Mechanomyographic and electromyographic responses during submaximal cycle ergometry


Housh, T.J.; Perry, S.R.; Bull, A.J.; Johnson, G.O.; Ebersole, K.T.; Housh, D.J.; deVries, H.A.

European Journal of Applied Physiology 83(4 -5): 381-387

2001


The purpose of this study was to examine the mechanomyographic (MMG) and electromyographic (EMG) responses during continuous, cycle ergometer workbouts performed at constant power outputs. Eight adults [mean (SD) age, 21.5 (1.6) years] volunteered to perform an incremental test to exhaustion for the determination of peak power (Wpeak) and four, 15-min (or to exhaustion) rides at constant power outputs of 50%, 65%, 80%, and 95% Wpeak. Piezoelectric crystal contact sensors were placed on the vastus lateralis (VL) and vastus medialis (VM) muscles to record the MMG signals. Bipolar surface electrode arrangements were placed on the VL and VM to record the EMG signals. Five-second samples of the MMG and EMG signals were recorded every 30 s at power outputs of 50%, 65%, and 80% Wpeak, and every 15 s at 95% Wpeak. The amplitudes of the selected portions of the signals were normalized to the first values recorded during the continuous rides, and regression analyses were used to determine whether the slope coefficients for the MMG and EMG versus time relationships were significantly (P < 0.05) different from zero. The results indicate that EMG amplitude increased (range of slope coefficients: 0.03-0.56) during the continuous rides for both muscles at all four power outputs (except the VM at 50% Wpeak), while MMG amplitude increased (slope coefficient at 95% Wpeak for VM = 0.19), decreased (range of slope coefficients for VL and VM at 50% and 65% Wpeak = -0.14 to -0.24), or remained unchanged (range of slope coefficients for VL and VM at 80% Wpeak and VL at 95% peak = -0.06 to 0.12) depending on the power output. The patterns of the MMG responses, however, were similar for the VL and VM muscles, except at 95% Wpeak. Fatigue-induced changes in motor-unit recruitment and discharge rates, or muscular compliance may explain the differences between power outputs in the patterns of the MMG amplitude responses.

Eur
J
Appl
Physiol
(2000)
83:
381-387
©
Springer-Verlag
2000
ORIGINAL
ARTICLE
Terry
J.
Housh
Sharon
R.
Perry
Anthony
J.
Bull
Glen
0.
Johnson
Kyle
T.
Ebersole
Dona
J.
Housh
Herbert
A.
deVries
Mechanomyographic
and
electromyographic
responses
during
submaximal
cycle
ergometry
Accepted:
7
August
2000
Abstract
The
purpose
of
this
study
was
to
examine
the
mechanomyographic
(MMG)
and
electromyographic
(EMG)
responses
during
continuous,
cycle
ergometer
workbouts
performed
at
constant
power
outputs.
Eight
adults
[mean
(SD)
age,
21.5
(1.6)
years]
volunteered
to
perform
an
incremental
test
to
exhaustion
for
the
deter-
mination
of
peak
power
(W
peak
)
and
four,
15-min
(or
to
exhaustion)
rides
at
constant
power
outputs
of
50%,
65%,
80%,
and
95%
W
peak
.
Piezoelectric
crystal
contact
sensors
were
placed
on
the
vastus
lateralis
(VL)
and
va-
stus
medialis
(VM)
muscles
to
record
the
MMG
signals.
Bipolar
surface
electrode
arrangements
were
placed
on
the
VL
and
VM
to
record
the
EMG
signals.
Five-second
samples
of
the
MMG
and
EMG
signals
were
recorded
every
30
sat
power
outputs
of
50%,
65%,
and
80%
W
peak
,
and
every
15
s
at
95%
W
peak
.
The
amplitudes
of
the
se-
lected
portions
of
the
signals
were
normalized
to
the
first
values
recorded
during
the
continuous
rides,
and
regres-
sion
analyses
were
used
to
determine
whether
the
slope
coefficients
for
the
MMG
and
EMG
versus
time
rela-
tionships
were
significantly
(P
<
0.05)
different
from
zero.
The
results
indicate
that
EMG
amplitude
increased
(range
of
slope
coefficients:
0.03-0.56)
during
the
con-
tinuous
rides
for
both
muscles
at
all
four
power
outputs
(except
the
VM
at
50%
W
peak
),
while
MMG
amplitude
increased
(slope
coefficient
at
95%
W
peak
for
VM
=
0.19),
decreased
(range
of
slope
coefficients
for
VL
and
VM
at
50%
and
65%
W
peak
=
—0.14
to
—0.24),
or
remained
un-
changed
(range
of
slope
coefficients
for
VL
and
VM
at
80%
W
peak
and
VL
at
95%
W
peak
=
—0.06
to
0.12)
de-
pending
on
the
power
output.
The
patterns
of
the
MMG
responses,
however,
were
similar
for
the
VL
and
VM
muscles,
except
at
95%
W
peak
.
Fatigue-induced
changes
in
S.
R.
Perry
A.
J.
Bull
G.
0.
Johnson
K.
T.
Ebersole
D.
J.
Housh
H.
A.
deVries
T.
J.
