Content uploaded by Yen-Po Huang
Author content
All content in this area was uploaded by Yen-Po Huang on Jun 02, 2015
Content may be subject to copyright.
Journal of Medical and Biological Engineering, 31(4): 289-293
289
Effect of Push-up Speed on Upper Extremity Training until
Fatigue
Hsiu-Hao Hsu1 You-Li Chou2 Yen-Po Huang1
Ming-Jer Huang1,3 Shu-Zon Lou4 Paul Pei-Hsi Chou5,6,7,*
1Department of Engineering Science, National Cheng-Kung University, Tainan 701, Taiwan, ROC
2Institute of Biomedical Engineering, National Cheng-Kung University, Tainan 701, Taiwan, ROC
3Department of Logistics and Technology Management, Leader University, Tainan 709, Taiwan, ROC
4School of Occupational Therapy, Chung Shan Medical University, Taichung 402, Taiwan, ROC
5Faculty of Sports Medicine, Kaohsiung Medical University, Kaohsiung 807, Taiwan, ROC
6Department of Orthopedic Surgery, Kaohsiung Medical University Hospital, Kaohsiung 807, Taiwan, ROC
7Department of Orthopedic Surgery, Kaohsiung Municipal Hsiao-Kang Hospital, Kaohsiung 812, Taiwan, ROC
Received 9 Sep 2010; Accepted 17 Nov 2010; doi: 10.5405/jmbe.844
Abstract
Push-up exercises are commonly performed to strengthen the upper extremity muscles. However, the relationship
between the push-up speed and upper extremity fatigue is not well understood. Accordingly, the present study
investigated the effect of the push-up speed on the maximum possible number of push-up repetitions until fatigue and
the upper-extremity muscle activity, respectively, in order to identify suitable push-up strategies for upper-extremity
muscular strengthening. Fifteen healthy males participated in the study. Each subject performed push-ups at three
different speeds (i.e., fast: 7 push-ups/10 s; regular: 5 push-ups/10 s; and slow: 4 push-ups/10 s) until fatigued. The
muscle activity signals were measured during the push-up tests via surface electromyography. The strengthening effect
of the push-up exercises was evaluated by measuring the myodynamic decline rate at the shoulder, elbow and wrist
joints using an isokinetic dynamometer. The results showed that the maximum possible number of push-up repetitions
at the fast push-up speed was around 1.34 and 1.33 times higher than that at the regular push-up speed or slow push-up
speed, respectively. However, the endurance time (i.e., the time to fatigue) at the slow push-up speed was around 1.20
and 1.24 times longer than that at the fast push-up speed or regular push-up speed, respectively. Finally, at the slow
push-up speed, the total muscle activations in the triceps brachii, biceps brachii, anterior deltoid, pectoralis major, and
posterior deltoid, respectively, were 1.47, 2.43, 1.42, 1.48, and 1.91 times higher than those at the fast push-up speed.
Therefore, the experimental results suggest that push-ups should be performed at a faster speed when the aim is to
achieve a certain number of repetitions, but should be performed at a slower speed when the aim is to strengthen the
upper extremity muscles.
Keywords: Push-up, Upper extremity, Electromyography (EMG), Isokinetic dynamometer, Muscular strengthening
1. Introduction
Push-up exercises are convenient, easily learned, and
readily adapted to various levels of difficulty. As a result, they
are commonly performed by health-conscious individuals and
athletes to strengthen the upper extremity muscles [1]. When
performing upper extremity movements, stability of the joints
is ensured not only by the surrounding tissue (e.g., the
ligaments and capsules), but also by the muscular contraction
strength. As a result, maintaining and improving the muscular
* Corresponding author: Paul Pei-Hsi Chou
Tel: 886-7-3208209; Fax: 886-7-3119544
E-mail: pc.arthroscopy@gmail.com
strength is essential in enhancing an individual’s performance
ability and preventing movement-related injuries. Of all the
training exercises available for the upper extremity, push-ups
are among the most common since they yield a notable
improvement in both the muscle strength and the muscle
endurance.
