Eur J Appl Physiol (2003) 90: 601–607
DOI 10.1007/s00421-003-0914-3
O R I GI N A L A R T IC L E
Nicola A. Maffiuletti Æ Manuela Pensini Æ Gil Scaglioni
Alessandra Ferri Æ Yves Ballay Æ Alain Martin
Effect of electromyostimulation training on soleus and gastrocnemii
H- and T-reflex properties
Accepted: 20 June 2003 / Published online: 16 August 2003

Springer-Verlag 2003
Abstract When muscle is artificially activated, as with
electromyostimulation (EMS), action potentials are
evoked in both intramuscular nerve branches and cuta-
neous receptors, therefore activating spinal motoneu-
rons reflexively. Maximal soleus and gastrocnemii
H- and T-reflex and the respective mechanical output
were thus quantified to examine possible neural adap-
tations induced at the spinal level by EMS resistance
training. Eight subjects completed 16 sessions of iso-
metric EMS (75 Hz) over a 4-week period. Maximal
soleus and gastrocnemii M wave (Mmax), H reflex (Hmax)
and T reflex (Tmax) were compared between before and
after training, together with the corresponding plantar
flexor peak twitch torque. No significant changes were
observed for electromechanical properties of Hmax reflex
following EMS. On the other hand, peak twitch torque
produced by Tmax, but not by equal-amplitude H reflex,
significantly increased as a result of training (+21%,
P<0.05). These changes were associated with a trend
towards a significant increase for normalized gastro-
cnemii (+21%, P=0.07) but not soleus Tmax reflex. It is
concluded that, contrary to results previously obtained
after voluntary physical training, EMS training of the
plantar flexor muscles did not affect alpha motoneuron
excitability and/or presynaptic inhibition, as indicated
by H-reflex results. On the other hand, in the absence of
change in a control group, Tmax electromechanical
findings indicated that: (1) equal-amplitude H- and
T-reflex adapted differently to EMS resistance training;
and (2) EMS had an effect on gastrocnemii but not
on soleus muscle, perhaps because of the differences
N. A. Maffiuletti (&) Æ M. Pensini
G. Scaglioni Æ A. Ferri Æ Y. Ballay
A. Martin
INSERM/ERIT-M 0207 Motricité-Plasticité,
Faculté des Sciences du Sport,
Université de Bourgogne, BP 27877–21078
Dijon Cedex, France
E-mail: Nicola.Maffiuletti@u-bourgogne.fr
Tel.: +33-38-0396758
Fax: +33-38-0396702
in respective motor unit characteristics (e.g., axon
diameter).
Keywords Electrostimulation Æ H reflex Æ M wave Æ
Peak twitch Æ T reflex
Introduction
During the last few years, increasing attention has been
drawn to electromyostimulation (EMS) as a modality
for rehabilitation in patients (Enoka 1988; Morrissey
1988), and more recently as a modality for strength
training in healthy subjects. In fact, it is well established
that the application of electric shocks to a muscle results
in maximal voluntary strength gains (Eriksson et al.
1981; Halbach and Straus 1980; Kots et al. 1971, quoted
in Ward and Shkuratova 2002). Although the physio-
logical adaptations induced by this form of involuntary
training remain equivocal, many studies on healthy hu-
man muscles have proposed that neural factors, rather
than changes at the muscle level, largely account for the
strength gains observed after brief EMS training periods
(Colson et al. 2000; Enoka et al. 1988; Maffiuletti et al.
2002). This conclusion could appear paradoxical since
EMS is often considered a technique to activate muscles
without involving the nervous system. In intact systems,
however, the stimulus evokes action potentials in both
intramuscular nerve branches and cutaneous receptors,
thus generating force directly by activation of motor
axons and indirectly by ‘‘reflex’’ recruitment of spinal
motoneurons (Collins et al. 2001). The contribution of
such centrally propagating sensory feedback during
stimulated contractions is confirmed by the fact that
both neural activation and cross-education of the mus-
cles trained by EMS significantly increased in previous
reports (Hortobagyi et al. 1999; Maffiuletti et al. 2002).
As detailed in our previous work (Maffiuletti et al.