Housh
(El)
Department
of
Health
and
Human
Performance,
MABL
143,
University
of
Nebraska-Lincoln,
Lincoln,
NE
68588-0229,
USA
e-mail:
thoush@unlserve.unl.edu
Tel.:
+
1-402-4721160;
Fax:
+
1-402-4724305
motor-unit
recruitment
and
discharge
rates,
or
muscular
compliance
may
explain
the
differences
between
power
outputs
in
the
patterns
of
the
MMG
amplitude
responses.
Key
words
Muscle
wisdom
Fatigue
Muscular
compliance
Mechanomyogram
Electromyogram
Introduction
Mechanomyography
(MMG)
records
the
lateral
oscilla-
tions
of
contracting
skeletal
muscle
fibers
(Barry
1987,
1991;
Barry
and
Cole
1988;
Barry
et
al.
1985,
1990;
Cer-
quiglini
et
al.
1973;
Frangioni
et
al.
1987;
Orizio
et
al.
1989a;
Petitjean
and
Maton
1995;
Stokes
1993;
Zwarts
and
Keidel
1991).
Barry
and
Cole
(1988)
and
Orizio
(1993)
have
indicated
that
the
lateral
oscillations
recorded
as
MMG
are
a
function
of:
(1)
a
gross
lateral
movement
at
the
initiation
of
a
contraction
generated
by
the
non-simulta-
neous
activation
of
muscle
fibers,
(2)
smaller
subsequent
lateral
oscillations
generated
at
the
resonant
frequency
of
the
muscle,
and
(3)
dimensional
changes
of
the
active
fi-
bers.
The
amplitude
of
the
MMG
signal,
however,
is
in-
fluenced
by
many
factors,
including
muscle
temperature,
stiffness,
and
mass,
as
well
as
intramuscular
pressure
and
the
viscosity
of
the
intracellular
and
extracellular
fluid
mediums
(Barry
1987;
Barry
and
Cole
1988,
1990;
Fran-
gioni
et
al.
1987;
Marchetti
et
al.
1974;
Orizio
1993;
Orizio
et
al.
1989a;
Oster
and
Jaffe
1980;
Zhang
et
al.
1992).
Gordon
and
Holbourn
(1948)
suggested
that
the
MMG
reflects
the
"mechanical
counterpart"
of
motor-
unit
electrical
activity
as
measured
by
electromyography
(EMG).
Simultaneous
measurements
of
MMG
and
EMG
during
sustained
and
intermittent
muscle
actions
have
been
used
to
examine
the
electromechanical
aspects
of
neuromuscular
fatigue
(Orizio
1993;
Stokes
1993).
Stokes
(1993)
suggested
that
an
increase
in
EMG
amplitude,
but
declines
in
MMG
amplitude
and
force,
indicate
peripheral,
low-frequency
fatigue
due
to
excitation-contraction
coupling
failure.
Central
fatigue,
however,
is
characterized
by
a
decrease
in
EMG
amplitude
382
and
a
plateau
in
MMG
amplitude
(Stokes
1993).
Thus,
examination
of
the
patterns
of
MMG
and
EMG
ampli-
tude
responses
measured
simultaneously
during
fatiguing
tasks
may
provide
information
regarding
the
physiologi-
cal
mechanisms
underlying
neuromuscular
fatigue.
Most
studies
of
MMG
have
utilized
isometric
muscle
actions,
but
recently,
Shinohara
et
al.
(1997)
and
Stout
et
al.
(1997)
applied
MMG
measurements
to
incremen-
tal
cycle
ergometry.
No
previous
investigations,
how-
ever,
have
examined
the
MMG
responses
during
continuous
cycle
ergometer
workbouts
performed
at
constant
power
outputs.
Thus,
it
is
not
known
whether
the
patterns
of
MMG
amplitude
responses
observed
during
continuous
cycle
ergometry
are
the
same
as
those
observed
during
sustained
isometric
muscle
actions.
It
is
possible
that
during
cycle
ergometry
the
intermittent
nature
of
quadriceps
femoris
activation
affects
the
pat-
tern
of
the
MMG
amplitude
responses
to
a
fatiguing
task
in
a
manner
that
is
different
than
during
a
sustained
isometric
muscle
action.
Furthermore,
there
are
no
data
available
regarding
the
coherence
of
the
patterns
of
MMG
and
EMG
responses
during
continuous
cycle
ergometry.
Therefore,
the
purpose
of
the
present
study
was
to
examine
the
MMG
and
EMG
amplitude
re-
sponses
during
continuous
cycle
ergometer
workbouts
performed
at
constant
power
outputs.
Methods
Subjects
Eight
adults
[3
males
and
5
females,
mean
(SD)
age
=
21.5
(1.6)
years]
volunteered
for
this
investigation.
All
procedures
were
approved
by
the
University
Institutional
Review
Board
for
human
subjects
and
the
subjects
completed
a
health
history
questionnaire
and
gave
their
written
informed consent
to
participate,
prior
to
testing.
The
subjects
visited
the
laboratory
on
five
occasions.
The
first
visit
was
to
perform
an
incremental
test
to
exhaustion,
and
the
remaining
four
visits
were
to
perform
continuous,
constant-power-
output
workbouts.
Each
laboratory
visit
was
separated
by
24
h
and
was
performed
at
the
same
time
of
day.
Incremental
test
to
exhaustion
Each
subject
performed
an
incremental
test
to
exhaustion
on
a
cal-
ibrated
Quinton
(Corval
400)
electronically
braked
cycle
ergometer
at
a
pedal
cadence
of
70
rev
min
-1
.
