Journal of Electromyography and Kinesiology 13 (2003) 13–36

www.elsevier.com/locate/jelekin

Interpretation of EMG changes with fatigue: facts, pitfalls, and

fallacies
N.A. Dimitrova ∗, G.V. Dimitrov
Centre of Biomedical Engineering, Bulgarian Academy of Sciences, Acad. G. Bonchev. Str., Bl.105, Sofia 1113, Bulgaria

Abstract
Failure to maintain the required or expected force, defined as muscle fatigue, is accompanied by changes in muscle electrical
activity. Although studied for a long time, reasons for EMG changes in time and frequency domain have not been clear until now.
Many authors considered that theory predicted linear relation between the characteristic frequencies and muscle fibre propagation
velocity (MFPV), irrespective of the fact that spectral characteristics can drop even without any changes in MFPV, or in proportion
exceeding the MFPV changes. The amplitude changes seem to be more complicated and contradictory since data on increased,
almost unchanged, and decreased amplitude characteristics of the EMG, M-wave or motor unit potential (MUP) during fatigue can
be found in literature. Moreover, simultaneous decrease and increase in amplitude of MUP and M-wave, detected with indwelling
and surface electrodes, were referred to as paradoxical. In spite of this, EMG amplitude characteristics are predominantly used
when causes for fatigue are analysed. We aimed to demonstrate theoretical grounds for pitfalls and fallacies in analysis of experi-
mental results if changes in intracellular action potential (IAP), i.e. in peripheral factors of muscle fatigue, were not taken into
consideration. We based on convolution model of potentials produced by a motor unit and detected by a point or rectangular plate
electrode in a homogeneous anisotropic infinite volume conductor. Presentation of MUP in the convolution form gave us a chance
to consider power spectrum (PS) of MUP as a product of two terms. The first one, PS of the input signal, represented PS of the
first temporal derivative of intracellular action potential (IAP). The second term, PS of the impulse response, took into account
MFPV, differences in instants of activation of each fibre, MU anatomy, and MU position in the volume conductor in respect to
the detecting electrode. PS presentation through product means that not only changes in MFPV could be responsible for PS shift
as is usually assumed. Changes in IAP duration and IAP after-potential magnitude, affecting the first term of the product, influence
the product and thus MUP PS. Moreover, the interrelations between the two spectra and thus sensitivity of spectrum to different
parameters change with MU-electrode distance because the second term depends on it. Thus, we have demonstrated that theory
does not predict a linear relation between the characteristic frequencies (maximum, mean and median) and MFPV. IAP duration
and after-potential magnitude are among parameters affecting MUP or M-wave PS and thus, EMG PS detected by monopolar and
bipolar electrodes. Usage of single fibre action potential models instead of MUP ones can result in false dependencies of frequency
characteristics. The MUP amplitude characteristics are determined not only by amplitude of IAP, but also by the length of the IAP
profile and source-electrode distance. Due to the IAP profile lengthening and an increase in the negative after-potential, surface
detected EMG amplitude characteristics can increase even when IAP amplitude decreases considerably during fatigue. Increase in
surface detected MUP or M-wave amplitude should not be attributed to a weaker attenuation of the low-frequency components
with distance. Simultaneous decrease and increase in amplitude of MUP and M-wave detected with indwelling and surface electrodes
are regular, not paradoxical. Corner frequency of the high pass filter should be 0.5 or 1 Hz when muscle fatigue is analyzed. The
area of MUP or M-wave normalized in respect of the amplitude of the terminal phase (that is produced during extinction of the
depolarized zones at the ends of the fibres) could be useful as a fatigue index. Analysing literature data on IAP changes due to
Ca2+ increasing, we hypothesised that the ability of muscle fibres to uptake Ca2+ back into the sarcoplasmic reticulum could be the
limiting site for fatigue. If this hypothesis is valid, IAP changes are not a cause of fatigue; they are due to it.
 2002 Elsevier Science Ltd. All rights reserved.

1. Introduction
Failure to maintain the required or expected force,
defined as muscle fatigue, is accompanied by changes in
∗
Corresponding author. Tel.: +359-2-979-21-62; fax: +359-2-
723-787.
E-mail address: ngdim@iph.bio.bas.bg (N.A. Dimitrova).
muscle electrical activity. Piper [1] was the first who
observed a reduction in frequency of the surface EMG
(Piper rhythm) when a contraction was sustained.
Besides such a frequency shift, Cobb and Forbes [2]
found a consistent increase in the amplitude of surface
recorded EMG. Since then, the studies can be divided
into investigations directed at discovering signs of
fatigue and/or causes for fatigue.

1050-6411/03/$ - see front matter  2002 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S1050-6411(02)00083-4
14 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36

Nomenclature
IAP Intracellular action potential
Tin Duration of IAP
MFPV Muscle fibre propagation velocity
n Muscle fibre propagation velocity
IS Input signal
IR Impulse response
T Transfer function
OS Output signal
SF Single fibre
SFAP Single fibre action potential
MU Motor unit
MUP Motor unit potential
M-wave Myoelectric signal evoked in response to electrical stimulus
EMG Electromyogram
IEMG Integrated Electromyogram
RMS Root mean square
PS Power spectrum
Fmax Frequency at which power spectrum density reaches its maximum value
Fmean Mean frequency of the myoelectric signal power spectrum density function
Fmed Median frequency of the myoelectric signal power spectrum density function
x Longitudinal position of electrode in respect to end-plates
y Fibre-electrode or MU axis-electrode distance
LSD Longitudinal single differentiating (bipolar) electrode or filter
TSD Transversal single differentiating (bipolar) electrode or filter
IED Inter-electrode distance
SR Sarcoplasmic reticulum

Opinions agree that muscle fiber propagation velocity
(MFPV) decreases with fatigue and that EMG power
spectrum shifts during fatigue, mainly owing to a slow-
down of MFPV [3–16]. The fact that spectral character-
istics can drop even without any changes in MFPV
[17,18], or in proportion exceeding the MFPV changes
[10,11,19–24] has prompted many authors to conclude
that there must be other factors contributing to the
observed shift in the power spectrum too. These factors
and the way they could affect EMG spectral character-
istics have been so far unknown.
Voluntary and electrically elicited EMG signals have
a similar spectral shift during fatigue [13,25]. Simula-
tions performed by Lindström and Petersén [7] have
shown the power spectra of interference EMG signals
and of motor unit (MU) potential (MUP) as essentially
similar and dispersion in the action potentials produced
by individual fibers within MU as leading to dramatic
changes in the MUP power spectrum. Thus, an analysis
of factors affecting the spectral characteristics of MUPs
or M-waves can help us understand the reasons for EMG
power spectrum shift.
The amplitude changes seem to be more complicated
and contradictory since data on increased, almost
unchanged, and decreased amplitude characteristics of
the EMG, M-wave or MUP during fatigue can be found
in the literature. Moreover, simultaneous decrease and
increase in amplitude of MUP and M-wave, detected
with indwelling and surface electrodes, were referred to
as paradoxical. Analysis of changes in MUP or M-wave
size and shape with fatigue suggests peripheral factors
that contribute (together with the central factors) to
changes in amplitude and spectral characteristics of
EMG signals. The aim of the present lecture is to point
out the pitfalls and possible fallacies in the interpretation
of amplitude and spectral characteristics of experimental
results from a theoretical point of view and to analyze
possible reasons for EMG changes with fatigue.
2. Methods
2.1. First attempts to understand the properties of
EMG power spectra
Krakau [26] was the first to represent the nerve fibre
action potential as convolution between the input signal
(IS) and the corresponding transfer (T) function. In
accordance with Borel’s convolution theorem, the author
expressed Fourier’s transform of convolution (i.e. of the
 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36
output signal, OS) as a product of Fourier’s transform
of both (IS and T) functions. Nevertheless, Krakau drew
a conclusion about the effect of velocity on the fre-
quency maximum of an action potential power spectrum
following the expression for the transfer function only.
Thus, the property of one term was translated as a pro-
perty of the product.
Krakau did not pay attention to a possible effect of
IS power spectrum on the sensitivity of action potential
power spectrum. Such an effect can be significant in the
case of signals produced by a single infinite excitable
fibre [27,28] and muscle fibre [29,30].
To understand the complicated interplay between dif-
ferent physical and physiological processes that affect
the recorded EMG signal and its power spectrum, Lind-
ström [5] considered the signal filtering in the muscle as
a volume conductor. Attempting to describe the spec-
trum shape in terms of well-defined physical parameters,
the author related all expressions to the action potential
just outside the fibre. In line with Krakau [26], Lind-
ström noted that all formulas contained the factor 2πf/n,
which indicated that any change in the MFPV (n) was
accompanied by translation of the spectrum along the
frequency (f) axis. Lindström [5] gave also a mathemat-
ical model describing the effect of the average MFPV
(n) on the EMG power spectrum:
S(f) ⫽ v⫺2·G(f / v)
Following this line, Lindström et al. [31], Stulen and De
Luca [15], Lindström and Petersén [7], Basmajian and
De Luca [32], Arendt-Nielsen and Mills [20], Zwarts et
al. [33], Merletti et al. [10,11], Hägg [34], Merletti and
Lo Conte [35,36] considered that theory predicted linear
relation between all three characteristic frequencies (the
median, mean and maximum frequency) and MFPV.
This statement implies that changes in MFPV act only
as a scaling factor on potential duration and thus, on the
potential frequency distribution. However, changes in
MFPV reflect on the source, i.e. on a potential profile
existing over the fibre membrane and thus on the poten-
tial field produced by this source. The product of IAP
duration and MFPV determines the potential profile
length. We would like to demonstrate the effect of
changes not only in MFPV but also in IAP during
fatigue.
2.2. Fundamentals of SFAP, MUP, and M-wave
power spectra
It is well known from in vitro experiments that,
besides MFPV, duration (Tin) of IAP, amplitudes of IAP
and negative after-potential change during fatigue [37–
43]. Moreover, duration of different phases of IAP alters
non-simultaneously and in different proportion. Since
individual parameters cannot be affected independently
by an explorer and, as a rule, the source-electrode dis-
15
tance is unknown in experiments, to analyse the pro-
cesses, we have resorted to mathematical modelling of
muscle potentials that do not suffer from these problems.
Not only extracellular potential produced by infinite
fibre [27,28], but also the potential produced by finite
length skeletal muscle fibre, as well as by MUs, can be
considered [29,30,44–46] as convolution and thus as the
output signal (OS) of a linear timeshift-invariant system.
This gives us a chance to describe the power spectrum
in terms of physically well-defined parameters and also
to relate their changes to physiologically clear events.
First temporal derivative of IAP is the input signal
(IS) in the case of single fibre action potential (SFAP),
MUPs and M-waves. Impulse response (IR) in the case
of single fibre is the sum of potentials generated by two
dipoles (propagated in opposite directions from the end-
plate to the ends of the fibre) in the corresponding vol-
ume conductor and detected respectively by an electrode
or multi-electrode. In the case of MUP or M-wave, the
corresponding IR is the sum of impulse responses of all
individual active fibres. Thus IAP amplitude, shape, dur-
ation, and magnitude of negative after-potentials affect
IS, while distances between electrode and activated
fibres, arrangement and longitudinal position of the elec-
trode with respect to the end-plate area, volume conduc-
tor properties, location and scattering of the end plates
(1) and fibre ends, desynchronization of fibre activation,
diameters of the fibres, and propagation velocity along
individual fibres affect the corresponding IR.
The volume conductor is assumed to be infinite or
semi-infinite, homogeneous and anisotropic (Kan=5). We
have not taken into account the more complicated
restriction of the volume conductor (that increases MUP
and M-wave size), nor the presence of layers of weaker
conductivity (skin and fat) whose effect roughly corre-
sponds to an increase of equivalent source-electrode dis-
tances [47,48]. These parameters and their effects remain
unchanged during fatigue, but considerably complicate
description and solution of the problem.
IAP amplitude, duration, and shape during different
stages of fatigue have been varied mainly according to
parameters obtained in in vitro experiments by other
authors [37–43]. The result from in vivo experiment is
also taken into consideration [49,50]. Data on changes in
MFPV are taken from single fibre experiments [51,52].
Since SFAP, MUP or M-wave are expressed through
convolution, their power spectra are represented in
accordance with Borel’s convolution theorem as a pro-
duct of the power spectra of IS and the corresponding
IR. The source-electrode distance affects the IR power
spectrum, while the IS power spectrum is distance-inde-
pendent. The simulations represented in the paper are
performed for anotomical parameters of fibres and MUs
typical for M. biceps brachii.
16 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36
3. Effect of intracellular action potential
parameters and propagation velocity on spectral
and amplitude characteristics of motor unit
potentials
3.1. Spectral characteristics
3.1.1. Power spectrum of single fibre action potentials
When potentials are detected close to the fibre, IR
(simulated for various MFPV) is in a higher frequency
range (Fig. 1, left column, middle row) than IS (Fig.
1, upper row). Consequently, their product (i.e. power
spectrum of SFAP) is mainly determined (Fig. 1, left
column, lower row) by the frequency range of IS. Thus,
although the power spectrum of IS does not depend on
MFPV, its effect makes the shift of the resultant power
spectrum of OS (i.e. of SFAP) due to MFPV changes
insignificant [29,30] and the characteristic frequencies
(Fmax, Fmed, Fmean) change only slightly. On the contrary,
changes in the duration of IAP (Fig. 2, left column,
upper row) affect the power spectrum of OS strongly
(Fig. 3, left column). The characteristic frequencies
decrease considerably and in parallel with the increase
of duration. These results demonstrate that interrelations
between the power spectra of IS and IR can modify sen-
Fig. 1. Effect of changes in propagation velocity on power spectra
(PS) of impulse responses, IR (middle row), PS of SFAP (lower row,
OS, left), and PS of MUP (lower row, OS, right) upon 0.2 mm fibre-
or MU axis-electrode distance. (upper row) - PS of input signal, IS,
i.e. of the first temporal derivative of IAP. Longitudinal position of
the electrode in respect of the end-plate area was x ⫽ 25 mm. Different
curves are for different velocities (v ⫽ 2, 3, 4 and 5 m/sec). Curves
with dot are for v ⫽ 2 m / sec.
sitivity of OS power spectrum to changes in parameters
that affect not only IR but also IS power spectra.
With the increase of distance (y) from the fibre, the
IR power spectrum shifts to a lower frequency (Fig. 4,
left column, from top to bottom). As a result, SFAP
power spectrum becomes sensitive to changes in MFPV
(Fig. 4, second and third column). Nevertheless, changes
in the characteristic frequencies Fmax, Fmed and Fmean with
MFPV are quite different. Contrary to Fmax (Fig. 4,
fourth column, curves with ‘o’), Fmed and Fmean can even
increase with decrease of velocity (Fig. 4, fourth column,
second and third rows, curves with + and ∗). These dif-
ferences reflect increase in high frequency components
in the power spectrum of SFAP with distance. These
components correspond to stepwise discontinuities in IR
that occur when the propagated unit dipoles reach the
fibre ends [45]. The greater the fibre-electrode distance,
the greater the relative weight of discontinuities is. Irres-
pective of the fact that their effect could be invisible in
the IR power spectrum, they are visible in the SFAP
power spectrum due to their amplification by the IS
power spectrum (Fig. 1, upper row). In the time domain,
this is expressed by increased relative weight of positive
terminal phases (that reflect extinction of the excitation
waves at the fibre ends) with distance in respect to the
main ones (Fig. 5, left column, from top to bottom).
3.1.2. Power spectrum of motor unit potentials or M-
wave
Lindström and Petersén [7] predicted that spatial and
temporal dispersion of SFAPs produced by individual
fibers within MU, could lead to dramatic changes in the
power spectrum of MUP. Indeed, under intraterritorial
MUP detection, power spectrum of IR for MU (Figs. 1
and 3 middle row, right) is in a much lower frequency
range (due to spatial and temporal dispersion) than that
for SF (Figs. 1 and 3, middle row, left). Consequently,
the IR power spectrum mainly determines the MUP
power spectrum. The latter is also shifted to lower fre-
quency (Figs. 1 and 3, lower row, right). Moreover, it
additionally shifts to lower frequency when MFPV
decreases (Fig. 1, lower row, right). In contrast to SF,
changes in IAP duration practically do not affect the
MUP spectral distribution under intraterritorial MUP
detection (Fig. 3, lower row, compare left and right).
The sensitivity of IR power spectrum to spatial and
temporal dispersion of SFAPs decreases with the
increase of MU axis-electrode distance. Consequently,
IRs for single fibre and for MU are almost identical
under extraterritorial detection (compare Figs. 4 and 6,
left columns). The low frequency spectral components
of SFAP and MUPs that reflect the main phases are also
almost identical (compare Figs. 4 and 6, second and third
columns). The weight of high frequency components
(that reflect the excitation extinction), however, is
reduced in the MUP power spectrum. In the temporal
 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36 17