2002), a 4-week training program of 75-Hz EMS of the
plantar flexor muscles resulted in significant increases
602
(12%) of agonist electromyographic activity normal-
ized to the maximal M-wave and activation level, as
estimated by the twitch interpolation technique. Such an
EMS training-induced increase in neural activation was
associated with enhanced volitional drive from the
supraspinal centers inducing greater activation of the
muscles that assist the prime movers. Whether such
adaptation is also a consequence of spinal mechanisms
remains however unknown, since no longitudinal studies
have previously investigated the neural adaptations in-
duced by involuntary physical training (e.g., EMS)
through recordings of evoked reflex responses.
On the other hand, human studies have demonstrated
that reflex excitability can be modified after multiple
sessions of voluntary physical training. For example, the
electrically evoked H-reflex has been shown to increase
following resistance training (Aagaard et al. 2002) but
also endurance training (Pérot et al. 1991). In the same
way, the amplitude of reflex potentials elicited through
mechanical stimuli (i.e., tendon jerk or T-reflex) signifi-
cantly increased as a result of two-legged hopping (Voigt
et al. 1998) and endurance running training (Pérot et al.
1991). Collectively, these findings suggest that voluntary
chronic physical activity can change the efficacy of the
connections between afferents and motoneurons. How-
ever, artificial activation can also result in spinal cord
plasticity, because of the centrally propagating sensory
signals associated with contractions triggered by EMS
(see above). Consequently, changes in activation prop-
erties of the spinal neural circuitry could be hypothe-
sized following intensive EMS training.
The purpose of the present study, therefore, was to
examine the effects of a 4-week EMS program of the
plantar flexor muscles on the amplitude of the maximal
soleus and gastrocnemii reflex responses evoked by two
different techniques (i.e., stimulation of the tibial nerve,
H-reflex, and percussion of the Achilles tendon, T-reflex)
and the potential training-induced changes on the
associated plantar flexor mechanical output (i.e., peak
twitch torque). This study uses data collected, but not
analysed, in an earlier study of neural adaptations to
EMS resistance training (Maffiuletti et al. 2002).
Methods
Subjects
After receiving the approval from the local committee on human
research, 14 male subjects signed informed consent prior to
involvement in investigation. Eight healthy individuals were as-
signed to the EMS training group [age 20.4 (2.1) years; height 186.4
(8.0) cm; weight 83.5 (9.6) kg], with the remaining six subjects as-
signed to the control group [age 26.0 (5.1) years; height 176.7
(4.8) cm; weight 69.5 (4.6) kg].
EMS training program
Sixteen sessions of isometric bilateral EMS were completed with
four sessions per week. Subjects were seated in a calf machine and
the angle at the hip, knee and ankle joint was maintained at 90.
Each session was preceded by a standardized warm-up, consisting
of 5 min of sub-maximal EMS (5 Hz; pulses lasting 200 ls). Then,
the intensity of rectangular-wave pulsed currents (75 Hz) lasting
400 ls was individually increased (range 0–100 mA) to the maxi-
mum that could be comfortably tolerated. During the stimulation,
subjects were asked to relax and the level of electrically elicited
isometric force varied between 50 and 70% of the pretraining
maximal voluntary contraction. Forty-five contractions were car-
ried out during each session with an on-off ratio of 4 s to 20 s. EMS
was delivered by a portable stimulator (Compex Sport, Medi-
compex, Ecublens, Switzerland) with self-adhesive electrodes
placed over the triceps surae muscles. The two positive electrodes,
each measuring 25 cm2 (5 cm·5 cm), were placed over the super-
ficial aspect of the soleus muscle, about 5 cm distal from where the
two heads of the gastrocnemius join the Achilles tendon. The
negative electrode, measuring 50 cm2 (10 cm·5 cm), was placed
along the mid-dorsal line of the leg, over both gastrocnemii. The
maximally tolerated intensity varied between 30 and 90 mA
depending on differences among subjects in pain threshold. No
subjects reported serious discomfort from this current.
Study design
All subjects were tested at baseline. Experimental subjects were
tested again after 4 weeks of EMS training and the control group
was re-tested after 4 weeks of normal activity to assess the reli-
ability of the observations. The dependent variables were the peak-
to-peak electromyographic (EMG) amplitude of maximal M wave
(Mmax), H-reflex (Hmax) and T-reflex (Tmax) potentials for soleus
and gastrocnemii expressed both in millivolts and relative to Mmax,
together with the respective peak twitch tension, i.e., the maximal
amplitude of the plantar flexor twitch torque (in Newton-meters).
