Neuroscience Letters 424 (2007) 4146

Effect of electrostimulation trainingdetraining on

neuromuscular fatigue mechanisms
Marc Jubeau a, , Raphael Zory b , Julien Gondin c , Alain Martin a , Nicola A. Maffiuletti d
a

Laboratory INSERM U887, Faculty of Sport Sciences, University of Burgundy, BP 27877, 21078 Dijon, France
b Modelling in Sports Activities Laboratory, University of Savoie, UFR STAPS, Chambery, France
c Department of Sport Physiology, Faculty of Sport Sciences, IFR Marey, University of the Mediterranean, France
d Neuromuscular Research Laboratory, Schulthess Clinic, Zurich, Switzerland
Received 2 May 2007; received in revised form 30 June 2007; accepted 10 July 2007

Abstract
The aim of this study was to evaluate the effects of neuromuscular electrical stimulation (NMES) training and subsequent detraining on
neuromuscular fatigue mechanisms. Ten young healthy men completed one NMES fatigue protocol before and after a NMES training program of
4 weeks and again after 4 weeks of detraining. Muscle fatigue (maximal voluntary torque loss), central fatigue (activation failure), and peripheral
fatigue (transmission failure and contractile failure) of the plantar flexor muscles were assessed by using a series of electrically evoked and
voluntary contractions with concomitant electromyographic and torque recordings. At baseline, maximal voluntary torque decreased significantly
with fatigue (P < 0.001), due to both activation and transmission failure. After detraining, maximal voluntary torque loss was significantly reduced
(P < 0.05). In the same way, the relative decrease in muscle activation after training and detraining was significantly lower compared to baseline
values (P < 0.05). Short-term NMES trainingdetraining of the plantar flexor muscles significantly reduced the muscle fatigue associated to one
single NMES exercise session. This was mainly attributable to a reduction in activation failure, i.e., lower central fatigue, probably as a result of
subjects accommodation to pain and discomfort during NMES.
2007 Elsevier Ireland Ltd. All rights reserved.
Keywords: Central fatigue; Discomfort; Electrical stimulation; Transmission failure

Surface neuromuscular electrical stimulation (NMES) involves

artificially activating skeletal muscles with a protocol designed
to minimize the pain and discomfort associated with the stimulation [12]. In the last few years, neuromuscular adaptations
induced by NMES training have received particular attention,
and special emphasis has been devoted to chronic changes in
the function of the central nervous system [9,11,14]. For example, we have recently demonstrated that artificial resistance
training of the plantar flexor muscles resulted in early neural
adaptations (increased muscle activation and EMG activity),
likely mediating the improvement in maximal voluntary contraction (MVC) strength [11]. Moreover, late neural adaptations
(increased spinal reflex amplitude and decreased coactivation)
were observed following detraining, and probably contributed
to strength gains preservation [11]. On the other hand, only few
studies have adequately investigated the acute changes in neu-

Corresponding author. Tel.: +33 380396762; fax: +33 380396702.

E-mail address: Marc.Jubeau@u-bourgogne.fr (M. Jubeau).

0304-3940/$ see front matter 2007 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.neulet.2007.07.018

romuscular properties and fatigue after NMES exercise. This

information is valuable because the rapid onset of fatigue classically associated to NMES contractions represents one of the
main limitations of this technique [13].
Fatigue is an exercise-induced reduction in maximal voluntary muscle force [8]. Factors mediating neuromuscular fatigue
are commonly classified into two components with respect to
their locality: central and peripheral. Central fatigue refers to a
progressive reduction in voluntary activation of muscle during
exercise whereas peripheral fatigue is characterized by changes
at or distal to the neuromuscular junction [8]. Recently, we
have studied triceps surae muscle fatigue mechanisms after a
single bout of NMES resistance exercise [3]. In this experiment, the significant reduction of MVC torque was associated
to neuromuscular transmission failure, as witnessed by M-wave
amplitude reductions, and to a significant decrease in muscle
activation, i.e., central fatigue. To our knowledge, it is unclear
whether these central and peripheral components of NMES
exercise-induced fatigue can be modified after multiple sessions of training and subsequent detraining. Since the occurrence

