ferbert A, Priori A, Rothwell JC, Day BL, Colebatch JG, Marsden CD. Interhemispheric Inhibition of the human motor cortex. J Physiol. 1991;453:525–546. [PubMed]
69. Murase N, Duque J, Mazzaochio R, Cohen LG. Influence of interhemispheric interactions on motor function in chronic stroke. Ann Neurol. 2004;55:400–409. [PubMed]
70. Fregni F, Boggio PS, Mansur CG, et al. Transcranial direct current stimulation of the unaffected hemisphere in stroke patients. Neuroreport. 2005;16:1551–5. [PubMed]
71. Mansur CG, Fregni F, Boggio PS, et al. A sham stimulation-controlled trial of rTMS of the unaffected hemisphere in stroke patients. Neurology. 2005;64:1802–4. [PubMed]
72. Beaulieu LD, Masse-Alarie Brouwer B, Schneider C. Brain control of volitional ankle tasks in people with chronic stroke and in healthy individuals. J of the Neurol Sci. 2014:148–155. [PubMed]