Eur J Appl Physiol (2004) 92: 399406
DOI 10.1007/s00421-004-1072-y
O R I GI N A L A R T IC L E
Daniel R. Moore Kirsten A. Burgomaster
Lee M. Schoeld Martin J. Gibala Digby G. Sale
Stuart M. Phillips
Neuromuscular adaptations in human muscle following
low intensity resistance training with vascular occlusion
Accepted: 26 January 2004 / Published online: 17 June 2004

Springer-Verlag 2004
Abstract Low-intensity (50% of a single repetition
maximum1 RM) resistance training combined with
vascular occlusion results in increases in muscle strength
and cross-sectional area [Takarada et al. (2002) Eur J
Appl Physiol 86:308331]. The mechanisms responsible
for this hypertrophy and strength gain remain elusive
and no study has assessed the contribution of neuro-
muscular adaptations to these strength gains. We
examined the eect of low-intensity training (8 weeks of
unilateral elbow exion at 50% 1 RM) both with (OCC)
and without vascular occlusion (CON) on neuromus-
cular changes in the elbow exors of eight previously
untrained men [19.5 (0.4) years]. Following training,
maximal voluntary dynamic strength increased
(P<0.05) in OCC (22%) and CON (23%); however,
isometric maximal voluntary contraction (MVC)
strength increased in OCC only (8.3%, P<0.05). Motor
unit activation, assessed by interpolated twitch, was high
(98%) in OCC and CON both pre- and post-training.
Evoked resting twitch torque decreased 21% in OCC
(P<0.05) but was not altered in CON. Training resulted
in a reduction in the twitch:MVC ratio in OCC only
(29%, P<0.01). Post-activation potentiation (PAP)
signicantly increased by 51% in OCC (P<0.05) and
was not changed in CON. We conclude that low--
intensity resistance training in combination with vas-
cular occlusion produces an adequate stimulus for
increasing muscle strength and causes changes in indices
of neuromuscular function, such as depressed resting
twitch torque and enhanced PAP.
D. R. Moore K. A. Burgomaster L. M. Schoeld
M. J. Gibala D. G. Sale S. M. Phillips (&)
Department of Kinesiology,
Exercise and Metabolism Research Group,
McMaster University, 1280 Main St. W., Hamilton,
Ontario, L8S 4K1, Canada
E-mail: phillis@mcmaster.ca
Tel.: +1-905-5259140
Fax: +1-905-5234025
Keywords Twitch contraction Post-activation
potentiation Motor unit activation
Introduction
It is well established that muscle bres respond to
overload stresses such as resistance training with both an
increase in the cross-sectional area (CSA) of bres
(hypertrophy) and an increase in force generating
capacity. During resistance exercise, motor units (MU),
and hence muscle bres, are recruited according to the
size principle (Henneman et al. 1965). This principle
states that MU with smaller cell bodies, which are
associated with type I muscle bres, are activated rst
and at lower exercise intensities. At higher exercise
intensities, larger MU and their associated type II
muscle bres are recruited. For hypertrophy to be
maximized, it is important to activate type II muscle
bres during training since these bres respond to a
resistive exercise stress with greater hypertrophy than
type I bres (MacDougall et al. 1980; McCall et al.
1996). Since motor unit activation (MUA), and there-
fore bre recruitment, is linearly related to force pro-
duction (Sale 2003), many exercise programs aimed at
increasing muscle size and strength perform repetitions
at a target intensity of at least 7095% of an individuals
one repetition maximum (1 RM).
Muscle fatigue also plays a modulatory role in MUA
and bre recruitment. Muscular activity that results in
the onset of muscular fatigue causes a decrease in muscle
force production, which can be compensated for by an
increase in MUA in an eort to maintain force output
(Sale 1987). Hence, with fatigue there is an increased
MUA (Loscher et al. 1996; Suzuki et al. 2002) and,
according to the size principle (Henneman et al. 1965),
enhanced activation of the higher threshold MUs that
innervate type II bres.
Muscular fatigue may be hastened during low inten-
sity muscular contractions (i.e., less than 50% of 1 RM)
400
by occluding arterial blood ow (Moritani et al. 1992).
Training with reduced muscle blood ow has been
shown to result in an increase in the integrated electro-
myogram (EMG) of the active muscle, indicating that a
greater number of what would normally be inactive
muscle bres are required to lift relatively light loads
(Moritani et al. 1992; Takarada et al. 2000). Low
intensity occluded resistance training (LIORT) has been
shown to be eective at increasing muscle strength and
inducing hypertrophy in trained (Takarada et al. 2002)
and untrained (Shinohara et al. 1998; Takarada et al.
2000) individuals, despite the utilization of relatively low
