V O L U ME 4 , ISS U E 7 JULY 2 0 2 0

MASS
M ONTHLY A PPL ICATIO N S IN
STRE N G TH SPO R T

E R I C H E LMS | G R E G N UCK O LS | MIC HAEL ZO URDO S | ERIC T REXL E R

 The Reviewers
Eric Helms
Eric Helms is a coach, athlete, author, and educator. He is a coach for drug-free strength and
physique competitors at all levels as a part of team 3D Muscle Journey. Eric regularly publishes peer-
reviewed articles in exercise science and nutrition journals on physique and strength sport, in addition
to writing for commercial fitness publications. He’s taught undergraduate- and graduate-level nutrition
and exercise science and speaks internationally at academic and commercial conferences. He has a
B.S. in fitness and wellness, an M.S. in exercise science, a second Master’s in sports nutrition, a Ph.D.
in strength and conditioning, and is a research fellow for the Sports Performance Research Institute
New Zealand at Auckland University of Technology. Eric earned pro status as a natural bodybuilder
with the PNBA in 2011 and competes in unequipped powerlifting, weightlifting, and strongman.

Greg Nuckols
Greg Nuckols has over a decade of experience under the bar and a B.S. in exercise and sports
science. Greg earned his M.A. in exercise and sport science from the University of North Carolina
at Chapel Hill. He’s held three all-time world records in powerlifting in the 220lb and 242lb classes.
He’s trained hundreds of athletes and regular folks, both online and in-person. He’s written for many
of the major magazines and websites in the fitness industry, including Men’s Health, Men’s Fitness,
Muscle & Fitness, Bodybuilding.com, T-Nation, and Schwarzenegger.com. Furthermore, he’s had the
opportunity to work with and learn from numerous record holders, champion athletes, and collegiate
and professional strength and conditioning coaches through his previous job as Chief Content
Director for Juggernaut Training Systems and current full-time work on StrongerByScience.com.

Michael C. Zourdos
Michael (Mike) C. Zourdos, Ph.D., CSCS, has specializations in strength and conditioning and skeletal
muscle physiology.  He earned his Ph.D. in exercise physiology from The Florida State University (FSU)
in 2012 under the guidance of Dr. Jeong-Su Kim. Prior to attending FSU, Mike received his B.S. in
exercise science from Marietta College and M.S. in applied health physiology from Salisbury University.
Mike served as the head powerlifting coach of FSU’s 2011 and 2012 state championship teams. He
also competes as a powerlifter in the USAPL, and among his best competition lifts is a 230kg (507lbs)
raw squat at a body weight of 76kg. Mike owns the company Training Revolution, LLC., where he has
coached more than 100 lifters, including a USAPL open division national champion.

Eric Trexler
Eric Trexler is a pro natural bodybuilder and a sports nutrition researcher. Eric has a PhD in Human
Movement Science from UNC Chapel Hill, and has published dozens of peer-reviewed research
papers on various exercise and nutrition strategies for getting bigger, stronger, and leaner. In
addition, Eric has several years of University-level teaching experience, and has been involved in
coaching since 2009. Eric is the Director of Education at Stronger By Science.
 Table of Contents

6
BY GR EG NUCKOL S

The Menstrual Cycle’s Effects on Performance Are

Complicated
The menstrual cycle doesn’t seem to meaningfully affect performance when
measured in the lab. However, female athletes often report that their performance
fluctuates throughout the menstrual cycle. These discrepant findings can be
primarily explained by psychological factors.

20
BY MI CHAEL C. ZOUR DOS

Much Needed Research on Deadlift Velocity

A plethora of research has examined various aspects of velocity-based training
in the squat; however, information on the deadlift is lagging behind. This article
reviews a recent study that established a normative velocity profile in the deadlift,
investigated the reliability of deadlift velocity, and investigated the accuracy of
predicting deadlift 1RM using velocity.

32
BY ER I C HEL MS

A Progression Framework for Hypertrophy

Hypertrophy: It’s a goal listed next to strength, power, and muscular endurance in
your textbook or professional manual, complete with ranges for each variable of
training. What assumptions does this framework require? This article takes a “first
principles” perspective on hypertrophy, using it to provide a model for progression.

44
BY ER I C T R EXL ER

How Low Can You Go With Your Caffeine Dose?

Caffeine is great, but a 6mg/kg pre-workout dose in the afternoon can present
some jitteriness and sleep-related challenges for many people. A new study sought
to determine if we can still enjoy caffeine’s strength benefits using a fairly low dose
(<2mg/kg). Read on to find out if this low-dose strategy is a viable option.

57
BY GR EG NUCKOL S

More Evidence Against Fiber Type-Specific Muscle Adaptations

Coaches and scientists have suggested for years that high-rep training could target
slow-twitch muscle fibers, while heavier training could target fast-twitch muscle
fibers. The results of a recent study question that claim. Heavier and lighter training
both produced similar growth in the soleus (a predominantly slow-twitch muscle)
and gastrocnemius (a muscle with a balanced fiber type ratio), whereas low-load
training should have been more effective for promoting soleus growth if low-load
training could target slow-twitch fibers.
67
BY MI CHAEL C. ZOUR DOS

Penalty: Reduction in Gains for Interference

A new study showed that strength gains can still be made even with high amounts of
aerobic training. This article breaks down the recent data and provides a clear guide
on how you can implement cardio in a way that will have a minimal effect on your size
and strength.

81
BY ER I C T R EXL ER

A New Meta-Analysis Says… Arginine Works Now?

Arginine has largely been written off as an ineffective supplement for exercise
performance. However, a new meta-analysis suggests that it improves both aerobic
and anaerobic performance. Did we abandon arginine too quickly? Read on to find out.

93
BY GR EG NUCKOL S

The Evidence for Myonuclear Muscle Memory May be

Murkier Than Commonly Believed
It’s commonly believed that myonuclei – the “control centers” of muscle fibers – are
added to muscle fibers when fibers grow, but aren’t lost by fibers when they shrink,
facilitating muscle re-growth. This is a proposed mechanism for the phenomenon of
“muscle memory.” However, a recent review suggests that the data is less conclusive
than people may realize.

109
BY MI CHAEL C. ZOUR DOS

VIDEO: Executing Iron Supplementation

Back in Volume 2, we covered the effectiveness of iron supplementation; however,
iron supplements come in various forms and with potential side effects that must be
discussed. This video picks up where we left off to discuss the execution of taking an
iron supplement.

111
BY ER I C HEL MS

VIDEO: Evidence-Based Practice Part II

Being “evidence based” is something that most MASS subscribers strive for as
coaches, content producers, or as a way to get the best results in the kitchen or
gym. In part 1 of this series, we established what evidence-based practice actually
is. In part 2, we will discuss the types of evidence and their use in decision making.
Letter From the Reviewers

W
e passed the half-way mark of 2020 in this July issue, which covers a
range of topics.
On the lifting front, Greg reviewed a study on calf training - spoiler alert
- calves grow! More importantly, due to their fiber composition, they’re ideal for
studying fiber-type specific training with high or low reps. Greg also broke down
a review on muscle memory, exposing more complexity and less certainty than
previously thought. Finally, Greg reviewed a meta-analysis on the effects of the
menstrual cycle on performance, which overall, will be a relief to our readers who
have or coach those with a menstrual cycle. Working in the training department
this month, Dr. Helms wrote a concept review on progression for hypertrophy. He
examined the lenses through which we view hypertrophy training, and created a
framework for when and why to progress specific variables. To round out the training
content, Dr. Zourdos reviewed one of the few studies on the velocity profile of the
deadlift, assessing its reliability and utility for load selection. Finally, Dr. Zourdos
led the next MASS chapter on concurrent training, examining a recent study and
creating a guide for minimizing the interference effect.
Switch hitting for the nutrition department, Dr. Zourdos’ video discussed how
to implement iron supplementation. Dr. Trexler assessed a meta-analysis on the
performance enhancing effects of arginine. We learn the devil is in the methods and
the results aren’t straightforward. Dr. Trexler also reviewed a study on the minimum
effective caffeine dose for enhancing strength and endurance. Finally, Dr. Helms
finished his video series on evidence based practice, delving into the strengths and
weaknesses of different types of evidence.
We hope you enjoy this issue of MASS, and always we appreciate your commitment
to learning.

5
 Studies Reviewed:
Menstrual Cycle Effects on Exercise-Induced Fatigability. Pereira et al. (2020).
Variations in Strength-Related Measures During the Menstrual Cycle in Eumenorrheic
Women: A Systematic review and Meta-Analysis. Blagrove et al. (2020)

The Menstrual Cycle’s Effects on

Performance Are Complicated
BY GREG NUCKOLS

The menstrual cycle doesn’t seem to meaningfully affect

performance when measured in the lab. However, female athletes
often report that their performance fluctuates throughout the
menstrual cycle. These discrepant findings can be primarily
explained by psychological factors.

6
 KEY POINTS
1. The menstrual cycle doesn’t seem to reliably impact fatigability or muscular
performance in women with regular periods.
2. Conversely, many female athletes report that their sport performance is hindered
by menstrual cycle-related symptoms.
3. It’s possible that this apparent contradiction can be resolved by psychological
explanations.
4. Hormonal fluctuations during the menstrual cycle vary considerably between
women. While strength and fatigability during different phases of the cycle may
not vary too much on average, it’s plausible that meaningful differences could
exist on an individual level.

T
hroughout the menstrual cycle, the were similar (2). It doesn’t seem that strength
hormones estradiol and progesterone fluctuates much throughout the menstrual cy-
follow predictable cyclical patterns. cle, on average. However, since many female
Both hormones are known to have effects in athletes report that menstrual symptoms neg-
both the nervous system and skeletal muscles atively affect their performance, there’s quite
(3, 4). In fact, the muscular effects of these hor- a bit of nuance to explore.
mones seem to affect how well female athletes
respond to resistance training – training taking Purpose and Hypotheses
place during the follicular phase (between the
onset of menses and ovulation) seems to pro- Purpose
duce more muscle growth and larger strength The purpose of the review focusing on fatiga-
gains than training taking place during the lu- bility (1) was to “summarize the effects of the
teal phase (between ovulation and the onset ovarian hormonal fluctuations on the exer-
of menses; 5, 6, 7). Thus, it’s fair to wonder cise-induced reduction in force during fatigu-
whether acute performance also fluctuates ing contractions.” The purpose of the review
throughout the menstrual cycle. focusing on strength (2) was to “systemati-
A pair of reviews this month investigated the cally review the current body of research that
effects of the menstrual cycle on acute fati- has investigated changes in strength-related
gability and strength-related measures. The variables during different phases of the men-
review focusing on fatigability (1) uncovered strual cycle in eumenorrheic women.”
studies with results that were all over the place. Hypotheses
On net, it doesn’t seem that menstrual cycle
Interestingly, the review focusing on fatigue
phase has a large effect on most measures of
did provide a hypothesis (which is uncom-
acute fatigability, though. Strength findings
mon for review articles), though the hypoth-

7
esis isn’t very specific: “Our hypothesis is 1. Subjects were women with normal menstru-
that menstrual cycle phase will influence the al cycles (21-35 days for at least 6 months).
exercise-induced decline in force, but the
2. The study assessed at least one strength-
effects with vary according to the task per-
or power-related outcome.
formed and limb used.” The review focusing
on strength-related variables did not have a 3. The outcome measure was assessed at at least
hypothesis, as one would expect. two defined points in the menstrual cycle.
4. Menstrual phase was identified or verified
Subjects and Methods by some sort of physiological measure, such
Since these studies were both reviews, they as hormone levels or body temperature.
had no subjects. Rather, they reviewed and 5. The study was published in a peer-reviewed
summarized previously published research. journal.
The review on fatigue-related measures (1) It also excluded studies that had these char-
included studies that “describe metrics of acteristics:
motor output” and studies that, “report exer-
cise-induced reductions in force during both 1. Subjects used hormonal contraception.
the luteal and follicular phases” in women not 2. Subjects were on hormone replacement
using hormonal contraceptives. After identi- therapy.
fying 920 possible studies in a systematic da-
tabase search, the authors narrowed the initial 3. The measures taken were separated by
list to 46 studies to be included in the review. longer than one regular menstrual cycle.
There was no meta-analytic element of this 4. Subjects were dealing with something
review, meaning that all relevant studies were like an injury or illness that would affect
summarized, but their results weren’t fit with performance.
a statistical model to compute a pooled effect
size. The review of fatigue-related measures 5. The study didn’t adequately describe mea-
never identifies itself as a systematic review, surement protocols or results.
but it appears that it is one. The inclusion cri- 6. The subjects used ergogenic aids.
teria were more lax than you generally see
in most systematic reviews, but that doesn’t After identifying 1035 potential papers in
necessarily disqualify it. The primary hall- their search, the authors narrowed it to 21
mark of a systematic review (a comprehen- studies that met all criteria. From there, the
sive and systematic database search) is pres- authors performed a series of meta-analyses
ent and outlined. comparing strength and power performance
between the early follicular (within 5 days
The review on strength-related measures (2) of the onset of menstruation) and ovulatory
was a systematic review and meta-analysis. To phases (within 2 days of ovulation), ovula-
be included, a study had to meet these criteria: tory and mid-luteal phases (21-25 days fol-

8
lowing the onset of menstruation), and early well-trained, highly trained, Collegiate, or
follicular and mid-luteal phases. A handful elite athletes. Subjects were 19-33 years old.
of studies assessed performance at other time
Table 1 shows the meta-analytic results. Note
points (e.g. mid-follicular or late luteal), but
that the “n” columns tell you the total number
not enough to warrant meta-analysis. Out-
of subjects for each comparison, not the total
comes were separated by the strength qual-
number of studies. As you can see, there are
ity assessed: maximal isometric contraction
no significant differences for any compari-
strength, isokinetic peak torque, and explo-
son, and the mean effect for all comparisons
sive strength (like jump height, peak cycling
is trivial (|g| < 0.08).
power output, or rate of force development).
The fatigue review split results out by cate-
Findings gory of performance test, including time to
task failure, relative reduction in force fol-
The studies included in the strength me- lowing fatiguing contractions, cycling endur-
ta-analysis included a total of 232 partici- ance time, running endurance time, and rat-
pants, with a wide range of training statuses, ings of perceived exertion (effort-based, not
ranging from two studies on sedentary sub- RIR-based). Somewhat confusingly, “maxi-
jects to five studies on subjects described as mal voluntary strength” was one of the cate-

9
gories of performance included in the fatigue found that isometric handgrip time to task
review. I’m not going to include the table of failure was significantly longer in the follic-
strength studies since it would be redundant. ular phase compared to the luteal phase. No
other significant or noteworthy effects were
Nine studies with a total of 119 subjects
reported in the other six studies.
investigated the effects of menstrual cycle
phase on fatigability. One (reviewed last Six studies with a total of 187 subjects inves-
year in MASS) found that isometric knee tigated the effects of menstrual cycle phase
extension time to task failure was signifi- on strength reductions following fatiguing
cantly longer in the mid-luteal phase com- muscular contractions. One reported greater
pared to the early follicular phase, and two fatigue resistance during the follicular phase,

10
and one reported greater fatigue resistance certainly notable, but the effect size metric was
during the luteal phase after gripping exercis- also considerably inflated because the stan-
es. No other significant or noteworthy effects dard deviations were incredibly small, since g
were reported in the other four studies. is computed by dividing the mean difference
by the pooled standard deviation (small stan-
For cycling endurance performance, there was
dard deviations shrink the denominator, and
one finding in favor of the luteal phase and
thus inflate the effect size).
one finding in favor of the follicular phase out
of 10 studies. For running performance, there For perceived exertion, one study reported
was one finding in favor of the luteal phase greater perceived exertion during the mid-lu-
over the early follicular phase, and one find- teal phase than the early follicular phase, while
ing in favor of the late follicular phase over three studies reported greater RPEs during
the early follicular phase out of eight studies. the follicular phase (with two specifying that
The study that found improved running per- it was during the early follicular phase). One
formance during the luteal phase reported an study also reported higher leg RPEs (how hard
enormous Hedge’s g effect size (6.77), but cycling felt for the leg muscles specifically)
when I dug deeper, the actual effect wasn’t as during the ovulatory phase than the mid-fol-
large as the effect size would make it appear licular or mid-luteal phases. The other eight
(8). The subjects in the study completed a time reported no significant differences.
to exhaustion test at a fixed running speed, and
time to exhaustion was 37.8 ± 0.9 minutes in Interpretation
the luteal phase, compared to 31.3 ± 1.0 min-
utes in the early follicular phase. A difference Before getting into the results, it’s worth
of 6.5 minutes on a time to exhaustion test is discussing physiological reasons why one

11
might expect performance to vary through- It promotes glycogen sparing, and seems to
out the menstrual cycle. The menstrual cy- have a net excitatory effect in the central ner-
cle is characterized by cyclical fluctuations vous system. Progesterone, on the other hand,
in estrogen and progesterone. Both estrogen is a net negative for exercise performance. It
and progesterone levels are low at the onset has a net inhibitory effect in the central ner-
of menses, and estrogen levels gradually in- vous system, and it elevates the body’s thermo-
crease throughout the follicular phase, peak- regulatory set point, which could potentially
ing just before ovulation. Progesterone levels increase the risk of overheating during exer-
begin increasing around the time of ovula- cise (3, 4, 9, 10, 11). So, in theory, most types
tion, peaking during the middle of the lute- of exercise performance should be enhanced
al phase, and gradually decreasing until the during the mid-to-late follicular or ovulatory
onset of menses. Meanwhile, estrogen levels phases (when estrogen is high and progester-
decrease a bit from their pre-ovulation high, one is low), and possibly inhibited during the
but remain somewhat elevated throughout mid-luteal phase (when progesterone is at its
the luteal phase, decreasing near the end of highest) or during menses and throughout the
the luteal phase in anticipation of menses. early luteal phase (when estrogen is low).
Estrogen is a positive for exercise performance. Overall, however, that doesn’t seem to be

12
the case. There’s not consistent evidence if you’re exercising in a hot, humid environ-
suggesting that any facet of strength or pow- ment with little airflow, the thermoregulato-
er performance fluctuates much throughout ry effects of progesterone could accelerate
the menstrual cycle, and various measures heat-related fatigue, but otherwise, overheat-
of fatigability don’t seem to be strongly in- ing is unlikely to be the primary performance
fluenced by menstrual cycle phase either. So bottleneck. Under most normal circumstanc-
what gives? es, fluctuations of estrogen and progesterone
within normal physiological ranges probably
My read of the situation is that the mecha-
don’t have large effects on performance.
nisms by which estrogen and progesterone
can influence acute performance simply ar- However, these findings raise an interesting
en’t that potent in most circumstances. In question: Why do so many female athletes
other words, the sympathetic nervous system report that they perform worse during and
arousal you experience when exercising at around menses? For example, a new study
reasonably high intensities (either aerobical- on 15 female rugby athletes found that 14
ly or anaerobically) dwarfs any modest stim- reported menstrual cycle-related symp-
ulatory effect of estrogen or inhibitory effect toms, and 10 believed these symptoms im-
of progesterone. Except for elite athletes or paired their performance (12). This isn’t a
super long-duration exercise bouts, small new finding, however. For example, a 2016
shifts in carbohydrate versus fat utilization study on British female runners (13) found
probably have at most a very modest effect that approximately half of the runners felt
on acute aerobic performance (and virtually that their menstrual cycle impacted their
no impact on strength performance). Finally, training or performance. In fact, I’ve nev-

13
 games could cause distraction (12). However,
that effect may not necessarily show up in a
UNDER MOST NORMAL simple lab test of performance. It doesn’t take
too much strategizing or higher order thinking
CIRCUMSTANCES, to perform an isometric knee extension or to
perform a time to exhaustion test on a tread-
FLUCTUATIONS OF mill. Thus, the distractions caused by menstru-
ESTROGEN AND al symptoms might impact real-world sporting
performance without necessarily affecting
PROGESTERONE WITHIN lab-measured performance focused purely on
simple physiological capacities.
NORMAL PHYSIOLOGICAL
Second, athletes may conflate discomfort
RANGES PROBABLY DON’T with impaired performance. I don’t want it
to sound like I’m trying to gaslight female
HAVE LARGE EFFECTS athletes, because this is pretty universal. It’s
ON PERFORMANCE. not uncommon to believe your performance
is worse than it actually is when you don’t
feel your best. As a simple example, I used
to believe that I performed worse in the gym
er come across survey data from female
after a bad night of sleep. I certainly felt
athletes where some percentage of athletes
worse in the gym and after leaving the gym.
don’t report negative effects from menstru-
Then, I pulled a pretty big deadlift PR during
al symptoms. So, what’s going on here? We
a training session following a night where I
don’t reliably see performance decrements
only slept 3 or 4 hours, which caught me off
in the lab, but a non-negligible proportion
guard. After that, I started tracking my sleep
of female athletes perceive that their perfor-
in my training journal, and found out that
mance is worsened by menstrual symptoms.
there was virtually no relationship between
I think there are a few potential explanations. how much I slept and how I performed in
First, female athletes report that menstrual the gym (at least acutely – my performance
symptoms can be distracting. A key distinc- tanks after a few weeks of poor sleep). And
tion between sport performance and lab test- as it turns out, the research largely backs
ing is that most sports require considerable me up; exercise feels more challenging af-
amounts of strategy, planning, execution, and ter a night of shortened sleep, but resistance
focus. Any sort of physical discomfort could training performance and force output gen-
thus provide a distraction that negatively af- erally aren’t affected (14). So, there may be
fects performance; furthermore, in the recent a similar phenomenon at play here; female
study on rugby players, several subjects re- athletes who experience menstrual discom-
ported that fears of flooding during practice or fort may assume that they’re performing

14
worse because they don’t feel their best estrogen and progesterone levels are both very
when training or competing, when their ob- low, and just before ovulation, estrogen lev-
jective performance is actually unaffected. els are high while progesterone levels are still
low. That’s relevant because many of the stud-
Third, fluctuations in performance, especially
ies to date have simply compared performance
around menses, may not regularly show up in
during the follicular phase versus the luteal
the literature due to sampling bias. If you’re
phase, and only some have been more specific
a female college student who regularly expe-
about when testing was to occur. In the stud-
riences severe menstrual symptoms, and you
ies included in the fatigue review, there were
see a flier asking you to sign up for a study that
19 measures comparing fatigability or per-
will require you to perform maximal exercise
formance in the early follicular phase versus
during menses, you’d probably be disinclined
some other point in the menstrual cycle. Of
to sign up for the study. So, if we assume that
the 19 measures, 1 favored the early follicular
the severity of menstrual cycle symptoms is
phase, and 7 favored another phase (general-
associated with the detrimental effects on per-
ly luteal, and late follicular in one instance),
formance a female athlete experiences, it’s
with 11 null findings. In contrast, there were
possible that the people who experience the
36 measures that compared fatigability or per-
largest decrement in performance due to men-
formance during mid-follicular phase, late fol-
strual symptoms are also the people least like-
licular phase or just the “follicular phase” (not
ly to sign up for studies on the subject. If that’s
specifying early, mid-, or late) versus some
the case, the performance and fatigue litera-
other point in the menstrual cycle. Of the 36
ture would indicate that the decrements in per-
measures, 7 favored the mid-follicular phase,
formance during the early follicular phase are
late follicular phase or the “follicular phase,”
smaller than they really are.
and only 2 favored another phase, with 27 null
Finally, maybe there are fluctuations in perfor- findings. For both sets of measurements, most
mance throughout the menstrual cycle (specif- comparisons didn’t result in any significant
ically around menses) that the research hasn’t differences, but of the significant differences
been sensitive enough to pick up. While the that did exist, there’s a trend for performance
menstrual cycle can be split into two distinct to be worse during the early follicular phase
phases pretty easily (follicular and luteal), compared to other phases (1 for and 7 against),
for research purposes, it’s probably prudent and a trend for performance to be better during
to split it up even further. The phases matter the rest of the follicular phase compared to oth-
insofar as estrogen and progesterone levels er phases (7 for and 2 against). So, there might
are different in each phase. However, estro- be some slight fluctuations in performance,
gen and progesterone levels also change a lot with performance being slightly higher during
within each phase. Hormone levels early in the the mid- and late follicular phases, “normal”
follicular phase differ starkly from hormone during the luteal phase (and maybe ovulatory
levels late in the follicular phase, for example. phase, though there’s less data on the ovulato-
During menstruation (early follicular phase), ry phase), and slightly lower during the early

15
follicular phase. However, these fluctuations fluctuations throughout the menstrual cycle
may be too small or too variable for studies to haven’t yet been studied, so I can’t provide
reliably detect them. you with a range of what “normal” variability
looks like.
In fact, I think variability in the effects of the
menstrual cycle is an important factor to con- I just want to make two final notes before
sider. Menstrual symptoms vary considerably, wrapping things up. These reviews apply to
both in terms of how they manifest (cramps, female athletes with normal menstrual cycles,
fatigue, and headaches are all common, but not using hormonal contraceptives. Hormon-
not everyone who experiences cramps expe- al contraceptives are a huge can of worms
riences fatigue, and not everyone who experi- (monophasic vs. biphasic vs. triphasic, oral vs.
ences fatigue experiences headaches) and their injections vs. intrauterine, estrogen and pro-
severity (from being slightly annoying to near- gestin vs. progestin-only, older vs. newer gen-
ly debilitating). Furthermore, the magnitude of erations of progestins, etc.) that I couldn’t even
the fluctuations in estrogen and progesterone attempt to open in this article without doubling
throughout the menstrual cycle vary consider- or tripling the length. As for female athletes
ably between individuals. Specifically, female with irregular cycles or amenorrhea, I’d sus-
athletes who experience larger fluctuations in pect that performance wouldn’t fluctuate very
estrogen and progesterone (particularly es- much (however, most female athletes who are
trogen) may experience larger fluctuations in amenorrheic probably are experiencing re-
performance throughout the menstrual cycle – duced overall performance due to low energy
particularly a larger increase in endurance per- availability). Oligomenorrhea and amenorrhea
formance during the late follicular phase (15). are both generally characterized by low hor-
In other words, there’s a difference between mone levels, which almost inherently entails
saying “most measures of performance don’t smaller fluctuations of estrogen and progester-
seem to fluctuate throughout the menstrual cy- one. Both are conditions you should talk to a
cle on average” and saying, “you won’t expe- doctor and sports dietitian about.
rience fluctuations in performance throughout
the menstrual cycle.” You shouldn’t necessari- Next Steps
ly assume that a female athlete will experience
notable swings in performance each month, Most exercise science studies on the menstru-
but if you do notice that a female athlete expe- al cycle don’t actually measure estrogen and
riences changes in performance with her cycle progesterone levels at testing visits. By mea-
(or if you’ve noticed that you personally expe- suring hormone levels, we could see if larger
rience changes in performance with your cy- fluctuations in hormone levels truly are pre-
cle), that’s not something to be alarmed about, dictive of larger fluctuations in performance.
because it’s probably not incredibly uncom- In studies that assess performance during the
mon. Unfortunately, to the best of my knowl- early follicular phase, researchers should ask
edge, inter-individual differences in strength participants about the severity of their men-
strual symptoms to ascertain if decrements in

16
 APPLICATION AND TAKEAWAYS
Overall, one shouldn’t expect strength, power, strength endurance, or endurance
performance to fluctuate very much throughout the menstrual cycle. However,
performance may be slightly worse during the early follicular phase (menses),
which might be associated with the severity of menstrual symptoms. Finally, while
performance doesn’t vary much throughout the menstrual cycle on average, that
doesn’t preclude the possibility that some individuals may experience notable
fluctuations in performance.

performance during the early follicular phase

are larger or more consistent in female ath-
letes that experience more severe menstrual
symptoms. Finally, I’d be interested in see-
ing more studies that test performance at 3+
points in the menstrual cycle, since measur-
ing at just two points may be insufficient to
pick up on shorter and more subtle perfor-
mance fluctuations.