Seat
height
was
adjusted
for
near
full
extension
of
the
legs,
and
toe
clips
were
used
to
secure
the
subject's
feet
to
the
pedals.
The
subject
was
fitted
with
a
CIC
UNIQ
Cardiomonitor
system
(Leger
and
Thivierge
1988).
Following
a
5-min
warm-up
at
30
W,
the
power
output
was
increased
to
50
W.
Thereafter,
the
power
output
was
increased
by
30
W
every
2
min
until
voluntary
exhaustion
and
the
subject
could
no
longer
maintain
a
pedal
cadence
of
70
rev
min
-1
.
Peak
power
(W
peak
)
was
defined
as
the
highest
power
output
attained
during
the
incremental
test.
Continuous,
constant-power-output
workbouts
Each
subject
visited
the
laboratory
four
additional
times
to
per-
form
rides
at
a
pedal
cadence
of
70
rev
min
-1
to
exhaustion
or
a
maximum
of
15
min.
Five-second
samples
of
the
MMG
and
EMG
signals
(Fig.
1)
were
recorded
during
the
continuous,
constant-
power
output-workbouts
(randomly
ordered)
at
50%
[111
(30)
W],
65%
[144
(39)
W],
80%
[177
(48)
W],
and
95%
[210
(57)
W]
of
Wpeak.
At
50%,
65%,
and
80%
Wp
eak,
the
MMG
and
EMG
signals
were
sampled
at
30-s
intervals.
At
95%
Wp
ea
k
the
signals
were
sampled
every
15
s.
These
power
outputs
were
selected
to
compare
the
MMG
and
EMG
responses
at
four
different
power
outputs
that
represent
moderate-
to
high-intensity
exercise.
Mechanomyography
The
MMG
signals
were
detected
by
piezoelectric
crystal
contact
sensors
(Hewlett-Packard
21050A,
bandwidth
0.02-2000
Hz)
placed
on
the
vastus
lateralis
(VL)
and
vastus
medialis
(VM)
muscles
between
the
EMG
electrodes.
A
stabilizing
ring
and
dou-
ble-sided
adhesive
tape
were
used
to
assure
consistent
contact
pressure
of
the
sensor
(Bolton
et
al.
1989;
Orizio
1993).
The
MMG
signals
were
sampled
at
1000
points
s
-1
,
bandpass
filtered
at
5-
100
Hz,
and
expressed
as
root
mean
square
(rms;
mVrms)
ampli-
tude
(MP100,
Biopac
Systems,
Santa
Barbara,
Calif.,
USA).
The
contraction
period
for
each
pedal
thrust
was
determined
from
the
EMG
burst,
according
to
the
procedures
of
Shinohara
et
al.
(1997).
The
rms
values
for
the
MMG
and
EMG
signals
from
both
muscles
were
calculated
as
the
average
of
four
complete
pedal
thrusts
during
the
5-s
time
window.
Electromyography
Bipolar
surface
(7.62
cm
center-to-center)
electrode
(Quinton
Quick
prep
silver/silver
chloride)
arrangements
were
placed
on
the
dominant
thigh
over
the
VL
and
VM
muscles
according
to
the
procedures
of
Ebersole
et
al.
(1999).
For
the
VL,
the
electrodes
were
placed
midway
between
the
greater
trochanter
and
the
lateral
condyle
of
the
femur.
For
the
VM,
the
electrodes
were
placed
20%
of
the
distance
between
the
medial
gap
of
the
knee
joint
and
the
anterior
superior
spine
of
the
pelvis.
The
interelectrode
distance
was
selected
to
accommodate
placing
the
MMG
sensor
between
the
EMG
electrodes
(Ebersole
et
al.
1999;
Smith
et
al.
1998).
The
ref-
erence
electrode
was
placed
over
the
iliac
crest.
To
ensure
consis-
tent
placement
of
the
electrodes
for
each
testing
session,
marks
were
placed
around
the
circumference
of
the
electrodes
with
a
permanent
marker.
Interelectrode
impedance
was
kept
below
2000
Ohms
by
abrasion
of
the
skin.
The
EMG
signals
were
pre-
amplified
(gain:
x
1000)
using
a
differential
amplifier
(EMG
100,
Biopac
System;
bandwidth
=
10-4000
Hz).
The
EMG
signals
were
sampled
at
1000
points
s
-1
,
bandpass
filtered
(2nd-order
Black-
man
filter)
at
10-500
Hz,
and
expressed
as
the
rms
(µVrms)
am-
plitude
(MP100,
Biopac
Systems).
Statistical
analysis
The
amplitude
values
were
normalized
to
the
first
recorded
samples
of
the
MMG
and
EMG
signals
(initial
amplitude
values)
at
30
s
(for
50%,
65%,
and
80%
Wpeak)
and
15
s
(for
95%
W
peak
)
into
the
workbouts
(Figs.
2,
3,
4,
5),
similar
to
the
procedures
of
Rodriquez
et
al.
(1993).
Time
was
normalized
as
a
percentage
of
the
total
workbout
duration,
according
to
Orizio
and
Veicsteinas
(1992)
and
Rodriquez
et
al.
(1993).
Regression
analyses
(generalized
least-
squares
or
least-squares
when
appropriate
based
on
the
Durbin-
Watson
test
statistic;
Johnston
and
DiNardo
1997)
were
used
to
determine
whether
the
slope
coefficients
for
the
mean
normalized
MMG
amplitude
versus
time
and
EMG
amplitude
versus
time
relationships
were
significantly
different
from
zero.
