CHEST PRESS EXERCISES WITH DIFFERENT STABILITY
REQUIREMENTS RESULT IN SIMILAR MUSCLE DAMAGE
RECOVERY IN RESISTANCE-TRAINED MEN
DIOGO V. FERREIRA,1 JOÃO B. FERREIRA-JÚNIOR,2 SAULO R.S. SOARES,1 EDUARDO L. CADORE,3
MIKEL IZQUIERDO,4 LEE E. BROWN,5 AND MARTIM BOTTARO1
1
College of Physical Education, University of Brası´lia, Brası´lia, Brazil; 2Federal Institute of Southeastern Minas Gerais,
Campus Rio Pomba, Rio Pomba, Brazil; 3School of Physical Education, Federal University of Rio Grande do Sul, Porto Alegre,
Brazil; 4Department of Health Sciences, Public University of Navarre, Campus of Tudela, Navarra, Spain; and 5Department
of Kinesiology, California State University, Fullerton, California
ABSTRACT
Ferreira, DV, Ferreira-Júnior, JB, Soares, SRS, Cadore, EL,
Izquierdo, M, Brown, LE, and Bottaro, M. Chest press exercises
with different stability requirements result in similar muscle dam-
age recovery in resistance trained men. J Strength Cond Res 31
(1): 71–79, 2017—This study investigated the time course of 96
hours of muscle recovery after 3 different chest press exercises
with different stability requirements in resistance-trained men.
Twenty-seven men (23.5 6 3.8 years) were randomly assigned
to one of the 3 groups: (a) Smith machine bench press; (b)
barbell bench press; or (c) dumbbell bench press. Participants
performed 8 sets of 10 repetition maximum with 2 minutes rest
between sets. Muscle thickness, peak torque (PT), and soreness
were measured pre, post, 24, 48, 72, and 96 hours after exer-
cise. There were no differences in the time course of PT or
muscle thickness values of the pectoralis major (p = 0.98 and
p = 0.91, respectively) or elbow extensors (p = 0.07 and p =
0.86, respectively) between groups. Muscle soreness of the
pectoralis major was also not different between groups (p .
0.05). However, the Smith machine and barbell groups recov-
ered from triceps brachii muscle soreness by 72 hours after
exercise (p . 0.05), whereas the dumbbell group did not pres-
ent any triceps brachii muscle soreness after exercise (p .
0.05). In conclusion, resistance-trained men experience similar
muscle damage recovery after Smith machine, barbell, and
dumbbell chest press exercise. However, muscle soreness of
the elbow extensors takes a longer time to recover after using
a barbell chest press exercise.
KEY WORDS strength training, resistance exercise, free
weight, bench press, muscle soreness
Address correspondence to Diogo V. Ferreira, ferreira.diogov@gmail.com.
31(1)/71–79
Journal of Strength and Conditioning Research
Ó 2016 National Strength and Conditioning Association
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
INTRODUCTION
E
xercise choice is a critical variable that strength
and conditioning professionals must take into
consideration when designing resistance training
programs (3). Exercise mode can be manipulated
by changing body posture, number of muscles used (single-
or multi-joint exercises), grip, width of hands and feet, range
of motion, or type of device with different stability require-
ments (i.e., free weights, machines, elastic bands, etc.)
(3,4,13). These variations provide several options to design
resistance training programs. Consequently, preview studies
have evaluated the effects of different modes of resistance
exercise on acute and chronic responses to strength training
(3,20,26).
Free weight and machine exercises require distinct move-
ment control (15,26). Free weights must be stabilized in the
transverse, coronal, and sagittal planes, whereas machines use
unidirectional guided movements (15,26,28). These differen-
ces in stability requirements influence muscle force produc-
tion and neuromuscular activity (12,15,20–22,26,28). It has
been observed that higher loads can be lifted in the squat
when using a Smith machine compared with free weights
(12), whereas muscle activation is 34, 26, and 49% less in
the gastrocnemius, biceps femoris, and vastus medialis,
respectively, using a Smith machine (22). Conversely, stud-
ies have found higher loads lifted and higher muscle acti-
vation with a free weight barbell chest press when
compared with a Smith machine (15,21,26). There also
appears to be a difference within free weight exercises
because of stabilization requirements. Muscular force pro-
duction in a dumbbell chest press has been shown to be
17% less when compared with a barbell, whereas muscle
activation was higher with dumbbells (20). These results
indicate that levels of physiological stress and mechanical
strain may be different between free weights and machines
for the same muscle group.
Muscle performance may be temporarily reduced
for minutes, hours, or several days after resistance exercise
VOLUME 31 | NUMBER 1 | JANUARY 2017 | 71
 Recovery Modality and Strength Recovery