17
References

1. Pereira HM, Larson RD, Bemben DA. Menstrual cycle effects on exercise-induced
fatigability. Front Physiol. Published 2020 Apr 27. doi:10.3389/fphys.2020.00517
2. Blagrove RC, Bruinvels G, Pedlar CR. Variations in strength-related measures during the
menstrual cycle in eumenorrheic women: A systematic review and meta-analysis. J Sci
Med Sport. 2020;S1440-2440(19)30814-X. doi:10.1016/j.jsams.2020.04.022
3. Hansen M. Female hormones: do they influence muscle and tendon protein metabolism?
Proc Nutr Soc. 2018;77(1):32-41. doi:10.1017/S0029665117001951
4. Smith SS, Woolley CS. Cellular and molecular effects of steroid hormones on CNS
excitability. Cleve Clin J Med. 2004;71 Suppl 2:S4-S10. doi:10.3949/ccjm.71.suppl_2.s4
5. Reis E, Frick U, Schmidtbleicher D. Frequency variations of strength training sessions
triggered by the phases of the menstrual cycle. Int J Sports Med. 1995;16(8):545-550.
doi:10.1055/s-2007-973052
6. Sung E, Han A, Hinrichs T, Vorgerd M, Manchado C, Platen P. Effects of follicular versus
luteal phase-based strength training in young women. Springerplus. 2014;3:668. Published
2014 Nov 11. doi:10.1186/2193-1801-3-668
7. Wikström-Frisén L, Boraxbekk CJ, Henriksson-Larsén K. Effects on power, strength and
lean body mass of menstrual/oral contraceptive cycle based resistance training. J Sports
Med Phys Fitness. 2017;57(1-2):43-52. doi:10.23736/S0022-4707.16.05848-5
8. Bandyopadhyay A, Dalui R. Endurance capacity and cardiorespiratory responses in
sedentary females during different phases of menstrual cycle. Kathmandu Univ Med J
(KUMJ). 2012;10(40):25-29. doi:10.3126/kumj.v10i4.10990
9. Lundsgaard AM, Kiens B. Gender differences in skeletal muscle substrate metabolism -
molecular mechanisms and insulin sensitivity. Front Endocrinol (Lausanne). 2014;5:195.
Published 2014 Nov 13. doi:10.3389/fendo.2014.00195
10. Stephenson LA, Kolka MA. Thermoregulation in women. Exerc Sport Sci Rev.
1993;21:231-262.
11. Pivarnik JM, Marichal CJ, Spillman T, Morrow JR Jr. Menstrual cycle phase affects
temperature regulation during endurance exercise. J Appl Physiol (1985). 1992;72(2):543-
548. doi:10.1152/jappl.1992.72.2.543
12. Findlay RJ, Macrae EHR, Whyte IY, Easton C, Forrest Née Whyte LJ. How the menstrual
cycle and menstruation affect sporting performance: experiences and perceptions of
elite female rugby players. Br J Sports Med. 2020;bjsports-2019-101486. doi:10.1136/
bjsports-2019-101486

18
13. Bruinvels G, Burden R, Brown N, Richards T, Pedlar C. The Prevalence and Impact of
Heavy Menstrual Bleeding (Menorrhagia) in Elite and Non-Elite Athletes. PLoS One.
2016;11(2):e0149881. Published 2016 Feb 22. doi:10.1371/journal.pone.0149881
14. Fullagar HH, Skorski S, Duffield R, Hammes D, Coutts AJ, Meyer T. Sleep and athletic
performance: the effects of sleep loss on exercise performance, and physiological and
cognitive responses to exercise. Sports Med. 2015;45(2):161-186. doi:10.1007/s40279-
014-0260-0
15. Oosthuyse T, Bosch AN. The effect of the menstrual cycle on exercise metabolism:
implications for exercise performance in eumenorrhoeic women. Sports Med.
2010;40(3):207-227. doi:10.2165/11317090-000000000-00000

19
Study Reviewed: Analysis of the Load-Velocity Relationship in the Deadlift Exercise.
Benavides-Ubric et al. (2020)

Much Needed Research on

Deadlift Velocity
BY MICHAEL C. ZOURDOS

A plethora of research has examined various aspects of velocity-

based training in the squat; however, information on the
deadlift is lagging behind. This article reviews a recent study
that established a normative velocity profile in the deadlift,
investigated the reliability of deadlift velocity, and investigated
the accuracy of predicting deadlift 1RM using velocity.

20
 KEY POINTS
1. This study determined a normative load-velocity profile in the deadlift, examined
if deadlift velocity was reliable from session to session, and investigated the
accuracy of predicting deadlift 1RM from submaximal velocity.
2. A load-velocity profile was established and was different from previous
literature. In short, it seems that shorter and more well-trained lifters exhibit
slower deadlift velocities.
3. Velocity was deemed to be reliable from session to session, and it was
determined that deadlift 1RM could be predicted with reasonable accuracy from
submaximal velocity. These findings add to the thin body of literature on velocity
and deadlifting, and suggest that various aspects of velocity-based training can
be implemented on the deadlift effectively.

M
ASS has now covered some aspect 100% of 1RM in the deadlift. The purpos-
of velocity-based training about a es of the study were to create a normative
dozen times. This coverage has load-velocity profile on the deadlift, deter-
revealed the benefits of long-term load pre- mine if deadlift velocity was reliable between
scription with velocity (2 – MASS review), the two testing sessions, and to predict 1RM
the relationship between velocity loss and from submaximal velocity. In brief, a nor-
repetitions in reserve (RIR)-based rating of mative load-velocity profile was established
perceived exertion (RPE) (3 – MASS re- (specifics presented later), but this profile
view), the reliability of squat velocity from was considerably different than the load-ve-
session-to-session (4 – MASS review), the locity profile found in previous literature (5).
load-velocity relationships among various This study also found that average veloci-
exercises (5 – MASS review), and the ability ty did not change at the same percentage of
of submaximal velocity to predict one-repe- 1RM between testing sessions, which means
tition maximum (1RM) (6 – MASS review). velocity was reliable. Lastly, the researchers
Despite all this coverage in MASS, only one reported that a two-point linear regression
of these reviews has examined the deadlift, forecast over-predicted deadlift 1RM by an
which is critically under-researched in veloc- average of only 3.6kg, which was statistically
ity-based training. Fortunately, the presently similar (p>0.05) to the actual 1RM. This arti-
reviewed study (1) provides some velocity cle will discuss why these findings may have
data on the deadlift. This study had 50 trained been similar or different from previous liter-
men perform a deadlift 1RM test during two ature and will provide an up-to-date guide on
different testing sessions six weeks apart. In implementing velocity-based training on the
each session, the load-velocity profile was deadlift into your program.
established at every 5% increment from 40-

21
 in which subjects had height and body mass
Purpose and Research assessed, completed basic health history pa-
Questions perwork, and then deadlifted with light loads
to ensure appropriate technique. On the sec-
Purpose
ond visit, exactly one week later, subjects
The purposes of this study were: 1) to establish performed the first testing session. Then 42
a normative load-velocity profile in the deadlift, of the 50 subjects returned six weeks later for
2) to determine the reliability of an individual’s the second testing session. In each session,
deadlift velocity from session to session, and 3) only the conventional deadlift was used.
to examine if deadlift 1RM could be accurately
predicted from submaximal velocity. Testing Sessions
Both testing sessions were identical and in-
Hypotheses
volved working up to a 1RM deadlift. Sub-
The authors hypothesized that there would jects worked up to a deadlift 1RM by per-
be a linear relationship between load and forming reps with a load corresponding to
velocity, and that velocity would be reliable every 5% increment from 30-100% of 1RM.
between the two tests performed six weeks From 30-60%, three reps were performed at
apart. However, no hypothesis was provided each load. Two reps were performed at each
regarding 1RM prediction accuracy. load from 65-75%. One rep was performed at
loads between 80-100% of 1RM. When more
Subjects and Methods than one rep was performed, the highest aver-
age concentric velocity was used for analysis.
Subjects
Outcome Measures
50 men with at least two years of training ex-
perience and a deadlift 1RM of at least 1.5 The researchers used the data from the first
times body mass participated. The available testing session to see if using velocities
subject details are in Table 1. achieved at loads corresponding to 30 and
80% of 1RM could accurately predict 1RM,
Overall Design and to report a normative load-velocity pro-
This study was completed over three visits. file for the deadlift for percentages between
The first visit was a familiarization session 40-100% of 1RM. Using both testing ses-

22
sions, the researchers examined if the veloci- predicted 1RM pretty accurately, and the in-
ties at each percentage of 1RM were reliable dividual predictions tended to be a bit better.
(i.e. similar to each other). As seen in Table 2, the mean difference and
effect sizes between the group 1RM predic-
Additionally, the researchers established ve-
tion and actual 1RM are larger than that of
locity profiles using peak concentric veloc-
the individual predictions, and the r-squared
ity and mean propulsive velocity. Neither of
value is higher for the individual predictions.
these metrics are particularly useful for us, and
Further, the authors presented Bland-Altman
mean propulsive velocity is the same as aver-
plots (agreement plots to further interpret
age concentric velocity for all intents and pur-
the data). In the Bland-Altman plots (Figure
poses (7). In the findings and interpretation,
1), we can see that the limits of agreement,
we’ll focus on average concentric velocity.
which are represented by the top and bottom
horizontal lines, are smaller (i.e. the top and
Findings bottom lines are closer together). The indi-
Normative Load-Velocity Profile in the Deadlift vidual data points are more tightly conformed
to the zero-difference line (i.e. center hori-
As expected, a very strong relationship was
found between average concentric velocity and
percentage of 1RM in the deadlift (r=0.913).
The average concentric velocities and associ-
ated 95% confidence intervals at each intensi-
ty for the deadlift can be seen in Table 2. We’ll
elaborate later, but this normative profile is
considerably different from a previous deadlift
profile reported by Fahs et al (5), but I think
those differences are easily explained.
Reliability of Velocity
Between the two testing sessions, the aver-
age velocity across all percentages of 1RM
was deemed to be reliable. To determine this,
an intraclass correlation coefficient of 0.870
was observed, which is classified as “good”
reliability (vv). When 1RM velocity was ana-
lyzed by itself, it was found to only have “ac-
ceptable” reliability (intraclass correlation of
0.763) between sessions.
1RM Prediction Accuracy
Both group and individual linear regressions

23
zontal line) in the individual prediction plot while also revisiting some established prin-
compared to the group plot (more description ciples. For the established principles, the
in the figure’s caption). In other words, these findings show a strong inverse relationship
plots suggest better agreement for the indi- between velocity and percentage of 1RM in
vidual versus the group predictions. the deadlift (r = -0.91), which is almost iden-
tical (r = -0.92) to the relationship reported
Interpretation by Dr. Helms (8). This study is also the sec-
ond to present normative average velocity
The reviewed study provides novel insight, values across the spectrum of intensities for

24
 at a similar proximity to failure (~5-7RPE);
however, that is unlikely to be the case since
the load-velocity relationship has consider-
able inter-individual variation. So if using
normative profiles, which one from Table 4
should you use? First, we know that a higher
training age leads to slower 1RM velocities
in the squat (9) and bench press (10). The
aforementioned Fahs study (5), which was
not cited by the authors in the presently re-
viewed study, utilized men and women with
a training age of 5.5 years, a deadlift 1RM of
157.5 ± 52.3 kg, and relative strength of 1.87
times body mass. They observed that higher
relative strength was related to slower veloc-
the deadlift; however, the velocities present-
ities. Although the subjects in the currently
ed here are considerably different than Fahs
reviewed study had a relative strength of 1.79
et al (5). To illustrate this, the present veloc-
times body mass on the deadlift, the average
ity profiles are shown side by side in Table
1RM was 17kg lower, and all subjects were
4 with the data from Fahs for all similar per-
males. Thus, I think that the higher training
centages used.
status in Fahs et al probably had some con-
Upon inspection of Table 4, it is immediately tribution to the slower velocities at the higher
apparent that the values are different between percentages of 1RM. Additionally, Helms et
the two studies. Although velocity profiles al (8) reported an even lower 1RM deadlift
should be individualized, normative profiles velocity than Fahs (0.14 ± 0.05 m/s) in trained
are useful as a starting point or to program ve- powerlifters (12 men and 3 women) with an
locity-based training across a large group. In average 1RM deadlift of 2.6 times body mass,
fact, applying velocity across a group has led further indicating that a greater training status
to greater strength gains over six weeks when may be related to a slower 1RM deadlift ve-
compared to group-based percentage loading locity. Fahs also found that being taller was
(2 – MASS review). In a group setting, you associated with faster deadlift velocities, and
could advise athletes to perform singles on the subjects in the presently reviewed study
the deadlift with a load corresponding to 0.51 were an average of 6cm taller than Fahs’
m/s (I would use a range of 0.48-0.54 m/s), subjects. Thus, if you are well-trained and
which is supposedly ~85%, and do as many shorter, then I would steer toward using the
sets as possible (I would also cap the number velocities in Fahs’ study as a starting point.
of sets) until average velocity dropped below One other note is that of the 51 subjects in
the normative velocity range. In that exam- the Fahs study, 6 used the sumo deadlift with
ple, this would theoretically keep all athletes the other 45 performing conventional dead-

25
lift, and that study reported no difference in The presently reviewed study is the first to
relative strength or velocity between the two demonstrate the reliability of velocity at sub-
deadlift styles; thus, it doesn’t seem that in- maximal intensities in the deadlift. In other
clusion of the sumo deadlift in the Fahs study words, the velocity values at each percentage
accounts for the between-study differences. of 1RM remained similar between the two
testing sessions. Specifically, across all ve-
If using normative velocity profiles, we have
locities, the intraclass correlation coefficient,
previously recommended you cross-refer-
a measure of reliability, was 0.87 for average
ence these values with RPE (one, two) to cre-
velocity between the two testing sessions,
ate your own individual load-velocity profile.
and the coefficient of variation was 2.0%. An
If 0.51 m/s is supposed to be a ~5-7RPE after
intraclass correlation coefficient of this value
one rep, but someone continues to failure and
is classified as “good” reliability, and typi-
gets eight reps, then 0.51 m/s probably does
cally anything over 0.70 is deemed “accept-
not equal 85% of 1RM. For that individual,
able.” The velocities being the same between
the load-velocity profile should be adjusted
the two sessions six weeks apart is support-
accordingly. Likewise, if one rep at 0.51 m/s
ed by the fact that 1RM did not change from
is equivalent to an 8RPE for someone else,
testing session 1 (139.3 ± 16.4 kg) to session
then it’s likely that 0.51 m/s corresponds to
2 (140 ± 16.1 kg). It may seem funny that
~90% of 1RM. In brief, gathering RPE along
1RM didn’t change because we see six-week
with velocity can help in establishing indi-
training studies all the time where a 1RM
vidual load-velocity profiles and control for
goes up significantly; however, subjects here
proximity to failure.
were just training on their own, not on a study
program. While this is the first time that ve-
locity reliability has been shown in the dead-
lift, Banyard et al (4 - MASS review) did ob-
GATHERING RPE ALONG serve reliability previously in the squat. Over
three visits, Banyard found squat velocity to
WITH VELOCITY CAN be reliable between 20-90% of 1RM, with
intraclass correlation coefficient values of
HELP IN ESTABLISHING 0.81-0.96 (i.e. good to excellent). However,
at 100% of 1RM, Banyard found that squat
INDIVIDUAL LOAD- velocity was not reliable between sessions
(intraclass correlation = 0.55). The present-
VELOCITY PROFILES ly reviewed study from Benavides-Ubric (1)
AND CONTROL FOR found “acceptable” reliability (intraclass cor-
relation = 0.763) for 1RM velocity; howev-
PROXIMITY TO FAILURE. er, this finding is similar to Banyard in that
1RM velocity was not as reliable as velocity
at other percentages of 1RM. With this infor-

26
mation, we can be pretty sure that as long as
you have a similar degree of readiness from
session to session, you can use the same ve- YOUR AVERAGE VELOCITY
locity profile for submaximal percentages.
In other words, if a velocity of 0.52 m/s cor- CAN FLUCTUATE JUST A
responds to 70% of 1RM on the deadlift for
you in one session, then it probably corre-
LITTLE BIT AND STILL BE
sponds with a practically similar percentage PRACTICALLY THE SAME FROM
in the next session. If that 70% happens to be
150kg, and you later start performing 160kg SESSION TO SESSION.
at 0.52 m/s consistently, then 160kg is like-
ly your new 70%. Importantly, the Banyard
study (4) determined that the smallest worth- for these sessions. So, while 1RM velocity
while change in velocity was 0.06 m/s; thus, may not technically be reliable, it probably
you could give yourself a range of 0.47-0.57 doesn’t change that much once a lifter has
m/s for your 70% velocity in the above ex- achieved a high training age, assuming a true
ample. I don’t think you actually need to be 1RM is reached each time. In short, using
that specific with it, but the main point is that velocity-based training on the deadlift seems
your average velocity can fluctuate just a lit- feasible from session to session. I would
tle bit and still be practically the same from cross-reference the velocity with an honest
session to session. So, I would wait until you RPE/RIR to be sure the velocity has you at
see a consistent change in velocity at a spe- the appropriate proximity to failure.
cific load before you reach a determination
that clear progress has been made. I would The other piece of information to discuss is
also track RPE along with velocity to further 1RM prediction, and this study is only the
validate progress. second to use submaximal velocities to fore-
cast deadlift 1RM. The first study to do so,
In terms of 1RM velocity being unreliable, Ruf et al (12), showed that 3-, 4-, and 5-point
I think this is a bit misleading from a practi- submaximal velocity equations over-pre-
cal perspective. For example, Ruf et al (12) dicted 1RM by ~5-10 kg. Ruf had subjects
observed an unacceptable intraclass correla- test 1RM deadlifts during one session and
tion (0.63) for reliability of deadlift velocity then use the velocity from that 1RM to pre-
at a 1RM; however, the velocity at 1RM is dict 1RM in another session 48 hours later.
so low that it doesn’t take much variation for This means that if subjects recorded an aver-
the reliability value to be lower. For exam- age velocity of 0.15 m/s at 1RM in the first
ple, in the Banyard study (4), between three session, then the predicted load which cor-
squat 1RM days, the mean of the average ve- responded to 0.15 m/s in the second session
locity on all three days was 0.24 m/s, yet the was used as the prediction (example here).
intraclass correlation for velocity was 0.55 During the second session in the Ruf study,

27
subjects also performed a 1RM and the ve- singles. I realize that the Ruf predictions had
locity was found to be unreliable (intraclass a greater error than the presently reviewed
correlation = 0.63) compared to the first ses- study, but I don’t think that error is hugely
sion. So, I believe the findings from Ruf et relevant to get a gauge of 1RM changes over
al support our assertion above that the “un- time. If you are looking to base your pow-
reliability” in 1RM velocity is statistically erlifting competition attempts solely off of
true, but that unreliability may not be that a 1RM prediction that uses velocity from a
big of a deal practically. In other words, the previous session, or if you want to use this
5-10 kg inaccuracy of the 1RM predictions 1RM prediction to program percentages,
in Ruf’s study could be chalked up to some- then that may not be the best idea. Howev-
one’s 1RM velocity changing from 0.15 m/s er, if every month you do a 1RM forecast
to 0.20 m/s from session-to-session. But at using a previous 1RM velocity as the pre-
the same time, compared to velocity predic- diction point, and the 1RM gradually goes
tions of 1RM in other studies, being off by up, then the prediction is practically useful
5-10 kg isn’t too bad. For example, another from the perspective of gauging progress.
study from Banyard (13) reported a ~20kg If during that period, you are gradually see-
over-prediction of squat 1RM from 3-, 4-, ing velocity and reps performed improve at
and 5-point submaximal velocity forecasts. similar submaximal loads, then you can be
The presently reviewed study only used doubly sure the increase in predicted 1RM is
a 2-point equation (velocities achieved at representing progress to a decently accurate
loads close to 30 and 80% of 1RM) to pre- degree.
dict 1RM and reported an over-prediction of
Brief Summary
only 3.6kg (Table 3). However, these pre-
dictions were made retroactively on testing In summary, these data show that deadlift veloc-
day 1. This means that once a 1RM was de- ities are reliable from session to session, which
termined, the authors worked backward to is good news for implementing velocity-based
find the velocities associated with 30 and training. This study also provides normative val-
80% of 1RM and then predicted 1RM based ues for intermediate lifters on the deadlift to use
upon the velocity they just achieved (demon- as a starting point for velocity-based training,
strated here). This likely explains why pre- although I would recommend cross-referencing
dictions were more accurate in the present those values with RPE to quickly individualize
study than in the study by Ruf et al (12) or a load-velocity profile for the deadlift. Lastly,
Banyard et al (13). Using a previously es- this study shows that submaximal velocity can
tablished 1RM velocity as the prediction pretty accurately predict 1RM in the deadlift.
point such as in the Ruf study is potentially Taken together with previous data, if you use a
more practically useful than the retroactive previous deadlift 1RM velocity to predict your
calculation. If you know a previous 1RM max, then that prediction is probably practically
velocity, you can attempt to predict a future useful over time.
1RM by just doing a couple of submaximal

28
 APPLICATION AND TAKEAWAYS
1. This study provided a normative load-velocity profile in the deadlift. These average
velocities were much faster than shown in a previous study. Taken together, it
seems that more well-trained individuals and shorter individuals exhibit slower
velocities in the deadlift.
2. Based on the findings presented here, it seems that deadlift velocities are reliable
from session to session, which means that theoretically, velocity-based training can
indeed be implemented on the deadlift.
3. A linear regression forecast of deadlift 1RM from only two submaximal velocities
predicted 1RM with pretty good accuracy across a group level (over-prediction of
only 3.6kg on average). It does seem that velocity can be utilized to predict deadlift
1RM; however, in practice, the lifter should be aware that this prediction is unlikely
to be perfectly accurate.

Next Steps
We still need to fully uncover why velocity
profiles are individual. Fahs et al (5) found
significant relationships between submax-
imal deadlift velocity and both relative
strength and height, but a vast majority of
the variance in velocity is still unexplained.
It’s possible that limb length explains some
of the variance, but Fahs failed to observe a
relationship between femur length and veloc-
ity in the deadlift, and both Fahs et al (5) and
Cooke et al (14) failed to observe a relation-
ship between femur length and velocity in the
squat. To date, there are three (2, 15, 16) lon-
gitudinal studies on velocity-based training,
and none of them have examined the dead-
lift. Thus, a study looking at velocity-based
training in the deadlift versus a fixed loading
model (i.e. percentages or RM zones) would
suffice as an appropriate next step.