An
alpha
level
of
P
<
0.05
was
used
for
all
analyses.
Results
The
mean
(SD)
W
P
B
and
maximal
heart
rate
from
the
incremental
test
to
exhaustion
were
221
(60)
W
(range
=
150-300
W)
and
187
(8)
beats
min
-I
(range
=
177-200
beats
min
-1
),
respectively.
All
subjects
com-
Fig.
1
Examples
of
raw
mech-
anomyographic
(MMG)
and
electromyographic
(EMG)
sig-
nals
measured
simultaneously
from
the
vastus
lateralis
(VL)
and
vastus
medialis
(VM)
muscles
1000
-
383
500
-
4
11#,
-14
VL
EMG
0
"''
(11.V)
-1000
t00
50
VL
MMG
0
(
mV
)
-50
-
-100
-
1000
-
500
-
VM
EMG
0
(µV)
500
-
1000
-
100
-
50
-
VM
MMG
0
(mV)
.50
-
tA•
(
f\T'V
N;
100
1
1.0
2.0
3.0
Time
(sec)
00
4.0
5.0
pleted
the
15-min
workbouts
at
50%
and
65%
Wpeak.
Only
three
subjects,
however,
completed
15-min
workbouts
at
80%
W
pea
k,
and
none
of
the
subjects
rode
for
the
full
15
min
at
95%
Wpeak.
The
mean
(SD)
times
to
exhaustion
at
80%
(n
=
5)
and
95%
(n
=
8)
W
peak
were
9.2
(2.2)
min
and
4.3
(1.4)
min,
respectively.
The
slope
coefficients
for
the
mean
normalized
MMG
amplitude/time
relationships
for
the
VL
at
50%
W
pea
k
(-0.24)
and
65%
W
peak
(-0.20)
were
significantly
less
than
zero.
The
slope
coefficients
at
80%
W
peak
(0.12)
and
95%
W
peak
(0.07),
however,
were
not
different
from
zero.
Figure
2
shows
the
mean
normalized
MMG
amplitude
(as
a
percentage
of
the
initial
amplitude
value)
versus
time
(as
a
percentage
of
the
total
workbout
duration)
relationships
for
the
VL
for
the
four
continuous,
constant-power-output
workbouts.
The
slope
coefficients
for
the
mean
normalized
MMG
amplitude/time
relationships
for
the
VM
at
50%
Wp
ea
k
(-0.24)
and
65%
W
peak
(-0.14)
were
significantly
less
than
zero.
The
slope
coefficient
at
80%
W
pea
k
(-0.06),
however,
was
not
different
from
zero,
while
at
95%
W
peak
,
the
slope
coefficient
(0.19)
was
significantly
greater
than
zero.
Figure
3
shows
the
mean
normalized
MMG
amplitude
(as
a
percentage
of
the
initial
ampli-
tude
value)
versus
time
(as
a
percentage
of
the
total
workbout
duration)
relationships
for
the
VM
for
the
four
continuous,
constant-power-output
workbouts.
The
slope
coefficients
for
the
mean
normalized
EMG
amplitude/time
relationships
for
the
VL
at
50%
(0.03),
65%
(0.08),
80%
(0.20),
and
95%
Wpeak
(0.47)
were
all
significantly
greater
than
zero.
Figure
4
shows
the
mean
normalized
EMG
amplitude
(as
a
percentage
of
the
140
MMG
VM
95%
Ppeak
(slope=
0.19)
Fig.
3
Mean
normalized
MMG
amplitude
(percentage
of
the
initial
amplitude
value)
ver-
sus
time
(percentage
of
the
total
workbout
duration)
relation-
ships
for
the
VM
for
the
four
continuous,
constant-power-
output
workbouts
given
in
Fig.
2
120
80%
Ppeak
(slope
-0.06)
Ppeak
(slope=
-0.14)
50%
Ppeak
(slope=
-0.24)
60
40
20
90
100
10
0
0
20
3b
,
fo
50
60
70
80
Normalized
time
(%
of
total
workout
duration)
384
Fig.
2
Mean
normalized
MMG
amplitude
(percentage
of
the
initial
amplitude
value)
ver-
sus
time
(percentage
of
the
total
workbout
duration)
relation-
ships
for
the
VL
for
the
four
continuous,
constant-power-
output
workbouts
(i.e.
50%,
65%,
80%
and
95%
peak
power,
Wpeak)
140
MMG
VL
120
E
-
3
80
7.
Ea
6
-d
0
,72,
40
Z
20
80%
Ppeak
lopc=0.12)
95%
Ppeak
(slope=0.07)
50%
Ppeak
(slope
-0.24)
65%
Ppeak
(slope=
-0.20)
0
0
10
20
30
40
50
60
70
80
90
100
Normalized
time
(%
of
total
workout
duration)
initial
amplitude
value)
versus
time
(as
a
percentage
of
the
total
workbout
duration)
relationships
for
the
VL
for
the
four
continuous,
constant-power-output
workbouts.
The
slope
coefficient
for
the
mean
normalized
EMG
amplitude/time
relationship
for
the
VM
at
50%
W
pea
k
(0.00)
was
not
different
from
zero,
while
the
slope
co-
efficients
at
65%
(0.05),
80%
(0.23),
and
95%
W
peak
(0.56)
were
all
significantly
greater
than
zero.
