Comparison of Muscle Growth and Dynamic Strength Adaptations Induced by Unilateral and Bilateral Resistance Training: A Systematic Review and Meta-analysis

Abstract

Background
Currently, great debate exists over the proposed superiority of some resistance exercises to induce muscular adaptations. For example, some argue that unilateral exercise (meaning one limb at a time) is superior to bilateral exercises (meaning both limbs). Of note, an evidence-based answer to this question is yet to be determined, particularly regarding muscle hypertrophy.

Objective
This systematic review and meta-analysis aimed to compare the effects of unilateral versus bilateral resistance training on muscle hypertrophy and strength gains.

Methods
A thorough literature search was performed using PubMed, Scopus, and Web of Science databases. The Cochrane Risk of Bias tool 2 (RoBII) tool was used to judge the risk of bias. Meta-analyses were performed using robust variance estimation with small-sample corrections.

Results
After retrieving 703 studies, 9 met the criteria and were included in the meta-analyses. We found no significant differences in muscle hypertrophy between bilateral and unilateral training [effect size (ES): − 0.21, 95% confidence interval (95% CI): − 3.56 to 3.13, P = 0.57]. Bilateral training induced a superior increase in bilateral strength (ES: 0.56, 95% CI: 0.16–0.96, P = 0.01). In contrast, unilateral training elicited a superior increase in unilateral strength (ES: − 0.65, 95% CI: − 0.93 to − 0.37, P = 0.001). Overall, studies presented moderate risk of bias.

Conclusion
On the basis of the limited literature on the topic, we found no evidence of differential muscle hypertrophy between the two exercise selections. Strength gains appear to follow the principle of specificity.

Full text

1
Comparison of Muscle Growth and Dynamic Strength Adaptations Induced by Unilateral
1
and Bilateral Resistance Training: A Systematic Review and Meta-Analysis
2
Short title: Amount of muscle mass and muscular adaptations
3
Authors: Witalo Kassiano1,2, João Pedro Nunes1,3, Bruna Costa1, Alex S. Ribeiro4, Jeremy P.
4
Loenneke2, Edilson S. Cyrino1
5
Affiliations: 1Metabolism, Nutrition and Exercise Laboratory, Physical Education and Sport
6
Center, State University of Londrina, Londrina, Paraná, Brazil; 2Kevser Ermin Applied
7
Physiology Laboratory, The University of Mississippi, University, Mississippi, United States of
8
America; 3School of Medical and Health Sciences, Edith Cowan University, Joondalup,
9
Australia; 4University of Coimbra, FCDEF, Coimbra, Portugal.
10
*Correspondence author: Witalo Kassiano. E-mail: acc.witalo@gmail.com | ORCID number:
11
0000-0002-0868-8634. Metabolism, Nutrition, and Exercise Laboratory. Physical Education and
12
Sport Center, State University of Londrina, Rodovia Celso Garcia, km 380, 86057-970,
13
Londrina, Brazil.
14
15

2
ABSTRACT
16
Background Nowadays, great debate exists over the proposed superiority of some resistance
17
exercises to induce muscular adaptations. For example, some argue that unilateral exercise
18
(meaning one limb at a time) is superior to bilateral exercises (meaning both limbs). Of note, an
19
evidence-based answer to this question is yet to be determined, particularly on muscle
20
hypertrophy.
21
Objective This systematic review and meta-analysis aimed to compare the effects of unilateral
22
versus bilateral resistance training on muscle hypertrophy and strength gains.
23
Methods A thorough literature search was performed using PubMed, Scopus, and Web of
24
Science databases. The RoBII tool was used to judge the risk of bias. Meta-analyses were
25
performed using robust variance estimation with small-sample corrections.
26
Results After retrieving 703 studies, nine met the criteria and were included in the meta-
27
analyses. We found no significant differences in muscle hypertrophy between bilateral and
28
unilateral training [Effect size (ES) = -0.21, 95% confidence interval (95% CI): -3.56 to 3.13, P =
29
0.57]. Bilateral training induced superior increase in bilateral strength (ES = 0.56, 95% CI: 0.16
30
to 0.96, P = 0.01). In contrast, unilateral training elicited superior increase in unilateral strength
31
(ES = -0.65, 95% CI: -0.93 to -0.37, P = 0.001). Overall, studies presented moderate risk of bias.
32
Conclusion Based on the limited literature on the topic, we found no evidence of differential
33
muscle hypertrophy between the two exercise selections. Strength gains appear to follow the
34
principle of specificity.
35
Key points:
36

3
- Bilateral exercises (meaning both limbs) are the standard choice when writing resistance
37
training program for inducing muscular adaptations (e.g., increased strength and muscle
38
hypertrophy).
39
- However, recently, some have suggested that unilateral exercises (meaning one limb at time)
40
may induce greater adaptations. Of note, an evidence-based answer to this question is yet to be
41
determined, particularly on muscle hypertrophy.
42
- We conducted a systematic review and meta-analysis and, based on the limited literature on the
43
topic, we found no evidence of differential muscle hypertrophy between the two exercise
44
selections. Strength gains appear to follow the principle of specificity.
45

4
1 INTRODUCTION
46
Exercise selection is a fundamental aspect when designing resistance training programs aiming
47
to optimize muscular adaptations (e.g., muscle hypertrophy and strength) [1]. Among the several
48
features that differentiate resistance exercises, one of the main ones that may affect adaptations is
49
whether it is performed bilaterally (meaning both limbs simultaneously) or unilaterally (meaning
50
one limb at time) [2, 3]. Traditionally, bilateral exercises (e.g., back squat, bilateral knee
51
extension, barbell biceps curl) are selected as the main exercises. In contrast, equivalent
52
unilateral exercises (e.g., rear elevated split squat, one-leg knee extension, one-arm biceps curl)
53
are selected as supplementary, in the context of reducing inter-limb asymmetry or rehabilitation
54
[2]. Of note, nowadays some researchers and coaches have been arguing that unilateral/single-
55
limb exercises may be as or, in some cases, more effective than bilateral/double-limb exercises to
56
induce muscular adaptations [2-6]. The reason for this may lie in the differences between the two
57
types of resistance exercises⎯e.g., differences in the total amount of muscle mass involved and
58
force production⎯and physiological-related consequences [2, 5, 7].
59
Different lines of investigation have shown that large muscle mass exercise induces a
60
greater cardiovascular response [8, 9], a quicker time to task failure [4, 10], and less local muscle
61
fatigue [5] than small muscle mass exercises in response to a similar maximal workload. This
62
indicates that even in maximal effort, the stimulus to a target muscle may be limited during large
63
muscle mass exercises [4, 5]. Parallel to this, the production of force generated when both limbs
64
contract simultaneously is frequently lower than the summed forces from each limb⎯a
65
phenomenon referred to as a bilateral force deficit. Although the mechanisms explaining bilateral
66
deficit are unclear, one hypothesis is that there is a reduced neural drive to the target muscle
67
when contracting two limbs simultaneously [11]. Based on these observations, it has been
68