42

M. Jubeau et al. / Neuroscience Letters 424 (2007) 4146

of central fatigue in unaccustomed subjects is probably due to

unusual pain and discomfort associated with NMES, we hypothesized reduced activation failure after the trainingdetraining
period, mainly as a result of the increased tolerance to NMES
with practice [2].
The purpose of the present study was to investigate the effects
of NMES training and subsequent detraining on neuromuscular
fatigue mechanisms, with particular emphasis on central fatigue.
Seventeen healthy male subjects were arbitrarily assigned to
the NMES training group (TG) (n = 10; age, 24 1 years; height
178 2 cm; weight, 71 2 kg) and to the (non-training) control group (CG) (n = 7; age, 24 1 years; height 178 2 cm;
weight, 75 3 kg). Subjects were normally active and volunteered to participate in the study. None of them had experienced
NMES in the past. The local ethics committee approved the
experimental procedures, and the experiments were performed
in accordance with the Declaration of Helsinki (last modified in
2000). Before the participation to the study, all the subjects gave
written informed consent.
The TG was tested before (pre) and after (post) a 4-week
NMES training program (frequency: 4 sessions/weeks; duration: 18 min/session; volume: 40 contractions/session) and then
again, after 4 weeks of detraining. During each NMES session,
current amplitude (range of 21120 mA; on-line control) was
gradually increased during the first 45 NMES contractions until
the maximum tolerated level (pain threshold) and was then kept
constant until the last contraction. Tolerance to NMES significantly increased throughout the training program, since maximal
tolerated stimulation intensity linearly increased from the first
(31 13 mA) to the last (75 7 mA) NMES training session.
The level of electrically evoked torque was measured during
a representative training session and ranged between 42 and
104% of the pre-training MVC (mean S.D.: 82 19%) [11].
The CG was tested (pre) and retested (post) after 4 weeks. Due
to scheduling difficulties, we did not include a third testing session for the CG (8 weeks after baseline). Subjects were asked
not to perform any strenuous exercise for at least 48 h prior to
all testing sessions.
The testing session consisted of a series of supramaximal electrically evoked contractions (stimulation of the tibial
nerve) and MVCs of the plantar flexor muscles, with concomitant electromyographic (EMG) and torque recordings. The
same measurements were completed before (unfatigued conditions) and immediately after (fatigued conditions) a standardized
fatiguing task to assess muscle fatigue (MVC torque loss),
central fatigue (activation failure), and peripheral fatigue (transmission failure and contractile failure). The fatiguing task
consisted of one single session of NMES, i.e., 40 NMES contractions identical to those realized during training. In the present
study, activation failure, transmission failure, and contractile
failure were respectively associated to impairments in muscle
activation, M-wave amplitude, and muscle contractile properties.
Subjects were tested in the supine position. The hip, knee,
and ankle joints were positioned at 90 . All the contractions
were performed under isometric conditions by using an isokinetic machine (Biodex, Shirley, USA). The foot was fixed into