(4050% 1 RM) exercise intensities. Enhanced muscle
strength following LIORT is likely partially mediated by
an increase in muscle CSA (Takarada et al. 2000, 2002).
Therefore, it seems that training with a large external
load (i.e., 80% of 1 RM or greater), which can lead to
joint and ligament damage, is not necessary in order to
obtain signicant muscle strength gains and/or hyper-
trophy.
It has already been established that resistance train-
ing with vascular occlusion can lead to increases in
strength (Shinohara et al. 1998; Takarada et al. 2000,
2002); furthermore, at least part of the increased
strength can be attributed to muscle hypertrophy
(Takarada et al. 2000, 2002). However, increased
strength could also result from neural adaptations, such
as increased agonist muscle activation (Sale 2003)
through increased motoneuron ring frequency and/or
motoneuron excitability (Aagaard et al. 2002). Muscular
adaptations in addition to hypertrophy, such as altered
twitch contractile properties, might also occur. Post-
activation potentiation (PAP) refers to a phenomenon
whereby a conditioning stimulus, such as a 10-s maximal
voluntary contraction (MVC), increases the muscles
responsiveness to calcium, which results in a greater
twitch torque and shorter twitch duration. PAP has been
shown to be greater in type II bres (Hamada et al.
2000b), is enhanced in endurance athletes (Hamada et al.
2000a) and unaltered after resistance training (Rice et al.
1993). At present, no information exists on how LIORT
aects neuromuscular adaptations; therefore, the pur-
pose of the present study was to examine the eect of
resistance training with vascular occlusion on neuro-
muscular adaptations, including muscle activation and
twitch contractile properties, as well as post-activation
potentiation (PAP).
Methods
Subjects
Eight healthy males [age, 19.5 (0.4) years; height, 180.0
(1.0) cm; weight, 84.0 (4.5) kg; means (SE)] with no
previous weight training experience volunteered for the
study. All subjects were informed about the experimental
procedure to be used as well as the purpose of the study
and all potential risks. Written consent was obtained
from the subjects. The study conformed to all standards
for the use of human subjects in research as outlined in
the Helsinki declaration and was approved by the local
Research Ethics Board of McMaster University and
Hamilton Health Sciences.
General design
Subjects trained the elbow exors of both arms three
times per week for a duration of 8 weeks at an intensity
of 50% of their previously determined 1 RM. One arm
was randomly assigned to perform the training program
with reduced muscle blood ow (OCC), which was ob-
tained through the application of a blood pressure cu
around the upper arm, while the other was not occluded
and hence served as a trained, non-occluded control
(CON). Isometric strength tests were performed both
pre- and post-training and the subjects voluntary dy-
namic 1 RM was measured bi-weekly to identify chan-
ges in muscle strength and to increase the resistive
stimulus, if necessary, to maintain 50% of 1 RM. Neu-
romuscular adaptations, including MUA and evoked
twitch characteristics, were also assessed pre- and post-
training. All subjects received a familiarization session at
least 1 week prior to pre-testing in order to become well
acquainted with the testing apparatus and protocol.
Testing apparatus
Evoked twitch characteristics, maximal voluntary iso-
metric force production (MVC), MUA, and degree of
PAP were measured while the subjects sat in an adjust-
able chair with their upper arm resting comfortably in
the horizontal plane on an aluminum plate of a custom-
made apparatus. The subjects forearm was strapped
with their hand in the supinated position to a second
aluminum plate with Velcro straps. The apparatus
maintained the elbow at an angle of 120, which was
seen to be the optimum angle for isometric torque gen-
eration in untrained men (Tsunoda et al. 1993). Seat
height was adjusted to ensure that the upper arm was at
a 90 angle to the chest and the elbow was at the pivot
point of the apparatus. Shoulder movement was limited
by a second set of Velcro straps to ensure isolation of
torque development by the elbow exors.
Resting twitch
Resting twitch properties were assessed through percu-
taneous electrical stimulation by lead electrodes prior to
strength measurements to ensure results were not con-
taminated by PAP. Two lead electrodes (14 cm,
24 cm), which were wrapped with gauze and moistened
with conducting gel, were placed over the motor point of
the biceps and on the ventral surface of the forearm just
distal to the elbow, respectively. Stimuli were rectangular

