Journal of Strength and Conditioning Research, 2007, 21(1), 157–163

䉷 2007 National Strength & Conditioning Association

DISSIMILAR EFFECTS OF ONE- AND THREE-SET

STRENGTH TRAINING ON STRENGTH AND MUSCLE
MASS GAINS IN UPPER AND LOWER BODY IN
UNTRAINED SUBJECTS
BENT R. RøNNESTAD,1,4 WILHELM EGELAND,1 NILS H. KVAMME,1 PER E. REFSNES,2 FAWZI KADI,3
AND TRULS RAASTAD1

Norwegian College of Sport Sciences, Oslo, Norway; 2National Sports Center, Oslo, Norway; 3Department of
1

Health Sciences, University of Örebro, Örebro, Sweden; 4Centre for Continuing Education, Lillehammer
University College, Lillehammer, Norway.

ABSTRACT. Rønnestad, B.R., W. Egeland, N.H. Kvamme, P.E. has been thoroughly reviewed with some discrepancies
Refsnes, F. Kadi, and T. Raastad. Dissimilar effects of one- and among the reviewers on the effect on strength improve-
three-set strength training on strength and muscle mass gains ment (4–6, 10, 24, 26, 27). Results in the literature range
in upper and lower body in untrained subjects. J. Strength Cond.
from no differences in strength gain between single and
Res. 21(1):157–163. 2007.—The purpose of this study was to
compare the effects of single- and multiple-set strength training multiple sets (33) to a significant superiority of multiple
on hypertrophy and strength gains in untrained men. Twenty- sets (29). There seem to be several methodological expla-
one young men were randomly assigned to either the 3L-1UB nations to this discrepancy. The pretest strength mea-
group (trained 3 sets in leg exercises and 1 set in upper-body sures were sometimes performed only once (12, 18, 20,
exercises; n ⫽ 11), or the 1L-3UB (trained 1 set in leg exercises 28, 32), thus saying nothing about the reliability of the
and 3 sets in upper-body exercises; n ⫽ 10). Subjects trained 3 baseline data. In several single- vs. triple-set studies
days per week for 11 weeks and each workout consisted of 3 leg there is a difference in training intensity (17, 18, 20, 28)
exercises and 5 upper-body exercises. Training intensity varied and type of exercises (18, 20) between training groups.
between 10 repetition maximum (RM) and 7RM. Strength (1RM)
When the goal is to examine the effects of different train-
was tested in all leg and upper-body exercises and in 2 isokinetic
tests before training, and after 3, 6, 9, and 11 weeks of training. ing volumes on strength increases, all other training var-
Cross sectional area (CSA) of thigh muscles and the trapezius iables should be held constant to attribute any differences
muscle and body composition measures were performed before in strength increase to the training volume.
training, and after 5 and 11 weeks of training. The increase in Other studies are missing data on the amount of rest
1RM from week 0 to 11 in the lower-body exercises was signif- between sets given to the subjects who performed 3 sets
icantly higher in the 3L-1UB group than in the 1L-3UB group (1, 28) and the subjects’ strength training experience (36).
(41 vs. 21%; p ⬍ 0.001), while no difference existed between There are numerous reports on strength-trained subjects
groups in upper-body exercises. Peak torque in maximal isoki- gaining more from a larger training volume compared
netic knee-extension and thigh CSA increased more in the 3L-
1UB group than in the 1L-3UB group (16 vs. 8%; p ⫽ 0.03 and
with untrained subjects (15, 24, 26, 27). Therefore, mul-
11 vs. 7%; p ⫽ 0.01, respectively), while there was no significant tiple sets seem to be superior to single-set training in
difference between groups in upper trapezius muscle CSA. The strength trained subjects. Thus, it is essential to know
results demonstrate that 3-set strength training is superior to the training experience of the subjects when comparing
1-set strength training with regard to strength and muscle mass the strength-enhancing effects of single- and triple-set
gains in the leg muscles, while no difference exists between 1- strength training.
and 3-set training in upper-body muscles in untrained men. Each human muscle seems to be unique on the basis
KEY WORDS. single set, multiple sets, leg muscles, upper-body of its muscle fiber composition, fiber diameter, and func-
muscles, training volume tion (34). Fleck and Kraemer (9) report that strength
gains with different strength training regimens can vary
dramatically from one muscle group to another. There-
INTRODUCTION fore, the response to single- vs. multiple-set strength
strength-training program mainly consists of training could possibly be different in diverse muscle

A 3 different variables: volume, intensity, and

frequency. In addition, the order of exercises,
rest period between sets and exercise, con-
traction type, contraction velocity (15, 23), and
nutrition (8) may affect the adaptations to strength train-
groups. This is supported by Paulsen et al. (22) who ob-
served that leg muscles responded to a greater extent to
3-set strength training when compared with single-set
strength training, while there were no differences in up-
per-body muscle response between the 2 training regi-
ing. These variables can be manipulated in numerous mens in untrained men. This indicates a lower threshold
ways, resulting in an almost endless continuum of differ- of training volume in upper-body muscles compared with
ent strength training programs. However, of the various lower-body muscles. In the study of Paulsen et al. (22) the
training variables, volume has received the most atten- subjects trained the same exercises with the same inten-
tion in the past 7 years, with focus mainly on the debate sity and two pre- and posttests were conducted. The study
concerning single-set vs. multiple-set strength training lasted only 6 weeks, and no measurement of changes in
programs (5, 6, 10). muscle mass or nutrition control was carried out.
The question of single- vs. triple-set strength training Therefore, the purpose of this study was to further
157
158 RøNNESTAD, EGELAND, KVAMME ET AL.