Many studies have established biomechanical kinematic
and kinetic models of the upper extremity [2-6]. Furthermore,
the effects of different types of push-ups on the degree of
muscle activation have also been reported. For example, a
narrow base position results in significantly higher
electromyography (EMG) activities of the pectoralis major and
triceps brachii muscle groups than a wide base position [7].
Similarly, the pectoralis major muscle activation in posterior
J. Med. Biol. Eng., Vol. 31 No. 4 2011
290
push-ups is higher than normal, whereas the triceps muscle
activation is lower than normal [8]. However, the correlation
between the push-up speed and the strengthening effect of
push-up exercises is not yet clear. Therefore, the implications
of the push-up speed on the muscular performance and the
maximum possible number of repetitions are also not fully
understood. Accordingly, this study investigated the effect of
the push-up speed (fast, regular and slow) on the maximum
possible number of repetitions, the endurance time, the
upper-extremity muscle activation, and the myodynamic
decline rate. The myodynamic decline rate in different
isometric test conditions was measured using an isokinetic
dynamometer and the muscle activity at different push-up
speeds is measured via surface electromyography. The study
provides an insight into the different usage mechanisms of the
muscle groups when performing push-ups at different speeds
and enables the identification of appropriate push-up strategies
for upper extremity training.
2. Materials and methods
2.1 Participants and experimental protocol
Fifteen physically healthy male students participated in the
investigation. The subjects ranged from 22 to 27 yrs of age
(24.27 ± 1.22 yrs), 60 to 84 kg in weight (72.47 ± 5.93 kg), and
170 to 180 cm in height (174.67 ± 2.87 cm). The BMI of the
participants ranged from 20 to 26 kg/m2 (23.7 ± 1.8 kg/m2). All
of the participants were right-hand dominant and free of
upper-extremity disorders.
The effect of the push-up speed on the myodynamic (i.e.,
muscle strength) decline rate was examined by measuring the
torque at the shoulder, elbow and wrist joints before and after
the push-up exercises using an isokinetic dynamometer (Kin
Com KC125AF, Kin Com Isokinetic International Corp.,
Harrison, TN). As shown in Fig. 1, each subject was asked to
perform ten isometric tests, namely shoulder extension (SE),
shoulder flexion (SF), shoulder abduction (SAB), shoulder
adduction (SAD), shoulder external rotation (SRE), shoulder
internal rotation (SRI), elbow extension (EE), elbow flexion
(EF), forearm supination (FS) and forearm pronation (FP). In
each case, the myodynamic decline rate was calculated as
(Tpre–Tpost)/Tpre, where Tpre and Tpost are the measured torque
values before and after the push-up test, respectively.
The muscle activity signals at the different push-up speeds
were measured using a surface electromyography (sEMG)
system (MA300, Motion Analysis Corp.) at a sampling rate of
1000 Hz. For each subject, EMG sensors were attached to the
supinator, pronator teres, triceps brachii, middle deltoid, biceps
brachii, anterior deltoid, pectoralis major, posterior deltoid,
infraspinatus and teres minor muscle groups [9,10]. Having
attached the EMG electrodes, the subjects performed a series of
3-second maximum voluntary isometric contractions (MVIC)
of the relative muscle group in order to obtain a datum with
which to normalize the EMG data acquired during the push-up
tests [10]. The raw sEMG data collected during the tests were
exported to Matlab (Mathworks Inc., Natick, MA, USA) for
Shoulder
Extension
Shoulder
Flexion
Shoulder
Abduction
Shoulder
Adduction Shoulder
Internal Rotation
Shoulder
External Rotation
Elbow
Extension
Elbow
Flexion
Wrist
Supination Wrist
Pronation
Figure 1. Isometric test conditions used to evaluate rate of myodynamic
decline following push-up tests.
further analysis and processing. The data were initially rectified
by converting the negative voltage signals to positive signals,
and a linear envelope was then used to estimate the volume of
the muscle activation. The sEMG data were divided by the
corresponding MVIC value in order to obtain a normalized
MVIC value (%MVIC) in the range 0~100% [10,11]. It should
be noted that the actual muscle activation during the push-up
exercises was determined from the vertical displacement
history of a reflexive marker attached to the 4th thoracic
vertebrae rather than from the EMG data. In addition, the
duration over which the volume of muscle activation was
evaluated in this study was defined as the total duration of the
push-up test (i.e., from the start of the test until the point of
fatigue). The total muscle activation (TMA) in each push-up
test was computed as
0
()
TMA ( ) 100%
TEMG t
EMG t dt
MVIC
(1)
where T is the total duration of the test.