voltage pulses of 100 ls duration, produced by a Grass
S11 stimulator (Grass Instruments, Quincy, Mass.,
USA). Signals from the custom design OM2160 torque
transducer (Omega Engineering, Stamford, Conn.,
USA), which was mounted in the shaft of the apparatus
described above, were amplied and ltered at 200 Hz
with a lowpass 2nd order Butterworth lter before being
digitized by ACODAS hardware (Dataq Instruments,
Akron, Ohio, USA). Torque signals were sampled at
2 kHz and were stored, analyzed, and digitized by a
Compaq 46 PC computer system and displayed on screen
in real time. Maximal twitch contractions were evoked in
the resting muscle by progressively increasing the stim-
ulation intensity until increases failed to increase twitch
torque further. The stimulus intensity that elicited
peak twitch torque was used throughout the duration of
the tests. Automated twitch analysis was done with
custom designed software. Twitch properties examined
were peak torque (PT), torque rise time (RT, time
from 10% to 90% PT), half-relaxation time (HRT), peak
rate of torque development (RTD), and peak rate of
torque relaxation (RTR). Peak rates of twitch torque
development and torque relaxation were measured as the
greatest change in torque over a 3-ms period on the
rising and falling phases of the twitch contraction,
respectively.
MVC, MUA, and PAP
MVC, MUA, and PAP were measured through a
modied interpolated twitch protocol. Since the sub-
jects resting twitch characteristics were assessed prior to
the interpolated twitch protocol, but after having been
strapped into the testing apparatus (see above), subjects
did not warm up the elbow exors before the MVC to were used since they were both preceded by an eccentric
ensure that the resting twitch was not contaminated by
PAP and to maintain the placement of the stimulating
electrodes. The protocol began with an electrical stim-
ulation at a predetermined voltage (see above) to
establish a baseline twitch value. After 5 s, subjects, who
performed the protocol on at least one other day to
practice the task and become familiar with the testing
equipment, were instructed to perform a single 10-s
isometric MVC accompanied by verbal encouragement.
The subjects MVC was dened as the maximal torque
produced at any point throughout the 10-s contraction.
Halfway through the MVC, at the 5-s mark, a second
stimulus was administered to elicit an interpolated
twitch (IT). Five seconds after the cessation of the 10-s
MVC, a resting twitch was again evoked to assess the
extent of PAP. To examine the time course of the PAP,
twitches were evoked every 30 s for 3 min following the
MVC. Thus, the 10-s MVC served a three-fold function.
First, it measured the subjects maximal voluntary iso-
metric strength. Second, it acted as a conditioning
stimulus to elicit PAP. Finally, the 10-s MVC allowed
determination of the extent of MUA, as dened by the
equation: MUA=(1 IT/PAP)100%, (where IT is
401
the^interpolated twitch torque and PAP is the rst
potentiated twitch torque after the MVC). PAP twitch
characteristics were analyzed according to the previously
described technique and MVC values were manually
analyzed using a cursor-based waveform scrolling tech-
nique.
EMG testing
In order to examine the extent of MUA during a
training session, EMG data were collected in the nal
training session for sets 1 and 3 of the second training
block (described below). Subjects skin overlying the
biceps and medial epicondyle was cleaned with dispos-
able alcohol wipes prior to the application of 1 cm
diameter disposable Kendall Meditrace 130 electrodes
(The Ludlow Company, Chicopee, Mass., USA). Two
electrodes were placed approximately 1 cm apart over
the belly of the biceps and one ground electrode was
applied to the medial epicondyle of the arm. Raw EMG
was sampled at 3 kHz using a custom made bioamplier
with a highpass lter with a cuto frequency of 7 Hz and
a lowpass lter with a cuto frequency of 1,000 Hz.
Data were collected with acquisition software from
Dataq (Akron, Ohio), digitized, and stored on an IBM
compatible computer. All EMG signals were then full-
wave rectied and integrated over the interval of the
entire concentric contraction phase to determine mean
EMG. Absolute EMG output for the second repetition
of set 1 was compared to the nal repetition to examine
the rise in EMG, and therefore MUA, throughout the
set. The rise in EMG activity throughout set 1 was also
expressed as the percentage change from repetition 2.
The second and nal repetitions (i.e., concentric phase)
contraction and could therefore be easily compared. The
rst repetition of both sets 1 and 3 were compared to
examine the level of EMG output required to lift the
50% 1 RM load, which could be used to infer the extent
of fatigue, and were also expressed as percentage change
from set 1.
Training protocol
Each subject was required to complete an 8-week pro-
gressive resistance training program at an intensity of
50% of their previously, and repeatedly, determined
1 RM [described previously by (Burgomaster et al.
2003)]. Briey, single arm curls using a weighted cable
pulley attached to a handle were used to train the elbow
exors. Training positions were standardized using a
seated preacher bench. Subjects were instructed to
remain in an upright seated position with their back
straight and both feet at on the oor. The seat height
was adjusted so that the upper arm was perpendicular to
the chest. The upper arm rested comfortably on the
preacher bench pad and the shoulders remained square
402