Knee angle was consistently maintained at 90 of flexion during
both training and testing. Considered that in this position the (not
negligible) contribution of the gastrocnemii to the plantar flexor
torque is reduced (Fugl-Meyer et al. 1979), and the large majority
of the evoked torque originates from the soleus muscle (see Shields
et al. 1997), the peak twitch torque associated with soleus Mmax,
Hmax and Tmax potentials was taken into consideration during the
entire experiment. The concomitant amplitude of EMG potentials
from medial and lateral gastrocnemius muscles was then summed
and considered as gastrocnemii Mmax, Hmax and Tmax responses
throughout the analysis. These two latter reflex responses were also
normalized with respect to gastrocnemii Mmax.
Torque and EMG recordings
During the testing session, subjects were comfortably seated with
both knee and ankle joint angles of 90. The right foot was secured
to a pedal equipped with strain gauges (linear range 0–250 Nm;
sensitivity 40 mV/Nm), which was developed by the local engi-
neering school, in order to record the plantar flexor mechanical
response.
Pairs of silver-chloride surface electrodes (recording diameter
10 mm) were pasted over the soleus, medial and lateral gastroc-
nemius with an interelectrode (centre-to-centre) distance of 2 cm,
to record EMG activity. The low resistance (<2 kW) between the
two electrodes was obtained by abrading the skin with emery paper
and cleaning with alcohol. For the soleus, recording electrodes were
placed along the mid-dorsal line of the leg, about 5 cm distal from
where the two heads of the gastrocnemius join the Achilles tendon.
Medial and lateral gastrocnemius electrodes were fixed lengthwise
over the middle of the muscle belly (De Luca 1997). Signals were
amplified with a bandwidth frequency ranging from 1.5 Hz to
2 kHz (common mode rejection ratio=90 dB; Z input=100 MW;
gain=1000). EMG and mechanical traces were simultaneously
digitized on-line (sampling frequency 2 kHz), averaged and stored
on a PC for further analysis with commercially available software
(Tida, Heka Elektronik, Lambrecht/Pfalz, Germany).
 603

M-wave, H- and T-reflex recordings
The posterior tibial nerve was stimulated at different intensities
using a cathode ball electrode pressed in the popliteal fossa to elicit
H reflexes and M waves. The percutaneous electrical stimulus was a
rectangular pulse (1 ms duration) delivered by a Digitimer stimu-
lator (DS7, Hertfordshire, UK). The anode was a large electrode
(50 cm2), placed on the anterior surface of the knee. The stimu-
lating cathode was positioned in such a way that the M response
was minimal at stimulus intensities around and below soleus Hmax,
more likely for the gastrocnemii muscles. For each subject, the
amplitude of the small M wave preceding the Hmax response was
comparable for soleus and gastrocnemii (10% of the respective
Mmax) both pre- and post-test (see Maffiuletti et al. 2001). Each
subject was initially familiarized with several submaximal electrical
stimuli over a period of 10–15 min.
Testing procedure and recordings started by progressively
increasing (1 mA) the current intensity from 0 until there was no
further increase in peak-to-peak soleus M response. Four stimuli
were delivered at each intensity, with a 3-s interval between stimuli
(i.e., 0.33 Hz). The average EMG and mechanical signals obtained
at each intensity were stored to obtain later H-M recruitment
curves for soleus and gastrocnemii muscles. Once soleus Mmax
amplitude remained stable during three to five consecutive step-
wise intensity increases, Mmax for respective muscles as well as
related peak twitch torque were computed from the traces associ-
ated with the highest intensity [mean (SD), 45.4 (12.2) mA] series
(see Mmax in Fig. 1). The stimulus intensity appropriate for
obtaining the maximal soleus H reflex was then carefully searched
for and, once it was found, the associated Hmax response for the
respective muscles and Hmax peak twitch were determined (see
Hmax in Fig. 1).