Before starting the push-up tests, the subjects were asked
to extend their elbows fully and to position both hands in a
forearm axially non-rotated posture. The hand width was set to
1.5 times the shoulder width and the feet were set to one
shoulder-width apart. The subjects were asked to perform
push-ups at three different speeds, namely fast, regular and
slow. In every case, the up and down stages of the push-up
cycle were indicated audibly by an electronic metronome. For
the fast push-up repetitions, the metronome was set to 84 beats
per minute (bpm), i.e., 42 cycles per minute (equivalent to
7 push-ups/10s). Meanwhile, for the regular and slow push-up
repetitions, the metronome beat was set to 60 bpm
(5 push-ups/10 s) and 48 bpm (4 push-ups/10 s), respectively.
The investigation commenced with the fast push-up tests. The
subjects were instructed to perform push-ups for 15 seconds in
accordance with the instructed cadence. After 15 seconds, the
subjects were told to wait for around 5 seconds to allow for
data recording, and were then requested to repeat the same
procedure (i.e., push-ups for 15 seconds followed by a 5 second
pause) until they were completely fatigued, i.e., they had
completely exhausted their energy and stamina, and were
Push-up Speed on Upper Extremity Strengthening
291
physically unable to perform any more repetitions. Following a
rest period of two weeks, the experimental procedure was
repeated at the regular push-up speed. Finally, following a
further two-week rest period, the experimental procedure was
repeated once again at the slow push-up speed.
2.2 Statistical analysis
The number of push-up repetitions, the endurance time, the
myodynamic decline data, and the sEMG data were analyzed
using SPSS statistical software (SPSS Inc., Chicago, Illinois,
USA). In addition, the myodynamic decline data and sEMG data
were analyzed via repeated-measure one-way analysis of
variance (rmANOVA) tests using a significance level of
P < 0.05. In performing the tests, the push-up speed was treated
as the independent variable and the myodynamic decline rate
and the TMA were treated as dependent variables. A post-hoc
analysis of the effect of the push-up speed on the dependent
variables was performed using the Bonferroni method.
3. Results
3.1 Total number of push-up repetitions and endurance time
In performing the push-up tests, the subjects were asked
to try and keep up with the designated push-up speed as best as
they could, even as they became tired. The average cycle times
of the fast, regular and slow push-ups were found to be
1.67 ± 0.14 s, 2.14 ± 0.09 s and 2.63 ± 0.07 s, respectively.
Even though the average cycle time was longer than the
instructed cadence as a result of the subjects becoming tired, a
significant difference existed in the average cycle times of the
tests performed at the three different push-up speeds. Table 1
presents the statistical results for the maximum number of
push-ups before fatigue and the endurance time at each of the
three push-up speeds. As shown, a significant difference
(P = 0.012) existed in the maximum number of push-ups
performed at the three different speeds. In addition, it is
observed that the maximum number of push-ups was obtained
at the highest push-up speed. Finally, it is seen that the
endurance time at the slow push-up speed (101.2 s) was
significantly longer (p = 0.038) than that at the fast or regular
push-up speed.
Table 1. Maximum number of push-up repetitions and endurance time
for push-ups performed at various speeds until fatigue.
Push-up
speed
Fast¶
mean (SD)
Regular†
mean (SD)
Slow‡
mean (SD)
P§
Post hoc
Number of
times
51.3 (13.9)
38.2 (8.5)
38.6 (7.5)
0.012*
F>R, S
duration
time (sec)
84.2 (17.3)
81.3 (16.7)
101.2 (18.9)
0.038*
S>F, R
¶ F: fast push-up speed
† R: regular push-up speed
‡ S: slow push-up speed
§ P value is significance of one-way ANOVA.