to the cable pulley. Subjects contracted their elbow
exors and moved the weight in the sagittal plane in an
eort to maximally isolate the elbow exor muscle group
and limit extraneous movements. The training weight of
each arm was selected as 50% 1 RM, measured by a
three-trial average on the same apparatus, and was ad-
justed according to bi-weekly maximal strength assess-
ments. Subjects served as their own controls as their
arms were randomly assigned to either the OCC or CON
condition. Blood ow occlusion was applied with a
pneumatic blood pressure cu (width, 7 cm; length,
53 cm) that was placed on the upper arm just proximal
to the belly of the biceps. The cu was inated to
100 mmHg and was maintained at this pressure
throughout the duration of the training block, as dened
by the training protocol (details given in Table 1). A cu
pressure of 100 mmHg was selected for the occlusive
stimulus as this pressure has been suggested to restrict
venous blood ow and cause pooling of blood in
capacitance vessels distal to the cu, ultimately
restricting arterial blood ow (Takarada et al. 2000).
Each training block consisted of a dened number of
sets separated by a 1-min rest period, and a 5-min rest
period was allowed between training blocks, in which
the cu was removed in OCC. The number of repetitions
for the non-occluded arm was matched to that of the
occluded arm to ensure the same relative amount of
work was performed by each arm. Each repetition
consisted of a concentric (muscle shortening) contrac-
tion (elbow joint angle that progressed from 180 to 45,
where 180 is full extension) and an eccentric (muscle
lengthening) contraction (return to starting position).
Subjects were encouraged to maintain a repetition ca-
dence of a 2-s concentric phase and a 2-s eccentric phase
according to a metronome. Verbal encouragement was
provided to ensure maximum eort.
for dependent samples. For all analyses, when signicant
interactions were found, the Tukey post hoc test was
used to test dierences among mean values. Signicance
was accepted at P 0.05. All data are presented as
means (SE).
Results
Training session
During the rst training session, 10.0 (0), 5.8 (0.4), and
3.0 (0.6) repetitions were performed in sets 1, 2 and 3,
respectively. During the last training session, 10 (0), 7.5
(0.9), and 5.9 (1.6) repetitions in training block 1 and 9.4
(0.5), 5.5 (0.8), and 3.6 (1.0) repetitions in training block
2 were performed in sets 1, 2, and 3, respectively.
Strength
The MVC was similar between conditions at baseline
[OCC, 61.6 (5.2) Nm; CON, 64.3 (4.5) Nm] and sig-
nicantly increased by 10% in the OCC condition only
(interaction, P<0.05; Fig. 1). Maximal voluntary dy-
namic 1 RM was not dierent between limbs at baseline
[OCC, 42.4 (1.9) kg; CON, 43.2 (2.0) kg] and increased
by 22 and 23% for the OCC and CON conditions,
respectively (main eect for time, P<0.05), with no
dierences between conditions (Fig. 2).
Motor unit activation
MUA during the MVC was high in both arms pre-
training [OCC= 97.6 (1.0)%; CON= 98.0 (0.4)%] and
was not altered following training.

Statistics
Twitch characteristics
MVC, MUA and resting twitch data were analyzed with
a two (within) factor ANOVA (training condition, OCC Resting twitch PT was similar between arms at baseline
vs CON; time,pre- vs post-training). Post-MVC (PAP) and decreased 21% in the OCC arm only following
data were analyzed with a three (within) factor ANOVA
[training condition, time, and timing (time post-MVC)].
The absolute EMG data recorded during the nal
training session were analyzed with a two (within) factor
ANOVA (training condition, repetitions). Data for rel-
ative increase in EMG were analyzed with a paired t-test

Table 1 Training program. Training expressed as number of sets

target number of repetitions per set. F Set to failure

Week Block 1 Rest interval Block 2

1 210, 1F N/A N/A

2 210, 1F 5 min 110 Fig. 1 Maximal voluntary isometric torque (mvc) of the elbow
3 210, 1F 5 min 210 exors in the occluded (OCC) and non-occluded (CON) conditions,
48 210, 1F 5 min 210, 1F both pre- and post-training. Values are means (SE). *P<0.05,
dierent from pre-training and CON
 403

Fig. 2 Percentage increase from pre-training in isotonic (1 RM)

strength for both OCC and CON. *P<0.05, dierent from pre-
training;
P<0.05, dierent from pre-training and week 2; 
P<0.05, dierent from pre-training and weeks 2 and 4. There was
no dierence between groups

training (interaction, P<0.05; Table 2; Fig. 3a).

