PHD MARCELO SALDANHA - ARTIGOS

Prévia do material em texto

ARTIGOS 
 
 
RECOMENDAÇÕES NUTRICIONAIS 
PARA ATIVIDADE FÍSICA E 
FUNDAMENTOS PARA A PRESCRIÇÃO 
DIETÉTICA 
 
Nutrition and Athletic
Performance
ABSTRACT
It is the position of the Academy of Nutrition and Dietetics, Dietitians of
Canada, and the American College of Sports Medicine that the performance
of, and recovery from, sporting activities are enhanced by well-chosen nu-
trition strategies. These organizations provide guidelines for the appropriate
type, amount, and timing of intake of food, fluids, and supplements to
promote optimal health and performance across different scenarios of
training and competitive sport. This position paper was prepared for mem-
bers of the Academy of Nutrition and Dietetics, Dietitians of Canada (DC),
and American College of Sports Medicine (ACSM), other professional as-
sociations, government agencies, industry, and the public. It outlines the
Academy_s, DC_s and ACSM_s stance on nutrition factors that have been
determined to influence athletic performance and emerging trends in the
field of sports nutrition. Athletes should be referred to a registered dietitian/
nutritionist for a personalized nutrition plan. In the United States and in
Canada, the Certified Specialist in Sports Dietetics (CSSD) is a registered
dietitian/nutritionist and a credentialed sports nutrition expert.
POSITION STATEMENT
It is the position of the Academy of Nutrition and Dietet-
ics, Dietitians of Canada, and the American College of
Sports Medicine that the performance of, and recovery from,
sporting activities are enhanced by well-chosen nutrition
strategies. These organizations provide guidelines for the
appropriate type, amount and timing of intake of food, fluids
and dietary supplements to promote optimal health and
sport performance across different scenarios of training and
competitive sport.
This paper outlines the current energy, nutrient, and fluid
recommendations for active adults and competitive athletes.
These general recommendations can be adjusted by sports
dietitians to accommodate the unique issues of individual
athletes regarding health, nutrient needs, performance goals,
physique characteristics (ie, body size, shape, growth, and
composition), practical challenges and food preferences. Since
credentialing practices vary internationally, the term ‘‘sports
dietitian’’ will be used throughout this paper to encompass all
terms of accreditation, including RDN, RD, CSSD, or PDt.
This Academy position paper includes the authors_ inde-
pendent review of the literature in addition to systematic
review conducted using the Academy_s Evidence Analysis
Process and information from the Academy Evidence
Analysis Library (EAL). Topics from the EAL are clearly
delineated. The use of an evidence-based approach provides
important added benefits to earlier review methods. The
major advantage of the approach is the more rigorous stan-
dardization of review criteria, which minimizes the likeli-
hood of reviewer bias and increases the ease with which
disparate articles may be compared. For a detailed descrip-
tion of the methods used in the evidence analysis process,
access the Academy_s Evidence Analysis Process at https://
www.andevidencelibrary.com/eaprocess.
Conclusion Statements are assigned a grade by an expert
work group based on the systematic analysis and evaluation
of the supporting research evidence. Grade I = Good; Grade
II = Fair; Grade III = Limited; Grade IV = Expert Opinion
Only; and Grade V = Not Assignable (because there is no
evidence to support or refute the conclusion). See grade
definitions at www.andevidencelibrary.com/.
Evidence-based information for this and other topics can be
found at https://www.andevidencelibrary.com and sub-
scriptions for non-members are purchasable at https://
www.andevidencelibrary.com/store.cfm.
EVIDENCE-BASED ANALYSIS
This paper was developed using the Academy of Nutrition
andDietetics Evidence Analysis Library (EAL) andwill outline
some key themes related to nutrition and athletic performance.
The EAL is a synthesis of relevant nutritional research on im-
portant dietetic practice questions. The publication range for
SPECIAL COMMUNICATIONS
This joint position statement is authored by the Academy of Nutrition and
Dietetics (AND), Dietitians of Canada (DC), and American College of
Sports Medicine (ACSM). The content appears in AND style. This paper is
being published concurrently in Medicine & Science in Sports & Exercise�
and in the Journal of the Academy of Nutrition and Dietetics, and the
Canadian Journal of Dietetic Practice and Research. Individual name rec-
ognition is reflected in the acknowledgments at the end of the statement.
Submitted for publication December 2015.
Accepted for publication December 2015.
0195-9131/16/4803-0543/0
MEDICINE & SCIENCE IN SPORTS & EXERCISE�
Copyright � 2016 by the American College of Sports Medicine, Academy
of Nutrition and Dietetics, and Dietitians of Canada
DOI: 10.1249/MSS.0000000000000852
ACADEMY OF NUTRITION AND DIETETICS
DIETITIANS OF CANADA
JOINT POSITION STATEMENT
543
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
http://www.andevidencelibrary.com/
http://www.andevidencelibrary.com/
http://www.andevidencelibrary.com/
https://www.andevidencelibrary.com
https://www.andevidencelibrary.com/store.cfm
https://www.andevidencelibrary.com/store.cfm
the evidence-based analysis spanned March 2006–November
2014. For the details on the systematic review andmethodology
go to www.andevidencelibrary.com. Table 1 presents the
evidence analysis questions used in this position paper.
NEW PERSPECTIVES IN SPORTS NUTRITION
The past decade has seen an increase in the number and
topics of publications of original research and review, consen-
sus statements from sporting organizations, and opportunities
for qualification and accreditation related to sports nutrition
and dietetics. This bears witness to sports nutrition as a
dynamic area of science and practice that continues to flourish
in both the scope of support it offers to athletes and the
strength of evidence that underpins its guidelines. Before
embarking on a discussion of individual topics, it is valu-
able to identify a range of themes in contemporary sports nu-
trition that corroborate and unify the recommendations in
this paper.
1. Nutrition goals and requirements are not static. Athletes
undertake a periodized program in which preparation for
peak performance in targeted events is achieved by
integrating different types of workouts in the various
TABLE 1. Evidence analysis questions included in the position statement Refer to http://www.andevidencelibrary.com/ for a complete list of evidence analysis citations.
EAL Question Conclusion and Evidence Grade
Energy Balance and Body Composition
#1: In adult athletes, what effect does negative energy balance have on exercise
performance?
In three out of six studies of male and female athletes, negative energy balance (losses of
0.02% to 5.8% body mass; over five 30-day periods) was not associated with decreased
performance. In the remaining three studies where decrements in both anaerobic and aerobic
performance were observed, slow rates of weight loss (0.7% reduction body mass) were
more beneficial to performance compared to fast (1.4% reduction body mass) and one study
showed that self-selected energy restriction resulted in decreased hormone levels.
Grade II- Fair
#2: In adult athletes, what is the time, energy, and macronutrient requirement
to gain lean body mass?
Over periods of 4 to 12 weeks, increasing protein intake during hypocaloric conditions maintains
lean body mass in male and female resistance-trained athletes. When adequate energy is
provided or weight loss is gradual, an increase in lean body mass may be observed.
Grade III- Limited
Recovery
#3: In adult athletes, what is the effect of consuming carbohydrate on
carbohydrate and protein-specific metabolic responses and/orof key interest: Calcium. Calcium
is especially important for growth, maintenance, and repair
of bone tissue; regulation of muscle contraction; nerve
conduction; and normal blood clotting. The risk of low
bone-mineral density and stress fractures is increased by low
energy availability, and in the case of female athletes,
menstrual dysfunction, with low dietary calcium intake
contributing further to the risk.78,99,100 Low calcium intakes
are associated with restricted energy intake, disordered eat-
ing and/or the specific avoidance of dairy products or other
calcium-rich foods. Calcium supplementation should be
determined after a thorough assessment of usual dietary in-
take. Calcium intakes of 1,500 mg/d and 1,500–2,000 IU/day
of vitamin D are needed to optimize bone health in athletes
with low energy availability or menstrual dysfunctions.12
Micronutrients of key interest: Antioxidants. Anti-
oxidant nutrients play important roles in protecting cell
membranes from oxidative damage. Because exercise can
increase oxygen consumption by 10- to 15-fold, it has been
hypothesized that chronic training contributes a constant
‘‘oxidative stress’’ on cells.101 Acute exercise is known to
increase levels of lipid peroxide by-products,101 but also re-
sults in a net increase in native antioxidant system functions
and reduced lipid peroxidation.102 Thus, a well-trained athlete
may have a more developed endogenous antioxidant system
than a less-active individual and may not benefit from anti-
oxidant supplementation, especially if consuming a diet high
in antioxidant rich foods. There is little evidence that anti-
oxidant supplements enhance athletic performance101 and the
interpretation of existing data is confounded by issues of
study design (eg, a large variability in subject characteristics,
training protocols, and the doses and combinations of anti-
oxidant supplements; the scarcity of crossover designs).
There is also some evidence that antioxidant supplementation
may negatively influence training adaptations.103
The safest and most effective strategy regarding micro-
nutrient antioxidants is to consume a well-chosen diet
containing antioxidant-rich foods. The importance of reac-
tive oxygen species in stimulating optimal adaptation to
training merits further investigation, but the current literature
does not support antioxidant supplementation as a means to
prevent exercise induced oxidative stress. If athletes decide
to pursue supplementation, they should be advised not to
exceed the Tolerable Upper Intake Levels since higher doses
could be pro-oxidative.101 Athletes at greatest risk for poor
antioxidant intakes are those who restrict energy intake,
follow a chronic low-fat diet, or limit dietary intake of fruits,
vegetables, and whole grains.46
In summary of the micronutrients, athletes should be edu-
cated that the intake of vitamin and mineral supplements does
not improve performance unless reversing a pre-existing
deficiency78,79and the literature to support micronutrient sup-
plementation is often marred with equivocal findings and weak
evidence. Despite this, many athletes unnecessarily consume
micronutrient supplements even when dietary intake meets
micronutrient needs. Rather than self-diagnosing the need for
micronutrient supplementation, when relevant, athletes should
seek clinical assessment of their micronutrient status within a
larger assessment of their overall dietary practices. Sports
dietitians can offer several strategies for assessing micro-
nutrient status based on collection of a nutrient intake his-
tory along with observing signs and symptoms associated
with micronutrient deficiency. This is particularly important
for iron, vitamin D, calcium, and antioxidants. By encour-
aging athletes to consume a well-chosen diet focused on
food variety, sports dietitians can help athletes avoid mi-
cronutrient deficiencies and gain the benefits of many other
performance-promoting eating strategies. Public health guide-
lines such as the DRIs provide micronutrient intake recom-
mendations for sports dietitians to help athletes avoid both
deficiency and safety concerns associated with excessive in-
take. Micronutrient intake from dietary sources and fortified
foods should be assessed alongside micronutrient intake from
all other dietary supplements.
THEME 2: PERFORMANCE NUTRITION:
STRATEGIES TO OPTIMIZE PERFORMANCE
AND RECOVERY FOR COMPETITION AND
KEY TRAINING SESSIONS
Pre-, During and Post-Event Eating
Strategies implemented in pre-, during, and post-exercise
periods must address a number of goals. First they should
support or promote optimal performance by addressing various
factors related to nutrition that can cause fatigue and deteriora-
tion in the outputs of performance (eg, power, strength, agility,
skill, and concentration) throughout or towards the end of the
sporting event. These factors include, but are not limited to,
dehydration, electrolyte imbalances, glycogen depletion, hy-
poglycemia, gastrointestinal discomfort/upset, and disturbances
http://www.acsm-msse.org554 Official Journal of the American College of Sports Medicine
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
to acid–base balance. Fluids or supplements consumed before,
during, or in the recovery between sessions can reduce or delay
the onset of these factors. Strategies include increasing or re-
placing key exercise fuels and providing substrates to return the
body to homeostasis or further adapt to the stress incurred
during a previous exercise session. In some cases, pre-event
nutrition may need to redress the effects of other activities un-
dertaken by the athlete during event preparation such as dehy-
dration or restrictive eating associated with ’’making weight’’ in
weight category sports. A secondary goal is to achieve gut
comfort throughout the event, avoiding feelings of hunger or
discomfort and gastrointestinal upsets that may directly reduce
the enjoyment and performance of exercise and interfere with
ongoing nutritional support. A final goal is to continue to
provide nutritional support for health and further adaptation to
exercise, particularly in the case of competitive events that
span days and weeks (eg, tournaments and stage races).
Nutrient needs and the practical strategies for meeting
them pre, during, and post exercise depend on a variety of
factors including the event (mode, intensity, duration of
exercise), the environment, carryover effects from previous
exercise, appetite, and individual responses and preferences.
In competitive situations, rules of the event and access to
nutritional support may also govern the opportunities for
food intake. It is beyond the scope of this review to provide
further discussion other than to comment that solutions to
feeding challenges around exercise require experimentation
and habituation by the athlete, and are often an area in which
the food knowledge, creativity, and practical experiences of
the sports dietitian make valuable contributions to the athlete_s
nutrition plan. Such scenarios are also where the use of sports
foods and supplements are often most valuable, since well-
formulated products can often provide a practical form of
nutritional support to meet specialized nutrient needs.
Hydration Guidelines: Fluid and
Electrolyte Balance
Being appropriately hydrated contributes to optimal health
and exercise performance. In addition to the usual daily water
losses from respiration, gastrointestinal, renal, and sweat
sources, athletes need to replace sweat losses. Sweating assists
with the dissipation of heat, generated as a byproduct of
muscular work but is often exacerbated by environmental
conditions, and thus helps maintain body temperature within
acceptable ranges.104 Dehydration refers to the process of
losing body water and leads to hypohydration. Although it is
common to interchange these terms, there are subtle differ-
ences since they reflect process and outcome.Through a cascade of events, the metabolic heat generated
by muscle contractions during exercise can eventually lead
to hypovolemia (decreased plasma/blood volume) and thus,
cardiovascular strain, increased glycogen utilization, altered
metabolic and CNS function, and a greater rise in body
temperature.104–106 Although it is possible to be hypohy-
drated but not hyperthermic (defined as core body temper-
ature exceeding 40-C; 104-F),107 in some scenarios the extra
thermal strain associated with hypohydration can contribute
to an increased risk of life-threatening exertional heat illness
(heatstroke). In addition to water, sweat contains substantial
but variable amounts of sodium, with lesser amounts of
potassium, calcium, and magnesium.104 To preserve ho-
meostasis, optimal body function, performance, and per-
ception of well-being, athletes should strive to undertake
strategies of fluid management before, during, and after
exercise that maintain euhydration. Depending on the ath-
lete, the type of exercise, and the environment, there are
situations when this goal is more or less important.
Although there is complexity and individuality in the
response to dehydration, fluid deficits of 92% body weight
can compromise cognitive function and aerobic exer-
cise performances, particularly in hot weather.104,105,108,109
Decrements in the performance of anaerobic or high-
intensity activities, sport-specific technical skills, and aero-
bic exercise in a cool environment are more commonly
seen when 3%–5% of BW is lost due to dehydration.104,105
Severe hypohydration with water deficits of 6%–10% BW
has more pronounced effects on exercise tolerance, de-
creases in cardiac output, sweat production, skin and muscle
blood flow.107
Assuming an athlete is in energy balance, daily hydration
status may be estimated by tracking early morning body
weight (measured upon waking and after voiding) since
acute changes in body weight generally reflect shifts in body
water. Urinary specific gravity and urine osmolality can also
be used to approximate hydration status by measuring the
concentration of the solutes in urine. When assessed from a
midstream collection of the first morning urine sample, a uri-
nary specific gravity of G 1.020, perhaps ranging to G 1.025 to
account for individual variability,106 is generally indicative
of euhydration. Urinary osmolality reflects hypohydration
when 9900 mOsmol/kg, while euhydration is considered as
G700 mOsmol/kg.104,106
Before exercise. Some athletes begin exercise in a
hypohydrated state, which may adversely affect athletic per-
formances.105,110 Purposeful dehydration to ‘‘make weight’’
may result in a significant fluid deficit, which may be difficult
to restore between ‘‘weigh-in’’ and start of competition.
Similarly, athletes may be hypohydrated at the onset of ex-
ercise due to recent, prolonged training sessions in the heat or
to multiple events in a day.104,105,108,110
Athletes may achieve euhydration prior to exercise by
consuming a fluid volume equivalent to 5–10 ml/kg BW
(~2–4 ml/lb) in the 2 to 4 hours before exercise to achieve
urine that is pale yellow in color while allowing for suffi-
cient time for excess fluid to be voided.104,108 Sodium con-
sumed in pre-exercise fluids and foods may help with fluid
retention. Although some athletes attempt to hyper-hydrate
prior to exercise in hot conditions where the rates of sweat
loss or restrictions on fluid intake inevitably lead to a sig-
nificant fluid deficit, the use of glycerol and other plasma
expanders for this purpose is now prohibited by the World
Anti-Doping Agency (www.wada-ama.org).