(5). This reduction is partially due to the physiological
stress and mechanical strain on muscle tissue given the
volume and intensity during heavy resistance exercise.
Longer-lasting impairment in muscle function character-
ized by a reduction in range of motion, an increase in
inflammatory response, muscle swelling, and delayed onset
muscle soreness is termed as exercise-induced muscle dam-
age (5,10,19). Accordingly, the time course of muscle dam-
age recovery is an important factor that can affect the
volume and intensity of subsequent training sessions. In
addition, an optimal balance between training stimulus
and recovery is essential to avoid overtraining and maxi-
mize training adaptations (5,8).
Investigating the effects of exercise with different
stability requirements on the time course of muscle
damage recovery may help strength and conditioning
professionals to better design strength training programs
across periodization cycles. However, to the best of
the authors’ knowledge, there are no studies that have
investigated this on the long-term time course of muscle
recovery (up to 96 hours) in resistance-trained men. We
hypothesized that using barbells would result in higher
muscle damage and a longer recovery. The rationale for
this is based on the higher load lifted during a barbell
chest press than with a machine or dumbbells (20,26),
as the magnitude of muscle damage seems to be
associated with exercise load (6,9). Therefore, the
purpose of this study was to compare the time course
of muscle damage recovery between 3 modes of chest
press exercise with different stability requirements in
resistance-trained men.
METHODS
Experimental Approach to the Problem
Volunteers were randomly assigned, using a random
number table, into one of the 3 groups: (a) Smith machine
bench press, (b) barbell bench press, or (c) dumbbell
bench press. To avoid the repeated bout effect on muscle
damage levels and the time course of recovery, the groups
were independent and each volunteer only participated in
one of the 3 groups. Participants came to the laboratory
on 7 occasions. The first visit consisted of familiarization
with the experimental procedures and anthropometric
and velocity-controlled 10 repetition maximum (RM) load
assessments. On the second visit, after 72 hours, a veloc-
ity-controlled 10RM re-test was performed. On the third
visit, 72 hours after the second, volunteers performed their
specific exercise training protocol. Indirect markers of
muscle damage were assessed before (pre), immediately
after, 24, 48, 72, and 96 hours after the training protocol.
To minimize circadian influences, subjects visited the
laboratory at the same time each day. Volunteers were
instructed to not engage in any vigorous physical activity
or unaccustomed exercise and to not take medications or
food supplements during the entire study period.
Subjects
Twenty-seven resistance-trained men (age: 23.5 6 3.8
years [ranging from 18 to 34 years]; height: 175 6 6.0
cm; mass: 80.11 6 7.54 kg) volunteered to participate
(Table 1). Volunteers had to be involved with strength
training for at least 1 year (4.36 6 3.12 years) without
interruption to be included in the study. In addition, they
had to have a bench press 1RM higher than their body
weight. Participants were excluded if they had any history
of neuromuscular, metabolic, hormonal, or cardiovascular
disease or if they were taking any medication that could
influence hormonal or neuromuscular function. Partici-
pants were informed about the design and experimental
procedures of the study and all possible risks and discom-
forts related to the procedures. They all signed an
informed consent form approved by the local institutional
review board and was performed in accordance with the
Declaration of Helsinki.
Velocity-Controlled 10 Repetition Maximum Assessment
The load used by each training group was determined by
velocity-controlled 10RM testing according to Kraemer and
Fry (14). Each group was tested according to their specific

TABLE 1. Subjects’ physical characteristics and training status of each experimental group.*

Smith machine Barbell Dumbbell

group (n = 9) group (n = 9) group (n = 9) p

Age (y) 23.7 6 3.5 23.0 6 3.1 23.9 6 4.9 0.88

Body mass (kg) 79.1 6 5.2 83.2 6 8.4 78.8 6 8.6 0.54
Height (cm) 175 6 5 178 6 8 174 6 4 0.32
PT shoulder (N$m) 116.5 6 18.1 121.9 6 22.9 127.2 6 23.4 0.58
PT elbow (N$m) 59.4 6 11.7 65.3 6 16.3 65.3 6 15.4 0.61
Resistance training experience (y) 4.2 6 3.1 4.8 6 3.5 4.0 6 3.1 0.86
Chest training sets per session 10.0 6 4.0 8.0 6 3.0 10.0 6 3.0 0.48

*PT shoulder = peak torque of shoulder horizontal adductors; PT elbow = peak torque of elbow flexors.

