Content uploaded by Richard G. Lomax
Author content
All content in this area was uploaded by Richard G. Lomax on Oct 05, 2021
Content may be subject to copyright.
259
Journal of Strength and Conditioning Research, 2003, 17(2), 259–273
q2003 National Strength & Conditioning Association
A Series of Studies—A Practical Protocol for
Testing Muscular Endurance Recovery
J
OHN
R. M
C
L
ESTER
,
1
P
HILLIP
A. B
ISHOP
,
2
J
OE
S
MITH
,
2
L
ANA
W
YERS
,
3
B
ARRY
D
ALE
,
4
J
OSEPH
K
OZUSKO
,
5
M
ARK
R
ICHARDSON
,
2
M
ICHAEL
E. N
EVETT
,
2
AND
R
ICHARD
L
OMAX
6
1
Department of Physical Education and Recreation, Western Kentucky University, Bowling Green, Kentucky
42101;
2
Department of Kinesiology, The University of Alabama, Tuscaloosa, Alabama 35487;
3
Northriver Fitness
Center, Tuscaloosa, Alabama 35404;
4
Department of Physical Therapy, University of South Alabama, Mobile,
Alabama 36688;
5
Department of Family, Nutrition, and Excercise Science, Queens College, Flushing, New York
11367;
6
Department of Educational Studies in Psychology, Research Methodology, and Counseling, The
University of Alabama, Tuscaloosa, Alabama 35487.
ABSTRACT
The purpose of this series of studies was to use a practical
measure to examine the course of muscular endurance re-
covery after 3 sets to failure in 10 men (ages 18 to 30 years)
and then compare those results with 10 men (ages 18 to 30
years) who performed 7 sets and 10 older men (ages 50 to
65 years) who performed 3 sets. Recovery as indicated by
number of repetitions performed was observed at 24, 48, 72,
and 96 hours. Repeated-measures ANOVA was used to in-
vestigate differences in recovery over time. For group means,
performance was significantly lower in all 3 groups after 24
hours (p,0.05). At 48 hours, performance of the groups
was not significantly different from baseline (p.0.05).
Number of repetitions performed at 72 hours was signifi-
cantly higher than that in session 1 (10.2 61.4 reps in session
1 vs. 11.2 62.3 at 72 hours, p50.022) in the young 3-sets
group, but not in the other groups. After 96 hours, only the
young 7-sets group was found to be performing at a level
approaching significance (10.3 61.2 reps in session 1 vs. 11.1
62.0 at 96 hours, p50.051). No significant difference was
found between the young 3-sets and 7-sets groups at any
time (p.0.05). The young 3-sets group was found to be
performing at a significantly higher level than the older
group at 72 hours (11.2 62.3 reps in the younger vs. 9.9 6
1.7 in the older group, p50.008), a difference that also ap-
proached significance at 96 hours (p50.06). Large intersub-
ject variability was observed at all time points. The results
suggest that individual recovery testing before exercise pre-
scription is practical, and this protocol may be sensitive to
differences in training volume and subject age.
Key Words: aging and exercise, exercise testing pro-
tocols, resistance training, training volume, strength
training
Reference Data: McLester, J.R., P.A. Bishop, J. Smith, L.
Wyers, B. Dale, J. Kozusko, M. Richardson, M.E. Nev-
ett, and R. Lomax. A series of studies—A practical
protocol for testing muscular endurance recovery. J.
Strength Cond. Res. 17(2):259–273. 2003.
Introduction
A
t present, many typical resistance-training pro-
grams consist of 2–6 days per week of upper- and
lower-body training on alternate days, allowing 48
hours of recovery between sessions. However, great va-
riety exists in the exercise programs of both the lay-
person and the athlete. In addition, large variation ex-
ists in individual responses to the same training pro-
tocol.
Recovery from resistance exercise has been studied
extensively. Muscle force has been found to recover to
within normal range anywhere from 24 hours to 7
days after the exercise bout depending on the protocol
used to fatigue the subjects (13, 15, 27–29, 35, 36).
However, the topic is often explored through the use
of isokinetic machines, electromyography (EMG), or
indirectly through frequency studies. Therefore, the se-
ries of studies presented here attempts to explore the
use of a common fatiguing stimulus and a practical
protocol for endurance recovery testing in 3 different
situations.
Although some have shown evidence that muscle
soreness and fatigue can have independent causes (35,
36), delayed-onset muscle soreness (DOMS) must be
considered along with the topic of recovery time. Stud-
ies of DOMS have found it to peak at various times
between 24 and 72 hours after the exercise bout (2, 40).
As strength has been shown to be less than optimal
during DOMS (6–8, 20, 23, 26, 35, 40, 41), training with
DOMS, or training in the presence of an effector mech-
260 McLester, Bishop, Smith, Wyers, Dale, Kozusko, Richardson, Nevett, and Lomax
anism for DOMS, may not be as effective as waiting
until it has subsided. Allowing the optimal recovery
period between training sessions should result in the
ability to train at a higher intensity while preventing
detraining.
Though these studies are of great use in under-
standing muscle recovery, the protocols for exhaustion
differ and the methods of study normally involve ec-
centric exercise only. There exists the need for a prac-
tical means of assessing individual muscular endur-
ance recovery from a given bout of resistance exercise.
In addition, there is probably large interathlete var-
iability in recovery due to individual variations in
training experience, muscle composition, fuel storage,
and other aspects of physiology. This may not only
occur between sports but also within, as different
sports positions are often trained in varying ways in
accordance with specificity. Each individual’s personal
exercise habits create a unique recovery situation,
which probably requires individualized recovery pre-
scription.
For example, currently many combinations of fre-
quency, intensity, and volume of training are being uti-
lized both by experienced athletes and laymen. These
various combinations of training variables are usually
based upon some personal belief about necessary re-
covery time.
Studies have been conducted that compare various
frequencies of training (3, 11, 21, 33). For example, 1
study has shown relatively little difference between 2
and 3 days per week (3), whereas another has shown
comparable strength increases from 1-day-per-week
and 3-day-per-week protocols (33). Still others have
found that better results are obtained with frequencies
of training higher than 3 days per week (11, 21).
Though these studies investigated frequency, in es-
sence they are recovery studies dealing with various
daily and weekly volume doses. In addition, most of
the frequency studies thus far have involved novice
lifters. The neural gains in strength that occur in in-
experienced lifters may mask any decrement in per-
formance due to inadequate recovery.
Of those studies that investigate recovery directly,
several have shown acute muscle fatigue to be related
to both the volume and intensity of the exercise session
(13, 15, 27–29, 35, 36). However, many studies are lim-
ited to only 1 muscle group, or recovery is investigated
with protocols or equipment that are not commonly
utilized by or available to typical strength trainers. It
would seem logical that recovery would also have an
inverse relation with training volume. For example,
short-term power of the leg extensors during recovery
has been found to be inversely proportional to the in-
tensity of the prior exercise bout (19). In addition, Hak-
kinen (13) has found muscle recovery after the squat-
lift to be 97% complete in 48 hours. However, the fa-
tiguing protocol for this study (13) was 20 sets of 1
repetition with 100% of 1 repetition maximum (1RM),
a very difficult protocol to interpret in terms of appli-
cation in the average exercise setting. In other words,
the average strength trainer does not perform 20 sets
of a single exercise, and they typically perform more
than 1 repetition on each of those sets (although each
is not 100% of 1RM).
Kroon and Naeije (29) fatigued subjects using var-
ious types of contractions and percentages of maximal
voluntary contraction force (MVC): isometric (50% of
MVC on the first set and 40% of MVC on the second
set), concentric (40% of MVC on the first set and 30%
of MVC on the second set), or eccentric contractions
(40% of MVC on the first set and 30% of MVC on the
second set). The contractions were of 3 seconds dura-
tion and each was followed by 2 seconds of rest. After
each set of 10 repetitions, 1 minute of rest was given.
The subject was considered to be exhausted when he
could not perform the desired contraction 3 times in
succession. The concentric and isometric actions
caused EMG responses and muscular endurance per-
formance that were disrupted for 1 to 2 days, but ec-
centric actions caused disruptions that did not statis-
tically return to normal for 7 days. Once again, al-
though it appears that recovery may be slower than
many people assume, this protocol is very hard to in-
terpret in terms of the average resistance-training pro-
gram because of the involvement of both concentric
and eccentric actions in the typical training protocol.
Also within the realm of muscular endurance re-
covery, there has been a surge of interest as of late in
research concerning aged populations. Most of the
studies thus far have focused on trainability of the
aged and the benefits of such training. One area of
research that is lacking is that pertaining to recovery
from resistance training in older individuals. Much an-
ecdotal evidence has supported the notion that older
individuals may need an extended period of recovery.