NUTRITION AND ATHLETIC PERFORMANCE Medicine & Science in Sports & Exercised 555
SPEC
IA
L
C
O
M
M
U
N
IC
ATIO
N
S
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
http://www.wada-ama.org
During exercise. Sweat rates vary during exercise from
0.3–2.4 L/h dependent on exercise intensity, duration, fitness,
heat acclimatization, altitude, and other environmental con-
ditions (heat, humidity, etc.).104,106,111,112 Ideally, athletes
should drink sufficient fluids during exercise to replace
sweat losses such that the total body fluid deficit is limited to
G2% BW. Various factors may impair the availability of fluid
or opportunities to consume it during exercise and for most
competitive, high caliber athletes, sweat loss generally ex-
ceeds fluid intake. However, individual differences are seen
in drinking behavior and sweat rates in sport, and result in a
range of changes in fluid status from substantial dehydration
to over-hydration.110
Routine measurement of pre- and post-exercise BW, ac-
counting for urinary losses and drink volume, can help the
athlete estimate sweat losses during sporting activities to
customize their fluid replacement strategies.104 In the ab-
sence of other factors that alter body mass during exercise
(eg, the significant loss of substrate which may occur during
very prolonged events), a loss of 1 kg BW represents ap-
proximately 1 L sweat loss. The fluid plan that suits most
athletes and athletic events will typically achieve an intake
of 0.4 to 0.8 L/h,104 although this needs to be customized to
the athlete_s tolerance and experience, their opportunities for
drinking fluids and the benefits of consuming other nutrients
(eg, carbohydrate) in drink form. Ingestion of cold beverages
(0.5 -C) may help reduce core temperature and thus improve
performance in the heat. The presence of flavor in a beverage
may increase palatability and voluntary fluid intake.
Although the typical outcome for competitive athletes is to
develop a fluid deficit over the course of an exercise session,
over the past 2 decades there has been an increasing awareness
that some recreational athletes drink at rates that exceed their
sweat losses and over-hydrate. Over-drinking fluids in ex-
cess of sweat and urinary losses is the primary cause of
hyponatremia (blood sodium G135 mmol/L), also known as
water intoxication, although this can be exacerbated in cases
where there are excessive losses of sodium in sweat and fluid
replacement involving low-sodium beverages.113,114 It can
also be compounded by excessive fluid intake in the hours
or days leading up to the event. Over-hydration is typically
seen in recreational athletes since their work outputs and
sweat rates are lower than competitive athletes, while their
opportunities and belief in the need to drink may be greater.
Women generally have a smaller body size and lower sweat
rates than males and appear to be at greater risk of over-
drinking and possible hyponatremia.104 Symptoms of hypo-
natremia during exercise occur particularly when plasma
sodium levels fall below 130 mmol/L and include bloating,
puffiness, weight gain, nausea, vomiting, headache, confu-
sion, delirium, seizures, respiratory distress, loss of con-
sciousness, and possibly death if untreated. While the
prevalence of hypohydration and hypernatremia is thought
to be greater than reports of hyperhydration and hypo-
natremia, the latter are more dangerous and require prompt
medical attention.104,106,114
Sodium should be ingested during exercise when large
sweat sodium losses occur. Scenarios include athletes with
high sweat rates (91.2 L/h), ‘‘salty sweat,’’ or prolonged
exercise exceeding 2 hours in duration.105,106,109Although
highly variable, the average concentration of sodium in
sweat approximates 50 mmol/L (~1 g/L) and is hypotonic in
comparison to blood sodium content. Thirst sensation is
often dictated by changes in plasma osmolality and is usu-
ally a good indication of the need to drink but not that the
athlete is dehydrated.108 Older athletes may present with
age-related decreases in thirst sensation and may need en-
couragement to drink during and post-exercise.104
Although skeletal muscle cramps are typically caused by mus-
cle fatigue, they can occur with athletes from all types of sports in
a range of environmental conditions104and may be associated
with hypohydration and electrolyte imbalances. Athletes who
sweat profusely, especially when overlaid with a high sweat
sodium concentration, may be at greater risk for cramping, par-
ticularly when not acclimatized to the heat and environment.115
After exercise. Most athletes finish exercising with a
fluid deficit and may need to restore euhydration during the
recovery period.104,110 Rehydration strategies should primar-
ily involve the consumption of water and sodium at a modest
rate that minimizes diuresis/urinary losses.105 The presence of
dietary sodium/sodium chloride (from foods or fluids) helps
to retain ingested fluids, especially extracellular fluids, in-
cluding plasma volume. Therefore, athletes should not be
advised to restrict sodium in their post-exercise nutrition
particularly when large sodium losses have been incurred.
Since sweat losses and obligatory urine losses continue dur-
ing the post-exercise phase, effective rehydration requires the
intake of a greater volume of fluid (eg, 125%–150%) than the
final fluid deficit (eg, 1.25–1.5 L fluid for every 1 kg BW
lost).104,106 Excessive intake of alcohol in the recovery period
is discouraged due to its diuretic effects. However, the previ-
ous warnings about caffeine as a diuretic appear to be
overstated when it is habitually consumed in moderate (e.g.
G 180 mg) amounts.104
Carbohydrate Intake Guidelines
Because of its role as an important fuel for the muscle and
central nervous system, the availability of carbohydrate stores
is limiting for the performance of prolonged continuous or
intermittent exercise, and is permissive for the performance of
sustained high-intensity sport. The depletion of muscle gly-
cogen is associated with fatigue and a reduction in the inten-
sity of sustained exercise, while inadequate carbohydrate for
the central nervous system impairs performance-influencing
factors such as pacing, perceptions of fatigue, motor skill, and
concentration.3,116 As such, a key strategy in promoting op-
timal performance in competitive events or key workouts is
matching of body carbohydrate stores with the fuel demands
of the session. Strategies to promote carbohydrate availability
should be undertaken before, during, or in the recovery be-
tween events or high-quality training sessions.
http://www.acsm-msse.org556 Official Journal of the American College of Sports Medicine
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Achieving adequate muscle glycogen stores. Ma-
nipulating nutrition and exercise in the hours and days prior to
an important exercise bout allows an athlete to commence the
session with glycogen stores that are commensurate with the
estimated fuel costs of the event. In the absence of severe
muscle damage, glycogen stores can be normalised with 24 h
of reduced training and adequate fuel intake 117 (Table 2).
Events 990 minutes in duration may benefit from higher
glycogen stores,118 which can be achieved by a technique
known as carbohydrate loading. This protocol of achieving
supercompensation of muscle glycogen evolved from the
original studies of glycogen storage in the 1960s and, at least
in the case of trained athletes, can be achieved by extending
the period of a carbohydrate-rich diet and tapering training
over 48 h36 (Table 2).
Carbohydrate consumed in meals and/or snacks during
the 1–4 hours pre-exercise may continue to increase body
glycogen stores, particularly liver glycogen levels that have
been depleted by the overnight fast.117 It may also provide a
source of gut glucose release during exercise.117 Carbohydrate
intakes of 1–4 g/kg, with timing, amount, and food choices
suited to the individual, have been shown to enhance endur-
ance or performance of prolonged exercise (Table 2).117,119 Gen-
erally, foods with a low-fat, low-fiber, and low–moderate
protein content are the preferred choice for this pre-event
menu since they are less prone to cause gastrointestinal prob-
lems and promote gastric emptying.120 Liquid meal supple-
ments are useful for athletes who suffer from pre-event
nerves or an uncertain pre-event timetable and thus prefer a
more quickly digested option. Above all, the individual ath-
lete should choose a strategy that suits their situation and
their past experiences and can be fine-tuned with further
experimentation.
The intake of carbohydrate prior to exercise is not always
straightforward since the metabolic effects of the resulting
insulin response include a reduction in fat mobilization and
utilization and concomitant increase in carbohydrate utili-
zation.119 In some individuals, this can cause premature fa-
tigue.121 Strategies to circumvent this problem include
ensuring at least 1 g/kg carbohydrate in the pre-event meal
to compensate for the increased carbohydrate oxidation, in-
cluding a protein source at the meal, including some high-
intensity efforts in the pre-exercise warm up to stimulate
hepatic gluconeogenesis, and consuming carbohydrate dur-
ing the exercise.122 Another approach has been suggested in
the form of choosing pre-exercise meals from carbohydrate-
rich foods with a low glycemic index, which might reduce the
metabolic changes associated with carbohydrate ingestion as
well as providing a more sustained carbohydrate release dur-
ing exercise. Although occasional studies have shown that
such a strategy enhances subsequent exercise capacity,123 as
summarized by the EAL (Table 1 Question #11) and others,119
pre-exercise intake of low glycemic index carbohydrate
choices has not been found to provide a universal benefit to
performance even when the metabolic perturbations of pre-
exercise carbohydrate intake are attenuated. Furthermore,
consumption of carbohydrate during exercise, as further ad-
vised in Table 2, dampens any effects of pre-exercise carbo-
hydrate intake on metabolism and performance.124
Depending on characteristics including the type of exer-
cise, the environment, and the athlete_s preparation and
carbohydrate tolerance, the intake of carbohydrate during
exercise provides a number of benefits to exercise capacity
and performance via mechanisms including glycogen spar-
ing, provision of an exogenous muscle substrate, prevention
of hypoglycemia, and activation of reward centers in the
central nervous system.116 Robust literature on exercise
carbohydrate feeding has led to the recognition that different
amounts, timing and types of carbohydrate are needed to
achieve these different effects, and that the different effects
may overlap in various events.36,125 Table 2 summarizes the
current guidelines for exercise fueling, noting opportunities
where it may play a metabolic role (events of 960–90 min)
and the newer concept of ‘‘mouth sensing’’ where frequent
exposure of the mouth and oral cavity to carbohydrate is
likely to be effective in enhancing workout and pacing strat-
egies via a CNS effect.126 Of course, the practical achieve-
ment of these guidelines needs to fit the personal preferences
and experiences of the individual athlete, and the practical
opportunities provided in an event or workout to obtain and
consume carbohydrate-containing fluids or foods. A range of
everyday foods and fluids and formulated sports products that
include sports beverages may be chosen to meet these
guidelines; this includes newer products containing mixtures
of glucose and fructose (the so-called ‘‘multiple transporta-
ble carbohydrates’’), which aim to increase total intestinal
absorption of carbohydrates.127 Although this could be of use
to situations of prolonged exercise where higher rates of ex-
ogenous carbohydrate oxidation might sustain work intensity
in the face of dwindling muscle glycogen stores, the EAL
found that evidence for benefits is currently equivocal (Table 1,
Question #9).
Glycogen restoration is one of the goals of post-exercise re-
covery, particularly between bouts of carbohydrate-dependent
exercise where there is a priority on performancein the second
session. Refueling requires adequate carbohydrate intake
(Table 2) and time. Since the rate of glycogen resynthesis is
only ~ 5% per hour, early intake of carbohydrate in the recovery
period (~1–1.2 g/kg/h during the first 4–6 hours) is useful in
maximizing the effective refueling time.117 As long as total
intake of carbohydrate and energy is adequate and overall
nutritional goals are met, meals and snacks can be chosen from
a variety of foods and fluids according to personal preferences
of type and timing of intake.36,117 More research is needed to
investigate how glycogen storage might be enhanced when
energy and carbohydrate intakes are sub-optimal.
Protein intake guidelines. Protein consumption in the
immediate pre- and post-exercise period is often intertwined
with carbohydrate consumption as most athletes consume
foods, beverages, and supplements that contain both macro-
nutrients. Dietary protein consumed in scenarios of low-
carbohydrate availability128 and/or restricted energy intake53
NUTRITION AND ATHLETIC PERFORMANCE Medicine & Science in Sports & Exercised 557
SPEC
IA
L
C
O
M
M
U
N
IC
ATIO
N
S
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
in the early post-exercise recovery period has been found to
enhance and accelerate glycogen repletion. For example, it
has been established that recovery of performance129 and
glycogen repletion rates53 were similar in athletes consuming
0.8 g carbohydrate/kg/BW + 0.4 g protein/kg/BW compared to
athletes consuming only carbohydrate (1.2 g/kg/BW). This may
support exercise performance and benefit athletes frequently
involved in multiple training or competitive sessions over same
or successive days.
Although protein intake may support glycogen resynthesis
and, when consumed in close proximity to strength and en-
durance exercise, enhancesMPS,59,130 there is a lack of evidence
from well-controlled studies that protein supplementation di-
rectly improves athletic performance.131,132 However, a modest
number of studies have reported that ingesting ~50–100 g of
protein during the recovery period leads to accelerated recovery
of static force and dynamic power production during delayed
onset muscle soreness.133,134 Despite these findings other studies
show no performance effects from acute ingestion of protein at
intake levels that are much more practical to consume on a
regular basis. Furthermore, studies that imply positive findings
when the control group receives a flavored water placebo133 or
a placebo that is not isocaloric are unable to rule out the impact
of post-exercise energy provision on the observed effect.134
Protein ingestion during exercise and during the pre-
exercise period seems to have less of an impact on MPS than
the post-exercise provision of protein but may still enhance
muscle reconditioning depending on the type of training that
takes place. Co-ingestion of protein and carbohydrate during
2 hours of intermittent resistance-type exercise has been
shown to stimulate MPS during the exercise period135 and
may extend the metabolic adaption window particularly
during ultra–endurance-type exercise bouts.136 Potential
benefits of consuming protein before and during exercise
may be targeted to athletes focused on the MPS response to
resistance exercise and those looking to enhanced recovery
from ultra-endurance exercise.
Table 1, EAL Questions 5–7 summarizes the literature on
consuming protein alone or in combination with carbohydrate
during recovery on several outcomes. More work is needed to
elucidate the relevance and practicality of protein consump-
tion on subsequent exercise performance and if mechanisms
in this context are exclusive to accelerating muscle glycogen
synthesis. The utility of a protein supplement should also be
measured against the benefits of consuming protein or amino
acids from meals and snacks that are already part of a sports
nutrition plan to meet other performance goals.
Dietary supplements and ergogenic aids. Exter-
nal and internal motives to enhance performance often
encourage athletes to consider the enticing marketing
and testimonials surrounding supplements and sports foods.
Sports supplements represent an ever growing industry, but a
lack of regulation of manufacture and marketing means that
athletes can fall victim to false advertising and unsubstantiated
claims.137 The prevalence of supplementation among athletes
has been estimated internationally at 37%–89%, with greater
frequencies being reported among elite and older athletes.
Motivations for use include enhancement of performance or
recovery, improvement or maintenance of health, an increase
in energy, compensation for poor nutrition, immune support,
and manipulation of body composition,138,139 yet few athletes
undertake professional assessment of their baseline nutritional
habits. Furthermore, athletes_ supplementation practices are
often guided by family, friends, teammates, coaches, the in-
ternet, and retailers, rather than sports dietitians and other
sport science professionals.138
Considerations regarding the use of sports foods and
supplements include an assessment of efficacy and potency.
In addition, there are safety concerns due to the presence of
overt and hidden ingredients that are toxic and the poor
practices of athletes in consuming inappropriately large
doses or problematic combinations of products. The issue of
compliance to anti-doping codes remains a concern with
potential contamination with banned or non-permissible
substances. This carries significant implications for athletes
who compete under anti-doping codes (eg, National Colle-
giate Athletic Association, World Anti-Doping Agency).139
A supplement manufacturer_s claim of ‘‘100% pure,’’ ‘‘phar-
maceutical grade,’’ ‘‘free of banned substances,’’ ‘‘Natural
Health Product – NHPN/NPN’’ (in Canada) or possessing a
drug identification number are not reliable indications that
guarantee a supplement is free of banned substances. How-
ever, commercial, third-party auditing programs can indepen-
dently screen dietary supplements for banned and restricted
substances in testing facilities (ISO 17025 accreditation stan-
dard)140 thereby providing a greater assurance of supplement
purity for athletes concerned about avoiding doping violations
and eligibility.
