- Email: [email protected]
Cortical facilitation of cutaneous reflexes in leg muscles during human gait
Brain Research 787 Ž1998. 149–153
Short communication
Cortical facilitation of cutaneous reflexes in leg muscles during human gait M. Pijnappels a
a,1
, B.M.H. Van Wezel a , G. Colombo b , V. Dietz b , J. Duysens
a, )
Department of Medical Physics and Biophysics, UniÕersity of Nijmegen, Geert Grooteplein 21, 6525 EZ Nijmegen, Netherlands Swiss Paraplegic Center, UniÕersity Hospital Balgrist, UniÕersity of Zurich, Forchstrasse 340, CH-8008 Zurich, Switzerland ¨ ¨
b
Accepted 9 December 1997
Abstract During human gait, cortical convergence on sural nerve reflex pathways was investigated by means of transcranial magnetic stimulation ŽTMS. of the cortex in five phases of the step cycle during human walking on a treadmill. Muscular responses to paired electrical and magnetic stimulation were compared with the linear summation of the individual stimuli. For both the tibialis anterior ŽTA. and biceps femoris ŽBF. muscles, the averaged data of four subjects showed a significant facilitation mainly in the swing phase of the step cycle. It is suggested that facilitation of corticospinal input onto cutaneous reflex pathways is enhanced specifically in these periods of the step cycle. q 1998 Elsevier Science B.V. Keywords: Human gait; Cutaneous reflexes; Sural nerve; Cortical facilitation; Magnetic stimulation; Tibialis anterior; Biceps femoris
In many leg muscles the stimulation of cutaneous afferents from the foot elicits reflex responses with an amplitude and sign which depends on the phase of the step cycle during gait. This is functionally meaningful since reflexes may be useful in some phases but unwanted in others Žfor example, a flexor reflex is appropriate at the onset of the swing phase when the leg is flexed, but the same reflex is not convenient at end swing when the foot is ready to take up body weight.. Such phase-dependent modulation of reflexes has been demonstrated both in cat w7,3x and in man w15,4,13x, but at present it is still largely unknown how this modulation is achieved. Some of the modulation may be provided by a locomotor pattern generator Žfor most recent evidence, see Ref. w6x. but it is also possible that interaction between afferent inputs is important Žfor a review, see Ref. w2x. or that supraspinal sources play a role. To examine the latter possibility one should combine cutaneous input with activation of supraspinal centres to see whether there is facilitation or suppression in those phases of the cycle where one normally observes enhanced or suppressed reflexes. Facilitatory reflex responses in leg muscles following stimulation of the sural nerve Žinnervating the lateral side )
Corresponding author. Fax: q 31-243-54-1435; E-mail: [email protected] 1 Current address: Department of Movement Sciences, University of Maastricht, P.O. Box 616, 6200 MD Maastricht, the Netherlands. 0006-8993r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 7 . 0 1 5 5 7 - 6
of the foot. are larger during gait than at rest w5x and in ankle flexors such as tibialis anterior ŽTA. the responses are largest at the onset of the swing phase. Such enhancement of cutaneous reflexes might have a cortical origin since it has long been known that the motor cortex can influence transmission in cutaneous reflex pathways both in man w1x and in cat w9x. Recently, transcranial magnetic stimulation ŽTMS. has been used in humans to examine cortical facilitation of cutaneous reflexes. In immobile humans, Nielsen et al. w8x showed that during tonic voluntary contractions the discharge probability of TA motor units following combined electrical sural and magnetic cortical stimulation was larger than the algebraic sum of the discharge probability following each of the two stimuli separately. This facilitation was absent when cortical stimulation was applied by electrical Ždirect. instead of magnetical