Subsequently, several reflex responses were elicited by tapping
the Achilles tendon at a frequency of 1 Hz with a Babinski
hammer. The intensity of the mechanical stimuli and the point of
application of the tap over the tendon were modulated by the
experimenter that handled the hammer, in order to obtain Tmax
reflexes. For each subject, EMG and mechanical traces associated
with 40 T-reflexes were stored for off-line analysis. The four
bigger soleus T-reflex potentials were then selected and averaged, in
order to calculate Tmax for the respective muscles and corre-
sponding peak twitch (see Tmax in Fig. 1).
The stimulus repetition intervals adopted here are rather short
for consecutive reflex measurements (i.e., 3 s for the H-reflex and
1 s for the T-reflex). However, according to Pierrot-Deseilligny and
Mazevet (2000), H-reflex explorations at 0.2–0.3 Hz are allowed,
considering that one needs to find a compromise between post-
activation depression and the necessity of collecting a large number
of reflexes because of variability. It must also be considered that,
according to the findings of Van Boxtel (1986), similar post-acti-
vation depressions comprised between 75 and 80% are obtained for
H-reflexes evoked at intervals of 3 s and for T-reflexes separated by
1 s (see Fig. 2 in Van Boxtel 1986). The electromechanical prop-
erties of soleus H and T responses were thus compared in this
study, all other things being equal.
Statistical analyses
Differences between the values before and after the 4-week period
were tested for significance by Student’s paired t-tests, for both
experimental and control groups. Independent two-tailed t-tests
were used to analyse differences between means of variables for the
experimental and control groups. In each case the level of signifi-
cance was established at P £ 0.05.
Results
At baseline, no significant difference was observed be-
tween experimental and control group for the ensemble
of the variables considered. EMS training did not sig-
nificantly affect the crude amplitude of the Mmax, Hmax
and Tmax potentials for soleus and gastrocnemii muscles
(Table 1). Peak twitch values associated with Mmax were

Fig. 1 Example of EMG traces

(left) from soleus (thick line)
and medial gastrocnemius
muscle (thin line) and related
plantar flexor twitch torque
(right) associated with maximal
M wave (Mmax), maximal H-
reflex (Hmax), maximal T-reflex
(Tmax) and H-reflex matching
baseline soleus Tmax (H=Tmax)
responses (average of four
acquisitions for one
representative subject)
604

Fig. 3A–D Peak-to-peak amplitude of soleus and gastrocnemii

Fig. 2A–D Peak-to-peak amplitude of soleus and gastrocnemii
maximal H reflex (Hmax) normalized to the respective maximal M
wave (A, C) and concomitant plantar flexor peak twitch torque (B,
D) before and after the 4-week period for experimental (A, B) and
control groups (C, D). Values are means (SE)
maximal T-reflex (Tmax) normalized to the respective maximal M
wave (A, C) and concomitant plantar flexor peak twitch torque (B,
D) before and after the 4-week period for experimental (A, B) and
control groups (C, D). Values are means (SE). *Values after
training are significantly higher than pre-training at P<0.05

not significantly different before and after the 4-week
period for the experimental subjects [21.0 (4.6) Nm to
19.9 (4.5) Nm] and for the control group [17.5 (3.6) Nm
to 16.3 (5.5) Nm].
Normalized soleus and gastrocnemii Hmax amplitude
increased insignificantly (P=0.22 and 0.11, respectively)
after training (Fig. 2A), while the associated twitch
torque output decreased 10% (Fig. 2B), though not
significantly so. When considering normalized Tmax
potentials, EMS did not affect soleus amplitude
(Fig. 3A) while a trend towards a significant increase
was observed for gastrocnemii (+21%, P=0.07). The
peak twitch produced by such a reflex response was 21%
higher (P<0.05) following EMS training (Fig. 3B).