TABLE 1. Subjects’ characteristics at baseline in group 1L- Procedures

3UB (1 set in leg exercises and 3 sets in upper-body exercises)
and 3L-1UB (3 sets in leg exercises and 1 set in upper-body During 2 familiarization sessions, subjects were instruct-
exercises). Values are mean ⫾ SE.* ed in proper lifting technique and testing procedures.
Variable 1L-3UB 3L-1UB Thereafter, subjects completed 2 testing sessions before
the start of the training period on separate days, each
Age (y) 26.6 ⫾ 1.6 26.5 ⫾ 1.3
separated by approximately 72 hours.
Height (cm) 181.8 ⫾ 3.0 181.8 ⫾ 2.2
Weight (kg) 82.6 ⫾ 3.4 80.4 ⫾ 4.2 Before the testing, subjects conducted a 10-minute
Fat percentage 24.1 ⫾ 1.9 22.1 ⫾ 3.0 warm-up on a cycle at a workload of 60–70 W. The iso-
1RMupper-body (kg) 69 ⫾ 1.9 69 ⫾ 2.1 kinetic knee-extension and knee-flexion was performed in
1RMleg (kg) 118 ⫾ 13.2 107 ⫾ 11.5 a REV9000 dynamometer (Technogym, Gambettola, Ita-
Knee flexion (N·m)† 128 ⫾ 5.4 125 ⫾ 7.3 ly). The range of motion was set from a knee angle of 90⬚
Knee extension (N·m)† 228 ⫾ 9.5 217 ⫾ 14.4 to 20⬚ from full extension. Subjects performed 4 warm-up
CSAquadriceps (cm2) 71.4 ⫾ 3.2 73.2 ⫾ 3.8 contractions followed by 3 maximal contractions at a
CSAhamstrings (cm2) 61.9 ⫾ 4.4 62.6 ⫾ 3.4
speed of 60⬚·s⫺1. Peak torque was used in the data treat-
CSAupper trapezius (cm2) 10.5 ⫾ 0.7 13.2 ⫾ 1.4
Lean body massupper-body (kg) 33.1 ⫾ 1.6 35.5 ⫾ 3 ment. Coefficients of variation were 6.2 and 5.5% for the
Lean body masslower-body (kg) 19.9 ⫾ 1.4 20.0 ⫾ 1.5 knee-extension and -flexion, respectively. Only the right
leg was tested. Immediately following the isokinetic tests,
* CSA ⫽ cross sectional area; 1RM ⫽ 1 repetition maximum. the subjects continued with the 1RM tests.
† Peak torque at 60⬚·s⫺1.
The 1RM testing was performed in leg press, leg ex-
tension, leg curl, seated chest press, seated rowing, latis-
investigate the differences in adaptation between upper- simus pull-down, biceps curl, and shoulder press. In all
and lower-body muscles to 1- and 3-set strength training exercises, the subjects performed a standardized warm-
protocols. In this study we have added measures of up consisting of 3 sets with gradually increasing load (40,
change in muscle mass as a possible mechanism behind 75, and 85% of expected 1RM) and decreasing number of
the different strength gains and extended the training repetitions, 12, 7, and 3. The first attempt in all 1RM
period to 11 weeks. tests was performed with a load approximately 5% below
the expected 1RM. After each successful attempt, the load
METHODS was increased by 2–5% until failure of lifting the load in
Experimental Approach to the Problem 2–3 following attempts. The rest period between each at-
tempt was 4 minutes. The order of tests was similar in
The current study was conducted to compare the effect of all testing sessions. All 1RM testing was overseen by the
single- and triple-set strength training on strength gains same investigator and conducted on the same equipment
and changes in muscle mass during the first 11 weeks of with identical subject/equipment positioning. The coeffi-
training in untrained men. In an effort to reduce any pos- cient of variation was ⬍5.4% in all 1RM tests. The high-
sible effects of differences in total training volume, 1 est value from the 2 pre- and posttests was used in sta-
group performed 3 sets in leg exercises and 1 set in upper- tistical analysis. Only one 1RM test was conducted after
body exercises, while the other group performed 1 set in 3, 6, and 9 training weeks. All tests were accomplished
leg exercises and 3 sets in upper-body exercises. Effects at the same time of day and in the same order.
on strength gains were tested in 1 repetition maximum Muscle Cross Sectional Area. Cross sectional area
(1RM) tests in 3 leg exercises and 5 upper-body exercises (CSA) of the thigh muscles and trapezius muscle was
and in 2 isokinetic tests before training, and after 3, 6, 9, measured using magnetic resonance imaging technique
and 11 weeks of training. Cross sectional area of thigh
(MRI; GE Signa 1.5 Tesla EchoSpeed, GE Medical Sys-
muscles and trapezius muscle and body composition mea-
tems, Milwaukee, WI). Scans of the thigh muscles were
sures were performed before training, and after 5 and 11
taken while the subjects were relaxed in a supine posi-
weeks of training. Dietary intake was assessed before
training, and after 5 and 10 weeks of strength training. tion. Axial transverse images were obtained through the
Training intensity was equated and physical activity out- thigh at ¼, ½, and ¾ of the length of the femur from the
side the training program, as well as diet and protein condyle. After an axial localization scan in parallel with
consumption, was controlled. both acromioclavicular joints, 9 sagittal images of the up-
per trapezius muscle were taken from acromioclavicular
Subjects articulation with an 8-mm space between each image.
Twenty-four healthy men (see Table 1 for descriptive Due to methodical errors, the CSA of the upper trapezius
data) with no regular strength training (⬍3 times per muscle was measured in only 5 subjects from each group.
month) during the last year volunteered to participate in The identities of the scans were blinded and analyzed by
this study. Three subjects withdrew before completion of the same person. The thigh muscles were divided into
the study for reasons unrelated to the study. The study extensors (quadriceps muscles) together with the sarto-
was approved by the Regional Ethics Committee of Nor- rius muscle, and flexors/adductors compartments using a
way. tracer function in the software. Using similar methods,
Subjects were randomly divided into 2 groups. Group coefficient of variation has been calculated to 2% from
1L-3UB (n ⫽ 10) trained 1 set in leg exercises and 3 sets repeated examinations in 8 subjects (21).
in upper-body exercises. Group 3L-1UB (n ⫽ 11) trained Dual Energy X-Ray Absorptiometry. Lean body mass
3 sets in leg exercises and 1 set in upper-body exercises. and fat mass were determined by dual energy x-ray ab-
There were no differences between groups in anthropo- sorptiometry (DXA) using a Lunar Prodigy densiometer
metric parameters, 1RM, peak isokinetic strength, or (GE Medical Systems, Madison, WI). Prior to the DXA
measures of muscle mass before the training period (Ta- scan, subjects were requested to avoid training for 24
ble 1). hours and to avoid any ingestion of liquids 2 hours before
 EFFECTS OF ONE- AND THREE-SET STRENGTH TRAINING 159