29
References

1. Benavides-Ubric A, Díez-Fernández DM, Rodríguez-Pérez MA, Ortega-Becerra M,

Pareja-Blanco F. Analysis of the Load-Velocity Relationship in Deadlift Exercise. Journal
of Sports Science and Medicine. 2020 Sep 1;19(3):452-9.
2. Dorrell HF, Smith MF, Gee TI. Comparison of velocity-based and traditional percentage-
based loading methods on maximal strength and power adaptations. The Journal of
Strength & Conditioning Research. 2020 Jan 1;34(1):46-53.
3. Rodríguez-Rosell D, Yáñez-García JM, Sánchez-Medina L, Mora-Custodio R, González-
Badillo JJ. Relationship Between Velocity Loss and Repetitions in Reserve in the Bench
Press and Back Squat Exercises. Journal of strength and conditioning research. 2019 Apr.
4. Banyard HG, Nosaka K, Vernon AD, Haff GG. The reliability of individualized load–
velocity profiles. International journal of sports physiology and performance. 2018 Jul
1;13(6):763-9.
5. Fahs CA, Blumkaitis JC, Rossow LM. Factors related to average concentric velocity of
four barbell exercises at various loads. The Journal of Strength & Conditioning Research.
2019 Mar 1;33(3):597-605.
6. Pérez-Castilla A, Suzovic D, Domanovic A, Fernandes JF, García-Ramos A. Validity of
different velocity-based methods and repetitions-to-failure equations for predicting the
one-repetition maximum during two upper-body pulling exercises. Journal of strength and
conditioning research. 2019.
7. García-Ramos A, Pestaña-Melero FL, Pérez-Castilla A, Rojas FJ, Haff GG. Mean velocity
vs. mean propulsive velocity vs. peak velocity: which variable determines bench press
relative load with higher reliability?. The Journal of Strength & Conditioning Research.
2018 May 1;32(5):1273-9.
8. Helms ER, Storey A, Cross MR, Brown SR, Lenetsky S, Ramsay H, Dillen C, Zourdos
MC. RPE and velocity relationships for the back squat, bench press, and deadlift in
powerlifters. The Journal of Strength & Conditioning Research. 2017 Feb 1;31(2):292-7.
9. Zourdos MC, Klemp A, Dolan C, Quiles JM, Schau KA, Jo E, Helms E, Esgro B, Duncan
S, Merino SG, Blanco R. Novel resistance training–specific rating of perceived exertion
scale measuring repetitions in reserve. The Journal of Strength & Conditioning Research.
2016 Jan 1;30(1):267-75.
10. Ormsbee MJ, Carzoli JP, Klemp A, Allman BR, Zourdos MC, Kim JS, Panton LB.
Efficacy of the repetitions in reserve-based rating of perceived exertion for the bench press
in experienced and novice benchers. The Journal of Strength & Conditioning Research.
2019 Feb 1;33(2):337-45.

30
11. Koo TK, Li MY. A guideline of selecting and reporting intraclass correlation coefficients
for reliability research. Journal of chiropractic medicine. 2016 Jun 1;15(2):155-63.
12. Ruf L, Chéry C, Taylor KL. Validity and reliability of the load-velocity relationship to
predict the one-repetition maximum in deadlift. The Journal of Strength & Conditioning
Research. 2018 Mar 1;32(3):681-9.
13. Banyard HG, Nosaka K, Haff GG. Reliability and validity of the load–velocity
relationship to predict the 1RM back squat. The Journal of Strength & Conditioning
Research. 2017 Jul 1;31(7):1897-904.
14. Cooke DM, Haischer MH, Carzoli JP, Bazyler CD, Johnson TK, Varieur R, Zoeller RF,
Whitehurst M, Zourdos MC. Body Mass and Femur Length Are Inversely Related to
Repetitions Performed in the Back Squat in Well-Trained Lifters. The Journal of Strength
& Conditioning Research. 2019 Mar 1;33(3):890-5.
15. Orange ST, Metcalfe JW, Robinson A, Applegarth MJ, Liefeith A. Effects of In-
Season Velocity-Versus Percentage-Based Training in Academy Rugby League Players.
International journal of sports physiology and performance. 2019 Oct 30;1(aop):1-8.
16. Shattock K, Tee JC. Autoregulation in Resistance Training: A Comparison of Subjective
Versus Objective Methods. Journal of Strength and Conditioning Research. 2020 Feb 13.

31
 Concept Review:
Why and When to Progress Training Volume for Hypertrophy

A Progression Framework for

Hypertrophy
BY ERIC HELMS

Hypertrophy: It’s a goal listed next to strength, power, and

muscular endurance in your textbook or professional manual,
complete with ranges for each variable of training. What
assumptions does this framework require? This article takes a
“first principles” perspective on hypertrophy, using it to provide a
model for progression.

32
 KEY POINTS
1. Whole muscle hypertrophy is the structural outcome of performing resistance
training which causes specific adaptations in components of skeletal muscle.
However, hypertrophy is not a specific performance adaptation itself.
Progression based on the recovery of training which causes hypertrophy will
improve your ability to recover, but not necessarily hypertrophy in all cases.
2. Set volume has a dose-response, diminishing returns relationship with
hypertrophy, following an inverted U. Eventually, as sets are added, hypertrophy
rates plateau, and as more sets are added, rates of growth slow, eventually
ceasing, then regressing. Hypertrophy increases strength, allowing increases in
load or reps, which maintains the effectiveness (rep range and RPE) of each set.
3. While an appropriate initial set-volume optimizes hypertrophy, it is not known
when and if sets should increase to maintain the fastest rate of muscle growth
possible. But, increasing sets too quickly can slow hypertrophy. Thus, it may be
better to increase sets when load or reps are no longer going up, if recovery is
also occurring.

I
n this concept review, I’ll approach hyper- In this article, I’ll discuss hypertrophy from
trophy training progression from a con- a “first principles” perspective. Meaning, I’ll
ceptual standpoint. I know, we’ve spent a start at a point before most articles or text-
lot of time discussing progression already in books on training theory start, and thus, be-
MASS, and Dr. Zourdos and I are proud to say fore some of the assumptions that are often
most of this work comes from the video de- made. It’s not that these assumptions are nec-
partment (seriously, step it up Greg and Trex). essarily wrong, but they can lead to practices
For example, Dr. Zourdos has a two-part series that might be suboptimal if we accept them
(Part 1, Part 2) on load progression strategies, uncritically. I’ll discuss the nature of hyper-
a video on how to progress assistance work, trophy, what we know about inducing a stim-
and I have a video on making progress using ulus to keep muscle growth going, and final-
muscle group specialization cycles. There’s ly, I’ll lay out a framework for progression
even more on progression in our content on that hopefully provides more pros than cons.
setting up training, autoregulation, and period-
ization, even when progression itself isn’t the
main focus. However, almost all of it is about What’s a concept review?
how to progress; we haven’t done much on A written concept review is similar to our
conceptualizing why and when to progress dif- signature video reviews. The aim of this
article type is to review a cornerstone topic
ferent variables (such as sets, reps, load, and in physiology or applied science research.
RPE) for the goal of hypertrophy.

33
 a collection of adaptations to different com-
Whole Muscle Hypertrophy ponents of skeletal muscle. The appearance
Physique competitors want big muscles, re- of increased muscle size in two individuals
gardless of what contributes to that size, to who train quite differently – or in the same
achieve a certain look. For example, when they individual following two distinctly different
pump up, that temporary increase in fullness training phases at different times – may ap-
counts just as much as anything else that con- pear outwardly similar at the macro level, but
tributes to muscle size. Strength athletes, on may be the result of differing proportional
the other hand, want bigger muscles because increases to these components at the micro
more contractile tissue means more force out- level. As outlined in a recent review by Haun
put. However, an increase in contractile tissue and colleagues (1), hypertrophy can occur in
is just one potential contributor to whole mus- three broad components (Figure 1):
cle hypertrophy. Further, the vast majority of 1. Myofibrillar hypertrophy, which is the in-
muscle growth research uses metrics of whole crease in the size and/or possibly the num-
muscle hypertrophy like muscle thickness, ber of myofibrils.
cross sectional area, or lean body mass, which
can’t distinguish which components of muscle 2. Connective tissue hypertrophy, which
grew in what proportion. includes increases in the volume of mus-
cles’ extracellular matrix with increases
As I alluded to, whole muscle hypertrophy in its mineral or protein content.
as a gross outcome is actually the result of
3. Sarcoplasmic hypertrophy, which is a
non-transient increase in the volume of the
sarcoplasm, which may include increases
in the volume of the mitochondria, sarco-
plasmic reticulum, t-tubules, and/or en-
zyme or substrate content.
Conceptually, unlike strength or muscular
endurance, whole muscle hypertrophy is not
a specific performance adaptation. Rather,
change in whole muscle size is the gross struc-
tural consequence of specific structures adapt-
ing in response to stimuli. A change in whole
muscle size can occur in response to a broad
range of training styles, varying widely from
low-load high-rep to high-load low-rep train-
ing (2), low-to-high total volumes (3), short-
to-long rest periods (4), training short of (5) or
to failure (6), and with various modalities from

34
free weights to machines (7). Thus, a change
in whole muscle size reflects the adaptive de-
mands that could be caused by, but are not lim-
ited to: 1) The need for more contractile tissue
to produce force; 2) The need for more con-
nective tissue to transmit force or to “house”
hypertrophy when it occurs in other parts of
skeletal muscle; and/or 3) The need for more
cellular organelles, enzymes, or substrates re-
lated to energy production for the completion
of more work. Therefore, it may be imperfect
to conceptualize “hypertrophy” as a training
type in the same way we think of strength,
power, or muscular endurance. Training par-
adigms for these performance outcomes are
designed to balance the stimulus, recovery,
and adaptation of training for a specific perfor- (11). Also in line with many other physio-
mance adaptation; yet, whole muscle hyper- logical adaptations, some work suggests this
trophy is a structural consequence of specific relationship follows an “inverted U” (12),
adaptations, not a specific adaptation itself (al- such that at a certain point, further increases
though again, it can absolutely contribute to in the number of sets reduces the rate of mus-
force production, and thus strength and power, cle growth, presumably as the recovery de-
which I will discuss more later in this article). mand of training saps the adaptive resources
While a wide array of variables can be manip- for hypertrophy (Figure 2). Indeed, in some
ulated to cause whole muscle hypertrophy, applied studies, groups doing more volume
applied research has narrowed the variables than a lower volume comparison group have
to some degree to get closer to “optimal,” es- actually not experienced measurable hyper-
tablishing that loads equal to or greater than trophy, or even experienced a small loss of
30-40% of 1RM (8), when trained at least rea- lean mass (13, 14).
sonably close to momentary muscle fatigue
for a sufficient number of sets that last suffi- Is Set Progression the
ciently long (~5-6 reps+) (9), should produce
robust hypertrophy, if there is a sufficient
Logical Conclusion?
supply of energy and protein for its creation This dose response relationship between vol-
(10). Perhaps unsurprisingly, there is also a ume and muscle growth has led to the creation
dose-response relationship between the total of systems of hypertrophy training which pri-
number of sets performed (when sets are at a oritize manipulating sets as a means of pro-
sufficient load and proximity to failure) and gressing the specific stimulus (set volume)
the magnitude of whole muscle hypertrophy most related to hypertrophy. Such a system

35
is seemingly built on the following premise: Secondly, you could argue based on the prin-
ciple of specificity, that the only thing we can
As your ability to recover from more sets in-
be strictly confident in regarding doing more
creases, the number of sets you can benefit
sets, is that doing more sets gets you better
from increases as well.
at doing more sets. To truly assess the prem-
At first glance, this is not an unreasonable ise of whether or not increases in set volume
premise given the dose-response relationship will result in more growth if you are recov-
between hypertrophy and set volume. Howev- ering from your current set volume, we have
er, there are actually a number of issues with to assess the meaning of recovery (and its
this idea. First, the dose-response relationship time course). Depending on who you talk to,
is based on comparisons between groups doing recovery is defined differently. Most often,
different total volumes, not different week-to- speaking colloquially, recovery is described
week progressions of volume across training cy- subjectively. As an aside, “subjective recov-
cles. To my knowledge, there aren’t any studies ery” has merit despite it not being objective.
that match for total volume, while comparing Believe it or not, validated questionnaires
a group that increases sets each week against which use quantitative scales to rate subjec-
a group that maintains a fixed number of sets. tive recovery outperform objective markers
Therefore, while we can say doing enough sets for monitoring and mirroring training load
produces more muscle growth than not doing (15). More pertinent to the present discus-
enough sets in a mesocycle, we can’t necessari- sion, in sports science, recovery is most sim-
ly conclude sets should be increased in a meso- ply defined as returning to one’s baseline
cycle. Simply put, while this premise could be performance (16). While using this objective,
true, we don’t have data to support it. sports science definition of recovery is more
black and white, using it in the context of hy-
pertrophy training comes back to the problem
of hypertrophy not being a performance ad-
aptation. The importance and utility of stimu-
lus, recovery, and adaptation are self-evident
THE ONLY THING WE CAN when applied to specific training for a spe-
BE STRICTLY CONFIDENT IN cific performance adaptation. If you are try-
ing to lift a very heavy load, or do many reps
REGARDING DOING MORE with a moderate load, you train for that task
by doing it, and you monitor progress and re-
SETS, IS THAT DOING MORE covery by your performance. You lift heavy
loads, or do many reps with moderate loads,
SETS GETS YOU BETTER respectively, with added complexity existing
AT DOING MORE SETS. only to manage recovery and accomplish the
task more efficiently. While both of these
tasks can result in hypertrophy, they aren’t

36
 sions. If you had a high workload capacity
and were adapted to a high exercise frequen-
cy, the repeated bout effect would enable
GETTING BETTER AT you to perform a decent amount of volume
at a reasonable level of effort on Wednesday.
RECOVERING FROM THE However, if you were already at what was
WORK WHICH STIMULATES optimal, this additional training wouldn’t
help you grow; just because you can do more
HYPERTROPHY IS NOT volume, doesn’t mean you always should.
Truly, for the premise that recovery indicates
NECESSARILY THE SAME a need for more stimulus, we’d have to define
THING AS BEING ABLE recovery as “the recovery of all physiological
processes related to hypertrophy.” Unfortu-
TO STIMULATE MORE nately, that isn’t something we can feasibly
measure at this point in time.
HYPERTROPHY BY
DOING MORE WORK. Volume Load Progression
The conversation about volume and hy-
pertrophy is not new, but it has changed.
perfectly synonymous with hypertrophy. If I For years, experts have said volume should
was to put this another way, getting better at progress at some stage if the goal is hyper-
recovering from the work which stimulates trophy. I’ve said this, however, at the time
hypertrophy is not necessarily the same thing most people meant volume load (sets x reps
as being able to stimulate more hypertrophy x load). To my knowledge, the 2015 Stron-
by doing more work. ger By Science article “The New Approach
to Training Volume” by Nathan Jones was
Conceptually, your “optimal volume” for the first to connect the dots and conclude that
hypertrophy is only clearly related to your perhaps instead of volume load (sets x reps x
recovery if said recovery is poor enough to load), we should be counting the number of
prevent you from performing that amount “hard sets.” He made a good case that hard
of volume. Likewise, there’s no reason you sets are a simpler, more accurate representa-
couldn’t be adapted to performing and recov- tion of how stimulative training volume is for
ering from very high volumes, far in excess hypertrophy. In the peer reviewed literature,
of what would be optimal for hypertrophy Baz-Valle and colleagues argued a similar
(think CrossFit). For example, imagine we position in a 2018 systematic review we cov-
had perfect knowledge of your optimal vol- ered in MASS, concluding “the total number
ume. Then, imagine you could complete it of sets to failure, or near to, seems to be an
all across Monday, Friday, and Saturday ses- adequate method to quantify training volume

37
when the repetition range lies between 6 and ume load from strength gains (doing more
20+ if all the other variables are kept con- reps or lifting more weight) are a confirma-
stant” (9). I agree with these conclusions, as tion that overload occurred. Therefore, vol-
they’re based on the fact that most studies re- ume load eventually has to increase if you
port greater hypertrophy in groups perform- are actually growing, if all else is equal.
ing more sets that meet this criteria. Indeed,
While the need to increase volume load is
in the second edition of my books I gave vol-
arguably self evident, we can’t say the same
ume recommendations as the number of sets
for the number of sets. There are no data I’m
for this reason.
aware of indicating that set volumes neces-
However, conceptualizing volume for hy- sarily increase with training age. As stated,
pertrophy as the number of sets has differ- there are plenty of studies comparing groups
ent implications for progression than when doing different set volumes, establishing the
conceptualizing it as volume load, which I aforementioned dose response relationship,
don’t think most people consider. Specifi- but there aren’t studies tracking individuals
cally, while I’m confident that volume load through multiple points over their lifting ca-
should steadily progress if your goal is hy- reers to establish optimal set volume. With
pertrophy, I’m not confident the same can that said, we do have evidence that quickly
be said about the number of sets. Why? As increasing sets on a weekly basis can coin-
discussed earlier, bigger muscles are general- cide with a slowed rate of hypertrophy. Spe-
ly stronger muscles. If you are growing, that cifically, Haun and colleagues increased sets
should eventually show up on the bar (or ca- by two, every week from 10 to 32 sets, but
ble stack). Even if you keep reps and sets the only observed significant increases in ex-
same, this means volume load will increase tracellular water-corrected lean body mass
(sets x reps x load). Therefore, if you are from pre- to mid-testing, but not from mid-
growing, it leads to strength increases, and to post-testing (18). Meaning, muscle growth
volume load must go up. slowed as sets continued to increase to a very
high number (technically, we can’t know the
This ties into how you view progressive over-
set increases caused this slow down, but they
load and the relationship between strength
didn’t prevent it).
and hypertrophy. Increasing your strength
doesn’t make you grow; rather, when strength This shouldn’t be surprising, as out of all the
increases on movements in the same rep range variables influencing volume load, adding sets
after the novice stage, it is at least in part due increases it the most. Going from 10 to 32 sets,
to the contribution of bigger muscles to force if reps and load remains the same, is 320% of
production (17). From this perspective, “pro- the original work completed! Such a rapid in-
gressive overload” isn’t something you must crease can overwhelm one’s ability to adapt
proactively do by progressing the overload. and recover. Consider how much more aggres-
Rather, your progress occurs because you sive adding sets is compared to adding reps
caused overload. Meaning, increases in vol- or load. Going from 3 sets of 10 with 100kg

38
(3000kg), to 3 sets of 10 with 105kg (3150kg)
or to 3 sets of 11 with 105kg (3300kg) is a
modest 5 or 10% increase, respectively, com-
pared to a 33% increase going to 4 sets of 10 INCREASES IN LOAD
with 100kg (4000kg). More importantly, since
“hard sets” of at least ~5-6 reps are most rep-
OR REPS MAINTAIN THE
resentative of the hypertrophy stimulus, I’d ar- RELATIVE STIMULUS OF
gue that load and repetition increases serve a
different function than increasing sets. As you EACH SET. ADDING SETS
get stronger, if you don’t increase load or do
more reps, your sets get further from failure.
ON THE OTHER HAND IS,
Thus, increases in load or reps maintain the
relative stimulus of each set. Adding sets on
WELL, ADDING SETS. IT
the other hand is, well, adding sets. It could be COULD BE VIEWED AS A
viewed as a true increase in volume for hyper-
trophy, rather than ensuring your current vol- TRUE INCREASE IN VOLUME
ume remains effective.
FOR HYPERTROPHY.
When Set Progression is
Warranted ly, doesn’t mean it’s always a good idea to
I’ve argued that set progression is an aggres- increase sets in the first place.
sive move, but that doesn’t mean you should Indeed, there is evidence that decreasing
never do it. Further, there is some data we your volume in some cases can increase your
can use to help us make decisions about how rate of hypertrophy. In a recent study assess-
to increase sets. Most studies assign a certain ing hypertrophy, Aube and colleagues com-
number of sets for “high” and “low” volume pared multiple groups performing different
groups without consideration of how much set-volumes and also reported the change in
volume the participants were doing before set-volume among the participants compared
the study. Greg reviewed a study that found to what they were doing habitually. Interest-
arbitrary assignment to 22 sets/week pro- ingly, among the individuals who grew the
duced less hypertrophy than increasing ha- most (the top tertile of responders), they in-
bitual volume by 20%, even though at the creased their number of sets by 6.6 ± 12.4 per
group level, volume between conditions was week (20). That standard deviation, which is
similar (19). This indicates if you are to in- almost twice the mean, indicates a few of the
crease sets, it should be a relative increase individuals who grew the most in this study
from where you were. However, just because likely reduced the number of sets they were
it’s better to make a relative increase in set previously doing. Strictly speaking, it’s pos-
volume from what you were doing previous-

39
sible that those who decreased their sets in the subjective and objective terms while reps and
fastest growing tertile were actually growing load are also plateaued (Figure 3).
even faster before they enrolled in the study.
Another conceptual use of set progression is
Perhaps they were such genetic specimens
specialization cycles. Dr. Zourdos discussed
they would end up in the top tertile of hyper-
the concept in relation to high volume train-
trophy no matter what, but hypothetically, the
ing here in his Interpretation, Next Steps, and
point remains given our understanding of the
Application and Takeaways. As established,
inverted U relationship between volume and
sets are an aggressive increase in volume load,
hypertrophy. Simply put, if you’re already
so it might make more sense to “free up” the
doing sets in excess of what is optimal for
resources to adapt and recover from high vol-
you, decreasing your number of sets could
make you grow faster, or start growing again
if you were plateaued.
Practically, you want to set a reasonable start-
ing volume which you think will be ideal for
hypertrophy, based on your current knowl-
edge. Then, adjust volume over time to en-
sure you stay as close to that goal as possible.
When first starting out, if you don’t have his-
torical data indicating that a certain set volume
seems to result in your best growth, it’s not
a bad idea to use the existing published data
and start at ~8-12 hard sets per muscle group
per week (11). With an appropriate initial vol-
ume, you should be able to increase load and/
or reps over time if other variables are proper-
ly in place. These variables include sufficient
proximity to failure, rest, your rep range and
load falling within specific (but broad) ranges,
a reasonable training organization, adequate
nutritional support, and sleep. From this point,
increases in reps and load indicate overload is
occurring and serve to maintain the efficacy
of your sets. Then, when load or reps are no
longer progressing, it’s worth considering if a
“true” volume increase to the number of sets
is warranted. Specifically, this is likely appro-
priate if you are also recovering well in both

40
umes by only using high volume training on didn’t start too high in the first place). Further,
one or two body parts at a time while sets are just because recovery isn’t directly related to
reduced on your other muscle groups. This hypertrophy, doesn’t mean it can’t be useful
gets around the problem of doing too much when assessing the cause of a plateau.
too quickly, and allows an advanced trainee
Therefore, the premises which I questioned
to increase the volume on a muscle-group-
should not be wholly rejected, but rather up-
by-muscle-group basis, without increasing
dated with added nuance. Many might have
total volume to counterproductive levels. For
simply concluded before reading this article:
a more in depth look at how to do this, see my
“Volume is the variable most related to hyper-
video specifically on this topic.
trophy and best quantified as hard sets. There-
fore, we should progress sets in our training
Final Thoughts cycles.” My hope is now you understand that
As we’ve discussed numerous times in MASS, progressing reps or load shouldn’t be com-
despite volume having a known and import- pared to increasing sets, as increasing reps or
ant relationship with hypertrophy, you can do load can serve a different purpose. Further, you
too much. You can also increase it too quick- physically won’t always be able to increase
ly. While I’ve called into question certain as- reps or load, but you can always add another
sumptions, it doesn’t mean some don’t con- set. Meaning, in addition to ensuring the effec-
tain elements of utility or truth. For example, tiveness of your existing set volume, increases
I mentioned we don’t have data to support the in reps or load serve diagnostically to help you
idea that set volume must increase with train- assess the effectiveness of your training and
ing age. However, absence of evidence is not your objective recovery. Therefore, increases
evidence of absence. In my anecdotal experi- in reps or load is not in contrast with increas-
ence, going from post-newbie levels of muscle es in sets. Rather, they can go hand in hand,
mass to growing noticeably again often does with plateaus in reps or load helping to inform
require a sizable increase in set volume (if you when sets should increase.

APPLICATION AND TAKEAWAYS

Steady improvements in reps or load are a practical gauge of whether growth is
likely occurring. While strength gains indicate hypertrophy, the acute recovery of
performance is not necessarily tied to hypertrophy. Work capacity and recovery can
improve without necessarily changing your current optimal volume for hypertrophy.
Of all variables, sets increase total work the most, and too much work over short
time frames can impede muscle growth. Thus, rep or load progression can indicate
overload occurred, and serves to maintain the relative effectiveness of each set. When
reps or load are not progressing, recovery can be assessed to determine if an increase
in sets is warranted. This framework may maximize the upsides and minimize the
downsides of progressing each of these variables for the goal of hypertrophy.

41
References

1. Haun CT, Vann CG, Roberts BM, Vigotsky AD, Schoenfeld BJ, Roberts MD. A critical
evaluation of the biological construct skeletal muscle hypertrophy: size matters but so does
the measurement. Frontiers in Physiology. 2019;10:247.
2. Schoenfeld BJ, Grgic J, Ogborn D, Krieger JW. Strength and Hypertrophy Adaptations
Between Low- vs. High-Load Resistance Training: A Systematic Review and Meta-
analysis. J Strength Cond Res. 2017;31(12):3508-3523.
3. Schoenfeld BJ, Contreras B, Krieger J, et al. Resistance Training Volume Enhances
Muscle Hypertrophy but Not Strength in Trained Men. Med Sci Sports Exerc.
2019;51(1):94-103.
4. Grgic J, Lazinica B, Mikulic P, Krieger JW, Schoenfeld BJ. The effects of short versus
long inter-set rest intervals in resistance training on measures of muscle hypertrophy: A
systematic review. Eur J Sport Sci. 2017;17(8):983-993.
5. Carroll KM, Bazyler CD, Bernards JR, et al. Skeletal Muscle Fiber Adaptations Following
Resistance Training Using Repetition Maximums or Relative Intensity. Sports (Basel).
2019;7(7):169. Published 2019 Jul 11.
6. Karsten B, Fu YL, Larumbe-Zabala E, Seijo M, Naclerio F. Impact of Two High-Volume
Set Configuration Workouts on Resistance Training Outcomes in Recreationally Trained
Men [published online ahead of print, 2019 Jul 29]. J Strength Cond Res.
7. Schwanbeck SR, Cornish SM, Barss T, Chilibeck PD. Effects of Training With Free
Weights Versus Machines on Muscle Mass, Strength, Free Testosterone, and Free Cortisol
Levels [published online ahead of print, 2020 Apr 30]. J Strength Cond Res.
8. Lasevicius T, Ugrinowitsch C, Schoenfeld BJ, et al. Effects of different intensities of
resistance training with equated volume load on muscle strength and hypertrophy. Eur J
Sport Sci. 2018;18(6):772-780.
9. Baz-Valle E, Fontes-Villalba M, Santos-Concejero J. Total Number of Sets as a Training
Volume Quantification Method for Muscle Hypertrophy: A Systematic Review [published
online ahead of print, 2018 Jul 30]. J Strength Cond Res.
10. Slater GJ, Dieter BP, Marsh DJ, Helms ER, Shaw G, Iraki J. Is an Energy Surplus
Required to Maximize Skeletal Muscle Hypertrophy Associated With Resistance Training.
Front Nutr. 2019;6:131. Published 2019 Aug 20.
11. Schoenfeld BJ, Ogborn D, Krieger JW. Dose-response relationship between weekly
resistance training volume and increases in muscle mass: A systematic review and meta-
analysis. J Sports Sci. 2017;35(11):1073-1082.