Žindirect. impulses. Hence, the authors concluded that the magnetic facilitation was due to increased susceptibility of the corticomotoneuronal cells rather than to an interaction between the cutaneous and corticospinal volleys at a subcortical site. If the facilitation of sural nerve evoked responses during gait is also related to cortical input then a first requirement is that magnetic stimulation should yield extra facilitation of the responses during gait, similar to the facilitations seen during voluntary contractions. This was investigated in four healthy adult subjects. Muscle activity of the TA and biceps femoris ŽBF. mus-
150
M. Pijnappels et al.r Brain Research 787 (1998) 149–153
cles was measured by means of surface EMG. During locomotion on a treadmill at a comfortable speed of 4 kmrh, cutaneous reflex responses were elicited by electrical stimulation of the sural nerve. Electrical stimulation was given through a bipolar stimulation electrode, positioned on the sural nerve at the ankle Žstimulated side is ipsilateral leg.. The electrical stimulus consisted of a brief pulse train Ž5 pulses of 1 ms duration at a frequency of 200 Hz., as described in previous studies w4,12,13x. During quiet standing, the perception threshold ŽPT. was determined. During gait the electrical stimulus intensity was set at 2 PT. Transcranial magnetic stimulation ŽTMS. of the motor cortex was applied to provoke motor evoked potentials ŽMEP. in the lower limb muscles. Typically, MEPs were observed at a latency of about 30 ms and duration of 20 ms. TMS of the motor cortex was delivered by a magnetic stimulator ŽMagstim Quadropulse TM Model 500, Magstim. through a double cone coil. The coil was positioned centrally over Cz Žvertex. and the optimal current direction Žposterior or anterior. was determined for every subject individually to yield the best MEP responses. The magnetic stimulus intensity was set at motor threshold ŽMT., which was determined in the TA muscle in an
upright standing position. To ensure that the stimulation site was kept stable during walking, a harness was used to fix the head of the subject with respect to the trunk Žsee Ref. w11x.. The stimulation coil was mounted on the harness with an adjustable clamp. The effect of the magnetic stimulation was tested on P2 responses, which typically appear in TA and BF within a time window of 80–130 ms and 70–110 ms respectively, following sural nerve stimulation w6x. The magnetic stimulation was timed such that the onset of the magnetic evoked motor response fell on one of five phases corresponding to 20%, 40%, 60%, 80% and 100% of the step cycle Žsee Fig. 1A.. For both muscles the MEPs were measured within a window of 30 and 50 ms following magnetic stimulation. To study the interactions, the following four conditions were analysed: control condition without stimulus ŽCO., only electrical stimulus ŽE., only magnetic stimulus ŽM. and both electrical and magnetic stimuli ŽEM.. As there is some variability in the latencies of the responses and facilitations, four different time intervals were taken for the EM condition Želectrical stimulus preceding the magnetic stimulation with 40, 50, 60 or 70 ms.. For comparison, the same four conditions were repeated without magnetic stimulation Žfour E conditions resulting
Fig. 1. ŽA. Averaged EMG pattern of a stride cycle of the ipsilateral TA muscle for one representative subject. The bar at the top indicates the stance phase; the triangles at the bottom represent the five phases Žin percentages of the step cycle. at which the onset of the stimulus responses occurred. ŽB. Graph of results of averaged Ž10 trials; qS.D.. response level, expressed as percentage of maximum voluntary contraction of the TA muscle for the respective subject, for each of the five phases of the step cycle. CO s control condition; E s electrical stimulation; M s magnetic stimulation; EM s both electrical and magnetic stimulation.