the scan. Subjects were lying in a standardized position

in the machine.
Dietary Intake. The subjects recorded their daily die-
tary intake for a 4-day period (Wednesday to Saturday)
using a weighted food intake method. When subjects are
not supervised 24 hours a day, the weighted food intake
method is recognized as a valid method (2, 3). The sub-
jects were given food record journals and digital food
weighing scales (Vera 67002; Soehnle-Waagen GmbH &
Co, Murrhardt, Germany; precision 1 g). They were also
given detailed oral and written guidelines about how to FIGURE 1. Relative changes in 1 repetition maximum (1RM)
carry out this method. Dietary assessment data were an- during the 11-week training intervention in the leg exercises
alyzed using a nutrient analysis program (Mat på data (left) and upper-body exercises (right). 1L-3UB ⫽ 1 set in leg
4.1; LKH, Oslo, Norway). exercises and 3 sets in upper-body exercises; 3L-1UB ⫽ 3 sets
Training. The 11-week training period consisted of 3 in leg exercises and 1 set in upper-body exercises. * Significant
workouts per week. Each workout consisted of leg press, difference from baseline (p ⬍ 0.001). # Significant differences
leg extension, leg curl, seated chest press, seated rowing, between groups (p ⬍ 0.001).
latissimus pull-down, biceps curl, and shoulder press for
all subjects and in the same order. After a 10-minute
warm-up with light jogging or cycling, subjects performed
2 warm-up sets before the leg exercises and another 2
warm-up sets before the upper body exercises. All sub-
jects were supervised by one of the investigators on all
workouts during the first 3 weeks, and thereafter at least
once a week during the entire training period.
Both groups trained 3 times per week on nonconsec-
utive days. Training intensity (number of RM) was al-
tered similarly for the 2 groups. During the first 2 weeks
both groups trained with 10RM sets in all exercises; dur-
ing the third and fourth training weeks they increased
the intensity to 8RM sets, and during the final 6 weeks
they trained with 7RM sets. Subjects were encouraged to
continuously increase their RM loads during the inter-
vention. Subjects were allowed assistance on the last rep-
etition. However, to achieve a modified daily undulating
FIGURE 2. Percentage increase in peak isokinetic knee-
periodization, the subjects were told to reduce their train- extension and knee-flexion torque from pre- to posttest in the
ing load with ⬃10% in the second exercise bout in each 1L-3UB (1 set in leg exercises and 3 sets in upper-body
week (this was coordinated between the 2 training exercises) and 3L-1UB group (3 sets in leg exercises and 1 set
groups). in upper-body exercises). * Significant difference from baseline
In an effort to reduce any possible effects of differenc- (p ⬍ 0.05). # Significant differences between groups (p ⫽ 0.03).
es in whole-body training volume, 1 group performed 3
sets in upper-body exercises and 1 set in leg exercises,
whereas the other group performed 3 sets in leg exercises groups from before to after training. The level of signifi-
cance was set at p ⱕ 0.05.
and 1 set in upper-body exercises. All exercises were per-
formed in an explosive manner in the concentric phase, RESULTS
while the eccentric phase had a slower speed (approxi-
mately 2–3 seconds). The increase in 1RM from week 0 to 11 in the lower-body
Subjects were allowed to complete no more than 1 exercises was significantly higher in the 3L-1UB group
bout of endurance training per week during the interven- than in the 1L-3UB group (41 vs. 21%; p ⬍ 0.001; Figure
tion. This was controlled with a training diary. 1 left panel). There were no significant differences be-
It has been indicated that protein administration be- tween groups in 1RM changes in upper-body exercises
fore strength training may stimulate the protein synthe- (Figure 1 right panel). Peak torque in maximal isokinetic
sis and thus positively affect the strength and hypertro- knee extensions increased significantly more in the 3L-
phy response to strength training (35). In an attempt to 1UB group than in the 1L-3UB group (16 vs. 8%; p ⫽
standardize the protein and energy consumption in the 0.03), while the increase in knee-flexion peak torque was
critical time around each bout of exercise, all subjects ate not statistically different between groups (p ⫽ 0.08; Fig-
a protein chocolate bar before each bout of exercise, and ure 2).
consumed energy drinks during each bout of exercise There were no differences between groups in total
throughout the intervention. All subjects completed at training volume (kg ⫻ reps ⫻ set). During the first train-
ing week, both the 1L-3UB group and 3L-1UB group had
least 91% of the workouts.
the same total training volume (sum of all exercises),
Statistical Analyses 9,201 ⫾ 438 kg and 9,676 ⫾ 532 kg, respectively. Due to
reduction in repetitions and increase in load, the total
All values given in the text and figures are mean ⫾ SE. training volume was the same in the last training week
Unpaired t-tests were used to compare the relative chang- with no significant difference between groups (9,310 ⫾
es from before to after training between groups. Paired t- 346 kg and 10,174 ⫾ 581 kg in the 1L-3UB group and
tests were used to test for significant changes within 3L-1UB group, respectively). The 3L-1UB group in-
160 RøNNESTAD, EGELAND, KVAMME ET AL.

creased their training load in the leg exercises to a great-

er extent than the 1L-3UB group in both absolute and
relative terms (p ⱕ 0.02; Figure 3), while there was no
difference between groups in upper-body exercises.
The CSA of the thigh muscles increased significantly
more in the 3L-1UB group than in the 1L-3UB group (11
vs. 7%; p ⫽ 0.01). When the thigh was divided into knee-
extensors and knee-flexors, the increase in CSA in the 3L-
1UB group was still superior to the 1L-3UB group (p ⱕ
0.05; Figure 4). Independent of groups, the largest in-
crease in CSA of the knee-extensors was observed in the
distal region, and for the knee-flexors the largest increase
in CSA was observed in the proximal region (data not FIGURE 4. Percentage increase in cross sectional area (CSA)
shown). Regarding CSA changes in the trapezius muscle, of knee-extensors and knee-flexors from pre- to posttest in the
there was no difference between the 1L-3UB group and 1L-3UB (1 set in leg exercises and 3 sets in upper-body
the 3L-1UB group (13.9 ⫾ 2.5% and 9.7 ⫾ 1.4%, respec- exercises) and 3L-1UB group (3 sets in leg exercises and 1 set
tively). in upper-body exercises). * Significant different from baseline
There were no significant differences between groups (p ⬍ 0.01). # Significant differences between groups (p ⱕ 0.05).
regarding changes in lean body mass and fat mass, but
it was only 5 subjects in each group (Figures 5 and 6).
The 3L-1UB group increased their body weight to a great- ing, and strength test specificity were similar in the 2
er extent than the 1L-3UB group (p ⬍ 0.03, Figure 7). training groups, the additional improvements in strength
During the intervention there was no difference be- are likely the result of a more effective stimulus induced
tween groups in intake of energy, protein, carbohydrate, by multiple sets in the leg muscles. Although differences
or fat (Table 2). in adaptation between upper- and lower-body have not
been subject to detailed examination in earlier studies, a
DISCUSSION closer examination reveals no differences between 1 and
Both the 1L-3UB and 3L-1UB groups significantly im-
proved 1RM in all exercises during the training interven-
tion. However, 3-set strength training was superior to 1-
set regarding 1RM increase in leg exercises, whereas the
increase in 1RM for the upper-body exercises was iden-
tical with 1- and 3-set training. The 1RM results are sup-
ported by the larger increase in peak isokinetic knee-ex-
tension and knee-flexion torque with 3-set training com-
pared to 1-set training. In addition, the greater strength
increase in leg exercises with 3 sets compared to 1 set
was concomitant with a greater increase in leg muscle
mass with 3 sets.
Our results show that leg muscles respond with a larg- FIGURE 5. Relative changes in lean body mass after 5 and 11
er gain in 1RM and muscle mass when a moderate weeks of strength training in the lower body (left) and upper
strength training volume (3 sets) is compared with a low body (right). 1L-3UB ⫽ 1 set in leg exercises and 3 sets in
upper-body exercises (n ⫽ 5); 3L-1UB ⫽ 3 sets in leg exercises
training volume (1 set) in the early phase of adaptation and 1 set in upper-body exercises (n ⫽ 5). * Significant
to strength training. However, this difference in response difference from baseline (p ⬍ 0.05).
to 1- and 3-set strength training was not evident in up-
per-body muscles. Because training variables such as in-
tensity of training, repetition velocity, frequency of train-