5
suggested that unilateral resistance exercises (i.e., smaller muscle mass exercise) would enhance
69
muscular adaptations in comparison to bilateral exercises (i.e., larger muscle mass exercise) [2,
70
4, 5]. However, findings on muscle hypertrophy are controversial, with studies reporting similar
71
muscle growth [12], or results more favorable for unilateral training [13]. This makes it difficult
72
to know if bilateral and unilateral resistance training has a differential effect on muscle
73
hypertrophy.
74
Regarding strength, the most parsimonious hypothesis is that adaptations would follow
75
the principle of specificity; that is, bilateral training induces greater bilateral strength gains than
76
unilateral training, and unilateral training elicits greater unilateral strength gains than bilateral
77
training. Alternatively, it has been suggested that the existence of bilateral force deficit may
78
affect the magnitude of strength adaptations induced by bilateral and unilateral resistance
79
training [2, 3]. Specifically, by observing that force production during unilateral movements can
80
account for more than 50% of the total force produced during equivalent bilateral exercise, some
81
argue that unilateral training would confer an advantage, in part, due to possibility of training
82
with higher load per limb [2]. Based on this, it has been suggested that unilateral training would
83
elicit more favorable strength gains [2]; e.g., greater unilateral strength gains parallel to similar
84
bilateral strength improvements compared to bilateral training.
85
In fact, a recent meta-analysis gives some support to this hypothesis [14]. Zhang et al.
86
[14] found that unilateral training increased unilateral strength more than bilateral training,
87
whereas bilateral training did not result in greater bilateral strength gains than unilateral training
88
[14]. However, another meta-analysis on this topic observed opposite findings [15]; Liao et al.
89
[15] found that bilateral training increased bilateral strength more than unilateral training, but
90
unilateral training did not result in greater unilateral strength gains than bilateral training [15].
91

6
The conflicting data between studies may be attributed to differences in the variables analyzed as
92
a proxy for maximum strength⎯Zhang et al. [14] included exclusively isotonic strength
93
measurements whereas Liao et al. [15] included different strength measurements (e.g., one-
94
repetition maximum, isokinetic peak torque)⎯and how the authors dealt with correlated effect
95
sizes; both studies did not use meta-analysis model that took into account the dependence
96
between effect sizes when certain studies presented more than one strength measure [14, 15].
97
Thus, it remains to be determined if there is differential muscle growth and dynamic
98
maximum strength adaptations between bilateral and unilateral resistance training. In parallel to
99
a scientific inquiry perspective, determining potential differences between the two resistance
100
exercise modes is fundamental to enable coaches and practitioners to make evidence-based
101
decisions with exercise selection. In this context, a synthesis of evidence on the effects of
102
bilateral and unilateral training on muscle growth is imperative, as well as a meta-analysis that
103
overcomes the delimitations of previous meta-analytical studies (i.e., include more homogeneous
104
dynamic strength measurements and adopt a meta-analysis model that considers the dependence
105
between effect sizes when this is the case) would be of great value. Thus, in this systematic
106
review and meta-analysis, we aimed to compare the effects of unilateral versus bilateral
107
resistance training on muscle hypertrophy and maximum strength changes.
108
109
2 METHODS
110
2.1 Research question
111
This systematic review was conducted under Preferred Reporting Items for Systematic Review
112
and Meta-Analysis Protocol (PRISMA) and Prisma in Exercise, Rehabilitation, Sport Medicine,
113
and Sports Science (PERSiST) [16, 17]. This meta-analysis was not pre-registered. The research
114

7
questions were defined according to the population, intervention, comparator, and outcomes (PICO)
115
framework, as follows:
116
Population: Individuals with or without resistance training experience, with no restrictions on sex or
117
age. Investigations including individuals with chronic diseases, musculoskeletal disorders, or
118
injuries were excluded.
119
Intervention: Longitudinal randomized trials employing parallel-group design comparing unilateral
120
(i.e., training one limb at a time) and bilateral (i.e., training both limbs at the same time) resistance
121
training programs lasting ≥ 3 weeks.
122
Comparator: An experimental trial comparing unilateral versus bilateral dynamic resistance
123
exercises.
124
Outcomes: Changes in muscle hypertrophy (assessed by muscle thickness, cross-sectional area,
125
volume, or muscle mass) and/or changes in dynamic strength (assessed by repetition-maximum
126
[1−5RM] strength tests). Studies including only non-specific strength measures (e.g., isometric, or
127
isokinetic strength) were not included.
128
2.2 Literature search
129
To conduct the review, we searched PubMed/MEDLINE, Web of Science, and Scopus electronic
130
databases up to December 2023. Only peer-reviewed articles in English were selected for
131
inclusion; citations from scientific conferences were excluded from analysis. We used “subject-
132
key-term + free word” search format. The following keywords inclusive of five main terms as
133
unilateral, bilateral, resistance training, muscle hypertrophy, and strength were used and
134
combined under Boolean’s language with operators AND and OR. Term 1: “unilateral”, “single
135
limb”, “one limb”, “one leg”, “one arm”. Term 2: “bilateral”, “double limb”, “two limbs”, “two
136
legs”, “two arms”. Term 3: “resistance training”, “resistance exercise”, “strength training”,
137

8
“strength exercise”. Term 4: “muscle hypertrophy”, “muscle thickness”, “cross-sectional area”,
138
“muscle mass”, “muscle size”, “muscle volume”. Term 5: “maximum strength”, “maximum
139
force”, “one-repetition maximum”, “repetition maximum”. The title and abstract of each study
140
were inspected for relevance, and the full texts were then scrutinized for those initially appearing
141
to meet inclusion criteria. Studies in which the abstracts did not provide enough information
142
according to our inclusion criteria were retrieved for full-text evaluation. In the selected articles, the
143
reference lists and Google Scholar citations were screened for additional manuscripts. In addition,
144
the list of articles that cited the included studies were screened. The first author WK completed the
145
search.
146
2.3 Study coding, data extraction, and risk of bias
147
The following data were extracted from the included studies: study characteristics (author, year,
148
sample size, and study design), participants demographics (age, sex, and resistance training
149
experience), resistance training program (duration, frequency, resistance exercise, number of sets,
150
rest interval), and outcome measures. We then coded data for the included studies’ pretraining and
151
posttraining means and standard deviations. WK and JPN independently extracted the data from the
152
included papers. After the data extraction, BC confirmed the precision of the extracted data. The
153
quality assessment of included articles was performed using the Cochrane the risk of bias tool 2
154
(RoBII). Articles were assessed for hypertrophy/strength outcomes for bias: 1) arising from the
155
randomization process, 2) due to deviations from intended intervention, 3) due to missing outcome
156
data, 4) in the measurement of the outcome and 5) in the selection of reported result. Each domain
157
was determined to be of high, moderate, or low risk of bias. Then the studies were given an overall
158
classification of high, moderate, or low risk of bias. Traffic light and weighted summary risk of bias
159
plots for included studies were produced by the online risk of bias (robvis) tool
160