M. Pijnappels et al.r Brain Research 787 (1998) 149–153
in electrical stimuli given 70 to 100 ms before the five phase points at 20%, 40%, 60%, 80% and 100% of the step cycle.. As can be seen in Fig. 1B this resulted in a total of 10 conditions for each of the five phases. All 50 conditions were presented 10 times in random order. Successive stimulus presentations were separated by two or three non-stimulated step cycles. Data were sampled in blocks of 1600 ms at 2000 Hz. The EMG signals were amplified and bandpass filtered Ž30–1000 Hz., fullwave rectified and stored on hard disk along with the other signals Žipsi- and contralateral ground contact forces, triggering of heel strike and codes referring to the electrical and magnetic stimulus output.. The average response of the 10 trials was calculated within a window of 20 ms, the onset of which corresponded to the five phases mentioned above. The resulting data underwent an amplitude-normalization procedure by scaling the EMG values with respect to the measured maximum voluntary contraction value ŽFig. 1.. To test whether sural nerve reflex pathways and cortical descending pathways converge during human gait, the responses to the combined stimulation ŽEM. were compared with the individual responses. If these two neural systems are independent, one would expect that EM responses are equal to the algebraic sum of the separate E and M responses Žpredicted responses.. On the other hand,
151
if EM responses are larger than the summation of E and M responses, then one has to assume that there is convergence in the pathways for the two types of responses Žsee above.. Fig. 1A shows the averaged EMG of 10 stride cycles of the ipsilateral TA muscle for one of the four investigated subjects. In Fig. 1B, for each phase, the results of the averaged response levels for the various stimulus conditions are represented. Electrical stimulation Žopen bars. yielded facilitatory responses that were well above background EMG Žfirst bar of each group. at early Ž60%. and middle swing Ž80%., while suppressive responses could be discerned at end swing Ž100% of the step cycle.. This phase-dependent reflex reversal conformed to the one described previously w15,4,13x. The same type of modulation was seen in the EM conditions, except that the facilitatory responses were even larger than those observed for the E condition because of the additional magnetic response. Fig. 2 shows the same type of analysis for the ipsilateral BF muscle. The magnetic responses were much smaller in this muscle than in TA but cutaneous facilitatory responses were much larger and present in each phase of the step cycle. For comparison of the predicted Žlinear summation of E and M responses. with the measured responses to EM combined stimulation, the data for each phase were pooled
Fig. 2. ŽA. Averaged EMG pattern of a stride cycle of the ipsilateral BF muscle of the same subject as in Fig. 1. ŽB. Graph of results of averaged responses of the BF muscle, same as in Fig. 1.
152
M. Pijnappels et al.r Brain Research 787 (1998) 149–153
Fig. 3. ŽA. Predicted and measured responses for the TA muscle Žaverages of four subjects and four interstimulus intervals.; Ž"S.D.. for each phase of the step cycle Žin percentages of the step cycle.. The measured responses are normalized with respect to the predicted responses, which are set at 100% for each phase. Significant differences Ž p - 0,05. between the predicted and measured response are indicated with ). ŽB. Same for the BF muscle.
Ž40 trials for the four interstimulus intervals.. This was justified as the reflex responses showed the same pattern of facilitation or suppression in each interval at any given phase. Pooling of the data of all four subjects thus yielded 160 predicted and 160 measured responses for each phase. The statistical significance of the difference between predicted and measured response was tested with a multivariate ANOVA Žsignificance at p - 0.05.. Three out of four subjects showed significant facilitations of the measured response in the TA muscle in several phases Žranging from 1 to 3. of the step cycle. One of them showed also a significant suppression at the 100%-phase of the step cycle. For the BF muscle, all four subjects showed significant facilitations in at least two phases of the step cycle. Fig. 3 shows the averaged predicted and measured responses for each phase of the step cycle for both TA and BF muscles for all four subjects. It shows that at the population level, the measured TA responses were significantly larger than the predicted responses for three phases of the step cycle. For BF a significant facilitation occurred at early swing Žas for TA. and at end swing. This BF facilitation is important since it is the first time that EM facilitation could be demonstrated in an upper leg muscle Žfor a review, see Ref. w10x..
The finding of TA facilitation after combined EM stimulation is in line with the results of others. Nielsen et al. w8x found facilitatory effects during slight voluntary leg muscle contraction while Wolfe and Hayes w14x reported facilitation during relaxation. The present data show that a similar facilitation occurs during gait, but not equally strong in all phases of the step cycle. The strong EM facilitation during early swing Ž60%. in both muscles is of interest since sural nerve stimulation yields prominent phase-dependent facilitation in both of these muscles during this period of the step cycle w6x. As mentioned in the introduction, it is possible that this facilitation is either due to CPG actions, to interaction between afferents or to supraspinal input from the motor cortex. The present data support the latter possibility.