the TM

72 Journal of Strength and Conditioning Research

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 the TM

Journal of Strength and Conditioning Research | www.nsca.com

the velocity of each repeti-

tion, with 1–2 seconds for
the concentric and 2–3 sec-
onds for the eccentric phase.
Training Protocols
The training protocols were
composed of 8 sets of 10 repeti-
tions with 2 minutes of rest
between sets. To avoid a severe
drop in the number of repeti-
tions, the initial load was 90% of
10RM. Moreover, in the fourth
set, the load was reduced by
20%. Each repetition lasted
approximately 4 seconds, with
1–2 seconds for the concentric
and 2–3 seconds for the eccen-
tric phase. Range of motion was
also controlled for all groups so
that for the eccentric phase, they
had to touch their chest and re-
Figure 1. Shoulder horizontal adductor peak torque assessment. A) Initial position. B) Final position.
turn to a position with their el-
bows fully extended at the end of
the concentric phase. During the
exercise, and the load was adjusted with weight plates dumbbell chest press, a plastic stick was placed in each dumb-
starting at 1 kg. Volunteers warmed up by performing 10 bell bar and the sticks had to touch the chest at the end of the
repetitions at 40% of their estimated 10RM, then 10 rep- eccentric phase. In addition, their neck, head, shoulders, and
etitions at 60%, with 60 seconds of rest. Finally, their hips were kept in contact with the bench throughout the exer-
velocity-controlled 10RM load was determined with no cise, with their feet on the floor.
more than 3 attempts with 5 minutes of rest between Muscle Thickness Assessment
attempts. An electronic metronome was used to control Muscle thickness of the right pectoralis major and triceps
brachii were measured by ultra-
sonography using B-Mode
ultrasound (Model BF; Philips-
VMI, Ultra Vision Flip, New
York, NY, USA). A water-
soluble transmission gel was
applied to the measurement site,
and a 7.5-MHz ultrasound
probe was placed perpendicular
to the surface, without depress-
ing the skin. Volunteers were
measured supine for the pector-
alis major and prone for the
triceps brachii, after resting 5 mi-
nutes. The pectoralis major was
measured at the point between
the third and fourth rib under
the midpoint of the clavicle (29).
Triceps brachii was measured at
60% of the distance from the
acromial process of the scapula
Figure 2. Elbow extensor peak torque assessment. A) Initial position. B) Final position. to the lateral epicondyle of
the humerus (29). Once the

VOLUME 31 | NUMBER 1 | JANUARY 2017 | 73

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 Recovery Modality and Strength Recovery

horizontal adduction, volunteers

were positioned supine with
belts fastened across their trunk,
pelvis, and calf to minimize
extraneous body movements
(Figure 1). The acromial process
was used as a marker to align
the shoulder with the dyna-
mometer’s lever arm, allowing
a physiological range of motion
from 908 of horizontal abduction
to 08 of horizontal adduction
(908 total range of motion).
These procedures were in accor-
dance with work by Silva
et al. (23).
For the elbow extensors, vol-
Figure 3. Number of repetitions in each set for Smith machine, barbell, and dumbbell groups.
unteers were seated with their
arms placed over a Scott Bench
technician was satisfied with the quality of the image, it was positioned close to the dynamometer, allowing a range of
frozen on the monitor (7) then digitized and later analyzed motion from 1258 flexion to 58 of extension (1208 total range
with Image-J software (Version 1.37; National Institute of of motion; Figure 2). The lateral epicondyle of the humerus
Health, Bethesda, MD, USA). The measurement area was was used to align elbow rotation to the dynamometer’s lever
marked at baseline to assure that the same location was as- arm. The forearm remained in a neutral position throughout
sessed at each time point. Volunteers were asked to not clean the test. Gravity correction was obtained by measuring the
the mark. Muscle thickness was calculated as the distance from torque exerted by the lever arm and the participant’s relaxed
the subcutaneous adipose tissue-muscle interface to the arm at full extension for both tests. Values for the isokinetic
muscle-bone interface (1). variables were automatically adjusted for gravity by the Bio-
dex Advantage software. Researchers provided verbal encour-
Peak Torque Assessment
agement during all tests. Maximal peak torque was defined as
Unilateral peak torque of the shoulder horizontal adductors the highest torque value (N$m) recorded during the 2 sets.
and elbow extensors were measured by an isokinetic dyna-
mometer (Biodex Medical, Inc., Shirley, NY, USA). Volunteers Muscle Soreness and Subjective Perception of Muscular
performed 2 sets of 4 repetitions at 608 per second for each Fatigue and Recovery
exercise with 2 minutes of rest between sets. For shoulder Muscle soreness of the pectoralis major and triceps brachii
were assessed using a 100-mm
visual analog scale with “no
soreness” (0 mm) on one end
and “severe soreness” (100
mm) on the other. Volunteers
rated their muscle soreness
when the muscle was palpated
by the examiner, who applied
pressure to the medial part of
the pectoralis major and triceps
brachii with the third and
fourth finger for approximately
3 seconds (27). The same
examiner performed all test
procedures for all subjects.
To measure subjective per-
ceived muscular fatigue and
Figure 4. Mean 6 SD of normalized shoulder horizontal adduction peak torque before (pre), immediately post, recovery, subjects rated their
and 24–96 hours after exercise in each group. *p # 0.05, lower than pre. PT = peak torque. perception of physical fitness
using a visual analog scale from
the TM