One study has found a decrement in strength train-
ability with age (32). However, most researchers be-
lieve that the capability to gain strength is intact in
older men (1, 9, 10, 34) and women (1, 5, 9, 24, 39).
Some of the work even shows strength gains compa-
rable with younger subjects (10). Most believe that
strength increases in aged populations are at least in
part due to a retained ability to increase muscle mass
(5, 9, 10). In some studies, however, the strength gains
in older populations have been attributed to neural fac-
tors as opposed to hypertrophic factors (30, 34). This
is not a surprising finding as most aged are relatively
untrained and would be expected to respond as un-
trained subjects to a novel training stimulus. Also, it
should be noted that their much lower initial strength
relative to younger cohorts means that smaller gains
present as larger percentages. However, Moritani and
deVries (34) found that neural factors played a large
role throughout the training process (8 weeks) for old-
Muscular Endurance Testing
261
er subjects, whereas in younger subjects the neurolog-
ical component only dominated for the first 4 weeks.
There are a couple of reasonable explanations for a
possibly reduced capability of aged muscle to hyper-
trophy. One reason for a potentially blunted hypertro-
phic effect could simply be the reported loss of muscle
fibers with age (31). Another explanation could be the
need for extra recovery time between sessions, possi-
bly stemming from the observed decrements in both
growth hormone and testosterone (16, 17, 37, 38) with
age or a blunted response of these hormones to an
exercise bout even when resting levels are normal (12).
These hormones are known to be anabolic and lower
concentrations could increase the need for recovery
time. Also, some speculate that there could be an in-
ability to train at high enough intensity to observe the
normal hormonal spikes seen with a training session
in younger subjects (17). It is hypothesized that this
reduced capacity could be due to neural inhibition (14,
30), or it could be due to lack of aggression because of
the above-mentioned loss of testosterone, or simply to
fatigue.
One of the most obvious applications of individual
recovery testing is simply to increase the effectiveness
of exercise protocols currently used by the general
population. Exercise protocols vary so much between
individuals, and individual responses to training are
so different that it should seem obvious that blanket
prescription of resistance training frequency is not the
most effective means of optimizing training. People
need to know their own optimal frequency of training
not only to avoid detraining but also to prevent under-
recovery, which may maximize training responses. In
addition, some trainers may want to know the mini-
mum number of workouts they should perform per
week. Therefore, the current series of studies attempts
to explore the use of a practical protocol for individ-
ualized recovery testing in 3 situations.
Experiment 1: Endurance Recovery After 3
Sets to Failure of 8 Exercises
Research is needed to establish a practical protocol for
determining an individual’s recovery-from-training
curve, especially in those who are already experienced
with resistance training to varying degrees. Therefore,
the purpose of this study was to examine the course
of muscular endurance recovery using a common fa-
tiguing protocol and a very practical measure of per-
formance. If a practical protocol for the determination
of recovery can be sensitive to daily individual chang-
es in 10RM performance, it could be a useful tool in
individualizing training prescriptions.
Methods
Subjects
Ten healthy men (18 to 30 years of age) were recruited
for the study. Ten subjects were utilized because this
was the first use of this paradigm and it was uncertain
if individual differences among the subjects would be
large or small. Only men with strength training ex-
perience were recruited, to provide as much homoge-
neity as possible. To be eligible, each was required to
have resistance-trained consistently with a protocol of
3 to 4 days per week or such that each major muscle
group has been trained at least twice per week, for at
least 12 weeks (24 workouts) immediately before ini-
tiation of the study. This requirement was instituted
to remove as much effect due to neural factors as pos-
sible. No subjects were competitive lifters.
Subjects were asked not to participate in any other
forms of exercise other than the testing protocol dur-
ing the study and were also asked to maintain dietary
habits as close as possible to those before the study.
Written informed consent in accordance with the pro-
cedures for the protection of human subjects set forth
by the local Institutional Review Board was obtained
before testing.
Experimental Approach to the Problem
A pretest questionnaire was administered to the sub-
jects regarding previous and present exercise experi-
ence to assure that the subject population was homo-
geneous and that all were at least as experienced as
the criterion without requiring that they be advanced
lifters. Weight, body density, percent body fat, and
subsequently, percent lean tissue were estimated
through the use of medical scales (Detecto), skinfold
calipers (Lange), and the 3-skinfold-site Jackson and
Pollock equation (chest, abdomen, and thigh) (25). All
measurements were taken by the same experienced
technician in accordance with the American College of
Sports Medicine’s guidelines (25).
Also during the pretest session, a 10RM for each
exercise in the protocol was established. A light load
(easily allowing 15 repetitions) was used as a warm-
up before the 10RM test. The 10RM was established
for each lift by increasing the load by 2.3 to 9.1 kg
(depending on the level of difficulty) after each suc-
cessful attempt until the 10RM was reached for each
of the 8 exercises. Two to 4 minutes were allowed be-
tween trials to allow for adequate recovery. All 10RM
tests were supervised by the same experienced tech-
nician to insure that the 10RM was found within 3
trials and that the testing was consistent across sub-
jects. This 10RM weight was the load used consistently
for all subsequent test sessions, with subjects encour-
aged to do as many repetitions as possible. The 10RM
weight load (as opposed to using a percentage of
1RM) was chosen because of its common application
in the average recreational setting. This paradigm
should be safer for nonathletes, and does not require
as much assistance; therefore it was the measurement
of choice for this particular study. Also, it is believed
that the number of reps with 10RM weight is a more
262 McLester, Bishop, Smith, Wyers, Dale, Kozusko, Richardson, Nevett, and Lomax
specific measure of resistance training performance
since many athletes typically train in the 6- to 12-rep-
etition range.
Once the treatment had begun, subjects were asked
before each exercise session their global level of DOMS
on a scale of 1 to 10 (1 being little or no muscle sore-
ness and 10 being extreme muscle soreness) produced
by the previous session. Also, subjects were asked after
each exercise to rate the perception of difficulty with
that bout (1 being extremely easy and 10 being ex-
tremely difficult). The topic of soreness was explored
because although the subjects were experienced, they
may not have been performing sets to momentary
muscular failure.
Experimental Design
The 10RM weight load established in the pretest was
the load used for each set of exercise in the testing
sessions. Before the first set of each exercise, a warm-
up set of 10 repetitions with approximately 60% of the
subject’s 10RM weight was performed. For the first ex-
ercise session, the subjects performed 3 sets to mo-
mentary muscular failure of each exercise before mov-
ing to the next. The number of successful repetitions
accomplished by the subject during this first exercise
session was the reference standard for the subsequent
sessions. Subjects were required to perform their max-
imum number of repetitions for each set of each lift in
subsequent test sessions. The subjects returned for 4
subsequent bouts of exercise (24, 48, 72, and 96 hours
after the previous session in randomized order among
subjects) in which they were asked to perform all sets
to momentary muscular failure. The original randomly
established order of exercises for the individual subject
was held constant for each session. Subjects exercised
at the same time of day for the initial and subsequent
sessions.
Only the first set of exercise for each of the bouts
was used for analysis to avoid potential variations in
rest intervals between sets among test sessions (rest
between sets ranged from 30 seconds to 1 minute).
Therefore, the first set was assumed to be the purest
measure of performance. The second and third sets
were performed to achieve the degree of fatigue typi-
cal of many recreational resistance-training regimens.
Randomization among subjects was used to partially
cancel any training effects that might accrue from any
particular timing of workouts (tests).
All subjects were instructed to perform each set to
momentary muscular failure, and all tests were closely
supervised. However, because of the fact that the num-
ber of supervisors was limited, they were aware of the
testing order. As in all studies with supervisors that
are not blinded, this can be a source of some bias due
to expectations. The exercises were performed using
Cybex machines (Lumax, Ronkonkoma, NY) and the
targeted muscle groups were as follows: (a) seated up-
right bench press—pectoralis major (machine model
#4015); (b) seated upright lateral arm raise—deltoid
(4031); (c) seated elbow extension (triceps press)—tri-
ceps brachii (4035); (d) seated arm pull-down (lat
pull)—latissimus dorsi (4005); (e) seated elbow flexion
(bicep curl)—biceps brachii (4040); (f) supine hip ex-
tension (leg press)—gluteus maximus and quadriceps
(4100); (g) prone knee flexion (leg curl)—hamstrings
(4110); (h) seated knee extension (leg extension)—
quadriceps (4105).
These particular lifts were chosen because they rep-
resent the major muscle groups and because of their
common usage among most lifters. The first 3 exercis-
es were always performed in the same order, as were
lifts d and e and lifts f through h. Grouping in this
manner kept fatigue of smaller muscles from interfer-
ing with large muscle group performance. However,
the order of those groups of exercises was randomized
to control for any ordering effects.