The ethical use of sports supplements is a personal choice
and remains controversial. It is the role of qualified health
professionals, such as a sports dietitian, to build rapport with
athletes and provide credible, evidence-based information
regarding the appropriateness, efficacy and dosage for the
use of sports foods and supplements. After completing a
thorough assessment of an athlete_s nutritional practices
and dietary intake, sports dietitians should assist the athlete
to determine a cost to benefit analysis of their use of a
product, noting that the athlete is responsible for products
ingested and any subsequent consequences (ie, legal, health,
safety issues).139
The benefits of the use of supplements and sports foods
include practical assistance to meet sports nutritional goals,
prevention or treatment of nutrient deficiencies, a placebo
effect, and in some cases, a direct ergogenic effect. How-
ever, this must be carefully balanced against risks, and the
expense and potential for ergolytic effects.139,141 Factors to
consider in the analysis include a theoretical analysis of the
nutritional goal or performance benefit that the product is to
address within the athlete_s specific training or competition
program, the quality of the evidence that the product can
address these goals, previous experience regarding individ-
ual responsiveness, and the health and legal consequences.
http://www.acsm-msse.org558 Official Journal of the American College of Sports Medicine
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Relatively few supplements that claim ergogenicbenefits
are supported by sound evidence.139,141 Research method-
ologies on the efficacy of sports supplements are often lim-
ited by small sample sizes, enrollment of untrained subjects,
poor representation of athlete sub-populations (females,
older athletes, athletes with disabilities, etc.), performance
tests that are unreliable or irrelevant, poor control of
confounding variables, and failure to include recommended
sports nutrition practices or the interaction with other sup-
plements.139,141 Even when there is a robust literature on a
sports supplement, it may not cover all applications that are
specific to an event, environment, or individual athlete.
Supplement use is best undertaken as an adjunct to a well-
chosen nutrition plan. It is rarely effective outside these
conditions and not justified in the case of young athletes
who can make significant performance gains via maturation
in age, sports experience, and the development of a sports
nutrition plan.
It is beyond the scope of this paper to address the multi-
tude of sports supplements used by athletes and caveats
surrounding sport-specific rules allowing their use. The
Australian Institute of Sport has developed a classification
system that ranks sports foods and supplement ingredients
based on significance of scientific evidence and whether a
product is safe, legal, and effective in improving sports
performance.142 Table 3 serves as a general guide to de-
scribe the ergogenic and physiological effects of potentially
beneficial supplements and sport foods.141,143–148 This guide
is not meant to advocate specific supplement use by athletes
and should only be considered in well-defined situations.
THEME 3: SPECIAL POPULATIONS
AND ENVIRONMENTS
Vegetarian Athlete
Athletes may opt for a vegetarian diet for various reasons
from ethnic, religious, and philosophical beliefs to health,
food aversions, and financial constraints or to disguise dis-
ordered eating. As with any self-induced dietary restriction,
it would be prudent to explore whether the vegetarian athlete
also presents with disordered eating or a frank eating disor-
der.13,14 A vegetarian diet can be nutritionally adequate
containing high intakes of fruits, vegetables, whole grains,
nuts, soy products, fiber, phytochemicals, and antioxi-
dants.149 Currently, research is lacking regarding the impact
on athletic performance from long-term vegetarianism
among athletic populations.150
Depending on the extent of dietary limitations, nutrient
concerns for vegetarianism may include energy, protein, fat,
iron, zinc, vitamin B-12,, calcium, n-3 fatty acids,149 and
low intakes of creatine and carnosine.151 Vegetarian athletes
may have an increased risk of lower bone mineral density
and stress fractures.152 Additional practical challenges in-
clude gaining access to suitable foods during travel, restau-
rant dining, and at training camps and competition venues.
Vegetarian athletes may benefit from comprehensive dietary
assessments and education to ensure their diets are nutri-
tionally sound to support training and competition demands.
Altitude
Altitude exposure (ie, daily or intermittent exposure to
92,000 m, 9 6,600 ft) may be a specialized strategy within
an athlete_s training program or simply their daily training
environment.153 One of the goals of specialized altitude
training blocks is to naturally increase red blood cell mass
(erythropoiesis) so that greater amounts of oxygen can be
carried in the blood to enhance subsequent athletic perfor-
mances.112 Initial exposure to altitude leads to a decrease in
plasma volume with corresponding increases in hemoglobin
concentration. Over time there is a net increase in red cell
mass and blood volume therefore greater oxygen carrying
capacity.154 However, possessing sufficient iron stores prior
to altitude training is essential to enable hematological ad-
aptations.154 Consumption of iron-rich foods with or without
iron supplementation may be required by athletes before and
during altitude exposure.
Specific or chronic exposure to a high altitude environment
may increase the risk of illness, infection, and sub-optimal
adaptation to exercise due to direct effects of hypobaric hyp-
oxic conditions, an unaccustomed volume and intensity of
training, interrupted sleep, and increased UV light exposure.155
The effects are greater with higher elevation and require more
acclimatization to minimize the risk of specific altitude illness.
Adequate nutrition is essential to maximize the desired effect
from altitude training or to support more chronic exposure to a
high altitude environment. Key nutritional concerns include
the adequacy of intake of energy, carbohydrate, protein, fluids,
iron, and antioxidant-rich foods.112 An increased risk of de-
hydration at altitude is associated with initial diuresis, in-
creased ventilation, and low humidity, and exercise sweat
losses. Some experts suggest daily fluid needs as high as 4–5 L
with altitude training and competition, while others encourage
individual monitoring of hydration status to determine fluid
requirements at altitude.112
Extreme Environments
Extreme environmental challenges (heat, cold, humidity,
altitude) require physiological, behavioral, and technological
adaptations to ensure athletes are capable of performing at
their best. Changes in environmental conditions stimulate
thermoregulatory neuronal activity in the brain to increase
heat loss (sweating and skin vasodilation), prevent heat loss
(skin vasoconstriction), or induce heat gain (shivering).
Sympathetic neural activation triggers changes in skin blood
flow to vary convective heat transfer from the core to the
skin (or vice versa) as required for maintaining an optimal
core temperature. Unique considerations of nutrition-
related concerns are presented when exercising in hot or
cold environments.107,155,156
NUTRITION AND ATHLETIC PERFORMANCE Medicine & Science in Sports & Exercised 559
SPEC
IA
L
C
O
M
M
U
N
IC
ATIO
N
S
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Hot environments. When ambient temperature exceeds
body temperature, heat cannot be dissipated by radiation;
furthermore, the potential to dissipate heat by evaporation of
sweat is substantially reduced when the relative humidity is
high.107,156 Heat illness from extreme heat exposure can re-
sult in appetite changes and serious health implications (ie,
heat exhaustion and exertional heat stroke). Heat exhaus-
tion is characterized by the inability to sustain cardiac out-
put related to exercise-heat stress causing elevated skin
temperatures with or without hyperthermia (938.5-C).
Symptoms of heat exhaustion can include anxiety, dizziness,
fainting. Exertional heat stroke (body core hyperthermia,
typically 940-C) is the most serious and leads to multi-organ
dysfunction, including brain swelling, with symptoms of central
nervous system abnormalities, delirium, and convulsions, thus
can be life-threatening.107,156
Athletes competing in lengthy events conducted in hot
conditions (eg, tennis match or marathon) and those forced
TABLE 3. Dietary supplements and sports foods with evidence-based uses in sports nutrition.
Category Examples Use Concerns Evidence
Sports food Sports drinks Practical choice to meet sports
nutritional goals especially when
access to food, opportunities to
consume nutrients or gastrointestinal
concerns make it difficult to consume
traditional food and beverages
Cost is greater than whole foods Burke
Sports bars May be used unnecessarily or in
inappropriate protocols
(2015)141
Sports confectionery
Sports gels
Electrolyte supplements
Protein supplements
Liquid meal supplements
Medical supplements Iron supplements Prevention or treatment of nutrient
deficiency under the supervision of
appropriate medical/nutritional expert
May be self-prescribed unnecessarily
without appropriate supervision
or monitoring
Burke (2015)141
Calcium supplements
Vitamin D supplementsMulti-vitamin/mineral
n-3 fatty acids
Specific performance
supplements Ergogenic effects
Physiological effects/mechanism
of ergogenic effect Concerns regarding usea Evidence
Creatine Improves performance of
repeated bouts of
high-intensity exercise with
short recovery periods
Increases Creatine and Phosphocreatine
concentrations
Associated with acute weight gain
(0.6–1 kg) which may be problematic
in weight sensitive sports
Tarnopolsky
(2010)143
- Direct effect on competition
performance
May also have other effects such as
enhancement of glycogen storage and
direct effect on muscle protein synthesis
May cause gastrointestinal discomfort
- Enhanced capacity for training
Some products may not contain
appropriate amounts or forms
of creatine
Caffeine Reduces perception of fatigue Adenosine antagonist with effects on
many body targets including central
nervous system
Causes side-effects (tremor, anxiety,
increased heart rate, etc.) when
consumed in high doses
Astorino (2010)144
Allows exercise to be sustained
at optimal intensity/output
for longer Promotes Ca2+ release from sarcoplasmic
reticulum
Toxic when consumed in very
large doses
Tarnopolsky
(2010)143
Rules of National Collegiate Athletic
Association competition prohibit
the intake of large doses that
produce urinary caffeine levels
exceeding 15 ug/ml
Burke (2013)145
Some products do not disclose caffeine
dose or may contain other stimulants
Sodium bicarbonate Improves performance of events
that would otherwise be
limited by acid–base disturbances
associated with high rates of
anaerobic glycolysis
When taken as an acute dose pre-exercise,
increases extracellular buffering capacity
May cause gastrointestinal side-effects
which cause performance impairment
rather than benefit
Carr (2011)146
- High intensity events of
1–7 minutes
- Repeated high-intensity sprints
- Capacity for high-intensity ‘‘sprint’’
during endurance exercise
Beta-alanine Improves performance of events
that would otherwise be
limited by acid–base disturbances
associated with high rates of
anaerobic glycolysis
When taken in a chronic protocol,
achieves increase in muscle carnosine
(intracellular buffer)
Some products with rapid absorption
may cause paresthesia (tingling sensation)
Quesnele
(2014)147
- Mostly targeted at high-intensity
exercise lasting 60–240 seconds
- May enhance training capacity
Nitrate Improves exercise tolerance and
economy
Increases plasma nitrite concentrations to
increase production of nitric oxide with
various vascular and metabolic effects
that reduces O2 cost of exercise
Consumption in concentrated food sources
(eg, beetroot juice) may cause gut
discomfort and discoloration of urine
Jones (2014)148
Improves performance in endurance
exercise at least in non-elite athletes Efficacy seems less clear cut in high
caliber athletes
These supplements may perform as claimed but does not imply endorsement by this position stand. a Athletes should be assisted to undertake a cost to benefit analysis (141) before
using any sports food and supplements with consideration of potential nutritional, physiological, and psychological benefits for their specific event weighed against potential disad-
vantages. Specific protocols of use should be tailored to the individual scenario (see references for further information) and specific products should be chosen with consideration of the
risk of contamination with unsafe or illegal chemicals.
http://www.acsm-msse.org560 Official Journal of the American College of Sports Medicine
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
to wear excessive clothing (eg, American football players or
BMX competitors) are at greatest risk of heat illness.111
Strategies to reduce high skin temperatures and large sweat
(fluid and electrolyte) losses are required to minimize car-
diovascular and hyperthermic challenges that may impair
athletic performance when exercising in the heat; athletes
should be regularly monitored when at risk for heat-related
illness.107,156 Specific strategies should include: acclimati-
zation, individualized hydration plans, regular monitoring of
hydration status, beginning exercise euhydrated, consuming
cold fluids during exercise, and possibly the inclusion of
electrolyte sources.107,156
Cold environments. Athletic performance in cold en-
vironments may present several dietary challenges that re-
quire careful planning for optimal nutritional support. A
large number of sports train and compete in the cold ranging
from endurance athletes (eg, Nordic skiers) through to judged
events (eg, free style ski). Furthermore, drastic, unexpected
environmental changes can turn a warm-weather event (eg,
cross country mountain bike race or triathlon) into extreme
cold conditions in a short period of time leaving unprepared
athletes confronted with performing in the cold.
Primary concerns of exercising in a cold environment
are maintenance of euhydration and body temperature.156
However, exercise-induced heat production and appropriate
clothing are generally sufficient to minimize heat loss.155,156
When adequately prepared (eg, removing wet clothing,
keeping muscles warm after exercise warm-up) athletes
can tolerate severe cold in pursuit of athletic success.
Smaller, leaner athletes are at greater risk of hypothermia
due to increased heat production required to maintain core
temperature and decreased insulation from lower body fat.
Metabolically, energy requirements (from carbohydrates)
are increased, especially when shivering, to maintain core
temperature.155,156
Several factors can increase the risk of hypohydration
when exercising in the cold, such as: cold-induced diuresis,
impaired thirst sensation, reduced desire to drink, limited
access to fluids, self-restricted fluid intake to minimize uri-
nation, sweat losses from over-dressing and increased res-
piration with high altitude exposure.
In the cold, hypohydration of 2%–3% BW loss is less
detrimental to endurance performances than similar losses
occurring in the heat.104,155,156 Severe cold exposure may be
problematic on training versus competition days since
training duration may exceed competition duration and of-
ficials may delay competitions in inclement weather yet
athletes may continue to train in similar conditions. Athletes_
energy, macronutrient, and fluid intakes should be regularly
assessed and changes in body weight and hydration status
when exercising in both hot and cold environments. Edu-
cating athletes about modifying their energy, carbohydrate
intakes, and recovery strategies according to training and
competition demands promotes optimal training adaptation
and maintenance of health. Practical advice for preparation
and selection of appropriate foods and fluids that withstand
cold exposure will ensure athletes are equipped to cope with
weather extremes.
THEME 4: ROLES AND RESPONSIBILITIES OF
THE SPORTS DIETITIAN
Sport nutrition practice requires combined knowledge in
several topics: clinical nutrition, nutrition science, exercise
physiology, and application of evidence-based research. In-
creasingly, athletes and active individuals seek professionals
to guide them in making optimal food and fluid choices to
support and enhance their physical performances. An expe-
rienced sports dietitian demonstrates the knowledge, skills,
and expertise necessary to help athletes and teams work to-
wards their performance-related goals.
The Commission on Dietetic Registration (the credentialing
agency for the Academy of Nutrition and Dietetics) has cre-
ated a unique credential for registered dietitian nutritionists
who specialize in sports dietetic practice with extensive ex-
perience working with athletes. The Board Certified Specialist
in Sports Dietetics (CSSD) credential is designed as the pre-
mier professional sports nutrition credential in the United
States and is available internationally, including Canada.
Specialists in sports dieteticsprovide safe, effective, evidence-
based nutrition assessments, guidance, and counseling for
health and performance for athletes, sport organizations, and
physically active individuals and groups. For CSSD certifi-
cation details refer to the Commission on Dietetic registration:
www.cdrnet.org. Enhancement of sport nutrition knowledge
and continuing education can also be achieved by completing
recognized post-graduate qualifications such as the 2-year
distance learning diploma in sports nutrition offered by the
International Olympic Committee. For more information refer
to Sports Oracle: www.sportsoracle.com/Nutrition/Home/.
The Academy of Nutrition and Dietetics157 describes the
competencies of the sports dietitian to ‘‘provide medical
nutrition therapy in direct care and design, implement, and
manage safe and effective nutrition strategies that enhance
lifelong health, fitness, and optimal physical performance.’’
Roles and responsibilities of sports dietitians working with
athletes are outlined in Table 4.
SUMMARY
The following summarizes the evidence presented in this
position paper:
Í Athletes need to consume energy that is adequate in amount
and timing of intake during periods of high-intensity and/or
long duration training to maintain health and maximize
training outcomes. Low energy availability can result in
unwanted loss of muscle mass; menstrual dysfunction and
hormonal disturbances; sub-optimal bone density; an in-
creased risk of fatigue, injury, and illness; impaired adapta-
tion and a prolonged recovery process.
Í The primary goal of the training diet is to provide nutri-
tional support to allow the athlete to stay healthy and
NUTRITION AND ATHLETIC PERFORMANCE Medicine & Science in Sports & Exercised 561
SPEC
IA
L
C
O
M
M
U
N
IC
ATIO
N
S
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
http://www.cdrnet.org
http://www.sportsoracle.com/Nutrition/Home/
injury-free while maximizing the functional and metabolic
adaptations to a periodized exercise program that prepares
him or her to better achieve the performance demands of
their event. While some nutrition strategies allow the ath-
lete to train hard and recover quickly, others may target an
enhanced training stimulus or adaptation.
Í The optimal physique, including body size, shape and com-
position (eg, muscle mass and body fat levels), depends upon
the sex, age, and heredity of the athlete, andmay be sport- and
event-specific. Physique assessment techniques have inherent
limitations of reliability and validity, but with standardized
measurement protocols and careful interpretation of results,
they may provide useful information. Where significant ma-
nipulation of body composition is required, it should ideally
take place well before the competitive season to minimize
the impact on event performance or reliance on rapid weight
loss techniques.
Í Body carbohydrate stores provide an important fuel source
for the brain andmuscle during exercise, and are manipulated
by exercise and dietary intake. Recommendations for car-
bohydrate intake typically range from 3–10 g/kg BW/d (and
up to 12 g/kg BW/d for extreme and prolonged activities),
depending on the fuel demands of training or competition,
the balance between performance and training adaptation
goals, the athlete_s total energy requirements and body
composition goals. Targets should be individualized to the
athlete and his or her event, and also periodized over the
week, and training cycles of the seasonal calendar according
to changes in exercise volume and the importance of high
carbohydrate availability for different exercise sessions.