Fig. 2. Effect of increase in IAP duration (upper row), in negative after-potential amplitude (middle row) and of simultaneous increase in IAP
duration and after-potential amplitude (lower row) on IS power spectrum (middle and right column). The left column represents IAPs in time
domain. Curves with dot are for initial IAP, i.e. IAP with the shortest (1 ms) IAP duration, and smallest after-potential. The right column represents
the same as the middle one but a lower frequency range is displayed.

domain, this corresponds to a longer duration and

Fig. 3. Effect of increase in IAP duration from 1 to 5 ms on power
spectra (PS) of IS (upper row), PS of SFAP (lower row, OS, left) and
PS of MUP (lower row, OS, right) upon 0.2 mm fiber- or MU axis-
electrode distance. (middle row) – PS of IR for SF (left) and MU
(right). Longitudinal position of the electrode in respect of the end-
plate area was x ⫽ 25 mm. Different curves are for different IAP dur-
ation. Curves with dot are for the shortest (1 ms) IAP duration.
smaller amplitude of MUP terminal phases (Fig. 5, com-
pare terminal positive phases in left and right columns).
Consequently, all characteristic frequencies of the MUP
power spectrum change in parallel (Fig. 6, fourth
column), contrary to the case of SFAP (Fig. 4, fourth
column). Thus, usage of single fibre action potential
models instead of MUP ones can result in false depen-
dencies of frequency characteristics.
To estimate the effect of changes in IS upon extraterri-
torial MUP detection, let us analyse the effect of increase
in IAP duration (Fig. 2, left column, upper row), in IAP
negative after-potential amplitude (Fig. 2, left column,
middle row), and the effect of simultaneous increase in
both parameters (Fig. 2, left column, low row). Such
changes in IAP could be observed during fatigue. All
these IAP changes result in a considerable enhancement
of low frequency components in the IS power spectrum
(Fig. 2, second column, from top to bottom). The third
column represents the same power spectra but in the fre-
quency range corresponding to that of the power spectra
of IR and MUP under extraterritorial detection. When
combined with velocity decrease observed during
fatigue, these changes in IS power spectrum affect all
the characteristic frequencies in the same direction as
velocity does. This is valid for signal detection with
monopolar (Fig. 7) as well as bipolar electrode oriented
transversally (Fig. 8) or longitudinally (Fig. 9). Irrespec-
tive of the type of detection, increase in the IAP duration
(lines with ‘o’) or in IAP after-potential amplitude (lines
with ‘∗’), as well as their combination (lines with ‘+’)
cause a stronger decrease of the characteristic fre-
18 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36

Fig. 4. Effect of changes in velocity (from 2 to 5 m/s in 1m/s step) on power spectra (PS) of IR of single fibre (left column) and PS of SFAP
represented in absolute (second column) and relative units (third column). Fourth column represents change in characteristic frequencies Fmax
(curves with ‘o’), Fmean (curves with ‘∗’), and Fmed (curves with ‘+’). Different rows are for different fibre-electrode distances: y ⫽ 6 mm (upper
row), y ⫽ 12 mm (middle row) and y ⫽ 18 mm (lower row). Curves with dot are v=2 m/s.

Fig. 5. Effect of changes in velocity (from 2 to 5 m/s in 1m/s step) on SFAP (left column) and MUP (right column) detected at different fibre-
or MU axis-electrode distances (y). Longitudinal position of the electrode from the end-plate region was 25 mm. The fibres were with asymmetrical
position of the end-plates in respect to the end-plates (mean fibre length was 120 mm, shorter fibre semilength was 55 mm). Curves with dot are
for v ⫽ 2 m /s. Negativity is upward. The terminal phases are positive (downward).

quencies than when induced only by MFPV decrease 3.1.3. Effect of the longitudinal position of electrode
(lines without markers). The effect is prominent at any on characteristic frequencies
studied distance from MU axis (different columns in As Lindström and Petersen [7], Roy et al. [53], Lateva
Figs. 7, 8 and 9). et al. [54], Dimitrov and Dimitrova [29], Hogrel et al.
 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36 19

Fig. 6. The same as Fig. 4 but for motor unit.

Fig. 7. Reduction of the characteristic frequencies Fmax (upper row), Fmean (middle row) and Fmed (lower row) of MUP detected by a monopolar
electrode at 25 mm from the end-plate region. Different curves represent changes induced by reduction in MFPV (solid lines without mark);
reduction in MFPV combined with increase of IAP duration (lines with ‘o’); reduction in MFPV combined with increase of negative after-potential
(lines with ‘∗’) as well as reduction in MFPV combined with increase of IAP duration and negative after-potential amplitude (lines with ‘+’).
Different columns are for different MU axis – electrode distances (y): y ⫽ 6 mm (left column); y ⫽ 12 mm (middle column); y ⫽ 18 mm (right
column).

[55], Dimitrova et al. [46], and Farina et al. [56,57] have
found out, longitudinal position of the electrode affects
SFAP and MUP power spectra. The range of frequency
deviations due to moving the electrode along the muscle
can be considerably (about 3–4 times) greater than that
induced by the maximum change in the negative after-
potential amplitude (Fig. 10) or in IAP duration. This
fact, and the effect of fibre length on power spectrum
(Lindström et al., [6]; Bazzy et al. [58]; Inbar et al. [59];
Okada [60]; Dimitrova et al. [61]; Lateva et al. [54];
Doud and Walsh [62]; Mannion and Dolan [63]; Dimi-
trov and Dimitrova [45]; Dimitrova et al. [46]; Cechetto
et al. [64]) can compromise the results obtained under
dynamic contraction, when detecting electrode can move
together with the skin. To overcome this problem at least
partially, one could use a long plate electrode (oriented
with its longer side along the fibre length) for monopolar
detection, or two plate electrodes for a belly-tendon
20 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36

Fig. 8. The same as Fig. 7 but for signal detected by transversal single differentiating electrode (TSD) with IED ⫽ 10 mm.

Fig. 9. The same as Fig. 7 but for signal detected by longitudinal single differentiating electrode (LSD) with IED ⫽ 10 mm.

detection with one long (for example 80 mm in the case
of biceps brachii) pole at the muscle belly and the other
pole at the tendon beyond any end of the active fibres
(Fig. 11).
3.2. Amplitude characteristics
3.2.1. Literature data on amplitude characteristics and
causes of fatigue
Contrary to spectral characteristics that change
roughly in parallel with the force decline during fatigue,
increase, slight (insignificant) change and decrease of
EMG amplitude with fatigue are reported in literature
(see for example De Luca [65], and Fitts [66]). Data on
differences in the character of EMG amplitude changes
have prompted many investigators to conclude that
amplitude characteristics are not reliable [3,16,67]. In
spite of this, EMG signals in the time domain and signal
amplitude characteristics are predominantly used when
causes for fatigue are analysed [3,4,9,12,14,49,50,68–
91].
Comparing voluntary and electrically excited tension
 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36 21

Fig. 10. Longitudinal variation of Fmean of power spectra of signals detected by point monopolar (upper row) or longitudinal single differentiating
(bipolar) electrode with IED ⫽ 10 mm (lower row) located at different MU axis-electrode distances (columns from left to right are for y ⫽ 0,6,12
and 18 mm, respectively). Different curves are for different negative after-potentials. Curves with dot are for the smallest negative after-potential.

Fig. 11. Longitudinal variation of Fmean of power spectra of signals detected at 18 mm from MU axis by monopolar plate electrodes (upper row)
or belly-tendon electrodes (lower row) of different length (L) oriented along the fibres. The electrode width was 5 mm. Position of the electrode
corresponded to geometrical centre of the plate located at the belly. The second plate of the belly-tendon electrode was located at the tendon. Its
dimensions (40 mm x 5 mm) were maintained unchanged. Different curves are for different negative after-potentials. Curves with dot are for the
smallest negative after-potential.