42
12. Heaselgrave SR, Blacker J, Smeuninx B, McKendry J, Breen L. Dose-Response
Relationship of Weekly Resistance-Training Volume and Frequency on Muscular
Adaptations in Trained Men. Int J Sports Physiol Perform. 2019;14(3):360-368.
13. Amirthalingam T, Mavros Y, Wilson GC, Clarke JL, Mitchell L, Hackett DA. Effects of
a Modified German Volume Training Program on Muscular Hypertrophy and Strength. J
Strength Cond Res. 2017;31(11):3109-3119.
14. Hackett DA, Amirthalingam T, Mitchell L, Mavros Y, Wilson GC, Halaki M. Effects
of a 12-Week Modified German Volume Training Program on Muscle Strength and
Hypertrophy-A Pilot Study. Sports (Basel). 2018;6(1):7. Published 2018 Jan 29.
15. Saw AE, Main LC, Gastin PB. Monitoring the athlete training response: subjective self-
reported measures trump commonly used objective measures: a systematic review. Br J
Sports Med. 2016;50(5):281-291.
16. Sands WA, Apostolopoulos N, Kavanaugh AA, Stone MH. Recovery-adaptation. Strength
& Conditioning Journal. 2016 Dec 1;38(6):10-26.
17. Erskine RM, Fletcher G, Folland JP. The contribution of muscle hypertrophy to strength
changes following resistance training. Eur J Appl Physiol. 2014;114(6):1239-1249.
18. Haun CT, Vann CG, Mobley CB, et al. Effects of Graded Whey Supplementation During
Extreme-Volume Resistance Training. Front Nutr. 2018;5:84. Published 2018 Sep 11.
19. Scarpelli MC, Nóbrega SR, Santanielo N, et al. Muscle Hypertrophy Response Is Affected
by Previous Resistance Training Volume in Trained Individuals [published online ahead of
print, 2020 Feb 27]. J Strength Cond Res.
20. Aube D, Wadhi T, Rauch J, et al. Progressive Resistance Training Volume: Effects on
Muscle Thickness, Mass, and Strength Adaptations in Resistance-Trained Individuals
[published online ahead of print, 2020 Feb 13]. J Strength Cond Res.

43
 Study Reviewed: A Low Caffeine Dose Improves Maximal Strength, but Not Relative
Muscular Endurance in Either Heavier-or Lighter-loads, or Perceptions of Effort or Discomfort
at Task Failure in Females. Waller et al. (2020)

How Low Can You Go With Your

Caffeine Dose?
BY ERIC TREXLER

Caffeine is great, but a 6mg/kg pre-workout dose in the afternoon

can present some jitteriness and sleep-related challenges for many
people. A new study sought to determine if we can still enjoy
caffeine’s strength benefits using a fairly low dose (<2mg/kg). Read
on to find out if this low-dose strategy is a viable option.

44
 KEY POINTS
1. The current study (1) sought to evaluate the effect of low-dose (100mg; <2mg/
kg body mass) caffeine supplementation on maximal strength, muscular
endurance, perceived effort, and perceived discomfort in physically active
female participants.
2. Caffeine significantly improved isometric maximum voluntary contraction (MVC)
performance (placebo = 82.6 Nm [95% confidence interval: 68.4 to 96.8 Nm];
caffeine = 92.8 Nm [78.5 to 107.0 Nm]), but not muscular endurance, perceived
effort, or perceived discomfort.
3. Growing evidence suggests that males and females enjoy similar ergogenic
effects from caffeine supplementation. Caffeine can improve performance at
doses much lower than the 5-6 mg/kg doses most frequently studied, but the
optimal dose may vary from person to person.

I
’m probably not alone when I say that unwanted effects like jitteriness, headaches,
caffeine is near and dear to my heart. or sleep disturbance. When it comes to the
Caffeine can be easily found in afford- sex disparity in the research, it’s pretty clear
able and delicious food and beverage sources, that we lack sufficient research with female
and the currently available research literature participants. In fact, a 2019 paper found that
suggests that caffeine enhances a variety of up to that point in time, female participants
performance outcomes, ranging from strength accounted for only 13.2% of the published lit-
and power to endurance (2). However, the erature assessing caffeine’s ergogenic effects
caffeine literature is far from complete, and (3). While Greg recently reviewed a study
there are a couple of inconvenient realities to indicating that males and females experience
address: studies showing performance bene- similar ergogenic effects in response to caf-
fits from caffeine tend to utilize pretty large feine, another recent study suggested that sex
doses, and there’s a general lack of caffeine differences are possible.
research conducted with females. When it
The presently reviewed study (1) sheds some
comes to doses used in the literature, Grgic
light on both of these caffeine-related con-
et al noted in a giant review of 21 caffeine
siderations by investigating the performance
meta-analyses (2) that the most common dos-
effects of low-dose caffeine (100mg, equiv-
age used was 6 mg/kg of body mass; for a
alent to <2mg/kg body mass) in a sample
100kg person, we’re talking about 600mg, or
of 19 female subjects. More specifically,
the equivalent of approximately six cups of
these participants were habitual caffeine us-
coffee. If we could get an ergogenic effect at
ers (>100mg/day but <300mg/day) between
lower doses, that’d be great in terms of conve-
the ages of 18 and 35, and were not current-
nience and minimizing the likelihood of any
ly involved in resistance training. Caffeine

45
significantly increased isometric maximum Hypotheses
voluntary contraction (MVC) performance The authors hypothesized that caffeine would
(placebo = 82.6 Nm [95% confidence inter- positively impact leg extension strength and
val: 68.4 to 96.8 Nm]; caffeine = 92.8 Nm muscular endurance at 30% of 1RM, and
[78.5 to 107.0 Nm]). However, caffeine did that this effect would be related to caffeine’s
not appear to significantly improve muscular “perceptual” effects (this presumably refers
endurance, which was measured as the dura- to both perceived effort and discomfort). The
tion of time that participants could maintain authors did not state a hypothesis regard-
an isometric leg extension at 30% and 70% of ing caffeine’s potential effects on tasks per-
MVC torque. Caffeine also failed to signifi- formed at 70% MVC.
cantly impact ratings of perceived effort and
perceived discomfort. The landscape of caf-
feine research always seems to be shifting to Subjects and Methods
some extent; read on to find out exactly what Subjects
these results add to the literature, and where
19 physically active female participants be-
our understanding of caffeine’s performance
tween the ages of 18 and 35 were recruited for
effects currently rests.
this study. Their baseline characteristics are
presented in Table 1. Notably, all participants
Purpose and Hypotheses were habitual caffeine drinkers (defined as av-
Purpose eraging over 100mg/day of caffeine based on
a four-day food diary), but none of the partic-
The purpose of the presently reviewed study ipants consumed over 300mg/day of caffeine.
was to evaluate the effects of low-dose caf- Participants were excluded if they were preg-
feine supplementation (100mg, which was nant, breast feeding, using steroids, diabetic,
<2mg/kg of body mass) on maximal strength, hypertensive, or currently engaged in structured
along with muscular endurance, perceived resistance training. The authors did not mention
effort, and discomfort using both heavy (70% whether or not hormonal contraception was
of maximum voluntary contraction [MVC]) permissible, so it seems safe to assume that this
and light (30% of MVC) loads in physically was not required for inclusion, but also was not
active women. a reason for exclusion from the study.

46
Methods discomfort were assessed on scales ranging
The current study compared a 100mg dose of from 0 (no effort/discomfort) to 10 (maximal
caffeine, which was equal to about 1.5mg/kg effort/discomfort).
of body mass (1.1mg/kg for the participant Participants completed one familiarization
with the highest body mass, and 1.85mg/ visit to get used to the leg extension task,
kg for the participant with the lowest body followed by four testing visits. For the four
mass). Vitamin D2 capsules were used as the visits, they completed each combination of
placebo; this is an atypical choice, but ulti- supplement (placebo or caffeine) and leg ex-
mately not a big deal, because vitamin D2 tension intensity (30% or 70%) in random
supplementation wouldn’t be expected to in- order, with at least five days of rest between
fluence the outcomes tested in this study. The all visits. Participants were instructed to
authors stated that they chose this placebo avoid strenuous exercise within 48 hours of
because it was similar in size and appearance each visit, and not to consume caffeine, al-
to the caffeine capsules used. cohol, dietary supplements, or pain relievers
Performance was assessed a few different on the morning of testing. Participants were
ways using an isokinetic dynamometer. For also asked to consistently replicate their food
the maximal voluntary contraction test, the intake prior to each visit, but didn’t neces-
dynamometer was set up so that the machine sarily have to show up to visits in a fasted
was locked in place with the right knee posi-
tioned at a 135-degree angle (45 degrees of
flexion); at this knee angle, participants tried
to perform a leg extension with maximal
force. Thirty seconds after the MVC, partici-
pants performed either heavy (30% MVC) or
light (70% MVC) isometric leg extensions to
failure. With a screen displaying their torque
level, they pressed against the pad until the
desired level of torque was generated, and
then they continued producing this amount
of torque until they could no longer do so.
It’s important to note that participants were
instructed to shoot for 30% or 70% of the
MVC from that visit; so, if a subject had a
big increase in MVC at a particular visit,
they would have to complete the muscular
endurance test at a higher absolute intensity
that day. Immediately after this muscular en-
durance test, perceived effort and perceived

47
state. All visits occurred between 8 and 11
am, and testing time was kept constant for
each participant (so, they could pick any time
between 8 am and 11 am for their first visit,
but all visits would have to occur at that same
time of day).
Finally, a quick note about the stats. In this
study, the authors used some traditional hy-
pothesis testing that you’re likely familiar
with, but they also used something called
“equivalence testing.” Typically, we use sta-
tistical tests to determine if, for example, caf-
feine values were significantly different from
placebo values. However, just because they
weren’t significantly different doesn’t mean
they’re necessarily equivalent, and doesn’t
mean we can say that there was no effect
whatsoever. Equivalence testing calculates
a 90% confidence interval for the difference
between treatments (in this case, caffeine and
placebo), and sets up boundaries that spec-
ify how large of a difference is considered
notable or “worthwhile.” If the entire 90%
confidence interval of the difference between
treatments falls within the boundaries of this
“too small to care about” zone, we can con-
clude that the conditions are equivalent, and
that no meaningful effect was observed.

Findings
Isometric maximum voluntary contraction
(MVC) performance was improved by caf-
feine (placebo = 82.6 Nm [95% confidence
interval: 68.4 to 96.8 Nm], caffeine = 92.8 Nm
[78.5 to 107.0 Nm]; Figure 1). Equivalence
testing confirmed that the caffeine and place-
bo conditions were not statistically equivalent.

48
As one would expect, subjects were able to = 0.769). Caffeine did not significantly impact
maintain a contraction at 30% of MVC (around perceived effort, whether looking at the main
191 seconds, on average) much longer than at effect (averaged across both intensities; p =
70% of MVC (around 64 seconds). Caffeine 0.189) or the supplement by intensity interac-
did not significantly improve muscular endur- tion effect (p = 0.305). Values and confidence
ance, assessed as time to task failure, at 30% intervals were as follows: placebo at 30% MVC
of MVC (placebo = 178.2 s [146.2 to 210.2 = 7.16 [6.42 to 7.89], caffeine at 30% MVC =
s], caffeine = 204.1 s [172.1 to 236.1 s]; Fig- 8.00 [7.26 to 8.74], placebo at 70% MVC =
ure 2). Similarly, caffeine did not significantly 7.42 [6.69 to 8.16], and caffeine at 70% MVC =
improve muscular endurance at 70% of MVC 7.53 [6.79 to 8.26]. Equivalence testing indicat-
(placebo = 62.1 s [30.0 to 94.1 s], caffeine = ed that caffeine and placebo were statistically
65.3 s [33.3 to 97.3 s]; Figure 3). While values equivalent for the 70% MVC condition, but not
for placebo and caffeine were not significant- the 30% MVC condition.
ly different, equivalence tests did not indicate
that caffeine and placebo were statistically Interpretation
equivalent either. So, caffeine’s effect was not
pronounced enough to confidently say that it The straightforward interpretation of these re-
improved performance, but the caffeine and sults is that low-dose caffeine can significant-
placebo values weren’t close enough togeth- ly improve maximal strength, but not muscu-
er to confidently say that caffeine definitely lar endurance, in women. That alone gives us
failed to improve performance. some very valuable information about the vi-
ability of low-dose caffeine supplementation,
Perceived effort did not differ between the and the effects of caffeine in the female pop-
30% and 70% muscular endurance tests (p = ulation, which is notably underrepresented in
0.086). Caffeine did not significantly impact the caffeine literature. However, these results
perceived effort, whether looking at the main are a little bit surprising; a giant review that
effect (averaged across both intensities; p = Grgic et al (2) recently published indicated
0.475) or the supplement by intensity inter- that the effects of caffeine on muscular en-
action effect (p = 0.759). Values and confi- durance tend to be as large, if not a little bit
dence intervals were as follows: placebo at larger, than effects on maximal strength. So,
30% MVC = 8.00 [7.39 to 8.61], caffeine at I think we should take a closer look at a few
30% MVC = 8.26 [7.65 to 8.87], placebo at important considerations before concluding
70% MVC = 8.53 [7.92 to 9.14], and caffeine that caffeine “didn’t work” for the muscular
at 70% MVC = 8.63 [8.02 to 8.24]. Equiva- endurance tests.
lence testing indicated that caffeine and pla-
cebo were statistically equivalent for both the As the equivalence tests indicated, caffeine
30% MVC and 70% MVC conditions. and placebo were not significantly equiva-
lent. So the authors themselves have given
Perceived discomfort did not differ between us a very straightforward reason to hesitate
the 30% and 70% muscular endurance tests (p before suggesting that caffeine “didn’t work”

49
for muscular endurance in the current study, endurance at 30% MVC increased by about
but there’s a bit more to it than that. First, let’s 14.5%. In contrast, muscular endurance at
look at the statistical tests involved (these au- 70% MVC only increased by 5% in response
thors did a tremendous job with their stats, by to caffeine; it’s interesting to note that most
the way). Subjects completed four visits: two of the subjects tended to have a modest in-
placebo visits, and two caffeine visits. They crease in time to task failure in the caffeine
did an MVC test at all four visits, followed condition compared to placebo, whereas
by a muscular endurance test that involved four subjects had pretty large reductions in
either a higher load (70% MVC) or lower response to caffeine intake (Figure 3). It’s
load (30% MVC). So, when the researchers worth noting this apparent variability be-
compared the effects of caffeine and place- tween subjects, but we shouldn’t get too car-
bo on MVC strength, each subject had two ried away with interpreting that. Every now
caffeine values and two placebo values, and and then you see researchers fall into the trap
the model only had one predictor variable in where they over-interpret these trends, cut
the mix (supplement). In contrast, when they out the “non-responders,” then conclude that
compared the effects of caffeine and place- the treatment works as long as you remove
bo on muscular endurance, each subject had all the subjects for which it didn’t work. We
only one value for each combination of sup- don’t want to get caught in that trap, so we’ll
plement (caffeine or placebo) and load (30% just note the variability and move on.
or 70% MVC), and the model included three
Another thing to consider is the way loads
predictor variables (supplement, load, and
were determined for the muscular endurance
the supplement by load interaction). I could
tests. When studies use a more practical ex-
be mistaken (linear mixed models are com-
ercise modality (such as bench press reps to
plicated, and statistical power for interaction
failure) to assess muscular endurance, they
effects can also be complicated), but I believe
typically use a fixed load. For example, they
the MVC comparisons had a higher degree
might throw 70% of the subject’s 1RM on
of statistical power than the muscular en-
the bar for the placebo condition, and use the
durance comparisons, which would make it
exact same weight when they come back to
more likely to observe statistically significant
complete the caffeine condition. In the cur-
effects for MVC performance than muscular
rent study, maximal isometric strength was
endurance (I should be clear that this would
measured at each individual visit with the
not be a limitation of the study, but rather an
MVC testing. When subjects moved on to
unavoidable mathematical reality).
their muscular endurance test, the intensity
This seems to be reflected by a superficial (30% or 70% of MVC) was based on that
glance at the patterns in the MVC results and day’s MVC, not their baseline MVC. There
the 30% load muscular endurance results: is no correct or incorrect way to approach the
maximal strength (MVC) increased by about question of whether or not to use a fixed load
12% in response to caffeine, and muscular or an updated load at each visit, but we have

50
to interpret the results accordingly. Let’s say quite as promising. For example, Forbes et al
you were a subject in this study, and caffeine reported that an energy drink (2 mg/kg caf-
increased your MVC value (based on the feine) enhanced bench press reps to fatigue
group-level results, it likely did). When you across three sets to failure, but did not improve
proceeded to your muscular endurance test, repeated sprint performance (5). A study by
you’re now using 30% or 70% of an MVC Del Coso and colleagues tested two different
value that was inflated due to the ergogen- treatments (energy drinks with 1 mg/kg and 3
ic effect of caffeine; this essentially makes mg/kg caffeine) in comparison to a placebo;
the endurance task more challenging than it while the 3 mg/kg treatment significantly en-
would’ve been had your MVC strength been hanced bench press and half-squat power, the
unaffected by caffeine. So, you could very 1 mg/kg treatment did not (6).
justifiably argue that the lack of significant
Of course, you could argue that the energy
improvements in muscular endurance are to
drink literature is confounded by the presence
some degree related to the fact that maximal
of other ingredients, such as carbohydrates,
strength increased during the caffeine trials.
taurine, and tyrosine, among others. Howev-
The authors noted this consideration in their
er, the literature looking at lower doses of caf-
discussion, so they did a really nice job con-
feine anhydrous generally reports similar find-
textualizing their findings.
ings. In 2014, Dr. Lawrence Spriet published
While caffeine doses used in the research an excellent open-access review paper on the
literature are typically 3-6mg/kg (and often efficacy of low-dose caffeine supplementation
skewed toward the top end of this range), this for exercise performance (7). For this review,
is far from the first study to show ergogenic ef- he defined “low dose” as ≤3 mg/kg. Based on
fects with lower doses. Perhaps the best place the literature using endurance exercise tests, it
to look for this research is the energy drink seems pretty clear that super-high doses (9 mg/
literature, as commercially available energy kg and above) are no more effective than doses
drinks tend to provide caffeine doses below 3 in the 5-6 mg/kg range. It’s quite intuitive that
mg/kg. This body of literature is summarized most studies tend to use doses around 5-6 mg/
quite effectively in an open-access review kg in their experiments; you’d hate to put all the
paper published in 2013, so MASS readers time, energy, and resources into a study, only to
are referred to this paper (4) if they’d like to find out you cut the caffeine dose just a little bit
see a study-by-study assessment of the liter- too short. However, lifters and athletes might
ature. In short, there are a number of studies wish to shoot for the lowest effective dose, rath-
showing that energy drinks providing around er than the dose that is least likely to result in
2 mg/kg of caffeine may acutely improve up- a failed experiment, especially given the sleep
per-body and lower-body lifting volume and issues and minor (but uncomfortable) side ef-
endurance performance when ingested 10- fects that can be experienced when the caffeine
60 minutes prior to exercise, although find- dose is just a bit too high. The evidence from
ings for sprint and agility performance aren’t endurance-focused studies also suggests that 3

51
mg/kg doses tend to promote ergogenic effects cern for people wishing to use caffeine as a
quite similar to those observed with doses up regular pre-exercise supplementation strate-
in the 5-9 mg/kg range. Once the dose gets be- gy, participants in the current study enjoyed a
low 3 mg/kg, things get a little bit less clear and performance-enhancing benefit from ingest-
consistent. Overall, there’s some pretty solid ing a caffeine dose that was fairly congru-
data supporting endurance performance bene- ent with their typical caffeine intake, which
fits from caffeine doses as low as 3 mg/kg. The would be a good sign for people hoping to
data still look pretty good, although a little bit utilize caffeine consistently before workouts.
less consistent, around 2 mg/kg, and the data
Authors of the presently reviewed study not-
generally appear to suggest that 1mg/kg prob-
ed that individual responses tended to be
ably isn’t enough.
quite variable (see Figures 1-3). It’s difficult
Unfortunately there isn’t a great deal of re- to say how much of this is just random vari-
search looking at doses below 3 mg/kg with ation and measurement error, and how much
resistance training outcomes. One study from might truly represent divergent responses to
Astorino et al reported that 5 mg/kg led to caffeine (or divergent characteristics related
larger performance improvements than a 2 to habituation or withdrawal), which the au-
mg/kg dose when ingested an hour before ex- thors rightfully acknowledged. It’s also pos-
ercise (8). Grgic et al found that 2 mg/kg of sible that some degree of variation could be
caffeine ingested an hour before exercise en- explained by differences in menstrual cycle
hanced lower-body strength, whereas 4 mg/ phase or the use of hormonal contraceptives,
kg and 6 mg/kg doses did not (9). In the same as caffeine metabolism can be influenced by
study, all doses appeared to be effective for estrogen levels. However, the authors also
enhancing lower-body strength endurance, proposed that genetic factors might contrib-
while only 4 mg/kg and 6 mg/kg doses en- ute, at least in part, to the variability in the
hanced upper-body strength (9). In light of observed responses. With that in mind, I think
these mixed outcomes, it seems that caffeine it’s important to discuss a couple of recent
doses as low as 2 mg/kg can be viewed as findings related to how our genes influence
a viable caffeine supplementation strategy our responses to caffeine.
for now (pending further confirmation from
The CYP1A2 gene encodes the liver enzyme
future research), and the presently reviewed
that is responsible for about 95% of caffeine
study lends more support for lower dosing
metabolism. If you inherited the AA geno-
strategies in the 1.5-2.0 mg/kg range. An-
type, you metabolize caffeine quickly; if you
other promising aspect of this study relates
inherited the AC or CC genotype, you me-
to the habitual caffeine intake of participants.
tabolize caffeine more slowly. Some of the
All subjects habitually consumed somewhere
first research looking at caffeine genotypes
between 100-300 mg/day of caffeine, and
suggested that slow metabolizers might have
the supplement provided a caffeine dose of
higher long-term cardiovascular health risks if
100mg. So, while habituation may be a con-
they are heavy caffeine users. As Grgic et al

52
summarized in their recent paper (2), the liter- of participants with the CC or CT genotype.
ature evaluating the effect of CYP1A2 geno- Taken together, these recent studies on caf-
type on performance responses to caffeine has feine genotypes is somewhat promising; while
been fairly mixed. Two studies reported that the matter is very, very far from settled, these
caffeine enhanced endurance performance in new studies cast some doubt on the assertion
subjects with AA genotypes, but not in sub- that individuals with the “wrong” CYP1A2
jects carrying one or more C alleles. One study or ADORA2A genotype are unable to enjoy
also found that caffeine only improved mus- strength and power improvements in response
cular endurance (repetitions to fatigue) for to caffeine supplementation. Ultimately, more
subjects with the AA genotype, whereas two research is needed to confirm these emerging
other studies found that CYP1A2 genotype findings, and to determine what factors truly
did not substantially alter caffeine’s effects on dictate the variability of individual responses
vertical jump performance or Wingate sprint to caffeine supplementation.
performance. The newest study on the topic
In conclusion, the current findings support the
by Grgic et al found that 3 mg/kg of caffeine
results from previous research indicating that
significantly improved strength, power, and
female participants enjoy ergogenic effects
sprint performance, without major differences
from caffeine that are similar to those previ-
between genotype groups (10).
ously observed in male samples. While most
The ADORA2A gene encodes one of the ad- caffeine studies use doses as high as 5-6mg/
enosine receptors that caffeine binds to; ul- kg, there’s some research indicating that low-
timately, binding to the adenosine receptor er caffeine doses might be enough to enhance
is responsible for most of the effects that we strength performance. The current findings also
associate with caffeine. You can inherit the agree with previous research showing that indi-
TT, CT, or CC genotype for the ADORA2A vidual responses to caffeine can be quite vari-
gene. One study (11) has evaluated the effect
of this genotype on ergogenic responses to
caffeine, with results indicating that 5 mg/kg
of caffeine improved endurance performance LIFTERS WHO WISH TO IDENTIFY
in subjects with the AA genotype, but not in THEIR LOWEST EFFECTIVE CAFFEINE
subjects carrying one or more C alleles (those
with CT or CC genotypes). As a result, it has DOSE MIGHT CONSIDER AIMING
been suggested that carriers of the C allele for FOR THE LOW END OF THE
the ADORA2A gene are “non-responders” in
terms of caffeine’s performance-enhancing ef-
ERGOGENIC RANGE FOR CAFFEINE
fects. In contrast, a new study by Grgic et al SUPPLEMENTATION (~1.5 TO
(12) found that 3 mg/kg of caffeine improved 2 MG/KG), THEN INCREASING
performance on a wide range of strength,
power, and sprint outcomes in a sample full THEIR DOSE FROM THERE.