9
(https://mcguinlu.shinyapps.io/robvis/) [18]. WK and JPN independently evaluated the quality of
161
the included studies, and any disagreement was resolved by consensus.
162
2.4 Statistical analysis
163
All data were analyzed by two investigators (WK and JL) in an effort to maximize accuracy.
164
Because we were interested in capturing the magnitude of the variability within the intervention
165
itself, we calculated the effect sizes for each study using the mean difference and the standard
166
deviation (SD) of the difference (commonly known as Cohen’s dz) [19]. However, we also
167
calculated the effect sizes as mean difference divided by the pooled SD of the pre and reran the data
168
in order to provide the size of the effect relative to the spread of the sample [20]. If the SD of the
169
difference was not presented but exact P value was, we calculated the t value using the inverse of
170
the cumulative distribution function. The t value was then used to calculate the change score
171
deviation. If the variability of the change was not provided (and could not be obtained from
172
available data), the SD of the change was estimated using the following formula:
173
174
SDof change = [(SDPretest)2 + (SDPosttest)2 - 󰇛2r × SDPretest × SDPosttest󰇜]
175
176
SD represents the standard deviation, and r represents the correlation coefficient between the
177
pretest and the posttest values. We used 0.8 as the pre-post correlation. The standardized effect size
178
and the standard error of this effect size were computed as follows [21]:
179
180
Standardized ES = changebilateral- changeunilateral
󰇛N1 - 1󰇜v1+󰇛N2 - 1󰇜v2
N1 + N2-2
181
182

10
Standardized SE = 󰇛N1+N2󰇜
N1× N2
+ ES2
2(N1+ N2)
183
184
ES represents the effect size, N1 represents the sample size of the bilateral group, N2
185
represents the size of the unilateral group, v1 represents the variance of the bilateral group, v2
186
represents the variance of the unilateral group, and SE represents the standard error.
187
A robust variance meta-analysis model was used to account for correlated effect sizes within
188
studies [22]. This meta-analysis model is specifically designed and used when dealing with
189
dependent effect sizes (e.g., several measures for a specific outcome assessed within a single study).
190
Statistics were performed using the robumeta package (version 2.1) within R statistical software
191
(version 3.6.3, R foundation for Statistical Computing, Vienna, Austria) and R Studio (version
192
1.4.1103, RStudio, Inc., Boston, MA, USA). In robumeta we performed a correlated effects model
193
with small-sample corrections. We adopted the default correlation of 0.8. Model weights were
194
determined using the default setting (CORR) and effect sizes are presented in standardized units.
195
Also, we performed a sensitivity analysis to determine the effect of rho and tau squared. Tau
196
squared represents the between study variance component in the correlated effects meta-regression
197
model and is calculated using the method-of-moments estimator provided in Hedges et al. [23]. I2
198
was also provided and is used to quantify the amount of variability in effect size estimates due to
199
effect size heterogeneity. We also implemented the metafor package (version 3.0-2, Restricted ML)
200
in order to report the prediction intervals. To reduce issues associated with using a normal
201
distribution, we used the argument tdist = TRUE with rma.mv function, which applies the Knapp
202
and Hartung adjustment to the analysis. Three separate comparisons were made: 1) hypertrophic
203
effects of bilateral vs. unilateral training; 2) bilateral muscular strength changes of bilateral vs.
204

11
unilateral training; 3) unilateral muscular strength of bilateral vs. unilateral training. Data are
205
presented as effect size, standard error, and 95% confidence interval and prediction intervals. An
206
Eggers test for publication bias was not performed due to the small number of studies included (a
207
general rule of thumb is to have at least 10 studies).
208
209
3 RESULTS
210
The search strategy initially yielded 699 articles plus four studies found through citations searching
211
and other source. Following the deletion of duplicates, the literature search yielded 303 records. A
212
total of 288 articles were excluded based on the title and abstract screen; 19 articles were obtained in
213
full text (15 from initial searching and four from citation searching and another source), and the
214
selection criteria were applied. Reasons for exclusion included: article did not include outcomes of
215
interest, contained data from the same sample, did not present retrievable data, employed isometric
216
training, compared exercises that differ more than in the number of limbs exercised, training and
217
strength testing did not match, or had participants changing exercises at each training session and
218
each participant progressing to the next exercise at different times within the training program (in
219
one group but not the other). Finally, nine studies were included in the meta-analyses. The
220
PRISMA flow diagram is presented in Figure 1.
221
222
*** PLEASE INSERT FIGURE 1 NEAR HERE ***
223
224
3.1 Study characteristics
225
Details from the 9 studies (n = 200 participants) included in the final analysis are presented in Table
226
1. Resistance training interventions were performed on male-only samples in seven studies [24-30],
227

12
and female-only samples in two studies [12, 13]. Seven studies investigated lower-limb exercises
228
[12, 24, 25, 27-30], one investigated upper-limb [26], and one investigated upper- and lower-limb
229
exercises [13]. The most frequent resistance exercises investigated were bilateral back squat versus
230
equivalent unilateral variations such as rear elevated split squat, Bulgarian split squat, and step-up
231
followed by bilateral versus unilateral knee extension, and bilateral versus unilateral leg press. One
232
study assessed changes in muscle hypertrophy using muscle thickness [12] and one upper and
233
lower-limb lean tissue mass [13]. Strength outcome was analyzed by 1RM tests in eight studies [12,
234
13, 24-29] and 5RM tests in one study [30]. Eight studies assessed bilateral and unilateral strength
235
[12, 13, 24-26, 28-30], while two assessed bilateral strength only [27]. For unilateral strength tests,
236
two studies provided data from the dominant limb [28, 30], two studies presented the average values
237
between both limbs [24, 29], two presented the values for both limbs [13, 26], one used the sum of
238
the values of both limbs [12], and one did not specify [25].
239
240
*** PLEASE INSERT TABLE 1 NEAR HERE ***
241
242
3.2 Muscle hypertrophy: bilateral training vs. unilateral training
243
The overall between-group effect on muscle hypertrophy using change score variability was -0.21,
244
with a standard error of 0.26, and a 95% confidence interval of -3.56 to 3.13 (Figure 2, P = 0.570, df
245
= 1). I2 was 55.4 and Tau2 was 0.19. Sensitivity analysis demonstrated that this effect was stable
246
across different Rho values. However, this effect should be interpreted with caution due to the few
247
studies available. The 95% prediction intervals from metafor ranged from -6.09 to 5.51.
248
249