In order to emphasize electromechanical Tmax find-
ings, for each subject the peak twitch associated with the
soleus H-reflex matching baseline soleus Tmax amplitude
(see H=Tmax in Fig. 1) was compared before and after
the experimental period. These values were obtained
from the individual soleus H-M recruitment curves (and
concomitant peak twitch values) and, consequently, a
submaximal H-reflex was adapted for each subject to his

Table 1 Peak-to-peak
amplitude of soleus and Muscle Experimental group (n=8) Control group (n=6)
gastrocnemii maximal M wave
(Mmax), H-reflex (Hmax) and Pre Post Pre Post
T-reflex (Tmax) before and after
the 4-week period for Mmax Soleus (mV) 9.5 (2.8) 9.0 (1.9) 10.9 (2.8) 10.3 (3.0)
experimental and control Gastrocnemii (mV) 12.8 (4.9) 12.3 (3.5) 12.9 (4.5) 13.3 (5.4)
groups. Values are means (SD) Hmax Soleus (mV) 5.1 (2.0) 5.2 (0.8) 5.8 (2.1) 5.9 (1.7)
Gastrocnemii (mV) 2.8 (1.4) 3.0 (1.0) 3.3 (1.0) 4.5 (2.2)
Tmax Soleus (mV) 2.1 (1.1) 2.0 (1.0) 3.6 (2.2) 2.7 (1.1)
Gastrocnemii (mV) 1.2 (0.8) 1.2 (0.6) 2.1 (1.2) 2.2 (1.4)

Fig. 4A–D Peak-to-peak amplitude of soleus and gastrocnemii H-
reflex matching baseline soleus Tmax (H=Tmax) normalized to the
respective maximal M wave (A, C) and concomitant plantar flexor
peak twitch torque (B, D) before and after the 4-week period for
experimental (A, B) and control groups (C, D). Values are means
(SE)
personal Mmax response. For the ensemble of the
experimental subjects, soleus H=Tmax amplitudes ex-
pressed relative to Mmax averaged 21.7 (10.8)%. Sur-
prisingly, peak twitch (Fig. 4B) and also H-reflex
amplitude for gastrocnemii (Fig. 4A) were unchanged
after training, therefore indicating that the mechanical
output associated with equal-amplitude H- and T-re-
flexes did not adapt similarly to EMS.
Normalized Hmax, Tmax and H=Tmax reflexes for the
respective muscles and the associated peak twitch values
were not significantly different before and after the 4-
week period for the control group (Figs. 2, 3, 4).
Discussion
The present involuntary training program had no sig-
nificant effect on the amplitude of the maximal M wave
and on the associated twitch contractile properties (see
also Maffiuletti et al. 2002), indicating that EMS did not
modify the excitation-contraction coupling properties of
605
the triceps surae muscle. On the other hand, both EMG
activity and activation level significantly increased by
12% after training, and thus we argued that this arti-
ficial training modality likely enhanced volitional drive
from the supraspinal centers (Maffiuletti et al. 2002).
However, such activation increases could also be as-
cribed to changes occurring in spinal neural circuitry,
considered that action potentials are evoked in both
intramuscular nerve branches and cutaneous receptors
during EMS, but this hypothesis was not verified in our
previous work.
The amplitudes of maximal soleus and gastrocnemii
H potentials expressed relative to respective Mmax (i.e.,
Hmax/Mmax ratio) did not change significantly following
the current EMS intervention. Training-induced modi-
fications in H-reflex amplitude are usually ascribed to
changes in alpha motoneuron excitability and/or chan-
ges in presynaptic inhibition of Ia afferents (for review
see Zehr 2002), but none of these adaptations is likely to
have occurred here. The present results are in general
agreement with previous studies that have examined the
influence of voluntary training protocols on H-reflex
amplitude obtained at rest (Aagaard et al. 2002; Voigt
et al. 1998). It must nevertheless be considered that, al-
though not investigated with the present methodology,
H-reflexes evoked during actual voluntary contractions
(Aagaard et al. 2002) and/or during the performance of
the training exercise (Voigt et al. 1998; see also Zehr
2002) likely represent a more functional estimate of the
training-induced adaptations at the spinal cord level.
For example, Aagaard et al. (2002) found no changes in
resting soleus H-reflex amplitude following 14 weeks of
heavy weight-lifting exercises, while the maximal H-re-
flex and V-wave recorded during maximal isometric
ramp contractions significantly increased with training,
by 20% and 50% respectively. Similarly, 4 weeks of
progressive two-legged hopping training resulted in lar-
ger H-reflexes during hopping but not during standing
rest (Voigt et al. 1998). Future EMS training studies
should examine the properties of spinal reflexes elicited
during voluntary isometric contractions and/or electri-
cally evoked contractions, in order to better ascertain
the specificity of neural adaptations.