53
able. With these considerations in mind, lifters
who wish to identify their lowest effective caf-
feine dose might consider aiming for the low
end of the ergogenic range for caffeine supple- INDIVIDUAL RESPONSES
mentation (~1.5 to 2 mg/kg), then increasing
their dose from there (if necessary) based on
TO CAFFEINE VARY, SO
how well they tolerate caffeine intake and how THE OPTIMAL DOSE
favorably their performance is affected. On the
other hand, lifters who don’t have any issues MIGHT NOT BE THE
tolerating higher doses might just opt for a dose
in the 3-6 mg/kg range, as this is the most wide- SAME FOR EVERYONE.
ly studied dosing range and tends to most reli-
ably enhance performance when ingested about
60 minutes prior to exercise.
lower doses of caffeine (or better yet, looking
at multiple doses within the same study) us-
Next Steps ing strength assessments with a high degree of
There are still many questions pertaining to caf- measurement precision, with the ultimate goal
feine that require further exploration, but I think of identifying the lowest caffeine dose that re-
the findings from this particular study relate to liably enhances strength. Finally, we need way
three of the big topics. While it’s nice to see more research evaluating how caffeine-relat-
more caffeine studies using female samples, we ed genes like CYP1A2 and ADORA2A affect
still need more research to build up the body strength and power outcomes, as we can’t con-
of female-specific caffeine literature if we want fidently assume that the effects of a specific
to confidently generalize all that we’ve learned genotype on cardiovascular risk or aerobic ex-
in the male-dominated studies to date. It would ercise responses to caffeine necessarily apply to
also be nice to see more studies looking at some strength and power outcomes as well.

APPLICATION AND TAKEAWAYS

These recommendations probably don’t need to be tailored in a sex-specific manner,
as males and females appear to have pretty similar ergogenic responses to caffeine
supplementation. If a lifter doesn’t have any issues tolerating moderately high caffeine
doses, research suggests that doses ranging from 3-6 mg/kg most reliably enhance
performance when ingested about 60 minutes prior to exercise. However, lifters who
don’t tolerate high doses of caffeine well might aim to identify their lowest effective
caffeine dose by starting around 1.5-2 mg/kg and increasing from there (as needed)
based on performance responses and perceived tolerability. Individual responses to
caffeine vary, so the optimal dose might not be the same for everyone.

54
References

1. Waller G, Dolby M, Steele J, Fisher JP. A low caffeine dose improves maximal strength,
but not relative muscular endurance in either heavier-or lighter-loads, or perceptions of
effort or discomfort at task failure in females. PeerJ. 2020 May 14;8:e9144.
2. Grgic J, Grgic I, Pickering C, Schoenfeld BJ, Bishop DJ, Pedisic Z. Wake up and smell
the coffee: caffeine supplementation and exercise performance-an umbrella review of 21
published meta-analyses. Br J Sports Med. 2019 Jun;54(11):681:688.
3. Salinero JJ, Lara B, Jiménez-Ormeño E, Romero-Moraleda B, Giráldez-Costas V,
Baltazar-Martins G, et al. More Research Is Necessary to Establish the Ergogenic Effect of
Caffeine in Female Athletes. Nutrients. 2019 Jul 15;11(7):1600.
4. Campbell B, Wilborn C, La Bounty P, Taylor L, Nelson MT, Greenwood M, et al.
International Society of Sports Nutrition position stand: energy drinks. J Int Soc Sports
Nutr. 2013 Jan 3;10(1):1.
5. Forbes SC, Candow DG, Little JP, Magnus C, Chilibeck PD. Effect of Red Bull energy
drink on repeated Wingate cycle performance and bench-press muscle endurance. Int J
Sport Nutr Exerc Metab. 2007 Oct;17(5):433–44.
6. Del Coso J, Salinero JJ, González-Millán C, Abián-Vicén J, Pérez-González B. Dose
response effects of a caffeine-containing energy drink on muscle performance: a repeated
measures design. J Int Soc Sports Nutr. 2012 May 8;9:21.
7. Spriet LL. Exercise and Sport Performance with Low Doses of Caffeine. Sports Med.
2014;44(Suppl 2):175–84.
8. Astorino TA, Terzi MN, Roberson DW, Burnett TR. Effect of two doses of caffeine
on muscular function during isokinetic exercise. Med Sci Sports Exerc. 2010
Dec;42(12):2205–10.
9. Grgic J, Sabol F, Venier S, Mikulic I, Bratkovic N, Schoenfeld BJ, et al. What Dose of
Caffeine to Use: Acute Effects of 3 Doses of Caffeine on Muscle Endurance and Strength.
Int J Sports Physiol Perform. 2019 Sep 9;1–8.
10. Grgic J, Pickering C, Bishop DJ, Schoenfeld BJ, Mikulic P, Pedisic Z. CYP1A2 genotype
and acute effects of caffeine on resistance exercise, jumping, and sprinting performance. J
Int Soc Sports Nutr. 2020 Apr 15;17(1):21.
11. Loy BD, O’Connor PJ, Lindheimer JB, Covert SF. Caffeine Is Ergogenic for Adenosine
A2A Receptor Gene (ADORA2A) T Allele Homozygotes: A Pilot Study. J Caffeine Res.
2015 Mar 31;5(2):73–81.

55
12. Grgic J, Pickering C, Bishop DJ, Del Coso J, Schoenfeld BJ, Tinsley GM, et al.
ADORA2A C Allele Carriers Exhibit Ergogenic Responses to Caffeine Supplementation.
Nutrients. 2020 Mar 11;12(3):741.

56
 Study Reviewed: Do the Anatomical and Physiological Properties of a Muscle Determine its
Adaptive Response to Different Loading Protocols? Schoenfeld et al. (2020)

More Evidence Against Fiber Type-

Specific Muscle Adaptations
BY GREG NUCKOLS

Coaches and scientists have suggested for years that high-rep

training could target slow-twitch muscle fibers, while heavier training
could target fast-twitch muscle fibers. The results of a recent study
question that claim. Heavier and lighter training both produced
similar growth in the soleus (a predominantly slow-twitch muscle) and
gastrocnemius (a muscle with a balanced fiber type ratio), whereas
low-load training should have been more effective for promoting
soleus growth if low-load training could target slow-twitch fibers.

57
 KEY POINTS
1. Using a within-subject design in untrained male subjects, heavy (~8RM) and
light (~25RM) calf training produced similar growth in the soleus and both
heads of the gastrocnemius.
2. Interestingly, the lateral gastroc grew more than the medial gastroc or the
soleus, regardless of loading condition.
3. According to the notion of load-dependent fiber type-specific hypertrophy
(slow-twitch fibers growing more with low-load training, and fast-twitch fibers
growing more with high-load training), low-load training should have produced
more soleus growth than high-load training, since the soleus is predominantly
composed of slow-twitch fibers. Thus, this study provides more evidence
against the notion of load-dependent fiber type-specific hypertrophy.

Y
our muscles are composed of two (4), while the gastrocnemius generally has
broad categories of muscle fibers: more balanced proportions of type I and type
Type I (slow-twitch) and type II II fibers. Thus, if one assumes the existence
(fast-twitch) fibers. Type I fibers are recruited of load-dependent fiber type-specific hyper-
before type II fibers, and they’re less powerful trophy, one would expect low-load training
than their type II counterparts, but they’re also to produce more soleus growth, but not more
less fatigable. It’s been proposed that low-load gastrocnemius growth. However, that’s not
training may promote more type I fiber growth, what the present study found (1). Low-load
and heavier training may promote more type II (20-30RM) and high-load (6-10RM) training
fiber growth (2). We’ve previously reviewed produced similar growth in the soleus and both
a systematic review on fiber type-specific hy- heads of the gastrocnemius. Interestingly, re-
pertrophy in MASS (3), and its findings were gardless of training intensity, the lateral head
very amenable to a variety of interpretations. If of the gastrocnemius grew more than either the
you wanted to make a case for load-dependent medial head and the soleus. Gains in isomet-
fiber type-specific hypertrophy, you could ric plantar flexion strength were also similar
make that case, and if you wanted to make a between conditions. Thus, this study provides
case against load-dependent fiber type-specif- further evidence against the idea of load-de-
ic hypertrophy, you could also make that case. pendent fiber type-specific hypertrophy.
I think it’s safe to say that there’s not yet a
consensus in the literature.
Purpose and Hypotheses
A recent study (1) helps add to this body of
literature by examining the effects of training Purpose
intensity on calf growth. The soleus is pre- The primary purpose of this study was to
dominantly (~80%) composed of type I fibers “evaluate longitudinal changes in muscle

58
strength and hypertrophy of the calf mus- The study started with a familiarization week
cles between light (20-30RM) and heavy (6- to get the subjects accustomed to the exercis-
10RM) resistance training routines consist- es (unilateral seated and standing calf raises).
ing of plantarflexion exercise.” A secondary Following familiarization, baseline strength
purpose was to assess and compare growth of testing (isometric plantar flexion strength,
the different calf muscles. assessed via a dynamometer) was performed,
and the researchers assessed the thickness
Hypotheses
of the gastrocnemius and soleus using ultra-
The authors hypothesized that strength gains sound. The muscle thickness measurements
would be greater in the high-load group, and were taken pretty high on the calf – about
that low-load training would promote more 25% of the distance between the back of the
soleus growth and high-load training would knee and the heel.
promote more gastrocnemius growth.
Following baseline testing, the subjects trained
their calves twice per week for eight weeks.
Subjects and Methods They performed four sets per exercise per ses-
Subjects sion, training to momentary concentric fail-
ure, with 90 seconds of rest between sets and
26 young, untrained male subjects completed
three minutes of rest between exercises. Loads
this study.
were adjusted as necessary to ensure that all
Experimental Design sets stayed in the desired rep max zones (6-
This study used a within-subject design, 10RM for one leg, and 20-30RM for the other
meaning that for each subject, one leg was leg). Reps were completed with a controlled
randomized to perform low-load (20-30RM) concentric contraction and a ~2 second ec-
training, and the other leg performed high- centric contraction to ensure that the subjects
load (6-10RM) training. weren’t “bouncing” the reps. The order of

59
the exercises and the leg trained first varied
workout-to-workout to ensure that one leg or Interpretation
muscle group didn’t systematically receive a While this study is framed as a way to assess
greater stimulus by always being trained first. the impact of fiber typing on muscle growth,
The orders of the training sessions can be it’s worth making clear that the researchers
seen in Table 1. Following the eight weeks of didn’t actually take biopsies to assess the fi-
training, isometric plantar flexion strength and ber type proportions of the subjects’ muscles,
muscle thicknesses were remeasured. nor did they take biopsies to assess growth in
each fiber type to see if fiber type specific hy-
Findings pertrophy occurred. However, I don’t think
that’s a major drawback. Pre- and post-train-
High-load and low-load training caused sim- ing biopsies would have certainly been nice,
ilar hypertrophy in all calf muscles, and also but it’s well-understood that the soleus is in-
caused similar increases in isometric plantar sanely type-I dominant. Thus, if fiber type
flexion strength. In general, the soleus (+1.3- specific hypertrophy occurred, with more
1.5mm) grew less than the lateral gastrocne- type-I growth in the low-load condition, it
mius (+2.1-2.3mm), with the medial gastroc- would have almost certainly resulted in more
nemius in the middle (+1.5-1.8mm). total soleus growth in the low-load condition.

60
With that being said, it would still be a small training at very low intensities (say, almost
stretch to count this as direct evidence against exclusively doing sets of 20-30+ reps), high-
load-dependent fiber type specific hypertro- load training could cause some preferential
phy; however, I think it’s about the strongest type II growth, but I just don’t think it’s a
indirect evidence one could hope for. phenomenon that has much of an impact for
most lifters in most circumstances.
If you’ll recall from an older MASS article,
the evidence regarding fiber type specific
hypertrophy is ambiguous at best, with the
most methodologically sound (and, quite
frankly, trustworthy) studies not finding I’M OPEN TO THE IDEA THAT
clear evidence for its existence. Since that
time, one other high-quality study did find
TRAINING INTENSITY COULD
preferential type I fiber growth in powerlift- CAUSE PREFERENTIAL
ers doing low-load training with blood flow
restriction (5). In general, I’d say I’m open GROWTH OF DIFFERENT
to the idea that training intensity could cause MUSCLE FIBER TYPES IN
preferential growth of different muscle fiber
types in certain circumstances, but I think the CERTAIN CIRCUMSTANCES,
evidence for the phenomenon is much weak-
er and more inconsistent than most people
BUT I THINK THE EVIDENCE
realize, and the presently reviewed study fur- FOR THE PHENOMENON
ther reinforces that idea. I think it’s entirely
possible that, for lifters who habitually do IS MUCH WEAKER AND
most of their training at very high intensities, MORE INCONSISTENT THAN
low-load training (~30-40% of 1RM) could
cause some preferential type I growth, and MOST PEOPLE REALIZE.
for lifters who habitually do most of their

61
One finding that might be surprising at first muscle group. I think most people just don’t
is that strength gains were similar in both train their calves with as much focus and ef-
conditions. After all, doesn’t heavier train- fort as they train other muscle groups. For
ing generally cause larger strength gains? calf training, I think using a controlled tem-
It does, with the caveat that those larger po (like they used in the presently reviewed
strength gains are primarily seen in dynam- study) is key. Your achilles tendon is made
ic strength tasks. Training intensity actually for elastic recoil, so if you bounce your reps,
doesn’t seem to have much of an effect on I wouldn’t be shocked if the tendon is doing
isometric strength gains (6). While heavier way more of the work than the muscles are.
training increases 1RM strength to a great-
It’s worth pointing out that the individual
er degree, the most recent meta-analysis on
results to each intensity range varied drasti-
the topic found that high-load and low-load
cally. Notice how erratic the lines represent-
training have similar effects on isometric
ing within-individual results are in Figure 2
strength. This suggests that high-load and
above. A fair amount of subjects experienced
low-load training cause similar increases in
drastically better results with either high-load
actual contractile protein content, and the
or low-load training. In general, within-indi-
difference in dynamic strength development
vidual results in studies using within-subject
is primarily due to larger skill improvements
designs look more orderly, indicating that
with heavier training. Since this study used
the individual matters quite a bit more than
a within-subject design, the crossover effect
the difference in stimulus. In other words,
may have contributed as well (7).
some people simply grow more than others,
One thing I’d like to point out about this
study is that the rate of calf growth observed
in this study is in line with the rates of mus-
cle growth commonly seen in studies on un-
trained lifters. An increase in muscle thickness
TRAINING INTENSITY
of 7-15% in untrained lifters in eight weeks is
pretty similar to the rate of growth seen in
FOR CALF GROWTH IS A
other muscle groups in studies on untrained
lifters (8). People joke that the calves are the
VARIABLE THAT MIGHT PAY
hardest muscle group to build, but sometimes PRETTY BIG DIVIDENDS
it doesn’t seem like they’re joking. And who
knows – if this study was conducted on sub- IF YOU TAKE THE TIME
jects with five years of training experience,
maybe their calves wouldn’t have grown at TO TROUBLESHOOT
all, but I can’t think of a good a priori reason
to expect that growing the calves is funda- AND PERSONALIZE IT.
mentally different from growing any other

62
 shot. If you’ve been doing super high-rep sets
(they’re what Arnold recommended, after all)
with little to show for it, give some heavier
calf raises a shot.
Finally, it’s worth noting that the soleus grew
the least, and the lateral gastroc grew the
most. While I’m somewhat skeptical of the
notion of fiber type specific hypertrophy, it
is pretty well-established that type II fibers
generally grow more than type I fibers fol-
lowing resistance training (10). So, while I’m
not shocked that high-load and low-load had
similar effects on soleus growth, I’m also not
surprised that the soleus generally grew less
than the gastrocnemius, which generally has
a more balanced ratio of type I and type II fi-
bers. Regarding the lateral gastroc tending to

and the training protocol used doesn’t make

an enormous difference. For example, Figure
3 shows the results of a previously reviewed FOR PEOPLE WHO’VE
study comparing constant versus varied pro-
gramming (9). The subjects who got pretty NEVER DONE CALF
good results with the “constant” program
also got pretty good results with the “var- TRAINING, YOUR LATERAL
ied” program, and subjects that experienced
minimal growth on the “constant” program GASTROC MAY BE EVEN
also didn’t grow very much with the “varied”
program. In the present study (1), however,
MORE UNTRAINED THAN
there were quite a few subjects that barely
grew with low-load training but experienced
YOUR MEDIAL GASTROC
very robust growth with high-load training,
or vice versa. This suggests that training in-
IN A RELATIVE SENSE, AND
tensity for calf growth is a variable that might THUS HAVE GREATER
pay pretty big dividends if you take the time
to troubleshoot and personalize it. If you’ve POTENTIAL FOR GROWTH.
been doing sets of ~10 reps and you’re not
happy with your results, give higher reps a

63
 APPLICATION AND TAKEAWAYS
Calf growth seems to occur similarly well with relatively high (6-10RM) and
relatively low (20-30RM) loads. Furthermore, the lateral gastrocnemius may have a
greater capacity for hypertrophy than the soleus does.

grow more than the medial gastroc, the lateral

gastroc displays less muscle activation during
day-to-day tasks like standing and walking
when compared to the medial gastroc (11,
12). So, for people who’ve never done calf
training, your lateral gastroc may be even
more untrained than your medial gastroc in a
relative sense, and thus have greater potential
for growth in order to “catch up.”

Next Steps
We need more studies on calf growth in gen-
eral, ideally in trained lifters. I’d also be in-
terested in seeing a study on the hypertrophy
effects of controlled versus bouncing calf
raises. I’d be interested in seeing volume
dose-response research on the calf muscles
as well; anecdotally, calves seem to tolerate
much higher training volumes than other ma-
jor muscle groups, but it would be interesting
to see if that observation could be validated by
research. Regarding the possibility of prefer-
ential type I growth with load-load training,
it would be interesting to see if the results of
the present study would have been different
if the low-load group was training with, say,
50-60RM loads instead of 20-30RM loads.

64
References

1. Schoenfeld BJ, Vigotsky AD, Grgic J, et al. Do the anatomical and physiological
properties of a muscle determine its adaptive response to different loading protocols?
Physiol Rep. 2020;8(9):e14427. doi:10.14814/phy2.14427
2. Ogborn D, Schoenfeld BJ. The Role of Fiber Types in Muscle Hypertrophy: Implications
for Loading Strategies. Str Cond Jour. April 2014 - Volume 36 - Issue 2 - p 20-25 doi:
10.1519/SSC.0000000000000030
3. Grgic J, Schoenfeld BJ. Are the Hypertrophic Adaptations to High and Low-Load
Resistance Training Muscle Fiber Type Specific? Front Physiol. 2018;9:402. Published
2018 Apr 18. doi:10.3389/fphys.2018.00402
4. Gollnick PD, Sjödin B, Karlsson J, Jansson E, Saltin B. Human soleus muscle: a
comparison of fiber composition and enzyme activities with other leg muscles. Pflugers
Arch. 1974;348(3):247-255. doi:10.1007/BF00587415
5. Bjørnsen T, Wernbom M, Kirketeig A, et al. Type 1 Muscle Fiber Hypertrophy after Blood
Flow-restricted Training in Powerlifters. Med Sci Sports Exerc. 2019;51(2):288-298.
doi:10.1249/MSS.0000000000001775
6. Schoenfeld BJ, Grgic J, Ogborn D, Krieger JW. Strength and Hypertrophy
Adaptations Between Low- vs. High-Load Resistance Training: A Systematic Review
and Meta-analysis. J Strength Cond Res. 2017;31(12):3508-3523. doi:10.1519/
JSC.0000000000002200
7. Cirer-Sastre R, Beltrán-Garrido JV, Corbi F. Contralateral Effects After Unilateral Strength
Training: A Meta-Analysis Comparing Training Loads. J Sports Sci Med. 2017;16(2):180-
186. Published 2017 Jun 1.
8. Wernbom M, Augustsson J, Thomeé R. The influence of frequency, intensity, volume and
mode of strength training on whole muscle cross-sectional area in humans. Sports Med.
2007;37(3):225-264. doi:10.2165/00007256-200737030-00004
9. Damas F, Angleri V, Phillips SM, et al. Myofibrillar protein synthesis and muscle
hypertrophy individualized responses to systematically changing resistance training
variables in trained young men. J Appl Physiol (1985). 2019;127(3):806-815. doi:10.1152/
japplphysiol.00350.2019
10. van Wessel T, de Haan A, van der Laarse WJ, Jaspers RT. The muscle fiber type-fiber size
paradox: hypertrophy or oxidative metabolism? Eur J Appl Physiol. 2010;110(4):665-694.
doi:10.1007/s00421-010-1545-0

65
11. Duysens J, van Wezel BM, Prokop T, Berger W. Medial gastrocnemius is more activated
than lateral gastrocnemius in sural nerve induced reflexes during human gait. Brain Res.
1996;727(1-2):230-232. doi:10.1016/0006-8993(96)00525-2
12. Héroux ME, Dakin CJ, Luu BL, Inglis JT, Blouin JS. Absence of lateral gastrocnemius
activity and differential motor unit behavior in soleus and medial gastrocnemius
during standing balance. J Appl Physiol (1985). 2014;116(2):140-148. doi:10.1152/
japplphysiol.00906.2013

66
Study Reviewed: Impact of Low Volume Concurrent Strength Training Distribution on
Muscular Adaptation. Kilen et al. (2020)

Penalty: Reduction in Gains

for Interference
BY MICHAEL C. ZOURDOS

A new study showed that strength gains can still be made even
with high amounts of aerobic training. This article breaks down
the recent data and provides a clear guide on how you can
implement cardio in a way that will have a minimal effect on your
size and strength.

67
 KEY POINTS
1. Danish military conscripts performed three different concurrent training protocols
over nine weeks.
2. One group, micro-training, performed 15-minute strength and running sessions
that were separated by two hours. A classic training group performed 30 minutes
of running immediately followed by 30 minutes of lifting twice per week. A
standard group performed two 60-minute sessions per week, both of which
combined running and lifting.
3. Overall, there were not particularly meaningful differences between groups, but
strength and hypertrophy still tended to occur. This study shows that minimal
resistance training and high amounts of running can still lead to gains. This
article provides all the pertinent details on how to implement cardio in a way that
minimizes the interference effect.

I
f performing concurrent training atten- bic sessions and four 15-minute strength ses-
uates your rate of gains (i.e. the inter- sions per week with each session separated by
ference effect), it’s unlikely that you’ll a minimum of two hours, and no more than
wind up in the penalty box for interference; three total training sessions were allowed on
however, it is true that the magnitude of the one day. A “classic training” group also per-
interference effect is generally overstated. In formed 15-minute workouts, but performed
reality, many studies that show the interfer- four of them (two aerobic followed by two
ence effect are designed to cause it. In other lifting) consecutively. A “standard” group
words, the amount and mode of aerobic ex- performed two 60-minute sessions per week,
ercise determines the magnitude of the inter- both of which had a mix of running and lift-
ference. For example, if you run for an hour a ing. For each group, the aerobic training mo-
day, 4-6 days per week, then the interference dality was always running, and the running
effect will be pretty pronounced. However, progressed from moderate intensity to inter-
if you cycle 3 days per week for 30 minutes vals. Additionally, all groups performed ~20
each time, the interference with hypertrophy hours of regular military training. Strength
and strength will be pretty minimal. The pres- improved more consistently in the micro and
ently reviewed study (1) had a sample size of classic groups than in the standard group, and
290 members of the Danish military; all 290 the micro group also increased type IIa fiber
subjects performed regular military training type proportion by 5%. No individual group
and then were split into three groups for two improved vastus lateralis (lateral quadriceps)
hours of additional training per week. One size, but there was an increase in type II fi-
concurrent training group, the “micro-train- ber area when all groups were combined. In
ing” group, performed four 15-minute aero- general, the magnitude of change from pre-

68
to post-study was pretty low; however, that Hypotheses
shouldn’t be too surprising when we consider The authors predicted that strength gains
the low total volume of lifting and the high would be greater with the 15-minute exercise
amount of additional military training. How- bouts compared to the hour-long combined
ever, this study does have a few gems that can training sessions.
further our knowledge of concurrent training.
This article will point out those gems and will
provide a clear and comprehensive view on Subjects and Methods
how to incorporate cardio in a way that will Subjects
minimize the interference effect.
Danish military members (men, n = 285;
women, n = 6) had pull-up, shot put, and long
Purpose and Hypotheses jump performance tested. Further, a subgroup
Purpose of 87 men consented to strength testing and
muscle biopsies. Descriptive details of the
The purpose of this study was to investigate entire sample of the subgroup are in Table 1.
whether splitting up concurrent training ses-
sions into eight 15-minute sessions (two total Study Design
hours per week) affected strength, muscle fi- 290 members of the Danish military were
ber type distribution, and satellite cell num- split into three groups. All engaged in con-
ber differently than two 60-minute sessions current training for nine weeks; however,
of combined aerobic and resistance training only 87 subjects had strength and hypertro-
(two total hours per week) over nine weeks. phy assessed. All groups trained for a total

69
of two hours per week as part of the study. as multi-joint exercises performed with short
In addition to the study-specific training, all rest. The available details of the 15-minute
subjects performed normal military training strength and running sessions used in the mi-
which was described as, “a high amount of cro-training and classic training groups can
lower extremity low-intensity aerobic activ- be seen in Table 2.
ity (e.g. standing or marching with 0–10 kg
Outcome Measures
loads) for up till 20 h weekly.” One group,
the micro-training group, performed eight All individuals had their standing long jump,
15-minute sessions per week (four lifting shot put performance, and number of body
and four running) with no more than three weight pull-ups performed tested; however,
sessions on the same day, and daily sessions we won’t focus on the long jump and shot
separated by at least two hours. The classic put results in our interpretation. Additional-
training group performed four consecutive ly, a subsample of 87 subjects consented to
15-minute sessions twice per week. The testing maximal voluntary contraction force
paper is not entirely clear, but it seems that of the knee extensors, maximum voluntary
the first two 15-minute sessions in the clas- contraction force of the elbow flexors, and
sic group were both running and the last two hand grip strength, and also to have a muscle
both lifting, which essentially amounts to 30 biopsy taken on their vastus lateralis (later-
minutes of running followed by 30 minutes al quadriceps). Assessments from the biop-
of lifting for one hour of concurrent train- sy were: fiber type distribution percentages,
ing twice per week. Therefore, the micro muscle fiber area, and the number of satellite
and classic groups each performed exactly cells per fiber. All tests were done before and
one hour of lifting and one hour of running after the nine weeks.
per week as part of the study. Lastly, a stan-
dard group performed two one-hour sessions
twice per week, both of which combined run-
ning and lifting in the same session; however,
training in the standard group was not strictly
controlled. The paper stated that ~40% of the
standard group’s one-hour session consisted
of resistance training and the other ~60% was
“running or muscular-endurance training.”
The paper also notes that the standard train-
ing varied each week to fit within the mili-
tary basic training program, while noting that
the micro-training and classic training groups
were strictly controlled. The only other de-
tail regarding the standard group is that the
muscular endurance training was described