with EMG signals in the time domain and signal ampli-
tude, Merton [81] obtained results that became an
important landmark in the studies of fatigue. The author
has shown that reduced force production can be without
a decline of signal amplitude. Merton concluded then
that the site of fatigue was peripheral and the reduced
central drive, as well as the impaired neuromuscular
transmission was unimportant. Naess and Stom-Mathi-
son [83], Krnjevič and Miledi [78,79], Stephens and
Taylor [91] have challenged the latter conclusion.
Bigland-Ritchie [92] remarked that, “… traditionally, if
force loss is accompanied by a parallel decline in electri-
cal activity, fatigue is attributed to failure of excitation;
but if the electrical activity is undiminished, the failure is
attributed to the contractile system”. Many authors have
shared this opinion [7,68,74,82,87,93]. An increase in
the muscle fibre excitation threshold obtained in a num-
ber of studies [41,76,78,94–97] has supported this tra-
dition. It is unclear, however, whether there are other
factors that could evoke decline in electrical activity.
The increase of EMG amplitude is more difficult to
explain. Following Edwards and Lippold [98], many
authors attribute the increased EMG amplitude and area
to recruitment of additional MUs (necessary to compen-
sate for the decrease in force of contraction). Others attri-
bute the increased EMG amplitude to an increased firing
frequency, and/or synchronization of MU recruitment
[67,99–106]. Central mechanisms may contribute to the
increase of EMG amplitude (RMS) and area, but they
cannot explain an increase of MUP or M-wave ampli-
tude. Many investigators [9,14,17,49,50,70,71,75,107–
114] have found a decrease in MUP or M-wave ampli-
tude, preceded by its increase. Only peripheral mech-
anisms should be responsible for this.
Duchateau and Hainaut [108] have explained the
initial increase in amplitude by a reduced dispersion
between potentials generated by individual fibres in the
result of presynaptic and/or end-plate potentials facili-
tation. Fitch and McComas [109] have assumed that the
initial increase in amplitude is most probably a function
of the supernormal period of excitability of muscle
fibres. Later on, Hicks and McComas [49], Hicks et al.
[75], and McComas et al. [50] have attributed aug-
mentation of the M-wave amplitude to enhanced elec-
trogenic Na+ - K+ pumping, resulting in the increase of
IAP amplitude. In line with this, Hicks and McComas
[49] have proposed to use the change in M-wave ampli-
tude as an indirect index of Na+ - K+ pumping in intact
22 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36
human muscles. Celichovski et al. [107] have maintained
that increase in action potential amplitude in individual
muscle fibres seems improbable, and similarly to Duch-
ateau and Hainaut [108], explained the augmentation of
M-wave amplitude with a better synchronization of
potentials from individual fibres. Furthermore, Cupido et
al. [71], who supported idea on enhanced pumping, have
realised that the progressive rise in extracellular [K+]
during contraction should precede the enhanced Na+ -
K+ pumping and, consequently, the M-wave would begin
to diminish from the onset of contraction due to depolar-
ization of the muscle fibres. The authors were surprised
by the lack of any immediate decline in M-wave in their
experiments. Thus they assumed as a more attractive
explanation of paradoxical augmentation of M-wave the
more effective summation of individual potentials due
to increase of IAP duration and prominent negative after-
potentials. Nevertheless, Chan et al. [70] also explained
enlargement of M-wave with the increase of IAP ampli-
tude due to membrane hyperpolarization and with the
prolongation of negative peak duration of the individual
muscle fiber potentials as a result of propagation velo-
city slowdown.
Simultaneous increase in RMS of EMG recorded by
surface electrodes and decrease in RMS of EMG
detected by needle electrodes is classified [32,115,116]
as a ‘paradox’, although it has been predicted earlier that
increase in the length of depolarization zone could lead
to similar changes in SFAP amplitude, depending on
fibre-electrode distance [117–120]. Furthermore, Gydi-
kov et al. [110] have used this dependence to explain an
increase in the amplitude of MUP detected with a more
distant surface electrode and a simultaneous decrease in
the amplitude of MUP detected with a closer surface
electrode. Opposite changes in the amplitudes of M-
waves detected by intramuscular and surface electrodes
can be also seen in Moritani et al. [14] (compare their
Fig. 3 with their Figs. 5 and 6). The authors have noticed
that surface detected M-wave in the beginning of stimu-
lation with 20 and 50 Hz increases, while M-wave simul-
taneously detected by intramuscular electrode, decreases.
The authors have not explained the increase, while the
decline of intramuscular M-wave has been attributed to
reduction in muscle membrane excitability. Zijdewind et
al. [114] also considered as paradoxical the simultaneous
rise and fall of the EMG signal amplitude measured at
different sites of the human first dorsal interosseus mus-
cle during sustained contractions, at 50% maximum vol-
untary contraction.
3.2.2. Effect of IAP spike lengthening on SFAP and
MUP amplitude
Besides in the form of convolution, MUP or M-wave
can be also represented as a linear summation of tem-
porally and spatially dispersed single fibre action poten-
tials. SFAPs have phases reflecting the excitation arising
at the end plates (leading edge) and its extinction at the
fibre ends (terminal phases), as well as phases that reflect
propagation of depolarized zones from the end plates to
the fibre ends. Since the formation of an electric field
and its magnitude is essentially a spatial problem, the
size of individual SFAP is determined by the IAP profile
whose effect is distance-dependent. So are the MUP or
M-wave amplitudes.
To understand changes in MUP, M-wave and EMG
signal amplitude with fatigue, let us consider changes in
SFAP or MUP amplitude with IAP profile (or depolar-
ized zone over the fibre membrane) lengthening due to
increase of IAP duration or MFPV. The potential gener-
ated by an individual depolarized zone could be con-
sidered as that produced by two oppositely directed
stacks of distributed dipoles located at the fibre axis
[121]. The stacks correspond to the fibre portions that
are in phase of de- and repolarization, respectively. Since
each stack produces a dipole field [121], the source could
be schematically represented by two oppositely directed
equivalent point dipoles (Fig. 12). At small fibre-elec-
trode distances (Fig.12, y1), the distance between the two
dipoles is much greater than y1. Thus, interaction
between the opposite in sign potential fields produced
by these dipoles is relatively weak. Nevertheless, max-
ima of the negative fields (Fig. 12, dashed lines) from
opposite dipoles are in the region between the dipoles
and the negative fields enforce each other. Increasing the
distance between the dipoles (Fig. 12, right scheme) will
result in reduction of the additional negative potential
from the other dipole and thus in decrease of the extra-
cellular potentials detected at these short distances. The
resulting potential field has dipole character.
When the fibre-electrode distance is not very small
(Fig. 12, y3), interaction between the two oppositely
directed fields becomes significant. In this case the aver-
age fibre-electrode distance (y3) for the studied SF or
MU is greater than the distance at which the sites of
maximum negative potentials produced by each equival-
ent dipole coincide (Fig. 12, left scheme, y2 = b/√2,
where b is the inter-dipole distance). Distance (d)
between the points with maximum negative potential
(Fig. 12, left scheme, y3) decreases with the increase of
distance between the two equivalent dipoles (Fig. 12,
right scheme, y3). Consequently, if the strength of
dipoles in both schemes is maintained unchanged, the
negative potential produced by the two dipoles at y3
should be higher in the right-hand scheme where nega-
tive maxima coincide.
At larger distances from the fibre (Fig. 12, left
scheme, y4), the negative maximum of one dipole (Fig.
12, dashed lines) is produced in the region where the
other dipole produces a high positive potential. Inter-
acting, the overlapping fields diminish each other every-
where, except in the region between the zero lines (Fig.
12, dotted lines) where both potentials are negative, but
 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36 23

Fig. 12. Schematic presentation of differences in interaction between dipole fields produced by short (left scheme) and long (right scheme) IAP
spike profiles at different radial distances y (dashed-dotted lines) of the electrode. In both schemes, the left and right point equivalent dipoles
substitute for the distributed dipoles corresponding to the IAP phase of de- and repolarization, respectively. Different lines represent zero lines
(dotted lines), positions of the positive maxima (solid lines) and negative maxima (dashed lines) of the individual dipole fields. See text for
more details.

their magnitude is rather small and close to zero. At such
distances from active fibres a quadruple character of the
potential field prevails. Increasing the IAP profile length
and thus the distance between the dipoles (Fig. 12, right
scheme), reduces interaction between the opposite fields
(that diminishes them) and thus increases the extracellu-
lar potentials at these source-electrode distances. The
increase will continue approximately up to length of the
depolarized zone b=y4∗√2.
Thus, SFAP amplitude (Fig. 13, left column)
decreases with lengthening of IAP spike profile when
the observation point is close to the fibre (curves with
dot), and increases at large distances from the fibre. This
corresponds to the theoretically predicted dependence
[119].
In the case of MUP or M-wave, the character of
dependence is almost the same at large distances (Fig.
13, middle and right columns, lines without marker).
Also, it is almost the same in cases of detection by
bipolar electrode, irrespective whether the electrode is
oriented transversally (TSD) or longitudinally (LSD) to
the fibre (Fig. 14). The greater the MU axis-electrode
distance, the greater the MUP amplitude relative
increase. Difference in changes of SFAP and MUP

Fig. 13. Effect of increasing in the length of depolarized zone (i.e. of IAP spatial profile) on amplitudes of SFAP (left column) and MUP detected
at different longitudinal distance (x) of a monopolar electrode from the end-plate area. MUPs above end-plate area, x ⫽ 0 mm (middle column)
or at x ⫽ 25 mm (right column) are represented for two cases of the length increase: at the expense of increase in MFPV (upper row) and at the
expense of increase in IAP spike duration (lower row). SFAP was not affected by the way of increasing in the length of depolarized zone. Different
curves are for different fibre-electrode or MU axis-electrode distance (from top to bottom y ⫽ 18,12,6, and 0 or 0.2 mm). Curves with dot are for
the shortest distance from the fibre (y ⫽ 0.2 mm) or MU axis (y ⫽ 0 mm).
24 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36

Fig. 14. Effect of increasing in the length of depolarized zone (i.e. of IAP spatial profile) on amplitudes of MU signals detected by transversal
(TSD) and longitudinal (LSD) single differentiating electrodes. (left and middle columns) are for TSD electrode located at longitudinal distance
x ⫽ 0 and x ⫽ 25 mm from the end-plate area, respectively. (right column) is for LSD electrode located at longitudinal distance x ⫽ 25 mm from
the end-plate area. The signals are represented for two cases of the length increase: at the expense of increase in MFPV (upper row) and at the
expense of increase in IAP spike duration (lower row). Different curves are for different MU axis-electrode distances (from top to bottom y=18,12,6,
and 0 mm). Curves with dot are for the shortest distance from MU axis (y ⫽ 0 mm).

amplitudes with IAP profile lengthening is more con-
siderable under intraterritorial detection. Radial decline
of SFAP decreases with IAP profile lengthening
[119,122,123]. Consequently, MUP amplitude (in con-
trast to SFAP amplitude) can increase with IAP profile
lengthening even under intraterritorial detection (Fig. 13,
middle and right columns, lines with dot) due to the
enhancing effect of more distantly located muscle fibres.
When IAP profile lengthening is due to MFPV increase,
however, the slopes of the curves are lower than under
IAP duration increase. This is due to the higher temporal
dispersion of potentials from individual fibres under
velocity increase.
The obtained results show that the MFPV decrease
has to shorten IAP profile length and thus to result in a
lower amplitude of MUP detected from skin surface (at
large MU-electrode distance). This is irrespective of sim-
ultaneous shift in the signal power spectrum to lower
frequencies due to MFPV decrease (Figs. 7, 8 and 9).
Thus, the increase in surface detected MUP or M-wave
amplitude during fatigue should not be attributed to a
weaker attenuation of the lower-frequency components
with distance, as Stulen and De Luca [116], De Luca
[115], and Basmajian and De Luca [32] believed.
The increase in negative after-potential amplitude
(Fig. 2, left column, second row) can be considered as
lengthening of IAP duration and thus of IAP spatial pro-
file. It could be represented schematically as an
additional distance increase between the two oppositely
directed equivalent dipoles (Fig. 12). The greater the
MU-electrode distance, the greater the effect of negative
after-potential increase on amplitude characteristics is
(Fig. 15). This is valid for SFAP (Fig. 15, first and
second rows) and MUP detected monopolarly (Fig. 15,
third and fourth rows) or bipolarly (Fig. 15, fifth row).
4. Simulation of peripheral muscle fatigue
4.1. Simultaneous variations of IAP parameters and
MFPV during fatigue
Since not only MFPV decreases during fatigue, let us
analyze the changes in MUPs due to the corresponding
simultaneous variations of IAP amplitude, duration and
shape (including after-potentials), MFPV and desynch-
ronization in activation of individual muscle fibres.
As direct in vivo data on changes in IAP
accompanying fatigue in humans are lacking, we have
relied on data obtained in vitro. As such experiments
have shown [37–43], few stages of IAP alteration can
be distinguished during repetitive stimulation of muscle
fibres. Initially IAP spike duration increases mainly at
the expense of a slowdown of IAP repolarization phase.
The rate of IAP rise remains practically unchanged,
while IAP amplitude slightly decreases. On the contrary,
the amplitude of the early negative after-potential
increases. Subsequently, the absolute value of the resting
potential, spike amplitude, and rate of IAP rise decrease,
along with a further slowdown of the IAP falling phase
and an amplitude increase of the early negative after-
potential. We have divided the in vitro observed IAP
changes during fatigue into four stages (Fig. 16). The
initial parameters characterizing IAP before fatigue are:
 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36 25

Fig. 15. Effect of increase in negative after-potential amplitude (percent in respect to IAP spike amplitude) on relative changes in amplitude (left
column), RMS (middle column) and area (right column) of signals produced by a single fibre (SF) or motor unit (MU). Different rows are for
monopolar detection from SF at longitudinal location of the electrode, x ⫽ 0 mm (upper row) and x ⫽ 25 mm (second row), monopolar detection
from MU at x ⫽ 0 mm (third row) and x ⫽ 25 mm (fourth row), and for detection of MU signals with bipolar LSD electrode (IED ⫽ 10 mm) at
x ⫽ 25 mm (lower row). Different curves are for different fibre-electrode or MU axis-electrode distances (from top to bottom y ⫽ 0 or (0.2), 6,
12, and 18 mm). Curves with dot are for the shortest distance from the fibre (y ⫽ 0.2 mm) or MU (y ⫽ 0 mm) axis.

characterizing the different stages of fatigue are rep-

resented in Table 1.
Data on simultaneous MFPV changes are necessary
to determine changes in IAP spatial profile during
fatigue. We have taken 4.0 m/s for initial MFPV corre-
sponding to the muscle state before fatigue. During indi-
vidual stages of fatigue it was 3.5, 3.25, 3.1, 3.05 m/s
respectively. These values correspond to those obtained
in human single muscle fibre experiments [51]. Since
the increase of IAP spike duration exceeds the relative
decrease in MFPV, IAP spike profile length increases
during fatigue. This corresponds to the results reported

Table 1
Changes in the IAP parameters with fatigue in respect to initial ones
(100%)

Fig. 16. Intracellular action potential (IAP) in temporal domain for Stage IAP Depolarization Repolarization After-potential
different stages of fatigue. Curve with dot is for initial stage. of amplitude duration (%) duration (%) amplitude (%)
fatigue (%)

IAP amplitude -120 mV; depolarization duration -0.7 1 98 100 130 200
ms; repolarization duration -1 ms, and negative after- 2 93 120 170 300
3 80 150 220 400
potential amplitude -3.5 mV. We have assumed them as 4 50 200 300 600
totalling 100%. The relative values of IAP parameters
26 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36

by Juel [76] from in vitro experiments of mouse ever, at 18 mm, and IED ⫽ 10 mm, the amplitudes
fatiguing. decrease during the last stage of fatigue.