Twitch:MVC ratio was similar between limbs at baseline
[OCC=0.17 (0.02); CON=0.16 (0.03)] and decreased
29% in the OCC condition only (interaction, P<0.01;
Fig. 4). Training did not signicantly (P>0.05) alter
RT, HRT, RTD, or RTR in either the OCC or CON
arms (Table 2).
Post-activation potentiation
When compared to the baseline twitch, the degree of
PAP evident in the twitch torque was similar at baseline
between arms and was increased by 52% following
training only in the OCC condition (interaction,
P<0.05; Table 3; Fig. 3b), however, the absolute PAP
twitch torque was similar between the conditions both
pre- and post-training (Fig. 3a). Training did not sig-
nicantly (P>0.05) alter RT, HRT, RTD, or RTR in
the rst PAP twitch (Table 3) or in any of the later PAP
twitches (data not shown) in either the OCC or CON
conditions.
EMG activity
In training block 2 of the nal training session, there was
a main eect for time (P<0.05) but no signicant
Table 2 Involuntary resting twitch characteristics in OCC and
CON conditions
Fig. 3 A Peak resting (Rest) and post-activation potentiation
(PAP) absolute twitch torque in OCC and CON conditions, both
pre- and post-training. *P<0.05, dierent from pre-training and
CON. B Percentage change in twitch torque, versus baseline, of the
rst PAP twitch. *P<0.05, dierent from pre-training and CON
(P>0.05) dierences in absolute EMG activity between
the OCC and CON conditions (Table 4); however OCC
experienced a signicantly (P<0.05) greater relative
increase from set 1 to set 3 in the EMG of the rst
repetition (Fig. 5).
Discussion
The present study demonstrated that training at an
intensity of 50% 1 RM is eective at increasing volun-
tary dynamic elbow exor strength independent of the
application of a blood ow occlusion stimulus. Our
results are similar to that of Shinohara and colleagues
(1998) who demonstrated that 4 weeks of isometric
training at 40% MVC with tourniquet ischemia signi-
cantly improved isometric strength, while training at the
same intensity without occlusion yielded strength
increases that approached statistical signicance
(P=0.056). Movement pattern specicity (i.e., similarity

Parametera OCC CON

Pre Post Pre Post

PT (Nm) 10.1 (2.1) 8.0 (1.3)* 9.8 (1.8) 9.8 (1.5)

RT (ms) 36.6 (2.4) 32.6 (6.9) 34.8 (5.7) 35.6 (6.4)
HRT (ms) 73.9 (8.9) 75.0 (23.2) 83.1 (19.8) 74.6 (24.0)
RTD (Nms1) 287.4 (16.6) 236.3 (17.8) 290.8 (17.6) 273.5 (9.6)
RTR (Nms1) 116.2 (19.6) 104.6 (17.3) 101.6 (8.3) 132.6 (21.5)

*Signicantly dierent from pre-training and CON (P<0.05)

a
OCC With vascular occlusion, CON without vascular occlusion, Fig. 4 Twitch:maximum voluntary contraction (MVC) ratio in
PT peak torque,RT rise time, HRT half-relaxation time, RTD peak OCC and CON conditions, both pre- and post-training. *P<0.05,
rate of torque development, RTR peak rate of torque relaxation dierent from pre-training and CON
404

Table 3 Involuntary PAP twitch characteristics in OCC and CON

conditions immediately following a 10-s MVC

Parameter OCC CON

Pre Post Pre Post

PT (Nm) 16.3 (0.7) 15.7 (0.7) 16.6 (0.7) 16.8 (0.9)