a shoe that was mounted on the dynamometer footplate. The

alignment between the center of rotation of the dynamometer shaft and the ankle joint rotation axis was consistently
checked. A portable stimulator (Compex Sport-P, Medicompex SA, Ecublens, Switzerland) was used to deliver biphasic
symmetric rectangular pulses with the following characteristics: frequency 75 Hz, pulse duration 400 s, duty cycle 24%
(6.25 s on, 20 s off), rise time 1.5 s and fall time 0.75 s. We used
standard NMES parameters [9,11,15] with long pulse durations
to produce powerful and painless contractions and high frequency modulation to ensure tetanic contractions [12]. Three
self-adhesive electrodes (thickness 2 mm) were placed over the
triceps surae muscle. The two positive electrodes (5 cm 5 cm)
were positioned over the superficial aspect of the soleus muscle,
5 cm distal from where the two heads of the gastrocnemius join
the Achilles tendon. The negative electrode (10 cm 5 cm) was
placed along the middorsal line of the leg, over both medial and
lateral gastrocnemius. At pre, the fatiguing task was realized at
the maximal tolerated stimulation intensity, i.e., pain threshold
(26 3 mA). The NMES fatiguing task at post and detraining
was performed at the same relative torque level as baseline, since
training-induced increases in MVC were anticipated. The relative torque recorded at the beginning of each fatigue task was
49 4, 48 4, and 49 3% of MVC at pre, post and detraining
for TG, and 41 3 and 41 3% of MVC at pre and post for
CG. No significant differences were observed between the two
groups. At post and detraining, the stimulation intensity necessary to evoke the same relative torque was 33 3 and 30 2 mA,
respectively.
Before the beginning of the each testing session, subjects
completed a standardized warm-up consisting of 5 min of submaximal NMES (frequency 5 Hz, pulse duration 350 s) and
submaximal voluntary contractions of the plantar flexor muscles. The testing session included: (i) supramaximal electrical
nerve stimulation (single stimuli) at rest (three trials each), to
record plantar flexors M-waves and twitch peak torque and (ii)
MVC of the plantar flexor muscles with paired stimuli delivered over the isometric plateau (superimposed doublet) and 2 s
after the MVC (two trials), to study muscle activation [17] and
potentiated doublet peak torque.
The posterior tibial nerve was stimulated using a cathode
electrode (silver-chloride electrode, diameter 10 mm) pressed
in the poplitea fossa. The anode was a large rectangular electrode (10 cm 5 cm), placed on the anterior surface of the knee.
The percutaneous electrical stimulus was a rectangular pulse
(1 ms duration) delivered by a Digitimer stimulator (DS7, Herthfordshire, UK). Before the warm-up, the current intensity was
progressively increased until plantar-flexor twitch torque and
M-wave amplitude reached a maximal value. This intensity was
further increased by 10% and then maintained for the entire
testing procedure.
The EMG activity of the soleus and lateral gastrocnemius
muscles was recorded by means of two silver-chloride surface
electrodes (recording diameter: 10 mm; interelectrode distance:
2 cm). Low interelectrode impedance (<2 k) was obtained by
abrading the skin with emery paper and cleaning with alcohol.
For the soleus, recording electrodes were placed along the mid-

M. Jubeau et al. / Neuroscience Letters 424 (2007) 4146

43

Table 1
MVC torque, activation level, Mmax amplitude for soleus and lateral gastrocnemius (LG) muscles and peak torque associated to twitch and doublet in unfatigued and
fatigued conditions at pre, post, and detraining for TG and CG
TG

CG

Pre

Post

Detraining

Pre

Post

MVC torque (Nm)

Unfatigued
Fatigued

90.9 5.7
80.9 5.9

107.8 7.2
93.2 7.4

108.6 4.8
102.9 6.5

100.0 9.3
86.2 8.3

96.9 9.3
84.8 7.7

Activation level (%)

Unfatigued
Fatigued

90.6 3.4
82.4 5.0

98.0 0.4
96.0 1.0

97.8 0.5
95.0 1.7

96.6 1.8
87.0 5.8

97.9 0.9
93.7 3.3

Soleus Mmax (mV)

Unfatigued
Fatigued

8.6 0.8
6.2 0.6

7.7 0.8
5.5 0.6

9.3 0.8
6.4 0.7

12.6 1.2
9.9 1.5

11.1 1.1
7.6 1.3

LG Mmax (mV)

Unfatigued
Fatigued

8.4 1.1
7.3 1.0

6.4 1.0
5.7 0.7

6.0 0.9
5.3 0.6

7.3 0.7
6.6 1.0

6.9 0.9
6.3 1.3

Twitch peak torque (Nm)

Unfatigued
Fatigued

11.5 0.8
10.5 0.6

11.1 0.7
9.8 0.6

12.5 0.8
12.2 0.3

12.4 1.2
10.2 0.9

11.6 1.0
9.5 0.9

Doublet peak torque (Nm)

Unfatigued
Fatigued

28.2 1.8
25.1 1.6

27.4 1.5
23.5 1.2

30.7 1.8
29.5 1.4

28.1 1.9
23.4 1.6

26.9 2.1
23.0 1.9

Mean values S.E. A significant fatigue time interaction was noted for MVC torque and activation level for the TG. A significant fatigue effect was observed for
soleus Mmax and doublet peak torque for both TG and CG, and for twitch peak torque for CG. A significant time effect was observed for MVC torque, activation
level and doublet peak torque for TG, and only for soleus Mmax for CG.