4.2. Changes in signal amplitude during fatigue
Amplitudes of MUP (detected by a monopolar
electrode) corresponding to the first two stages of fatigue
are greater than the amplitude during the initial phase,
even at y ⫽ 6 mm from MU axis (Fig. 17, left column,
upper row and Fig. 18, left column, upper row, the curve
with marker). This is due to the IAP spike lengthening
and negative after-potential increase, irrespective of the
smaller IAP amplitude (Fig. 16) and MFPV. However,
this effect is insufficient to compensate for a greater
decrease of IAP amplitude during subsequent stages
(third and fourth) of fatigue. At 6 mm from MU axis,
MUP amplitude decreases during the third stage and in
the last stage of fatigue reaches a lower value than that
before the fatigue (Fig. 17, left column, upper row and
Fig. 18, left column, upper row, the curve with marker).
The effect of the negative after-potential increases
with MU-electrode distance (Fig. 15). Thus MUP ampli-
tude increases during the third stage of fatigue at 12 and
18 mm and even during the fourth stage at 18 mm from
the MU axis (Fig. 17, left column, second and third rows
and Fig. 18, left column, upper row). The character of
changes in amplitude of MU signal detected by TSD
(Fig. 17, second and third columns; Fig. 18, left column,
second and third rows) and LSD (Fig. 17, fourth column;
Fig. 18, first column, fourth row) electrodes is similar
to that described for MUP monopolar detection. How-
4.3. Change in duration of different MUP phases
during fatigue
Alterations in duration of IAP (Tin) or/and MFPV
determine the changes in duration of MUP negative
phase. The effects are also distance dependent. In gen-
eral, at small and medium distances, changes in Tin
mainly affect the duration of MUP negative phase. At
large distances, changes in MFPV and negative after-
potentials are the main factors affecting MUP or M-wave
duration. However, there are differences between the
effects depending on the position of the detecting elec-
trode: midway between the end plate area and tendons,
or above the end plate region. In the latter case, Tin
affects negative MUP duration even at large distances.
Terminal phases of SFAPs repeat the shape and dur-
ation of IAP and do not depend on MFPV. Their polarity
is positive above the fibres (Fig. 5, left column) and
negative behind the fibre ends [117]. In MUPs or M-
waves, the shape of terminal phases represents the sum
of asynchronous IAPs according to the arrival times at
the fibre ends, which depend on scattering of the end-
plates and ends of the fibres and on asynchrony in fibres’
activation and MFPV along individual fibres (Fig. 5,
right column). Also, duration of the terminal phase does
not change with distance from the source, although its
relative weight in detected potentials increases with the
increase of source-electrode distance.

Fig. 17. Simulated changes in MU signals detected by monopolar and bipolar electrodes during different stages of fatigue at different distances
(y) from MU axis (different rows). Different columns are for monopolar electrode located at 25 mm longitudinal distance (x) from end-plates area
(left column), for transversal single differentiating (TSD) electrode located at x ⫽ 0 mm (second column) or at x ⫽ 25 mm (third column), for
longitudinal single differentiating electrode (LSD) located at x ⫽ 25 mm (right column). Different curves correspond to different stages of fatigue.
Curves with dot are for initial stage. IED ⫽ 10 mm for bipolar detection.
 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36 27

Fig. 18. Relative changes in amplitude (left column), RMS (middle column) and area (right column) of the MU signals detected at different
distances (y) from MU axis. Different rows are for monopolar electrode located at 25 mm longitudinal distance (x) from end-plates area (upper
row), for transversal single differentiating (TSD) electrode located at x ⫽ 0 mm (second row) or at x ⫽ 25 mm (third row), for longitudinal single
differentiating electrode (LSD) located at x ⫽ 25 mm (lower row). Different curves are for different distances (y) from MU axis (from top to
bottom y ⫽ 18, 12, and 6 mm). Curves with dot are for the smallest MU axis-electrode distance (y ⫽ 6 mm).

MUP duration increases with fatigue. At 6 mm from
the MU axis this is due to an increasing IAP spike dur-
ation and to decreasing propagation velocity. At 12 and
18 mm from MU axis, far from the innervation zone and
tendons, duration of the MUP negative phase is practi-
cally unaffected by the IAP spike duration, contrary to
the effect of after-potential enlargement. This is like in
the case of infinite fibres [118,124]. The terminal phase
duration increases both with IAP spike duration and
after-potentials increase.
4.4. Changes in MUP area and RMS during fatigue
Amplitude characteristics, such as MUP or M-wave
area, and RMS, combine the effects of changes in MUP
or M-wave amplitude and duration that are not correlated
directly. Consequently, amplitude characteristics (Fig.
18, middle and right columns) can alter in a way differ-
ent from that of amplitude changes (Fig. 18, left
column). Nevertheless, the character of changes is dis-
tance-dependent and MUP area or RMS could also
increase, remain almost unchanged, or decrease during
fatigue. As a result of simultaneous changes in MUP
amplitude and duration, MUP area increases more
strongly than MUP amplitude and its value at 18 mm
from the MU-axis continues to increase even during the
last stage of fatigue (Fig. 18, right column, upper curves)
irrespective of 50% reduction in amplitude of IAP. How-
ever, RMS changes non-monotonously like MUP ampli-
tude (Fig. 18, middle column).
The presented results show that simultaneous decrease
and increase in amplitude of MUP and M-wave detected
with indwelling and surface electrodes are not paradoxi-
cal. Distance-dependent amplitude characteristics of
SFAP, MUP and M-wave can explain this, taking into
account a possible lengthening of IAP profile during
fatigue. The character of changes in amplitude character-
istics discussed for monopolar detection of MUP and M-
wave [125] is valid for bipolar detection, irrespective of
longitudinal or transversal electrode orientation (Figs. 17
and 18 for TSD and LSD).
Since not only fatigue, but the level of muscle force
also affects EMG amplitude and frequency character-
istics, Luttmann et al. [8] and Hägg et al. [4] have pro-
posed the Joint Analysis of EMG Spectrum and Ampli-
tude (JASA) method. In accordance with this method,
however, muscle fatigue could be misclassified as ‘force
decrease’ in smaller or more superficial muscles, as well
as in the final stages of fatigue or during maximum vol-
ume contraction, when EMG amplitude decreases [14].
The results presented above demonstrate that changes
of EMG amplitude characteristics with fatigue can be
28 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36

contradictory and misleading, if the distance-dependent

effects of changes in the IAP spike length, duration,
propagation velocity, and negative after-potentials are
not taken into account and the distance from active fibres
is not controlled. This can be a reason why Gerdle et al.
[3] failed to find a close correlation between RMS and
peak torque, in contrast to simultaneously decreasing
mean frequency and peak torque.

4.5. Fatigue index

To overcome the contradictory changes in EMG

amplitude characteristics with fatigue due to distance-
dependent effects, one could try using the properties of
MUP or M-wave terminal phases. Since the shape of the
terminal phases represent that of the sum of asynchro-
nous IAPs according to the arrival times at the fibre ends,
their duration does not change with distance from the
source. The area of terminal phase normalized in respect
of their amplitude increases not only under larger tem-
poral dispersion in fibres activation as in the case of rein-
nervated MU [126], but also under increasing IAP dur-
ation, negative after-potentials, and under MFPV
decrease. The area to amplitude ratio of the terminal
phase does not depend on the source-electrode distance
and volume conductor properties but it could be difficult
to determine its onset.
To overcome this problem, one could use the signal
positive area or signal area instead of the terminal phase
area. This is because the signal should be without DC
component [127]. The obtained index also monoton-
ically increases with fatigue (Fig. 19), except in the case
of monopolar or transversal single differentiated elec-
trode, when there is a positive phase before the negative
one. This problem could be avoided, if monopolar or Fig. 20. Effect of type of IAP after-potential and of high—pass filter
transversal single differentiated electrode is positioned on shape of MUPs detected by a monopolar electrode. (upper row)—
Intracellular action potentials (IAP) with negative after-potential (curve
above the end-plate area, or if a belly-tendon electrode
with circle), without after-potential (curve without mark) and with
is used. However, an additional masking problem could positive after-potential (curve with filled circle); (middle row) - MUPs
appear, if the corner frequency of the high pass filter is produced by the corresponding IAPs and (lower row) the same MUPs
20 Hz, as is generally recommended [32]. In this case, after applying a filter that cuts off frequencies within the 0–20 Hz
the positive terminal phase could be seen as biphasic range.
(positive-negative), even in the case when IAP has a
negative after-potential (Fig. 20, lower row).

Fig. 19. Index of fatigue, i.e. relative changes of normalized (in respect of the terminal phase amplitude) area of MU signal detected by monopolar
(left column), transversal single differentiating, TSD (second and third columns) and longitudinal single differentiating LSD (fourth column)
electrodes under different stages of fatigue. Different curves correspond to different MU axis-electrode distances (y ⫽ 6, 12, and 18 mm). Curves
with dot are for the smallest MU axis-electrode distance (y ⫽ 6 mm). (x) denotes longitudinal position of the electrode from the end-plate area.
 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36 29

4.6. Changes in spectral characteristics during fatigue
As it could be expected, the spectral characteristics
decrease during fatigue in proportion exceeding the
MFPV changes, irrespective of electrode type, arraign-
ment and distance of the electrode from the end-plate
region (x) and from MU axis (Fig. 21).
4.7. The corner frequency of the high pass filter
necessary to register signs of fatigue
4.7.1. Literature recommendations
Basmajian and De Luca [32] recommended 20 Hz as
a corner frequency of the high pass filter for surface
detected EMG signal. This is because motion artifacts
are generally in a lower frequency range and because
the frequency components of the EMG signal below 20
Hz are unstable and fluctuate unpredictably. These
reasons are also confirmed in the European recommen-
dations on signal processing methods [128]. The rec-
ommended low frequency cut-off for EMG spectral
analysis is, however, reduced there to 10 Hz. This is
motivated by the fact that, in some cases, the EMG spec-
trum includes relevant information concerning the fre-
quency rates of active MUs within this frequency range.
The unwanted low frequency components in EMG signal
could be also due to changes in the negative after-poten-
tial during fatigue. A considerable widening of the ter-
minal positive phase (reflecting increase of the negative
after-potentials) could be noticed in recordings perfor-
med by Hainaut and Duchateau [129], Fig. 1 and Mer-
letti and Lo Conte [130,36], Fig. 1 during fatigue.
However, the effect of the negative after-potentials is
analysed only in a few studies [29,124,127,131,132],
although it has been known that it could change with
fatigue [37,38,42]. The effect of negative after-potentials
on EMG signal shape is also analyzed [30,46,133,134]
after the publication of European recommendations.
Since the EMG spectrum includes relevant information
concerning the changes of negative after-potentials
within a frequency range below 10 Hz [127,132], the
corner frequency of the high pass filter should be 0.5 or
1 Hz when muscle fatigue is analyzed. The use of active
electrodes made of non-polarizable material [135,136]
should solve (at least partially) the problems with motion
artifacts and polarization of the electrode.
4.7.2. Improper filtering or IAP with a positive after-
potential?
Shape of terminal phases in signals detected monopol-
arly is closely related to that of IAP. If the electrode is
located above the end-plate region, the terminal phase