IPT (%) 67.2 (13.7) 101.3 (15.7)* 72.9 (11.8)
RT (ms) 36.6 (0.7) 33.1 (2.4) 36.8 (0.9)
HRT (ms) 55.9 (6.1) 52.6 (6.8) 58.0 (6.2)
RTD (Nms1) 472.5 (26.9) 454.5 (40.8) 467.6 (23.7)
RTR (Nms1) 264.4 (57.7) 261.8 (44.9) 255.0 (30.4)
*Signicantly dierent from pre-training and CON (P<0.05)
a
PAPPost-activation potentiation, MVC maximal voluntary con-
traction, IPT increase in peak torque from rest
between training and testing modes) can have a large
inuence on an individuals ability to produce force
(Thorstensson et al. 1976). It has been suggested that
resistance training may alter force production through
task-specic MU activation within a muscle and among
synergistic muscles, eectively improving muscle coor-
dination to facilitate strength gains (Sale 1987, 2003).
Similarly, resistance training and the practice it aords
may reduce the amount of antagonist co-contraction,
allowing a greater net force production in the practiced
maneuver (Sale 2003). Therefore the increase in volun-
tary dynamic strength in both OCC and CON in the
present study may have primarily been the result of
neuromuscular adaptations such as enhanced task-spe-
cic MU activation and/or improved muscle coordina-
tion.
LIORT was eective at increasing strength in an
unpracticed situation, as evidenced by the signicant
increase in isometric MVC strength in the OCC condi-
tion only. Our results are in agreement with the work of
Takarada and colleagues (Takarada et al. 2000) who
demonstrated that dynamic resistance training at 50%
1 RM with an occlusion pressure of 110 mmHg resulted
in an 18.4% increase, when collapsed across all tested
isokinetic contraction velocities, in voluntary elbow
exor strength, which was signicantly greater than
training at the same intensity in the absence of the
occlusive stimulus (1.0%). Moreover, the strength in-
crease as a result of low-intensity occlusion training was
comparable to the increase seen while training without
occlusion at 80% of 1 RM (22.6%) (Takarada et al.
2000), an intensity of contraction more commonly used
to stimulate hypertrophy and maximize strength gains.
Table 4 EMG activity in OCC and CON conditions during train-
ing block 2 of the nal training session
OCC CON
Set 1 Set 3 Set 1 Set 3
Repetition 1 50122 785222 59827 66533
EMG (mV)
Repetition 2 Repetition 10 Repetition 2 Repetition 10
Set 1 686.552.7 961.068.3 751.150.4 878.4214.0
EMG (mV)
74.5 (11.2)
39.3 (1.6)
60.2 (4.0)
438.4 (6.1)
244.8 (25.9) Fig. 5 Percentage increase in EMG during block 2 of the nal
training session of the rst repetition of set 3, compared to the rst
repetition of set 1 (set 1 set 3) and the 10th repetition of set 1
compared to the 2nd repetition of set 1 (rep 2 rep 10) of the elbow
exors. EMG data were collected during the rst training block of
the last training session. *P<0.05, dierent from CON
More importantly, our 8.3% increase in isometric
strength was similar to the previously reported 13%
increase in isometric strength, despite our training
duration being half of that used by Takarada et al.
(2000). Therefore, our results demonstrate that training
with a relatively light load with vascular occlusion can
signicantly increase muscle strength in a novel (i.e.,
unpracticed) force generating maneuver.
In the present study, the MUA of the elbow exors
(measured during the isometric MVC) was high (98%)
in both the OCC and CON groups prior to the onset of
training, which has been shown previously (Allen et al.
1995), and was not altered following training. Strength
gains independent of increases in MUA suggests mus-
cular adaptations, such as bre hypertrophy, may have
been responsible for the moderate strength gains seen in
the present study. However, if muscle hypertrophy was
responsible for the observed increase in isometric
strength, it is unclear why dynamic maximal strength
was not similarly aected. It should also be noted that
the interpolated twitch technique is unable to detect
more specic neuromuscular changes, such as a decrease
in antagonist co-activation or an increase in muscle
coordination, which have also been suggested to
inuence strength gains (Carolan and Cafarelli 1992;
Rutherford and Jones 1986).
The presence of the blood pressure cu in the OCC
condition would likely accelerate the onset of fatigue
and reduce muscle force production, necessitating an
increase in bre recruitment in order to move the 50%
1 RM load (Sale 1987). While absolute EMG activity
during the nal training session increased in both the
OCC and CON conditions, it was slightly exaggerated in
OCC, as demonstrated through a trend for a greater rise
during the rst training set (P=0.076). This suggests
that the OCC condition required a greater amount of
muscle activity than CON to lift the 50% 1 RM load. In
addition, while the absolute EMG for the rst repetition
of sets 1 and 3 showed a trend towards being greater in
OCC (P=0.085), the relative increase in the EMG
during the rst repetition was signicantly greater in
OCC compared to CON. Greater EMG activity during
the rst repetition of the OCC condition would suggest