Í Recommendations for protein intake typically range from
1.2–2.0 g/kg BW/d, but have more recently been ex-
pressed in terms of the regular spacing of intakes of
modest amounts of high quality protein (0.3 g/kg body
weight) after exercise and throughout the day. Such in-
takes can generally be met from food sources. Adequate
energy is needed to optimize protein metabolism, and
when energy availability is reduced (eg, to reduce body
weight/fat), higher protein intakes are needed to support
MPS and retention of fat-free mass.
Í For most athletes, fat intakes associated with eating styles
that accommodate dietary goals typically range from 20%–
35% of total energy intake. Consuming e20% of energy
intake from fat does not benefit performance and extreme
restriction of fat intake may limit the food range needed to
meet overall health and performance goals. Claims that
extremely high-fat, carbohydrate-restricted diets provide a
benefit to the performance of competitive athletes are not
supported by current literature.
Í Athletes should consume diets that provide at least the
Recommended Dietary Allowance (RDA)/Adequate Intake
(AI) for all micronutrients. Athletes who restrict energy
intake or use severe weight-loss practices, eliminate com-
plete food groups from their diet, or follow other extreme
dietary philosophies are at greatest risk of micronutrient
deficiencies.
Í A primary goal of competition nutrition is to address
nutrition-related factors that may limit performance by
causing fatigue and a deterioration in skill or concentration
over the course of the event. For example, in events that are
dependent on muscle carbohydrate availability, meals eaten in
the day(s) leading up to an event should provide sufficient
carbohydrate to achieve glycogen stores that are commensu-
rate with the fuel needs of the event. Exercise taper and a
carbohydrate-rich diet (7–12 g/kg BW/d) can normalize mus-
cle glycogen levels within ~ 24 hours, while extending this to
48 hours can achieve glycogen super-compensation.
Í Foods and fluids consumed in the 1–4 hours prior to an
event should contribute to body carbohydrate stores
(particularly, in the case of early morning events to re-
store liver glycogen after the overnight fast), ensure ap-
propriate hydration status and maintain gastrointestinal
comfort throughout the event. The type, timing and amount
of foods and fluids included in this pre-event meal and/or
snack should be well trialed and individualized according to
the preferences, tolerance, and experiences of each athlete.
Í Dehydration/hypohydration can increase the perception of
effort and impair exercise performance; thus, appropriate
fluid intake before, during, and after exercise is important
for health and optimal performance. The goal of drinking
during exercise is to address sweat losses which occur to
assist thermoregulation. Individualized fluid plans should
be developed to use the opportunities to drink during a
workout or competitive event to replace as much of the
sweat loss as is practical; neither drinking in excess of
sweat rate nor allowing dehydration to reach problematic
levels. After exercise, the athlete should restore fluid bal-
ance by drinking a volume of fluid that is equivalent to
~125–150% of the remaining fluid deficit (eg, 1.25–1.5 L
fluid for every 1 kg BW lost).
Í An additional nutritional strategy for events of greater than
60 minutes duration is to consume carbohydrate according
to its potential to enhance performance. These benefits are
achieved via a variety of mechanisms which may occur
independently or simultaneously and are generally divided
into metabolic (providing fuel to the muscle) and central
(supporting the central nervous system). Typically, an intake
of 30–60 g/hour provides benefits by contributing to muscle
fuel needs and maintaining blood glucose concentrations,
although in very prolonged events (2.5+ h) or other sce-
narios where endogenous carbohydrate stores are substan-
tially depleted, higher intakes (up to 90 g/h) are associated
with better performance. Even in sustained high-intensity
events of 45–75 min where there is little need for carbo-
hydrate intake to play a metabolic role, frequent exposure
of the mouth and oral cavity to small amounts of carbo-
hydrate can still enhance performance via stimulation of
thebrain and central nervous system.
Í Rapid restoration of performance between physiologically
demanding training sessions or competitive events requires
appropriate intake of fluids, electrolytes, energy, and car-
bohydrates to promote rehydration and restore muscle
http://www.acsm-msse.org562 Official Journal of the American College of Sports Medicine
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
glycogen. A carbohydrate intake of ~1.0–1.2 g/kg/h, com-
mencing during the early recovery phase and continuing for
4 to 6 hours, will optimize rates of resynthesis of muscle
glycogen. The available evidence suggests that the early
intake of high quality protein sources (0.25–0.3 g/kg BW)
will provide amino acids to build and repair muscle tissue
and may enhance glycogen storage in situations where car-
bohydrate intake is sub-optimal.
Í In general, vitamin and mineral supplements are unneces-
sary for the athlete who consumes a diet providing high-
energy availability from a variety of nutrient-dense foods. A
multivitamin/mineral supplement may be appropriate in some
cases when these conditions do not exist; for example, if an
athlete is following an energy-restricted diet or is unwilling
or unable to consume sufficient dietary variety. Supple-
mentation recommendations should be individualized, re-
alizing that targeted supplementation may be indicated to
treat or prevent deficiency (eg, iron, vitamin D, etc.).
Í Athletes should be counseled regarding the appropriate use
of sports foods and nutritional ergogenic aids. Such products
should only be used after careful evaluation for safety, effi-
cacy, potency and compliance with relevant anti-doping
codes and legal requirements.
Í Vegetarian athletes may be at risk for low intakes of en-
ergy, protein, fat, creatine, carnosine, n-3 fatty acids, and
key micronutrients such as iron, calcium, riboflavin, zinc,
and vitamin B-12.
This Academy of Nutrition and Dietetics, Dietitians of
Canada (DC), and American College of Sports Medicine
(ACSM) position statement was adopted by the Academy
House of Delegates Leadership Team on July 12, 2000 and
reaffirmed on May 25, 2004 and February 15, 2011; ap-
proved by DC on November 17, 2015 and approved by the
ACSM Board of Trustees on November 20, 2015. This po-
sition statement is in effect until December 31, 2019. Posi-
tion papers should not be used to indicate endorsement of
products or services.
Authors
Academy: D. Travis Thomas, PhD, RDN, CSSD (College of Health Sci-
ences, University of Kentucky, Lexington, KY);
DC: Kelly Anne Erdman, MSc, RD, CSSD (Canadian Sport Institute
Calgary/University of Calgary Sport Medicine Centre, Calgary, AB,
Canada);
ACSM: Louise M. Burke, OAM, PhD, APD, FACSM (AIS Sports
Nutrition/Australian Institute of Sport Australia and Mary MacKillop
Institute of Health Research, Australian Catholic University).
Reviewers
Academy
Sports, Cardiovascular and Wellness Nutrition dietetic practice group
(Jackie Buell, PhD, RD, CSSD, ATC Ohio State University, Columbus, OH);
Amanda Carlson-Phillips, MS, RD, CSSD (EXOS - Phoenix, AZ);
Sharon Denny, MS, RD (Academy Knowledge Center, Chicago, IL);
D. Enette Larson-Meyer, PhD, RD, FACSM (University of Wyoming,
Laramie, WY);
TABLE 4. Sports dietitian roles and responsibilities.
Role of Sports Dietitian Responsibilities
Assessment of nutritional needs and current dietary practices � Energy intake, nutrients and fluids before, during and after training and competitions
� Nutrition-related health concerns (eating disorders, food allergies or intolerances, gastrointestinal
disturbances, injury management, muscle cramps, hypoglycemia, etc.) and body composition goals
� Food and fluid intake as well as estimated energy expenditure during rest, taper and travel days
� Nutritional needs during extreme conditions (eg, high altitude training, environmental concerns)
� Adequacy of athlete_s body weight and metabolic risk factors associated with low body weight
� Supplementation practices
� Basic measures of height, body weight, etc. with possible assessment of body composition
Interpretation of test results (eg, biochemistry, anthropometry) � Blood, urine analysis, body composition and physiological testing results, including hydration status
Dietary prescription and education � Dietary strategies to support behavior change for improvements with health, physical performance,
body composition goals and/or eating disorders
� Dietary recommendations prescribed relative to athlete_s personal goals and chief concerns related to
training, body composition, and/or competition nutrition, tapering, and/or periodized fat/weight loss
� Quantity, quality, and timing for food and fluid intake before, during and after training and/or
competition to enhance exercise training capacity, endurance and performance
� Medical nutritional therapeutic advice pertaining to unique dietary considerations (eating disorders,
food allergies, diabetes, gastrointestinal issues, etc.)
� Menu planning, time management, grocery shopping, food preparation, food storage, food budgeting,
food security, and recipe modification for training and/or competition days
� Food selection related to travel, restaurants, and training and competition venue choices
� Supplementation, ergogenic aids, fortified foods, etc. regarding legality, safety, and efficacy
� Sport nutrition education, resource development and support may be with individual athletes,
entire teams, and/or with coaches, athletic trainers, physiologists, food service staff, etc.
Collaboration and integration � Contribution as a member of a multidisciplinary team within sport settings to integrate nutrition
programming into a team or athlete_s annual training and competition plan
� Collaboration with the health care team/performance professionals (physicians, athletic trainer,
physiologists, psychologists, etc.) for the performance management of athletes
Evaluation and professionalism � Evaluation of scientific literature and provision of evidence-based assessment and application to
athletic performance
� Development of oversight of nutrition policies and procedures
� Documentation of measurable outcomes of nutrition services
� Recruitment and retention of clients and athletes in practice
� Provision of reimbursable services (eg, diabetes medical nutrition therapy)
� Promotion of career longevity for active individuals, collegiate and professional athletes
� Service as a mentor for developing sports dietetics professionals
� Maintenance of credential(s) by actively engaging in profession-specific continuing education activities
NUTRITION AND ATHLETIC PERFORMANCE Medicine & Science in Sports & Exercised 563
SPEC
IA
L
C
O
M
M
U
N
IC
ATIO
N
S
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Mary Pat Raimondi, MS, RD (Academy Policy Initiatives & Advocacy,
Washington DC);
DC
Ashley Armstrong, MS, RD (Canadian Sport Institute Pacific, Vancouver,
Victoria and Whistler, BC, Canada);
Susan Boegman, BSc, RD, IOC Dip Sport Nutrition (Canadian Sport
Institute Pacific, Victoria BC, Canada);
Susie Langley MS, RD, DS, FDC (Retired, Toronto, ON, Canada);
Marielle Ledoux, PhD, PDt (Professor, University of Montreal, Montreal,
QC, Canada);
Emma McCrudden, MSc (Canadian Sport Institute Pacific, Vancouver,
Victoria and Whistler, BC, Canada);
Pearle Nerenberg, MSc, PDt (Pearle Sports Nutrition, Montreal, QC, Canada);
Erik Sesbreno, BSc, RD, IOC Dip Sport Nutrition (Canadian Sport Institute
Ontario, Toronto, Ontario, Canada).
ACSM
Dan Benardot, PhD, RD, LD, FACSM (Georgia State University
Atlanta, GA);
Kristine Clark, PhD, RDN, FACSM (The Pennsylvania State University,
University Park, PA); Melinda M. Manore, PhD, RD, CSSD, FACSM (Oregon
State University, Corvallis, OR);
Emma Stevenson BSc, PhD (Newcastle University, Newcastle upon
Tyne, Tyne and Wear, UK).
Academy Positions CommitteeWorkgroup
Connie Diekman, Med, RD, CSSD, LD (chair) (Washington University,
St. Louis, MO);
Christine A. Rosenbloom, PhD, RDN, CSSD, FAND (Georgia State
University, Atlanta, GA); Roberta Anding, MS, RD/LD, CDE, CSSD,
FAND (content advisor) (Texas Children_s Hospital, Houston and Houston
Astros MLB Franchise, Houston,TX).
We thank the reviewers for their many constructive comments and sug-
gestions. The reviewers were not asked to endorse this position or the
supporting paper.
ACSM disclaimer: Care has been taken to confirm the accuracy of
the information present and to describe generally accepted practices.
However, the authors, editors, and publisher are not responsible for errors
or omissions or for any consequences from application of the information
in this publication and make no warranty, expressed or implied, with
respect to the currency, completeness, or accuracy of the contents of the
publication. Application of this information in a particular situation re-
mains the professional responsibility of the practitioner; the clinical treat-
ments described and recommended may not be considered absolute and
universal recommendations.
REFERENCES
1. Deakin V, Kerr D, Boushey C. Measuring nutritional status of
athletes: clinical and research perspectives. In: Burke L, Deakin V,
eds. Clinical Sports Nutrition. 5th eds. North Ryde, Australia:
McGraw-Hill; 2015:27–53.
2. Manore M, Thompson J. Energy requirements of the athlete:
assessment and evidence of energy efficiency. In: Burke L,
Deakin V, eds.Clinical Sports Nutrition. 5th eds. Sydney, Australia:
McGraw-Hill; 2015:114–139.
3. Spriet LL. New insights into the interaction of carbohydrate and fat
metabolism during exercise. Sports medicine. 2014;44(Suppl 1):S87–96.
4. Cunningham JJ. A reanalysis of the factors influencing basal
metabolic rate in normal adults. The American journal of Clinical
Nutrition. 1980;33(11):2372–2374.
5. Roza AM, Shizgal HM. The Harris Benedict equation reevaluated:
resting energy requirements and the body cell mass. The American
journal of clinical nutrition. 1984;40(1):168–182.
6. Guebels CP, Kam LC,Maddalozzo GF,ManoreMM. Active women
before/after an intervention designed to restore menstrual function:
resting metabolic rate and comparison of four methods to quantify
energy expenditure and energy availability. International Journal
of Sport Nutrition and Exercise Metabolism. 2014;24(1):37–46.
7. Ainsworth BE, Haskell WL, Whitt MC, et al. Compendium of
physical activities: an update of activity codes and MET in-
tensities. Medicine and Science in Sports and Exercise. 2000;
32(9 Suppl):S498–504.
8. U.S. Department of Health and Human Services, U.S. Department
of Agriculture. Dietary Guidelines for Americans, 2015–2020.
8th edition. Available from http://health.gov/dietaryguidelines/
2015/guidelines/.
9. Institute of Medicine FaNB. Dietary Reference intakes for Energy,
Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein and
Amino Acids. Washington, D.C.: National Academies Press; 2005.
10. Loucks AB. Energy balance and energy availability. In: Maughan
RJ, ed. Sports Nutrition, The Encyclopaedia of Sports Medicine,
an IOC Medical Commission Publication. West Sussex, UK:
John Wiley & Sons, Ltd.; 2013:72–87.
11. De Souza MJ, Nattiv A, Joy E, et al. 2014 Female Athlete Triad
Coalition Consensus Statement on Treatment and Return to Play
of the Female Athlete Triad: 1st International Conference held in
San Francisco, California, May 2012 and 2nd International
Conference held in Indianapolis, Indiana, May 2013. British
Journal of Sports Medicine. 2014;48(4):289.
12. Mountjoy M, Sundgot-Borgen J, Burke L, et al. The IOC con-
sensus statement: beyond the Female Athlete Triad–Relative
Energy Deficiency in Sport (RED-S). British Journal of Sports
Medicine. 2014;48(7):491–497.
13. Garner DM. Eating Disorder Inventory-3: Professional Manual.
Psychological Assessment Resources, Incorporated; 2004.
14. Association AP. Diagnostic and Statistical Manual of Mental
Disorders, 5th Edition: DSM 5.
15. Sundgot-Borgen J, Meyer NL, Lohman TG, et al. How to mini-
mise the health risks to athletes who compete in weight-sensitive
sports review and position statement on behalf of the Ad Hoc
Research Working Group on Body Composition, Health and
Performance, under the auspices of the IOC Medical Commission.
British Journal of Sports Medicine. 2013;47(16):1012–1022.
16. Stellingwerff T, Maughan RJ, Burke LM. Nutrition for power
sports: middle-distance running, track cycling, rowing, canoeing/
kayaking, and swimming. Journal of sports sciences. 2011;
29(Suppl 1):S79–89.
17. O_Connor H, Slater G. Losing, gaining and making weight for
athletes. In: Lanham-New S, Stear S, Sherriffs M, Collins A, eds.
Sport and Exercise Nutrition. West Sussex, UK: Wiley-Blackwell;
2011:210–232.
18. Sundgot-Borgen J, Garthe I. Elite athletes in aesthetic and
Olympic weight-class sports and the challenge of body weight and
body compositions. Journal of Sports Sciences. 2011;29(Suppl 1):
S101–114.
19. Steffes GD, Megura AE, Adams J, et al. Prevalence of metabolic
syndrome risk factors in high school and NCAA division I football
players. Journal of Strength and Conditioning Research / National
Strength & Conditioning Association. 2013;27(7):1749–1757.