74 Journal of Strength and Conditioning Research

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 the TM

Journal of Strength and Conditioning Research | www.nsca.com

to analyze peak torque and

muscle thickness. Physical
characteristics were analyzed
by one-way ANOVA, and
training volume differences
were analyzed by 2-way (3
groups 3 8 training sets)
mixed factor ANOVA. In the
case of significant differences,
a Fisher’s least significant dif-
ference post hoc test was used.
As muscle soreness and subjec-
tive physical fitness data did
not present a normal distribu-
tion, the nonparametric Mann-
Whitney (between groups)
and Friedman (within group)
Figure 5. Mean 6 SD percent change from baseline of pectoralis major muscle thickness before (pre),
immediately post, and 24–96 hours after exercise in each group. *p # 0.05, higher than pre.
tests were used to analyze
these variables. The signifi-
cance level was set a priori at
0 to 120 mm, where 0 mm was maximum fatigue and not p # 0.05. Reliability of all measurements was calculated by
recovered and 100 mm was maximal physical fitness intraclass correlation coefficient (ICC) values using single
recovery, compared to their fitness the week before the values. In addition, the effect size calculation was used to
training protocol (2). determine the magnitude of each condition effect. Cohen’s
(11) ranges of 0.1, 0.25, and 0.4 were used to define small,
Statistical Analyses medium, and large ƒ values, respectively. SPSS software
Data are presented as mean 6 standard deviation. The (version 17.0; SPSS, Inc., Chicago, IL, USA) was used for
Shapiro-Wilk test was used to check for a normal distribu- all data analyses.
tion. Taking into account that peak torque and muscle
thickness data were normally distributed, a 2-way 3 3 6 RESULTS
(group [Smith machine, barbell and dumbbell] 3 time Training Performance and Physical Characteristics
[pre, immediately post, 24, 48, 72, and 96 hours after exer- Physical characteristics and training status were not different
cise]) mixed factor analysis of variance (ANOVA) was used between groups (p . 0.05, Table 1). Test-retest reliability ICCs
for the Smith machine, barbell,
and dumbbell 10RM tests were
0.96, 0.97, and 0.89, respectively.
The load lifted by the dumbbell
group (62.8 6 9.5 kg) was 18.6%
lower than the barbell group
(74.5 6 12.5 kg, p = 0.042)
and 15.2% lower than the Smith
machine group (72.4 6 9.7 kg,
p = 0.05). However, the total
amount of weight lifted during
each training protocol (10RM
loading) did not differ between
the Smith machine (14,218.0 6
3743.0 kg), barbell (14,411.0 6
3986.9 kg), and dumbbell
groups (12,077.0 6 2915.0 kg)
(F = 3.0, p = 0.69). In addition,
there was no difference between
Figure 6. Mean 6 SD of pectoralis major muscle soreness before (pre), immediately post, and 24–96 hours after groups for the number of repe-
exercise in each group. *p # 0.05, higher than pre. titions performed (F = 0.81, p =
0.64; Figure 3).

VOLUME 31 | NUMBER 1 | JANUARY 2017 | 75

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 Recovery Modality and Strength Recovery

TABLE 2. Time course recovery of elbow flexors after exercise protocol in each experimental group.*

Group Baseline Post 24 h 48 h 72 h 96 h

PT (N$m) Smith 59.4 6 11.7 46.1 6 10.7† 56.7 6 9.5 58.7 6 8.4 61.5 6 8.4 60.7 6 7.7
Barbell 65.3 6 16.3 53.2 6 10.8† 62.6 6 14.6 61.7 6 12.5 60.1 6 11.2 61.4 6 13.8
Dumbbell 65.3 6 15.4 57.8 6 13.9† 61.9 6 15.8 61.6 6 12.5 61.7 6 13.6 65.1 6 14.9
MT (mm) Smith 44.0 6 10.5 49.7 6 8.6† 44.8 6 9.3 44.5 6 9.1 44.5 6 9.5 44.1 6 8.7
Barbell 44.2 6 8.8 50.8 6 8.8† 46.8 6 8.9 46.1 6 9.1 45.5 6 9.3 45.9 6 8.8
Dumbbell 37.2 6 5.1 43.4 6 5.9† 38.9 6 3.9 39.4 6 4.3 38.3 6 4.7 37.4 6 4.6
MS (cm) Smith 0 1.9 6 1.9† 1.3 6 1.9† 0.3 6 1.0 0.1 6 0.2
Barbell 0 2.2 6 1.5†z 1.7 6 1.7†z 0.4 6 0.5 0
Dumbbell 0 0.8 6 1.1 0.2 6 0.7 0 0

*PT = peak torque; MT = muscle thickness; MS = muscle soreness.