13
The overall between-group effect on muscle hypertrophy using SD of pre was -0.03, with a standard
250
error of 0.06, and a 95% confidence interval of -0.83 to 0.77 (P = 0.696, studies: df=1). I2 was 0.00
251
and Tau2 was 0.00. Sensitivity analysis demonstrated that this effect was stable across different Rho
252
values. However, this effect should be interpreted with caution due to the few studies available. The
253
95% prediction intervals from metafor ranged from -0.79 to 0.68.
254
255
*** PLEASE INSERT FIGURE 2 NEAR HERE ***
256
257
3.3 Bilateral dynamic strength: bilateral training vs. unilateral training
258
The overall between-group effect on bilateral strength using change score variability was 0.56, with
259
a standard error of 0.17, and a 95% confidence interval of 0.16 to 0.96 (Figure 3, P = 0.011, studies:
260
df = 7.72). I2 was 26.3 and Tau2 was 0.07. Sensitivity analysis demonstrated that this effect was
261
stable across different Rho values. The 95% prediction intervals from metafor ranged from -0.47 to
262
1.60.
263
264
The overall between-group effect on bilateral strength using SD of pre was 0.33, with a standard
265
error of 0.12, and a 95% confidence interval of 0.04 to 0.63 (P = 0.030, df = 7.55). I2 was 0.00 and
266
Tau2 was 0.00. Sensitivity analysis demonstrated that this effect was stable across different Rho
267
values. The 95% prediction intervals from metafor ranged from -0.38 to 1.05.
268
269
*** PLEASE INSERT FIGURE 3 NEAR HERE ***
270
271
3.4 Unilateral dynamic strength: bilateral training vs. unilateral training
272

14
The overall between-group effect on unilateral strength using change score variability was -0.65,
273
with a standard error of 0.11, and a 95% confidence interval of -0.93 to -0.37 (Figure 4, P = 0.001,
274
df = 6.56). I2 was 0.0 and Tau2 was 0.00. Sensitivity analysis demonstrated that this effect was stable
275
across different Rho values. The 95% prediction intervals from metafor ranged from -1.01 to -0.17.
276
277
The overall between-group effect on bilateral strength using SD of pre was -0.42, with a standard
278
error of 0.08, and a 95% confidence interval of -0.63 to -0.21 (P = 0.002, df = 6.56). I2 was 0.0 and
279
Tau2 was 0.00. Sensitivity analysis demonstrated that this effect was stable across different Rho
280
values. The 95% prediction intervals from metafor ranged from -0.78 to -0.15.
281
282
*** PLEASE INSERT FIGURE 4 NEAR HERE ***
283
284
3.5 Risk of bias
285
From the risk of bias assessment, we observed that the nine studies presented "some concerns". The
286
domains that most frequently presented some concerns were "randomization process" (7 studies),
287
"measurement outcome" (8 studies) and "selection of the reported result" (8 studies). Concerns
288
arose most due to the lack of information on allocation sequence concealment, whether evaluators
289
were aware of the intervention received, and prespecified analysis procedures. Figure 5 shows the
290
weighted summary risk of bias plots. Figure 6 shows the traffic light risk of bias plots.
291
292
*** PLEASE INSERT FIGURE 5 NEAR HERE ***
293
*** PLEASE INSERT FIGURE 6 NEAR HERE ***
294
295

15
4 DISCUSSION
296
This systematic review with meta-analysis aimed to compare the effects of bilateral versus
297
unilateral resistance training on muscle hypertrophy and strength adaptations. The current
298
findings suggest that exercise selection, whether bilateral or unilateral, appears to influence
299
strength changes, but there is great uncertainty on the hypertrophic responses. More specifically,
300
the main results were that: 1) from the limited studies available so far, there was no evidence of a
301
difference in the magnitude of muscle hypertrophy induced by bilateral and unilateral training
302
(though this model is not stable due to insufficient data); 2) bilateral resistance training induced a
303
superior effect on increasing bilateral dynamic strength in comparison to unilateral training; 3)
304
unilateral resistance training elicited a superior increase in unilateral dynamic strength in
305
comparison to bilateral training. In the ensuing paragraphs, we discuss these results in the
306
context of the available evidence, proposing potential explanations and the limitations of current
307
literature, as well as making suggestions for future studies on this topic.
308
4.1 Muscle Strength
309
The results of the present meta-analysis support the notion that strength increases are
310
greater in the task that individuals trained, following the principle of specificity. Interestingly,
311
the two meta-analytic studies on this topic observed ambiguous results. Zhang et al. [14]
312
observed greater gains in unilateral maximal strength after unilateral training while bilateral
313
strength gains were not different between unilateral and bilateral training, thus indicating more
314
favorable results for unilateral training. In contrast, Liao et al. [15] observed greater gains in
315
bilateral strength after bilateral training while unilateral strength gains were not different
316
between the two exercises selections. We speculate that the conflicting data between these
317
studies may be attributed to (i) differences in study inclusion criteria (e.g., inclusion of data from
318

16
theses and dissertations [14]); (ii) inclusion of different strength measures (e.g., isotonic,
319
isokinetic, power) to infer about strength [15]; (iii) failure to take into account correlated effect
320
sizes in the meta-analytic model [14]. In the present meta-analysis, we included only studies
321
published in peer-reviewed journals that assessed our maximal strength outcome through
322
isotonic strength measurements (i.e., 1-5RM). Also, we included all dynamic strength
323
measurements (e.g., 1RM leg press and knee extension) from the same study adopting a meta-
324
analytic model that allows us to take into account the dependence between effect sizes. Putting
325
this into perspective, such factors together may conceivably help to explain, in part, such
326
divergent data.
327
Previously, some suggested that unilateral training might result in more favorable
328
maximal strength gains (e.g., greater unilateral strength gains and similar bilateral strength gains
329
than bilateral training). This proposition was made based on the frequent, but not unanimous,
330
observation that force production during unilateral movements can account for more than 50% of
331
the total force produced during equivalent bilateral exercise [2, 3, 31]. This could confer an
332
advantage of unilateral training compared to bilateral training (i.e., training heavier per limb) [2].
333
However, the findings of the present meta-analysis do not support this hypothesis. Of note, in
334
addition to bilateral deficits not always being observed [31], some reports actually suggest that
335
bilateral facilitation (i.e., when the force produced during bilateral contraction is greater than the
336
sum of the unilateral forces of the two limbs) [2, 32, 33] and this observation could in theory
337
confer an advantage to bilateral training. But, again, the findings of the present meta-analysis do
338
not support this hypothesis.
339
We found that bilateral strength gains were greater when training was performed
340
bilaterally. Meanwhile, unilateral strength gains were greater when training was performed
341