Acknowledgements We would like to thank L. Jensen for excellent technical and programming assistance, P. Praamstra and D.F. Stegeman for their help with the preliminary experiments, H.W.A.A. Van de Crommert for participation in part of the experiments and I. Schillings for helpful suggestions.
M. Pijnappels et al.r Brain Research 787 (1998) 149–153
V. Dietz was supported by NATO grant 910574 and by a grant from the Swiss National Science Foundation ŽNo. 31-42899.95. and from the Schweizerische Bankgesellschaft on behalf of a client. References w1x J. Babinski, Du phenomene ´ ` des orteils et de sa valeur semiologique, ´ Semaine Medical 18 Ž1898. 321–322. ´ w2x J.D. Brooke, J. Cheng, D.F. Collins, W.E. Mcilroy, J.E. Misiaszek, W.R. Staines, Sensori–sensory afferent conditioning with leg movement: gain control in spinal reflex and ascending paths, Prog. Neurobiol. 51 Ž1997. 393–421. w3x J. Duysens, K.G. Pearson, The role of cutaneous afferents from the distal hindlimb in the regulation of the step cycle of thalamic cats, Exp. Brain Res. 24 Ž1976. 245–255. w4x J. Duysens, M. Trippel, G.A. Horstmann, V. Dietz, Gating and reversal of reflexes in ankle muscles during human walking, Exp. Brain Res. 82 Ž1990. 351–358. w5x J. Duysens, A.A.M. Tax, M. Trippel, V. Dietz, Increased amplitude of cutaneous reflexes during human running as compared to standing, Brain Res. 613 Ž1993. 230–238. w6x J. Duysens, A.A.M. Tax, L. Murrer, V. Dietz, Backward and forward walking use different patterns of phase-dependent modulation of cutaneous reflexes in humans, J. Neurophysiol. 76 Ž1996. 301–310.
153
w7x H. Forssberg, S. Grillner, S. Rossignol, Phase dependent reflex reversal during walking in chronic spinal cats, Brain Res. 85 Ž1. Ž1975. 103–107. w8x J. Nielsen, N. Petersen, B. Fedirchuk, Evidence suggesting a transcortical pathway from cutaneous foot afferents to tibialis anterior motoneurones in man, J. Physiol. 501 Ž2. Ž1997. 473–484. w9x M.J. Pinter, R.E. Burke, M.J. O’Donovan, R.P. Dum, Supraspinal facilitation of cutaneous polysynaptic EPSPs in cat medial gastrocnemius motoneurons, Exp. Brain Res. 45 Ž1982. 133–143. w10x J.C. Rothwell, Techniques and mechanisms of action of transcranial stimulation of the human motor cortex, J. Neurosci. Meth. 74 Ž1997. 113–122. w11x M. Schubert, A. Curt, L. Jensen, V. Dietz, Cortical input in gait: modulation of magnetically evoked motor responses, Exp. Brain Res. 115 Ž1997. 234–246. w12x A.A.M. Tax, B.M.H. Van Wezel, V. Dietz, Bipedal reflex coordination to tactile stimulation of the sural nerve during human running, J. Neurophysiol. 73 Ž1995. 1947–1964. w13x B.M.H. Van Wezel, F.A.M. Ottenhoff, J. Duysens, Dynamic control of location-specific information in tactile cutaneous reflexes from the foot during human walking, J. Neurosci. 17 Ž10. Ž1997. 3804– 3814. w14x D.L. Wolfe, K.C. Hayes, Conditioning effects of sural nerve stimulation on short and long latency motor evoked potentials in lower limb muscles, Electroencephalogr. Clin. Neurophysiol. 97 Ž1995. 11–17. w15x J.F. Yang, R.B. Stein, Phase-dependent reflex reversal in human leg muscles during walking, J. Neurophysiol. 63 Ž1990. 1109–1117.