Fig. 21. Relative changes in velocity (bold lines) and in characteristic frequencies (columns from left to right are for Fmax, Fmed and Fmean,
respectively) induced by combined changes in parameters with fatigue. Different rows are for MU signal detected by monopolar (upper row), TSD
(second and third rows) and LSD (fourth row) electrodes. Different thin curves are for different distances (y ⫽ 6, 12, and 18 mm) from MU axis.
Curves with dot are for the smallest MU axis-electrode distance (y ⫽ 6 mm). (x) denotes longitudinal position of the electrode from the end-
plate area.
30 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36
must be positive (Fig. 20, middle row, curves without
marker and with empty circle) when the IAP is without
after-potential or with a negative after-potential (Fig. 20,
upper row, curves without marker and with empty
circle). However, one can see (Fig. 1 in McComas et
al. [50], and Fig. 2 in McFadden and McComas [137])
positive-negative terminal phase in recordings performed
in the proximity of a motor point during fatigue. The
pattern of terminal phases in the latter figures is very
similar to that shown in our Fig. 20 (middle row, curve
with filled circle) when MUP is simulated for IAP with
positive after-potential (Fig. 20, upper row, curve with
filled circle). IAP detected by the authors after stimu-
lation (Fig. 1 in Hicks and McComas [49], and Fig. 5
in McComas et al. [50]) also demonstrates slight signs
of a positive after-potential. The control recording of M-
wave performed before stimulation, however, shows no
sign of positive after-potential. Does this mean that the
after-potential becomes positive in certain conditions
during fatigue or the amplifiers used by the authors were
not very stable? The amplifiers were reported to be with
pass-bands of 5–5kHz. However, if the corner frequency
of high-pass filter is about 20 Hz, MUPs for IAP with
any after-potential should demonstrate signs of the ‘pres-
ence’ of positive after-potential in IAP (Fig. 20, lower
row). Thus, if one could rely on recordings, it would be
possible to extract information about dynamics of IAP
processes basing on non-invasive methods of signal
detection, i.e. without microelectrode technique.
4.8. Interpretations of EMG changes during fatigue
could be misleading, if the effects of IAP profile
lengthening and negative after-potential increase are
not taken into account
The possibility of different distance-dependent
changes in MUP or M-wave amplitudes with fatigue
makes searching for the cause of muscle fatigue on the
basis of association or dissociation of force and
accompanying it electrical activity [58,74,77,82,87,
92,93,138,139] not very reliable, if the distance-depen-
dent effect of IAP profile changes is not taken into
account. The ratio between RMS-amplitude of each
EMG burst and the mean force generated in that burst
can hardly be a measure of neuromuscular efficiency
[102,140,141] since decline in electrical activity can
occur without failure of excitation. Increased
IEMG/force ratio can also hardly indicate that more neu-
ral energy than before is needed for the production of
certain muscle tension after fatigue loading [142]. The
same is valid also for the ratios between muscle tension
and M-wave parameters [143].
Increased EMG amplitude by itself can hardly be used
as a measure of enhanced neural input [144], as indirect
index of Na+ - K+ pumping [49], or as an indicator of
recruitment of larger MUs in response to fatigue [145],
because the EMG signal can increase (due to a lengthen-
ing of IAP profile and negative after-potential
increasing) without alterations in neural input and
recruitment of additional MUs. An increased EMG
amplitude as a measure or an index of fatigue [146–149]
can also be misleading because (irrespective of consider-
able IAP changes with fatigue) EMG amplitude could
change only slightly or even decrease at a rather small
or medium distance from the active fibres.
The results described show that IAP change must be
taken into account for a correct interpretation of EMG
changes with fatigue. On the other hand, a question on
sequence of events resulting in changes of IAP arises.
Analysis of literature data on relations between Ca2+
increase and changes in IAP, could result in the follow-
ing hypothesis on processes affecting muscle fibre poten-
tials during fatigue.
5. Hypothesis: Ability of muscle fibres to uptake
Ca2+ back into the sarcoplasmic reticulum could be
the limiting site for fatigue
Rates of Na+ influx in muscle fibres and K+ efflux
from them are increased during muscle activity. The
finding that K+ is accumulated extracellularly and Na+
intracellularly suggests that activation of the Na+ - K+
pump is insufficient [150], or that the pump capacity is
limited [150,151]. Thus, extracellular K+ accumulation
occurs [86,88,152–156] despite the fact that the Na+ -
K+ pumps are adapted to the increased demand during
activity [73,157].
Increased extracellular K+ induces a decrease in the
membrane potential, which in turn causes a reduced
amplitude and propagation velocity of IAP [76,158].
Reduction in transmembrane concentration gradients for
Na+ and K+ may interfere with excitability of the fibres
[69,76,80,84,88,156,159,160]. Excitability of muscle
fibres is suppressed at high extracellular K ⫹ [76,161–
165]. Since IAP is the first step in the chain of excitation-
contracting events [166,167], it has been suggested that
alteration in sarcolemma function induces muscle fatigue
by preventing cell activation [86,88,90]. Edwards [168]
has suggested that cell depolarization would provide a
safety mechanism to protect the cell against ATP
depletion and Ca2+ accumulation.
Reduction in transmembrane concentration gradients
for Na+ and K+ during muscle activity should take place
not only in fast fatigable, but also in slow, fatigue-resist-
ant muscle fibres. Thus, it can not explain more pro-
nounced fatigability of the fast muscle fibres. Besides,
the IAP changes are coupled with contractility of the
muscle fibres. The IAP in fast fatigable muscle fibres is
practically unchanged during repetitive activation when
the contractility is inhibited by hypertonic solutions
[40,169], i.e. when there is no Ca+ release from sarco-
 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36
plasmic reticulum (SR) but movement of Na+ and K+
ions continues. Consequently, reduction in transmem-
brane concentration gradients for Na+ and K+ due to the
muscle activity is hardly the main factor that determines
IAP changes during muscle fatigue.
Let us consider data on possible effect of Ca2+ on IAP
and then try to look for differences in Ca2+ regulation
in both types of the fibres. Ishiko and Sato [170] found
that IAP shape changed with the increase of Ca2+ con-
centration in extracellular solution. These changes
strongly resemble those in IAP during fatigue [37–43].
Later, Howell et al. [171] as well as Howell and Oetliker
[172] also analyzed effect of higher Ca2+ concentration
in extracellular solution on the IAP but in smaller range
of Ca2+ concentration. The slowing of IAP repolarization
phase was more pronounced under higher extracellular
Ca2+ concentration in their experiments as well as in
those of Ishiko and Sato [170]. On the other hand, basing
on the results of Adrian and Peachey [173], the authors
looked for a notch and hump in IAP as indicators for
propagation in the tubular system. When extracellular
Ca2+ concentration was elevated to 10 mM Ca2+, the
notch and hump were attenuated or abolished. The
authors concluded that, along with the elevated potass-
ium concentration, elevated Ca2+ concentration could
contribute to or cause propagation failure in the t-system
and thus decline of the tetanus tension. They did not pay
attention to the fact that the abolishing or attenuation of
the notch and hump could be due to the slowing of the
IAP repolarization phase as was described by Adrian and
Peachey [173]. Consequently, the abolishing of the notch
and hump should not be a reason for conclusion on
propagation failure in the t-system. It could be assumed
that, higher Ca2+ concentration in extracellular solution
leads to higher intracellular free Ca2+ concentration,
which affects IAP. Taking into account all these results,
we would assume that accumulation of Ca2+ should be
in the root not only of decrease in contractile force [174–
182], but also of the changes in IAP shape observed dur-
ing fatigue.
As slow oxidative muscle fibres provide almost full
Ca2+ uptake into the SR under normal conditions [183–
185], they are more resistant to fatigue. However, they
cannot provide fast contraction. Increased quantity of
Ca2+ release could provide the fast contraction, but the
capacity of ATP-dependent pump to uptake Ca2+ into the
SR is limited. To overcome Ca2+ accumulation (at least
for a short time), there are Ca2+-binding buffers (like
parvalbumin) in the fast glycolytic fibres. In spite of this
and contrary to the slow fibres, exercise-induced Ca2+
accumulation occurs in the fast fibres [183–185], prob-
ably due to the limited capacity of the buffers. As a
result, force generation declines [177,178,180,182]. On
the other hand, K+ conductance increases due to acti-
vation of Ca2+-dependent K+ channels [80,86,88,89,186–
188]. Activation of Ca2+-dependent K+ channels leads to
31
a decrease in the membrane potential, which in turn
causes a reduced amplitude and propagation velocity of
IAP. Thus, the limiting factor of muscle fatigue could
be the ability of muscle fibres to uptake Ca2+ back into
the SR.
Furthermore, due to the presence of Ca2+-binding buf-
fers in the fast muscle fibres, intracellular free Ca2+ is
lower in rest and it is necessary to perform some contrac-
tions to approach the optimal level providing maximum
contraction [182]. This effect is known as potentiation
[189–193] and it is considered to be a property of fast-
twitch muscle fibres [72,193–197]. On the other hand,
lower intracellular free Ca2+ in rest leads to a weaker
activation of Ca2+-dependent K+ channels and thus to
more negative resting potential [198].
In the presence of ischaemia, Ca2+ uptake into the SR
becomes slower due to the reduced ability of ATP-
dependent pump. Thus, in the fast-twitch fatigable mus-
cle fibres, Ca2+ accumulation becomes faster. Moreover,
the exercise-induced Ca2+ accumulation should occur
also in slow oxidative muscle fibres. As a result, they
become also fatigable and the relative changes in MFPV
and other parameters of IAP and SFAP in slow fibres
are similar to those in fast fibres during continuous
activity under normal conditions [52,192,199]. Thus,
fatigue occurs more rapidly in this case. Besides, recov-
ery from fatigue does not take place in ischaemic con-
ditions [81] due to the reduced ability of ATP-dependent
Ca2+ pump.
6. Conclusions
1. Theory does not predict a linear relation between the
characteristic frequencies (maximum, mean and
median) and MFPV.
2. IAP duration and after-potential magnitude are among
parameters affecting MUP or M-wave power spec-
trum and thus, EMG power spectrum.
3. Usage of single fibre action potential models instead
of MUP ones can result in false dependencies of fre-
quency characteristics.
4. Due to the IAP profile lengthening and an increase
in the negative after-potential, surface detected EMG
amplitude characteristics can increase even when IAP
amplitude decreases considerably during fatigue.
5. Increase in surface detected MUP or M-wave ampli-
tude should not be attributed to a weaker attenuation
of the low-frequency components with distance.
6. Simultaneous decrease and increase in amplitude of
MUP and M-wave detected with indwelling and sur-
face electrodes are regular, not paradoxical.