dorsal line of the leg, about 5 cm distal from where the two heads
of the gastrocnemius join the Achilles tendon. Lateral gastrocnemius EMG electrodes were fixed lengthwise over the middle of
the muscle belly. Signals were amplified with a bandwidth frequency ranging from 15 Hz to 2 kHz (common mode rejection
ratio = 90 dB).
EMG and torque traces were digitized on-line (sampling frequency 2024 Hz), and stored for analysis with
commercially available software (Tida, Heka Elektronik, Lambrecht/Pfalz, Germany). Activation level of the plantar flexor
muscles was estimated according to the following formula [1]:
[1 (superimposed doublet/potentiated doublet)] 100. Peakto-peak amplitude of maximal M-wave (Mmax ) was calculated
for both soleus and lateral gastrocnemius muscles. Peak torque
was quantified from twitch and potentiated doublet torque traces.
Whatever the dependent variable, the average of two or three
trials was used for data analysis.
A two-way ANOVA (group fatigue) with repeated measures on the second factor was used to compare baseline (raw)
data between the TG and the CG, both in the fatigued and
unfatigued conditions. Another two-way ANOVA with repeated
measures was used to study the effect of time (pre, post
and detraining data) and fatigue (unfatigued versus fatigued
data) on dependent variables (raw data), for both TG (2 3)
and CG (2 2). If a significant time fatigue interaction was
observed, the fatigue-induced percent change (i.e., unfatigued
versus fatigued data) was calculated and a one-way ANOVA
(repeated measures) was used to study the effect of time. LSD
post hoc analyses were used when necessary. A two-tailed
unpaired Student t-test was used to compare fatigue-induced
percent changes between TG and CG. A probability level of
P 0.05 was accepted as significant.
For the ensemble of the dependent variables considered
(Table 1), no significant differences were observed at baseline
between the two subject groups, except for soleus Mmax .

For TG, a significant main effect of time was observed for

MVC torque, activation level and doublet peak torque during the
trainingdetraining period (P 0.01). For CG, a significant main
effect of time was observed only for soleus Mmax (P = 0.05). A
significant main effect of fatigue was observed for MVC torque,
activation level, soleus Mmax , and doublet peak torque in the
TG (P 0.01). Similarly, a significant main effect of fatigue
was observed for MVC torque, soleus Mmax , and twitch and
doublet peak torque in the CG (P 0.05). For TG, a significant
time fatigue interaction was observed for MVC torque and
activation level (P < 0.01), while no significant interaction was
observed for CG.
At pre, post and detraining, MVC torque of the plantar flexor
muscles decreased significantly (i.e., muscle fatigue) after the
NMES session (P < 0.01). MVC percent loss after detraining
was significantly reduced compared to pre (P = 0.04) and post
(P = 0.005), and it was significantly lower than CG at post
(P = 0.05) (Fig. 1).

Fig. 1. Plantar flexor MVC torque loss (%) at pre, post and detraining for TG
and CG. Mean values S.E. *,**significantly lower than pre- or post-values
at P < 0.05 and P < 0.01, respectively. significantly lower than control group
values of post at P 0.05.

44

M. Jubeau et al. / Neuroscience Letters 424 (2007) 4146

Fig. 2. Plantar flexor muscle activation loss (%) at pre, post and detraining for
TG and CG. Mean values S.E. *significantly lower than pre-values at P < 0.05.