†p # 0.05, lower than baseline.
zp# 0.05, higher than dumbbell group.

Time Course Recovery of Shoulder Horizontal Adductors torque (F = 30.11, p , 0.001, power = 1) and muscle thickness
Test-retest reliability ICCs for peak torque, muscle thickness, (F = 19.02, p , 0.001, power = 1). Each group recovered their
and muscle soreness of the triceps brachii were 0.93, 0.93, and shoulder horizontal adductor peak torque by 72 hours after
0.86, respectively. Peak torque of the shoulder horizontal the exercise protocol (p . 0.05), whereas pectoralis major
adductors decreased immediately after resistance exercise muscle thickness returned to baseline by 24 hours for all
with no differences between groups (17.1% for the Smith groups (p . 0.05).
machine, 16.3% for the barbell, and 16.9% for the dumbbell, There was no difference in pectoralis major muscle soreness
p = 0.94). In addition, there was no significant group-by-time between groups (p . 0.05, ƒ = 0.11; Figure 6). All groups
interaction for shoulder horizontal adductors peak torque recovered by 96 hours (Smith machine; x2 = 29.08, p ,
(F = 0.43, p = 0.93, power = 0.22, ƒ = 0.2) (Figure 4) or 0.001; barbell: x2 = 31.05, p , 0.001; dumbbell: x2 = 32.05,
pectoralis major muscle thickness (F = 0.26, p = 0.98, power = p , 0.001).
0.11, ƒ = 0.15) (Figure 5). There were also no significant main
effects for group for peak torque (F = 0.31, p = 0.74, power = Time Course Recovery of the Elbow Extensors
0.09) or muscle thickness (F = 0.13, p = 0.88, power = 0.07). Test-retest reliability ICCs for peak torque, muscle thickness,
However, there were significant main effects for time for peak and muscle soreness of the pectoralis major was 0.96.
Time course recovery of the
elbow extensors is presented in
Table 2. There were no signifi-
cant group-by-time interac-
tions for elbow extensor peak
torque (F = 1.63, p = 0.11,
power = 0.24, ƒ = 0.38) or mus-
cle thickness (F = 0.48, p =
0.90, power = 0.19, ƒ = 0.2).
There were also no significant
main effects for group for
elbow extensor peak torque
(F = 0.66, p = 0.52, power =
0.18) or muscle thickness (F =
2.08, p = 0.15, power = 0.38).
However, there were main ef-
fects for time for elbow exten-
sor peak torque (F = 14.99, p =
0.001, power = 0.96) and mus-
Figure 7. Mean 6 SD of subjective perception of physical fitness before (pre) and 24–96 hours after exercise in
each group. *p # 0.05, lower than pre. †p # 0.05, lower than pre within barbell group. #p # 0.05, higher than cle thickness (F = 35.64, p ,
barbell group. 0.001, power = 1). All groups
recovered their elbow extensor
the TM