Statistical Analyses
SPSS for windows statistical program (v. 10.0) was
used to perform all analyses. A general linear model
(GLM) with repeated measures was performed on all
variables. The GLM was followed by univariate anal-
yses for all significant effects. Polynomial contrasts
were used for each exercise to detect which trends
were significant for the recovery of each lift. Signifi-
cance was determined at an overall p#0.05. Data
were expressed as mean 6SD. Because of small sam-
ple size, one weakness of the study was low power to
detect a significant difference. However, most of the
inferences drawn from the study were not affected by
statistical power.
Results
Subjects’ physical characteristics are displayed in Table
1. Figure 1 depicts the changes in performance for each
subject since the previous test period, as represented
by the mean delta score for all 8 lifts combined.
Performance was lowest after the 24-hour recovery
period, with none of the subjects being fully recovered
(mean of 10.2 61.4 repetitions in session 1 vs. 8.4 6
2.1 at 24 hours, p50.001). After 48 hours of recovery,
the subjects’ performance was no longer significantly
different from baseline (mean of 10.2 61.4 repetitions
in session 1 vs. 9.9 62.1 at 48 hours, p50.313). Re-
sults from the 72-hour recovery period indicated that
the subjects performed above baseline values (mean of
10.2 61.4 repetitions in session 1 vs. 11.2 62.3 at 72
hours, p50.022). After 96 hours, the subjects’ perfor-
mance was no longer significantly different from ses-
sion one (mean of 10.2 61.4 repetitions in session 1
vs. 11.2 62.8 at 96 hours, p50.075).
Figure 2 depicts performance since the previous
test period as represented by means and standard de-
viations of the number of repetitions, for the upper-
Muscular Endurance Testing
263
Table 1. Subject characteristics for 3-sets males (n510) and 7-sets males (n510).
3-sets Younger
Mean SD
7-sets Younger
Mean SD
3-sets Older
Mean SD
Age (y)
Ht (cm)
Wt (kg)
Body fat (%)
22.6
175.8
75.6
10.5
4.6
6.5
10.8
3.4
24.9
179.6
87.1
14.9
3.2
7.5
7.5
5.4
56.4*
179.1
88.6
23.9
5.0
4.8
17.4
6.0
Training history
Training (y)
Frequency (days/wk)
Sets/exercise
Reps/set
5.9
3.4
3.4
10.4
3.9
0.8
1.0
3.6
7.1
3.1
3.1
9.6
4.0
0.9
0.4
1.6
5.4
2.75
2.4
10.7
6.2
0.4
0.8
1.8
* Significant difference from 3-sets Younger group (p,0.05).
Figure 1. Changes in performance (repetitions performed)
for each subject since the previous test period, as repre-
sented by the mean delta score for all 8 lifts combined.
Figure 2. Changes in performance (mean 6SD repeti-
tions performed) since the previous test period for upper-
and lower-body exercises. No significant difference (p.
0.05) between body regions. *, significant difference (p,
0.05) from session 1 for all 8 lifts combined.
Figure 3. Performance (mean delta score) for the group
by lift since the previous test session.
and lower-body exercises. No significant interaction of
time and body region was observed (p50.676). As
can be seen, the tendencies of the upper- and lower-
body performance were virtually the same at 24 hours
(mean of 10.5 61.2 repetitions in session 1 vs. 8.5 6
2.3 at 24 hours for the lower body; mean of 10.1 61.4
repetitions in session 1 vs. 8.3 62.0 at 24 hours for
the upper body) and neither was significantly different
from baseline at 48 hours (mean of 10.5 61.2 repeti-
tions in session 1 vs. 10.2 62.6 at 48 hours for the
lower body; mean of 10.1 61.4 repetitions in session
1 vs. 9.8 61.8 at 48 hours for the upper body). After
72 hours, the lower-body mean did rise slightly higher
above baseline than the upper body, though this was
not a significant difference (mean of 10.5 61.2 repe-
titions in session 1 vs. 11.8 62.9 at 72 hours for the
lower body; mean of 10.1 61.4 repetitions in session
1 vs. 10.9 61.8 at 72 hours for the upper body). After
the 96-hour recovery period, the lower body showed a
trend toward still rising above baseline, whereas the
upper body displayed a trend toward returning to ses-
sion 1 values, though these were not significantly dif-
264 McLester, Bishop, Smith, Wyers, Dale, Kozusko, Richardson, Nevett, and Lomax
ferent (mean of 10.5 61.2 repetitions in session 1 vs.
11.9 62.9 at 96 hours for the lower body; mean of 10.1
61.4 repetitions in session 1 vs. 10.8 62.6 at 96 hours
for the upper body).
Figure 3 displays the mean delta scores for the
group by lift since the previous test session. Polyno-
mial contrasts indicated that for the bench press (lin-
ear, p50.023; quadratic, p50.016; cubic, p50.011),
tricep extension (linear, p50.019; quadratic, p50.045;
cubic, p50.000), and leg extension (linear, p50.006;
quadratic, p50.018; cubic, p50.001), linear, qua-
dratic, and cubic trends were significant. For the lat
pull (linear, p50.008; cubic, p50.015), leg press (lin-
ear, p50.003; cubic, p50.009), and leg curl (linear,
p50.03; cubic, p50.000) exercises, the linear and
cubic trends were significant. The bicep curl showed
significant quadratic (p50.001) and cubic trends (p5
0.001). The lateral raise exercise showed a significant
(p50.032) cubic trend.
Soreness ratings at 24 (2.7 61.9), 48 (2.1 62.5), 72
(1.8 62.3), and 96 hours (1.5 61.8) were not analyzed
statistically but tended to be low and tended to de-
crease with each successive rest interval. Perception of
effort ratings were very consistent for the 24- (7.3 6
1.5), 48- (7.0 61.4), 72- (7.1 61.1), and 96-hour (6.6
61.0) sessions and also seemed to decline with in-
creasing recovery intervals.
Discussion
The purpose of this study was to investigate the time
course of muscular endurance recovery using a com-
mon fatiguing protocol and a practical measure of per-
formance. The results indicate that the protocol used
in this investigation seems to be sensitive to changes
in individual muscular endurance and may be useful
in individualized exercise prescription.
The finding that none of the subjects returned to
baseline values after 24 hours was not surprising.
However, the subsequent recovery periods do show
some unexpected results. Though the subjects were
not significantly different from baseline after 48 hours,
it is interesting to look at the individual data. For ex-
ample, after 48 hours of recovery, only 4 of the subjects
were fully recovered enough to duplicate their initial
performance. This finding is somewhat contradictory
to the commonly utilized every-other-day (every 48
hours) protocol. Many lifters do not perform all of
their sets to failure; however, many subjects are per-
forming a larger number of sets and exercises than
were used in our experimental paradigm. In addition,
of subjects that had not returned to baseline at 48
hours, 1 subject was still declining in strength. This is
similar to the finding of Johansson et al. (22), whose
subjects were found to experience their lowest levels
of force 48 hours after 10 sets of 10 eccentric actions.
However, in the Johansson et al. (22) study, 48 hours
is also when DOMS was found to peak, but unfortu-
nately the degree of experience of their subjects was
not elucidated upon. As mentioned previously, the bio-
chemical, biophysical, or biomechanical responses that
accompany DOMS in untrained lifters could still be
present to a less overt degree in experienced lifters and
therefore could cause delayed effects on force produc-
tion. In addition, some individuals could be more sus-
ceptible to these responses. Even though our study
does not purport to elucidate mechanisms for these
individual differences in recovery at 48 hours, it does
provide evidence that the typical every-other-day
training routine may not be optimal for some exercis-
ers.
By 72 hours, 8 of the subjects were back to baseline
and evidencing a significant increase in repetition per-
formance ability (p50.022 when compared with base-
line). This finding of a rise above baseline at 72 hours
as opposed to 48 hours may seem counter to common
exercise prescription, but it is very much in line with
previous research. Several investigations have found
muscular force production capacity to be disrupted for
3 to 4 days after training (4, 23, 36). Once again, how-
ever, direct comparisons cannot be made, as different
methods were used for muscle fatigue and testing
(percutaneous electrical myostimulation of eccentric
actions, for example [4]). It is also at this 72-hour ses-
sion that the subjects began to show more variation
among individuals, partially due to the fact that 1 of
the subjects was still not back to baseline values and
another was actually beginning to decrease his num-
ber of repetitions, suggestive of detraining. Even
though 8 of the subjects were fully recovered after 72
hours of rest, the fact that 1 subject was not and an-
other was apparently detraining again points to large
individual variation among subjects’ ability to recover.
It can only be assumed that the subject apparently be-
ginning to detrain may be representative of that por-
tion of the population that should engage in training
every 48 hours to provide the necessary stimulus for
progressive strength increases. However, the subject
still not fully recovered may represent an equal por-
tion of our small sample that needs a longer period of
recovery, at least after a workout similar to our test
protocol.