Figure
5
shows
the
mean
normalized
EMG
amplitude
(as
a
percentage
of
the
initial
amplitude
value)
versus
time
(as
a
percentage
of
the
total
workbout
duration)
relationships
for
the
VM
for
the
four
continuous,
constant-power-output
workbouts.
Discussion
Previous
studies
(Barry
et
al.
1985;
Goldenberg
et
al.
1991;
Keidel
and
Keidel
1989;
Orizio
1993;
Orizio
et
al.
1989b;
Stokes
1993;
Stokes
and
Dalton
1991;
Zwarts
and
Keidel
1991)
have
demonstrated
that
the
patterns
of
MMG
amplitude
responses
to
sustained
isometric
mus-
cle
actions
are
dependent
upon
the
level
of
force
pro-
duction
and
the
rate
at
which
fatigue
develops.
At
low
levels
of
force
production
(10-25%
maximum
voluntary
contraction,
MVC),
when
fatigue
develops
slowly,
MMG
amplitude
increases
with
time.
At
high
levels
of
isometric
force
(60-80%
MVC),
when
fatigue
develops
rapidly,
there
is
typically
a
reduction
in
MMG
ampli-
tude
with
time.
At
moderate
levels
of
isometric
force
production
(40-50%
MVC),
there
is
usually
no
change
in
MMG
amplitude
with
time.
Orizio
has
attributed
these
force-dependent
patterns
to
fatigue-induced
changes
in
motor-unit
recruitment
and/or
motor-unit
discharge
rates,
which
"may
determine
an
increase
or
a
decrease"
(Orizio
1993)
in
MMG
amplitude.
According
to
Sjogaard
(1978),
during
cycle
ergometry
at
100%
V0
2
.
and
a
pedal
cadence
of
60
rev
min
-1,
EMG
VL
160
140
.,
-7,-
,
120•
.a
.7Pt
100
7
E
SO
N
60•
E
,
,
;
„s?
:
40
20
0
0
95%
Ppeak
(slopc=(1.47)
80%
Ppeak
(slope=0.20)
65%
Ppeak
(slope-0.08)
50%
Ppeak
(slope=
0.03)
90
100
10
20
30
40
50
60
70
80
Normalized
time
(%
of
total
workout
duration)
EMG
VM
95%
Ppeak
(slope-0.56)
80%
Ppeak
(slope=0.23)
65%
Ppeak
(slope=0.05)
50%
Ppeak
(slope=0.00)
385
Fig.
4
Mean
normalized
EMG
amplitude
(percentage
of
the
initial
amplitude
value)
versus
time
(percentage
of
the
total
workbout
duration)
relation-
ships
for
the
VL
for
the
four
continuous,
constant-power-
output
workbouts
given
in
Fig.
2
Fig.
5
Mean
normalized
EMG
amplitude
(percentage
of
the
initial
amplitude
value)
versus
time
(percentage
of
the
total
workbout
duration)
relation-
ships
for
the
VM
for
the
four
continuous,
constant-power-
output
workbouts
given
in
Fig.
2
160
140
1-)
"120
L.7
80
"?.)
60
E
40
20
0
0
10
20
30
40
50
60
70
80
90
100
Normalized
time
(%
of
total
workout
dration)
the
force
exerted
against
the
pedals
represented
only
16%
of
the
isometric
MVC
and
40%
of
the
maximal
dynamic
strength.
These
findings
suggest
that
the
power
outputs
utilized
in
the
present
study
(50-95%
W
peak
)
were
characterized
by
low
levels
of
force
production
relative
to
maximal
capabilities
and,
therefore,
it
is
likely
that
at
the
beginning
of
the
workbouts
at
each
power
output
there
were
motor
units
available
that
had
not
been
recruited.
Like
previous
studies
of
isometric
muscle
actions
(Orizio
1993;
Stokes
1993),
the
results
of
the
present
investigation
indicate
that
the
patterns
of
the
MMG
amplitude
responses
to
the
continuous
cycle
ergometer
workbouts
were
dependent
upon
the
power
output
at
which
the
ride
was
performed.
Furthermore,
there
were
dissociations
between
the
patterns
of
the
MMG
and
EMG
amplitude
responses
at
the
various
power
outputs.
EMG
amplitude
increased
with
time
for
both
muscles
and
each
power
output
(except
the
VM
at
50%
Wpeak),
while
MMG
amplitude
increased,
decreased,
or
re-
mained
unchanged.
The
increases
in
EMG
amplitude
at
all
power
outputs
in
the
present
study
were
likely
to
be
due
to
peripheral,
low-frequency
fatigue
(Petrofsky
1979;
Stokes
1993).
Stokes
has
described
peripheral,
low-frequency
fatigue
as
a
situation
where
"...increased
activation
is
required
to
achieve
a
given
force
in
the
presence
of
excitation-
contraction
coupling
failure"
(Stokes
1993).
The
varia-
tions
in
the
patterns
of
the
MMG
responses
suggested,
however,
that
the
influence
of
peripheral,
low-frequency
fatigue
and
the
relative
importance
of
changes
in
motor-
unit
recruitment
and
discharge
rates
were
specific
to
the
power
output
at
which
the
ride
was
performed.
At
50%
and
65%
W
pea
k,
the
decreases
across
time
in
MMG
amplitude
may
be
explained
by
the
relationship
between
motor-unit
recruitment
and
the
effects
of
"muscle
wisdom",
as
described
by
Marsden
et
al.