Muscle activation significantly decreased (P < 0.001) with

fatigue only before training (i.e., central fatigue). The relative
decrease in muscle activation was significantly lower at post
(P = 0.02) and detraining (P = 0.04) compared to pre (Fig. 2).
The main objective of the present study was to ascertain
whether multiple sessions of NMES followed by detraining
could reduce muscle fatigue in able-bodied individuals. We
hypothesized a decrease of muscle fatigue due to reduced activation failure after the trainingdetraining period, mainly as a
result of the increased tolerance to NMES with practice [2].
In order to check this hypothesis, we performed a stimulated
fatiguing task before and after NMES training and detraining
to quantify muscle fatigue (MVC torque loss), central fatigue
(activation failure), and peripheral fatigue (transmission and
contractile failure). The main findings of the present investigation were that NMES training and subsequent detraining
significantly reduced plantar flexor muscle fatigue and central
fatigue associated to a series of NMES isometric contractions.
Boerio et al. [3] recently demonstrated significant muscle
activation failure after 30 NMES contractions of the plantar
flexor muscles, and suggested that such central fatigue probably contributed to the NMES exercise-induced MVC torque
loss (9%). In the present investigation, we also observed central fatigue after the first NMES session (at baseline) suggesting
that even if NMES is applied at the muscle level, artificial activation of intact muscles can produce significant central effects
[6]. Moreover, the present M-wave findings confirm that neuromuscular transmission failure is probably one of the main
mechanisms responsible for muscle fatigue after high frequency
electrical stimulation, in line with previous NMES research
[3,19]. Taken as a whole, neuromuscular fatigue results confirm
that the MVC torque loss observed after a typical NMES exercise
session of the plantar flexor muscles was attributable to both central (muscle activation failure) and peripheral (neuromuscular
transmission failure) mechanisms.
For the first time, we have demonstrated that MVC torque
losses induced by NMES exercise were significantly reduced
after NMES trainingdetraining. Moreover, we have examined possible effects of artificial trainingdetraining on muscle
activation and neuromuscular transmission failure associated

to one single NMES exercise session. Nevertheless, it should

be remembered that the fatiguing sessions after training and
detraining were not completed at the maximum tolerated level,
but at the same relative level of MVC torque compared to baseline (50% MVC), i.e., using a stimulation intensity of 30 mA
whatever the testing session.
Interestingly, central adaptations to fatigue probably took
place after the present NMES trainingdetraining program and
contributed, at least in part, to the reduction in MVC torque loss.
Indeed, NMES exercise-induced muscle activation failure was
significantly reduced after the 4-week training program, as witnessed by twitch interpolation results. To our knowledge, this
is the first time that a significant training-induced decrease in
central fatigue is reported using the twitch interpolation technique. Based on surface EMG results, Rube and Secher [16]
suggested an attenuation of central factors in fatigue after 5
weeks of voluntary isometric training of lower extremity muscles. They concluded that adaptations located at the central level,
like a diminution of central inhibition of motoneurons were at
the origin of this fatigue reduction.
Torque evoked by NMES is limited by pain and discomfort
[12,13]. The application of electrical stimuli over the skin could
excite depending on current amplitude, pain receptors, such as
free nerve endings (A-delta and C fibres) located in dermal and
subdermal connective and adipose tissues [2]. The pain message is then transmitted from primary sensory nociceptors to
higher brain centers via the dorsal horn of the spinal cord. For
instance, functional magnetic resonance imaging studies have
demonstrated significant activation of the primary sensory and
motor cortex during NMES of the knee extensor muscles [18].
Interestingly, the activation of these cortical regions was progressively enhanced with increasing stimulation intensity, i.e.,
a doseresponse relationship [18]. Therefore, the occurrence
of central fatigue (at baseline) in the present study could be
explained, at least in part, by unusual pain and discomfort associated to NMES at maximal tolerated level (pain threshold).
However, it has been clearly demonstrated that subjects
can accommodate for the discomfort after repeated exposure
to NMES [2,10], i.e., increased tolerance. Tolerance may be
defined as the maximal level of stimulation acceptable to the
subjects while producing a robust muscle contraction [2]. In the
present study, the maximal tolerated NMES intensity increased
from 31 mA (first training session) to 75 mA (last training session) throughout the program (i.e., a 160% increase); this could
be interpreted as an improvement of tolerance to NMES. We
are therefore justified in assuming that muscle activation failure
observed at baseline was significantly reduced after the present
training program because of the increased tolerance to NMES
with continuous practice, thus resulting in less pain during stimulated contractions. This assumption is in line with previous
studies in which the interaction between pain and fatigue was
examined [4,7]. Experimental muscle pain induced by injection
of hypertonic saline was associated to greater muscle fatigue and
larger decrease in surface EMG activity with respect to a nonpainful condition [4]. According to Farina et al. [7], the greater
central fatigue during painful contractions was most likely due
to spinal or supraspinal factors, such as a decrease of motor unit