After 96 hours of recovery, again a very large de-
gree of individual variation was observed. Though not
significantly different from baseline, 3 of the subjects
were still showing a trend toward increase in repeti-
tion performance ability (possible training effect) and
6 had either reached a plateau or were unable to attain
the same repetition number (possibly detraining). In
addition, 1 subject was still not fully back to baseline.
These results may indicate that for some subjects 96
hours may be too long to wait before the next bout of
training, at least after 3 sets to failure of each exercise.
However, the fact that 3 subjects were still receiving a
Muscular Endurance Testing
265
training effect and 1 was not yet fully recovered again
indicates individual subject variation that should be
taken into consideration before exercise prescription.
Indeed, 96 hours between training sessions may intu-
itively seem too long, which points to the need for
carefully controlled training studies using individu-
alized frequency prescriptions. It should be noted that
each time point was made up of the mean of the 8
different exercises. This could help to explain the sub-
ject who was never fully recovered. In other words,
the mean number of repetitions of 8 different exercises
may be lowered because of some of the body parts
recovering faster than others, and therefore detraining
as the others are becoming fully recovered. It should
also be considered that previous studies have found
similar responses. Kroon and Naeije (29) found force
to be depressed for 7 days after performing eccentric
exercise to exhaustion with 40% of maximum. In ad-
dition, Johansson et al. (22) used a protocol of 10 sets
of 10 maximal isokinetic eccentric actions and found a
decrement in performance that was still present at 96
hours. Although the above protocols each differ from
those of the present study and their subjects were in-
experienced, our study supports their observations
that relatively long recovery periods may be needed
in some individuals. But even if the cause of this long-
lasting force decrement was due to varying recovery
times for different exercises, it is further evidence of
interindividual differences in recovery requirements,
which suggests that the most effective training pro-
tocols may require preliminary subject testing. Having
a different recovery interval for different athletes or
muscle groups could result in a very complex training
schedule, but complexity seems less significant in view
of the great amount of work invested by even serious
recreational lifters.
The reader should be reminded that these results
are based upon the specific population of subjects test-
ed. Therefore, the results cannot be generalized to all
other populations (e.g., beginners, more experienced,
or competitive lifters). It should also be noted that
there was wide variation in the training experience of
these subjects. Though it could be said that the large
individual variation observed among subjects’ recov-
ery was due to varying training histories, it further
demonstrates that the protocol used in this study
could be important for individualizing training pro-
grams for lifters of varying backgrounds. It should
also be mentioned that there is inherent day-to-day
variation in any testing protocol (such as 10RM test-
ing), which could also account for the observed indi-
vidual variation.
The different recovery responses evident in Figure
1 may arise from the test order. Perhaps those showing
the unrelenting rise after 48 hours benefited from
some particular test order (e.g., perhaps the test se-
quence of 48 hours, then 72, then 96, then 24 offers
some advantage relative to some other order). To test
this speculation, we retrospectively examined the test
order for subjects showing similar responses. Surpris-
ingly, none of the subjects showing similar responses
was subjected to similar test sequences.
As Figure 2 indicates, no significant difference was
observed between upper- and lower-body recovery.
This gives some hope that ultimately individual train-
ing programs would not be too complex. Consistent
with the individual results, neither the upper- nor low-
er-body exercises were significantly different from
baseline at 48 hours, also being affected to virtually
the same degree at 24 hours. However, after 72 hours
the lines in the figures begin to separate slightly
though not significantly, with the lower body appar-
ently receiving a greater training effect. In addition,
after 96 hours of recovery the lower body still seemed
to be increasing above baseline while the upper-body
variables were on the decline, although they were not
significantly different. The slight separation in the
lines representing the 2 regions of the body could sim-
ply be due to inadvertent differences in training inten-
sity present in the subjects’ normal exercise protocols.
If the legs of these subjects were relatively untrained,
it would be expected that they would have been af-
fected to a greater degree than the upper-body vari-
ables at 24 and 48 hours. In addition, the legs may be
more active during daily activity and may possess a
greater natural ability to respond to a given stimulus
possibly due to active recovery. An alternative expla-
nation is that the larger muscles of the legs may re-
quire a larger stimulus to cause significant fatigue. But
again, it should be kept in mind that the upper- and
lower-body regions were not significantly different at
any time period. Also, this is the area of the study in
which low statistical power could have played a large
role.
The data from Figure 3 show no detectable differ-
ences between the individual lifts. All lifts demonstrat-
ed significant cubic trends, indicating 2 distinct bends
in the lines representing recovery. However, subse-
quent testing with more subjects may reveal differenc-
es between muscle groups as evidenced by differences
in performance on the exercises. For example, biceps
curl and hamstrings curl both show a similar pattern.
Likewise, the 3 leg exercises all show sharp increases,
and 2 of these could be expected to show improve-
ments beyond 96 hours. Indeed, when the polynomial
contrasts were run from the 24-hour point instead of
session 1, the trends for all lifts were still significant
linearly, but only the bicep curl and leg curl were sig-
nificantly quadratically and only the bicep curl was
significant cubically.
In Figures 2 and 3 the mean values may obscure
the large intersubject variability, as was seen in Figure
1. Although individual variations in 8 different exer-
cises over 96 hours are too complex for presentation
266 McLester, Bishop, Smith, Wyers, Dale, Kozusko, Richardson, Nevett, and Lomax
Figure 4. Changes in performance (repetitions performed)
for each subject since the previous test period, as repre-
sented by the mean delta score for all 8 lifts combined.
and interpretation here, we believe that the results for
any individual lifter could be useful in planning the
optimal training frequency.
In summary, the protocol used in this experiment
seems to be sensitive to individual changes in mus-
cular endurance capacity. In addition, great individual
variation exists in the ability to recover muscular en-
durance. Statistically, subjects were not different from
baseline after 48 hours, indicating full recovery within
that time period. However, 48 hours may not allow
some to completely recover to baseline muscular en-
durance capacity after a bout of exercise, as indicated
by individual subject data. Subject performance was
statistically higher than baseline after 72 hours, maybe
indicative of a slight training effect at that point. After
96 hours, the group mean was back to baseline, with
some subjects still increasing in strength and others
reaching a plateau or possibly detraining.
Future studies are warranted to test this initial
work. The reliability and sensitivity of the protocol
need to be established. Training studies that imple-
ment individual frequency prescription are needed to
test the effectiveness and practicality of varying fre-
quency among subjects. In addition, untrained sub-
jects could be used as representatives of the physical
rehabilitation population and compared to trained
subjects. The paradigm presented, if it proves to be
sufficiently reliable and valid, may be useful as a de-
sign for testing the efficacy of ergogenics designed to
speed recovery.
Practical Applications
The present study provides some preliminary infor-
mation on a practical protocol for testing of muscular
endurance recovery in experienced college-aged men.
We believe that this protocol is safe enough and simple
enough that any lifter could use this protocol for self-
assessment. The protocol could be substituted for nor-
mal workouts over a 2-week period and permit rela-
tively convenient reassessment of recovery over the
course of a training or sports season cycle. Potential
areas for application are training athletes, patients un-
dergoing physical rehabilitation, and astronauts before
missions.
Experiment 2: Endurance Recovery after 7
Sets of Exercise vs. 3 Sets
It is apparent that research is needed to establish the
effects of volume of exercise done on the recovery from
fatiguing resistance exercise that is interpretable from
the standpoint of typical resistance training programs
and utilizes a practical testing protocol. It would be
highly useful information, as there are many different
volumes of training currently being utilized.
Therefore, the purpose of this study was to exam-
ine muscular endurance recovery after 7 sets of exer-
cise while utilizing a very practical testing protocol. In
addition, the results of this study were compared with
the data from experiment 1 involving the use of 3 sets
of exercise. Because, on the basis of Kroon and Naeije
(29), many individuals potentially train in a constant
state of underrecovery, the information could prove
very useful in the individualization of exercise pre-
scription if the protocol seems to be sensitive to chang-
es in training volume.
Methods
Subjects
Ten healthy men (18 to 30 years of age) were recruited
for the study. The same subject requirements as ex-
periment 1 were applied.
Experimental Approach to the Problem
The experimental protocols, design, and statistical
analyses were the same as for experiment 1 with the
exception that the subjects performed 7 sets to mo-
mentary muscular failure instead of 3 sets before mov-
ing on to the next exercise.
Results
Subjects’ physical characteristics are shown in Table 1.
Figure 4 depicts the changes in performance for each
subject since the previous test period as represented
by the mean delta score for all 8 lifts combined over
the testing period.