(1983)
386
and
Enoka
and
Stuart
(1992).
Muscle
wisdom
provides
an
economical
activation
of
fatiguing
muscle
and
is
characterized
by
a
fatigue-induced
decrease
in
the
rate
of
muscle
relaxation,
elongation
of
the
muscle
twitch
re-
sponse,
a
decrease
in
motor-unit
discharge
rate,
and
an
increase
in
the
fusion
of
the
twitches
of
muscle
fibers
at
a
given
submaximal
motor-unit
discharge
rate
(Enoka
and
Stuart
1992;
Marsden
et
al.
1983).
Typically,
the
fatigue-
induced
recruitment
of
additional
motor
units
increases
MMG
amplitude,
while
the
effects
of
muscle
wisdom
decrease
the
oscillations
of
muscle
fibers
and
MMG
amplitude
(Esposito
et
al.
1998;
Goldenberg
et
al.
1991;
Orizio
1993;
Stokes
1993).
Recently,
Esposito
et
al.
(1998)
reported
a
decrease
in
MMG
amplitude
with
time
during
a
20-s,
sustained
isometric
muscle
action
at
80%
MVC.
It
was
hypothesized
that
the
effects
of
muscle
wisdom
resulted
in
a
reduction
in
the
number
of
pressure
waves
per
unit
of
time
and,
therefore,
a
decrease
in
MMG
amplitude.
Thus,
at
50%
and
65%
W
peak
,
the
present
findings
suggest
that
muscle
wisdom
had
a
greater
influence
on
the
MMG
signal
than
did
motor-
unit
recruitment
and,
therefore,
MMG
amplitude
de-
creased
with
time.
At
80%
W
peak
,
there
was
no
change
in
MMG
am-
plitude
with
time
for
either
the
VL
or
the
VM.
Orizio
(1993)
suggested
that
a
lack
of
change
in
MMG
ampli-
tude
throughout
a
sustained
muscle
action
reflects
a
balance
between
the
influences
of
fatigue-induced
changes
in
motor-unit
recruitment
and
discharge
rates.
In
this
regard,
the
concept
of
muscle
wisdom
suggests
that
at
80%
W
pea
k
there
was
a
balance
between
the
in-
fluences
of
recruitment
of
additional
motor
units
(which
can
increase
MMG
amplitude)
and
decreases
in
the
discharge
rates
of
already
activated
motor
units
(which
can
decrease
MMG
amplitude)
that
resulted
in
the
maintenance
of
MMG
amplitude
with
time.
At
95%
W
peak
,
MMG
amplitude
increased
for
the
VM
and
remained
unchanged
for
the
VL.
These
findings
were
consistent
with
those
of
a
recent
study
by
Ebersole
et
al.
(1999)
that
reported
differences
between
the
VL
and
VM
in
the
patterns
of
MMG
amplitude
responses
to
incremental
isometric
muscle
actions.
It
is
possible
that
the
differences
between
muscles
were
due
to
the
iliotibial
band
covering
the
VL,
which
may
have
affected
the
transmission
of
the
muscle
fiber
oscillations
to
the
sur-
face
of
the
skin
(Ebersole
et
al.
1999).
Orizio
has
hy-
pothesized
that
the
tissue
between
the
muscle
and
the
MMG
transducer
"...
may
act
as
a
low
pass
filter
for
the
mechanical
waves
traveling
from
the
muscle
to
the
skin
surface"
(Orizio
1993).
With
regard
to
the
VM,
the
results
of
the
present
study
suggest
that
at
95%
W
peak
,
the
recruitment
of
ad-
ditional
motor
units
had
a
greater
influence
on MMG
amplitude
than
did
decreases
in
motor-unit
discharge
rates.
At
this
high
power
output,
the
recruitment
of
additional
motor
units
would
likely
involve
large,
su-
perficially
located,
fast-twitch
muscle
fibers,
which
may
have
resulted
in
a
substantial
increase
in
the
amplitude
of
the
MMG
signal.
Thus,
it
is
possible
that
the
recruitment
of
these
motor
units
more
than
offset
the
attenuating
influence
of
decreases
in
the
discharge
rates
of
the
already
activated
motor
units
and,
therefore,
there
was
a
net
increase
in
MMG
amplitude
with
time.
In
addition
to
the
effects
of
motor
control
strategies,
it
is
possible
that
factors
related
to
muscular
compliance
affected
the
MMG
amplitude
responses
to
the
continu-
ous
cycle
ergometer
workbouts
that
were
performed
in
the
present
study
(Orizio
et
al.
1989a).
Previous
studies
(Bakke
et
al.
1996;
Jensen
et
al.
1994;
Sjogaard
and
Saltin
1982;
Sjogaard
et
al.
1988)
have
demonstrated
that
sustained
isometric
and
dynamic
muscle
actions
increase
muscle
thickness,
fluid
content,
and
intramus-
cular
pressure,
which
together
decrease
muscular
com-
pliance.
Decreases
in
muscular
compliance
can,
theoretically,
attenuate
the
amplitude
of
the
MMG
sig-
nal
(Orizio
1993;
Orizio
et
al.
1989a).
Barry
et
al.
(1985)
suggested,
however,
that
muscle
swelling
during
pro-
longed
exercise
enhances
the
transmission
of
muscle
fi-
ber
oscillations
to
the
surface
of
the
skin
and,
therefore,
increases
MMG
amplitude.