* Significant differences (P < 0.05) among three push-up speeds.
3.2 Effect of push-up speed on myodynamic decline rate
Table 2 shows the myodynamic decline rate for each of
the ten isometric conditions following completion of the fast,
regular and slow push-up tests, respectively. The results show
that a myodynamic decline of more than 45% occurred in the
SE, SF, SAB, SAD, EE and EF isometric tests. However, for a
given isometric test condition, there was no significant
difference in the myodynamic decline rate among the three
different push-up speeds.
Table 2. Rate of myodynamic decline following push-ups performed at
various speeds until fatigue.
Push-up
speed
Fast
Regular
Slow
P§
Mean decline rate
Mean decline
rate
Mean decline
rate
Shoulder
SE
50.62%
48.61%
51.11%
0.933
SF
49.27%
40.49%
47.30%
0.399
SAB
47.85%
48.65%
51.95%
0.721
SAD
50.86%
48.78%
49.92%
0.943
SRE
39.12%
42.28%
41.56%
0.812
SRI
37.74%
42.29%
40.36%
0.687
Elbow
EE
44.27%
43.07%
44.69%
0.938
EF
46.48%
40.94%
44.52%
0.549
Forearm
FS
37.75%
35.15%
38.21%
0.730
FP
35.01%
34.42%
37.17%
0.845
P§
0.007**
0.015*
0.236
Post hoc
SE > SRE, SRI, FS, FP
SF > SRI, FS, FP
SAB > FP
SAD > SRE, SRI, FS, FP
EF > FP
SE > FS, FP
SAB > FS, FP
SAD > FS, FP
SE > FP
SAB > FS, FP
§ P value shows significance by one-way ANOVA.
* Significant differences (P < 0.05) among ten isometric tests.
** Significant differences (P < 0.01) among ten isometric tests.
3.3 Va riation in myodynamic decline rate among different
isometric test conditions
Table 2 shows that for each push-up speed, a significant
difference existed in the myodynamic decline rates associated
with the different isometric conditions (i.e., P = 0.007,
0.015 and 0.236 for the fast, regular and slow push-up speeds,
respectively).
3.4 Muscle activity
Table 3 presents the TMA results for each of the
10 muscle groups over the full duration of the fast, regular and
slow push-up tests, respectively. It can be seen that for all
muscle groups, the TMA in the slow push-up tests was
significantly higher than that in the regular push-up tests or
fast push-up tests. The higher TMA was particularly apparent
in the biceps brachii, triceps brachii, (P < 0.05), anterior
deltoid, posterior deltoid, and posterior deltoid muscle groups
(P = 0.053~0.058).
4. Discussion
The experimental results presented in this study show that
push-ups have a significant effect on the upper-extremity
strengthening process. Table 2 shows that a myodynamic
decline occurred in each of the considered isometric test
conditions following the push-up exercises. However, for a
given isometric condition, the push-up speed had no significant
J. Med. Biol. Eng., Vol. 31 No. 4 2011
292
Table 3. Total muscle activation (TMA) over whole push-up cycle for push-ups performed at various speeds.
Push-up Speed
Fast¶
Mean (SD)
Regular†
Mean (SD)
Slow‡
Mean (SD)
P§
Post hoc
Supinator
1261.56 (752.15)
1149.39 (619.87)
1641.85 (1075.29)
0.425
Pronator teres
941.04 (367.58)
1119.98 (749.27)
1530.81 (858.11)
0.200
Triceps brachii
2138.91 (775.92)
2038.74 (526.01)
3145.29 (1044.76)
0.012**
S>F, R
Middle deltoid
1243.85 (535.12)
1568.75 (818.10)
2205.96 (1191.52)
0.104
Biceps brachii
714.37 (288.00)
806.07 (692.50)
1732.77 (775.09)
0.006**
S>F, R
Anterior deltoid
1612.20 (730.92)
1636.69 (449.21)
2295.36 (707.65)
0.053
Pectoralis major
2114.23 (814.05)
2249.87 (968.03)
3121.81 (988.68)
0.054
Posterior deltoid
1159.48 (517.40)
1378.25 (584.92)
2217.42 (1399.48)
0.058
Infraspinatus
1216.83 (691.58)
1381.67 (926.22)
1973.02 (1018.66)
0.207
Teres minor
1381.88 (994.21)
1619.22 (722.59)
2431.47 (1065.04)
0.055
unit: %MVIC·sec
¶ F: fast push-up speed
† R: regular push-up speed
‡ S: slow push-up speed
§ P value shows significance by one-way ANOVA.