The lowest performance values were observed after
the 24-hour recovery period, with none of the 10 sub-
jects returning to baseline values (mean of 10.3 61.2
repetitions in session 1 vs. 8.5 61.7 repetitions at 24
hours, p50.000). After 48 hours of recovery, subjects’
performance was no longer significantly different from
session 1 (mean of 10.3 61.2 repetitions in session 1
vs. 9.8 62.0 repetitions at 48 hours, p50.175). After
Muscular Endurance Testing
267
Figure 5. Changes in performance (mean 6SD repeti-
tions performed) since the previous test period for the up-
per- and lower-body exercises. No significant difference (p
.0.05) between body regions. *, significant difference (p,
0.05) from session 1 for all 8 lifts combined.
Figure 6. Performance (mean 6SD repetitions per-
formed) comparisons between the 7-set and 3-set groups.
No significant difference (p.0.05) between groups.
72 hours of recovery, the subjects were still not signif-
icantly different from baseline (mean of 10.3 61.2 rep-
etitions in session 1 vs. 10.6 61.8 repetitions at 72
hours, p50.465). Though not quite statistically sig-
nificant, at 96 hours of recovery the subjects ap-
proached statistical significance above baseline (mean
of 10.3 61.2 repetitions in session 1 vs. 11.1 62.0
repetitions at 96 hours, p50.051).
Figure 5 depicts the course of performance since
the previous test session, as represented by means and
standard deviation, for the upper- and lower-body ex-
ercises. There was no significant difference between
the upper- and lower-body regions over time (p5
0.408). Both the upper and lower regions of the body
were equally affected at 24 hours (mean of 10.1 61.3
repetitions in session 1 vs. 8.5 61.9 repetitions at 24
hours for the lower body; mean of 10.5 61.2 repeti-
tions in session 1 vs. 8.6 61.6 repetitions at 24 hours
for the upper body) and 48 hours (mean of 10.1 61.3
repetitions in session 1 vs. 9.9 62.2 repetitions at 48
hours for the lower body; mean of 10.5 61.2 repeti-
tions in session 1 vs. 9.7 61.8 repetitions at 48 hours
for the upper body). At 72 hours, the upper- and low-
er-body regions were still not statistically different
(mean of 10.1 61.3 repetitions in session 1 vs. 10.2 6
1.7 repetitions at 72 hours for the lower body; mean
of 10.5 61.2 repetitions in session 1 vs. 10.8 61.8
repetitions at 72 hours for the upper body). After 96
hours, both the upper and lower regions of the body
were demonstrating similar patterns of a slight trend
toward above-baseline values that were still not sig-
nificantly different (mean of 10.1 61.3 repetitions in
session 1 vs. 10.9 61.9 repetitions at 96 hours for the
lower body; mean of 10.5 61.2 repetitions in session
1 vs. 11.2 62.0 repetitions at 96 hours for the upper
body).
Again, soreness ratings at 24 (4.4 62.1), 48 (3.0 6
2.4), 72 (3.2 62.2), and 96 hours (2.7 62.5) tended to
be low and tended to decrease with each successive
rest interval. Perception of effort ratings were very
consistent for the 24- (7.8 61.3), 48- (7.1 60.7), 72-
(7.5 61.3), and 96-hour (7.3 61.1) sessions, being
lowest at 48 and 96 hours.
Figure 6 depicts performance (mean 6SD repeti-
tions performed) comparisons between the 7-set and
3-set groups. There was no significant difference be-
tween the 7-set and 3-set groups at any time period (p
50.337).
Discussion
The purpose of this study was to investigate the effects
of 7 sets of exercise on muscular endurance recovery
using a practical testing protocol and compare those
findings with our previous experiment utilizing 3 sets
of each exercise. Though there was no significant dif-
ference between the 7-set and 3-set group at any time
period (p50.337), the results provide evidence that
the utilized protocol was sensitive to the notion that
higher volumes of exercise may require relatively lon-
ger periods of recovery, and also that training effects
from a given bout of exercise may not peak for several
days after a training bout.
As expected, ability to perform repetitions com-
parable with session 1 dropped sharply at 24 hours.
Although not significantly different, 6 of the subjects
were still performing a lower number of repetitions
after 48 hours of recovery than at baseline. As can be
seen in Figure 6, these findings are consistent with our
previous research on subjects performing 3 sets of each
exercise in which the subjects’ performance was not
significantly different from baseline values after 48
hours, but 6 of the subjects were performing below
baseline values.
The individual subject data provides evidence that
some lifters may need more recuperation than is some-
268 McLester, Bishop, Smith, Wyers, Dale, Kozusko, Richardson, Nevett, and Lomax
times believed. Though many lifters do not perform
all of their sets to failure, as was required for this in-
vestigation, many exercisers are performing much
higher volumes than the 7 sets we required. Some lift-
ers perform 3 to 4 sets of an exercise and may engage
in as many as 3 to 5 exercises for the same muscle
group. This higher volume of exercise may also be per-
formed while the subject is concurrently engaged in
other forms of exercise (e.g., aerobic training), the ef-
fects of which on recovery are largely unexamined. It
should be kept in mind, however, that some subjects
were performing above baseline at 48 hours, and may
be representative of individuals who can tolerate a
higher frequency of higher volume training. In addi-
tion, one subject’s ability to perform repetitions was
still declining at 48 hours, consistent with the findings
of Johansson et al. (22), whose subjects displayed their
lowest force output values 48 hours after 10 sets of 10
eccentric actions.
After 72 hours of recovery, the muscular endurance
of the subjects was still not significantly different from
session 1 values, but 7 individual subjects seemed to
be able to perform more repetitions than during ses-
sion 1. This is similar to the 8 subjects who were able
to perform at or above baseline values by 72 hours in
our previous work using 3 sets of each exercise. But
again, there was no significant difference between the
7-set and 3-set group at any time period (p50.337).
At the 96-hour time point, the group was not sig-
nificantly different from baseline (p50.051), but 8 of
the subjects were at or above their baseline perfor-
mance. Of those 8 subjects, 5 still tended to show in-
creases in performance (may be indicative of a delayed
training effect) and 3 were still performing a number
of repetitions equal to that observed at 72 hours. Of
the 2 subjects that were still below their personal base-
line performance at 96 hours, 1 was still at 72-hour
values, but the other was still rising toward baseline.
An interesting comparison from our 3-sets study is
that for 3 sets, by 96 hours only 3 subjects were still
rising, 6 had either reached a plateau or were detrain-
ing, and 1 subject was not back to the original baseline
values, although the only statistical difference from
baseline was found at 72 hours. Although this rise
above baseline in the 7-sets group did not reach sta-
tistical significance (p50.051), it may be indicative of
a delayed (and possibly longer-lived) training effect,
especially when compared to the significant increase
above baseline at 72 hours (p50.022) found in our 3-
sets study. Either way, it does seem to indicate that the
protocol utilized in this study is sensitive to changes
in training volume.
The findings at 72 and 96 hours are consistent with
previous research showing disruptions in force gen-
eration capacity for up to 3 to 4 days after exercise (4,
23, 36). In addition, Kroon and Naeije (29) used a pro-
tocol involving eccentric exercise to exhaustion with
40% of maximum and found force to be depressed for
7 days. Also, Johansson et al. (22) found performance
to still be decreased at 96 hours after 10 sets of 10
maximal isokinetic eccentric actions. These findings
are also consistent with the commonly used exercise
protocol in which the body is divided into 3 regions,
with each being trained with a relatively high volume
twice per week.
As can be seen clearly in Figure 5, the upper and
lower bodies seemed to respond very similarly during
the testing protocol. The 2 regions were affected to
virtually the same extent at 24 and 48 hours. It is in-
teresting to note, however, that after 72 hours the up-
per body did show a slight trend toward faster recov-
ery than the lower body (although not statistically sig-
nificant). This is in contrast to the findings in the 3-set
study in which the lower body seemed to show a
slightly but not significantly faster recovery after 72
hours. This could merely be due to intra- and inter-
subject variability in our small sample. However, it
could also be that a higher volume of training could
overwhelm the lower body’s ability to recuperate
when compared to performing only 3 sets of each ex-
ercise. In other words, the use of the legs so much
during daily activity may provide them with a greater
ability to recover, as long as the training volume is
below a certain level. Beyond that point, this same dai-
ly usage of the legs could interfere with the recovery
process. In either case it will certainly take studies
with much larger sample sizes to definitely elucidate
differences in recovery that may exist between the up-
per and lower bodies.
Large differences in individual recovery ability
were observed at every data point. This large intra-
individual variability, which was also observed in our
previous study with 3 sets of each exercise, is indica-
tive of a need for individualized recovery testing to
optimize training benefits. Individuals should not be
expected to respond similarly to the same protocol.