17
unilaterally. Although it is beyond the scope of this review, we can speculate on some
342
mechanisms that may explain these findings. For example, it has been suggested that after a
343
period of resistance training, improvements in task-specific coordination occur, which facilitate
344
greater strength increases in the trained task [34, 35]. Other possible explanations may lie in
345
neural changes [e.g., electromyographic (EMG) signal amplitude and voluntary activation]. For
346
example, a correlation was observed between changes in the EMG of both legs and changes in
347
bilateral maximal strength induced by 12 weeks of bilateral knee extension training [32]; the
348
same was observed for changes in the EMG values of right leg and unilateral maximal strength
349
of right leg (same for left leg) [32]. In the same study [32], bilateral training increased EMG of
350
both legs more than unilateral training. However, unilateral training did not increase EMG of the
351
right or left legs more than bilateral training [32]. Reinforcing the conflicting scenario, in another
352
report there was an increase in quadriceps EMG amplitude signal only after unilateral knee
353
extension resistance training [12]. Therefore, it remains to be determined whether changes in
354
EMG⎯and other proxies of neural adaptations such as voluntary activation⎯are a mechanism to
355
help explain why strength gains induced by the two exercise selections follow the principle of
356
specificity.
357
4.2 Muscle hypertrophy
358
Two studies measured changes in muscle size. One measured muscle thickness using
359
ultrasound, and another measured changes in lean tissue mass using dual energy X-ray
360
absorptiometry. From results of our meta-analysis, we found no evidence for a difference
361
between unilateral and bilateral training on muscle size changes. Traditionally, bilateral exercises
362
have been considered more effective in inducing muscular adaptations [36], partly because the
363
individual can lift more weight per repetition than unilateral equivalent exercises, which is
364

18
frequently accompanied by greater electromyographic activity [36]. Of note, the hypothesis that
365
lifting high loads could result in greater hypertrophy has been refuted [37, 38], as well as the
366
relationship between greater electromyographic activity and hypertrophy has also been
367
challenged [39, 40]. In this sense, more recently, some have argued that unilateral exercises
368
could be similar or, in some cases, more effective in inducing muscular adaptations [2]. Among
369
the arguments, one that receives attention is the potential influence of the amount of muscle mass
370
involved.
371
Theoretically, larger muscle mass exercises (e.g., bilateral leg press and knee extension)
372
could result in interruption of the exercise due to several factors other than local fatigue of the
373
target muscle [4, 5]. In contrast, smaller muscle mass exercises (e.g., unilateral leg press and
374
knee extension) would be interrupted more specifically by local fatigue of the target muscle [4,
375
5]. However, although some differences in the acute response are often observed (e.g., greater
376
peripheral fatigue and time-to-task failure in unilateral exercises) [4, 5, 7], we have no evidence
377
of difference on muscle growth magnitude. Of note, we do not completely rule out this
378
hypothesis. For example, in the Janzen et al. [13] study, only the unilateral training group—
379
which performed leg press, knee extension and leg curl—increased lean tissue mass compared to
380
the control, the bilateral group did not. In contrast, Botton et al. [12] observed increases in
381
quadriceps thickness in both training groups compared to control when performing exclusively
382
unilateral and bilateral knee extension. In another study [32], not included in the meta-analysis
383
due to insufficient data, individuals performed knee extension exclusively, and the increase in
384
quadriceps cross-sectional area between bilateral and unilateral training was not different.
385
Therefore, from the inspection of individual studies, it is possible to suggest the amount of
386

19
muscle mass may play a role when exercises involve a greater difference in muscle mass (e.g.,
387
leg press, squat). However, that remains speculative and needs to be tested.
388
4.3 Gaps, limitations, and research recommendations
389
To our knowledge, this is the first systematic review and meta-analysis on muscle
390
hypertrophy in response to bilateral and unilateral resistance training. The synthesis of the
391
available literature aids in a better understanding of the role of exercise selection in muscular
392
adaptations and in identifying research gaps on this topic. Noteworthy, our meta-analysis has
393
limitations that deserve to be highlighted. The number of studies included in the muscle
394
hypertrophy meta-analysis was small. Also, most studies did not clearly describe some
395
characteristics of the training and the exercise execution. For example, studies rarely described
396
whether there was a rest interval between limbs in the unilateral group and whether the exercise
397
was performed to or close to task failure. Therefore, future studies should make the training
398
program description more detailed. Part of the studies included in the strength meta-analysis
399
estimated the 1RM. In this sense, it is difficult to know whether this affected muscle strength
400
results.
401
Another potential limitation of the available data is the duration of the intervention. For
402
example, most reports (i.e., 8 studies) were 4–12 weeks in duration, with only one study [13]
403
lasting >20 weeks. Some argue that there may be differences in the time course of adaptations
404
between the two types of exercise [3]. But this remains speculative. Furthermore, unilateral
405
exercises can vary substantially in terms of stability degree and contralateral limb contribution
406
[2, 3, 41]. Some unilateral exercises, such as unilateral leg extension, are performed on machines
407
with a higher degree of stability. While others are performed with free weights and a lower
408
degree of stability, such as rear elevated split squat with free barbell. But the potential influence
409

20
of these factors on muscle growth and strength adaptations are unknown and needs further
410
investigation. Moreover, future studies should consider having a group perform both types of
411
exercise and observe whether unilateral and bilateral strength gains are greater than selecting just
412
one or the other. Finally, based on the risk of bias assessment, all the studies presented some
413
concerns, particularly for not establishing whether allocation sequence was concealment,
414
whether evaluators were aware of the intervention received or not, and not presenting
415
information about prespecified analysis procedures. Thus, readers should be cautious with this
416
interpretation of current data. Future studies should focus on these points to reduce the risk of
417
bias.
418
419
5 CONCLUSIONS
420
Our findings suggest that exercise selection, whether bilateral or unilateral, appears to influence
421
dynamic strength adaptations. On muscle growth, there was no evidence of differential
422
hypertrophic adaptations, but the literature is almost non-existent. More specifically, bilateral
423
resistance training is more effective than unilateral training for increasing bilateral strength.
424
While unilateral resistance training is more effective than bilateral training for increasing
425
unilateral strength. Regarding muscle size changes, the magnitude of muscle growth does not
426
appear to differ between bilateral and unilateral training (but more data is needed). From a
427
practical point of view, coaches and practitioners should consider selecting bilateral exercises
428
when the objective is to optimize bilateral strength increases, while unilateral exercises can be
429
prioritized to increase unilateral strength. For muscle hypertrophy, both types of exercise should
430
be considered. Coaches and practitioners can take into account other factors to determine
431
exercise selection, such as time availability to train, personal preferences, individual needs, and
432