Short communication
Cortical facilitation of cutaneous reflexes in leg muscles during human gait M. Pijnappels a
a,1
, B.M.H. Van Wezel a , G. Colombo b , V. Dietz b , J. Duysens
a, )
Department of Medical Physics and Biophysics, UniÕersity of Nijmegen, Geert Grooteplein 21, 6525 EZ Nijmegen, Netherlands Swiss Paraplegic Center, UniÕersity Hospital Balgrist, UniÕersity of Zurich, Forchstrasse 340, CH-8008 Zurich, Switzerland ¨ ¨
b
Accepted 9 December 1997
Abstract During human gait, cortical convergence on sural nerve reflex pathways was investigated by means of transcranial magnetic stimulation ŽTMS. of the cortex in five phases of the step cycle during human walking on a treadmill. Muscular responses to paired electrical and magnetic stimulation were compared with the linear summation of the individual stimuli. For both the tibialis anterior ŽTA. and biceps femoris ŽBF. muscles, the averaged data of four subjects showed a significant facilitation mainly in the swing phase of the step cycle. It is suggested that facilitation of corticospinal input onto cutaneous reflex pathways is enhanced specifically in these periods of the step cycle. q 1998 Elsevier Science B.V. Keywords: Human gait; Cutaneous reflexes; Sural nerve; Cortical facilitation; Magnetic stimulation; Tibialis anterior; Biceps femoris
In many leg muscles the stimulation of cutaneous afferents from the foot elicits reflex responses with an amplitude and sign which depends on the phase of the step cycle during gait. This is functionally meaningful since reflexes may be useful in some phases but unwanted in others Žfor example, a flexor reflex is appropriate at the onset of the swing phase when the leg is flexed, but the same reflex is not convenient at end swing when the foot is ready to take up body weight.. Such phase-dependent modulation of reflexes has been demonstrated both in cat w7,3x and in man w15,4,13x, but at present it is still largely unknown how this modulation is achieved. Some of the modulation may be provided by a locomotor pattern generator Žfor most recent evidence, see Ref. w6x. but it is also possible that interaction between afferent inputs is important Žfor a review, see Ref. w2x. or that supraspinal sources play a role. To examine the latter possibility one should combine cutaneous input with activation of supraspinal centres to see whether there is facilitation or suppression in those phases of the cycle where one normally observes enhanced or suppressed reflexes. Facilitatory reflex responses in leg muscles following stimulation of the sural nerve Žinnervating the lateral side )
Corresponding author. Fax: q 31-243-54-1435; E-mail: [email protected] 1 Current address: Department of Movement Sciences, University of Maastricht, P.O. Box 616, 6200 MD Maastricht, the Netherlands. 0006-8993r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 7 . 0 1 5 5 7 - 6
of the foot. are larger during gait than at rest w5x and in ankle flexors such as tibialis anterior ŽTA. the responses are largest at the onset of the swing phase. Such enhancement of cutaneous reflexes might have a cortical origin since it has long been known that the motor cortex can influence transmission in cutaneous reflex pathways both in man w1x and in cat w9x. Recently, transcranial magnetic stimulation ŽTMS. has been used in humans to examine cortical facilitation of cutaneous reflexes. In immobile humans, Nielsen et al. w8x showed that during tonic voluntary contractions the discharge probability of TA motor units following combined electrical sural and magnetic cortical stimulation was larger than the algebraic sum of the discharge probability following each of the two stimuli separately. This facilitation was absent when cortical stimulation was applied by electrical Ždirect. instead of magnetical Žindirect. impulses. Hence, the authors concluded that the magnetic facilitation was due to increased susceptibility of the corticomotoneuronal cells rather than to an interaction between the cutaneous and corticospinal volleys at a subcortical site. If the facilitation of sural nerve evoked responses during gait is also related to cortical input then a first requirement is that magnetic stimulation should yield extra facilitation of the responses during gait, similar to the facilitations seen during voluntary contractions. This was investigated in four healthy adult subjects. Muscle activity of the TA and biceps femoris ŽBF. mus-