FIGURE 3. Mean training weight (kg) during the 11-week

training intervention in the leg exercises (left) and upper-body
exercises (right). 1L-3UB ⫽ 1 set in leg exercises and 3 sets in
upper-body exercises; 3L-1UB ⫽ 3 sets in leg exercises and 1 FIGURE 6. Relative changes in total fat mass after 5 and 11
set in upper-body exercises. * Significant difference from weeks of strength training. 1L-3UB ⫽ 1 set in leg exercises
baseline (p ⬍ 0.01). # Significantly larger increase in 3L-1UB and 3 sets in upper-body exercises (n ⫽ 5); 3L-1UB ⫽ 3 sets in
group compared with 1L-3UB group in both absolute and leg exercises and 1 set in upper-body exercises (n ⫽ 5).
relative terms (p ⱕ 0.02). * Significant difference from baseline (p ⬍ 0.05).
 EFFECTS OF ONE- AND THREE-SET STRENGTH TRAINING 161

3 sets in the upper-body muscles (7, 22, 25, 31, 33) and
superiority of 3 sets over 1 set in leg muscles (17, 22, 25,
28) regarding strength gains in relatively untrained sub-
jects. Unfortunately, in other studies (18, 29) the results
were not divided into upper- and lower-body muscles,
making it impossible to determine whether there was a
difference.
Based on the results of the present study, it is difficult
to point out the reasons why upper-body muscles are not
as responsive to differences in training volume as leg
muscles. One possible explanation is that we use leg mus-
cles in daily-life activities to a greater extent than our
upper-body muscles. As a consequence, some of the
growth potential in the leg muscles might already be
reached through daily activities, meaning they are better
trained than the upper-body muscles in subjects not com-
mitted to regular strength training. Furthermore, trained
muscles seem to benefit more from a larger training vol-
ume than relatively untrained muscles (15, 24, 26, 27).
This is in line with the way Paulsen et al. (22) explained FIGURE 7. Percentage change in body weight during the
strength training period in 1L-3UB (1 set in leg exercises and
similar findings. However, if the upper-body muscles are
3 sets in upper-body exercises) and 3L-1UB group (3 sets in
less trained than leg muscles, a greater increase in rela- leg exercises and 1 set in upper-body exercises). * Significant
tive strength should be expected in upper-body muscles difference from baseline (p ⬍ 0.01) # Significant differences
than in the leg muscles. This was, however, not the case between groups (p ⬍ 0.03).
since relative strength increased more in the leg muscles
than in the upper-body muscles. Based on this finding the
opposite conclusion may be drawn; the upper-body mus-
cles need a higher training volume (⬎3 sets) than leg in the lower body (13) and that the percentage of fat de-
muscles to benefit from multiple-set strength training creased to a similar extent in both groups.
protocols. In support of this view, McBride et al. (19) The mechanisms behind superior gains in muscle
found that 6 sets in biceps curl were superior to 1 set mass after 3-set strength training in the lower-body mus-
during 12 weeks of training in untrained subjects. How- cles, but not in the upper-body muscles, remain unclear.
ever, the translation of these results into a normal However, testosterone and growth hormone (GH) are
strength training setting may be questioned because the known to be involved in the anabolic processes in the
subjects conducted only 2 exercises (leg press and biceps muscle cell, and hypertrophy may, therefore, be stimu-
curl) per workout 2 times per week. lated by changes in these hormones (16). In the present
study we cannot exclude the possibility that increased se-
Greater increase in muscle mass seems to be the ma-
rum concentrations of anabolic hormones elicited by 3
jor explanation of superior strength gains in the 3-set
sets of the leg muscles, contributed to the strength in-
group compared with the 1 set in the leg muscles. How-
crease in the upper-body muscles. It has been shown that
ever, differences in neural adaptations have also been
acute GH and testosterone responses are larger in 3-set
suggested as an explanation for the superiority of 3-set
than in 1-set strength exercise protocols (11). However,
strength training compared with 1-set (12). We did not because total training volume, intensity, and rest be-
measure neural adaptations like changes in muscle acti- tween sets were similar in both groups, differences in an-
vation, so we cannot exclude this as a possible explana- abolic hormone secretion were probably negligible com-
tion for the difference between groups. However, in a re- pared to traditional 3- vs. 1-set protocols.
cent review, it was concluded that untrained subjects only Another possible explanation of the different respon-
have a minor activation deficit of their muscle in simple siveness to training volume between upper- and lower-
movements (30). Since all strength measurements were body muscles might be intrinsic differences. For example,
conducted in machines and the movements were in the content of androgen receptors is higher in upper-body
straight lines with small coordinative challenges, the ex- muscles (trapezius) than in leg muscles (vastus lateralis),
ercises used in the present study can be defined as sim- and androgen receptors are less sensitive to strength
ple. Since the window of neural adaptations in simple training in the lower-body muscles compared with the up-
movements seems to be relatively narrow, it is not likely per-body muscles (14). It might, therefore, be hypothe-
that differences in neural adaptations explain the supe- sized that muscles in the lower body to a greater extent
riority of 3 sets in the leg exercises. are dependent on training volume due to the apparent
The 3L-1UB group increased their body weight sig- lack of up-regulation of androgen receptors compared to
nificantly more than the 1L-3UB group. Since there were upper-body muscles. Furthermore, it is possible that after
no differences between the groups in upper-body muscle the early phase of adaptation, training volume becomes
hypertrophy, the superior hypertrophy of the lower-body more important also in upper-body muscles due to re-
muscles in the 3L-1UB group is likely to explain the dif- duced changes in androgen receptors.
ference in body weight gain between groups. The latter is In conclusion, 3 sets of strength training on lower-
supported by the tendency toward superior increase in body muscles is superior to 1 set during the first 11 weeks
lean body mass in the lower body in the 3L-1UB group of strength training in untrained men. There seems to be
compared with the 1L-3UB group, while no differences no difference between 1 set and 3 sets in the upper-body
were found in the upper body. This is also reinforced by muscles during this first phase of adaptation to strength
the fact that there is a greater part of total muscle mass training. The superiority of the 3-set protocol on the low-
162 RøNNESTAD, EGELAND, KVAMME ET AL.