that the blood pressure cu interfered with the clearance
of potentially fatiguing metabolites (eg., H+ ions) from
the active muscle during the rest interval, necessitating a
greater bre recruitment to lift the 50% 1 RM load.
Alternatively, accelerated muscle fatigue in the OCC
condition may have increased the degree of MU syn-
chronization leading to an increase in the overall EMG
interference amplitude (Yao et al. 2000). Once inte-
grated, this enhanced EMG interference amplitude
could then lead to a net increase in the EMG signal
independent of any appreciable rise in MUA. However,
Suzuki and colleagues (2002) reported that the increase
in surface EMG during sustained voluntary submaximal
contractions corresponds to an increase in the number of
active MUs. Furthermore, a sustained contraction at
50% of maximal force is associated with an increased
EMG activity and a decreased interpolated twitch tor-
que (Loscher et al. 1996), indicating that an increase in
MUA and a decrease in the number of inactive bres has
occurred. Therefore, the present results of greater EMG
during training with blood ow occlusion would suggest
that a greater proportion of muscle bres (likely type II
bres), were active in the OCC than the CON condition.
In the present study, resting isometric twitch torque
decreased in the OCC arm but was unaltered in the
CON arm following training. Previously, resistance
training has resulted in an increased (Rich and Cafarelli
2000), decreased (Sale et al. 1982), or unaltered (Alway
et al. 1989; Kitai and Sale 1989; Rice et al. 1993) resting
twitch torque. Decreased resting twitch tension has
previously been suggested to reect an increased exten-
sibility of the muscle-tendon complex (Sale et al. 1982);
however this cannot explain the results of the present
investigation since participants trained both arms at the
same relative intensity. While a depressed twitch torque
in OCC may be a reection of low-frequency fatigue in
response to the training regimen, our data does not al-
low us to investigate this possibility fully. Since we are
the rst to examine the eects of LIORT on resting
twitch properties, our nding of a depressed resting
twitch torque may be unique to the occlusive stimulus.
Therefore, further studies are needed to determine the
underlying mechanism for the decrease in twitch tension
following LIORT seen in the present investigation.
In examining the extent of PAP following training, it
was seen that the OCC condition demonstrated a sig-
nicantly greater PAP whereas the CON arm did not.
This increased PAP was the result of a lower resting
twitch torque in the OCC arm that was restored to the
pre-training PAP torque. It appears that a brief condi-
tioning stimulus, in this case a 10-s MVC, produces an
adequate level of PAP to counteract whatever is
responsible for the depressed resting twitch in the OCC
arm. A greater relative increase in PAP following
LIORT would be considered a positive outcome, since
many daily activities are submaximal in nature and are
performed with potentiated muscles. Since potentiated
muscles can generate greater force during submaximal
activities, this means fewer bres need be active in order
405
to perform a given task, conserving the energy of the
inactive bres and reducing situations in which full
MUA would be required. Furthermore, if the MUA for
a specic task is reduced, this would mean a greater
proportion of active bres will be the energy ecient
type I bres (in accordance with the size principle of
MU recruitment), which will have a further energy
conserving eect.
LIORT can be a valuable alternative to enhancing
muscle strength in the absence of a large external load.
Populations such as the elderly or post-surgery rehabil-
itation patients, who may have compromised strength
and/or joint stability, would benet from a low intensity
resistance training program. Exercise programs that
utilize a relatively light training load (i.e., equal to or less
than 50% of maximal strength) in combination with
vascular occlusion would reduce joint articular and lig-
ament stress forces, which could reduce the incidence of
injury, while still provide an adequate stimulus to induce
increases in muscle strength and muscle hypertrophy
(Shinohara et al. 1998; Takarada et al. 2000, 2002).
While LIORT may not be suitable for all elderly indi-
viduals, owing to the potential vascular problems that
can be associated with this population, it has previously
been shown to be eective in increasing muscle strength
and size in aged women (Takarada et al. 2000) and could
be considered a viable intervention in the future.
In summary, resistance training at an intensity of
50% 1 RM resulted in signicant increases in dynamic
voluntary strength, which may have primarily occurred
through neuromuscular adaptations such as improved
muscle coordination. In contrast, only training with
vascular occlusion at 50% 1 RM elicited signicant
gains in isometric strength without any alterations in
maximal MUA. Training with vascular occlusion sig-