Future
research
is
needed
to
examine
the
influences
of
changes
in
muscle
thickness,
fluid
content,
intramuscular
pressure,
and
muscular
compliance
on
the
patterns
of
the
MMG
amplitude
re-
sponses
during
continuous
cycle
ergometry
at
various
power
outputs.
It
should
also
be
noted
in
Fig.
1,
that
there
was
MMG
activity
between
the
EMG
bursts.
This
portion
of
the
MMG
activity
could
not
be
directly
related
to
motor-
unit
activity
since,
as
the
EMG
signals
indicated,
it
oc-
curred
during
the
relaxation
(ascending)
phase
of
the
pedaling
cycle.
Thus,
it
is
likely
that
the
MMG
amplitude
values
recorded
in
the
present
study
were
influenced,
to
some
degree,
by
the
cycling
movement
itself.
Additional
studies
are
needed
to
determine
the
magnitude
of
the
influence
of
the
cycling
movement
on
MMG
amplitude
as
well
as
the
factors
involved
in
the
generation
of
the
MMG
signal
during
the
relaxation
phase
of
cycling.
In
summary,
this
investigation
is
the
first
to
examine
the
MMG
amplitude
responses
during
continuous
cycle
ergometer
workbouts
performed
at
constant
power
outputs.
The
results
indicate
that
while
EMG
amplitude
increased
with
time,
MMG
amplitude
increased,
de-
creased,
or
remained
unchanged,
depending
on
the
power
output.
It
is
possible
that
"muscle
wisdom",
which
involves
fatigue-induced
changes
in
motor
control
strategies,
may
explain
the
differences
between
power
outputs
in
the
patterns
of
MMG
amplitude
responses.
It
is
also
possible
that
changes
in
muscular
compliance
across
the
workbouts
influenced
the
MMG
signals.
References
Bakke
M,
Thomsen
CE,
Vilmann
A,
Soneda
K,
Farella
M,
Moller
E
(1996)
Ultrasonographic
assessment
of
the
swelling
of
the
human
masseter
muscle
after
static
and
dynamic
activity.
Arch
Oral
Biol
41:
133-140
Barry
DT
(1987)
Acoustic
signals
from
frog
skeletal
muscle.
Bio-
phys
J
51:
769-773
387
Barry
DT
(1991)
Muscle
sounds
from
evoked
twitches
in
the
hand.
Arch
Phys
Med
Rehabil
72:
573-575
Barry
DT,
Cole
NM
(1988)
Fluid
mechanics
of
muscle
vibrations.
Biophys
J
53:
899-905
Barry
DT,
Cole
NM
(1990)
Muscle
sounds
are
emitted
at
the
res-
onant
frequencies
of
skeletal
muscle.
IEEE
Trans
Biomed
Eng
37:
525-531
Barry
DT,
Geiringer
SR,
Ball
RD
(1985)
Acoustic
myography:
a
noninvasive
monitor
of
motor
unit
fatigue.
Muscle
Nerve
8:
189-194
Barry
DT,
Gordon
KE,
Hinton
GG
(1990)
Acoustic
and
surface
EMG
diagnosis
of
pediatric
muscle
disease.
Muscle
Nerve
13:
286-290
Bolton
CF,
Parkes
A,
Thompson
TR,
Clark
MR,
Sterne
CJ
(1989)
Recording
sound
from
human
skeletal
muscle:
technical
and
physiological
aspects.
Muscle
Nerve
12:
126-134
Cerquiglini
S,
Figura
F,
Marchetti
M,
Salleo
A
(1973)
Evaluation
of
athletic
fitness
in
weight-lifters
through
biomechanical,
bio-
electrical,
and
bioacoustical
data.
In:
Jokl
J
(ed)
Biomechanics
III,
vol
8.
Karger,
Basel,
pp
189-195
Ebersole
KT,
Housh
TJ,
Johnson
GO,
Evetovich
TK,
Smith
DB,
Perry
SR
(1999)
MMG
and
EMG
responses
of
the
superficial
quadriceps
femoris
muscles.
J
Electromyogr
Kinesiol
9:
219-
227
Enoka
RM,
Stuart
DG
(1992)
Neurobiology
of
muscle
fatigue.
J
Appl
Physiol
72:
1631-1648
Esposito
F,
Orizio
C,
Veicsteinas
A
(1998)
Electromyogram
and
mechanomyogram
changes
in
fresh
and
fatigued
muscle
during
sustained
contraction
in
men.
Eur
J
Appl
Physiol
78:
494-501
Frangioni
JV,
Kwan-Gett
TS,
Dobrunz
LE,
McMahon
TA
(1987)
The
mechanism
of
low-frequency
sound
production
in
muscle.
Biophys
J
51:
775-783
Goldenberg
MS,
Yack
HJ,
Cerny
FJ,
Burton
HW
(1991)
Acoustic
myography
as
an
indicator
of
force
during
sustained
contrac-
tions
of
a
small
hand
muscle.
J
Appl
Physiol
70:
87-91
Gordon
G,
Holbourn
AHS
(1948)
The
sounds
from
single
motor
units
in
a
contracting
muscle.
J
Physiol
(Lond)
107:
456-464
Jensen
BR,
Jorgensen
K,
Sjogaard
G
(1994)
The
effect
of
pro-
longed
isometric
contractions
on
muscle
fluid
balance.