20. Ackland TR, Lohman TG, Sundgot-Borgen J, et al. Current sta-
tus of body composition assessment in sport: review and position
statement on behalf of the ad hoc research working group on
body composition health and performance, under the auspices of
http://www.acsm-msse.org564 Official Journal of the American College of Sports Medicine
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
http://health.gov/dietaryguidelines/2015/guidelines/
http://health.gov/dietaryguidelines/2015/guidelines/
the I.O.C. Medical Commission. Sports Medicine. 2012;42(3):
227–249.
21. Santos DA, Dawson JA, Matias CN, et al. Reference values for
body composition and anthropometric measurements in athletes.
PloS One. 2014;9(5):e97846.
22. Turocy PS, DePalma BF, Horswill CA, et al. National Athletic
Trainers_ Association position statement: safe weight loss and
maintenance practices in sport and exercise. Journal of Athletic
Training. 2011;46(3):322–336.
23. Slater G, Rice A, Jenkins D, Hahn A. Body mass management
of lightweight rowers: nutritional strategies and performance
implications. British Journal of Sports Medicine. 2014;48(21):
1529–1533.
24. Wilson G, Drust B, Morton JP, Close GL. Weight-making
strategies in professional jockeys: implications for physical and
mental health and well-being. Sports Medicine. 2014;44(6):
785–796.
25. Garthe I, Raastad T, Refsnes PE, Koivisto A, Sundgot-Borgen J.
Effect of two different weight-loss rates on body composition and
strength and power-related performance in elite athletes. Inter-
national Journal of Sport Nutrition and Exercise Metabolism.
2011;21(2):97–104.
26. Mettler S, Mitchell N, Tipton KD. Increased protein intake reduces
lean body mass loss during weight loss in athletes. Medicine and
Science in Sports and Exercise. 2010;42(2):326–337.
27. Thomas DM, Martin CK, Lettieri S, et al. Can a weight loss of one
pound a week be achieved with a 3500-kcal deficit? Commentary
on a commonly accepted rule. Int J Obes (Lond). 2013;37(12):
1611–1613.
28. Hopkins WG, Hawley JA, Burke LM. Design and analysis of
research on sport performance enhancement. Medicine and Sci-
ence in Sports and Exercise. 1999;31(3):472–485.
29. Maughan RJ, Gleeson M. The Biochemical Basis of Sports Per-
formance. OUP Oxford; 2010.
30. Hawley JA, Burke LM, Phillips SM, Spriet LL. Nutritional
modulation of training-induced skeletal muscle adaptations.
Journal of Applied Physiology. 2011;110(3):834–845.
31. Cole M, Coleman D, Hopker J, Wiles J. Improved gross effi-
ciency during long duration submaximal cycling following a
short-term high carbohydrate diet. International Journal of
Sports Medicine. 2014;35(3):265–269.32. Philp A, Hargreaves M, Baar K. More than a store: regulatory
roles for glycogen in skeletal muscle adaptation to exercise.
American Journal of Physiology, Endocrinology and Metabo-
lism. 2012;302(11):E1343–1351.
33. Bartlett JD, Hawley JA, Morton JP. Carbohydrate availability
and exercise training adaptation: Too much of a good thing? Eur
J Sport Sci. 2014:1–10.
34. Stellingwerff T. Contemporary nutrition approaches to optimize
elite marathon performance. International Journal of Sports
Physiology and Performance. 2013;8(5):573–578.
35. Lee JM, Kim Y, Welk GJ. Validity of consumer-based physical
activity monitors. Medicine and Science in Sports and Exercise.
2014;46(9):1840–1848.
36. Burke LM, Hawley JA, Wong SH, Jeukendrup AE. Carbohy-
drates for training and competition. Journal of Sports Sciences.
2011;29(Suppl 1):S17–27.
37. Cox GR, Clark SA, Cox AJ, et al. Daily training with high car-
bohydrate availability increases exogenous carbohydrate oxida-
tion during endurance cycling. Journal of Applied Physiology.
2010;109(1):126–134.
38. Burke LM. Fueling strategies to optimize performance: training
high or training low? Scandinavian Journal of Medicine & Sci-
ence in Sports. 2010;20(Suppl 2):48–58.
39. Phillips SM, Van Loon LJ. Dietary protein for athletes: from
requirements to optimum adaptation. Journal of Sports Sciences.
2011;29(Suppl 1):S29–38.
40. Phillips SM. Dietary protein requirements and adaptive advantages
in athletes. The British Journal of Nutrition. 2012;108(Suppl 2):
S158–167.
41. Miller BF, Olesen JL, Hansen M, et al. Coordinated collagen and
muscle protein synthesis in human patella tendon and quadriceps
muscle after exercise. The Journal of Physiology. 2005;567(Pt 3):
1021–1033.
42. Babraj J, Cuthbertson DJ, Rickhuss P, et al. Sequential extracts of
human bone show differing collagen synthetic rates. Biochemical
Society Transactions. 2002;30(2):61–65.
43. Churchward-Venne TA, Burd NA, Mitchell CJ, et al. Supple-
mentation of a suboptimal protein dose with leucine or essential
amino acids: effects on myofibrillar protein synthesis at rest and
following resistance exercise in men. The Journal of Physiology.
2012;590(Pt 11):2751–2765.
44. Burd NA, West DW, Moore DR, et al. Enhanced amino acid
sensitivity of myofibrillar protein synthesis persists for up to 24 h
after resistance exercise in young men. The Journal of Nutrition.
2011;141(4):568–573.
45. Joint WHOFAOUNUEC. Protein and amino acid requirements in
human nutrition. World Health Organization Technical Report
Series. 2007(935):1–265, back cover.
46. Rosenbloom CA, Coleman EJ. Sports Nutrition: A Practice Manual
for Professionals. Academy of Nutrition & Dietetics; 2012.
47. Moore DR, Phillips SM, Slater G. Protein. In: Deakin V, Burke L,
eds. Clinical Sports Nutrition. 5th ed: McGraw-Hill Education;
2015:94–113.
48. Areta JL, Burke LM, Camera DM, et al. Reduced resting skeletal
muscle protein synthesis is rescued by resistance exercise and
protein ingestion following short-term energy deficit. American
Journal of Physiology, Endocrinology and Metabolism. 2014;
306(8):E989–997.
49. Rodriguez NR, Vislocky LM, Gaine PC. Dietary protein, en-
durance exercise, and human skeletal-muscle protein turnover.
Current Opinion in Clinical Nutrition and Metabolic Care. 2007;
10(1):40–45.
50. Wall BT, Morton JP, van Loon LJ. Strategies to maintain skeletal
muscle mass in the injured athlete: nutritional considerations and
exercise mimetics. Eur J Sport Sci. 2015;15(1):53–62.
51. Phillips SM, Moore DR, Tang JE. A critical examination of di-
etary protein requirements, benefits, and excesses in athletes.
International Journal of Sport Nutrition and Exercise Metabo-
lism. 2007;(Suppl 17):S58–76.
52. Tipton KD, Witard OC. Protein requirements and recommen-
dations for athletes: relevance of ivory tower arguments for
practical recommendations. Clinics in Sports Medicine. 2007;
26(1):17–36.
53. Beelen M, Burke LM, Gibala MJ, van Loon LJ. Nutritional
strategies to promote postexercise recovery. International Journal
of Sport Nutrition and Exercise Metabolism. 2010;20(6):515–532.
54. Moore DR, Robinson MJ, Fry JL, et al. Ingested protein dose
response of muscle and albumin protein synthesis after resistance
exercise in young men. The American Journal of Clinical Nu-
trition. 2009;89(1):161–168.
55. Areta JL, Burke LM, Ross ML, et al. Timing and distribution of
protein ingestion during prolonged recovery from resistance ex-
ercise alters myofibrillar protein synthesis. The Journal of Phys-
iology. 2013;591(Pt 9):2319–2331.
56. Schoenfeld BJ, Ratamess NA, Peterson MD, Contreras B, Sonmez
GT, Alvar BA. Effects of different volume-equated resistance
training loading strategies on muscular adaptations in well-trained
men. Journal of Strength and Conditioning Research / National
Strength & Conditioning Association. 2014;28(10):2909–2918.
57. Josse AR, Tang JE, Tarnopolsky MA, Phillips SM. Body com-
position and strength changes in women with milk and resistance
exercise. Medicine and Science in Sports and Exercise. 2010;
42(6):1122–1130.
NUTRITION AND ATHLETIC PERFORMANCE Medicine & Science in Sports & Exercised 565
SPEC
IA
L
C
O
M
M
U
N
IC
ATIO
N
S
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
58. Phillips SM. A brief review of critical processes in exercise-
induced muscular hypertrophy. Sports Medicine. 2014;44(Suppl 1):
S71–77.
59. Tipton KD, Elliott TA, Cree MG, Aarsland AA, Sanford AP,
Wolfe RR. Stimulation of net muscle protein synthesis by whey
protein ingestion before and after exercise. American Journal of
Physiology. Endocrinology and metabolism. 2007;292(1):E71–76.
60. Hartman JW, Tang JE, Wilkinson SB, et al. Consumption of fat-
free fluid milk after resistance exercise promotes greater lean
mass accretion than does consumption of soy or carbohydrate in
young, novice, male weightlifters. The American Journal of
Clinical Nutrition. 2007;86(2):373–381.
61. Josse AR, Atkinson SA, Tarnopolsky MA, Phillips SM. In-
creased consumption of dairy foods and protein during diet- and
exercise-induced weight loss promotes fat mass loss and lean
mass gain in overweight and obese premenopausal women. The
Journal of Nutrition. 2011;141(9):1626–1634.
62. Pennings B, Boirie Y, Senden JM, Gijsen AP, Kuipers H, van
Loon LJ. Whey protein stimulates postprandial muscle protein
accretion more effectively than do casein and casein hydrolysate
in older men. The American Journal of Clinical Nutrition.
2011;93(5):997–1005.
63. Health Canada. Eating Well with Canada_s Fod Guide. http://
www.hc-sc.gc.ca/fn-an/food-guide-aliment/index-eng.php. Accessed
7 July, 2015.
64. Phinney SD, Bistrian BR, Evans WJ, Gervino E, Blackburn GL.
The human metabolic response to chronic ketosis without caloric
restriction: preservation of submaximal exercise capability with
reduced carbohydrate oxidation. Metabolism: Clinical and Ex-
perimental. 1983;32(8):769–776.
65. Volek JS, Noakes T, Phinney SD. Rethinking fat as a fuel for
endurance exercise. Eur J Sport Sci. 2014:1–8.
66. Havemann L, West SJ, Goedecke JH, et al. Fat adaptation
followed by carbohydrate loading compromises high-intensity
sprint performance. Journal of Applied Physiology. 2006;100(1):
194–202.
67. Stellingwerff T, Spriet LL, Watt MJ, et al. Decreased PDH acti-
vation and glycogenolysis during exercise following fat adaptation
with carbohydrate restoration. American Journal of Physiology.
Endocrinology and Metabolism. 2006;290(2):E380–388.
68. Burke LM. Re-examining high-fat diets for sports performance:
did we call the ‘‘nail in the coffin’’ too soon? Sports Medicine.
2015;45(1):33–49.
69. Barnes MJ. Alcohol: impact on sports performance and recovery
in male athletes. Sports Medicine. 2014;44(7):909–919.
70. Lourenco S, Oliveira A, Lopes C. The effect of current and
lifetime alcohol consumption on overall and central obesity.
European Journal of Clinical Nutrition. 2012;66(7):813–818.
71. Burke LM,exercise
performance during recovery?
Based on the limited evidence available, there were no clear effects of carbohydrate
supplementation during and after endurance exercise on carbohydrate and protein-specific
metabolic responses during recovery.
Grade III- Limited
#4: What is the effect of consuming CHO on exercise performance during
recovery?
Based on the limited evidence available, there were no clear effects of carbohydrate
supplementation during and after endurance exercise on endurance performance in adult
athletes during recovery.
Grade III- Limited
#5: In adult athletes, what is the effect of consuming carbohydrate and protein
together on carbohydrate and protein-specific metabolic responses
during recovery?
� Compared to ingestion of carbohydrate alone, coingestion of carbohydrate plus protein
together during the recovery period resulted in no difference in the rate of muscle
glycogen synthesis.
� Coingestion of protein with carbohydrate during the recovery period resulted in improved net
protein balance post-exercise.
� The effect of co-ingestion of protein with carbohydrate on creatine kinase levels is
inconclusive and shows no impact on muscle soreness post-exercise.
Grade I- Good
#6: In adult athletes, what is the effect of consuming carbohydrate and protein
together on carbohydrate and protein-specific metabolic responses
during recovery?
Co-ingestion of carbohydrate plus protein, together during the recovery period resulted in no
clear influence on subsequent strength or sprint power.
Grade II- Fair
#7: In adult athletes, what is the effect of consuming carbohydrate and protein
together on exercise performance during recovery?
Ingesting protein during the recovery period (post-exercise) led to accelerated recovery of static
force and dynamic power production during the delayed onset muscle soreness period and
more repetitions performed subsequent to intense resistance training.
Grade II- Fair
#8: In adult athletes, what is the effect of consuming protein on carbohydrate
and protein-specific metabolic responses during recovery?
Ingesting protein (approximately 20 g to 30 g total protein, or approximately 10 g of essential
amino acids) during exercise or the recovery period (post-exercise) led to increased whole
body and muscle protein synthesis as well as improved nitrogen balance.
Grade I- Good
Training
#9: In adult athletes, what is the optimal blend of carbohydrates for maximal
carbohydrate oxidation during exercise?
Based on the limited evidence available, carbohydrate oxidation was greater in carbohydrate
conditions (glucose and glucose + fructose) compared to water placebo, but no differences
between the two carbohydrate blends tested were observed in male cyclists. Exogenous
carbohydrate oxidation was greater in the glucose + fructose condition vs. glucose-only in
a single study.
Grade III- Limited
#10: In adult athletes, what effect does training with limited carbohydrate
availability have on metabolic adaptations that lead to performance
improvements?
Training with limited carbohydrate availability may lead to some metabolic adaptations during
training, but did not lead to performance improvements. Based on the evidence examined,
while there is insufficient evidence supporting a clear performance effect, training with
limited carbohydrate availability impaired training intensity and duration.
Grade II- Fair
#11: In adult athletes, what effect does consuming high or low glycemic
meals or foods have on training related metabolic responses and
exercise performance?
In the majority of studies examined, neither glycemic index nor glycemic load affected
endurance performance nor metabolic responses when conditions were matched for
carbohydrate and energy.
Grade I- Good
Evidence Grades: Grade I: Good; Grade II: Fair; Grade III: Limited; Grade IV: Expert Opinion Only; Grade V: Not Assignable.
http://www.acsm-msse.org544 Official Journal of the American College of Sports Medicine
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
http://www.andevidencelibrary.com
http://www.andevidencelibrary.com/
cycles of the training calendar. Nutrition support also
needs to be periodized, taking into account the needs
of daily training sessions (which can range from
minor in the case of ‘‘easy’’ workouts to substantial
in the case of high quality sessions (eg, high inten-
sity, strenuous, or highly skilled workouts) and
overall nutritional goals.
2. Nutrition plans need to be personalized to the indi-
vidual athlete to take into account the specificity and
uniqueness of the event, performance goals, practical
challenges, food preferences, and responses to various
strategies.
3. A key goal of training is to adapt the body to develop
metabolic efficiency and flexibility while competition
nutrition strategies focus on providing adequate sub-
strate stores to meet the fuel demands of the event and
support cognitive function.
4. Energy availability, which considers energy intake in
relation to the energy cost of exercise, sets an im-
portant foundation for health and the success of sports
nutrition strategies.
5. The achievement of the body composition associated
with optimal performance is now recognized as an
important but challenging goal that needs to be indi-
vidualized and periodized. Care should be taken to
preserve health and long term performance by avoiding
practices that create unacceptably low energy avail-
ability and psychological stress.
6. Training and nutrition have a strong interaction in
acclimating the body to develop functional and met-
abolic adaptations. Although optimal performance is
underpinned by the provision of pro-active nutrition
support, training adaptations may be enhanced in the
absence of such support.
7. Some nutrients (eg, energy, carbohydrate, and pro-
tein) should be expressed using guidelines per kg
body mass to allow recommendations to be scaled to
the large range in the body sizes of athletes. Sports
nutrition guidelines should also consider the impor-
tance of the timing of nutrient intake and nutritional
support over the day and in relation to sport rather
than general daily targets.
8. Highly trained athletes walk a tightrope between
training hard enough to achieve a maximal training
stimulus and avoiding the illness and injury risk as-
sociated with an excessive training volume.