21
equipment availability. Considering the minimum amount of information available at this point,
433
future information could alter the current conclusion and practical recommendations.
434
435
Declarations
436
Funding No funds, grants, or other support was received for this study.
437
Conflict of Interest Jeremy Loenneke is an Editorial Board member of Sports Medicine but was
438
not involved in the selection of peer reviewers for the manuscript nor any of the subsequent editorial
439
decisions. Witalo Kassiano, João Pedro Nunes, Bruna Costa, Alex S. Ribeiro and Edilson S. Cyrino
440
declare that they have no conflicts of interest relevant to the content of this review.
441
Data Accessibility All data and code are available on the Open Science Framework project page
442
(https://osf.io/rkgfm/).
443
Author Contributions WK conceived the idea for this review and conducted the literature search.
444
WK and JPL worked together in the acquisition, analyzation, and interpretation of the data for the
445
study. WK and JPL interpreted the data, carefully reviewed the results, and edited the manuscript.
446
WK wrote the first draft of the manuscript. JPN, BC, ASR, JPL, and ESC revised the original
447
manuscript. All authors read and approved the final version.
448
449

22
REFERENCES
450
1. Kassiano W, Nunes JP, Costa B, Ribeiro AS, Schoenfeld BJ, Cyrino ES. Does varying
451
resistance exercises promote superior muscle hypertrophy and strength gains? a systematic review.
452
J Strength Cond Res. 2022;36(6):1753-62.
453
2. Mullican K, Nijem R. Are unilateral exercises more effective than bilateral exercises?
454
Strength Cond J. 2016;38(1):68-70.
455
3. Nijem RM, Galpin AJ. Unilateral versus bilateral exercise and the role of the bilateral force
456
deficit. Strength Cond J. 2014;36(5):113-8.
457
4. Rossman MJ, Venturelli M, McDaniel J, Amann M, Richardson RS. Muscle mass and
458
peripheral fatigue: a potential role for afferent feedback? Acta Physiol (Oxf). 2012
459
Dec;206(4):242-50.
460
5. Rossman MJ, Garten RS, Venturelli M, Amann M, Richardson RS. The role of active
461
muscle mass in determining the magnitude of peripheral fatigue during dynamic exercise. Am J
462
Physiol Regul Integr Comp Physiol. 2014 Jun 15;306(12):R934-40.
463
6. Abbiss CR, Karagounis LG, Laursen PB, Peiffer JJ, Martin DT, Hawley JA, et al. Single-
464
leg cycle training is superior to double-leg cycling in improving the oxidative potential and
465
metabolic profile of trained skeletal muscle. J Appl Physiol (1985). 2011 May;110(5):1248-55.
466
7. Zhang J, Iannetta D, Alzeeby M, MacInnis MJ, Aboodarda SJ. Exercising muscle mass
467
influences neuromuscular, cardiorespiratory, and perceptual responses during and following ramp-
468
incremental cycling to task failure. Am J Physiol Regul Integr Comp Physiol. 2021 Aug
469
1;321(2):R238-r49.
470

23
8. Matos-Santos L, Farinatti P, J PB, Massaferri R, Monteiro W. Cardiovascular responses to
471
resistance exercise performed with large and small muscle mass. Int J Sports Med. 2017
472
Nov;38(12):883-9.
473
9. De Oliveira JLS, Lima LCR, Barreto RV, Kassiano W, Neto JOV, Filho JCCN, et al.
474
Cardiovascular responses to unilateral, bilateral, upper, and lower limbs resistance exercises. Int J
475
Exerc Sci. 2023;16(2):1154-64.
476
10. Matkowski B, Place N, Martin A, Lepers R. Neuromuscular fatigue differs following
477
unilateral vs bilateral sustained submaximal contractions. Scand J Med Sci Sports. 2011
478
Apr;21(2):268-76.
479
11. Gandevia SC. Spinal and supraspinal factors in human muscle fatigue. Physiol Rev. 2001
480
Oct;81(4):1725-89.
481
12. Botton CE, Radaelli R, Wilhelm EN, Rech A, Brown LE, Pinto RS. Neuromuscular
482
adaptations to unilateral vs. bilateral strength training in women. J Strength Cond Res. 2016
483
Jul;30(7):1924-32.
484
13. Janzen CL, Chilibeck PD, Davison KS. The effect of unilateral and bilateral strength
485
training on the bilateral deficit and lean tissue mass in post-menopausal women. Eur J Appl
486
Physiol. 2006 Jun;97(3):253-60.
487
14. Zhang W, Chen X, Xu K, Xie H, Li D, Ding S, et al. Effect of unilateral training and
488
bilateral training on physical performance: A meta-analysis. Front Physiol. 2023;14:1128250.
489
15. Liao KF, Nassis GP, Bishop C, Yang W, Bian C, Li YM. Effects of unilateral vs. bilateral
490
resistance training interventions on measures of strength, jump, linear and change of direction
491
speed: a systematic review and meta-analysis. Biol Sport. 2022 Sep;39(3):485-97.
492

24
16. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The
493
PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021 Mar
494
29;372:n71.
495
17. Ardern CL, Büttner F, Andrade R, Weir A, Ashe MC, Holden S, et al. Implementing the
496
27 PRISMA 2020 Statement items for systematic reviews in the sport and exercise medicine,
497
musculoskeletal rehabilitation and sports science fields: the PERSiST (implementing Prisma in
498
Exercise, Rehabilitation, Sport medicine and SporTs science) guidance. Br J Sports Med. 2021
499
Oct 8;Epub ahead of print.
500
18. McGuinness LA, Higgins JPT. Risk-of-bias VISualization (robvis): An R package and
501
Shiny web app for visualizing risk-of-bias assessments. Res Synth Methods. 2021 Jan;12(1):55-
502
61.
503
19. Dankel SJ, Loenneke JP. Effect sizes for paired data should use the change score variability
504
rather than the pre-test variability. J Strength Cond Res. 2021 Jun 1;35(6):1773-8.
505
20. Caldwell A, Vigotsky AD. A case against default effect sizes in sport and exercise science.
506
PeerJ. 2020;8:e10314.
507
21. Borenstein M. Introduction to meta-analysis. Chichester: John Wiley & Sons; 2009.
508
22. Fisher Z, Tipton E, Zhipeng H. Robumeta: robust variance meta-regression. R package
509
version 2.1. https://CRANR-projectorg/package=robumeta. 2023.
510
23. Hedges LV, Tipton E, Johnson MC. Robust variance estimation in meta-regression with
511
dependent effect size estimates. Res Synth Methods. 2010 Jan;1(1):39-65.
512
24. Appleby BB, Cormack SJ, Newton RU. Unilateral and bilateral lower-body resistance
513
training does not transfer equally to sprint and change of direction performance. J Strength Cond
514
Res. 2020 Jan;34(1):54-64.
515