7. Corner frequency of the high pass filter should be 0.5
or 1 Hz when muscle fatigue is analyzed.
8. The area of MUP or M-wave normalized in respect
of the terminal phase amplitude could be useful as a
fatigue index.
32 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36
9. Ability of muscle fibres to uptake Ca2+ back into the
sarcoplasmic reticulum is the limiting site for fatigue.
If this hypothesis is valid, IAP changes are not a cause
of fatigue; they are due to it.
References
[1] H.E. Piper, Electrophysiologie Menschlichen Muskeln, Springer,
Berlin, 1912.
[2] S. Cobb, A. Forbes, Electromyographic studies of muscle fatigue
in man, Am J Physiol 65 (1923) 234–251.
[3] B. Gerdle, B. Larsson, S. Karlsson, Criterion validation of surface
EMG variables as fatigue indicators using peak torque. A study
of repetitive maximum isokinetic knee extensions, J Elecromyogr
Kinesiol 10 (2000) 225–232.
[4] G.M. Hägg, A. Luttmann, M. Jäger, Methodologies for evaluating
electromyographic field data in ergonomics, J Electromyogr
Kinesiol 10 (2000) 301–312.
[5] L. Lindström. On the Frequency Spectrum of EMG Signals,
Techn Rep 7: 70, Res Lab Med Electr Chalmers Univ Technol
Göteborg Sweden. 1970;1-39.
[6] L. Lindström, R. Magnusson, I. Petersén, Muscular fatigue and
action potential conduction velocity changes studied with fre-
quency analysis of EMG signals, Electromygraphy 10 (1970)
341–356.
[7] L. Lindström, I. Petersén, Power spectrum analysis of EMG sig-
nals and its application. In: Desmedt JE, editor. Computer-aided
Electromyography. Prog Clin Neurophysiol Basel: Karger.
[8] A. Luttmann, M. Jäger, J. Sökeland, W. Laurig, Electromyo-
graphic study on surgeons in urology, part II: determination of
muscular fatigue, Ergonomics 39 (1996) 298–313.
[9] R. Merletti, D. Farina, M. Gazzoni, M.P. Schieroni, Effect of age
on muscle functions investigated with surface electromyography,
Muscle Nerve 25 (2002) 65–76.
[10] R. Merletti, M. Knaflitz, C.J. De Luca, Myoelectric manifes-
tations of fatigue in voluntary and electrically elicited contrac-
tions, J Appl Physiol 69 (1990) 1810–1819.
[11] R. Merletti, M. Knaflitz, C.J. De Luca, Electrically evoked myoe-
lectric signals, Crit Rev Biomed Eng 19 (1992) 293–340.
[12] R. Merletti, S. Roy, Myoelectric and mechanical manifestations
of muscle fatigue in voluntary contractions, J Orthop Sports Phys
Therapy 24 (1996) 342–353.
[13] K. Mills, Power spectral analysis of electromyogram and com-
pound muscle action potential during muscle fatigue and recov-
ery, J Physiol (Lond) 326 (1982) 401–409.
[14] T. Moritani, M. Muro, A. Nagata, Intramuscular and surface elec-
tromyogram changes during fatigue, J Appl Physiol 60 (1986)
1179–1185.
[15] F.R. Stulen, C.J. DeLuca, Frequency parameters of the myoe-
lectric signals as a measure of muscle conduction velocity, IEEE
Trans BioMed Eng 28 (1981) 515–523.
[16] N.K. Vøllestadt, Measurement of human muscle fatigue, J Neuro-
sci Meth 74 (1997) 219–227.
[17] E.H.J.P. Linssen, M. Jacobs, D.F. Stegeman, E.M.G. Joosten, J.
Moleman, Fatigue in McArdle’s disease. Muscle fiber conduction
velocity and surface EMG frequency spectrum during ischaemic
exercise, Brain 113 (1990) 1779–1793.
[18] M. Naeije, H. Zorn, Relation between EMG power spectrum
shifts and muscle fiber action potential conduction velocity
changes during local muscular fatigue in man, Eur J Appl Physiol
50 (1982) 23–33.
[19] L. Arendt-Nielsen, A. Forster, K.R. Mills, EMG power spectral
shift and muscle-fibre conduction velocity during human muscle
fatigue, J Physiol (Lond) 353 (1984) 77.
[20] L. Arendt-Nielsen, K.R. Mills, The relationships between mean
power frequency of the EMG spectrum and muscle fibre velocity,
Electroenceph Clin Neurophysiol 60 (1985) 130–134.
[21] B. Bigland-Ritchie, E.F. Donovan, C.S. Roussos, Conduction
velocity and EMG power spectrum changes in fatigue of sus-
tained maximal efforts, J Appl Physiol 51 (1981) 1300–1305.
[22] H. Broman, G. Bilotto, C.J. De Luca, Myoelectric signal conduc-
tion velocity and spectral parameters: Influence of force and time,
J Appl Physiol 58 (1985) 1428–1437.
[23] C. Krogh-Lund, K. Jørgensen, Changes in conduction velocity,
median frequency, and root mean square-amplitude of the electro-
myogram during 25% maximal voluntary contraction of the tri-
ceps brachii muscle, to limit endurance, J Appl Physiol 63 (1993)
60–69.
[24] M.L. Zwarts, T.W. Van Weerden, H.T.M. Haeven, The relation
between average muscle conduction velocity and EMG power
spectra during isometric contraction, recovery and applied
ischemia, Eur J Appl Physiol 59 (1987) 212–216.
[25] H. Kranz, A.M. Williams, J. Cassel, D.J. Caddy, R.B. Sijberstein,
Factors determining the frequency content of the electromyog-
ram, J Appl Physiol 55 (1983) 392–399.
[26] C.F.T. Krakau, A note on the Fourier transform of Lorente de
Nó’s potential function of the external field of a nerve in a vol-
ume conductor, Kungl Fysiogr Sällsk Lund Förhandl 27 (1957)
177–183.
[27] G.V. Dimitrov, Z.C. Lateva, N.A. Dimitrova, Power spectra of
extracellular potentials generated by an infinite, homogeneous
excitable fibre, Med Biol Eng Comput 28 (1990) 24–30.
[28] Z.C. Lateva, G.V. Dimitrov, N.A. Dimitrova, Power spectra of
single infinite fibre extracellular potentials recorded by a bipolar
electrode, Med Biol Eng Comput 28 (1990) 537–543.
[29] G.V. Dimitrov, N.A. Dimitrova, Fundamentals of power spectra
extracellular potentials produced by skeletal muscle fibre of finite
length. Part II: Effect of parameters altering with functional state,
Med Eng Phys 20 (1998) 702–707.
[30] G.V. Dimitrov, N.A. Dimitrova, Effect of parameters altering
with muscle fibre functional state on power spectra of spatially
filtered extracellular potentials, J Med Eng Technol 25 (2001)
74–78.
[31] L. Lindström, R. Kadefors, I. Petersén, An electromyographic
index for localized muscle fatigue, J Appl Physiol 43 (1977)
750–754.
[32] J.V. Basmajian, C.J. De Luca, Muscles alive; their functions
revealed by electromyography, 5th ed., William and Wilkins, Bal-
timore, MA, 1985.
[33] M.L. Zwarts, T.W. Van Weerden, T.P. Links, H.T.M. Haeven,
H.J.G.H. Oosterhuis, The muscle fiber conduction velocity and
power spectra in familial hypokalemic periodic paralysis, Muscle
Nerve 11 (1988) 166–173.
[34] G.M. Hägg, Interpretation of EMG spectral alterations and alter-
ation indexes at sustained contraction, J Appl Physiol 73 (1992)
1211–1217.
[35] R. Merletti, L. Lo Conte, Advances in processing of surface
myoelectric signals: Part 1, Med Biol Eng Comput 33 (1995)
362–372.
[36] R. Merletti, L. Lo Conte, Surface EMG signal processing during
isometric contractions, J Electromyogr Kinesiol 7 (1997) 241–
250.
[37] J. Hanson, The effects of repetitive stimulation on the action
potential and the twitch of rat muscle, Acta Physiol Scand 90
(1973) 387–400.
[38] J. Hanson, A. Persson, Changes in the action potential and con-
traction of isolated frog muscle after repetitive stimulation, Acta
Physiol Scand 81 (1971) 340–348.
[39] J. Lännergren, H. Westerblad, Action potential fatigue in single
skeletal muscle fibres of Xenopus, Acta Physiol Scand 129 (1987)
311–318.
 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36
[40] H.C. Lüttgau, The effect of metabolic inhibitors on the fatigue
of the action potential in single muscle fibres, J Physiol (Lond)
178 (1965) 45–67.
[41] J.M. Metzger, R.H. Fitts, Fatigue from high- and low frequency
muscle stimulation: Role of sarcolemma action potentials, Exp
Neurol 93 (1986) 320–333.
[42] A. Person, The negative after-potential of frog skeletal muscle
fibres, Acta Physiol Scand 58 (Suppl 205) (1963) 1–32.
[43] N. Radicheva, L. Gerilovsky, A. Gydikov, Changes in the muscle
fibre extracellular action potentials in long-lasting (fatiguing)
activity, Eur J Appl Physiol 55 (1986) 545–552.
[44] G.V. Dimitrov, N.A. Dimitrova, Precise and fast calculation of
the motor unit potentials detected by a point and rectangular plate
electrode, Med Eng Phys 20 (1998) 374–381.
[45] G.V. Dimitrov, N.A. Dimitrova, Fundamentals of power spectra
extracellular potentials produced by skeletal muscle fibre of finite
length. Part I: Effect of fibre anatomy, Med Eng Phys 20 (1998)
580–587.
[46] N.A. Dimitrova, G.V. Dimitrov, O.A. Nikitin, Longitudinal vari-
ations of characteristic frequencies of skeletal muscle fibre poten-
tials detected by a bipolar electrode or multi-electrode, J Med
Eng Techn. 25 (2001) 34–40.
[47] G.V. Dimitrov, C. Disselhorst-Klug, N.A. Dimitrova, A. Tracht-
erna, G. Rau, The presence of unknown layer of skin and fat is an
obstacle to correct estimation of the motor unit size from surface
detected potentials, Electromyogr Clin Neurophysiol 42 (2002)
231–241.
[48] D. Farina, C. Cescon, R. Merletti, Influence of anatomical, physi-
cal, and detection-system parameters on surface EMG, Biol Cyb-
ern 86 (2002) 445–456.
[49] A. Hicks, A. McComas, Increased sodium pump activity follow-
ing repetitive stimulation of rat soleus muscle, J Physiol (Lond)
414 (1989) 337–349.
[50] A. McComas, V. Galea, R.W. Einhorn, Pseudofacilitation: A mis-
leading term, Muscle Nerve 17 (1994) 599–607.
[51] A. Gydikov, P.G. Gatev, G.V. Dimitrov, L. Gerilovsky, Changes
in the parameters of human single muscle fiber potentials with
consecutive discharges, Exp Neurol 76 (1982) 12–24.
[52] E. Stålberg, Propagation velocity in human muscle fibers in situ,
Acta Physiol Scand 70 (Suppl 287) (1966) 1–112.
[53] S.H. Roy, C.J. De Luca, J. Schneider, Effects of electrode
location on myoelectric conduction velocity and median fre-
quency estimates, J Appl Physiol 61 (1986) 1510–1517.
[54] Z.C. Lateva, N.A. Dimitrova, G.V. Dimitrov, Effect of recording
electrode position along a muscle fiber on surface potential power
spectrum, J Electromyogr Kinesiol 3 (1993) 195–204.
[55] J.-Y. Hogrel, J. Duchêne, J.-F. Marini, Variability of some SEMG
parameter estimates with electrode location, J Electromyogr
Kinesiol 8 (1998) 305–315.
[56] D. Farina, P. Madeleine, T. Graven-Nielsen, R. Merletti, L.
Arent-Nielsen, Standardising surface electromyogram recordings
for assessment of activity and fatigue in the human upper trapez-
ius muscle, Eur J Appl Physiol 86 (2002) 469–478.
[57] D. Farina, R. Merletti, M. Nazzaro, I. Caruso, Effect of joint
angle on EMG variables in leg and thigh muscles, IEEE Eng Med
Biol Mag 20 (2001) 62–71.
[58] A.R. Bazzy, J.B. Korten, G.G. Haddad, Increase in electromyog-
ram low-frequency power in nonfatigued contracting skeletal
muscle, J Appl Physiol 61 (1986) 1012–1017.
[59] G.F. Inbar, J. Allin, H. Kranz, Surface EMG spectral changes
with muscle length, Med Biol Eng Comput 25 (1987) 683–689.
[60] M. Okada, Effect of muscle length on surface EMG wave forms
in isometric contractions, Eur J Appl Physiol 56 (1987) 462–466.
[61] N.A. Dimitrova, G.V. Dimitrov, Z.C. Lateva, Influence of the
fiber length on the power spectrum of single muscle fiber extra-
cellular potentials, Electromyogr Clin Neurophysiol 31 (1991)
387–398.
33
[62] J.R. Doud, J.M. Walsh, Muscle fatigue and muscle length interac-
tion: effect on the EMG frequency components, Electromyogr
Clin Neurophysiol 35 (1995) 331–339.
[63] A.F. Mannion, P. Dolan, The effects of muscle length and force
output on the EMG power spectrum of the erector spinae, J Elec-
tromyogr Kinesiol 6 (1996) 159–168.
[64] A.D. Cechetto, P.A. Parker, R.N. Scott, The effects of four time-
varying factors on the mean frequency of a myoelectric signal, J
Electromyogr Kinesiol 11 (2001) 347–354.
[65] C.J. De Luca, Myoelectrical manifestation of localized muscular
fatigue in humans, CRC Crit Rev Biomed Eng 11 (1984) 251–
279.
[66] R.H. Fitts, Cellular Mechanisms of Muscle Fatigue, Physiol Rev
74 (1994) 49–94.
[67] E. Hultman, H. Sjöholm, Electromyogram, force and relaxation