76 Journal of Strength and Conditioning Research

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

peak torque by 48 hours (p . 0.05), whereas triceps brachii
muscle thickness returned to baseline by 24 hours (p . 0.05).
The Smith machine and barbell groups recovered from
elbow extensor muscle soreness by 72 hours (Smith
machine; x2 = 17.49, p = 0.002; barbell: x2 = 22.46, p ,
0.001), whereas the dumbbell group did not change through-
out the entire 96 hours (x2 = 11.84, p = 0.06). Moreover,
muscle soreness was higher in the barbell group at 24 and 48
hours when compared with the dumbbell group (p # 0.05).
The effect size was medium (ƒ = 0.25).
Time Course Recovery of Subjective Physical
Fitness Recovery
For subjective perception of training, the Smith machine and
dumbbell groups recovered by 72 hours (Smith machine;
x2 = 22.67, p , 0.001; dumbbell: x2 = 25.41, p , 0.001)
(Figure 7), but the barbell group did not recover till 96 hours
(x2 = 31.33, p , 0.001). Finally, subjective physical fitness
recovery was higher in the dumbbell group at 24 and 72
hours when than in the barbell group (p # 0.05). The effect
size was small (ƒ = 0.18).
DISCUSSION
Tracking the time course of muscle recovery is an important
consideration when designing periodized resistance training
programs, especially in high-level sports, such as competitive
weightlifting, powerlifting, and bodybuilding, because of its
known negative relationship with workload, training volume
(5,8), and subsequent training stimulus. Thus, the purpose of
this study was to compare the time course of muscle damage
recovery between 3 modes of chest press exercise with dif-
ferent stability requirements in resistance-trained men. Our
initial hypothesis was not confirmed, as there were no differ-
ences between groups in recovery time or muscle thickness.
However, when examining the 2 free weight modes (barbell
vs. dumbbell), elbow extensor muscle soreness of the barbell
group was higher and took longer to recover than the Smith
machine and dumbbell groups. In addition, despite similar
muscle strength recovery of the shoulder horizontal adduc-
tors and elbow extensors, the barbell group demonstrated
lower readiness (i.e., subjective perception of physical fitness)
for training than the other 2 groups.
Several studies have shown higher force production
during free weight bench press than a Smith machine bench
press (12,20,26). Saeterbakken et al. (20) observed that mus-
cle strength (1RM) in a horizontal barbell chest press was
3% higher than that for a Smith machine and 17% higher
than that for a dumbbell chest press. In addition, muscle
strength of a dumbbell chest press was 14% less than for
a Smith machine. These results are consistent with research
by Tillaar et al. (26), who found that the highest 1RM chest
press strength was performed in the barbell chest press
(106.4 6 15.5 kg), followed by the Smith machine (103.6
6 14.8 kg) then dumbbells (89.5 6 13.7 kg). Another study
also reported greater 1RM strength in the barbell chest press
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
the TM
Journal of Strength and Conditioning Research | www.nsca.com
in comparison with the Smith machine (12). Although the
present study did not show significant differences in training
volume between groups, our results are somewhat in agree-
ment with those of previous studies, because the load lifted
(10RM) by the dumbbell group was 18.6% lower than the
barbell group and 15.2% lower than the Smith machine
group.
The magnitude of muscle damage caused by each exercise
in this study was in accordance with previous studies that
have evaluated muscle recovery after chest press exercise in
trained men (16,27). Meneghel et al. (16) investigated
changes in indirect markers of muscle damage after 2 bouts
of free weight eccentric bench press performed by
resistance-trained men. Their participants performed 4 sets
of 8 eccentric actions (3 seconds for each repetition) at 70%
of their eccentric 1RM with 2 minutes of rest between sets.
Similar to this study, they reported a reduction of 10% in
1RM strength and peak muscle soreness of 3.8 mm after
exercise. The results reported by Soares et al. (24) also cor-
roborate our findings on the time course of recovery of the
elbow extensors after a multijoint exercise. They observed
that elbow flexor peak torque recovered at 24 and 72 hours
after multijoint exercise (seated row machine), in which
elbow flexors were recruited as synergist muscles. Thus, ac-
cording to Paulsen et al. (19), the muscle damage caused by
each exercise in this study could be considered moderate, as
overall, volunteers took 72 hours to recover.
As this is the first study to compare the effects of a Smith
machine, barbell, and dumbbell chest press on the time
course of muscle recovery, the present findings cannot be
directly compared with previous studies. Taking into
account the fact that resistance load can affect muscle
damage recovery (17,18,25), it was expected that the higher
load in the barbell chest press reported by previous studies
(12,20,26) would result in higher muscle damage. However,
besides difference in load between modes in this study, the
volume lifted throughout the 8 sets was similar between
groups. These results are consistent with Uchida et al. (27)
who compared the effects of 4 different intensities of bench
press on indirect markers of muscle damage. They tested
trained men with 5 intensities: (a) 50% (1RM), (b) 75%, (c)
90%, (d) 110%, and (e) control, whereas total volume was
matched between groups. Muscle soreness and plasma cre-
atine kinase levels were not significantly different between
groups after exercise. According to the authors, the intensity
of the bench press exercise did not affect the magnitude of
muscle soreness or blood markers of muscle damage. Thus,
it appears that training volume has an important influence
on exercise-induced muscle damage independent of instabil-
ity and load requirements for the same muscle group. How-
ever, this topic requires further investigation.
The Smith machine and barbell groups recovered from
elbow extensor soreness 48 hours after exercise, whereas the
dumbbell group did not present any muscle soreness at all.
This difference may be related to the stability requirements
VOLUME 31 | NUMBER 1 | JANUARY 2017 | 77
 Recovery Modality and Strength Recovery