70
 significant increase in type II fiber area (p<0.05,
Findings +12.81%), but no change in type I fiber area
Strength and Performance (p>0.05). There was no significant change in
satellite cells per fiber (p>0.05) within any indi-
Results for pull-up performance and maximal
vidual group. However, when all groups were
voluntary contraction strength can be seen in
combined, there was a significant increase in
Table 3. Number of pull-ups performed in-
the number of satellite cells per type II fiber
creased in both the micro-training and classi-
(p<0.01, +33.33%) and an increase when type I
cal training groups, and that increase in both
and II fibers were combined (p<0.05, +22.22%);
groups was greater (p<0.05) than the standard
however, there was no significant change in the
group, which did not improve pullup perfor-
number of satellite cells per type I fiber. The
mance. All groups significantly (p<0.05) in-
muscle growth and satellite cell findings can be
creased knee extension strength with no group
seen in Table 4AB.
differences, while no group experienced a
statistically significant increase in hand grip The micro-training group experienced a 5.1%
strength, and only the micro-training group increase in their type IIa fiber proportion.
improved elbow flexor strength. There were no other statistically significant
changes within an individual group; howev-
Fiber Area, Satellite Cell Number, and Fi-
er, that was likely due to the low sample size
ber Type Distribution
for this measure, which can be seen in Fig-
Due to the smaller sample size associated with ure 1. When all groups were combined, and
the biopsy measures, these assessments were the statistical power was greater, there was
analyzed within each group and with all groups a 3.7% decrease in type I fiber distribution,
combined together as one large cohort. The which was statistically significant (p<0.05).
standard group had a significant (p<0.05) in-
crease in type I fiber area (+26.56%); however,
no other group experienced an increase in mus-
Interpretation
cle size nor were there any group differences. This study had a fair number of findings that
When all groups were combined, there was a may have caused the previous section to be-

71
come convoluted. So, in this interpretation, the standard group, and a significant increase
let’s first recap those findings before inter- in type II fiber area for all groups combined.
preting them. Afterward I will aim to provide Satellite cells per type II fiber and all fibers
a clear picture of how to program to mini- combined increased within the entire cohort.
mize the interference effect.
In general, the magnitude of changes in this
study were pretty modest regardless of group,
Findings Recap and and the researchers do indeed allude to why
Analysis this was the case. Toward the end of the pub-
Strength increased for most measures in the lished paper, the researchers state that “low
micro and classic training groups, while the amounts” of resistance training during basic
only strength improvement in the standard military training should be used by soldiers,
group was for knee extension. Further, the and that “this initial strength” program should
classic group had a significantly greater im- be followed by a much more comprehensive
provement in number of pull-ups performed program, which should result in more “ro-
compared to the standard group. Therefore, bust” gains. From these words, we can gather
it seems that the standard group did not con- that the military conscripts in this study had
sistently improve strength as much as the relatively little resistance training experience
other two groups. For hypertrophy, there was (i.e. initial program), they were in the middle
a significant increase in type I fiber area in of basic training, and that the resistance train-

72
ing volume was quite low, which it was. In- er strength gains, because their program was
deed, the volume was six total sets per week different than the micro and classic train-
for only five reps per set. The subjects did ing groups. Specifically, the standard group
make some progress, but this is just not a ro- trained one hour twice per week doing train-
bust training program. Now, consider that not ing that was described as ~40% resistance
only was one hour of running performed per and ~60% endurance training per session,
week as part of the study, but all subjects per- while the other two groups spent 50% of their
formed an additional 20 hours of combined time lifting weights. Other factors potentially
aerobic exercise, standing, and/or marching affecting the low rate of improvement were
every week. This additional training easily that the classic group performed running and
could have attenuated the rate of strength and lifting sessions back to back with running
hypertrophy; however, to be fair, we cannot performed first, and the micro-training group
say that happened with 100% certainty, or that performed running and lifting sessions within
the interference effect manifested due to any two hours of each other. Previous literature
of the study protocols, because the study did has clearly demonstrated that volume per-
not include a resistance training only group. formance declines by ~25% on lower body
Additionally, it shouldn’t come as a surprise training when lifting is performed four hours
that the standard group tended to have low- after aerobic exercise, and volume perfor-

73
mance was still affected at eight hours fol- imal voluntary contraction force in the pres-
lowing aerobic training (3). Additionally, as ently reviewed study was ~15%, and many
Greg noted last month, a 2012 meta-analysis changes of the within-group strength changes
(4) found that running caused a greater inter- were below that (Table 3).
ference effect than cycling.
All of the above is to say that a myriad of fac-
A final reason to consider for the low mag- tors (low resistance training volume, using
nitude of change in this study is that subjects running instead of cycling, lifting immediate-
may have been in a caloric deficit due to the ly following running, an enormous amount
extreme amount of additional military train- of additional military training, and a likely
ing. While post-study body mass values were caloric deficit) contributed to the low rate of
not presented to confirm this, it seems likely. adaptations. However, this is not necessarily
In fact, Murach and Bagley (5) suggested that a knock on the study; rather, the study was
the interference effect all but disappears when designed to see if the minimal strength pro-
a caloric deficit is avoided with concurrent gram could improve strength and muscle size
training. In the original concurrent training in military conscripts undergoing basic train-
study, Hickson 1980 (6), untrained subjects ing. In that case, the study succeeded and it
lost 0.8kg of body mass over 10 weeks, yet provides a good framework for the authors
still improved back squat strength by 25% going forward. Additionally, there were still
while performing six days of both endurance increases for most measures of strength, and
training and lifting. Therefore, there is a prec- when combining groups to increase statis-
edent for novice lifters to have large strength tical power, there was still hypertrophy de-
gains with concurrent training in the initial spite the study’s design. For hypertrophy, the
stages. However, the largest change in max- standard group actually improved type I fiber
area, which could be explained by the fact that
this group’s program was not monitored and
was more endurance-based than the micro
THE INTERFERENCE and classic groups. The significant increase in
type IIa fiber distribution in the micro-train-
EFFECT ALL BUT ing group, but no change in the classic group,
could be due to the classic group performing
DISAPPEARS WHEN lifting immediately after running instead of
getting a two-hour break. The fact that type
A CALORIC DEFICIT II fiber growth didn’t occur within any indi-
vidual group, and only when all groups were
IS AVOIDED WITH combined also shouldn’t be surprising. Hyper-
trophy takes longer to manifest than strength
CONCURRENT TRAINING in the initial stages of training (7); however,
the results do show that muscle growth tended

74
to occur even with low volume. The increase lifting after aerobic exercise, you could only
in satellite cells per fiber demonstrates a posi- perform assistance work in that spot. None
tive adaptation, as more satellite cells per fiber of that is the “optimal” approach, and it may
are associated with more myonuclei per fi- never come into play for you; however, the
ber, which could lead to the ability to support concept of micro-training is interesting, and it
a larger fiber area (8). We have covered this does give you the potential to split up training
concept previously, as more myonuclei per and ensure that the major lifts remain mostly
myofiber were shown to be related to a greater unaffected by cardio.
ability to gain muscle mass.
An interesting part of this study is that there Concurrent Training –
may be a practical benefit to the concept of In General
micro-training. The greater proportion of type
For this section, we are going to look at con-
IIa fibers at post-study in the micro-train-
current training from the perspective of the
ing group could be due to the fact that the
lifter. Presumably, the lifter is performing
micro group had a two-hour break between
aerobic exercise to lose weight for physique
running and lifting. On the other hand, the
or weight class reasons or for general health
classic group performed the training back to benefits, and the former necessitates a caloric
back and the standard group performed an deficit. In these cases, the lifter does not really
hour-long cardio session, both of which may care about the aerobic exercise performance,
have caused a greater reliance on type I fibers. rather the lifter cares about the additional
Thus, it makes sense that interconversions to- caloric expenditure and possibly the cardio-
ward type IIa fibers were only made in the mi- vascular benefit that accompanies the addi-
cro-training group. In practice, one potential tional exercise. With that in mind, the lifter
mistake with concurrent training is training wants to avoid the interference effect. The
lower body lifting in too close of proximity good news is, if this is you, there are some
to aerobic exercise due to time constraints. If straightforward strategies that you can use to
you have to train both modalities of exercise in implement cardio in a way that minimizes the
close proximity, it makes sense to perform the interference effect, even if you are in an ener-
lifting first (assuming lifting is your priority), gy deficit. To clarify, if you are interested in
so that performance is unaffected. However, maximizing performance on both an aerobic
if constrained for time, perhaps using a few training discipline and lifting, then eventu-
“micro” sessions of just 15 minutes of lifting ally something will have to give. Further, if
spread throughout the week could be of ben- you are an aerobic athlete, adding in appro-
efit. You could save the squats and deadlifts priate resistance training can indeed help to
for non-cardio days or perform a couple sets improve performance (9). Thus, Table 5 is a
of squats in a micro-session prior to aerobic one-stop-shop of sorts that illustrates every-
exercise. If you treat squats and deadlifts that thing you can do as a lifter to implement car-
way, then if you have to perform lower body dio in a logical and common sense manner.

75
I imagine that most of what’s in Table 5 isn’t a mechanistic point of view, using sprinting is
too surprising, as we have in part covered this better than moderate duration steady state car-
before (one, two, three); however, I always like dio to avoid the interference effect, then that
to clarify when discussing concurrent training would technically be correct. However, this
because it’s a topic that is unnecessarily com- is one of the cases where what’s optimal isn’t
plicated by many. One point that does need to always practical. For one, sprinting causes a
be addressed from the above table is the idea significant amount of muscle damage in ath-
of using sprinting or HIIT as the primary mode letes such as soccer and rugby players who are
of cardio. If someone were to say that, from already used to sprinting, with effects lasting

76
up to 48-72 hours (12). If your sprints are on A final note on incorporating cardio for the lift-
a bicycle, that is certainly a better option than er is that you can be quite flexible since aero-
running sprints, but the fatigue will still likely bic performance is generally not the goal. I’ve
be greater than it would be with moderate in- said in MASS before that hypertrophy train-
tensity steady state cycling. So, sure, HIIT is ing is more forgiving than strength training.
mechanistically similar to lifting, but if you get In other words, if you’re training for hypertro-
in two or three HIIT sessions per week and two phy, you don’t have a specific lifting perfor-
or three lower body lifting sessions per week, mance goal; rather, your performance is your
then at some point the fatigue from HIIT may physique. Therefore, you have more freedom
bleed into one of your lower body training ses- as a physique athlete to vary rep ranges and
sions and compromise performance. Indeed, choose exercises that you like, whereas a pow-
a meta-analysis shows that HIIT can indeed erlifter has to squat, bench press, and deadlift
attenuate strength gains (13 - MASS review), with low reps for a decent amount of the time.
which suggests that some of these practical So, if you are not concerned with your aerobic
limitations may come into play. Additional- performance, then you can mix in HIIT, steady
ly, HIIT is just much more mentally difficult state cycling, elliptical training, rowing, swim-
to complete (granted not for everyone) than ming, and really anything that doesn’t bend
moderate steady state cardio is. If you lift in the rules in Table 5 too much to achieve the
the morning every day and your goal is to get desired caloric expenditure or health benefits.
three cardio sessions in per week after work or
school when you are already fatigued, it’s not
going to be easy to adhere to three HIIT ses-
sions per week. However, it’s probably reason- YOU ARE NOT CONCERNED
able to knock out a couple 30-minute cycling
sessions at 50-60% VO2max. Besides, cycling
WITH YOUR AEROBIC
a couple times per week for 30 minutes at a PERFORMANCE, THEN YOU
moderate intensity has been shown to have ab-
solutely no negative effect on hypertrophy and
CAN MIX IN HIIT, STEADY STATE
strength gains (14). If that last sentence sounds CYCLING, ELLIPTICAL TRAINING,
surprising, remember that at its outset, this ar-
ticle noted that most studies are designed to in- ROWING, SWIMMING, AND
duce the interference effect, but when studies REALLY ANYTHING THAT DOESN’T
use minimal amounts of cardio, the effect is
negligible. Therefore, sprinting can be a good BEND THE RULES IN TABLE 5
idea, but always consider whether you or your TOO MUCH TO ACHIEVE THE
client can fully adhere to all the sprinting, espe-
cially if the individual is a physique athlete and DESIRED CALORIC EXPENDITURE
in a prolonged caloric deficit, in which case en- OR HEALTH BENEFITS.
ergy levels may be low.

77
 APPLICATION AND TAKEAWAYS
1. This study showed that minimal resistance training, even in the presence of high
amounts of aerobic exercise, can still cause modest strength and size gains.
2. Conceptually, the idea of micro-training may have some merit. Specifically, if
you consistently perform aerobic and resistance training back to back, then try
performing some lower body lifting in a very short session at another time to
avoid the back-to-back scenario.
3. Overall, if you are performing cardio as a typical lifter, refer to Table 5 above as
your cheat sheet for how to incorporate cardio and mostly erase the interference
effect. The only caveat to Table 5 is that even though HIIT may be ideal from
a mechanistic perspective, high amounts of it can be impractical, so including
some steady state cardio may be more sustainable for some

You can even use a flexible template with your sessions over two months and nine of them
aerobic exercise. For example, you could aim were HIIT and nine are moderate intensi-
to complete 12 cardio sessions within a month ty steady state, I’d bet that session rating of
and stipulate that six will be HIIT and the oth- perceived exertion and perhaps the enjoy-
er six can be any non-running mode of cardio, ment of training would be better compared
and then perform the sprints on days in which to a group who performed 18 HIIT sessions.
your energy levels are the highest. Lastly, and In that design, a group should also be includ-
I know this may be blasphemy, but although ed which performed 18 steady state sessions.
this article has talked about how running is a Two other aspects of concurrent training that
poor choice (and it is), if you like running then I’d like to see investigated are training status
you should do it. Running once or twice per and individualization. Specifically, much of
week for a few miles isn’t enough to cause sig- the concurrent training literature is on novice
nificant interference as long as it is not done to intermediate lifters; but how much cardio
prior to lifting, or potentially 24 hours before a causes the interference effect in well-trained
heavy lower body day; just be smart about it. lifters? In terms of individualization, I sus-
pect that high responders don’t see the inter-
Next Steps ference effect to the same degree. So, select-
ing individuals with a high, moderate, and
I honestly feel that the concurrent training low number of satellite cells per myofiber
literature is pretty well-established at this and running them through individual magni-
point. However, I think the proposal above tudes of resistance training volume, but the
of a flexible template to determine the mo- same magnitude of aerobic exercise, would
dality of aerobic exercise has merit. If lifters be a good start to see if high responders are
needed to complete 18 time-equated cardio more resistant to the interference effect.

78
References

1. Kilen A, Bay J, Bejder J, Andersen AB, Bonne TC, Larsen PD, Carlsen A, Egelund J,
Nybo L, Mackey AL, Olsen NV. Impact of low-volume concurrent strength training
distribution on muscular adaptation. Journal of Science and Medicine in Sport. 2020 Apr 4.
2. Borg GA. Psychophysical bases of perceived exertion. Medicine & Science in Sports &
Exercise. 1982.
3. Sporer BC, Wenger HA. Effects of aerobic exercise on strength performance following
various periods of recovery. The Journal of Strength & Conditioning Research. 2003 Nov
1;17(4):638-44.
4. Wilson JM, Marin PJ, Rhea MR, Wilson SM, Loenneke JP, Anderson JC. Concurrent
training: a meta-analysis examining interference of aerobic and resistance exercises. The
Journal of Strength & Conditioning Research. 2012 Aug 1;26(8):2293-307.
5. Murach KA, Bagley JR. Skeletal muscle hypertrophy with concurrent exercise training:
contrary evidence for an interference effect. Sports Medicine. 2016 Aug 1;46(8):1029-39.
6. Hickson RC. Interference of strength development by simultaneously training for strength
and endurance. European journal of applied physiology and occupational physiology. 1980
Dec 1;45(2-3):255-63.
7. Abernethy PJ, Jürimäe J, Logan PA, Taylor AW, Thayer RE. Acute and chronic response
of skeletal muscle to resistance exercise. Sports Medicine. 1994 Jan 1;17(1):22-38.
8. Petrella JK, Kim JS, Mayhew DL, Cross JM, Bamman MM. Potent myofiber hypertrophy
during resistance training in humans is associated with satellite cell-mediated myonuclear
addition: a cluster analysis. Journal of applied physiology. 2008 Jun;104(6):1736-42.
9. Blagrove RC, Howatson G, Hayes PR. Effects of strength training on the physiological
determinants of middle-and long-distance running performance: a systematic review.
Sports medicine. 2018 May 1;48(5):1117-49.
10. Murlasits Z, Kneffel Z, Thalib L. The physiological effects of concurrent strength and
endurance training sequence: A systematic review and meta-analysis. Journal of sports
sciences. 2018 Jun 3;36(11):1212-9.
11. Balabinis CP, Psarakis CH, Moukas M, Vassiliou MP, Behrakis PK. Early phase changes
by concurrent endurance and strength training. Journal of Strength and Conditioning
Research. 2003 May 1;17(2):393-401.
12. Howatson G, Milak A. Exercise-induced muscle damage following a bout of sport
specific repeated sprints. The Journal of Strength & Conditioning Research. 2009 Nov
1;23(8):2419-24.

79
13. Sabag A, Najafi A, Michael S, Esgin T, Halaki M, Hackett D. The compatibility of
concurrent high intensity interval training and resistance training for muscular strength and
hypertrophy: a systematic review and meta-analysis. Journal of sports sciences. 2018 Nov
2;36(21):2472-83.
14. Dolan C. The influence of time-equated training programs on muscle hypertrophy,
strength, and body composition. Florida Atlantic University; 2015.

80
Study Reviewed: Effects of Arginine Supplementation on Athletic Performance Based on
Energy Metabolism: A Systematic Review and Meta-Analysis. Viribay et al. (2020)

A New Meta-Analysis Says…

Arginine Works Now?
BY ERIC TREXLER

Arginine has largely been written off as an ineffective supplement

for exercise performance. However, a new meta-analysis suggests
that it improves both aerobic and anaerobic performance. Did we
abandon arginine too quickly? Read on to find out.

81
 KEY POINTS
1. The current meta-analysis (1) sought to summarize the effects of arginine
supplementation on anaerobic (<5 minutes) and aerobic (>5 minutes)
performance.
2. Results indicated that arginine supplementation significantly improved both
aerobic (effect size [d] = 0.84; 95% confidence interval: [0.12, 1.56]; p = 0.02)
and anaerobic performance (d = 0.24 [0.05, 0.43]; p = 0.01). However, these
findings lack face validity, and there are methodological reasons to doubt the
reliability of these results.
3. For lifters and athletes planning to utilize nitric oxide precursors, citrulline and
dietary nitrate are more advisable options than arginine.

M
eta-analyses are said to be the ab- proves both aerobic (effect size [d] = 0.84;
solute pinnacle of scientific re- 95% confidence interval: [0.12, 1.56]; p =
search, placed at the very top of 0.02) and anaerobic performance (d = 0.24
the hierarchy of evidence. As such, it’s no [0.05, 0.43]; p = 0.01). As MASS readers
surprise that meta-analyses get a lot of atten- might recall from my previous reviews about
tion when they get published, and their find- creatine, vitamin D, and carnitine, I like to
ings tend to carry more weight than an indi- take a deep dive when assessing meta-analy-
vidual study. However, we sometimes find ses, and this is bound to be another one. So,
instances where meta-analyses don’t seem to let’s get started.
line up with our less systematic, less quanti-
tative understanding of a body of literature. Purpose and Hypotheses
This month, we find ourselves in that very
position. Back in the early 2000s, arginine Purpose
was all over the supplement market. As the The primary purpose of this paper was to sys-
most widely known precursor to nitric ox- tematically review and summarize the litera-
ide synthesis, it was the main ingredient in ture evaluating the effect of arginine supple-
the first generation of purported nitric oxide mentation on performance. Specifically, the
boosters, promising skin-splitting pumps and authors categorized performance outcomes
improved performance. However, research based on the predominant energy system in-
generally showed that arginine did a lacklus- volved, resulting in one analysis for “aero-
ter job on both counts, and it has largely been bic” outcomes and another analysis for “an-
replaced by citrulline, nitrate, and other in- aerobic” outcomes. The authors also noted
gredients related to nitric oxide metabolism. that an additional purpose of their paper was
In contrast, the presently reviewed meta-anal- to discuss the impact of dosage and timing of
ysis (1) reports that arginine significantly im- supplementation on performance outcomes.

82
Hypotheses are direct quotes from the manuscript:
The authors did not state a hypothesis, which 1. Well-designed experiments that included
is very common for systematic reviews and Arg supplementation.
meta-analyses.
2. Identical experimental condition in the
placebo or control group.
Subjects and Methods
3. Testing the effects of Arg supplementa-
Subjects tion on sports performance.
The authors’ search of the literature initial-
4. Clinical trial.
ly identified 115 unique studies. After going
through all of the inclusion and exclusion cri- 5. Clear information concerning the Arg
teria, they whittled the list down to 18 studies supplementation administration (timing
that qualified for the systematic review. Ac- and dosage).
cording to the authors, the 18 studies includ-
6. Published in any language.
ed in the systematic review had a combined
total of 394 participants; of this pooled sam- 7. Clear information about funding sources.
ple, 282 were trained athletes, and the sex
8. Absence of conflict of interests.
breakdown was 386 males to only 8 females.
I will point out that if you add up the number It’s not clear what the authors meant by
of subjects they report from each study in Ta- “well-designed,” as this is an inherently sub-
bles 2 and 3, you get a sum of 364 (not 394); I jective criterion without a clear operational
looked into a couple possible reasons for this definition provided. I’m also not crazy about
discrepancy, but couldn’t confidently identi- the idea of excluding all studies with reported
fy a reason why these numbers are different. conflicts of interest; while it’s important to
When it comes to the actual quantitative part know about conflicts of interest, they don’t
of the meta-analysis (that is, calculation of a inherently invalidate the work, and I’d hate
pooled effect size), they reduced the list from for researchers to think that transparently re-
18 to 15, as 3 of the studies had “insufficient porting their conflicts of interest will cause
and confusing data for meta-analysis.” researchers to disregard their findings in fu-
ture meta-analyses. The researchers also ex-
Methods
cluded studies if the arginine treatment also
Just like any systematic review or meta-anal- contained other supplements or was part of a
ysis, the process begins with a systemat- multi-ingredient formula, or if participants in
ic search of the literature. They searched a the study reported having health problems or
bunch of relevant keywords, then narrowed a previous injury.
the search based on their pre-specified in-
clusion and exclusion criteria. In order to After identifying the studies that qualified for
be included in this analysis, papers were re- inclusion, they extracted the relevant perfor-
quired to meet the following criteria, which mance data from each of the studies and cal-

83
culated Hedges g effect sizes. These are quite at. I am going to present the findings of the
similar to Cohen’s d, which you might be fa- study as the authors originally reported them;
miliar with from previous MASS articles. Es- please be sure to read my interpretation of the
sentially, the effect size is presented as the dif- findings before taking them at face value or
ference between two means, divided by their making any important decisions based on the
pooled standard deviation. For this meta-anal- reported findings.
ysis, performance outcomes were categorized
The authors reported that arginine signifi-
as aerobic (tests lasting more than 5 minutes,
cantly enhanced anaerobic performance, with
performed at an intensity lower than that re-
a pooled effect size of 0.24 [95% confidence
quired to achieve maximal oxygen consump-
interval: 0.05-043]; p = 0.01. The forest plot
tion [VO2max]) or anaerobic (tests lasting
for this analysis is presented in Figure 1, and
less than 5 minutes, performed at an intensity
the funnel plot is presented in Figure 2. The
above VO2max). The authors used a random
test for heterogeneity was nonsignificant, and
effects model for the meta-analysis, and did
the funnel plot was sufficiently symmetrical.
statistical tests to check for funnel plot asym-
metry and heterogeneity between studies. The authors reported that arginine signifi-
cantly enhanced aerobic performance, with a
Findings pooled effect size of 0.84 [95% confidence in-
terval: 0.12-1.56]; p = 0.02. The forest plot for
This is a bit out of the ordinary, but I’m going this analysis is presented in Figure 3, and the
to begin the findings section with a big cave- funnel plot is presented in Figure 4. The test