25
25. Krajewski KT, Bansbach HM, McLean L, McKenzie C, Rawcliffe A, Graham SM, et al.
516
Effects of short-term unilateral strength training on measures of postural control when wearing
517
“operationally relevant" backpack loads. J Strength Cond Res. 2020 Oct;34(10):2743-50.
518
26. Lee ELY, Abd Malek NF, Tan K, Pratama RS, Mohamad NI, Nadzalan AM. The effects
519
of unilateral versus bilateral resistance training on bilateral deficit, unilateral and bilateral strength
520
adaptation among trained men. J Phys Conf Ser. 2021;1793(1):012057.
521
27. Ramirez-Campillo R, Sanchez-Sanchez J, Gonzalo-Skok O, Rodríguez-Fernandez A,
522
Carretero M, Nakamura FY. Specific changes in young soccer player's fitness after traditional
523
bilateral vs. unilateral combined strength and plyometric training. Front Physiol. 2018;9:265.
524
28. Speirs DE, Bennett MA, Finn CV, Turner AP. Unilateral vs. bilateral squat training for
525
strength, sprints, and agility in academy rugby players. J Strength Cond Res. 2016 Feb;30(2):386-
526
92.
527
29. Stern D, Gonzalo-Skok O, Loturco I, Turner A, Bishop C. A comparison of bilateral vs.
528
unilateral-biased strength and power training interventions on measures of physical performance
529
in elite youth soccer players. J Strength Cond Res. 2020 Aug;34(8):2105-11.
530
30. Zhao X, Turner AP, Sproule J, Phillips SM. The effect of unilateral and bilateral leg press
531
training on lower body strength and power and athletic performance in adolescent rugby players.
532
J Hum Kinet. 2023 Mar;86:235-46.
533
31. Jakobi JM, Chilibeck PD. Bilateral and unilateral contractions: possible differences in
534
maximal voluntary force. Can J Appl Physiol. 2001 Feb;26(1):12-33.
535
32. Häkkinen K, Kallinen M, Linnamo V, Pastinen UM, Newton RU, Kraemer WJ.
536
Neuromuscular adaptations during bilateral versus unilateral strength training in middle-aged and
537
elderly men and women. Acta Physiol Scand. 1996 Sep;158(1):77-88.
538

26
33. Howard JD, Enoka RM. Maximum bilateral contractions are modified by neurally
539
mediated interlimb effects. J Appl Physiol (1985). 1991 Jan;70(1):306-16.
540
34. Škarabot J, Brownstein CG, Casolo A, Del Vecchio A, Ansdell P. The knowns and
541
unknowns of neural adaptations to resistance training. Eur J Appl Physiol. 2021 Mar;121(3):675-
542
85.
543
35. Ansdell P, Brownstein CG, Skarabot J, Angius L, Kidgell D, Frazer A, et al. Task-specific
544
strength increases after lower-limb compound resistance training occurred in the absence of
545
corticospinal changes in vastus lateralis. Exp Physiol. 2020 Jul;105(7):1132-50.
546
36. ACSM. American College of Sports Medicine position stand. Progression models in
547
resistance training for healthy adults. Med Sci Sports Exerc. 2009;41(3):687-708.
548
37. Lopez P, Radaelli R, Taaffe DR, Newton RU, Galvão DA, Trajano GS, et al. Resistance
549
training load effects on muscle hypertrophy and strength gain: systematic review and network
550
meta-analysis. Med Sci Sports Exerc. 2021 Jun 1;53(6):1206-16.
551
38. Schoenfeld BJ, Grgic J, Ogborn D, Krieger JW. Strength and hypertrophy adaptations
552
between low- vs. high-load resistance training: a systematic review and meta-analysis. J Strength
553
Cond Res. 2017 Dec;31(12):3508-23.
554
39. Vigotsky A, Halperin I, Trajano GS, Vieira TM. Longing for a longitudinal proxy: acutely
555
measured surface EMG amplitude is not a validated predictor of muscle hypertrophy. Sports Med.
556
2022 Feb;52(2):193-9.
557
40. Vigotsky AD, Beardsley C, Contreras B, Steele J, Ogborn D, Phillips SM. Greater
558
electromyographic responses do not imply greater motor unit recruitment and 'hypertrophic
559
potential' cannot be inferred. J Strength Cond Res. 2015 Dec 11.
560

27
41. Maroto-Izquierdo S, Nosaka K, Blazevich AJ, González-Gallego J, de Paz JA. Cross-
561
education effects of unilateral accentuated eccentric isoinertial resistance training on lean mass
562
and function. Scand J Med Sci Sports. 2022 Apr;32(4):672-84.
563
564

28
FIGURES CAPTIONS
565
Figure 1. PRISMA flow diagram.
566
Figure 2. Forest plot showing comparative effect of bilateral and unilateral training on muscle
567
hypertrophy. Values represent Cohen’s d (95% confidence interval). Each study is listed on the
568
left side of the plot with squares representing the effect size for each study and 95% confidence
569
interval. The square size vary in size according to the weights assigned to the different studies.
570
The overall effect is included at the bottom of the plot as a diamond with a width corresponding
571
to the confidence interval for the estimated effect. MT = muscle thickness, LTM = lean tissue
572
mass.
573
Figure 3. Forest plot showing comparative effect of bilateral and unilateral training on bilateral
574
strength. Values represent Cohen’s d (95% confidence interval). Each study is listed on the left
575
side of the plot with squares representing the effect size for each study and 95% confidence
576
interval. The square size vary in size according to the weights assigned to the different studies.
577
The overall effect is included at the bottom of the plot as a diamond with a width corresponding
578
to the confidence interval for the estimated effect. 1RM = one-repetition maximum, SQ = squat,
579
KE = knee extension, LP = leg press, KF. = knee flexion, LPD = lat pull down, CP = chest
580
press, BC = biceps curl, 5RM = 5-repetition maximum.
581
Figure 4. Forest plot showing comparative effect of bilateral and unilateral training on unilateral
582
strength. Values represent Cohen’s d (95% confidence interval). Each study is listed on the left
583
side of the plot with squares representing the effect size for each study and 95% confidence
584
interval. The square size vary in size according to the weights assigned to the different studies.
585
The overall effect is included at the bottom of the plot as a diamond with a width corresponding
586
to the confidence interval for the estimated effect. 1RM = one-repetition maximum, KE = knee
587