To date, a small number of longitudinal studies have
examined the effects of chronic physical activity on the
electromechanical properties of spinally mediated re-
flexes in humans. For the first time, triceps surae H- and
T-reflex recordings were used to examine the neural
adaptation mechanisms induced by intensive EMS
resistance training. As previously detailed (see Study
design), we were mainly concerned with soleus reflex
modifications in the present study. However, maximal
soleus H- and T-reflexes were both unchanged following
intensive EMS training. These results are not unexpected
if one considers that the orderly recruitment of motor
units during reflex responses is different than that during
EMS. Indeed, electrically and mechanically evoked
reflexes both recruit the motoneurons in an orderly way
by the Ia input from the smallest to the largest ones,
606
according to the size principle (Hennemann et al. 1965).
As a consequence, the lower threshold and so-called
smaller, slow-twitch motor units predominate in the
human soleus H-reflex (Buchthal and Schmalbruch
1970). On the other hand, high-frequency electrical
stimulation applied over triceps surae muscles prefer-
entially activates motoneurons with large-diameter ax-
ons, although additional recruitment of the smaller
motoneurons that innervate more fatigue-resistant
muscle fibers would also take place (Collins et al. 2001).
The lack of changes in spinal reflexes following EMS is
also in accordance with our previous speculations and
confirms, at least in part, that the activation increases
observed after the current training protocol (Maffiuletti
et al. 2002) would be ascribed to adaptations involving
supraspinal centers (e.g., motor command and/or
descending drive). In this context, it is important to note
that a recent functional neuroimaging study has dem-
onstrated significant cerebral cortex activation during
electrically evoked contractions of human wrist extensor
muscles (Han et al. 2003).
In the present investigation, recordings of the plantar
flexor mechanical output (i.e., twitch torque) produced
by the motor units reflexively activated represent an
original feature, since no previous H- or T-reflex study
has focused on the plasticity of this parameter. Peak
twitch associated with maximal T reflex significantly
increased after EMS training, while equal-amplitude H-
reflex (i.e., H=Tmax) and also the maximal H-reflex
mechanical output did not change. These modifications
were associated with a trend towards increased gas-
trocnemii but not soleus T-reflex amplitude. In contrast,
Voigt et al. (1998) have shown that 4 weeks of hopping
training significantly increased the T-reflex amplitude
for soleus but not for medial gastrocnemius. However,
whether the mechanical output from the motor units
activated by tendon percussion also increased was not
ascertained by these authors. Our results could be ex-
plained by changes in the mechanical properties of the
elastic structures in series with the muscle spindles (e.g.,
increased stiffness of tendinous elements). Voluntary
isometric resistance training at 70% MVC recently re-
sulted in increased stiffness of the human tendon struc-
tures (Kubo et al. 2001). The intensity of our electrically
evoked isometric contractions was comprised between
50% and 70% MVC, and therefore the load applied to
tendon structures was very similar to that incurred in
this previous study. It is therefore probable that, al-
though Achilles tendon elasticity and rate of torque
development were not assessed with the present meth-
odology, both voluntary and EMS isometric
training enhance tendon stiffness. On the other hand,
changes in musculoarticular stiffness can also be related
to the T-reflex electromechanical results observed here.
In a recent microgravity study, normalized triceps
surae T-reflex and passive musculotendinous stiffness
increased by 57% and 25% respectively, 2–3 days after
a spaceflight of 3–6 months, while musculoarticular
stiffness determined by using sinusoidal perturbations
significantly decreased by 11% (Lambertz et al. 2003).
Interestingly, the authors found a significant correlation
between changes in normalized triceps surae T-reflex
and in passive musculoarticular stiffness, therefore
indicating that musculoarticular structures took part in
the reflex adaptation.
A simple hand-held reflex hammer was used in the
current investigation to elicit T-reflexes and, due to
methodological difficulties, the stimulus intensity was
not measured. The effect of changes in tap intensity
can nevertheless be excluded because the amplitude of
T-reflex potentials and also of the concomitant twitch
saturated at medium tap strength. As a matter of fact,
the results obtained in the control group clearly confirm
the validity of the present methodology.