150
M. Pijnappels et al.r Brain Research 787 (1998) 149–153
cles was measured by means of surface EMG. During locomotion on a treadmill at a comfortable speed of 4 kmrh, cutaneous reflex responses were elicited by electrical stimulation of the sural nerve. Electrical stimulation was given through a bipolar stimulation electrode, positioned on the sural nerve at the ankle Žstimulated side is ipsilateral leg.. The electrical stimulus consisted of a brief pulse train Ž5 pulses of 1 ms duration at a frequency of 200 Hz., as described in previous studies w4,12,13x. During quiet standing, the perception threshold ŽPT. was determined. During gait the electrical stimulus intensity was set at 2 PT. Transcranial magnetic stimulation ŽTMS. of the motor cortex was applied to provoke motor evoked potentials ŽMEP. in the lower limb muscles. Typically, MEPs were observed at a latency of about 30 ms and duration of 20 ms. TMS of the motor cortex was delivered by a magnetic stimulator ŽMagstim Quadropulse TM Model 500, Magstim. through a double cone coil. The coil was positioned centrally over Cz Žvertex. and the optimal current direction Žposterior or anterior. was determined for every subject individually to yield the best MEP responses. The magnetic stimulus intensity was set at motor threshold ŽMT., which was determined in the TA muscle in an
upright standing position. To ensure that the stimulation site was kept stable during walking, a harness was used to fix the head of the subject with respect to the trunk Žsee Ref. w11x.. The stimulation coil was mounted on the harness with an adjustable clamp. The effect of the magnetic stimulation was tested on P2 responses, which typically appear in TA and BF within a time window of 80–130 ms and 70–110 ms respectively, following sural nerve stimulation w6x. The magnetic stimulation was timed such that the onset of the magnetic evoked motor response fell on one of five phases corresponding to 20%, 40%, 60%, 80% and 100% of the step cycle Žsee Fig. 1A.. For both muscles the MEPs were measured within a window of 30 and 50 ms following magnetic stimulation. To study the interactions, the following four conditions were analysed: control condition without stimulus ŽCO., only electrical stimulus ŽE., only magnetic stimulus ŽM. and both electrical and magnetic stimuli ŽEM.. As there is some variability in the latencies of the responses and facilitations, four different time intervals were taken for the EM condition Želectrical stimulus preceding the magnetic stimulation with 40, 50, 60 or 70 ms.. For comparison, the same four conditions were repeated without magnetic stimulation Žfour E conditions resulting
Fig. 1. ŽA. Averaged EMG pattern of a stride cycle of the ipsilateral TA muscle for one representative subject. The bar at the top indicates the stance phase; the triangles at the bottom represent the five phases Žin percentages of the step cycle. at which the onset of the stimulus responses occurred. ŽB. Graph of results of averaged Ž10 trials; qS.D.. response level, expressed as percentage of maximum voluntary contraction of the TA muscle for the respective subject, for each of the five phases of the step cycle. CO s control condition; E s electrical stimulation; M s magnetic stimulation; EM s both electrical and magnetic stimulation.