29
extension, LP = leg press, R = right side, L = left side, LPD = lat pull down, CP = chest press,
588
BC = biceps curl, RESS = rear elevated split squat, 5RM = 5-repetition maximum.
589
Figure 5. Weighted summary risk of bias plots.
590
Figure 6. Traffic light risk of bias plots.
591

30
592
593
Figure 1. PRISMA flow diagram.
594

31
595
Figure 2. Forest plot showing comparative effect of bilateral and unilateral training on muscle
596
hypertrophy. Values represent Cohen’s d (95% confidence interval). Each study is listed on the
597
left side of the plot with squares representing the effect size for each study and 95% confidence
598
interval. The square size vary in size according to the weights assigned to the different studies.
599
The overall effect is included at the bottom of the plot as a diamond with a width corresponding
600
to the confidence interval for the estimated effect. MT = muscle thickness, LTM = lean tissue
601
mass.
602

32
603
Figure 3. Forest plot showing comparative effect of bilateral and unilateral training on bilateral
604
strength. Values represent Cohen’s d (95% confidence interval). Each study is listed on the left
605
side of the plot with squares representing the effect size for each study and 95% confidence
606
interval. The square size vary in size according to the weights assigned to the different studies.
607
The overall effect is included at the bottom of the plot as a diamond with a width corresponding
608
to the confidence interval for the estimated effect. 1RM = one-repetition maximum, SQ = squat,
609

33
KE = knee extension, LP = leg press, KF. = knee flexion, LPD = lat pull down, CP = chest
610
press, BC = biceps curl, 5RM = 5-repetition maximum.
611

34
612
Figure 4. Forest plot showing comparative effect of bilateral and unilateral training on unilateral
613
strength. Values represent Cohen’s d (95% confidence interval). Each study is listed on the left
614
side of the plot with squares representing the effect size for each study and 95% confidence
615
interval. The square size vary in size according to the weights assigned to the different studies.
616
The overall effect is included at the bottom of the plot as a diamond with a width corresponding
617

35
to the confidence interval for the estimated effect. 1RM = one-repetition maximum, KE = knee
618
extension, LP = leg press, R = right side, L = left side, LPD = lat pull down, CP = chest press,
619
BC = biceps curl, RESS = rear elevated split squat, 5RM = 5-repetition maximum.
620

36
621
Figure 5. Weighted summary risk of bias.
622

37
Figure 6. Traffic light risk of bias plots.

38
Table 1. Characteristics of included studies.
Study
Participants
Training Program
Outcomes
Identity
Duration/
Frequency
Sets per
exercise per
session
Repetitions
Bilateral resistance
exercises
Unilateral resistance
exercises
Muscle hypertrophy and
strength measurements
Appleby et
al. [24]
Resistance
trained young
men (n = 23)
8 weeks,
2d/week
6−8 sets
4−8 repetitionsb
Bilateral squat
Step-up
1RM step-up and 1RM squat
Botton et al.
[12]
Non-
resistance
trained young
women (n =
29)
12 weeks,
2d/week
2 sets in
weeks 1−3, 3
sets in weeks
4−9, and 4
sets in weeks
10−12
12−15RM in
weeks 1−3,
9−12RM in
weeks 4−6,
7−10RM in
weeks 7−9, and
5−8RM in weeks
10−12
Bilateral knee
extension
Unilateral knee
extension
Overall quadriceps muscle
thickness (sum of rectus
femoris, vastus lateralis,
medialis, and intermedius) for
both limbs followed by the
sum of the right and left
thighs
Bilateral and unilateral 1RM-
strength assessed in knee
extension exercise
Janzen et al.
[13]a
Non-
resistance
trained post-
menopausal
women (n =
26)
26 weeks,
3d/week
2 sets
8−10RM
Lower limbs: Leg
press, knee
extension, and
hamstring curl
Upper limbs: lat pull-
down, biceps curl,
shoulder press, and
chest press
Lower limbs: Leg
press, knee
extension, and
hamstring curl
Upper limbs: lat pull-
down, biceps curl,
shoulder press, and
chest press
Lean tissue mass of upper and
lower limbs measured by dual
energy X-ray absorptiometry
Bilateral and unilateral 1RM-
strength assessed in leg press,
lat pull-down, and knee
extension
Krajewski
et al. [25]
Non-
resistance
trained men
(n = 15)
4 weeks,
3d/week
3 sets
2−6 repetitions
Bilateral back squat
and stiff legged
deadlift
Bulgarian split squat
and single leg stiff
legged deadlift
1RM back squat and 1RM
Bulgarian split squat
Lee et al.
[26]
Resistance
trained men
(n = 30)
6 weeks,
3d/week
3 sets
10 repetitions
Bilateral barbell chest
press and barbell
bicep curl
Unilateral dumbbell
chest press and
dumbbell biceps curl
Bilateral and unilateral 1RM-
strength assessed in chest
press and biceps curl

39
Ramirez-
Campillo et
al. [27]
Young male
soccer players
(n = 18)
8 weeks,
1d/weeks
3 sets
10 repetitions
Bilateral knee
extension and knee
flexion
Unilateral knee
extension and knee
flexion
Bilateral 1RM assessed in knee
extension and knee flexion
Speirs et al.
[28]
Resistance
trained men
(n = 18)
5 weeks,
2d/week
4 sets
3−6 repetitions
Bilateral back squat
Rear elevated split
squat
1RM bilateral back squat and
1RM rear elevated split squat
Stern et al.
[29]
Resistance
trained youth
male soccer
players (n =
23)
6 weeks,
2d/week
4 sets
6 repetitions
Bilateral back squat
Rear elevated split
squat
1RM bilateral back squat and
1RM rear elevated split squat
Zhao et al.
[30]
Resistance
trained youth
male rugby
players (n
=18)
5 weeks,
2d/week
4 sets
2−7 repetitions
Bilateral leg press
Unilateral leg press
Bilateral and unilateral 5RM in
leg press
Notes. a Upper limbs exercises performed in machines that had split lever arms, which allowed the participants to do unilateral training. b For unilateral exercise, the repetitions are
the total for the set (eg., 4 repetitions indicate 2 on each leg). 1RM = one-repetitions maximum, RM = repetition maximum.