It is not surprising to observe changes in the ampli-
tude of gastrocnemii but not of soleus T-reflex potential
after EMS, since this latter muscle contains more fati-
gue-resistant fibers compared to the former (Johnson et
al. 1973). Again, according to the preferential activation
of the largest motor units during EMS (for review see
Enoka 1988), the present training protocol would have
resulted in a selective adaptation of the gastrocnemii vs.
soleus muscle, also in line with voluntary EMG results
(see Maffiuletti et al. 2002). As a methodological con-
sideration, due to well-known differences between soleus
and gastrocnemii (see Johnson et al. 1973; Kennedy and
Cresswell 2001), training-induced adaptations can occur
selectively within the triceps surae muscle group. Gas-
trocnemii muscles are often neglected, especially when
reflex responses are investigated, perhaps missing
important information about the plasticity of the human
motor system. It is then tempting to suggest that future
reflex studies should consistently record and process
EMG activity from soleus and both gastrocnemii mus-
cles concomitantly, as realized by Trimble and Koceja
(1994) and more recently by Lambertz et al. (2003).
The different adaptability of maximal H- vs. T-reflex
properties observed after the current training program is
not a novel finding (Pérot et al. 1991; Voigt et al. 1998).
The H-wave is the reflex discharge of the alpha moto-
neuron pool in response to the orthodromic afferent
volley traveling in the large-diameter Ia fibers originat-
ing in the muscle spindles. The same pathway as for the
H-reflex also seems responsible for the mechanically
elicited reflex (Magladery et al. 1952; see also Pierrot-
Deseilligny and Mazevet 2000), even though the former
reflex is not influenced by the sensitivity of the muscle
spindles that are bypassed by the electrical stimulation.
Therefore, the T-reflex amplitude also depends on the
gamma fusimotor drive controlling the sensitivity to
stretch of muscle spindle primary endings (Pierrot-Des-
eilligny and Mazevet 2000). However, growing evidence
indicates that the two responses differ in several other
respects, e.g., the temporal dispersion and composition
of afferent volleys in the tibial nerve (see Burke et al.
1983, 1984; Gassel and Diamantopoulos 1966; Van
Boxtel 1986), which makes comparisons of the H- and
T-reflexes unreliable as measures of gamma activity. As

a result, it remains unclear whether equal-amplitude H-
and T-reflexes activate similar (Van Boxtel 1986) or
different motor units (Homma et al. 1981; Voigt et al.
1998). Whatever the differences between these two re-
flexes, the present study furnished evidence that the
electromechanical properties of equal-amplitude H- and
T-reflexes did not adapt in the same way following
multiple sessions of contractions triggered by EMS. The
lack of an H-reflex increase suggests that the effects were
peripheral rather than central in origin.
In conclusion, EMS training applied over the triceps
surae muscle, likely recruiting spinal motoneurons
reflexively, had no apparent effect on alpha motoneuron
excitability and/or presynaptic inhibition but signifi-
cantly increased the plantar flexor mechanical output
produced by a maximal T-reflex.
Acknowledgements The authors gratefully acknowledge Dr. L.
Gerilovsky and Dr. A. Van Boxtel for useful comments on a pre-
vious version of the manuscript.
References
Aagaard P, Simonsen EB, Andersen JL, Magnusson P, Dyhre-
Poulsen P (2002) Neural adaptation to resistance training:
changes in evoked V-wave and H-reflex responses. J Appl
Physiol 92:2309–2318
Buchthal F, Schmalbruch H (1970) Contraction time of twitches
evoked by H-reflexes. Acta Physiol Scand 80:378–382
Burke D, Gandevia SC, McKeon B (1983) The afferent volleys
responsible for spinal proprioceptive reflexes in man. J Physiol
(Lond) 339:535–552
Burke D, Gandevia SC, McKeon B (1984) Monosynaptic and
oligosynaptic contributions to human ankle jerk and H-reflex. J
Neurophysiol 52:435–448
Collins DF, Burke D, Gandevia SC (2001) Large involuntary
forces consistent with plateau-like behavior of human moto-