M. Pijnappels et al.r Brain Research 787 (1998) 149–153
in electrical stimuli given 70 to 100 ms before the five phase points at 20%, 40%, 60%, 80% and 100% of the step cycle.. As can be seen in Fig. 1B this resulted in a total of 10 conditions for each of the five phases. All 50 conditions were presented 10 times in random order. Successive stimulus presentations were separated by two or three non-stimulated step cycles. Data were sampled in blocks of 1600 ms at 2000 Hz. The EMG signals were amplified and bandpass filtered Ž30–1000 Hz., fullwave rectified and stored on hard disk along with the other signals Žipsi- and contralateral ground contact forces, triggering of heel strike and codes referring to the electrical and magnetic stimulus output.. The average response of the 10 trials was calculated within a window of 20 ms, the onset of which corresponded to the five phases mentioned above. The resulting data underwent an amplitude-normalization procedure by scaling the EMG values with respect to the measured maximum voluntary contraction value ŽFig. 1.. To test whether sural nerve reflex pathways and cortical descending pathways converge during human gait, the responses to the combined stimulation ŽEM. were compared with the individual responses. If these two neural systems are independent, one would expect that EM responses are equal to the algebraic sum of the separate E and M responses Žpredicted responses.. On the other hand,
151
if EM responses are larger than the summation of E and M responses, then one has to assume that there is convergence in the pathways for the two types of responses Žsee above.. Fig. 1A shows the averaged EMG of 10 stride cycles of the ipsilateral TA muscle for one of the four investigated subjects. In Fig. 1B, for each phase, the results of the averaged response levels for the various stimulus conditions are represented. Electrical stimulation Žopen bars. yielded facilitatory responses that were well above background EMG Žfirst bar of each group. at early Ž60%. and middle swing Ž80%., while suppressive responses could be discerned at end swing Ž100% of the step cycle.. This phase-dependent reflex reversal conformed to the one described previously w15,4,13x. The same type of modulation was seen in the EM conditions, except that the facilitatory responses were even larger than those observed for the E condition because of the additional magnetic response. Fig. 2 shows the same type of analysis for the ipsilateral BF muscle. The magnetic responses were much smaller in this muscle than in TA but cutaneous facilitatory responses were much larger and present in each phase of the step cycle. For comparison of the predicted Žlinear summation of E and M responses. with the measured responses to EM combined stimulation, the data for each phase were pooled
Fig. 2. ŽA. Averaged EMG pattern of a stride cycle of the ipsilateral BF muscle of the same subject as in Fig. 1. ŽB. Graph of results of averaged responses of the BF muscle, same as in Fig. 1.
152
M. Pijnappels et al.r Brain Research 787 (1998) 149–153
Fig. 3. ŽA. Predicted and measured responses for the TA muscle Žaverages of four subjects and four interstimulus intervals.; Ž"S.D.. for each phase of the step cycle Žin percentages of the step cycle.. The measured responses are normalized with respect to the predicted responses, which are set at 100% for each phase. Significant differences Ž p - 0,05. between the predicted and measured response are indicated with ). ŽB. Same for the BF muscle.
Ž40 trials for the four interstimulus intervals.. This was justified as the reflex responses showed the same pattern of facilitation or suppression in each interval at any given phase. Pooling of the data of all four subjects thus yielded 160 predicted and 160 measured responses for each phase. The statistical significance of the difference between predicted and measured response was tested with a multivariate ANOVA Žsignificance at p - 0.05.. Three out of four subjects showed significant facilitations of the measured response in the TA muscle in several phases Žranging from 1 to 3. of the step cycle. One of them showed also a significant suppression at the 100%-phase of the step cycle. For the BF muscle, all four subjects showed significant facilitations in at least two phases of the step cycle. Fig. 3 shows the averaged predicted and measured responses for each phase of the step cycle for both TA and BF muscles for all four subjects. It shows that at the population level, the measured TA responses were significantly larger than the predicted responses for three phases of the step cycle. For BF a significant facilitation occurred at early swing Žas for TA. and at end swing. This BF facilitation is important since it is the first time that EM facilitation could be demonstrated in an upper leg muscle Žfor a review, see Ref. w10x..
The finding of TA facilitation after combined EM stimulation is in line with the results of others. Nielsen et al. w8x found facilitatory effects during slight voluntary leg muscle contraction while Wolfe and Hayes w14x reported facilitation during relaxation. The present data show that a similar facilitation occurs during gait, but not equally strong in all phases of the step cycle. The strong EM facilitation during early swing Ž60%. in both muscles is of interest since sural nerve stimulation yields prominent phase-dependent facilitation in both of these muscles during this period of the step cycle w6x. As mentioned in the introduction, it is possible that this facilitation is either due to CPG actions, to interaction between afferents or to supraspinal input from the motor cortex. The present data support the latter possibility.