time during and after continuous electrical stimulation of human
skeletal muscle in situ, J Physiol (Lond) 339 (1983) 33–40.
[68] A.R. Bazzy, D.F. Donnelly, Diaphragmatic failure during loaded
breathing: role of neuromuscular transmission, J Appl Physiol 74
(1993) 1679–1683.
[69] R. Bouclin, E. Charbonneau, J.M. Renaud, Na+ and K+ effect on
contractility of frog sartorius muscle: implication for mechanism
of fatigue, Am J Physiol 268 (1995) C1528–1536.
[70] K.M. Chan, L.P. Andres, Y. Polykovskaya, W.F. Brown, Dis-
sociation of the electrical and contractile properties in single
human motor units during fatigue, Muscle Nerve 21 (1998)
1786–1789.
[71] C.M. Cupido, V. Galea, A.J. McComas, Potentiation and
depression of the M-wave in human biceps brachii, Muscle Nerve
491 (2) (1996) 541–550.
[72] R.M. Enoka, N. Trayanova, Y. Laouris, L. Bevan, R.M. Rein-
king, D.G. Stuart, Fatigue-related changes in motor unit action
potentials of adult cats, Muscle Nerve 15 (1992) 138–150.
[73] M.E. Everts, T. Clausen, Excitation-induced activation of the
Na+-K+ pump in rat skeletal muscle, Am J Physiol 266 ((Cell
Physiol 35)) (1994) C925–934.
[74] L. Grimby, J. Hannerz, B. Hedman, The fatigue and voluntary
discharge properties of single motor units in man, J Physiol
(Lond) 316 (1981) 545–554.
[75] A. Hicks, J. Fenton, S. Garner, A.J. McComas, M wave potenti-
ation during and after muscle activity, J Appl Physiol 66 (1989)
2606–2610.
[76] C. Juel, Muscle action potential propagation velocity changes
during activity, Muscle Nerve 11 (1988) 714–719.
[77] D. Kernell, A.W. Monster, Motoneuron properties and motor
fatigue, Exp Brain Res 46 (1982) 197–206.
[78] K. Krnjevič, R. Miledi, Failure of neuromuscular propagation in
rats, J Physiol (Lond) 140 (1958) 440–461.
[79] K. Krnjevič, R. Miledi, Presynaptic failure of neuromuscular
propagation in rats, J Physiol (Lond) 149 (1959) 1–22.
[80] M.I. Lindinger, G. Sjøgaard, Potassium regulation during exer-
cise and recovery, Sports Med 11 (1991) 382–401.
[81] P.A. Merton, Voluntary strength and fatigue, J Physiol (Lond)
123 (1954) 553–564.
[82] K. Murali, D. Pagala, N. Tatsuji, A. Grob, Failure of neuromuscu-
lar transmission and contractility during muscle fatigue, Muscle
Nerve 7 (1984) 454–464.
[83] K. Naess, A. Stom-Mathison, Fatigue of sustained contractions,
Acta Physiol Scand 34 (1955) 351–366.
[84] K. Overgaard, O.B. Nielsen, J.A. Flatman, T. Clausen, Relation
between excitability and contractility in rat soleus muscle: role
of the Na+ - K+ pump and Na+/K+ gradients, J Physiol (Lond)
518 (1) (1999) 215–223.
[85] T.G. Sandercock, J.A. Faulkner, J.W. Albers, P.H. Abbrecht, Sin-
gle motor unit and fiber action potentials during fatigue, J Appl
Physiol 58 (1985) 1073–1079.
[86] O.M. Sejersted, G. Sjøgaard, Dynamics and consequences of pot-
34 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36
assium shifts in skeletal muscle and heart during exercise, Physiol
Rev 80 (2000) 1411–1481.
[87] G.C. Sieck, Y.S. Prakash, Fatigue at the neuromuscular junction,
in: S.C. Gandevia, R.M. Enoka, A.J. McComas, D.S. Stuart, C.K.
Thomas (Eds.), Fatigue. Neural and muscular mechanisms, Ple-
num Press, New York and London, 1995, pp. 83–100.
[88] G. Sjøgaard, Exercise-induced muscle fatigue: The significance
of potassium, Acta Physiol Scand 140 (Suppl.593) (1990) 1–64.
[89] G. Sjøgaard, Role of exercise-induced potassium fluxes underly-
ing muscle fatigue: a brief review, Can J Physiol Pharmacol 69
(1991) 238–245.
[90] G. Sjøgaard, Potassium and fatigue: the pros and cons, Acta Phy-
siol Scand 156 (1996) 157–164.
[91] J.A. Stephens, A. Taylor, Fatigue of maintained voluntary muscle
contraction in man, J Physiol (Lond) 220 (1972) 1–18.
[92] B. Bigland-Ritchie, EMG and fatigue of human voluntary and
stimulat contractions, in: R. Porter, J. Whelan (Eds.), Human
muscle fatigue: physiological mechanisms (Ciba Found Symp
82), Pitman Medical, London, 1981, pp. 130–156.
[93] R.M. Enoka, L.L. Rankin, M.J. Joyner, D.G. Stuart, Fatigability
of rat hindling muscle: associations between electromyogram and
force during a fatigue test, J Physiol (Lond) 408 (1989) 251–270.
[94] E.M. Balog, L.V. Thompson, R.H. Fitts, Role of sarcolemma
action potentials and excitability in muscle fatigue, J Appl Phy-
siol 76 (1994) 2157–2162.
[95] A.J. Fuglevand, The role of the sarcolemma action potential in
fatigue, in: S.C. Gandevia, R.M. Enoka, A.J. McComas, D.S. Stu-
art, C.K. Thomas (Eds.), Fatigue, neural and muscular mech-
anisms, Plenum Press, New York and London, 1995, pp. 101–
108.
[96] R.A. Gertler, N. Robbins, Differences in neuromuscular trans-
mission in red and while muscles, Brain Res 142 (1978) 160–164.
[97] A. Roed, Fatigue during continuous 20 Hz stimulation of the rat
phrenic nerve diaphragm preparation, Acta Physiol Scand 134
(1988) 217–221.
[98] R.G. Edwards, O.C.J. Lippold, Relation between force and inte-
grated electrical activity in fatigued muscle, J Physiol (Lond) 132
(1956) 677–681.
[99] F. Buchthal, A. Madsen, Synchronous activity in normal and
atrophic muscle, Electroenceph Clin Neurophysiol 2 (1950)
425–444.
[100] S.C. Gandevia, Spinal and supraspinal factors in human muscle
fatigue, Physiol Rev 81 (2001) 1725–1789.
[101] K. Jørgensen, N. Fallentin, C. Krogh-Lund, B. Jensen, Electro-
myography and fatigue during prolonged, low-level static con-
tractions, Eur J Appl Physiol 57 (1988) 316–321.
[102] W.H.J.P. Linssen, D.F. Stegeman, E.M.G. Joosten, M.A. Van’t
Hof, R.A. Binknorst, S.L.H. Notermans, Variability and inter-
relationships of surface EMG parameters during local muscle
fatigue, Muscle Nerve 16 (1993) 849–856.
[103] B. Maton, Motor unit differentiation and integrated surface
EMG in voluntary isometric contraction, Eur J Appl Physiol 35
(1976) 149–157.
[104] B. Maton, Human motor unit activity during the onset of muscle
fatigue in submaximal isometric isotonic contraction, Eur J Appl
Physiol 46 (1981) 271–281.
[105] B. Maton, D. Gamet, The fatigability of two agonistic muscles
in human isometric voluntary submaximal contraction: an EMG
study. II Motor unit firing rate and recruitment, Eur J Appl Phy-
siol 58 (1989) 369–374.
[106] W. Yao, R.J. Fuglevand, R.M. Enoka, Motor-unit synchroniz-
ation increases EMG amplitude and decreases force steadiness
of simulated contractions, J Neurophysiol 83 (2000) 441–452.
[107] J. Celichowski, K. Grottel, A. Rakowska, Changes in motor unit
action potentials during the fatigue test, Acta Neurobiol Exp 51
(1991) 145–155.
[108] J. Duchateau, K. Hainaut, Electrical and mechanical failures
during sustained and intermittent contractions in humans, J Appl
Physiol 58 (1985) 942–947.
[109] S. Fitch, A. McComas, Influence of human muscle length on
fatigue, J Physiol 36 (1985) 205–213.
[110] A. Gydikov, N.A. Dimitrova, D. Kosarov, G.V. Dimitrov,
Influence of frequency and duration of firing on the shape of
potentials from different types of motor units in human muscles,
Exp Neurol 52 (1976) 345–355.
[111] E. Rabischong, Surface action potentials related to torque output
in paraplegics’ electrically stimulated quadriceps muscle, Med
Eng Phys 18 (1996) 538–547.
[112] C.K. Thomas, Fatigue in human thenar muscles paralysed by
spinal cord injury, J Elecromyogr Kinesiol 7 (1997) 15–26.
[113] I. Zijdewind, D. Kernell, Fatigue associated EMG behavior of
the first dorsal interosseous and adductor pollicis muscles in dif-
ferent groups of subjects, Muscle Nerve 17 (1994) 1044–1054.
[114] I. Zijdewind, D. Kernell, C.G. Kukulka, Spatial difference in
fatigue-associated electromyographic behaviour of the human
first dorsal interosseus muscle, J Physiol (Lond) 483 (2) (1995)
499–509.
[115] C.J. De Luca, Physiology and mathematics of myoelectric sig-
nals, IEEE Trans Biomed Eng 26 (1979) 313–325.
[116] F.B. Stulen, C.J. De Luca, The relation between the myoelectric
signal and physiological properties of constant force isometric
contractions, Electroenceph Clin Neurophysiol 45 (1978) 681–
698.
[117] G.V. Dimitrov, N.A. Dimitrova, Extracellular potential field of
a single striated muscle fibre immersed in anisotropic volume
conductor, Electromyogr Clin Neurophysiol 14 (1974) 423–436.
[118] G.V. Dimitrov, N.A. Dimitrova, Bipolar recording of potentials
generated by excitable fibres in a volume conductor, Agressolo-
gie 18 (1977) 235–252.
[119] N. Dimitrova, Influence of the length of the depolarized zone
on the extracellular potential field of a single unmyelinated
nerve fibre, Electromyogr Clin Neurophysiol 13 (1973) 547–
558.
[120] N.A. Dimitrova, G.V. Dimitrov, Extracellular potential field of
a single striated muscle fibre, Electromyogr Clin Neurophysiol
14 (1974) 279–292.
[121] G.V. Dimitrov, N.A. Dimitrova, Extracellular potentials pro-
duced by a transition between inactive and active regions of
an excitable fibre, Electromyogr Clin Neurophysiol 29 (1989)
265–271.
[122] G.V. Dimitrov, C. Disselhorst-Klug, N.A. Dimitrova, E.
Schulte, G. Rau, Simulation analysis of the ability of different
types of multi-electrodes to increase selectivity of detection and
to reduce cross-talk, J Electromyogr Kinesiol 2002 (accepted).
[123] G.V. Dimitrov, Z.C. Lateva, N.A. Dimitrova, Radial changes of
extracellular potential amplitude and integral characteristics and
the inverse problem in electroneurography, Med Biol Eng Com-
put 30 (1992) 357–363.
[124] G.V. Dimitrov, N.A. Dimitrova, Influence of the after-potentials
on the shape and magnitude of the extracellular potentials gener-
ated under activation of excitable fibres, Electromyogr Clin
Neurophysiol 19 (1979) 249–267.
[125] N.A. Dimitrova, G.V. Dimitrov, Amplitude related character-
istics of motor unit and M-wave potentials during fatigue. A
simulation study using literature data on intracellular potential
changes found in vitro, J Elecromyogr Kinesiol 12 (2002)
339–350.
[126] N.A. Dimitrova, C. Disselhorst-Klug, G.V. Dimitrov, G. Rau,
Area to amplitude ratio of the terminal phase of the belly-tendon
detected motor unit potentials could be used to recognise rein-
nervated motor units, Electromyogr Clin Neurophysiol 42
(2002) 267–273.
[127] G.V. Dimitrov, Z.C. Lateva, N.A. Dimitrova, Model of the slow
 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36
components of skeletal muscle potentials, Med Biol Eng Com-
put 32 (1994) 432–436.
[128] R. Merletti, D. Farina, H. Hermens, B. Freriks, J. Harlaar, Euro-
pean recommendations for signal processing methods for sur-
face EMG, SENIAM 8: European recommendations for Surface
ElectroMyoGraphy 8 (1999) 57–68.
[129] K. Hainaut, J. Duchateau, Muscle fatigue, effects of training and
disuse, Muscle Nerve 12 (1989) 660–669.
[130] R. Merletti, L. Lo Conte, Surface EMG signal processing. In:
Hermens HJ, Nene AV, Zilvold G, editors. Proc 11th ISEK
Cong, Enschede, The Netherlands, 1996:33-42.
[131] G.V. Dimitrov, Z.C. Lateva, N.A. Dimitrova, Power spectra of
muscle fibre extracellular potentials, in: W. Wallinga, H.B.K.
Boom, J. de Vries (Eds.), Proc 7th ISEK Cong, 20-23 June,
1988, Enschede, The Netherlands, 1988, pp. 211–215.
[132] Z.C. Lateva, K.C. McGill, The physiological origin of the slow
afterwave in muscle action potentials, Electroencephalogr Clin
Neurophysiol 108 (1998) 462–469.
[133] K.C. McGill, Z.C. Lateva, A model of the muscle-fiber intra-
cellular action potential waveform, including the slow repolariz-
ation phase, IEEE Trans Biomed Eng 48 (2001) 1480–1483.
[134] K.C. McGill, Z.C. Lateva, S. Xiao, A model of the muscle
action potential for describing the leading edge, terminal wave,
and slow after-wave, IEEE Trans Biomed Eng 48 (2001)
1357–1365.
[135] C. Gonran, E. Siebert, P. Fabry, E. Novakov, P.Y. Gumery,
Non-polarisable dry electrode based on NASICON ceramic,
Med Biol Eng Comput 33 (1995) 452–457.