9. Competition nutrition should target specific strategies
that reduce or delay factors that would otherwise
cause fatigue in an event; these are specific to the
event, the environment/scenario in which it is un-
dertaken, and the individual athlete.
10. New performance nutrition options have emerged in
the light of developing but robust evidence that brain
sensing of the presence of carbohydrate, and poten-
tially other nutritional components, in the oral cavity
can enhance perceptions of well-being and increase
self-chosen work rates. Such findings present oppor-
tunities for intake during shorter events, in which
fluid or food intake was previously not considered to
offer a metabolic advantage, by enhancing perfor-
mance via a central effect.
11. A pragmatic approach to advice regarding the use of
supplements and sports foods is needed in the face of
the high prevalence of interest in, and use by, athletes
and the evidence that some products can usefully
contribute to a sports nutrition plan and/or directly
enhance performance. Athletes should be assisted to
undertake a cost-benefit analysis of the use of such
products and to recognize that they are of the greatest
value when added to a well-chosen eating plan.
THEME 1: NUTRITION FOR
ATHLETE PREPARATION
Energy Requirements, Energy Balance
and Energy Availability
An appropriate energy intake is the cornerstone of the
athlete_s diet since it supports optimal body function, de-
termines the capacity for intake of macronutrient and
micronutrients, and assists in manipulating body composi-
tion. An athlete_s energy intakeCollier GR, Broad EM, et al. Effect of alcohol intake
on muscle glycogen storage after prolonged exercise. Journal of
Applied Physiology. 2003;95(3):983–990.
72. Hobson RM, Maughan RJ. Hydration status and the diuretic ac-
tion of a small dose of alcohol. Alcohol and Alcoholism. 2010;
45(4):366–373.
73. Parr EB, Camera DM, Areta JL, et al. Alcohol ingestion impairs
maximal post-exercise rates of myofibrillar protein synthesis
following a single bout of concurrent training. PloS One. 2014;
9(2):e88384.
74. Burke LM, Read RS. A study of dietary patterns of elite Australian
football players. Canadian Journal of Sport Sciences = Journal
Canadien Des Sciences Du Sport. 1988;13(1):15–19.
75. Graham T. Alcohol ingestion and man_s ability to adapt to ex-
ercise in a cold environment. Canadian Journal of Applied Sport
Sciences. Journal Canadien Des Sciences Appliquees Au Sport.
1981;6(1):27–31.
76. Verster JC. The alcohol hangover–a puzzling phenomenon.
Alcohol and Alcoholism. 2008;43(2):124–126.
77. Farajian P, Kavouras SA, Yannakoulia M, Sidossis LS. Dietary
intake and nutritional practices of elite Greek aquatic athletes.
International Journal of Sport Nutrition and Exercise Metabo-
lism. 2004;14(5):574–585.
78. Lukaski HC. Vitamin and mineral status: effects on physical
performance. Nutrition. 2004;20(7–8):632–644.
79. Volpe SL, Bland E. Vitamins, Minerals, and Exercise. In:
Rosenbloom CA, Coleman EJ, ed. Sports Nutrition: A Practice
Manual for Professionals. 5th eds. Chicago: Academy of Nutri-
tion and Dietetics; 2012:75–105.
80. Woolf K, Manore MM. B-vitamins and exercise: does exercise
alter requirements? International Journal of Sport Nutrition and
Exercise Metabolism. 2006;16(5):453–484.
81. Haymes E. Iron. In: Driskell J, Wolinsky I, eds. Sports Nutrition:
Vitamins and Trace Elements. New York, NY: CRC/Taylor &
Francis; 2006:203–216.
82. Beard J, Tobin B. Iron status and exercise. The American Journal
of Clinical Nutrition. 2000;72(2 Suppl):594S–597S.
83. McClung JP, Karl JP, Cable SJ, et al. Randomized, double-blind,
placebo-controlled trial of iron supplementation in female soldiers
during military training: effects on iron status, physical perfor-
mance, and mood. The American Journal of Clinical Nutrition.
2009;90(1):124–131.
84. DellaValle DM. Iron supplementation for female athletes: effects
on iron status and performance outcomes. Current Sports Medi-
cine Reports. 2013;12(4):234–239.
85. Cowell BS, Rosenbloom CA, Skinner R, Summers SH. Policies
on screening female athletes for iron deficiency in NCAA divi-
sion I-A institutions. International Journal of Sport Nutrition and
Exercise Metabolism. 2003;13(3):277–285.
86. Peeling P, Dawson B, Goodman C, Landers G, Trinder D. Ath-
letic induced iron deficiency: new insights into the role of in-
flammation, cytokines and hormones. European Journal of
Applied Physiology. 2008;103(4):381–391.
87. SimM, Dawson B, Landers G, Trinder D, Peeling P. Iron regulation
in athletes: exploring the menstrual cycle and effects of different
exercise modalities on hepcidin production. International Journal
of Sport Nutrition and Exercise Metabolism. 2014;24(2):177–187.
88. Peeling P, Sim M, Badenhorst CE, et al. Iron status and the acute
post-exercise hepcidin response in athletes. PloS One. 2014;
9(3):e93002.
89. Burden RJ, Morton K, Richards T, Whyte GP, Pedlar CR. Is iron
treatment beneficial in, iron-deficient but non-anaemic (IDNA)
endurance athletes? A meta-analysis. British Journal of Sports
Medicine. 2015;49(21):1389–1397.
90. Pojednic RM, Ceglia L. The emerging biomolecular role of vi-
tamin D in skeletal muscle. Exercise and Sport Sciences Reviews.
2014;42(2):76–81.
91. Sinha A, Hollingsworth KG, Ball S, Cheetham T. Improving the
vitamin D status of vitamin D deficient adults is associated with
improved mitochondrial oxidative function in skeletal muscle.
The Journal of Clinical Endocrinology and Metabolism. 2013;
98(3):E509–513.
92. Ruohola JP, Laaksi I, Ylikomi T, et al. Association between se-
rum 25(OH)D concentrations and bone stress fractures in Finnish
young men. Journal of Bone and Mineral Research: The Official
Journal of the American Society for Bone and Mineral Research.
2006;21(9):1483–1488.
93. Larson-Meyer DE, Willis KS. Vitamin D and athletes. Current
Sports Medicine Reports. 2010;9(4):220–226.
94. Cannell JJ, Hollis BW, Sorenson MB, Taft TN, Anderson JJ.
Athletic performance and vitamin D. Medicine and Science in
Sports and Exercise. 2009;41(5):1102–1110.
95. Halliday TM, Peterson NJ, Thomas JJ, Kleppinger K, Hollis BW,
Larson-Meyer DE. Vitamin D status relative to diet, lifestyle,
injury, and illness in college athletes. Medicine and Science in
Sports and Exercise. 2011;43(2):335–343.
http://www.acsm-msse.org566 Official Journal of the American College of Sports Medicine
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
http://www.hc-sc.gc.ca/fn-an/food-guide-aliment/index-eng.php
http://www.hc-sc.gc.ca/fn-an/food-guide-aliment/index-eng.php
96. Lagowska K, Kapczuk K, Friebe Z, Bajerska J. Effects of die-
tary intervention in young female athletes with menstrual dis-
orders. Journal of the International Society of Sports Nutrition.
2014;11:21.
97. Lewis RM, Redzic M, Thomas DT. The effects of season-long
vitamin d supplementation on collegiate swimmers and divers.
International Journal of Sports Nutrition and Exercise Metabo-
lism. 2013;23(5):431–440.
98. Moran DS, McClung JP, Kohen T, Lieberman HR. Vitamin d
and physical performance. Sports Medicine. 2013;43(7):601–611.
99. Nickols-Richardson SM, Beiseigel JM, Gwazdauskas FC. Eat-
ing restraint is negatively associated with biomarkers of bone
turnover but not measurements of bone mineral density in young
women. Journal of the American Dietetic Association. 2006;
106(7):1095–1101.
100. Nattiv A, Loucks AB, Manore MM, et al. American College of
Sports Medicine position stand. The female athlete triad. Medicine
and Science in Sports and Exercise. 2007;39(10):1867–1882.
101. Peternelj TT, Coombes JS. Antioxidant supplementation during
exercise training: beneficial or detrimental? Sports Medicine. 2011;
41(12):1043–1069.
102. Watson TA, MacDonald-Wicks LK, Garg ML. Oxidative stress
and antioxidants in athletes undertaking regular exercise training.
International Journal of Sports Nutrition and Exercise Metabo-
lism. 2005;15(2):131–146.
103. Draeger CL, Naves A, Marques N, et al. Controversies of anti-
oxidant vitamins supplementation in exercise: ergogenic or ergo-
lytic effects in humans? Journal of the International Society of
Sports Nutrition. 2014;11(1):4.
104. American College of Sports M, Sawka MN, Burke LM, et al.
American College of Sports Medicine position stand. Exercise
and fluid replacement. Medicine and Science in Sports and Ex-
ercise. 2007;39(2):377–390.
105. Shirreffs SM, Sawka MN. Fluid and electrolyte needs for train-
ing, competition, and recovery. Journal of Sports Sciences. 2011;
29(Suppl 1):S39–46.
106. Kenefick RW, Cheuvront SN. Hydration for recreational sport and
physical activity. Nutrition Reviews. 2012;70(Suppl 2):S137–142.
107. American College of Sports M, Armstrong LE, Casa DJ, et al.
American College of Sports Medicine position stand. Exertional
heat illness during training and competition. Medicine and Sci-
ence in Sports and Exercise. 2007;39(3):556–572.
108. Goulet ED. Dehydration and endurance performance in compet-
itive athletes. Nutrition Reviews. 2012;70(Suppl 2):S132–136.
109. Jeukendrup A, Carter J, Maughan RJ. Competition fluid and fuel.
In: Burke L, Deakin V, eds. Clinical Sports Nutrition. 5th ed.
North Ryde NSW, Australia: McGraw-Hill Australia Pty Ltd;
2015:377–419.
110. Garth AK, Burke LM. What do athletes drink during competitive
sporting activities? Sports Medicine. 2013;43(7):539–564.
111. Mountjoy M, Alonso JM, Bergeron MF, et al. Hyperthermic-
related challenges in aquatics, athletics, football, tennis and triath-
lon. BritishJournal of Sports Medicine. 2012;46(11):800–804.
112. Koehle MS, Cheng I, Sporer B. Canadian Academy of Sport and
Exercise Medicine position statement: athletes at high altitude.
Clinical Journal of Sport Medicine: Official Journal of the Canadian
Academy of Sport Medicine. 2014;24(2):120–127.
113. Jeukendrup AE. Nutrition for endurance sports: marathon, triath-
lon, and road cycling. Journal of Sports Sciences. 2011;29(suppl 1):
S91–99.
114. Hew-Butler T, Rosner MH, Fowkes-Godek S, et al. Statement of
the Third International Exercise-Associated Hyponatremia Con-
sensus Development Conference, Carlsbad, California, 2015.
Clinical Journal of Sport Medicine: Official Journal of the
Canadian Academy of Sport Medicine. 2015;25(4):303–320.
115. Bergeron MF. Exertional heat cramps: recovery and return to
play. Journal of Sport Rehabilitation. 2007;16(3):190–196.
116. Cermak NM, van Loon LJ. The use of carbohydrates during exercise
as an ergogenic aid. Sports Medicine. 2013;43(11):1139–1155.
117. Burke LM, Kiens B, Ivy JL. Carbohydrates and fat for training
and recovery. Journal of Sports Sciences. 2004;22(1):15–30.
118. Hawley JA, Schabort EJ, Noakes TD, Dennis SC. Carbohydrate-
loading and exercise performance. An update. Sports Medicine.
1997;24(2):73–81.
119. Ormsbee MJ, Bach CW, Baur DA. Pre-exercise nutrition: the
role of macronutrients, modified starches and supplements on
metabolism and endurance performance. Nutrients. 2014;6(5):
1782–1808.
120. Rehrer NJ, van Kemenade M, Meester W, Brouns F, Saris WH.
Gastrointestinal complaints in relation to dietary intake in triathletes.
International Journal of Sport Nutrition. 1992;2(1):48–59.
121. Foster C, Costill DL, Fink WJ. Effects of preexercise feedings on
endurance performance. Med Sci Sports. 1979;11(1):1–5.
122. Coyle EF. Timing and method of increased carbohydrate intake
to cope with heavy training, competition and recovery. Journal of
Sports Sciences. 1991;9 Spec No:29–51; discussion 51–22.
123. Thomas DE, Brotherhood JR, Brand JC. Carbohydrate feeding
before exercise: effect of glycemic index. International Journal
of Sports Medicine. 1991;12(2):180–186.
124. Burke LM, Claassen A, Hawley JA, Noakes TD. Carbohydrate
intake during prolonged cycling minimizes effect of glycemic
index of preexercise meal. Journal of Applied Physiology. 1998;
85(6):2220–2226.
125. Stellingwerff T, Cox GR. Systematic review: Carbohydrate sup-
plementation on exercise performance or capacity of varying dura-
tions. Applied Physiology, Nutrition, and Metabolism = Physiologie
Appliquee, Nutrition et Metabolisme. 2014;39(9):998–1011.
126. Burke LM, Maughan RJ. The Governor has a sweet tooth -
Mouth sensing of nutrients to enhance sports performance. Eur J
Sport Sci. 2014:1–12.
127. Jeukendrup AE. Carbohydrate and exercise performance: the role
of multiple transportable carbohydrates. Current Opinion in
Clinical Nutrition and Metabolic Care. 2010;13(4):452–457.
128. Betts JA, Williams C. Short-term recovery from prolonged exer-
cise: exploring the potential for protein ingestion to accentuate the
benefits of carbohydrate supplements. Sports Medicine. 2010;
40(11):941–959.
129. Berardi JM, Noreen EE, Lemon PW. Recovery from a cycling
time trial is enhanced with carbohydrate-protein supplementation
vs. isoenergetic carbohydrate supplementation. Journal of the
International Society of Sports Nutrition. 2008;5:24.
130. Tipton KD, Rasmussen BB, Miller SL, et al. Timing of amino
acid-carbohydrate ingestion alters anabolic response of muscle to
resistance exercise. American Journal of Physiology. Endocri-
nology and Metabolism. 2001;281(2):E197–206.
131. van Essen M, Gibala MJ. Failure of protein to improve time trial
performance when added to a sports drink. Medicine and Science
in Sports and Exercise. 2006;38(8):1476–1483.
132. Ivy JL, Res PT, Sprague RC, Widzer MO. Effect of a carbohydrate-
protein supplement on endurance performance during exercise of
varying intensity. International Journal of Sports Nutrition and
Exercise Metabolism. 2003;13(3):382–395.
133. Etheridge T, Philp A, Watt PW. A single protein meal increases
recovery of muscle function following an acute eccentric exercise
bout. Applied Physiology, Nutrition, and Metabolism = Physiologie
Appliquee, Nutrition et Metabolisme. 2008;33(3):483–488.
134. Hoffman JR, Ratamess NA, Tranchina CP, Rashti SL, Kang J,
Faigenbaum AD. Effect of a proprietary protein supplement on
recovery indices following resistance exercise in strength/power
athletes. Amino Acids. 2010;38(3):771–778.
135. Beelen M, Koopman R, Gijsen AP, et al. Protein coingestion
stimulates muscle protein synthesis during resistance-type exer-
cise. American Journal of Physiology. Endocrinology and Me-
tabolism. 2008;295(1):E70–77.
NUTRITION AND ATHLETIC PERFORMANCE Medicine & Science in Sports & Exercised 567
SPEC
IA
L
C
O
M
M
U
N
IC
ATIO
N
S
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
136. van Loon LJ. Is there a need for protein ingestion during exer-
cise? Sports Medicine. 2014;44(suppl 1):S105–111.
137. Health Canada. Pathway for Licensing Natural Health Products
Making Modern Health Claims. http://www.hc-sc.gc.ca/dhp-mps/
prodnatur/legislation/docs/modern-eng.php#a11. Accessed August
19th, 2015.
138. Braun H, Koehler K, Geyer H, Kleiner J, Mester J, Schanzer W.
Dietary supplement use among elite young German athletes. In-
ternational Journal of Sport Nutrition and Exercise Metabolism.
2009;19(1):97–109.
139. Maughan RJ. Risks and rewards of dietary supplement use by
athletes. In: Maughan RJ, ed. Sports Nutrition, The Encyclopaedia
of Sports Medicine, an IOC Medical Commission, 1st Edition.
West Sussex, UK: John Wiley & Sons Ltd.; 2014.
140. International Organization for Standardization and International
Electrotechnical Commission. General requirements for the
competence of testing and calibration laboratories, Switzerland.
ISO/IEC 17025:2005 (E). ISO; 2005.