Pretesting of lifters before exercise prescription might
help to individualize sometimes broadly prescribed
exercise protocols. We would also point out that this
is not a training study. Further investigation is needed
to determine if pretesting before protocol initiation
does indeed produce superior results.
Again, these results are based upon the specific
population of subjects tested. Also, there was wide
variation in the responses of these subjects, just as
there are wide variations inherent in any testing pro-
tocol from day to day.
Again we retrospectively examined the test orders
for subjects showing similar responses and none of the
subjects showing similar responses was subjected to
similar test sequences.
Future studies in this area should no doubt include
larger sample sizes to fully elucidate differences in re-
covery not only between individuals, but also between
Muscular Endurance Testing
269
Figure 7. Changes in performance (repetitions performed)
for each subject since the previous test period, as repre-
sented by the mean delta score for all 8 lifts combined.
body regions. Clearly, training studies are needed to
test the influence of prolonged recovery intervals on
the rate of strength increase. In addition, studies
should be performed that investigate the differences in
recovery between individuals of varying training
states and utilizing different training volumes. If the
current paradigm shows validity and reliability in fu-
ture studies, it could be a useful protocol for investi-
gating the above areas of interest.
In summary, our investigation provides evidence
that the provided protocol seems to be sensitive to
changes in recovery of muscular endurance brought
about by differences in training volume in experienced
but noncompetitive college-aged men. Consistent with
previous research, after 48 hours subjects do not per-
form significantly differently from baseline values. In
addition, it seems that maximal recovery probably oc-
curs between 72 and 96 hours for these subjects be-
cause of the finding that there is a significant rise
above baseline at 72 hours after 3 sets of an exercise
and a near significant rise above baseline at 96 hours
after 7 sets of exercise. At each recovery period, large
individual variation was observed, indicative of a need
to pretest individuals before exercise initiation to op-
timize benefits.
Practical Applications
The current investigation provides some insight into
the use of a practical protocol for testing of muscular
endurance with individuals engaged in exercise pro-
tocols of higher volume. Many lifters are splitting their
routines into 5 to 6 days and performing a high vol-
ume of exercise for each muscle per week. The proto-
col we used for testing in this study could be used for
individual recovery testing to potentially maximize
training benefit. As training volume increases, recov-
ery time increases concomitantly. If lifters are engaged
in a high enough volume, the chances for underrecov-
ery syndrome increase. To prevent underrecovery, in-
dividualized testing of recovery may be a useful tool.
Experiment 3: Endurance Recovery in
Middle-Aged vs. Younger Men
If there are differences in trainability of older subjects,
information regarding recovery could be very useful
in exercise prescription for this population. Within the
context of this study, it is also important to be sure
that a protocol for testing muscular endurance is sen-
sitive to any population differences that may exist.
Therefore, the purpose of this study was to investigate
the course of muscular endurance recovery in 10
healthy experienced men ages 50 to 65 years vs. the
10 younger subjects (18 to 30 years) from experiment
1 using a common fatiguing stimulus and practical
testing protocol. Experienced recreational resistance
training subjects are of interest because they should
evidence minimal neural effects, minimal aging atro-
phy, and good levels of motivation. If the recovery
curve in older individuals has a different profile from
that we have previously observed in younger subjects,
it would provide evidence that the protocol is sensitive
to changes in age of population and could have im-
portant implications for designing training programs
for the older individual.
Methods
Subjects
Ten healthy men (50 to 65 years of age) were recruited
for the study. With the exception of age, the same sub-
ject requirements as experiment 1 were applied.
Experimental Approach to the Problem
The experimental protocols, design, and statistical
analyses were the same as for experiment 1.
Results
Subjects’ physical characteristics and training history
are displayed in Table 1. All subjects were very co-
operative and compliant. They appeared to provide
maximal effort on each set for each lift. Although some
were reluctant to give up all other exercise, voluntary
compliance with the study requirements appeared to
be good.
Figure 7 depicts the changes in individual perfor-
mance since the previous test session as represented
by the mean delta score for all 8 exercises combined
over the testing period. As can be seen in this figure,
the number of repetitions was significantly lower after
the 24-hour recovery period (mean of 10.1 61.4 rep-
etitions in baseline vs. 8.5 61.8 repetitions at 24 hours,
p50.000). After 48 hours of recovery, the subjects
were no longer significantly different from baseline
values (mean of 10.1 61.4 repetitions in baseline vs.
270 McLester, Bishop, Smith, Wyers, Dale, Kozusko, Richardson, Nevett, and Lomax
Figure 8. Changes in performance (mean 6SD repeti-
tions performed) since the previous test period for upper-
and lower-body exercises. No significant difference (p.
0.05) between body regions. *, significant difference (p,
0.05) from baseline for all 8 lifts combined.
Figure 9. Performance (mean 6SD repetitions per-
formed) comparisons between older and younger men. *,
significant difference (p,0.05) between groups.
9.9 62.2 repetitions at 48 hours, p50.624). At 72
hours of recovery, subjects were still not significantly
different from baseline values (mean of 10.1 61.4 rep-
etitions in baseline vs. 9.9 61.7 repetitions at 72 hours,
p50.273). At 96 hours of recovery, subjects were once
again not significantly different from baseline values
(mean of 10.1 61.4 repetitions in baseline vs. 10 61.8
repetitions at 96 hours, p50.736).
Figure 8 demonstrates the course of performance
since the previous test period as represented by means
and standard deviations of the number of repetitions
performed for the upper- and lower-body lifts. No sig-
nificant interaction of time and body region was ob-
served (p50.094). The 2 regions of the body were
affected to virtually the same degree at 24 hours
(mean of 10.3 61.6 repetitions in baseline vs. 8.7 6
2.1 repetitions at 24 hours for the lower body; mean
of 9.9 61.3 repetitions in baseline vs. 8.4 61.6 repe-
titions at 24 hours for the upper body). After 48 hours
of recovery, the regions of the body were still the same,
with the lower body showing a slightly higher degree
of recovery (mean of 10.3 61.6 repetitions in baseline
vs. 10.4 62.3 repetitions at 48 hours for the lower
body; mean of 9.9 61.3 repetitions in baseline vs. 9.6
62.1 repetitions at 48 hours for the upper body). The
72-hour recovery interval demonstrated the upper and
lower bodies to be close to 48-hour performance (mean
of 10.3 61.6 repetitions in baseline vs. 10.4 61.7 rep-
etitions at 72 hours for the lower body; mean of 9.9 6
1.3 repetitions in baseline vs. 9.6 61.7 repetitions at
72 hours for the upper body). After 96 hours of recov-
ery, the lines were still parallel, with the values being
very similar to 72-hour levels (mean of 10.3 61.6 rep-
etitions in baseline vs. 10.5 61.9 repetitions at 96
hours for the lower body; mean of 9.9 61.3 repetitions
in baseline vs. 9.7 61.7 repetitions at 96 hours for the
upper body).
Soreness ratings at 24 (1.0 61.1), 48 (1.8 61.5), 72
(1.6 61.5), and 96 hours (1.1 61.7) tended to be low
and tended to peak at 48 hours. Perception-of-effort
ratings were very consistent for the 24- (7.0 61.3), 48-
(6.8 61.3), 72- (7.0 61.6), and 96-hour (7.2 61.5)
sessions and also seemed to be slightly higher at 96
hours.
Figure 9 depicts performance (mean 6SD repeti-
tions performed) comparisons between older and
younger men. The young 3-sets group was found to
be performing at a significantly higher level than the
older group only at 72 hours (11.2 62.3 reps in the
younger vs. 9.9 61.7 in the older group, p50.008),
a difference that also approached significance at 96
hours (p50.06).
Discussion
The purpose of this study was to investigate the course
of muscular endurance recovery in 10 healthy men
ages 50 to 65 years vs. 10 younger subjects (18 to 30
years) using a common fatiguing stimulus and prac-
tical testing protocol. This study provides evidence
that this practical protocol for muscular recovery test-
ing is sensitive to changes in individual recovery that
may be attributable to age.
The findings after 24 hours of recovery were to be
expected, with almost all of the subjects below base-
line values at this time point. One subject was slightly
above baseline performance at 24 hours, but it should
be taken into account that these values were the means
of all 8 lifts. Therefore the mean at this point could
possibly have been elevated because of some learning
effect on a few exercises, although this subject was ex-
perienced with all lifts. After 48 hours of recovery only
3 of the 10 subjects were able to perform as many rep-
etitions as at baseline, with 2 of those demonstrating
performance of more repetitions than during baseline
(maybe indicative of a slight training effect). This find-
Muscular Endurance Testing
271
ing is somewhat consistent with our previous research
in subjects 30 years of age and younger in which there
was no significant difference from baseline at 48 hours,
with individual data showing 4 of the subjects per-
forming at or above baseline after 48 hours. The sub-
jects who were unable to perform as many repetitions
at 48 hours when compared to baseline may be rep-
resentative of a segment of the population that cannot
sustain the commonly prescribed every-48-hours
workout schedule. Though many exercisers may not be
training at the intensity required during this study,
many of them are engaged in other forms of exercise
simultaneously, and therefore their total volume of
training may be greater than that of this study.