M. Jubeau et al. / Neuroscience Letters 424 (2007) 4146

firing rate. In our study, reduced pain due to increased tolerance

of electrical current would have diminished the central effects
associated to NMES, thereby minimizing the extent of central
fatigue.
With the present methodology, it is extremely challenging
to address the neural mechanisms explaining reduced central
fatigue, but some speculations can be proposed. Supraspinal activation of cortical regions could be modulated after the NMES
training program. Indeed, Davis et al. [5] reported that painful
electrical stimuli induced a greater activation of cortical areas
(primary somatosensory and cingulate cortex) than non-painful
stimuli. Even if it is not known whether activation of supraspinal
centers with NMES could contribute to peripheral muscle contraction, we cannot exclude that the decreased supraspinal
activation (due to reduced pain) after trainingdetraining could
be responsible for the decreased central fatigue. It has also been
demonstrated that plantar flexor torque during NMES contractions could be generated by a central mechanism by which
sensory feedback excites motoneurons at the level of the spinal
cord [6]. It could be speculated that the contribution of this spinal
mechanism was reduced after the present training program, due
to decreased sensory feedback or reciprocal inhibition, thus limiting the extent of central fatigue. Further research is necessary
to evaluate the plasticity of central mechanisms to fatigue, e.g.,
by comparing the acute and chronic response to painful versus
non-painful stimulations.
The present treatment had no effect on the NMES exerciseinduced decrement of soleus M-wave amplitude (30%), and
therefore significant neuromuscular transmission failure was
still observed whatever the testing session. Disequilibrium
of ionic gradients, i.e., significant alterations of extracellular
potassium or intracellular sodium concentrations during high
frequency stimulation, as recently discussed [19], could be the
principal mechanisms responsible for the M-wave amplitude
decrement. The present findings demonstrated the lack of adaptability of these mechanisms with NMES training and subsequent
detraining.
Even if a non-significant time fatigue interaction was
observed on plantar flexor muscle contractile properties, we cannot exclude that contractile failure was reduced after detraining,
therefore contributing to diminish the MVC torque loss. Indeed,
both twitch and doublet peak torque decreased after the fatiguing session at pre and post, but not at detraining (see Table 1).
In the same way, doublet peak torque loss was higher than 10%
at baseline and after training while it was reduced to 3% after
detraining. These speculations remain to be studied in the future.
Surprisingly, muscle fatigue was only reduced after 4 weeks
of detraining and not immediately after training. Recently,
it has been suggested that a NMES training program of the
quadriceps femoris muscle could induce overreaching and possibly muscle damage [15]. Similarly to Maffiuletti et al. [15]
and to other NMES studies [9,14], we observed a marked
though non-significant decrease of both twitch and doublet peak
torque after training followed by a recovery to baseline levels
after detraining. These results would confirm the occurrence of
peripheral overreaching immediately after the end of the training program, and the need for a detraining period to make full