Eur
J
Appl
Physiol
69:
439-444
Johnston
J, DiNardo
J
(1997)
Econometric
methods,
4th
edn.
McGraw-Hill,
New
York,
pp
179-182
Keidel
M,
Keidel
WD
(1989)
The
computer-vibromyography
as
a
biometric
progress
in
studying
muscle
function.
Biomed
Tech-
nik
34:
107-116
Leger
L,
Thivierge
M
(1988)
Heart
rate
monitors:
validity,
stability,
and
functionality.
Phys
Sports
Med
16:
143-151
Marchetti
M,
Salleo
A,
Figura
F,
Del
Gaudio
V
(1974)
Electr-
omyographic
and
phonomyographic
patterns
in
muscle
atrophy
in
man.
In:
Nelson
RC,
Morehouse
CA
(eds)
Biomechanics
IV,
vol
1.
University
Park
Press,
Baltimore,
pp
388-393
Marsden
CD,
Meadows
JC,
Merton
PA
(1983)
"Muscular
wis-
dom"
that
minimizes
fatigue
during
prolonged
effort
in
man:
peak
rates
of
motoneuron
discharge
and
slowing
of
discharge
during
fatigue.
In:
Desmedt
JE
(ed)
Motor
control
mechanisms
in
health
and
disease.
Raven,
New
York,
pp
169-211
Orizio
C
(1993)
Muscle
sound:
bases
for
the
introduction
of
a
mechanomyographic
signal
in
muscle
studies.
Crit
Rev
Biomed
Eng
21:
201-243
Orizio
C,
Veicsteinas
A
(1992)
Soundmyogram
analysis
during
sustained
maximal
voluntary
contraction
in
sprinters
and
long
distance
runners.
Int
J
Sports
Med
13:
594-599
Orizio
C,
Perini
R,
Veicsteinas
A
(1989a)
Muscular
sound
and
force
relationship
during
isometric
contraction
in
man.
Eur
J
Appl
Physiol
58:
528-533
Orizio
C,
Perini
R,
Veicsteinas
A
(1989b)
Changes
of
muscular
sound
during
sustained
isometric
contraction
up
to
exhaustion.
J
Appl
Physiol
66:
1593-1598
Oster
G,
Jaffe
JS
(1980)
Low
frequency
sounds
from
sustained
contraction
of
human
skeletal
muscle.
Biophys
J
30:
119-128
Petitjean
M,
Maton
B
(1995)
Phonomyogram
from
single
motor
units
during
voluntary
isometric
contraction.
Eur
J
Appl
Physiol
71:
215-222
Petrofsky
JS
(1979)
Frequency
and
amplitude
analysis
of
the
EMG
during
exercise
on
the
bicycle
ergometer.
Eur
J
Appl
Physiol
41:
1-15
Rodriquez
AA,
Agre
JC,
Knudtson
ER,
Franke
TM,
Ng
AV
(1993)
Acoustic
myography
compared
to
electromyography
during
isometric
fatigue
and
recovery.
Muscle
Nerve
16:
188-
192
Shinohara
M,
Kouzaki
M,
Yoshihisa
T,
Fukunaga
T
(1997)
Mechanomyography
of
the
human
quadriceps
muscle
during
incremental
cycle
ergometry.
Eur
J
Appl
Physiol
76:
314-319
Sjogaard
G
(1978)
Force-velocity
curve
for
bicycle
work.
In:
As-
mussen
E,
Jorgensen
K
(eds).
Biomechanics
VI-A.
University
Park
Press,
Baltimore,
pp
93-99
Sjogaard
G,
Saltin
B
(1982)
Extra-
and
intracellular
water
spaces
in
muscles
of
man
at
rest
and
with
dynamic
exercise.
Am
J
Physiol
243:
R271—R280
Sjogaard
G,
Savard
G,
Juel
C
(1988)
Muscle
blood
flow
during
isometric
activity
and
its
relation
to
muscle
fatigue.
Eur
J
Appl
Physiol
57:
327-335
Smith
DB,
Housh
TJ,
Johnson
GO,
Evetovich
TK,
Ebersole
KT,
Perry
SR
(1998)
Mechanomyographic
and
electromyographic
responses
to
eccentric
and
concentric
isokinetic
muscle
actions
of
the
biceps
brachii.
Muscle
Nerve
21:
1438-1444
Stokes
MJ
(1993)
Acoustic
myography:
applications
and
consid-
erations
in
measuring
muscle
performance.
Isokin
Exerc
Sci
3:
4-15
Stokes
MJ,
Dalton
PA
(1991)
Acoustic
myography
for
investigat-
ing
human
skeletal
muscle
fatigue.
J
Appl
Physiol
71:
1422-
1426
Stout
JR,
Housh
TJ,
Johnson
GO,
Evetovich
TK,
Smith
DB
(1997)
Mechanomyography
and
oxygen
consumption
during
incre-
mental
cycle
ergometry.
Eur
J
Appl
Physiol
76:
363-367
Zhang YT,
Frank
CB,
Rangayyan
RM,
Bell
GD
(1992)
A
com-
parative
study
of
simultaneous
vibromyography
and
electr-
omyography
with
active
human
quadriceps.
IEEE
Trans
Biomed
Eng
39:
1045-1052
Zwarts
MJ,
Keidel
M
(1991)
Relationship
between
electrical
and
vibratory
output
of
muscle
during
voluntary
contraction
and
fatigue.
Muscle
Nerve
14:
756-761