nicantly depressed resting twitch torque, the negative
eects of which are abolished following a stimulus that
induces PAP. Therefore, we conclude that low-intensity
resistance training in combination with vascular occlu-
sion produces an adequate stimulus for increasing
muscle strength independent of a large external load,
and causes changes in indices of neuromuscular func-
tion, such as a depressed resting twitch torque and en-
hanced PAP.
Acknowledgements This study was supported by the Natural Sci-
ence and Engineering Research Council (NSERC) of Canada
both SMP and MJG. Daniel Moore is the recipient of an NSERC
PGS-A scholarship. Thanks to the subjects for their time and ef-
fort. Thanks as well to Mr. John Moroz for his expert technical
assistance. SMP is the recipient of a Premiers Research Excellence
Award and a CIHR New Investigator Award and acknowledges
these sources of funding in the completion of this work.
References
Aagaard P, Simonsen EB, Andersen JL, Magnusson P, Dyhre-
Poulsen P (2002) Neural adaptation to resistance training:
changes in evoked V-wave and H-reex responses. J Appl
Physiol 92:23092318
406
Allen GM, Gandevia SC, McKenzie DK (1995) Reliability of
measurements of muscle strength and voluntary activation
using twitch interpolation. Muscle Nerve 18:593600
Alway SE, MacDougall JD, Sale DG (1989) Contractile adapta-
tions in the human triceps surae after isometric exercise. J Appl
Physiol 66:27252732
Burgomaster KA, Moore DR, Schoeld LM, Phillips SM,
Sale DG, Gibala MJ (2003) Resistance training with vascular
occlusion: metabolic adaptations in human muscle. Med Sci
Sports Exerc 35:12031208
Carolan B, Cafarelli E (1992) Adaptations in coactivation after
isometric resistance training. J Appl Physiol 73:911917
Hamada T, Sale DG, MacDougall JD (2000a) Postactivation
potentiation in endurance-trained male athletes. Med Sci Sports
Exerc 32:403411
Hamada T, Sale DG, MacDougall JD, Tarnopolsky MA (2000b)
Postactivation potentiation, ber type, and twitch contraction
time in human knee extensor muscles. J Appl Physiol
88:21312137
Henneman E, Somjen G, Carpenter DO (1965) Functional signif-
icance of cell size in spinal motoneurons. J Neurophysiol
28:560580
Kitai TA, Sale DG (1989) Specicity of joint angle in isometric
training. Eur J Appl Physiol 58:744748
Loscher WN, Cresswell AG, Thorstensson A (1996) Central fatigue
during a long-lasting submaximal contraction of the triceps
surae. Exp Brain Res 108:305314
MacDougall JD, Elder GC, Sale DG, Moroz JR, Sutton JR
(1980) Eects of strength training and immobilization on
human muscle bres. Eur J Appl Physiol 43:2534
McCall GE, Byrnes WC, Dickinson A, Pattany PM, Fleck SJ
(1996) Muscle ber hypertrophy, hyperplasia, and capillary
density in college men after resistance training. J Appl Physiol
81:20042012
Moritani T, Sherman WM, Shibata M, Matsumoto T, Shinohara
M (1992) Oxygen availability and motor unit activity in hu-
mans. Eur J Appl Physiol 64:552556
Rice CL, Cunningham DA, Paterson DH, Dickinson JR (1993)
Strength training alters contractile properties of the triceps bra-
chii in men aged 6578 years. Eur J Appl Physiol 66:275280
Rich C, Cafarelli E (2000) Submaximal motor unit ring rates after 8
wk of isometric resistance training. Med Sci Sports Exerc 32:190
196
Rutherford OM, Jones DA (1986) The role of learning and coor-
dination in strength training. Eur J Appl Physiol 55:100105
Sale DG, McComas AJ, MacDougall JD, Upton AR (1982) Neu-
romuscular adaptation in human thenar muscles following
strength training and immobilization. J Appl Physiol 53:419424
Sale DG (1987) Inuence of exercise and training on motor unit
activation. Exerc Sport Sci Rev 15:95151
Sale DG (2003) Neural adaptation to strength training. In Komi PV
(ed) Strength and power in sport. Blackwell, Oxford, pp 281314
Shinohara M, Kouzaki M, Yoshihisa T, Fukunaga T (1998) Ecacy
of tourniquet ischemia for strength training with low resistance.
Eur J Appl Physiol 77:189191
Suzuki H, Conwit RA, Stashuk D, Santarsiero L, Metter EJ (2002)
Relationships between surface-detected EMG signals and mo-
tor unit activation. Med Sci Sports Exerc 34:15091517
Takarada Y, Takazawa H, Sato Y, Takebayashi S, Tanaka Y,
Ishii N (2000) Eects of resistance exercise combined with
moderate vascular occlusion on muscular function in humans.
J Appl Physiol 88:20972106
Takarada Y, Sato Y, Ishii N (2002) Eects of resistance exercise
combined with vascular occlusion on muscle function in ath-
letes. Eur J Appl Physiol 86:308331
Thorstensson A, Karlsson J, Viitasalo JH, Luhtanen P, Komi PV
(1976) Eect of strength training on EMG of human skeletal
muscle. Acta Physiol Scand 98:232236
Tsunoda N, OHagan F, Sale DG, MacDougall JD (1993) Elbow
exion strength curves in untrained men and women and male
bodybuilders. Eur J Appl Physiol 66:235239
Yao W, Fuglevand RJ, Enoka RM (2000) Motor-unit synchroni-
zation increases EMG amplitude and decreases force steadiness
of simulated contractions. J Neurophysiol 83:441452