neurons. J Neurosci 21:4059–4065
Colson S, Martin A, Van Hoecke J (2000) Re-examination of
training effects by electrostimulation in the human elbow
musculoskeletal system. Int J Sports Med 21:281–288
De Luca CJ (1997) The use of electromyography in biomechanics. J
Appl Biomech 13:135–163
Enoka RM (1988) Muscle strength and its development. New
perspectives. Sports Med 6:146–168
Eriksson E, Haggmark T, Kiessling KH, Karlsson J (1981) Effect
of electrical stimulation on human skeletal muscle. Int J Sports
Med 2:18–22
Fugl-Meyer AR, Sjöström M, Wählby L (1979) Human plantar
flexion strength and structure. Acta Physiol Scand 107:47–56
Gassel MM, Diamantopoulos E (1966) Mechanically and electri-
cally elicited monosynaptic reflexes in man. J Appl Physiol
2:1053–1058
Halbach JM, Straus D (1980) Comparison of electromyostimula-
tion to isokinetic training in increasing power of the knee
extensor mechanism. J Orthop Sports Phys Ther 2:20–24
607
Han BS, Jang SH, Chang Y, Byun WM, Lim SK, Kang DS (2003)
Functional magnetic resonance image finding of cortical acti-
vation by neuromuscular electrical stimulation on wrist exten-
sor muscles. Am J Phys Med Rehabil 82:17–20
Hennemann E, Somjen G, Carpenter DO (1965) Excitability and
inhibitability of motoneurons of different sizes. J Neurophysiol
28:599–620
Homma S, Nakajima Y, Hayashi K, Shito K, Sato K (1981) Area
display of H, M and T waves. In: Taylor A, Prochazka A (eds)
Muscle receptors and movement. MacMillian, London, pp 381–
387
Hortobagyi T, Scott K, Lambert J, Hamilton G, Tracy J (1999)
Cross-education of muscle strength is greater with stimulated
than voluntary contractions. Motor Control 3:205–219
Johnson MA, Polgar J, Weightman D, Appleton D (1973) Data on
the distribution of fiber types in thirty six human muscles: an
autopsy study. J Neurol Sci 18:111–129
Kennedy PM, Cresswell AG (2001) The effect of muscle length on
motor-unit recruitment during isometric plantar flexion in hu-
mans. Exp Brain Res 137:58–64
Kubo K, Kanehisa H, Ito M, Fukunaga T (2001) Effects of iso-
metric training on the elasticity of human tendon structures in
vivo. J Appl Physiol 91:26–32
Lambertz D, Goubel F, Kaspranski R, Pérot C (2003) Influence of
long-term spaceflight on neuromechanical properties of muscles
in humans. J Appl Physiol 94:490–498
Maffiuletti NA, Martin A, Babault N, Pensini M, Lucas B,
Schieppati M (2001) Electrical and mechanical Hmax-to-Mmax
ratio in power- and endurance-trained athletes. J Appl Physiol
90:3–9
Maffiuletti NA, Pensini M, Martin A (2002) Activation of human
plantar flexor muscles increases after electromyostimulation
training. J Appl Physiol 92:1383–1392
Magladery JW, Park AM, Porte WE, Teasdall RD (1952) Spinal
reflex patterns in man. Res Publ Assoc Res Nerv Ment Dis
30:118–151
Morrissey MC (1988) Electromyostimulation from a clinical per-
spective. A review. Sports Med 6:29–41
Pérot C, Goubel F, Mora I (1991) Quantification of T- and H-
responses before and after a period of endurance training. Eur J
Appl Physiol 63:368–375
Pierrot-Deseilligny E, Mazevet D (2000) The monosynaptic reflex:
a tool to investigate motor control in humans. Interests and
limits. Neurophysiol Clin 30:67–80
Shields RK, Frey L, Reiling B, Sass K, Wilmert J (1997) Effects of
electrically induced fatigue on the twitch and tetanus of para-
lyzed soleus muscle in humans. J Appl Physiol 82:1499–1507
Trimble MH, Koceja DM (1994) Modulation of the triceps surae
H-reflex with training. Int J Neurosci 76:293–303
Van Boxtel A (1986) Differential effects of low-frequency depres-
sion, vibration-induced inhibition, and post-tetanic potentia-
tion on H-reflexes and tendon jerks in the human soleus muscle.
J Neurophysiol 55:551–568
Voigt M, Chelli F, Frigo D (1998) Changes in the excitability of
soleus muscle short latency stretch reflexes during human
hopping after 4 weeks of hopping training. Eur J Appl Physiol
78:522–532
Ward AR, Shkuratova N (2002) Russian electrical stimulation: the
early experiments. Phys Ther 82:1019–1030
Zehr PE (2002) Considerations for use of the Hoffmann reflex in
exercise studies. Eur J Appl Physiol 86:455–468