141. Burke LM, Cato L. Supplements and Sports Foods. In: Burke LM,
Deakin V, eds. Clinical Sports Nutrition, 5th Edition. 5th ed. North
Ryde NSW, Australia: McGraw-Hill Pty Ltd.; 2015:493–591.
142. Australian Institute of Sport. Supplements. http://www.ausport.gov.
au/ais/nutrition/supplements. Accessed 7 July, 2015.
143. Tarnopolsky MA. Caffeine and creatine use in sport. Ann Nutr
Metab. 2010;57(Suppl 2):1–8.
144. Astorino TA, Roberson DW. Efficacy of acute caffeine ingestion
for short-term high-intensity exercise performance: a systematic
review. Journal of Strength and Conditioning Research / National
Strength & Conditioning Association. 2010;24(1):257–265.
145. Burke L, Desbrow B, Spriet L. Caffeine for Sports Performance.
Human Kinetics; 2013.
146. Carr AJ, Hopkins WG, Gore CJ. Effects of acute alkalosis and
acidosis on performance: a meta-analysis. Sports Medicine. 2011;
41(10):801–814.
147. Quesnele JJ, Laframboise MA, Wong JJ, Kim P, Wells GD. The
effects of beta-alanine supplementation on performance: a sys-
tematic review of the literature. International Journal of Sport
Nutrition and Exercise Metabolism. 2014;24(1):14–27.
148. Jones AM. Influence of dietary nitrate on the physiological de-
terminants of exercise performance: a critical review. Applied
Physiology, Nutrition, and Metabolism = Physiologie Appliquee,
Nutrition et Metabolisme. 2014;39(9):1019–1028.
149. Craig WJ, Mangels AR, American Dietetic A. Position of the
American Dietetic Association: vegetarian diets. Journal of the
American Dietetic Association. 2009;109(7):1266–1282.
150. Berning JR. The Vegetarian Athlete. In: Maughan RJ, ed. The
Encyclopaedia of Sports Medicine: An IOC Medical Commission
Publications, Sports Nutrition. West Sussux, UK: Wiley; 2014:
382–391.
151. Burke DG, Chilibeck PD, Parise G, Candow DG, Mahoney D,
Tarnopolsky M. Effect of creatine and weight training on muscle
creatine and performance in vegetarians. Medicine and Science in
Sports and Exercise. 2003;35(11):1946–1955.
152.Wentz L, Liu PY, Ilich JZ, Haymes EM. Dietary and training
predictors of stress fractures in female runners. International
Journal of Sport Nutrition and Exercise Metabolism. 2012;22(5):
374–382.
153. Ross ML, Martin DS. Heat and Altitude. In: Deakin V, Burke L, eds.
Clinical Sports Nutrition. 5th ed. McGraw-Hill Education 2015;
767–791.
154. Bergeron MF, Bahr R, Bartsch P, et al. International Olympic
Committee consensus statement on thermoregulatory and altitude
challenges for high-level athletes. British Journal of Sports
Medicine. 2012;46(11):770–779.
155. Meyer NL, Manore MM, Helle C. Nutrition for winter sports.
Journal of Sports Sciences. 2011;29(suppl 1):S127–136.
156. Cheuvront SN, Ely BR, Wilber RL. Environment and Exercise. In:
Maughan RJ, ed. Sports Nutrition, The Encyclopaedia of Sports
Medicine, an IOC Medical Commission Publication, 1st edition.
West Sussex, UK: John Wiley & Sons Ltd.; 2014:425–438.
157. Steinmuller PL, Kruskall LJ, Karpinski CA, Manore MM,
Macedonio MA, Meyer NL. Academy of nutrition and dietetics:
revised 2014 standards of practice and standards of professional
performance for registered dietitian nutritionists (competent, profi-
cient, and expert) in sports nutrition and dietetics. Journal of the
Academy of Nutrition and Dietetics. 2014;114(4):631–641 e643.
http://www.acsm-msse.org568 Official Journal of the American College of Sports Medicine
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
http://www.hc-sc.gc.ca/dhp-mps/prodnatur/legislation/docs/modern-eng.php#a11
http://www.hc-sc.gc.ca/dhp-mps/prodnatur/legislation/docs/modern-eng.php#a11
http://www.ausport.gov.au/ais/nutrition/supplements
http://www.ausport.gov.au/ais/nutrition/supplements
Journal of Visualized Experiments www.jove.com
Copyright © 2012 Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported
License
March 2012 | 61 | e3413 | Page 1 of 5
Video Article
Determining the Contribution of the Energy Systems During Exercise
Guilherme G. Artioli1, Rômulo C. Bertuzzi2, Hamilton Roschel1,3, Sandro H. Mendes1, Antonio H. Lancha Jr.1, Emerson Franchini4
1Laboratory of Applied Nutrition, School of Physical Education and Sport, University of Sao Paulo
2Aerobic Performance Research Group, School of Physical Education and Sport, University of Sao Paulo
3Laboratory of Neuromuscular Adaptations to Strength Training, School of Physical Education and Sport, University of Sao Paulo
4Martial Arts and Combat Sports Research Group, School of Physical Education and Sport, University of Sao Paulo
Correspondence to: Emerson Franchini at emersonfranchini@hotmail.com
URL: http://www.jove.com/video/3413
DOI: doi:10.3791/3413
Keywords: Physiology, Issue 61, aerobic metabolism, anaerobic alactic metabolism, anaerobic lactic metabolism, exercise, athletes, mathematical
model
Date Published: 3/20/2012
Citation: Artioli, G.G., Bertuzzi, R.C., Roschel, H., Mendes, S.H., Lancha, A.H., Franchini, E. Determining the Contribution of the Energy Systems
During Exercise. J. Vis. Exp. (61), e3413, doi:10.3791/3413 (2012).
Abstract
One of the most important aspects of the metabolic demand is the relative contribution of the energy systems to the total energy required for a
given physical activity. Although some sports are relatively easy to be reproduced in a laboratory (e.g., running and cycling), a number of sports
are much more difficult to be reproduced and studied in controlled situations. This method presents how to assess the differential contribution of
the energy systems in sports that are difficult to mimic in controlled laboratory conditions. The concepts shown here can be adapted to virtually
any sport.
The following physiologic variables will be needed: rest oxygen consumption, exercise oxygen consumption, post-exercise oxygen consumption,
rest plasma lactate concentration and post-exercise plasma peak lactate. To calculate the contribution of the aerobic metabolism, you will
need the oxygen consumption at rest and during the exercise. By using the trapezoidal method, calculate the area under the curve of oxygen
consumption during exercise, subtracting the area corresponding to the rest oxygen consumption. To calculate the contribution of the alactic
anaerobic metabolism, the post-exercise oxygen consumption curve has to be adjusted to a mono or a bi-exponential model (chosen by the one
that best fits). Then, use the terms of the fitted equation to calculate anaerobic alactic metabolism, as follows: ATP-CP metabolism = A1 (mL . s-1)
x t1 (s). Finally, to calculate the contribution of the lactic anaerobic system, multiply peak plasma lactate by 3 and by the athlete’s body mass (the
result in mL is then converted to L and into kJ).
The method can be used for both continuous and intermittent exercise. This is a very interesting approach as it can be adapted to exercises
and sports that are difficult to be mimicked in controlled environments. Also, this is the only available method capable of distinguishing the
contribution of three different energy systems. Thus, the method allows the study of sports with great similarity to real situations, providing
desirable ecological validity to the study.
Video Link
The video component of this article can be found at http://www.jove.com/video/3413/
Protocol
Introduction
The energy necessary for sustaining a physical effort comes from two metabolic sources: aerobic and anaerobic metabolism. While the aerobic
metabolism is more efficient than the anaerobic metabolism (i.e., it produces a higher amount of ATP per mol of substrate), producing energy
through anaerobic metabolism can provide a high amount of energy in a very short time period. This may be decisive for any situation that
requires extremely fast movements.
Each sport has specific characteristics in terms of motor skills that confer unique physiologic and metabolic demands for that particular sport.
The most important aspect of the metabolic demand is the relative contribution of the energy systems to the total energy required for the activity.
To determine the specific demand of each sport is crucial for developing optimized training models, nutritional strategies and ergogenic aids that
may maximize athletic performance.
Some sports are relatively easy to be reproduced in a laboratory setting, thus it is possible to create a controlled environment in which athletes
can be evaluated. This is the case of running and cycling, for example. Predictable movements compose these sports and, therefore, they
are easy to be studied. Using some simple equipment, it is possible to mimic quite exactly the same movements that athletes perform in real
http://www.jove.com
http://www.jove.com
http://www.jove.com
mailto:emersonfranchini@hotmail.com
http://www.jove.com/video/3413
http://dx.doi.org/10.3791/3413
http://www.jove.com/video/3413/
Journal of Visualized Experiments www.jove.com
Copyright © 2012 Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported
License
March 2012 | 61 | e3413 | Page 2 of 5
situations, such as training and competitions. Indeed, these sports have been more extensively studied by exercise scientists and benefited with
a more complete and reliable scientific literature.
On the other hand, a number of sports are much more difficult to be reproduced in laboratory. These sports are unpredictable and dependent
on the actions of the partner(s) and opponent(s). This leads to an inability to accurately reproduce the competitive conditions in the lab and an
inability to assess these athletes in field during either training or competition. Maybe because of these problems, they have received much less
attention from the scientists. This is the case of the majority of team sports and many individual sports1.
Considering these aspects, we aimed to describe how to assess the differential contribution of the energy systems in sports that are difficult to
reproduce in controlled laboratory conditions. Becausejudo is a very complex and unpredictable sport, we will use judo as an example. However,
the concepts shown here can be adapted to a number of different sports.
1. Physiological Measurements at Rest
1. Measure the athlete’s body mass before he/she initiates exercising.
2. Before initiating the exercise, collect a small resting blood sample from earlobe or fingertip and keep it on ice until the whole experimental
procedure is finished.
3. Following, place the calibrated portable gas analyser at the most convenient position, which depends on the movements that the athlete will
perform, and record resting or baseline oxygen consumption for five minutes. During the baseline measurement, the athlete has to stay quiet
standing on his/her feet (if the exercise will be performed in a standing position) or sat in the equipment that will be used (if the exercise will
be performed in a cycloergometer or in any similar equipment).
2. Physiological Measurements during Exercise
1. After collecting resting blood sample and resting oxygen consumption, you may ask the athlete to start the specific exercise that you are
studying. The portable gas analyser has to be placed in a position that will no interfere with exercise and that exercise will not damage the
equipment. Continue measuring oxygen consumption throughout the exercise period.
3. Physiological Measurements after Exercise
1. After collecting exercise oxygen consumption data, keep recording oxygen consumption for ten minutes before shutting the equipment down.
Always recalibrate the gas analyser if more than one athlete is being evaluated in the same day.
2. In order to identify the peak plasma lactate after exercise, collect small blood samples immediately after exercise, three, five and seven
minutes after exercise. Keep them on ice until analysis.
4. Blood Samples Processing and Peak Plasma Lactate Determination
1. All blood samples must be placed in microtubes containing a similar volume of a 2% NaF solution (i.e., if you are collecting 25 μL of blood,
place it in 25 μL of 2% NaF).
2. When data collection is finished, separate plasma from erythrocytes by spinning the samples for 5 minutes at 2000 g at 4°C.
3. Plasma lactate can be determined through a variety of methods2,3. In our lab, we use the electrochemical method with the aid of an
automatized lactate analyser (Yellow Springs 1500 Sport, Ohio).
5. Calculations
1. Calculate the net energy generated by the aerobic metabolism by subtracting rest oxygen consumption from exercise oxygen consumption.
Oxygen consumption at rest is obtained by multiplying the average of the last 30 seconds of baseline oxygen consumption by the total
exercise duration time. Then, calculate the area under the curve of exercise oxygen consumption by using the trapezoidal method. Finally,
subtract the resting oxygen consumption from exercise oxygen consumption.
2. The contribution of the anaerobic alactic metabolism (i.e., the ATP-CP pathway) can be considered as the fast component of excess post-
exercise oxygen consumption4-6, as illustrated in Figure 1. Calculate the energy produced by the alactic system by fitting the kinetics of post-
exercise oxygen consumption to a bi- or a monoexponential curve. This can be done with the aid of mathematics' software (e.g. Microcal
Origin version 7.0). Choose by the mono- or bi-exponential curve based on the model that best fits to your data set (i.e., the lowest residue).
Then, use the terms provided by the fitted equation (Equation 1) to calculate alactic contribution according to Equation 2.
http://www.jove.com
http://www.jove.com
http://www.jove.com
Journal of Visualized Experiments www.jove.com
Copyright © 2012 Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported
License
March 2012 | 61 | e3413 | Page 3 of 5
Figure 1. Schematic illustration of a typical oxygen consumption curve obtained at rest, during, and after exercise.
Equation 1:
Equation 2:
where VO2(t) is oxygen uptake at time t, VO2baseline is oxygen uptake at baseline, A is the amplitude, δ is the time delay, τ is a time constant,
and 1 and 2 denote the fast and slow components, respectively.
3. To calculate the contribution of the lactic anaerobic system, it is assumed that 1 mM of lactate above the resting values corresponds to 3 mL
of oxygen consumed per kilogram of body mass7. Thus, calculate delta peak plasma lactate (i.e., peak plasma lactate minus resting plasma
lactate) and multiply it by 3 and by the athlete’s body mass. The obtained value of oxygen in mL is then converted to L and to energy (kJ),
assuming that each 1 L of O2 is equal to 20.92 kJ.
4. Finally, the result obtained by each energy system is summed so you have the total energy expenditure during the activity and the relative
contribution of each of system can be calculated.
6. Representative Results
Figure 2 depicts a representative curve of oxygen consumption at rest, during exercise and after exercise. In the example used here, athletes
performed three different judo techniques (o-uchi-gari, harai-goshi and seoi-nage) for five minutes (one throw every 15 s)8. This is a typical
response to intermittent exercise. After the calculations, we obtained the final results on the contribution of the energy systems during judo
exercises (Table 1).
Additional representative results are displayed in Table 2. In this example, indoor rock climbers of different competitive levels (i.e., recreational
vs. elite) were assessed during a low-difficulty climb route. Individual results for one elite athlete and one recreational athlete are shown (Table
2).
Seoi-nague Harai-goshi O-uchi-gari 
kJ % kJ % kJ %
Anaerobic alactic 46 ± 20 16.3 ± 2.8 43 ± 21 16.1 ± 2.7 36 ± 22 14.6 ± 2.8
Aerobic 223 ± 66 82.2 ± 2.9 211 ± 66 82.3 ± 3.8 196 ± 74 84.0 ± 3.8
Anaerobic lactic 4 ± 2 1.5 ± 0.7 5 ± 5 1.6 ± 1.4 4 ± 4 1.5 ± 1.1
Total 273 ± 86 - 259 ± 91 - 237 ± 99 -
Total (kJ/min) 51.9 ± 8.7 - 49.4 ± 8.9 - 45.3 ± 19.6 -
Table 1. Representative results of total energy expenditure and the contribution of the energy systems during three different judo exercises.
Competitive Level Aerobic (%) Anaerobic Lactic (%) Anaerobic Alactic
(%)
Total (kJ) Total (kJ/s)
Elite 40 8 52 70.4 1.00
Recreational 40 15 45 96.1 1.15
Table 2. Representative individual data of total energy expenditure and the contribution of the energy systems during a low-difficulty climb route.
http://www.jove.com
http://www.jove.com
http://www.jove.com
Journal of Visualized Experiments www.jove.com
Copyright © 2012 Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported
License
March 2012 | 61 | e3413 | Page 4 of 5
Figure 2. Representative results obtained during a 5-minute judo exercise.
Discussion
The method we have shown hare can be used for both continuous and intermittent exercise. The great advantage of the method is that it can
be adapted to exercises and sports that are difficult to be mimicked in controlled laboratory settings. Furthermore, this is the only available
method capable of distinguishing the contribution of three different energy systems. Thus, the method allows the study of sports with great
similarity to real situations, providing desirable ecological validity to the study9. For example, a recent study by Mello et al.10 showed that the
glycolytic contribution in a 2000 m on water rowing race is of only 7%, which means that rowing performance is mainly dependent on the aerobic
metabolism. Similarly, a study by Beneke et al.4 confirmed that the main source of energy during one of the most used anaerobic tests, the
Wingate Anaerobic Test, is the anaerobic metabolism (20% aerobic; 30% alactic and 50% glycolytic). Recent studies by our group have also
characterized the energy contributions of indoor climbing6 and judo8, as reported in this example. Indeed, the knowledge on the energetic
contribution is critical for the development of ergogenic strategies, training organization or even for validating a test.
This method has somelimitations. First, the cost of the equipment is somewhat high, and specialized trained personnel are required. Second,
although most sports can be mimicked with this technique, it is not any type of exercise that can be studied using the portable gas analyzer.