After 72 hours of recovery, the subjects’ perfor-
mance was not significantly different from baseline. In
addition, the 3 subjects who were performing at or
above baseline at 48 hours evidenced at 72 hours a
decrease in the number of repetitions performed (may-
be indicative of detraining due to too long a recovery
period). As can be seen in Figure 9, these findings are
in contrast to our work with younger subjects (mean
age of 22.6 64.6) in which 80% were performing sig-
nificantly above baseline values at 72 hours. When
compared with the subjects of experiment 1, the dif-
ference in repetitions between the middle-aged sub-
jects and the subjects ages 18 to 30 years was signifi-
cant (p50.008) at 72 hours, even though values at 24
hours (p50.498) and 48 hours (p50.794) were not
significantly different between the younger and older
cohorts. This difference may be indicative of a longer
recovery time in middle-aged subjects. In addition, it
may provide evidence that subjects 50 years and older
may experience a reduced training effect for a given
exercise stimulus compared with subjects 30 years of
age and younger. However, these statements are purely
speculation, as statistical power in this study was an
issue.
For the 96-hour recovery interval, the subjects were
still not significantly different from baseline, with in-
dividual data showing only 1 subject still demonstrat-
ing performance above baseline, 8 back to near base-
line values, and 1 was still not recovered enough to
perform at baseline. There are 2 likely explanations for
the subject not being fully recovered after 96 hours of
rest. It may be that the intensity of training required
for the study may have been a stimulus greater than
the subject’s personal ability to recover. An alternative
explanation is that because the data points are means,
different rates of recovery among various body parts
could mask the total ability to recover. In other words,
some body parts may be fully recovered during 1 ses-
sion while others are not, and then as other exercises
are showing recovery, the first parts to have moved
toward baseline may be declining in performance. In
any case, it clearly implies great subject variability in
the ability to recover. When compared with the youn-
ger subjects from our previous investigation, the
groups were not significantly different at the 96-hour
time point (p50.06), probably because of the large
degree of variability demonstrated by the subjects 30
years and under. As can be seen, statistical power was
most likely the reason for lack of significance in this
case.
As Figure 8 demonstrates, there was no significant
difference between the recovery rates of the upper-
and lower-body muscles. There seemed to be a ten-
dency toward a small difference at 48 hours, with the
lower body tending to recover slightly faster than the
upper body, but this difference was not significant (p
50.094). After 48 hours, the lines representing
strength were parallel up through the 96-hour interval.
Though many researchers believe that the ability of
older subjects to gain strength is intact (1, 5, 9, 10, 24,
34, 39), it could be that there is some decrement (32),
or at least some slowing of the rate of improvement,
as indicated by this study. Our work, even though not
a training study, seems to indicate that there could be
some decrement in training effect for a given stimulus
when compared with younger subjects when testing
with the provided protocol. Also, it could be that the
strength increases in older subjects may take a longer
time period, as evidenced by the greater need for re-
covery time found in this investigation. Strength gains
in older populations have mainly been attributed to
neural effects (30, 34). Our study is somewhat in
agreement with these, as we used experienced subjects
who were likely well past the neural stage of training,
and we observed almost no increase in the number of
repetitions with the exception of only 3 subjects.
The prolonged recovery time evidenced by the sub-
jects in the present study may be due to the previously
mentioned decreased levels of both growth hormone
and testosterone observed in older populations (16, 17,
37, 38) or the blunted response of these hormones to
an exercise bout (12, 18). Though some believe that
older subjects may not be able to train at a high-
enough intensity to stimulate the hormonal spikes ob-
served in younger subjects (14), our subjects seem to
have trained at just as high a relative intensity as the
younger subjects in our previous investigation, as ev-
idenced by a similar decrement in the number of rep-
etitions after 24 hours of recovery.
In addition to the possibly extended recovery time
observed in these subjects, there was also a large de-
gree of intrasubject variability. This same phenomenon
was found in our previous investigation with subjects
30 years old and younger. These large individual dif-
ferences in recovery suggest that our test paradigm
might be helpful to insure proper frequency, especially
in light of the implication that people 50 years old and
over might need longer recovery intervals.
The sensitivity and reliability of our testing para-
digm has not been established. Whether or not test-
272 McLester, Bishop, Smith, Wyers, Dale, Kozusko, Richardson, Nevett, and Lomax
retest reliability for individuals exists, the group mean
data seem to show that the protocol is sensitive to
qualitative differences in the responses of middle-aged
and younger weight lifters.
Again, these results are based upon the specific
population of subjects tested. Also, there was wide
variation in the responses of these subjects, just as
there are wide variations inherent in any testing pro-
tocol from day to day.
Also, though the physical characteristics of the sub-
jects in this study were not statistically different from
those in the younger group (other than age) with
which we are making comparisons, it is obvious that
the younger subjects were different in weight (75.6 6
10.8 kg in younger vs. 88.6 617.4 kg in older) and
body fat percentage (10.5 63.4% in younger vs. 23.9
66.0% in older). These physical differences are pos-
sible sources of difference in recovery. For example,
being relatively fatter may be indicative of a difference
in fitness status that could translate into an inability
to recover as quickly as someone in better physical
shape. Also, in Table 1 it can be seen that the older
subjects may have been accustomed to a slightly dif-
ferent exercise routine. Again these differences could
affect the person’s training state and therefore his abil-
ity to recover from the 3-set protocol that was required
for the study. However, we would argue that the phys-
ical difference between these groups is the basis of
comparing them.
Again we retrospectively examined the test orders
for subjects showing similar responses and none of the
subjects showing similar responses was subjected to
similar test sequences.
Future studies should be directed at investigating
this population of individuals using a larger sample
size, and subjects of even older age could be compared
with groups similar to those in our research. In ad-
dition, training studies using experienced older sub-
jects could help to elucidate whether or not there is a
decreased ability to gain strength as one increases in
age or if the progression of strength is just slower.
This study provides evidence that this practical
protocol for muscular recovery testing is sensitive to
changes in individual recovery that may be attribut-
able to age in experienced 50- to 65-year-old men. In
addition, because of large individual variation in re-
covery observed in these subjects, testing of subjects’
recovery time before exercise protocol initiation may
be warranted to provide the most effective frequency
prescription.
Practical Applications
This study provides information on a practical proto-
col for testing muscular endurance recovery from a
single resistive exercise bout in subjects 50 to 65 years
of age. Older exercisers may need an extended recov-
ery interval relative to younger lifters after a given
bout of exercise. The protocol implemented in this
study appears to be a safe and practical tool that may
be useful before exercise prescription to help deter-
mine optimal exercise frequency.
References
1. A
NIANSSON
,A.,P.L
JUNGBERG
,A.R
UNDGREN
,
AND
H. W
ET
-
TERQVIST
. Effect of a training program for pensioners on con-
dition and muscular strength. Arch. Gerontol. Geriatr. 3:229–241.
1984.
2. A
RMSTRONG
, R.B. Mechanisms of exercise-induced delayed on-
set muscle soreness: A brief review. Med. Sci. Sports Exerc. 16:
529–538. 1984.
3. B
RAITH
, R.W., J.E. G
RAVES
, M.L. P
OLLOCK
, S.L. L
EGGETT
, D.M.
C
ARPENTER
,
AND
A.B. C
OLVIN
. Comparison of 2 vs 3 days/
week of variable resistance training during 10- and 18-week
programs. Int. J. Sports Med. 10:450–454. 1989.
4. B
ROWN
, J., R.B. C
HILD
, A.E. D
ONNELLY
, J.M. S
AXON
,
AND
S.H.
D
AY
. Changes in human skeletal muscle contractile function
following stimulated eccentric exercise. J. Appl. Physiol. 72:515–
521. 1996.
5. C
HARETTE
, S.L., L. M
C
E
VOY
,G.P
YKA
,C.S
NOW
-H
ARTER
,D.
G
UIDO
, R.A. W
ISWELL
,
AND
R. M
ARCUS
. Muscle hypertrophy
response to resistance training in older women. J. Appl. Physiol.
70:1912–1916. 1991.
6. D
AVIES
, C.T.M.,
AND
M.J. W
HITE
. Muscle weakness following
eccentric work in man. Pflugers Arch. 392:168–171. 1981.