TABLE 2. Energy, protein, carbohydrate, and fat intake before start of training and in training weeks 5 and 10. Values are means
⫾ SE.*
1L-3UB 3L-1UB
Nutrient Pre 5th week 10th week Pre 5th week 10th week
Energy intake (KJ·d ) 11,435 ⫾ 704
⫺1
12,114 ⫾ 948 12,009 ⫾ 878 11,783 ⫾ 802 12,854 ⫾ 768 11,983 ⫾ 754
Protein (g·d⫺1) 106 ⫾ 6.1 115 ⫾ 8.1 119 ⫾ 8.9 109 ⫾ 6.3 136 ⫾ 7.8 109 ⫾ 6.2
Protein (g·kg⫺1·d⫺1) 1.3 ⫾ 0.1 1.4 ⫾ 0.1 1.4 ⫾ 0.1 1.4 ⫾ 0.1 1.7 ⫾ 0.1 1.3 ⫾ 0.1
Carbohydrate (g·d⫺1) 302 ⫾ 18 360 ⫾ 25 335 ⫾ 24 307 ⫾ 25 328 ⫾ 20 356 ⫾ 20
Carbohydrate 3.7 ⫾ 0.3 4.3 ⫾ 0.3 4.0 ⫾ 0.4 3.9 ⫾ 0.4 4.1 ⫾ 0.4 4.4 ⫾ 0.5
(g·kg⫺1·d⫺1)
Fat (g·d⫺1) 94 ⫾ 5.9 94 ⫾ 11.3 96 ⫾ 9.8 97 ⫾ 9.9 111 ⫾ 10.6 88 ⫾ 11.3
* 1L-3UB ⫽ 1 set in leg exercises and 3 sets in upper-body exercises; 3L-1UB ⫽ 3 sets in leg exercises and 1 set in upper-body
exercises.