84
 cally convert the results reported in the orig-
inal papers to a more usable form for the
meta-analysis, which can lead to some very
common errors. When I review a meta-anal-
ysis, I typically look for some of the key an-
alytical choices that I might have made dif-
ferently, and I look out for some of the more
common calculation or conversion errors that
frequently occur. If I spot those, I tend to ad-
just the analysis myself and report the adjust-
ed findings. With the current meta-analysis, I
was unable to do so.
I’m going to run through some of the issues
I spotted in this meta-analysis; I’m not do-
ing this to be harsh on the authors, but rath-
er to explain why I was unable to adjust the
analysis and report alternate results. In the
process of reviewing the meta-analysis, I no-
for heterogeneity was statistically significant, ticed that at least two, but almost certainly
and the funnel plot had significant asymme- three, of the aerobic performance outcomes
try. It’s not hard to find the root cause of this were calculated using standard errors instead
heterogeneity and asymmetry; three of the ef- of standard deviations. If you’re trying to
fect sizes were absolutely enormous compared identify which ones, they’re the three effect
to the rest. If you remove these three studies sizes that are astronomically larger than all
from the funnel plot (Figure 4B), the rest of the others. That alone would be a quick and
the literature is quite symmetrical. easy fix, but I noticed a few other things. This
analysis also did what is commonly known
Interpretation as “double-counting” (or, in this case, qua-
druple-counting). For example, the study by
Meta-analyses are really hard to do. They
Hurst et al (2) reported four different aerobic
take a ton of time and effort, but they also
outcomes; if the analysis isn’t mathematical-
involve a number of key decisions regarding
ly adjusted to reflect this, then a single study
how the analysis will be performed. To make
impacts the analysis about four times more
the process even more challenging, there are
strongly than it should. Again, this is another
sometimes instances where there is no clear-
quick and relatively easy fix, but other things
ly “correct” answer for these analytical ques-
popped up during my review.
tions, but rather a series of options with pros
and cons for each. There are also instances It appears as if the analysis includes multi-
in which the meta-analyst must mathemati- ple different types of effect sizes, but treats

85
them as if they’re the same. For Cohen’s D your heart rate returns to normal after exer-
effect sizes (or in this case, Hedges’ g, which cise. While this might be a decent proxy for
is very similar), you have the option to stan- general cardiovascular fitness levels in the
dardize based on raw standard deviations or general population, I’d hardly consider this a
change score standard deviations. Think of it performance test in a sample of soccer (foot-
this way: a group might bench 100 ± 20 kg ball) players, so I think this study probably
at baseline, and a supplement might increase should’ve been excluded from consideration.
their bench press values by an average of 3 ± In addition, at least one study provided ar-
5 kg. This gain of 3kg could be presented as ginine alongside another supplement (piper-
an effect size of 3/20kg (raw score standard- ine) (4); while I doubt the piperine is likely
ization) or 3/5kg (change score standardiza- to have altered the results, the meta-analy-
tion). Obviously, these are different metrics sis exclusion criteria indicate that this study
with very different values. The current me- should’ve been excluded.
ta-analysis includes a mixture of both, which
is a problem. The analysis also includes at I also noticed that in the process of recreating
least one outcome that probably shouldn’t be the original analysis, I was unable to replicate
categorized as a “performance” outcome. In some of the calculated effect sizes. For some,
the study by Pahlavani (3), the Harvard Step I was able to figure out why; for example, one
Test was used to create a performance met- of the outcomes from Bailey et al (5) appears
ric for the sample of trained soccer (football) to be a comparison between arginine and the
players. For this test, you step up and down citrulline group, rather than arginine versus
from a 20cm box for 5 minutes, at a cadence the placebo group. For others, however, I
of 30 step-ups per minute. Your score for this was unable to figure why my replication at-
test is calculated solely based on how rapidly tempts were unsuccessful. One such example

86
is the results from Hurst et al (2). The authors
reported four effect sizes from this study; I
was able to replicate three of the four, but I
don’t understand how they got an effect size
THE OVERWHELMING
of zero for the fourth one. They also got an
effect size of zero for one of the four anaer-
MAJORITY OF
obic outcomes obtained from Campbell et al
(6); again, I don’t understand how this value
ARGININE STUDIES
was obtained based on the data reported in
the original study. FAIL TO IDENTIFY
That’s a lot of information, but it all boils
down to a singular conclusion: I wouldn’t feel
SIGNIFICANT
comfortable reporting any adjusted or updat-
ed results without going back and essentially
ERGOGENIC EFFECTS.
starting this meta-analysis from scratch. If the
research question were a bit more compel-
ling, I might have actually done that. Howev- number of efficacious supplements, arginine
er, for two main reasons, I think we’ve gotten wasn’t even mentioned in the paper. When
to the point where it’s simply really, really looking at the larger effect sizes within the
hard to justify arginine supplementation. The current meta-analysis, many of them are con-
first reason is that the studies evaluating ar- sequences of some of the issues discussed
ginine supplementation’s effects on perfor- previously, including the use of change score
mance generally report underwhelming re- standardization instead of raw score stan-
sults. A 2012 systematic review by Bescos et dardization, or the use of standard error in-
al (7) noted that based on the literature avail- stead of standard deviation when calculating
able at that time, arginine had pretty unim- the effect size (both of which massively in-
pressive effects on performance, particularly flate effect sizes in a manner that’s complete-
when it was ingested alone (in the absence ly unwarranted). So, while it wouldn’t be fair
of other ergogenic ingredients) or tested in to say that there are absolutely no studies
relatively well-trained individuals. More re- showing some positive performance effects
cently, a 2018 position stand published by of arginine, it would definitely be fair to say
the International Society of Sports Nutrition that the overwhelming majority of arginine
highlighted that the overwhelming majority studies fail to identify significant ergogenic
of arginine studies have failed to identify any effects, and that the few positive results are
notable performance improvements (8). That typically observed in relatively untrained
same year, the International Olympic Com- samples using supplements that contain in-
mittee released a consensus statement on di- gredients other than arginine alone. It’s also
etary supplements (9); while they outlined a true that there are some scenarios in which

87
arginine supplementation might be advisable shown citrulline to be well tolerated at doses
for specific clinical applications, but those as high as 15g (10).
scenarios are well beyond the scope of this
When it comes to the performance effects of
article, and not directly relevant to the over-
citrulline, the literature is full of mixed find-
whelming majority of the MASS readership.
ings. Way back before I was part of the MASS
The second reason that it’s hard to justi- team, I published a meta-analysis about ci-
fy arginine supplementation is because we trulline’s effects on strength and power out-
have other supplements that do exactly what comes, and Greg reviewed it in Volume 3.
arginine is supposed to do, but more effec- The analysis indicated that citrulline malate
tively. As I mentioned in the introduction, had a small but statistically significant effect
people had high hopes for arginine based on on strength and power outcomes, particularly
its role as a precursor to nitric oxide synthe- when it came to strength endurance. Howev-
sis. The theory was that oral arginine sup- er, as Dr. Helms pointed out all the way back
plementation would increase blood arginine in Volume 2, the citrulline literature has fol-
levels, which would increase nitric oxide lowed an interesting trajectory: Many of the
production, thereby enhancing a number of early studies on strength and power outcomes
exercise-related parameters including blood reported positive findings, while many of the
flow, exercise efficiency, and the contractile more recent studies have reported less prom-
function of muscle (Figure 5). Unfortunate- ising findings. Just a few months ago, I pub-
ly, arginine has some unfavorable character- lished a review paper (10) with Dr. Adam
istics that limit its utility as a supplement. Gonzalez, which looked at citrulline’s effects
Most notably, a significant portion of orally on exercise performance in general (that is,
ingested arginine is broken down by bacte- including aerobic outcomes in addition to
ria and the arginase enzyme in the gut and strength and power). Overall, the assessment
liver, and it is also susceptible to extraction of the broader literature was pretty similar;
from systemic circulation (10). As a result, there are certainly a number of studies re-
citrulline supplementation actually increas- porting some promising ergogenic effects,
es blood arginine levels to a greater extent but the literature is far from unanimous. At
than an equivalent dose of oral arginine this point in time, there’s enough positive ev-
(11). Plus, if you’re thinking you should just idence to justify supplementing with citrul-
bump up the arginine dose to make up for line, particularly for individuals interested in
its low bioavailability, it’s important to note strength endurance outcomes, but too many
that gastrointestinal discomfort tends to be null findings to guarantee its efficacy with
reported with arginine doses up around 13g 100% confidence. Having said that, it’s quite
(10). Some studies have also reported minor affordable in comparison to typical supple-
gastrointestinal discomfort with citrulline ment prices, well-tolerated with minimal risk
doses that are higher than the recommended of adverse effects for healthy individuals, and
ergogenic dose, although some research has its supporting evidence is way more promis-

88
ing than the research on arginine. In terms of it’s also stronger than the citrulline literature.
practical application, studies with positive re- However, when it comes to strength and pow-
sults typically utilize 3-6g doses of L-citrul- er outcomes, the comparison between nitrate
line or 6-8g doses of citrulline malate (with a and citrulline becomes less clear, simply due
2:1 or 1:1 ratio of citrulline to malate), with to a lack of nitrate data. I recently reviewed
the supplement ingested 60-90 minutes prior one of the very few studies assessing the ef-
to exercise. fects of beetroot juice on a real-world strength
test with a decent degree of ecological valid-
Of course, as you can see in Figure 5, citrul- ity. Beetroot juice supplementation increased
line isn’t our only viable alternative to argi- the number of repetitions performed across
nine. Dietary nitrate (NO3-), which is found in three sets of bench press to failure with 70%
foods like beetroot, celery, spinach, and other of 1RM, and the magnitude of the effect was
leafy greens, is readily converted to nitrite, and pretty similar to what we’ve seen in some of
then to nitric oxide. If you’re up to date with the citrulline malate literature. While it’s hard
your MASS content, you might recall that I to say if nitrate is necessarily a better or worse
previously wrote about nitrate in both Volume option than citrulline-based supplements for
3 and Volume 4. When it comes to blood flow strength outcomes, the literature indicates
and endurance exercise performance, the di- that beetroot juice and other sources of nitrate
etary nitrate literature is much stronger than have a much more pronounced effect on ni-
the arginine literature, and I would argue that tric oxide production, blood flow, and endur-

89
ance performance in comparison to arginine, the top of the hierarchy of evidence. How-
so it seems pretty safe to conclude that nitrate ever, that placement is based on a number of
makes more sense than arginine for strength very, very critical assumptions: the search
and power outcomes as well. I should also note identified all of the relevant studies, the au-
that while the comparison of nitrate versus ci- thors weeded out the studies that shouldn’t be
trulline is currently unclear in the context of included, the individual studies were carried
strength outcomes, we don’t necessarily have out and reported effectively, an appropriate
to choose one or the other. Both increase nitric selection of outcomes was extracted from the
oxide production via separate pathways, and studies, effect sizes were calculated effec-
ergogenic nitrate doses can very realistically tively, and the statistical approach to the me-
be obtained by simply eating more foods with ta-analysis was appropriate. If one or more
higher nitrate content. There are also some of these assumptions are violated, that place
well-documented quality control issues when atop the hierarchy of evidence doesn’t mean
it comes to standardizing the actual nitrate much anymore, and it’s quite common to see
content of many of the commercially available one or more of these assumptions violated.
beetroot-based supplements (12). As a result, I
think a pretty sensible approach is to get plen- Next Steps
ty of nitrate from vegetable and fruit sources
(aiming for 400-800mg per day) and opt for a I view arginine the way I’d view a hypothet-
citrulline malate supplement (with a 2:1 citrul- ical researcher who pioneered a field many,
line to malate ratio) for someone who is inter- many decades ago. None of their theories re-
ested in trying to utilize nitric oxide precursors ally hold up anymore, their research methods
for performance purposes.
In conclusion, I don’t think the results of the
current meta-analysis should be used to dic-
tate your supplementation habits. We’ve got THE BODY OF ARGININE
a decent body of arginine literature available,
and it generally fails to identify ergogenic LITERATURE IS PRETTY
effects that are large enough or consistent
enough to warrant supplementation, particu-
UNDERWHELMING, AND
larly in pretty well-trained folks. More impor-
tantly, we have other ingredients that are eas-
BOTH CITRULLINE AND
ily accessible, well tolerated, and seem to do DIETARY NITRATE HAVE
arginine’s job better than arginine does. This
study also serves as another example of why EMERGED AS MUCH MORE
we can’t uncritically accept the results of me-
ta-analyses at face value. Of course, I would PROMISING ALTERNATIVES.
agree in theory that meta-analyses belong at

90
 APPLICATION AND TAKEAWAYS
It’s pretty tough to justify the use of arginine as a performance-enhancing
supplement. The body of arginine literature is pretty underwhelming, and both
citrulline and dietary nitrate have emerged as much more promising alternatives.
While we don’t yet have sufficient evidence to definitively conclude that citrulline
or nitrate will improve performance with 100% certainty, there’s enough positive
research (and low enough risk of adverse events or ergolytic effects) to give us
some optimism. For those wishing to utilize nitric oxide precursors for performance
purposes, a pretty cost-effective and practical approach would be to supplement
with 6-8g of citrulline malate around 60-90 minutes prior to training sessions, and
to aim for 400-800mg/day of nitrate intake by increasing the intake of nitrate-rich
vegetables and fruits.

look shockingly simplistic in hindsight, and

our current understanding of their research
topic is miles ahead of where they got with
it, but they innovated and helped lay a foun-
dation for people to build upon. Arginine
isn’t an ideal ingredient, but it paved the way
for nitric oxide precursor supplements, and
brought us closer to more viable alternatives
like citrulline and nitrate. We have plenty
more to learn about how nitric oxide precur-
sor supplements may affect strength and hy-
pertrophy, but I hope resources are dedicated
to research on citrulline and nitrate rather than
arginine. We still need more research on the
acute effects of citrulline, but we have sub-
stantially fewer studies looking at the acute
effects of beetroot juice and other dietary ni-
trate sources, so I hope we’ll get more of both
in the years to come. In addition, we really
need some longitudinal data to determine if
the day-to-day boost in acute performance
actually translates to better strength, power,
and hypertrophy gains over the course of a
2-3 month (at minimum) training program.

91
References
1. Viribay A, Burgos J, Fernández-Landa J, Seco-Calvo J, Mielgo-Ayuso J. Effects of
Arginine Supplementation on Athletic Performance Based on Energy Metabolism: A
Systematic Review and Meta-Analysis. Nutrients. 2020 May 2;12(5).
2. Hurst H, Sinclair J, Beenham M. Influence of Absolute versus relative L-arginine Dosage
on 1 km and 16.1 km time trial performance in trained cyclists. J Sci Cycl. 2014;3:2–8.
3. Pahlavani N, Entezari MH, Nasiri M, Miri A, Rezaie M, Bagheri-Bidakhavidi M, et al.
The effect of l-arginine supplementation on body composition and performance in male
athletes: a double-blinded randomized clinical trial. Eur J Clin Nutr. 2017;71(4):544–8.
4. Greer BK, Jones BT. Acute arginine supplementation fails to improve muscle endurance or affect
blood pressure responses to resistance training. J Strength Cond Res. 2011 Jul;25(7):1789–94.
5. Bailey SJ, Blackwell JR, Lord T, Vanhatalo A, Winyard PG, Jones AM. l-Citrulline
supplementation improves O2 uptake kinetics and high-intensity exercise performance in
humans. J Appl Physiol. 2015 Aug 15;119(4):385–95.
6. Campbell B, Roberts M, Kerksick C, Wilborn C, Marcello B, Taylor L, et al.
Pharmacokinetics, safety, and effects on exercise performance of L-arginine alpha-
ketoglutarate in trained adult men. Nutr. 2006 Sep;22(9):872–81.
7. Bescós R, Sureda A, Tur JA, Pons A. The effect of nitric-oxide-related supplements on
human performance. Sports Med. 2012 Feb 1;42(2):99–117.
8. Kerksick CM, Wilborn CD, Roberts MD, Smith-Ryan A, Kleiner SM, Jäger R, et al. ISSN
exercise & sports nutrition review update: research & recommendations. J Int Soc Sports
Nutr. 2018 Aug 1;15(1):38.
9. Maughan RJ, Burke LM, Dvorak J, Larson-Meyer DE, Peeling P, Phillips SM, et al. IOC
Consensus Statement: Dietary Supplements and the High-Performance Athlete. Int J Sport
Nutr Exerc Metab. 2018 Mar 1;28(2):104–25.
10. Gonzalez AM, Trexler ET. Effects of Citrulline Supplementation on Exercise
Performance in Humans: A Review of the Current Literature. J Strength Cond Res. 2020
May;34(5):1480–95.
11. Schwedhelm E, Maas R, Freese R, Jung D, Lukacs Z, Jambrecina A, et al. Pharmacokinetic
and pharmacodynamic properties of oral L-citrulline and L-arginine: impact on nitric oxide
metabolism. Br J Clin Pharmacol. 2008 Jan;65(1):51–9.
12. Gallardo EJ, Coggan AR. What’s in Your Beet Juice? Nitrate and Nitrite Content of Beet
Juice Products Marketed to Athletes. Int J Sport Nutr Exerc Metab. 2019 01;29(4):345–349.

92
Study Reviewed: The concept of Skeletal Muscle Memory: Evidence from Animal and
Human Studies. Snijders et al. (2020)

The Evidence for Myonuclear

Muscle Memory May be Murkier
Than Commonly Believed
BY GREG NUCKOLS

It’s commonly believed that myonuclei – the “control centers”

of muscle fibers – are added to muscle fibers when fibers grow,
but aren’t lost by fibers when they shrink, facilitating muscle re-
growth. This is a proposed mechanism for the phenomenon of
“muscle memory.” However, a recent review suggests that the
data is less conclusive than people may realize.

93
 KEY POINTS
1. Evidence is fairly clear that muscle fibers accrue more myonuclei as they
grow enough.
2. Whether myonuclei are lost following muscle fiber atrophy is less clear. Some
studies show a loss of myonuclei, while others show that myonuclei are retained.
3. The only direct evidence we have for myonuclei-mediated muscle memory
comes from rodent studies.
4. Retention of myonuclei following muscle atrophy may contribute to muscle
memory, but the relationship isn’t as straightforward as it’s often presented.

M
uscle fibers are very large individ- cle faster than a control group of mice that did
ual cells, and unlike smaller cells not have elevated levels of myonuclei. This
types that have a single nucleus, finding (in addition to other studies that had
muscle fibers have multiple nuclei, referred observed no loss in myonuclei in different an-
to as myonuclei. It’s thought that myonuclei imal atrophy models) led people to theorize
can only “oversee” a finite, relatively fixed that myonuclei aren’t lost upon detraining
volume of sarcoplasm (the “stuff” inside and atrophy, leading to elevated myonucle-
muscle fibers); the sarcoplasm overseen by a ar density; increased myonuclear density, in
particular myonucleus is its “myonuclear do- turn, allowing for faster muscle hypertrophy.
main.” As fibers get larger, and myonuclear Each myonucleus has a finite “transcriptional
domains approach their limits, muscle fibers capacity” (the amount of protein it can direct
must gain new myonuclei from surrounding ribosomes to build), so a greater density of
cells (satellite cells) to continue growing. myonuclei following detraining allows muscle
This collection of observations is often re- tissue to be re-built faster than it can be built
ferred to as “myonuclear domain theory.” originally. This mechanism is thought to un-
derpin the observation of “muscle memory” –
While muscle fibers must gain more myonu-
rapid regain of muscle following a layoff from
clei to continue growing, it’s not clear that they
training. I’m sure most people have seen a ver-
lose myonuclei when they atrophy. A ground-
sion of this iconic image (Figure 1) illustrating
breaking 2013 study found that mice gained
this concept (3).
muscle and accrued more myonuclei when
given testosterone (2). When testosterone This is a nice narrative, but how strong and
treatment was removed, the mice lost muscle consistent is the evidence for it? As it turns
tissue, but didn’t lose myonuclei. Following a out, many people (myself included) may
period of tension overload, the mice that had have gotten too excited, too soon. The pres-
previously received testosterone gained mus- ently reviewed publication (1) examines the

94
animal and human evidence for myonuclear
permanence (not losing myonuclei following Subjects and Methods
muscle atrophy) and myonuclei-facilitated Subjects
muscle memory. The evidence for both of
these concepts is somewhat weak and incon- The “subjects” in the present review were the
sistent, though I think the general concept findings of previous studies. There were 60
may still have some degree of merit. findings from animal studies and 16 findings
from human studies included in the review.
Purpose and Hypotheses Experimental Design
Purpose This was not a systematic review, which
The purpose of this review was to, “showcase means that there wasn’t a pre-specified
any evidence for the presence or absence of search strategy for identifying potential stud-
the proposed ‘muscle memory’ in both ani- ies to include, nor were there defined inclu-
mal and human models of muscle fiber atro- sion and exclusion criteria. However, in this
phy and hypertrophy.” Since this was a re- particular instance, I don’t think that’s a huge
view article, there were no hypotheses. drawback. This particular niche of muscle

95
physiology is pretty small, so I suspect the sessing the same region of muscle fibers over
authors of this review simply knew of all of time in vivo (i.e. in a living animal, without
the relevant studies in the area. The fine folks removing tissue). The second method is argu-
at Maastricht University that authored this re- ably better than the first, because there’s no
view certainly know this area of the research guarantee that muscle cross-sections are go-
better than I do, but for what it’s worth, I’m ing to be perfectly consistent; when you’re
not aware of any significant papers in this simply observing the same fibers and same
niche that weren’t included in this review (1). myonuclei over time, you eliminate that
I do think it’s essentially a systematic review, drawback. However, both of these methods
for all intents and purposes (not in form, but are still relatively new, so studies using them
in function). comprise a minority of the total research in
the area.
The review itself starts by explaining the re-
search methods people have used to study History
myonuclei, and the pros and cons of those The history of this line of research is interesting.
methods. Then, it explains how we arrived at I didn’t become aware of myonuclei research
our current conception of myonuclei-mediat- until after the concept of myonuclei-mediated
ed muscle memory. Finally, it summarizes the muscle memory was already reasonably popu-
evidence for and against myonuclei-mediated lar. However, the work that largely established
muscle memory in animal and human studies. that idea was published in 2012 and 2013 by
Bruusgaard, Egner, and Gundersen (2, 4). Prior
Findings to that point, it was generally believed that mus-
cle fibers gained myonuclei when they grew,
Research Methods
and lost myonuclei when they atrophied. And
Since MASS isn’t a review targeted at mus- in fact, there was good reason to believe that to
cle physiologists, I won’t explain the meth- be the case. Many studies had found evidence
ods used to assess myonuclei in excruciating of nuclear apoptosis (death) during atrophy, and
detail. However, it’s worth noting that there’s a good chunk of the early studies in the area did
one major issue associated with studying find that the total number of nuclei present in
myonuclei: Care needs to be taken to ensure a muscle cross-section decreased as muscle fi-
myonuclei are distinguishable from other nu- bers decreased in size (summarized in Table 1).
clei (like satellite cell nuclei). The consensus However, the groundbreaking studies in 2012
seems to be that there are two ideal methods and 2013 used the in-vivo method of myonucle-
for studying myonuclei. The first method in- ar assessment mentioned above, which clearly
volves staining muscle fibers with dyes that delineates between myonuclei and the nuclei
allow you to distinguish myonuclei, cell bor- of other cells in the area. The authors of those
ders, and satellite cells. The second method studies argued that previous studies were mere-
involves using gene transfer to label myonu- ly observing a loss in other nuclei (like satel-
clei with a fluorescent green protein, and as- lite cell and stromal cell nuclei), not myonuclei.