7. E
STON
, R.G., S. F
INNEY
,S.B
AKER
,
AND
V. B
ALTZOPOULOS
. Mus-
cle tenderness and peak torque changes after downhill running
following a prior bout of isokinetic eccentric exercise. J. of Sports
Sci. 14:291–299. 1996.
8. E
VANS
, D.T., L.L. S
MITH
, T.C. C
HENIER
, R.G. I
SRAEL
,
AND
M.R.
M
C
C
AMMON
. Changes in peak torque, limb volume, and de-
layed onset muscle soreness following repetitive eccentric con-
tractions [Abstract]. Int. J. Sports Med. 11:403. 1990.
9. F
RONTERA
, W.R., V.A. H
UGHES
, K.J. L
UTZ
,
AND
W. J . E
VANS
.A
cross-sectional study of muscle strength and mass in 48- to 78-
yr-old men and women. J. Appl. Physiol. 71:644–650. 1991.
10. F
RONTERA
, W.R., C.N. M
EREDITH
, K.P. O’R
EILLY
, H.G. K
NUTT
-
GEN
,
AND
W. J . E
VANS
. Strength conditioning in older men: Skel-
etal muscle hypertrophy and improved function. J. Appl. Phy-
siol. 64:1038–1044. 1988.
11. G
ILLAM
, G.M. Effects of frequency of weight training on muscle
strength enhancement. J. Sports Med. 21:432–436. 1981.
12. H
AGBERG
, J.M., D.R. S
EALS
, J.E. Y
ERG
,J.G
AVIN
,R.G
INGERICH
,
B. P
REMACHANDRA
,
AND
J.O. H
OLLOSZY
. Metabolic responses
to exercise in young and older athletes and sedentary men. J.
Appl. Phys. 65:900–908. 1988.
13. H
AKKINEN
, K. Neuromuscular fatigue and recovery in male
and female athletes during heavy resistance exercise. Int. J.
Sports Med. 14:53–59. 1993.
14. H
AKKINEN
,K.,
AND
A. H
AKKINEN
. Muscle cross-sectional area,
force production and relaxation characteristics in women at dif-
ferent ages. Eur. J. Appl. Physiol. 62:410–414. 1991.
15. H
AKKINEN
, K., H. K
AUHANEN
,
AND
P. V. K
OMI
. Effects of fa-
tiguing loading with variable resistance equipment on neural
activation and force production of the knee extensor muscles.
Electromyogr. Clin. Neurophysiol. 28:79–87. 1988.
16. H
AKKINEN
,K.,
AND
A. P
AKARINEN
. Acute hormone responses
to heavy resistance exercise in men and women at different
ages. Int. J. Sports Med. 16:507–513. 1995.
17. H
AKKINEN
, K., A. P
AKARINEN
,M.A
LEN
,H.K
AUHANEN
,
AND
P. V. K
OMI
. Neuromuscular and hormonal responses in elite ath-
letes to two successive strength training sessions in one day.
Eur. J. Appl. Physiol. 57:133–139. 1988.
Muscular Endurance Testing
273
18. H
AKKINEN
, K., A. P
AKARINEN
, R.U. N
EWTON
,
AND
W. J . K
RAE
-
MER
. Acute hormone responses to heavy resistance lower and
upper extremity exercise in young versus old men. Eur. J. Appl.
Physiol. 77:312–319. 1998.
19. H
ITCHCOCK
, C. Recovery of short-term power after dynamic
exercise. J. Appl. Physiol. 67:677–681. 1989.
20. H
OUGH
, T. Ergographic studies in muscular soreness. Am. J.
Physiol. 7:76–92. 1902.
21. H
UNTER
, G.R. Changes in body composition, body build and
performance associated with different weight training frequen-
cies in males and females. Natl. Strength Cond. Assoc. J. 7:26–28.
1985.
22. J
OHANSSON
,P.H.,L.L
INDSTROM
,G.S
UNDELIN
,
AND
B. L
IND
-
STROM
. The effects of preexercise stretching on muscular sore-
ness and force loss following heavy eccentric exercise. Scand. J.
Med. Sci. Sports 9:219–225. 1999.
23. J
ONES
, D.A., D.J. N
EWHAM
,
AND
C. T
ORGAN
. Mechanical influ-
ences on long-lasting human muscle fatigue and delayed-onset
pain. J. Physiol. 412:415–427. 1989.
24. K
AUFFMAN
, T.L. Strength training effect in young and aged
women. Arch. Phys. Med. Rehabil. 66:223–226. 1985.
25. K
ENNEY
, W.L. (ed.) American College of Sports Medicine Guidelines
for Exercise Testing and Prescription (5th ed.). Media, PA: Wil-
liams & Wilkins, 1995. pp. 56–57.
26. K
OMI
,P.V.,
AND
E.R. B
USKIRK
. Effect of eccentric and concentric
muscle conditioning on tension and electrical activity of human
muscle. Ergonomics 15:417–434. 1972.
27. K
OMI
,P.V.,
AND
J.T. V
IITASALO
. Changes in motor unit activity
and metabolism in human skeletal muscle during and after
repeated eccentric and concentric contractions. Acta Physiol.
Scand. 100:246–254. 1977.
28. K
RAEMER
, W., S. G
ORDON
,S.F
LECK
, L.J. M
ARCHITELLI
,R.M
EL
-
LO
, J.E. D
ZIADOS
,K.F
RIEDL
,E.H
ARMAN
,C.M
ARESH
,
AND
A.C.
F
RY
. Endogenous anabolic hormonal and growth factor re-
sponses to heavy resistance exercise in males and females. Int.
J. Sports Med. 12:228–235. 1991.
29. K
ROON
,G.W.,
AND
M. N
AEIJE
. Recovery of human biceps elec-
tromyogram after heavy eccentric, concentric or isometric ex-
ercise. Eur. J. Appl. Physiol. 63:444–448. 1991.
30. L
ARSSON
, L. Physical training effects on muscle morphology in
sedentary males at different ages. Med. Sci. Sports Exerc. 14:203–
206. 1982.
31. L
EXELL
, J., C.C. T
AYLOR
,
AND
M. S
JOSTROM
. What is the cause
of aging atrophy? Total number, size and proportion of differ-
ent fiber types studied in whole vastus lateralis muscle from
15- to 83-year-old men. J. Neurol. Sci. 84:275–294. 1988.
32. L
IEMOHN
, W.P. Strength and aging: An exploratory study. Int.
J. Aging Hum. Dev. 6:347–357. 1975.
33. M
C
L
ESTER
,J.R.,J
R
., P. B
ISHOP
,
AND
M.E. G
UILLIAMS
. Compari-
son of 1 day and 3 days per week of equal-volume resistance
training in experienced subjects. J. Strength Cond. Res. 14:273–
281. 2000.
34. M
ORITANI
,T.,
AND
H.A.
DE
V
RIES
. Potential for gross muscle
hypertrophy in older men. J. Gerontol. 35:672–682. 1980.
35. N
EWHAM
, D.J., K.R. M
ILLS
, B.M. Q
UIGLEY
,
AND
R.H.T. E
D
-
WARDS
. Pain and fatigue after concentric and eccentric muscle
contractions. Clin. Sci. 64:45–52. 1983.
36. P
IERRYNOWSKI
, M.R., P.M. T
UDUS
,
AND
M.J. P
LYLEY
. Effects of
downhill or uphill training prior to a downhill run. Eur. J. Appl.
Physiol. 56:668–672. 1987.
37. R
UDMAN
, D., M.H. K
UTNER
, C.M. R
OGERS
, M.F. L
UBIN
, G.A.
F
LEMING
,
AND
R.P. B
AIN
. Impaired growth hormone secretion
in the adult population: relation to age and adiposity. J. Clin.
Invest. 67:1361. 1981.
38. S
IMONSON
, E., W.M. K
EARNS
,
AND
N. E
NZER
. Effect of methyl-
testosterone treatment on muscular performance and the cen-
tral nervous system of older men. J. Clin. Endocrinol. 4:528–534.
1944.
39. S
IPILIA
, S., M. E
LORINNE
,M.A
LEN
,H.S
UOMINEN
,
AND
V. K
O
-
VANEN
. Effects of strength and endurance training on muscle
fibre characteristics in elderly women. Clin. Physiol. 17:459–474.
1997.
40. T
IIDUS
, P.M.,
AND
C.D. I
ANUZZO
. Effects of intensity and du-
ration of muscular exercise on delayed soreness and serum en-
zyme activities. Med. Sci. Sports Exerc. 15:461–465. 1983.
41. Y
ATES
,J.W.,
AND
W. J . A
RMBRUSTER
. Concentric and eccentric
strength loss and recovery following exercise-induced muscle
soreness [Abstract]. Int. J. Sports Med. 11:403. 1990.
Address correspondence to John R. McLester, Jr.,
john.mclester@wku.edu.