Acknowledgements We would like to thank L. Jensen for excellent technical and programming assistance, P. Praamstra and D.F. Stegeman for their help with the preliminary experiments, H.W.A.A. Van de Crommert for participation in part of the experiments and I. Schillings for helpful suggestions.
M. Pijnappels et al.r Brain Research 787 (1998) 149–153
V. Dietz was supported by NATO grant 910574 and by a grant from the Swiss National Science Foundation ŽNo. 31-42899.95. and from the Schweizerische Bankgesellschaft on behalf of a client. References w1x J. Babinski, Du phenomene ´ ` des orteils et de sa valeur semiologique, ´ Semaine Medical 18 Ž1898. 321–322. ´ w2x J.D. Brooke, J. Cheng, D.F. Collins, W.E. Mcilroy, J.E. Misiaszek, W.R. Staines, Sensori–sensory afferent conditioning with leg movement: gain control in spinal reflex and ascending paths, Prog. Neurobiol. 51 Ž1997. 393–421. w3x J. Duysens, K.G. Pearson, The role of cutaneous afferents from the distal hindlimb in the regulation of the step cycle of thalamic cats, Exp. Brain Res. 24 Ž1976. 245–255. w4x J. Duysens, M. Trippel, G.A. Horstmann, V. Dietz, Gating and reversal of reflexes in ankle muscles during human walking, Exp. Brain Res. 82 Ž1990. 351–358. w5x J. Duysens, A.A.M. Tax, M. Trippel, V. Dietz, Increased amplitude of cutaneous reflexes during human running as compared to standing, Brain Res. 613 Ž1993. 230–238. w6x J. Duysens, A.A.M. Tax, L. Murrer, V. Dietz, Backward and forward walking use different patterns of phase-dependent modulation of cutaneous reflexes in humans, J. Neurophysiol. 76 Ž1996. 301–310.
153
w7x H. Forssberg, S. Grillner, S. Rossignol, Phase dependent reflex reversal during walking in chronic spinal cats, Brain Res. 85 Ž1. Ž1975. 103–107. w8x J. Nielsen, N. Petersen, B. Fedirchuk, Evidence suggesting a transcortical pathway from cutaneous foot afferents to tibialis anterior motoneurones in man, J. Physiol. 501 Ž2. Ž1997. 473–484. w9x M.J. Pinter, R.E. Burke, M.J. O’Donovan, R.P. Dum, Supraspinal facilitation of cutaneous polysynaptic EPSPs in cat medial gastrocnemius motoneurons, Exp. Brain Res. 45 Ž1982. 133–143. w10x J.C. Rothwell, Techniques and mechanisms of action of transcranial stimulation of the human motor cortex, J. Neurosci. Meth. 74 Ž1997. 113–122. w11x M. Schubert, A. Curt, L. Jensen, V. Dietz, Cortical input in gait: modulation of magnetically evoked motor responses, Exp. Brain Res. 115 Ž1997. 234–246. w12x A.A.M. Tax, B.M.H. Van Wezel, V. Dietz, Bipedal reflex coordination to tactile stimulation of the sural nerve during human running, J. Neurophysiol. 73 Ž1995. 1947–1964. w13x B.M.H. Van Wezel, F.A.M. Ottenhoff, J. Duysens, Dynamic control of location-specific information in tactile cutaneous reflexes from the foot during human walking, J. Neurosci. 17 Ž10. Ž1997. 3804– 3814. w14x D.L. Wolfe, K.C. Hayes, Conditioning effects of sural nerve stimulation on short and long latency motor evoked potentials in lower limb muscles, Electroencephalogr. Clin. Neurophysiol. 97 Ž1995. 11–17. w15x J.F. Yang, R.B. Stein, Phase-dependent reflex reversal in human leg muscles during walking, J. Neurophysiol. 63 Ž1990. 1109–1117.