** Significant differences (P < 0.01) among three push-up speed.
effect on the myodynamic decline rate. This result is to be
expected since the myodynamic decline rate was measured
once the subjects were completely fatigued, irrespective of the
speed at which the repetitions were performed.
However, for a given push-up speed, the myodynamic
decline rates measured under the different isometric conditions
were significantly different. As shown in the lower row of
Table 2, the difference in the myodynamic decline rate among
the different isometric conditions was more significant
following the push-up tests performed at a fast speed
(P = 0.007) than following the tests performed at the regular
speed (P = 0.015) or the slow speed (P = 0.236). In other
words, the difference in the effort exerted by the different upper
extremity muscle groups increased as the push-up speed
increased, but reduced as the push-up speed reduced. This
finding can be explained by considering the effect of the
push-up speed on the different usage of the muscle groups.
As shown in Fig. 2, the peak EMG activity of the triceps
brachii muscle group occurred at the lowest position of the
push-up cycle at all three push-up speeds. However, the peaks
in the EMG curve obtained in the slow push-up test were lower
and broader than those in the curves obtained in the fast
push-up test. In the fast push-up tests, the muscle groups did
not need to support the body weight for a prolonged period of
time during the “descending” stage and “ascending” stages.
Instead, they were used predominantly to control the
deceleration of the body at the end of the “descending” stage
and to control the acceleration of the body at the beginning of
the “ascending stage”. By contrast, in the slow push-up tests,
the muscle groups were required to drive the body at a more
consistent speed throughout the entire push-up cycle. In
general, the change in acceleration when switching from the
“descending” stage of the repetition to the “ascending” stage is
accomplished using a subset of the upper extremity muscle
groups, i.e., the posterior deltoid, anterior deltoid, middle
deltoid, pectoralis major, triceps brachii and biceps brachii
[9,10]. However, supporting the body weight over the entire
push-up cycle involves all of the muscle groups. As a result, the
difference in the myodynamic decline rate observed under
different isometric conditions was more noticeable following
the high-speed tests than after the regular or slow-speed tests.
Table 2 shows that the largest myodynamic decline rates
occurred in the SE, SF, SAB, SAD, EE, and EF isometric test
conditions, which involved the posterior deltoid, anterior
deltoid, middle deltoid, pectoralis major, triceps brachii and
biceps brachii muscle groups, respectively [12]. These muscle
groups correspond exactly with those groups responsible for
controlling the change in acceleration during the push-up cycle.
Therefore, it can be inferred that irrespective of the speed at
which the push-ups are performed, the process of controlling
the change in acceleration of the body weight is responsible for
most of the energy consumed in each push-up cycle.
Figure 2. Mean normalized EMG activity of triceps brachii during a
single push-up cycle.
Table 1 shows that the maximum number of repetitions at
a fast push-up speed was significant higher than that at a
regular push-up speed or slow push-up speed. However,
Table 3 shows that a faster push-up speed did not result in a
greater muscle activation. Among the three push-up speeds, the
slow push-up speed resulted in a significantly larger TMA than
the regular or fast push-up speed. Since, a similar number of
repetitions were performed at the slow and regular push-up
speeds, respectively, the larger TMA at a slow push-up speed
Push-up Speed on Upper Extremity Strengthening
293
was most likely the result of a longer endurance time. That is,
the subjects spent a greater amount of time supporting their
body weight prior to fatigue when performing the push-ups at a
slow speed than when performing the push-ups at a regular
speed.