[136] C. Gonran, E. Siebert, S. Yacoub, E. Novakov, Noise of surface
bio-potential electrodes based on NASICON ceramic and Ag-
AgCl, Med Biol Eng Comput 34 (1996) 460–466.
[137] L.K. McFadden, A.J. McComas, Late depression of muscle
excitability in humans after fatiguing stimulation, J Physiol
(Lond) 496 (3) (1996) 851–855.
[138] F. Bellemare, N. Garzaniti, Failure of neuromuscular propa-
gation during human maximal voluntary contraction, J Appl
Physiol 64 (1988) 1084–1093.
[139] R.M. Enoka, L.L. Rankin, D.G. Stuart, K.A. Volz, Fatigue-
related changes in neuromuscular excitability of rat hindlimb
muscles, Muscle Nerve 11 (1988) 1123–1132.
[140] W.H.J.P. Linssen, D.F. Stegeman, E.M.G. Joosten, R.A. Bin-
knorst, M.J.H. Merks, H.J. Ter Laak, S.L.H. Notermans, Fatigue
in type I fiber predominance: A muscle force and surface EMG
study on the relative role of type I and type II muscle fibers,
Muscle Nerve 14 (1991) 829–837.
[141] J.H. Van der Hoeven, T.W. van Weerden, M.J. Zwarts, Long-
lasting supernormal conduction velocity after sustained maximal
isometric contraction in human muscle, Muscle Nerve 16 (1993)
312–320.
[142] K. Häkkinen, P.V. Komi, Effects of fatigue and recovery on
electromyographic and isometric force- and relaxation-time
characteristics of human skeletal muscle, Eur J Appl Physiol 55
(1986) 588–596.
[143] J.M. Heasman, T.R.D. Scott, V.A. Vare, R.Y. Flynn, C.R.
Gschwind, J.W. Middleton, S.B. Rutkowski, Detection of
fatigue in the isometric electrical activation of paralyzed hand
muscle of persons with tetraplegia, IEEE Trans Rehabil Eng 8
(2000) 286–296.
[144] J.H.T. Viitasalo, P.V. Komi, Signal characteristics of EMG dur-
ing fatigue, Eur J Appl Physiol 37 (1977) 111–121.
[145] B.R. Jensen, M. Pilegaard, G. Sjøgaard, Motor control during
fatiguing contractions, in: H. Christensen, G. Sjøgaard (Eds.),
PROCID Symposium: Muscular disorders in computer users:
mechanisms and models, Nat Inst Occup Health, Copenhagen,
1999, pp. 101–103.
[146] H. Christensen, K. Søgaard, B.R. Jensen, L. Finsen, G. Sjøgaard,
35
Intramuscular and surface EMG power spectrum from dynamic
and static contractions, J Electromyogr Kinesiol 5 (1995) 27–36.
[147] D.P. Currier, Measurement of muscle fatigue, J Am Phys Ther
Assoc 49 (1969) 724–730.
[148] H.A. De Vries, Efficiency of electrical activity as a physiologi-
cal measure of the functional state of muscle tissue, Am J Phys
Med 47 (1968) 10–22.
[149] H.A. De Vries, Method for evaluation of muscle fatigue and
endurance from electromyographic fatigue curves, Am J Phys
Med 47 (1968) 125–135.
[150] E. Verburg, J. Hallen, O.M. Sejersted, N.K. Vøllestad, Loss of
potassium from muscle during moderate exercise in humans: a
result of insufficient activation of the Na+ - K+ pump?, Acta
Physiol Scand 165 (1999) 357–367.
[151] C. Juel, J.J. Nielsen, J. Bangsbo, Exercise-induced translocation
of Na+ - K+ pump subunits to the plasma membrane in human
skeletal muscle, Am J Physiol 278 (2000) R1107–1110.
[152] W.O. Fenn, Loss of potassium in voluntary contraction, Am J
Physiol 120 (1937) 675–680.
[153] H. Hirche, E. Schumacher, H. Hagemann, Extracellular K+ con-
centration and K+ balance of the gastrocnemius of the dog dur-
ing exercise, Pflügers Arch 387 (1980) 231–237.
[154] P. Hnik, M. Holas, I. Krekule, N. Friz, S. Mejsnar, V. Smiesko,
E. Ujec, F. Vyskocil, Work-induced potassium changes in skel-
etal muscle and effluent venous blood assessed by liquid ion
exchanger micro-electrode, Pflügers Arch 362 (1976) 85–94.
[155] J.I. Medbø, O.M. Sejersted, Plasma potassium changes with
high intensity exercise, J Physiol (Lond) 421 (1990) 105–122.
[156] G. Sjøgaard, Electrolytes in slow and fast muscle fibers of
humans at rest and dynamic exercise, Am J Physiol 245 (1983)
R25–31.
[157] T. Clausen, Regulation of active Na+, K+ transport in skeletal
muscle, Physiol Rev 66 (1986) 542–580.
[158] A.L. Hodgkin, P. Horowicz, The influence of potassium and
chloride ions on the membrane potential of single muscle fibres,
J Physiol (Lond) 148 (1959) 127–160.
[159] B. Bigland-Ritchie, D.A. Jones, J.J. Woods, Excitation fre-
quency and muscle fatigue: electrical responses during human
voluntary and stimulated contractions, Exp Neurol 64 (1979)
14–27.
[160] D.A. Jones, B. Bigland-Ritchie, Electrical and contractile
changes in muscle fatigue, Biochem Exercise 16 (1986) 372–
392.
[161] S.P. Cairns, J.A. Flatman, T. Clausen, Relation between extra-
cellular [K+], membrane potential and contraction in rat soleus
muscle: modulation by the Na+ - K+ pump, Pflügers Arch 730
(1995) 909–915.
[162] T. Clausen, S.L.V. Andersen, J.A. Flatman, Na+ - K+ pump
stimulation elicits recovery of contractility in K+-paralysed rat
muscle, J Physiol (Lond) 472 (1993) 521–536.
[163] C. Juel, The effect of beta2-adrenoreceptor activation on ion-
shifts and fatigue in mouse soleus muscles stimulated in vitro,
Acta Physiol Scand 134 (1988) 209–216.
[164] J. Lännergren, H. Westerblad, Force and membrane potential
during and after fatiguing, continuous high-frequency stimu-
lation of single Xenopus muscle fibres, Acta Physiol Scand 128
(1986) 359–368.
[165] J.M. Renaud, P. Light, Effect of K+ on the twitch and tetanic
contraction in the sartorius muscle of the frog, Rana Pipiens.
Implication for fatigue in vivo, Can J Physiol Pharmacol 70
(1992) 1236–1246.
[166] A. Sandow, S.R. Taylor, H. Preiser, Role of the action potential
in excitation-contraction coupling, Fed Proc 111 (1965) 6–23.
[167] S.R. Taylor, H. Preiser, A. Sandow, Action potential parameters
affecting excitation-contraction coupling, J Gen Physiol 52
(1972) 421–436.
[168] R.H.T. Edwards, Human muscle function and fatigue, in: R.
36 N.A. Dimitrova, G.V. Dimitrov / Journal of Electromyography and Kinesiology 13 (2003) 13–36
Porter, J. Whelan (Eds.), Human muscle fatigue: physiological
mechanisms (Ciba Found Symp 82), Pitman Medical, London,
1981, pp. 1–18.
[169] A. Rauf Khan, B. Bengtsson, Effects of fatigue on electrical
responses of frog muscle fibres, Acta Physiol Scand 124 (Suppl
542) (1985) 324.
[170] N. Ishiko, M. Sato, The effect of calcium ions on electrical
properties of striated muscle fibres, Jap J Physiol 7 (1957)
51–63.
[171] J.N. Howell, A. Shankar, S.G. Howell, F. Wei, Evidence for t-
tubular conduction failure in frog skeletal muscle induced by
elevated extracellular calcium concentration, J Muscle Res Cell
Motil 8 (1987) 229–241.
[172] J.N. Howell, H. Oetliker, Effects of repetitive activity,
ruthenium red, and elevated extracellular calcium on frog skel-
etal muscle: implications for t-tubule conduction, Can J Physiol
Pharmacol 65 (1987) 691–696.
[173] R.H. Adrian, L.D. Peachey, Reconstruction of the action poten-
tial of frog sartorius muscle, J Physiol (Lond) 235 (1973)
103–131.
[174] E.R. Chin, D.G. Allen, The role of elevations in intracellular
Ca2+ in the development of low frequency fatigue in mouse sin-
gle muscle fibres, J Physiol (Lond) 491 (1996) 813–824.
[175] E.R. Chin, C.D. Balnave, D.G. Allen, Role of intracellular cal-
cium and metabolites in low frequency fatigue in mouse skeletal
muscle, Am J Physiol 272 (1997) C550–559.
[176] T.G. Favero, Sarcoplasmic reticulum Ca2+ release and muscle
fatigue, J Appl Physiol 87 (1999) 471–483.
[177] H. Westerblad, D.G. Allen, Changes in myoplasmic calcium
concentration during fatigue in single mouse muscle fibers, J
Gen Physiol. 98 (1991) 615–635.
[178] H. Westerblad, J.D. Bruton, D.G. Allen, J. Lännergren, Func-
tional significance of Ca2+ in long-lasting fatigue of skeletal
muscle, Eur J Appl Physiol 83 (2000) 166–174.
[179] H. Westerblad, S. Duty, D.G. Allen, Intracellular calcium con-
centration during low-frequency fatigue in isolated single fibers
of mouse skeletal muscle, J Appl Physiol 75 (1993) 382–388.
[180] H. Westerblad, J.A. Lee, J. Lännergren, D.G. Allen, Cellular
mechanisms of fatigue in skeletal muscle, Am J Physiol 261
(1991) C195–209.
[181] J.H. Williams, Contractile apparatus and sarcoplasmic reticulum
function: effects of fatigue, recovery and elevated Ca2+, J Appl
Physiol 83 (1997) 444–450.
[182] J.H. Williams, G.A. Klug, Calcium exchange hypothesis of skel-
etal muscle fatigue: a brief review, Muscle Nerve 18 (1995)
421–434.
[183] H. Gissel, Ca2+ accumulation and cell damage during muscle
contraction, in: H. Christensen, G. Sjøgaard (Eds.), PROCID
Symposium: Muscular disorders in computer users: mechanisms
and models, Nat Inst Occup Health, Copenhagen, 1999, pp.
145–149.
[184] H. Gissel, Ca2+ accumulation and cell damage during low fre-
quency stimulation, Eur J Appl Physiol 83 (2000) 175–180.
[185] H. Gissel, T. Clausen, Excitation-induced Ca2+ uptake in rat
skeletal muscle, Am J Physiol 45 (1999) R331–339.
[186] R. Fink, S. Hase, H.C. Lüttgau, E. Wettwer, The effect of cellu-
lar energy reserves and internal calcium ions on the potassium
conductance in skeletal muscle of the frog, J Physiol (Lond) 336
(1983) 211–228.
[187] H.C. Lüttgau, E. Wettwer, Ca2+-activated potassium conduc-
tance in metabolically exhausted skeletal muscle fibres, Cell
Calcium 4 (1983) 331–341.
[188] B.C. Palotta, Calcium activated potassium channels in rat mus-
cle inactivated from a short open state, J Physiol (Lond) 363
(1985) 501–506.
[189] R.I. Close, Dynamic properties of mammalian skeletal muscles,
Physiol Rev 52 (1972) 129–197.
[190] F. Colomo, P. Rocchi, Staircase effect and post-tetanic potenti-
ation in frog nerve-single muscle fibre preparation, Arch Fisiol
64 (1965) 189–266.
[191] J.E. Desmedt, K. Hainaut, Kinetics of myofilament activation in
potentiated contraction: staircase phenomenon in human skeletal
muscle, Nature 217 (1968) 529–532.
[192] K. Hainaut, M. Golde, Effect of ischaemia on contractile pro-
cesses in human skeletal muscle, Electromyogr Clin Neurophy-
siol 16 (1976) 67–74.
[193] C. Krarup, Enhancement and diminution of mechanical tension
evoked by staircase and by tetanus in rat muscle, J Physiol
(Lond) 311 (1981) 355–372.
[194] S.A. Binder-McLeod, J.C. Dean, J. Ding, Electrical stimulation
factors in potentiation of human quadriceps femoris, Muscle
Nerve 25 (2002) 271–279.
[195] R.W. Grange, R. Vandenboom, J. Xeni, M.E. Houston, Potenti-
ation of in vitro concentric work in mouse fast muscle, J Appl
Physiol 84 (1998) 236–243.
[196] M.E. Houston, J.H. Green, J.T. Stull, Myosin light chain phos-
phorylation and isometric twitch potentiation in intact human
muscle, Pflügers Arch 403 (1885) 348–352.
[197] S.C. Small, M.J. Stokes, Stimulation frequency and force
potentiation in the human adductor pollicis muscle, Eur J Appl
Physiol 65 (1992) 229–233.
[198] C. Juel, Potassium and sodium shifts during in vitro isometric
muscle contraction, and the ion-gradient recovery, Pflügers Arch
406 (1986) 458–463.
[199] P. Gatev, G.V. Dimitrov, A. Gydikov, L. Gerilovsky, Effect of
ischaemia on the potentials of human single muscle fibres, Acta
Physiol Pharmacol Bulg 7 (1981) 3–11.
Nonna Dimitrova was born in Leningrad,
USSR. She received her MS in Biomedical
Engineering from the Institute of Electrical
Engineering, Leningrad, and PhD from the Insti-
tute of Physiology, Bulgarian Academy of
Sciences in 1975. She is Associate Professor in
biophysics at the centre of Biomedical Engineer-
ing, Bulgarian Academy of Sciences. Her
research interests are in the field of mathemat-
ical modelling and bioelectrical phenomena.
George Dimitrov was born in Sofia, Bulgaria.
He received his MS in Biomedical Engineering
from the Institute of Electrical Engineering,
Leningrad, PhD from the Institute of Physi-
ology, Bulgarian Academy of Sciences in 1974,
and DSc in biophysics from Bulgarian Academy
of Sciences in 1999. He is Professor and Head
of Department at the Centre of Biomedical
Engineering, Bulgarian Academy of Sciences.
His research interests are in the field of bioe-
lectrical phenomena.