of each exercise. Increased stability is required with dumbbells
compared with barbells, which is most likely because of the
separate movement of the elbow joint, which can increase
instability in the frontal and sagittal planes, thus shifting the
muscle activation pattern (26). Higher muscle activation in the
elbow flexors has been seen in the chest press with different
stability requirements (dumbbell higher than barbell), whereas
higher elbow extensor activity was observed in a barbell than
in a dumbbell chest press (20,26). Thus, the load imposed on
elbow flexors (i.e., biceps) may be greater with dumbbells than
with a barbell, leading to a distinct and specific time course of
muscle soreness recovery.
Another interesting result of the present study was that
the subjective perception of physical fitness (Figure 7) was
higher in the barbell than the dumbbell group. The barbell
group took up to 96 hours to recover, whereas the Smith
machine and dumbbell groups took 72 hours. These results
may suggest that the barbell imposed a higher physiological
demand and are similar to those of Ahtiainen et al. (2) where
their trained volunteers took up to 144 hours to recover from
5 sets of 10RM leg press and 4 sets of 10RM squat exercises.
However, these findings require further investigation. More-
over, the present study is not without limitation as muscular
enzymes and biochemical inflammatory markers were not
measured. In addition, only young resistance-trained men
were evaluated.
In conclusion, different modes of chest press exercise
required similar recovery time for peak torque and muscle
thickness in the shoulder horizontal adductors and elbow
extensors. In addition, pectoralis muscle soreness recovery
was not different between modes. However, the time course
of elbow extensor muscle soreness recovery was greater after
barbell compared with dumbbell exercise. Whether the
similar muscle recovery observed in the present study would
result in similar muscle adaptations remains unknown. Thus,
further study is necessary to understand what effects free
weights vs. machines have on muscle strength and hyper-
trophy gains. It is also important to evaluate muscle damage
recovery in other populations, such as untrained volunteers,
women, and the elderly.
PRACTICAL APPLICATIONS
The results of the current study suggest that coaches and
strength professionals can expect similar muscle damage
recovery after Smith machine, barbell, or dumbbell chest
press resistance exercise in trained men. However, they might
not be able to perform strength training at their maximal
intensity until 72 hours after 10RM dumbbell or Smith
machine and 96 hours after barbell bench press. Thus,
strength and conditioning professionals should consider
focusing on regenerative training or lower load protocols,
such as muscle power or muscular endurance training, during
the recovery period after heavy resistance exercise when
maximal muscle force remains reduced. In addition, they can
target other muscle groups during this recovery time.
the TM
78 Journal of Strength and Conditioning Research
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
ACKNOWLEDGMENTS
This study received no external funding. The authors
declare no conflicts of interest. The results of the present
study do not constitute endorsement of any product by the
authors or the National Strength and Conditioning Asso-
ciation.
REFERENCES
1. Abe, T, DeHoyos, DV, Pollock, ML, and Garzarella, L. Time course
for strength and muscle thickness changes following upper and
lower body resistance training in men and women. Eur J Appl
Physiol 81: 174–180, 2000.
2. Ahtiainen, JP, Lehti, M, Hulmi, JJ, Kraemer, WJ, Alen, M, Nyman, K,
Selänne, H, Pakarinen, A, Komulainen, A, Kovanen, V, Mero, AA,
and Häkkinen, K. Recovery after heavy resistance exercise and
skeletal muscle androgen receptor and insulin-like growth factor-I
isoform expression in strength trained men. J Strength Cond Res 25:
767–777, 2011.
3. ACSM. American College of Sports Medicine position stand.
Progression models in resistance training for healthy adults. Med Sci
Sports Exerc 41: 687–708, 2009.
4. Andersen, V, Fimland, MS, Wiik, E, Skoglund, A, and
Saeterbakken, AH. Effects of grip width on muscle strength and
activation in the lat pull-down. J Strength Cond Res 28: 1135–1142,
2014.
5. Barnett, A. Using recovery modalities between training sessions in
elite athletes - Does it help?. Sports Med 36: 781–796, 2006.
6. Barroso, R, Roschel, H, Gil, S, Ugrinowitsch, C, and Tricoli, V.
Effect of the number and the intensity of eccentric muscle actions
on muscle damage markers. Rev Bras Med Esporte 17: 397–400,
2011.
7. Bemben, MG. Use of diagnostic ultrasound for assessing muscle size.
J Strength Cond Res 16: 103–108, 2002.
8. Bishop, PA, Jones, E, and Woods, AK. Recovery from training: A
brief review. J Strength Cond Res 22: 1015–1024, 2008.
9. Chen, TC, Nosaka, K, and Sacco, P. Intensity of eccentric exercise,
shift of optimum angle, and the magnitude of repeated-bout effect. J
Appl Physiol 102: 992–999, 2007.
10. Clarkson, PM and Hubal, MJ. Exercise-induced muscle damage in
humans. Am J Phys Med Rehab 81: S52–S69, 2002.
11. Cohen, J. Statistical Power Analysis for the Behavioral Sciences.
Hillsdale, NJ: Lawrence Erlbaum Associates, 1988.
12. Cotterman, ML, Darby, LA, and Skelly, WA. Comparison of muscle
force production using the Smith machine and free weights for
bench press and squat exercises. J Strength Cond Res 19: 169–176,
2005.
13. Glass, SC and Armstrong, T. Electromyographical activity of the
pectoralis muscle during incline and decline bench presses. J
Strength Cond Res 11: 163–167, 1997.
14. Kraemer, WJ and Fry, AC. Strength testing: Development and
evaluation of methodology. In: Physiological Assessment of Human
Fitness. P. Maud and C. Foster, eds. Champaign, IL: Human Kinetics,
2006, pp. 119–150.
15. McCaw, ST and Friday, JJ. A comparison of muscle activity between
a free weight and machine bench press. J Strength Cond Res 8: 259–
264, 1994.
16. Meneghel, AJ, Verlengia, R, Crisp, AH, Aoki, MS, Nosaka, K, da
Mota, GR, and Lopes, CR. Muscle damage of resistance-trained
men after two bouts of eccentric bench press exercise. J Strength
Cond Res 28: 2961–2966, 2014.
17. Nosaka, K and Newton, M. Difference in the magnitude of muscle
damage between maximal and submaximal eccentric loading. J
Strength Cond Res 16: 202–208, 2002.