45

benefit of the training-induced modifications in neuromuscular

features [11].
In conclusion, short-term NMES trainingdetraining of the
plantar flexor muscles, which is known to induce early and
late neural adaptations [9,11,14], additionally induced a significant reduction of muscle fatigue, mainly attributable to central
mechanisms. This NMES training-induced reduction in central
fatigue was probably the result of subjects accommodation to
pain and discomfort. The present findings offer new perspectives
for realistic applications of NMES in both sports and patient
populations in which fatigue is considered as a limiting factor.
Moreover, the present detraining results should be useful in various clinical situations and in particular in rehabilitation settings
for improving the conception of successful NMES programs for
immobilized subjects.
Acknowledgements
The authors gratefully acknowledge Elisabeth Gavignet,
Yves Ballay, Charles Mc Cammon, Yannick Georges, and all
the volunteers who participated in the study.
References
[1] G.M. Allen, S.C. Gandevia, D.K. McKenzie, Reliability of measurements
of muscle strength and voluntary activation using twitch interpolation,
Muscle Nerve 18 (1995) 593600.
[2] G. Alon, G.V. Smith, Tolerance and conditioning to neuro-muscular electrical stimulation within and between sessions and gender, J. Sports Sci.
Med. 4 (2005) 395405.
[3] D. Boerio, M. Jubeau, R. Zory, N.A. Maffiuletti, Central and peripheral fatigue after electrostimulation-induced resistance exercise, Med. Sci.
Sports Exerc. 37 (2005) 973978.
[4] A. Ciubotariu, L. Arendt-Nielsen, T. Graven-Nielsen, The influence of muscle pain and fatigue on the activity of synergistic muscles of the leg, Eur.
J. Appl. Physiol. 91 (2004) 604614.
[5] K.D. Davis, M.L. Wood, A.P. Crawley, D.J. Mikulis, fMRI of human
somatosensory and cingulate cortex during painful electrical nerve stimulation, Neuroreport 7 (1995) 321325.
[6] J.C. Dean, L.M. Yates, D.F. Collins, Turning on the central contribution
to contractions evoked by neuromuscular electrical stimulation, J. Appl.
Physiol. 103 (2007) 170176.
[7] D. Farina, L. Arendt-Nielsen, T. Graven-Nielsen, Experimental muscle pain
decreases voluntary EMG activity but does not affect the muscle potential
evoked by transcutaneous electrical stimulation, Clin. Neurophysiol. 116
(2005) 15581565.
[8] S.C. Gandevia, Spinal and supraspinal factors in human muscle fatigue,
Physiol. Rev. 81 (2001) 17251789.
[9] J. Gondin, M. Guette, Y. Ballay, A. Martin, Neural and muscular changes to
detraining after electrostimulation training, Eur. J. Appl. Physiol. 97 (2006)
165173.
[10] J.W. Halbach, D. Strauss, Comparison of electro-myo stimulation to isokinetic training in increasing power of knee extensor mechanism, J. Orthop.
Sports Phys. Ther. 2 (1980) 2024.
[11] M. Jubeau, R. Zory, J. Gondin, A. Martin, N.A. Maffiuletti, Late neural
adaptations to electrostimulation resistance training of the plantar flexor
muscles, Eur. J. Appl. Physiol. 98 (2006) 202211.
[12] D.A. Lake, Neuromuscular electrical stimulation. An overview and its
application in the treatment of sports injuries, Sports Med. 13 (1992)
320336.
[13] T. Lloyd, G. De Domenico, G.R. Strauss, K. Singer, A review of the use of
electro-motor stimulation human muscles, Aust. J. Physiother. 32 (1986)
1830.

46

M. Jubeau et al. / Neuroscience Letters 424 (2007) 4146

[14] N.A. Maffiuletti, M. Pensini, A. Martin, Activation of human plantar flexor

muscles increases after electromyostimulation training, J. Appl. Physiol.
92 (2002) 13831392.
[15] N.A. Maffiuletti, R. Zory, D. Miotti, M.A. Pellegrino, M. Jubeau, R. Bottinelli, Neuromuscular adaptations to electrostimulation resistance training,
Am. J. Phys. Med. Rehabil. 85 (2006) 167175.
[16] N. Rube, N.H. Secher, Effect of training on central factors in fatigue following two- and one-leg static exercise in man, Acta Physiol. Scand. 141
(1991) 8795.

[17] A. Shield, S. Zhou, Assessing voluntary muscle activation with the twitch
interpolation technique, Sports Med. 34 (2004) 253267.
[18] G.V. Smith, G. Alon, S.R. Roys, R.P. Gullapalli, Functional MRI determination of a doseresponse relationship to lower extremity neuromuscular
electrical stimulation in healthy subjects, Exp. Brain Res. 150 (2003)
3339.
[19] R. Zory, D. Boerio, M. Jubeau, N.A. Maffiuletti, Central and peripheral
fatigue of the knee extensor muscles induced by electromyostimulation,
Int. J. Sports Med. 26 (2005) 847853.