er-body muscles to improve strength was mainly caused 14. KADI, F., P. BONNERUD, A. ERIKSSON, AND L.E. THORNELL. The expres-
sion of androgen receptors in human neck and limb muscles: Effects of
by a greater increase in muscle mass.
training and self-administration of androgenic-anabolic steroids. Histo-
chem. Cell Biol. 113:25–29. 2000.
PRACTICAL APPLICATIONS 15. KRAEMER, W.J., K. ADAMS, E. CAFARELLI, G.A. DUDLEY, C. DOOLY, M.S.
In the first 11 weeks of strength training there seems to FEIGENBAUM, S.J. FLECK, B. FRANKLIN, A.C. FRY, J.R. HOFFMAN, R.U.
be no difference between 1-set and 3-set protocols regard- NEWTON, J. POTTEIGER, M.H. STONE, N.A. RATAMESS, AND T. TRIPLETT-
MCBRIDE. American College of Sports Medicine position stand. Progres-
ing strength gain and hypertrophy in the upper-body
sion models in resistance training for healthy adults. Med. Sci. Sports
muscles. However, 3 sets seem to be superior to 1 set in Exerc. 34:364–380. 2002.
the lower-body muscles regarding the same parameters. 16. KRAEMER, W.J., AND S.A. MAZZETTI. Hormonal mechanisms related to
Therefore, if the aim is to get optimal strength and hy- the expression of muscle strength and power. In: Strength and Power in
pertrophy in previously untrained subjects, and the op- Sport. P.V. Komi, ed. Oxford: Blackwell Science, 2003. pp. 73–95.
tion is between 1 and 3 sets, we recommend 1 set on the 17. KRAMER, J.B., M.H. STONE, H.S. O’BRYANT, M.S. CONLEY, R.L. JOHNSON,
D.C. NIEMAN, D.R. HONEYCUTT, AND T.P. HOKE. Effects of Single vs.
upper-body muscles and 3 sets strength training on the multiple sets of weight training: Impact of volume, intensity, and vari-
lower-body muscles during the first 11 weeks of training. ation. J. Strength Cond. Res. 11:143–147. 1997.
Based on the principle of overload and progression, it is 18. MARX, J.O., N.A. RATAMESS, B.C. NINDL, L.A. GOTSHALK, J.S. VOLEK, K.
likely that after the first period of adaptation, multiple DOHI, J.A. BUSH, A.L. GOMEZ, S.A. MAZZETTI, S.J. FLECK, K. HÄKKINEN,
sets will be superior also in the upper-body muscles. In R.U. NEWTON, AND W.J. KRAEMER. Low-volume circuit versus high-vol-
ume periodized resistance training in women. Med. Sci. Sports Exerc. 33:
addition, it might be that the upper-body muscles need
635–643. 2001.
more than 3 sets to benefit from multiple-set strength 19. MCBRIDE, J.M., J.B. BLAAK, AND T. TRIPPLETT-MCBRIDE. Effects of re-
training programs. sistance exercise volume and complexity on EMG, strength, and regional
body composition. Eur. J. Appl. Physiol. 90:626–632. 2003.
REFERENCES 20. MESSIER, S.P., AND M.E. DILL. Alterations in strength and maximal ox-
1. BERGER, R.A. Effect of varied weight training programs on strength. Res. ygen uptake consequent to Nautilus circuit weight training. Res. Q. 56:
Q. 33:168–181. 1962. 345–351. 1985.
2. BINGHAM, S.A. The dietary assessment of individuals: Methods, accura- 21. MOSS, B.M., P.E. REFSNES, A. ABILDGAARD, K. NICOLAYSEN, AND J. JEN-
cy, new techniques and recommendations. Nutr. Abstr. Rev. 57:705–743. SEN. Effects of maximal effort strength training with different loads on
1987. dynamic strength, cross-sectional area, load-power and load-velocity re-
3. BLACK, A.E., G.R. GOLDBERG, G.E. JEBB, M.B.E. LIVINGSTONE, T.J. COLE, lationships. Eur. J. Appl. Physiol. Occup. Physiol. 75:193–199. 1997.
AND A.M. PRENTICE. Critical evaluation of energy intake data using fun- 22. PAULSEN, G., D. MYKLESTAD, AND T. RAASTAD. The influence of volume