96
After those seminal studies were published, muscle atrophy; they didn’t know the meth-
the popular conception of myonuclei research ods being used were problematic, and since
started changing – myonuclei are accrued when there was plentiful evidence that nuclei were
muscle fibers grow, but they aren’t lost when being lost, what were the odds that all myo-
fibers atrophy. Thus, by this interpretation, nuclei were being spared?
myonuclei you accrue are effectively perma-
There are 13 studies that are a step up method-
nent, providing a mechanism for rapid muscle
ologically, but still don’t use ideal methods.
regrowth following atrophy.
In these studies, fibers are stained for both
Animal Research nuclei and the cell membrane (sarcolemma).
With the exception of one study on quail and All nuclei inside a fiber are myonuclei, and
one study on cats, all of the animal-based all of the nuclei outside a fiber are not myo-
myonuclei research has been conducted on nuclei. I’ve never looked down a microscope
mice and rats. Out of 60 research findings, and tried to count myonuclei, but apparently
39 used methods that are now known to be this method still leaves you with some judge-
problematic (only using nuclear staining, or ment calls, since most nuclei are clustered
running assays on muscle homogenate), due around the sarcolemma. In these 13 studies,
to difficulties distinguishing between myo- the research is almost perfectly split; seven
nuclei and the nuclei of other cells in the area. studies report a loss of myonuclei following
Of those 39 studies, 30 found evidence that muscle atrophy, while six studies find that
nuclei were lost following muscle atrophy. 34 myonuclei are maintained following atrophy.
of those 39 studies were carried out before Finally, just eight studies either use the ideal
2013, which explains why the field initially histological approach (staining for myonuclei,
believed that myonuclei were lost following satellite cells, and cell membranes), or assess

97
98
myonuclei in vivo. In the four studies using four found that myonuclear domains were
a histological approach, two report a loss of smaller in older adults, while seven found that
myonuclei following atrophy, and two report myonuclear domains were similar sizes in
no change. In the four studies using in vivo im- young and older adults. In studies that didn’t
aging, none report a loss of myonuclei. assess type I and type II fibers separately, five
found that myonuclear domains were smaller
Thus, as research methods improve, more and
in older adults, four found that myonuclear do-
more studies fail to find a decrease in myo-
mains were similar sizes in young and older
nuclei following atrophy. As such, it’s likely
adults, and two actually found that myonucle-
that during muscle atrophy, satellite cells and
ar domains were larger in older adults.
stromal cells decrease in density, but myonu-
clei are more likely to stick around. However, On balance, the picture painted by these stud-
there are studies that still indicate that myo- ies is fairly similar to the picture painted by
nuclei are lost following muscle atrophy, even animal research. 9 of 22 findings suggest that
when using high-quality methodology, which myonuclei are either preserved or lost at a rate
suggests that myonuclei aren’t truly perma- that’s slower than the rate of fiber atrophy,
nent, though they’re probably lost at a slower 11 suggest that fiber atrophy and myonuclei
rate than the rate at which atrophy occurs. loss occur in similar proportions, and 2 sug-
gest that myonuclei are lost to a greater ex-
Cross-sectional human research
tent than atrophy occurs. The two age-based
The vast majority of the human research in cohorts in most of these studies are separated
this area is cross-sectional. Rather than in- by 40-50 years, so the fact that almost half of
duce muscle atrophy, researchers compare the studies find that myonuclear domains are
muscle fiber size, myonuclear content, and larger in the older cohort may suggest that,
myonuclear domain size between young and while myonuclei can be lost during the atro-
older adults. Older adults (60+ years old) phy process, they may not be lost at quite the
generally have smaller muscle fibers than same rate as atrophy occurs. However, these
younger adults – especially type II fibers. If a studies also suggest that “myonuclear perma-
study finds that older adults have smaller fi- nence” probably does not occur – over a time
bers than young adults but similar myonucle- span of decades, some degree of myonucle-
ar content, and thus smaller myonuclear do- ar apoptosis probably does occur. Note that
mains, that suggests that myonuclei weren’t I’m using guarded language, however. Since
lost following age-related muscle atrophy. these studies are cross-sectional in nature,
we can’t necessarily assume causation, and
The evidence is mixed. I’ll ignore findings
we also can’t necessarily assume that muscle
specific to type I fibers (for the most part,
atrophy due to aging has the same effect on
younger and older adults have type I fibers of
myonuclei as muscle atrophy due to detrain-
similar size with similar myonuclear content
ing.
and similar myonuclear domain sizes), but of
the groups where type II fibers were assessed, Longitudinal Human Research

99
There are four longitudinal studies worth induced both muscle growth and myonuclear
mentioning. Only one attempted to repli- addition. Then, testosterone administration
cate the mouse study demonstrating myonu- was withdrawn, and the mice lost muscle,
clei-related muscle memory, but all four are returning to their baseline levels of muscu-
relevant to the proposed myonuclei-mediated larity. However, myonuclei content remained
muscle memory model. elevated in spite of muscle atrophy. Then, the
original group of mice underwent a 14-day
As mentioned in the intro, a 2013 study by overload period (induced by synergist abla-
Egner et al found evidence of myonuclei-re- tion), along with another group of mice that
lated muscle memory in mice (2). They ad- had not received testosterone treatment and
ministered testosterone to the mice, which had not accrued additional myonuclei. The

100
group of mice that had previously received within-subject design, and hypertrophy was
testosterone treatment and accrued more initially induced via resistance training in-
myonuclei experienced twice as much hyper- stead of testosterone administration. Subjects
trophy as the group of mice that had not pre- trained one leg for 10 weeks, detrained for 20
viously received testosterone treatment. weeks, and then trained both legs for 5 weeks.
However, Psilander’s study suffered from
A recent study by Psilander and colleagues two major issues: While muscle fibers in the
attempted to replicate Egner’s results (5). In- trained leg grew during the initial 10 weeks
stead of using a control group, they used a of training, the subjects didn’t accrue more

101
myonuclei, and the fibers didn’t atrophy back domains at baseline experienced significant in-
to their initial size during the 20 week de- creases in muscle fiber size following training,
training period. This wasn’t Psilander’s fault, while subjects with larger myonuclear domains
naturally, but it rendered the study unable failed to experience significant hypertrophy.
to actually assess myonuclei-related muscle
memory since neither myonuclear accretion Interpretation
nor detraining-induced fiber atrophy went ac-
cording to plan. To start this section, I want to make one thing
clear: The question here isn’t, “is ‘muscle
Another study in older adults had more suc- memory’ real,” but rather, “does this partic-
cess with manipulating myonuclei content ular mechanism seem to be a primary driver
(6). It involved six months of training and of the ‘muscle memory’ phenomenon?” The
one year of detraining. It wasn’t designed phenomenon of “muscle memory” – regain-
to fully test the idea of muscle memory be- ing muscle post-atrophy faster than building
cause it didn’t include a retraining period, it initially – was first scientifically document-
but it did test the first part of the theory of ed in the early 90s (9), and both research (10)
myonuclei-related muscle memory – gaining and thousands of lifers’ anecdotes testify to
myonuclei following resistance training, and its existence. The issue at hand is whether
not losing those myonuclei following muscle myonuclear permanence (not losing myo-
atrophy. Subjects did experience hypertrophy nuclei following muscle atrophy) a) is a real
and an increase in myonuclear content during phenomenon and b) partially explains the
the six-month training period, but following phenomenon of “muscle memory.”
a year of detraining, both muscle fiber size
and myonuclear content had regressed back Regarding myonuclear permanence, I think
to baseline. there’s sufficient evidence to claim that it is, at
minimum, too simplistic. On one hand, the ab-
Finally, two studies have provided evidence re- solute highest quality studies we have (utiliz-
lated to the last part of the myonuclei-related ing longitudinal in vivo imaging) all report that
muscle memory concept – if your myonuclear myonuclei aren’t lost following atrophy in ro-
domains are smaller, you’ll build muscle more dent models. On the other hand, half of the an-
rapidly due to enhanced transcriptional capaci- imal studies using the second best method for
ty. A study in younger adults (7) found that sub- studying myonuclei (staining myonuclei, cell
jects with smaller myonuclear domains at base- membranes, and satellite cells) report a decrease
line did not gain muscle faster than subjects in myonuclei following atrophy, and more than
with larger myonuclear domains at baseline. half of the cross-sectional human research is
The subjects with larger myonuclear domains at consistent with a loss in myonuclei during
baseline simply accrued new myonuclei faster, age-related atrophy. “Myonuclear permanence”
and were thus able to build muscle at the same implies that all myonuclei stick around forever,
rate. However, a similar study in older adults (8) but my interpretation of these findings is that at
did find that subjects with smaller myonuclear least some myonuclei can be lost during or fol-

102
lowing significant muscle atrophy. as someone gets older. Maybe you work out
in high school and college, build a lot of mus-
However, “myonuclei can be lost during or
cle, accrue a lot of myonuclei, and then stop
following atrophy” is not necessarily synon-
training for a decade. You lose all the mus-
ymous with “myonuclei are lost at the same
cle you built, but you retain all of the myo-
rate that fibers atrophy.” I think one of two
nuclei you gained, helping you rebuild size
potential explanations (or a combination of
and strength when you get back in the gym
both) would explain the bulk of the seeming-
in your 30s. However, you may start losing
ly disparate findings:
those myonuclei in your 50s, and have myo-
1. Myonuclei can be lost following atrophy, nuclear content matching your muscle fiber
but the loss of myonuclei lags behind other size (i.e. have the same myonuclear domain
atrophic processes to a significant degree. size as someone who’d never worked out) by
the time you turn 70. Again, these numbers
2. Myonuclei are not lost following atrophy
are for illustrative purposes only.
in young people, but can be lost following
atrophy in older adults A blend of these two models might look some-
thing like this: You work out in high school
To unpack the first potential explanation,
and college, build a lot of muscle, accrue a
let’s assume someone’s muscle fibers de-
lot of myonuclei, and then stop training. Over
crease in size by 20% following six months
the next decade, you may lose myonuclei fol-
of detraining. It may be the case that the loss
lowing muscle atrophy, but the process is
of myonuclei lags by a few months, or even
slow. If your myonuclear content increased
a few years. After those same six months of
by 20% when you were training, maybe it’s
detraining, all of the myonuclei may still be
still elevated by 10% if you take 10 years off
sticking around, meaning that the myonu-
of training in your 20s or 30s. However, if
clear domain size has decreased. However,
you stopped training at 70 years old instead
after a year of detraining, maybe myonucle-
of 22 years old, you may lose myonuclei at
ar content has decreased by 10%, and after
a much faster rate. If it takes you six months
two years, myonuclear content has decreased
to lose all the muscle you built, maybe it only
20%, matching the loss of total fiber size. Un-
takes a year for all of those new myonuclei to
der this model, myonuclei aren’t permanent,
be lost as well.
but perhaps they could assist somewhat with
“muscle memory” if the detraining period is I’m not sure which of these three models is
less than two years long. To be clear, I’m not the closest to being completely correct, but
proposing that this is the actual time course I think all three are closer to the “truth” than
of myonuclei loss; these time increments are either simplistic model – myonuclear perma-
just being used to illustrate the model. nence, or myonuclear apoptosis occurring at
the same rate as muscle fiber atrophy.
The second explanation is more straightfor-
ward – myonuclei are more likely to be lost The next obvious question is, “if those myo-

103
nuclei do stick around for a pretty long time, harder to gain, so a smaller initial myonucle-
does that actually matter?” Remember, one ar domain was beneficial. This leads me to
idea underpinning the myonuclei-mediat- believe that, all else being equal, this piece of
ed muscle memory concept is the idea that the myonuclei-related muscle memory model
greater myonuclear density (smaller myonu- makes some sense – if your muscles atrophy
clear domains) is equated with enhanced tran- but you retain your myonuclei, that boost in
scriptional capacity, leading to more muscle transcriptional capacity might increase mus-
proteins being built following a resistance cle growth to some degree. However, once
training stimulus. As mentioned previously, you add in one more variable – the ability to
two studies have examined this idea (7, 8). A easily accrue more myonuclei (as was seen
study with young subjects found that people in the study on younger subjects) – it doesn’t
with initially smaller myonuclear domains seem that smaller myonuclear domains mat-
did not build muscle faster (7), while a study ter much. Thus, mechanistically, the “muscle
with older subjects found that subjects with memory” effects of myonuclei may actually
initially smaller myonuclear domains were be pretty small for most people. They may
able to build muscle faster (8). The difference play a bigger role in people who have a hard-
seems to relate to the subjects’ ability to accrue er time accruing more myonuclei (like elder-
more myonuclei. In the young subjects, folks ly people and folks who struggle with mus-
with larger myonuclear domains just accrued cle growth generally), but if you’re someone
more myonuclei and grew without a hitch. who’s relatively young and if you’ve previ-
In the older subjects, new myonuclei were ously had decent success with building mus-

104
cle, having more myonuclei sticking around another benefit to list when promoting re-
after a layoff may not actually contribute a sistance training. The second is athletics. If
huge amount to the “muscle memory” re- steroids allow athletes to accrue WAY more
sponse you see when you start training again. myonuclei than they’d be able to accrue natu-
rally, those athletes may wind up with a very
As a matter of fact, after re-reading the sem-
long-lasting advantage over lifetime drug-
inal Egner study (2) demonstrating muscle
free athletes. For example, if steroids help
memory in mice with a more critical eye,
someone build 50% more muscle than they’d
it became clear that the myonuclei likely
have been able to build drug-free, and most of
couldn’t fully explain the “muscle memo-
those myonuclei stick around after they stop
ry” response in that study. Following the
using steroids, they may be able to maintain
three weeks of muscle atrophy in the group
enough enough of those gains that they wind
of mice that previously received testoster-
up with 20% more muscle than they’d have
one, their muscle fibers were the same size as
been able to build drug-free. If those myo-
the mice that had received a sham treatment,
nuclei all stick around for, say, 10 years (or
and their myonuclear density (myonuclei per
longer), it would make the current WADA
mm of fiber length) was approximately 15%
doping bans (2 years for a first offense) seem
greater. Following two weeks of overload,
woefully inadequate, opening up the discus-
fiber growth was roughly two-fold great-
sion of lifetime bans for athletes caught using
er in the mice that had previously received
anabolic steroids.
testosterone. Just using rough numbers, if a
15% greater myonuclear density can increase So, if myonuclei can’t fully explain the mus-
transcriptional capacity by 15%, but you’re cle memory phenomenon, what are some oth-
attempting to explain a 100% difference in er possible mechanisms?
hypertrophy, some other mechanism(s) must
One leading candidate is epigenetic modifi-
be accounting for the remaining 85% differ-
cation (10). We’ve written about that previ-
ence in hypertrophy. Thus, retaining myonu-
ously in MASS. I’m honestly not sure about
clei following atrophy may be a mechanism
other mechanisms, because muscle tissue
of “muscle memory,” but it can’t be the only
is highly plastic, meaning it has the capac-
mechanism, and it’s likely not the most im-
ity to change and adapt rapidly (relative to
portant mechanism.
most other tissues, at least), but mechanisms
However, I do think there are two instanc- of muscle memory would need to involve
es where maintenance of myonuclei follow- changes that persist for at least months, if not
ing atrophy could still be relevant. The first years. Proteins are replaced regularly, mito-
is aging. Since elderly people have a harder chondrial density decreases rapidly following
time accruing new myonuclei, if we found detraining, capillarization decreases within a
that elderly people have smaller myonuclear few weeks, etc. One possibility is ribosome
domains if they’d previously lifted weights density. Ribosomal content may be associ-
when they were younger, that would give us ated with hypertrophy (11), and I honestly

105
 APPLICATION AND TAKEAWAYS
The role of myonuclei in muscle memory is … hazy. It does seem that myonuclei
are lost at a slower rate than muscle fibers atrophy, but your myonuclei probably
aren’t actually permanent. Myonuclei may play a role in “muscle memory,” but other
mechanisms likely explain the lion’s share of the phenomenon.

have no idea how quickly ribosomal content enough to build a fair amount of muscle and
drops following detraining. Another possi- accrue a fair amount of myonuclei), and take
bility could be a neural mechanism of some biopsies periodically to observe how myo-
sort. Instead of needing to completely relearn nuclear content changes over a long period
motor patterns, you still retain some of your of time. Finally, I’d love to see a detraining
motor skills following a layoff, perhaps al- study on athletes who stop taking steroids,
lowing you to simply present your muscles but keep training hard. How much muscle
with a greater stimulus than a brand new do they end up losing? How does their myo-
lifter would be capable of inducing. Muscle nuclear content compare to people who’ve
memory is an interesting concept, and I be- never used steroids? Do they lose myonuclei
lieve the current research in the area is barely when they stop using steroids?
scratching the surface.

Next Steps
We need some longitudinal studies in humans.
A study designed like the Psilander study, but
that actually induced enough hypertrophy to
cause myonuclear accretion would be a good
start. For cross-sectional studies, we need
some studies with intermediate ages included.
Most of the research is looking for differenc-
es between 25 year olds and 70 year olds, but
those studies can’t tell us anything about the
time course of age-related myonuclear adap-
tations. Including a cohort of 40-50 year olds
would help considerably. I’d also be interest-
ed in a longitudinal study simply cataloging
the rate of myonuclei loss in humans follow-
ing 5-10 years of detraining. Have a group of
college students train for 6-12 months (long

106
References

1. Snijders T, Aussieker T, Holwerda A, Parise G, van Loon LJC, Verdijk LB. The concept
of skeletal muscle memory: Evidence from animal and human studies. Acta Physiol (Oxf).
2020;229(3):e13465. doi:10.1111/apha.13465
2. Egner IM, Bruusgaard JC, Eftestøl E, Gundersen K. A cellular memory mechanism aids
overload hypertrophy in muscle long after an episodic exposure to anabolic steroids. J
Physiol. 2013;591(24):6221-6230. doi:10.1113/jphysiol.2013.264457
3. Gundersen K. Muscle memory and a new cellular model for muscle atrophy and
hypertrophy. J Exp Biol. 2016;219(Pt 2):235-242. doi:10.1242/jeb.124495
4. Bruusgaard JC, Egner IM, Larsen TK, Dupre-Aucouturier S, Desplanches D, Gundersen K.
No change in myonuclear number during muscle unloading and reloading. J Appl Physiol
(1985). 2012;113(2):290-296. doi:10.1152/japplphysiol.00436.2012
5. Psilander N, Eftestøl E, Cumming KT, et al. Effects of training, detraining, and retraining
on strength, hypertrophy, and myonuclear number in human skeletal muscle. J Appl
Physiol (1985). 2019;126(6):1636-1645. doi:10.1152/japplphysiol.00917.2018
6. Snijders T, Leenders M, de Groot LCPGM, van Loon LJC, Verdijk LB. Muscle mass
and strength gains following 6 months of resistance type exercise training are only partly
preserved within one year with autonomous exercise continuation in older adults. Exp
Gerontol. 2019;121:71-78. doi:10.1016/j.exger.2019.04.002
7. Snijders T, Smeets JS, van Kranenburg J, Kies AK, van Loon LJ, Verdijk LB. Changes in
myonuclear domain size do not precede muscle hypertrophy during prolonged resistance-
type exercise training. Acta Physiol (Oxf). 2016;216(2):231-239. doi:10.1111/apha.12609
8. Holwerda AM, Overkamp M, Paulussen KJM, et al. Protein Supplementation after
Exercise and before Sleep Does Not Further Augment Muscle Mass and Strength Gains
during Resistance Exercise Training in Active Older Men. J Nutr. 2018;148(11):1723-
1732. doi:10.1093/jn/nxy169
9. Staron RS, Leonardi MJ, Karapondo DL, et al. Strength and skeletal muscle adaptations
in heavy-resistance-trained women after detraining and retraining. J Appl Physiol (1985).
1991;70(2):631-640. doi:10.1152/jappl.1991.70.2.631
10. Seaborne RA, Strauss J, Cocks M, et al. Human Skeletal Muscle Possesses an Epigenetic
Memory of Hypertrophy. Sci Rep. 2018;8(1):1898. Published 2018 Jan 30. doi:10.1038/
s41598-018-20287-3

107
11. Roberts MD, Haun CT, Mobley CB, et al. Physiological Differences Between Low Versus
High Skeletal Muscle Hypertrophic Responders to Resistance Exercise Training: Current
Perspectives and Future Research Directions. Front Physiol. 2018;9:834. Published 2018
Jul 4. doi:10.3389/fphys.2018.00834

108
 VIDEO: Executing Iron
Supplementation
BY MICHAEL C. ZOURDOS

Back in Volume 2, we covered the effectiveness of iron supplementation;

however, iron supplements come in various forms and with potential side
effects that must be discussed. This video picks up where we left off to discuss
the execution of taking an iron supplement.
Click to watch Michael's presentation.

109
Relevant MASS Videos and Articles
1. Iron Supplementation in Strength Sport (Volume 2 Issue 11)
2. Hepcidin, Iron’s Master Regulator, May Be Indicative of Your Energy Availability (Volume 3
Issue 8)

References
1. Pasricha SR, Low M, Thompson J, Farrell A, De-Regil LM. Iron Supplementation Benefits
Physical Performance in Women of Reproductive Age: A Systematic Review and Meta-
Analysis–3. The Journal of nutrition. 2014 Apr 9;144(6):906-14.
2. Mielgo-Ayuso J, Zourdos MC, Calleja-González J, Urdampilleta A, Ostojic S. Iron
supplementation prevents a decline in iron stores and enhances strength performance in elite
female volleyball players during the competitive season. Applied Physiology, Nutrition, and
Metabolism. 2015 Feb 10;40(6):615-22.
3. Rybo G, Sölvell L. Side-effect studies on a new sustained release iron preparation. Scandinavian
journal of haematology. 1971 Aug;8(4):257-64.
4. Ems T, Huecker MR. Biochemistry, Iron Absorption. InStatPearls [Internet] 2019 Apr 21.
StatPearls Publishing.
5. Trumbo P, Yates AA, Schlicker S, Poos M. Dietary reference intakes: vitamin A, vitamin
K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon,
vanadium, and zinc. Journal of the Academy of Nutrition and Dietetics. 2001 Mar 1;101(3):294.
6. Tolkien Z, Stecher L, Mander AP, Pereira DI, Powell JJ. Ferrous sulfate supplementation causes
significant gastrointestinal side-effects in adults: a systematic review and meta-analysis. PloS
one. 2015;10(2).
7. Stoffel NU, Cercamondi CI, Brittenham G, Zeder C, Geurts-Moespot AJ, Swinkels DW, Moretti D,
Zimmermann MB. Iron absorption from oral iron supplements given on consecutive versus alternate
days and as single morning doses versus twice-daily split dosing in iron-depleted women: two open-
label, randomised controlled trials. The Lancet Haematology. 2017 Nov 1;4(11):e524-33.
8. Hall R, Peeling P, Nemeth E, Bergland D, McCluskey WT, Stellingwerff T. Single versus split
dose of iron optimizes hemoglobin mass gains at 2106 m altitude. Medicine & Science in Sports
& Exercise. 2019 Apr 1;51(4):751-9.
9. McCormick R, Dreyer A, Dawson B, et al. The Effectiveness of Daily and Alternate Day Oral
Iron Supplementation in Athletes With Suboptimal Iron Status (Part 2). Int J Sport Nutr Exerc
Metab. 2020 Mar 1-6.

110
 VIDEO: Evidence-Based Practice
Part II
BY ERIC HELMS

Being “evidence based” is something that most MASS subscribers strive for as
coaches, content producers, or as a way to get the best results in the kitchen
or gym. In part 1 of this series, we established what evidence-based practice
actually is. In part 2, we will discuss the types of evidence and their use in
decision making
Click to watch Eric's presentation.

111
Relevant MASS Videos and Articles
1. VIDEO: Evidence Based Practice Part 1 (Volume 4, Issue 6)
2. The Basics of Interpreting Research: A How-to Guide for Each Section (Bonus)
3. VIDEO: Understanding the Research Process Part 1 (Volume 3, Issue 7)
4. VIDEO: Understanding the Research Process Part 2 (Volume 3, Issue 8)
5. VIDEO: Understanding the Research Process Part 3 (Volume 3, Issue 9)

References

1. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS. Evidence based medicine:
what it is and what it isn’t. BMJ: British Medical Journal. 1996 Jan 13;312(7023):71.
2. Ng C., Benedetto U. (2016) Chapter 2: Evidence Hierarchy. In: Biondi-Zoccai G. Umbrella
Reviews. Evidence Synthesis with Overviews of Reviews and Meta-Epidemiologic Studies.
Cham, Switzerland: Springer International. 2016.

112
 Just Missed the Cut
Every month, we consider hundreds of new papers, and they can’t all be included in MASS.
Therefore, we’re happy to share a few pieces of research that just missed the cut. It’s our
hope that with the knowledge gained from reading MASS, along with our interpreting research
guide, you’ll be able to tackle these on your own.

1. Aasa and Berglund. A descriptive analysis of functional impairments and patho-anatomical

findings in eight powerlifters
2. Blaauw. Activity‐dependent increases of protein synthesis in skeletal muscle: Sensing
the energy levels?
3. Jukic and Tufano. Acute effects of shorter but more frequent rest periods on mechanical
and perceptual fatigue during a weightlifting derivative at different loads in strength-
trained men
4. Di Rienzi and Britton. Adaptation of the Gut Microbiota to Modern Dietary Sugars and
Sweeteners
5. Vikmoen et al. Adaptations to strength training differ between endurance-trained and
untrained women
6. Tinline-Goodfellow et al. An Acute Reduction in Habitual Protein Intake Attenuates Post
Exercise Anabolism and May Bias Oxidation-Derived Protein Requirements in Resistance
Trained Men
7. Neto et al. Are the flat and arched bench press really similar?
8. Estafanos et al. Can you out‐exercise a bad sleep? A muscle‐centric view
9. Knudsen et al. Contraction‐regulated mTORC1 and protein synthesis: Influence of AMPK
and glycogen
10. Lundberg et al. Early accentuated muscle hypertrophy is strongly associated with
myonuclear accretion
11. Khoddam-Khorasani et al. Effect of changes in the lumbar posture in lifting on trunk
muscle and spinal loads: A combined in vivo, musculoskeletal, and finite element model
study
12. Pearson et al. Effect of Competition Frequency on Strength Performance of Powerlifting
Athletes
13. Hansen et al. Effects of alternating blood flow restricted training and heavy-load resistance
training on myofiber morphology and mechanical muscle function
14. Correia et al. Effects of Intermittent Fasting on Specific Exercise Performance Outcomes:
A Systematic Review Including Meta-Analysis
15. Chaves et al. Effects of resistance training with controlled versus self-selected repetition
duration on muscle mass and strength in untrained men

113
16. Schwanbeck et al. Effects of Training With Free Weights Versus Machines on
Muscle Mass, Strength, Free Testosterone, and Free Cortisol Levels
17. Solagna et al. Exercise‐dependent increases in protein synthesis are accompanied
by chromatin modifications and increased MRTF‐SRF signalling
18. Przewłócka et al. Gut-Muscle Axis Exists and May Affect Skeletal Muscle Adaptation
to Training
19. Gardiner et al. Injury risk and injury incidence rates in crossfit: a brief review
20. Boutros et al. Is a vegan diet detrimental to endurance and muscle strength?
21. Steele et al. Is Competitive Body-Building Pathological? Survey of 984 Male
Strength Trainers
22. Zaras et al. Lean Body Mass, Muscle Architecture, and Performance in Well-Trained
Female Weightlifters
23. Farrow et al. Lighter-Load Exercise Produces Greater Acute- And Prolonged-
Fatigue in Exercised and Non-Exercised Limbs
24. Gómez-Carmona et al. Lower-limb Dynamics of Muscle Oxygen Saturation During
the Back-squat Exercise: Effects of Training Load and Effort Level
25. Jukic et al. Magnitude and Reliability of Velocity and Power Variables During
Deadlifts Performed With and Without Lifting Straps
26. Martins et al. Metabolic adaptation is not a major barrier to weight-loss maintenance
27. Cristina-Souza et al. Panax ginseng Supplementation Increases Muscle
Recruitment, Attenuates Perceived Effort, and Accelerates Muscle Force Recovery
After an Eccentric-Based Exercise in Athletes
28. Longstrom et al. Physiological, Psychological and Performance-Related Changes
Following Physique Competition: A Case-Series
29. Sun and Kober. Regulating food craving: From mechanisms to interventions
30. Kim et al. Skeletal muscle adaptations to heat therapy
31. MacKenzie-Shalders et al. The effect of exercise interventions on resting metabolic
rate: A systematic review and meta-analysis

114
 Thanks for
reading MASS.
The next issue will be released to
subscribers on August 1, 2020.

Graphics and layout by Kat Whitfield

115