5. Conclusions
This study examined the effect of the push-up speed (fast,
regular and slow) on the myodynamic decline rate and
activation of the upper extremity muscle groups. At a fast
push-up speed, the maximum number of push-up repetitions
prior to fatigue was found to be 1.34 and 1.33 times higher than
that at a regular push-up speed or slow push-up speed,
respectively. However, the endurance time (i.e. the time to
fatigue) at a slow push-up speed was around 1.20 and
1.24 times longer than that at a fast push-up speed or regular
push-up speed, respectively. In addition, at a slow push-up
speed, the TMAs of the triceps brachii, biceps brachii, anterior
deltoid, pectoralis major, and posterior deltoid muscle groups
were 1.47, 2.43, 1.42, 1.48, and 1.91 times higher than those at
a fast push-up speed, respectively. Finally, the myodynamic
decline rate of the upper extremity muscles was found to be
independent of the push-up speed. Overall, the results suggest
that a slow push-up speed delays the occurrence of fatigue and
increases the muscle activation. By contrast, a fast push-up
speed increases the maximum number of push-up repetitions,
but reduces the muscle activation. Accordingly, the present
findings suggest two different push-up strategies. That is, when
a certain number of push-up repetitions are to be performed
(e.g., as part of military training), the repetitions should be
performed at a faster speed since this requires a lower muscle
activation and less effort. Conversely, when the aim is to
develop upper-body strength (e.g., in athletic training), the
push-ups should be performed more slowly since this increases
the muscle activation.
Acknowledgement
This study was supported by the National Science Council
of Taiwan under Contract No. NSC 96-221-E-037-002-MY3.
References
[1] K. N. An, E. Y. Chao, B. F. Morrey and M. J. Donkers,
“Intersegmental elbow joint load during pushup,” Biomed. Sci.
Instrum., 28: 69-74, 1992.
[2] K. N. An, S. L. Korinek, T. Kilpela and S. Edis, “Kinematic and
kinetic analysis of push-up exercise,” Biomed. Sci. Instrum., 26:
53-57, 1990.
[3] S. Z. Lou, C. J. Lin, P. H. Chou, Y. L. Chou and F. C. Su,
“Elbow load during pushup at various forearm rotations,” Clin.
Biomech., 16: 408-414, 2001.
[4] P. H. Chou, C. J. Lin, Y. L. Chou, S. Z. Lou, F. C. Su and G. F.
Huang, “Elbow load with various forearm positions during
one-handed pushup exercise,” Int. J. Sports Med., 23: 457-462,
2002.
[5] P. H. Chou, S. Z. Lou, S. K. Chen, H. C. Chen, T. H. Hsia, T. L.
Liao and Y. L. Chou, “Elbow load during different types of
bench-press exercise,” Biomed. Eng. Appl. Basis Commun., 20:
185-189, 2008.
[6] P. H. Chou, H. H. Hsu, S. K. Chen, S. K. Yang, C. M. Kuo and Y.
L. Chou, “Effect of push-up speed on elbow joint loading,” J.
Med. Biol. Eng., 31: 161-168, 2011.
[7] R. M. Cogley, T. A. Archambault, J. F. Fibeger, M. M.
Koverman, J. W. Youdas and J. H. Hollman, “Comparison of
muscle activation using various hand positions during the
push-up exercise,” J. Strength Cond. Res., 19: 628-633, 2005.
[8] M. K. Gouvali and K. Boudolos, “Dynamic and
electromyographical analysis in variants of push-up exercise,” J.
Strength Cond. Res., 19: 146-151, 2005.
[9] E. F. Delagi, A. Perotto, J. Iazzetti and D. Morrison (Eds.),
Anatomical Guide for the Electromyographer: The Limbs, New
York: Charles C. Thomas, 1981.
[10] J. R. Cram, G. S. Kasman and J. Holtz (Eds.), Introduction to
Surface Electromyography, Gaithersburg, Maryland: Aspen
Publishers, 1998.
[11] D. G. E. Robertson (Ed.), Research Methods in Biomechanics,
Champaign, IL: Human Kinetics, 2004.
[12] Chattanooga Group, Inc., The Balance System: Clinical Desk
Reference, Chattanooga, TN: Chattecx Corporation, 1992.
J. Med. Biol. Eng., Vol. 31 No. 4 2011
294