damental principles of energy physiology: 2. Evaluating the results of of exercise on early adaptations to strength training. J. Strength Cond.
public surveys. Eur. J. Clin. Nutr. 45:583–599. 1991. Res. 17:115–120. 2003.
4. CARPINELLI, R.N. Science versus opinion. Br. J. Sports Med. 38:240–242. 23. PEREIRA, M.I.R., AND P.S.C. GOMES. Movement velocity in resistance
2004.
training. Sports Med. 33:427–438. 2003.
5. CARPINELLI, R.N., AND R.M. OTTO. Strength training: Single versus mul-
24. PETERSON, M.D., M.R. RHEA, AND B.A. ALVAR. Maximizing strength de-
tiple sets. Sports Med. 26:73–84. 1998.
velopment in athletes: A meta-analysis to determine the dose-response
6. CARPINELLI, R.N., AND R.M. OTTO. Strength training: Single versus mul-
tiple sets. Sports Med. 27:412–416. 1999. relationship. J. Strength Cond. Res. 18:377–382. 2004.
7. CURTO, M.A., AND M.M. FISHER. The effect of single vs. multiple sets of 25. RHEA, M.R., B.A. ALVAR, S.D. BALL, AND L.N. BURKETT. Three sets of
resistance exercise on strength in trained males [Abstract]. Med. Sci. weight training superior to 1 set with equal intensity for eliciting
Sports Exerc. 31:S114. 1999. strength. J. Strength Cond. Res. 16:525–529. 2002b.
8. ESMARCK, B., J.L. ANDERSEN, S. OLSEN, E.A. RICHTER, M. MIZUNO, AND 26. RHEA, M.R., B.A. ALVAR, L.N. BURKETT, AND S.D. BALL. A meta-analysis
M. KJAER. Timing of postexercise protein intake is important for muscle to determine the dose response for strength development. Med. Sci.
hypertrophy with resistance training in elderly humans. J. Physiol. 535: Sports Exerc. 35:456–464. 2003.
301–311. 2001. 27. RHEA, M.R., A.A. BRENT, AND L.N. BURKETT. Single versus multiple sets
9. FLECK, S.J., AND W.J. KRAEMER. Designing Resistance Training Pro- for strength: A meta-analysis to address the controversy. Res. Q. Exerc.
grams (3rd ed.). Champaign, IL: Human Kinetics, 2004. Sport 73:485–488. 2002a.
10. GALVAO, D.A., AND D.R. TAAFFE. Single- vs. multiple-set resistance train- 28. SANBORN, K., R. BOROS, J. HRUBY, B. SCHILLING, H.S. O’BRYANT, R.L.
ing: Recent developments in the controversy. J. Strength Cond. Res. 18: JOHNSTON, T. HOKE, M.E. STONE, AND M.H. STONE. Short-term perfor-
660–667. 2004. mance effects of weight training with multiple sets not to failure vs. a
11. GOTSHALK, L.A., C.C. LOEBEL, B.C. NINDL, M. PUTUKIAN, W.J. SEBAS-
single set to failure in women. J. Strength Cond. Res. 14:328–331. 2000.
TIANELLI, R.U. NEWTON, K. HÄKKINEN, AND W.J. KRAEMER. Hormonal
29. SCHLUMBERGER, A., J. STEC, AND A. SCHMIDTBLEICHER. Single vs. mul-
responses of multiset versus single-set heavy-resistance exercise proto-
tiple-set strength training in women. J. Strength Cond. Res. 15:284–289.
cols. Can. J. Appl. Physiol. 22:244–255. 1997.
12. HASS, C.J., L. GARZARELLA, D. DEHOYOS, AND M.L. POLLOCK. Single ver- 2001.
sus multiple sets in long-term recreational weightlifters. Med. Sci. Sports 30. SHIELD, A., AND S. ZHOU. Assessing voluntary muscle activation with the
Exerc. 32:235–242. 2000. twitch interpolation technique. Sports Med. 34:253–267. 2004.
13. JANSSEN, I., S.B. HEYMSFIELD, Z. WANG, AND R. ROSS. Skeletal muscle 31. SILVESTER, L.J., C. STIGGINGS, C. MCGOWN, AND G.B. BRYCE. The effect
mass distribution in 468 men and women aged 18–88 yr. J. Appl. Physiol. of variable resistance and free-weight training programs on strength and
89:81–88. 2000. vertical jump. Natl. Strength Cond. Assoc. J. 3:30–33. 1982.
 EFFECTS OF ONE- AND THREE-SET STRENGTH TRAINING 163