18. Paschalis, V, Koutedakis, Y, Jamurtas, AZ, Mougios, V, and
Baltzopoulos, V. Equal volumes of high and low intensity of
eccentric exercise in relation to muscle damage and performance. J
Strength Cond Res 19: 184–188, 2005.
19. Paulsen, G, Mikkelsen, UR, Raastad, T, and Peake, JM. Leucocytes,
cytokines and satellite cells: What role do they play in muscle
damage and regeneration following eccentric exercise?. Exerc
Immunol Rev 18: 42–97, 2012.
20. Saeterbakken, AH, van den Tillaar, R, and Fimland, MS. A
comparison of muscle activity and 1-RM strength of three chest-
press exercises with different stability requirements. J Sports Sci 29:
533–538, 2011.
21. Schick, EE, Coburn, JW, Brown, LE, Judelson, DA, Khamoui, AV,
Tran, TT, and Uribe, BP. A comparison of muscle activation
between a Smith machine and free weight bench press. J Strength
Cond Res 24: 779–784, 2010.
22. Schwanbeck, S, Chilibeck, PD, and Binsted, G. A comparison of free
weight squat to Smith machine squat using electromyography. J
Strength Cond Res 23: 2588–2591, 2009.
23. Silva, RT, Gracitelli, GC, Saccol, MF, Laurino, CF, Silva, AC, and
Braga-Silva, JL. Shoulder strength profile in elite junior tennis
players: Horizontal adduction and abduction isokinetic evaluation.
Br J Sports Med 40: 513–517, 2006.
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
the TM
Journal of Strength and Conditioning Research | www.nsca.com
24. Soares, S, Ferreira-Junior, JB, Pereira, MC, Cleto, VA,
Castanheira, RP, Cadore, EL, Brown, LE, Gentil, P, Bemben, MG,
and Bottaro, M. Dissociated time course of muscle damage recovery
between single- and multi-joint exercises in highly resistance-trained
men. J Strength Cond Res 29: 2594–2599, 2015.
25. Tiidus, PM and Ianuzzo, CD. Effects of intensity and duration of
muscular exercise on delayed soreness and serum enzyme-activities.
Med Sci Sport Exer 15: 461–465, 1983.
26. Tillaar, RV and Saeterbakken, A. The sticking region in three chest-
press exercises with increasing degrees of freedom. J Strength Cond
Res 26: 2962–2969, 2012.
27. Uchida, MC, Nosaka, K, Ugrinowitsch, C, Yamashita, A, Martins, E
Jr, Moriscot, AS, and Aoki, MS. Effect of bench press exercise
intensity on muscle soreness and inflammatory mediators. J Sports
Sci 27: 499–507, 2009.
28. Uribe, BP, Coburn, JW, Brown, LE, Judelson, DA, Khamoui, AV,
and Nguyen, D. Muscle activation when performing the chest press
and shoulder press on a stable bench vs. a Swiss ball. J Strength Cond
Res 24: 1028–1033, 2010.
29. Yasuda, T, Fujita, S, Ogasawara, R, Sato, Y, and Abe, T. Effects of
low-intensity bench press training with restricted arm muscle blood
flow on chest muscle hypertrophy: A pilot study. Clin Physiol Funct
Imaging 30: 338–343, 2010.
VOLUME 31 | NUMBER 1 | JANUARY 2017 | 79