32. STARKEY, D.B., M.L. POLLOCK, Y. ISHIDA, M.A. WELSCH, W.F. BRECHUE, 36. WELSCH, M.A., W.F. BREUCHE, M.L. POLLOCK, D.B. STARKEY, AND J.E.
J.E. GRAVES, AND M.S. FEIGENBAUM. Effect of resistance training volume GRAVES. Effect of reduced training volume on bilateral isometric knee
on strength and muscle thickness. Med. Sci. Sports Exerc. 28:1311–1320. flexion/extension torque [Abstract]. Med. Sci. Sports Exerc. 26:S189.
1996. 1994.
33. STOWERS, T., J. MCMILLAN, D. SCALA, V. DAVIS, G.D. WILSON, AND M.H.
STONE. The short-term effects of three different strength-power training
methods. Natl. Strength Cond. Assoc. J. 5:24–27. 1983. Acknowledgments
34. THORNELL, L.E., M. LINDSTRøM, V. RENAULT, V. MOULY, AND G.S. BUT-
LER-BROWNE. Satellite cells and training in the elderly. Scand. J. Med.
This study was supported by grants from the Swedish National
Sci. Sports 13:48–55. 2003. Center for Research in Sports (68/04 and 80/05).
35. TIPTON, K.D., B.B. RASMUSSEN, S.L. MILLER, S.E. WOLF, S.K. OWENS-
STOWALL, B.E. PETRINI, AND R.R. WOLFE. Timing of amino acid-carbo-
hydrate ingestion alters anabolic response of muscle to resistance exer- Address correspondence to Truls Raastad, truls.
cise. Am. J. Physiol. Endocrinol. Metab. 281:E197–E206. 2001. [email protected].