Finally, as plasma lactate does not exactly represent the total lactate produced by skeletal muscle during activity, the results obtained by this
procedure can be considered as an estimative of the metabolic demand during exercise, rather than a precise quantification of the energetic
contribution. Nonetheless, this is the only validated method available11 capable of distinguishing the contribution of the three different energy
systems.
Disclosures
The authors declare they have no conflict of interest regarding this study.
Acknowledgements
We thank to Fabiana Benatti for her kind cooperation in the video. We also thank FAPESP (#2007/51228-0) and CNPq (#300133/2008-1) for the
support to our researches on this area.
References
1. Franchini, E., Del Vecchio, F.B., Matsushigue, K.A., et al. Physiological profiles of elite judo athletes. Sports Med. 41, 147-166 (2011).
2. Bergmeyer, H.U., Bergmeyer, J., & Grassl, M. Methods of enzymatic analysis. Academic Press, New York, (1983).
3. Passonneau, J.V. & Lowry, O.H. Enzymatic Analysis. A Practical Guide. Humana Press, Totowa, New Jersey, (1993).
4. Beneke, R., Pollmann, C., Bleif, I., et al. How anaerobic is the Wingate Anaerobic Test for humans? Eur. J. Appl. Physiol. 87, 388-392 (2002).
5. Beneke, R., Beyer, T., Jachner, C., et al. Energetics of karate kumite. Eur. J. Appl. Physiol. 92, 518-523 (2004).
6. Bertuzzi, R.C.D., Franchini, E., Kokubun, E., et al. Energy system contributions in indoor rock climbing. Eur. J. Appl. Physiol. 101, 293-300
(2007).
http://www.jove.com
http://www.jove.com
http://www.jove.com
Journal of Visualized Experiments www.jove.com
Copyright © 2012 Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported
License
March 2012 | 61 | e3413 | Page 5 of 5
7. di Prampero, P.E. & Ferretti, G. The energetics of anaerobic muscle metabolism: a reappraisal of older and recent concepts. Respir. Physiol.
118, 103-115 (1999).
8. Franchini, E., Bertuzzi, R.C.D., Degaki, E., et al. Energy Expenditure in Different Judo Throwing Techniques. Proceedings of first joint
international pre-Olympic conference of sports science and sports engineering, vol II. Bio-mechanics and sports engineering., 55-60 (2008).
9. Calmet, M. Developing ecological research in judo. Percept. Mot. Skills. 105, 646-648 (2007).
10. Mello, F.D., Bertuzzi, R.C., Grangeiro, P.M., et al. Energy systems contributions in 2,000 m race simulation: a comparison among rowing
ergometers and water. Eur. J. Appl. Physiol. 107, 615-619 (2009).
11. Bertuzzi, R.C., Franchini, E., Ugrinowitsch, C., et al. Predicting MAOD using only a supramaximal exhaustive test. Int. J. Sports Med. 31,
477-481 (2010).
http://www.jove.com
http://www.jove.com
http://www.jove.com
nutrients
Review
Caffeine Supplementation and Physical Performance,
Muscle Damage and Perception of Fatigue in Soccer
Players: A Systematic Review
Juan Mielgo-Ayuso 1,* , Julio Calleja-Gonzalez 2, Juan Del Coso 3 , Aritz Urdampilleta 4,
Patxi León-Guereño 5 and Diego Fernández-Lázaro 6
1 Department of Biochemistry and Physiology, School of Physical Therapy, University of Valladolid,
42004 Soria, Spain
2 Laboratory of Human Performance, Department of Physical Education and Sport, Faculty of Physical
Activity and Sport, University of the Basque Country, 01007 Vitoria, Spain; julio.calleja.gonzalez@gmail.com
3 Exercise Physiology Laboratory, Camilo José Cela University, 28692 Madrid, Spain; jdelcoso@ucjc.edu
4 Elikaesport, Nutrition, Innovation & Sport, 08290 Barcelona, Spain; a.urdampilleta@drurdampilleta.com
5 Faculty of Psychology and Education, University of Deusto, Campus of Donostia-San Sebastián,
20012 San Sebastián, Guipúzcoa, Spain; patxi.leon@deusto.es
6 Department of Cellular Biology, Histology and Pharmacology. Faculty of Physical Therapy,
University of Valladolid. Campus de Soria, 42004 Soria, Spain; diego.fernandez.lazaro@uva.es
* Correspondence: juanfrancisco.mielgo@uva.es; Tel.: +34-975-129-187
Received: 21 January 2019; Accepted: 15 February 2019; Published: 20 February 2019
����������
�������
Abstract: Soccer is a complex team sport and success in this discipline depends on different factors
such as physical fitness, player technique and team tactics, among others. In the last few years,
several studies have described the impact of caffeine intake on soccer physical performance, but the
results of these investigations have not been properly reviewed and summarized. The main objective
of this review was to evaluate critically the effectiveness of a moderate dose of caffeine on soccer
physical performance. A structured search was carried out following the Preferred Reporting Items
for Systematic Review and Meta-Analyses (PRISMA) guidelines in the Medline/PubMed and Web of
Science databases from January 2007 to November 2018. The search included studies with a cross-over
and randomized experimental design in which the intake of caffeine (either from caffeinated drinks
or pills) was compared to an identical placebo situation. There were no filters applied to the soccer
players’ level, gender or age. This review included 17 articles that investigated the effects of caffeine
on soccer-specific abilities (n = 12) or on muscle damage (n = 5). The review concluded that 5
investigations (100% of the number of investigations on this topic) had found ergogenic effects of
caffeine on jump performance, 4 (100%) on repeated sprint ability and 2 (100%) on running distance
during a simulated soccer game. However, only 1 investigation (25%) found as an effect of caffeine to
increase serum markers of muscle damage, while no investigation reported an effect of caffeine to
reduce perceived fatigue after soccer practice. In conclusion, a single and moderate dose of caffeine,
ingested 5–60 min before a soccer practice, might produce valuable improvements in certain abilities
related to enhanced soccer physical performance. However, caffeine does not seem to cause increased
markers of muscle damage or changes in perceived exertion during soccer practice.
Keywords: football; RPE; DOMS; sport performance; supplementation; ergogenic aids
1. Introduction
Soccer is considered one of the most popular sports worldwide. According to the Fédération
Internationale de Football Association (FIFA) Big Count survey, there are 265 million active soccer
Nutrients 2019, 11, 440; doi:10.3390/nu11020440 www.mdpi.com/journal/nutrients
http://www.mdpi.com/journal/nutrients
http://www.mdpi.com
https://orcid.org/0000-0002-6554-4602
https://orcid.org/0000-0002-5785-984X
http://www.mdpi.com/2072-6643/11/2/440?type=check_update&version=1
http://dx.doi.org/10.3390/nu11020440
http://www.mdpi.com/journal/nutrients
Nutrients 2019, 11, 440 2 of 15
players and the number is progressively increasing, especially in women’s football [1]. In addition,
soccer attracts millions of television spectators while the socio-economic impact of elite soccer affects
almost every culture worldwide [2]. Thus, the study of soccer and the variables that affect performance
in this complex team sport can have a great impact on sport sciences. Briefly, modern soccer is
characterized by the continuous combination of short sprints, rapid accelerations/decelerations and
changes of direction interspersed with jumping, kicking, tackling and informal times for recovery [3].
In addition to these physical fitness variables, players’ techniques and cognitive capacity, team tactics,
and psychological factors might also have an impact on overall soccer performance [4,5]. Unlike other
team sports, such as basketball or handball, soccer is a low-scoring game and, thus, the margins of
victory are close/reduced, particularly at the elite level. In consequence, the study of the effects of
ergogenic aidson performance have become an important subject for players, coaches and sport
scientists associated with soccer because it has the potential to increase success in the game [6].
Caffeine (1, 3, 7-trimethylxanthine) is one of the most popular supplements among athletes for
its potent stimulant effects and due to its easy availability in the market in different commercial
forms (energy drinks, caffeinated beverages, pills, pre-workout and thermogenic supplements, etc.).
In addition, the ergogenic effects of the acute ingestion of caffeine have been widely reported on
different forms of exercise, although most of the classic studies focused on endurance performance [7,8].
In the last few years, several investigations have found that caffeine can also increase anaerobic and
sprint performance, although the direct application of these research outcomes to the complexity
of soccer is complicated [9–11]. According to the Australian Institute of Sport (AIS), the potential
ergogenicity of caffeine reflects level 1 evidence, which allocates it as a safe supplement to use
in sport [12]. In addition, the International Olympic Committee indicates, in its recent consensus
statement for dietary supplements, that caffeine intake results in performance gains when ingested
before exercise in doses ranging from 3 to 6 mg/kg. Finally, two recent systematic reviews have
concluded that caffeine might be ergogenic in team sport athletes [6,13]. With this background, one
might suppose that caffeine is also ergogenic in soccer although the information regarding this sport
has not been summarized. In the last few years, several studies have investigated the effects of
caffeine intake on soccer physical performance [14–21] and in the opinion of the authors, the results of
these investigations need to be objectively reviewed and summarized. Therefore, the objective of this
systematic review was to critically evaluate the effectiveness of a moderate dose of caffeine on soccer
physical performance, muscle damage and perception of fatigue in order to provide more objective
and comprehensive information about the positive and negative impact of caffeine on soccer players.
2. Methods
2.1. Search Strategies
The present article is a systematic review focusing on the impact of caffeine intake on soccer
physical performance and it was conducted following the Preferred Reporting Items for Systematic
Review and Meta-Analyses (PRISMA) guidelines and the PICOS model for the definition of the
inclusion criteria: P (Population): “soccer players”, I (Intervention): “impact of caffeine on soccer
physical performance, muscle damage and perception of fatigue”, C (Comparators): “same conditions
with placebo”, O (Outcome): “soccer-specific abilities, serum markers of muscle damage and perceived
fatigue (RPE) and heart rate”, and S (study design): “double-blind and randomized cross-over
design” [22].
A structured search was carried out in the Medline (PubMed) database and in the Web of
Science (WOS) which includes other databases such as BCI, BIOSIS, CCC, DIIDW, INSPEC, KJD,
MEDLINE, RSCI, SCIELO, both high quality databases which guarantee good bibliographic support.
The search covered from July 2006, when Hespel et al., [23] suggested the use of caffeine as an effective
supplement for soccer athletic performance, to November 2018. Search terms included a mix of
Medical Subject Headings (MeSH) and free-text words for key concepts related to caffeine and soccer
Nutrients 2019, 11, 440 3 of 15
physical performance, muscle damage or perceived fatigue as follows: (“football”(All Fields) OR
“soccer”(All Fields)) AND (“caffeine”(All Fields) OR “energy drink”(All Fields)) AND ((“physical
performance”(All Fields) OR performance(All Fields))) OR ((“muscles”(MeSH Terms) OR “muscles”
(All Fields) OR “muscle” (All Fields))) OR damage(All Fields) OR (RPE(All Fields) OR “perceived
fatigue “(All Fields)). Through this search, relevant articles in the field were obtained applying the
snowball strategy. All titles and abstracts from the search were cross-referenced to identify duplicates
and any potential missing studies. Titles and abstracts were then screened for a subsequent full-text
review. The search for published studies was independently performed by two authors (JMA and JCG)
and disagreements about physical parameters were resolved through discussion.
2.2. Inclusion and Exclusion Criteria
For the articles obtained in the search, the following inclusion criteria were applied to select
studies: articles (1) depicting a well-designed experiment that included the ingestion of an acute dose
of caffeine—or a caffeine-containing product—before and/or during exercise in humans; (2) with
an identical experimental situation related to the ingestion of a placebo performed on a different
day; (3) testing the effects of caffeine on soccer-specific tests and/or real or simulated matches;
(4) with a double-blind, and randomized cross-over design; (5) with clear information regarding the
administration of caffeine (relative dose of caffeine per kg of body mass and/or absolute dose of
caffeine with information about body mass; timing of caffeine intake before the onset of performance
measurements, etc.); (6) where caffeine was administered in the form of a beverage, coffee gum or pills;
(7) on soccer players with previous training backgrounds in soccer; (8) the languages were restricted
to English, German, French, Italian, Spanish and Portuguese. The following exclusion criteria were
applied to the experimental protocols of the investigation: (1) the use of caffeine doses below 1 mg/kg
or above 9 mg/kg; (2) the absence of a true placebo condition; (3) the absence of pre-experimental
standardizations such as elimination of dietary sources of caffeine 24 h before testing; (4) carried
out in participants with a previous condition or injury. There were no filters applied to the soccer
players’ level, sex or age to increase the power of the analysis. Moreover, the Physiotherapy Evidence
Database scale (PEDro), the key factors of which assess eligibility criteria, random allocation, baseline
values, success of the blinding procedures, power of the key outcomes, correct statistical analysis
and measurement of participants’ distribution of studies, was used to evaluate whether the selected
randomized controlled trials were scientifically sound: 9–10 = excellent, 6–8 = good, 4–5 = fair, and
on soccer physical performance. The topics
and number of studies that were excluded were: 1 because of lack of information on body mass,
1 because the caffeine content was found in nutritional supplements with other drugs, 1 because it
was a suggestion for future research, 1 because the sport investigated was not specified, 4 because
they dealt with recovery or sleep processes, 4 because they investigated other team sports (1 rugby,
1 volleyball, 1 tennis, 1 Gaelic football), and the remaining 11 articles because they studied other
subjects unrelated to the focus of this systematic review. Thus, the current systematic review includes
17 studies.
Nutrients 2019, 11, Firstpage-Lastpage FOR PEER REVIEW 4 of 15 
of dates included in the inclusion criteria. From the remaining 40 articles, another 23 papers were 
removed because they were unrelated to the effects of caffeine on soccer physical performance. The 
topics and number of studies that were excluded were: 1 because of lack of information on body 
mass, 1 because the caffeine content was found in nutritional supplements with other drugs, 1 
because it was a suggestion for future research, 1 because the sport investigated was not specified, 4 
because they dealt with recovery or sleep processes, 4 because they investigated other team sports (1 
rugby, 1 volleyball, 1 tennis, 1 Gaelic football), and the remaining 11 articles because they studied 
other subjects unrelated to the focus of this systematic review. Thus, the current systematic review 
includes 17 studies. 
 
Figure 1. Selection of studies. 
3.2. Caffeine Supplementation 
The participants’ samples included players of both genders (241 males and 33 females), who 
competed in professional or elite (n = 108), semi-professional (n = 19) and amateur teams (n = 147). In 
addition, 70 players were adolescents. Out of the 17 investigations, only 2 studies included female 
soccer players. In 12 out of 17 studies, caffeine was administered based on the soccer player’s body 
mass, while an absolute dose was provided for all participants in 5 studies. In 2 studies the caffeine 
dose employed was less than 3 mg/kg, 3 studies used a caffeine dose of around 3 mg/kg, in 2 studies 
it was 4.5 mg/kg, in 3 studies it was around 5 mg/kg, in 4 studies the dose was 6 mg/kg and 2 studies 
included a dose above 6 mg/kg (i.e., 7.2 mg/kg). In 1 study, soccer players took different doses (1, 2 
 
Records identified through 
database searching 
(n = 135) 
Sc
re
en
in
g 
In
cl
u
d
ed
 
El
ig
ib
ili
ty
 
Id
en
ti
fi
ca
ti
o
n
 
Additional records identified 
through other sources 
(n = 0) 
Records after duplicates removed 
(n = 61) 
Records screened 
(n = 61) 
Records excluded 
(n = 21) 
Full-text articles assessed 
for eligibility 
(n = 40) 
Full-text articles excluded, 
with reasons 
(n = 23) 
 
Unsuitable Outcomes = 1 
Unsuitable methodology = 3 
Other sports = 4 
Subjects unrelated =15 
Studies included in 
qualitative synthesis 
(n = 17) 
Figure 1. Selection of studies.
3.2. Caffeine Supplementation
The participants’ samples included players of both genders (241 males and 33 females), who
competed in professional or elite (n = 108), semi-professional (n = 19) and amateur teams (n = 147).
In addition, 70 players were adolescents. Out of the 17 investigations, only 2 studies included female
soccer players. In 12 out of 17 studies, caffeine was administered based on the soccer player’s body
mass, while an absolute dose was provided for all participants in 5 studies. In 2 studies the caffeine
dose employed was less than 3 mg/kg, 3 studies used a caffeine dose of around 3 mg/kg, in 2 studies
it was 4.5 mg/kg, in 3 studies it was around 5 mg/kg, in 4 studies the dose was 6 mg/kg and 2
studies included a dose above 6 mg/kg (i.e., 7.2 mg/kg). In 1 study, soccer players took different
doses (1, 2 and 3 mg/kg). Regarding the form of administration, 9 investigations used capsules filled
with caffeine, 3 investigations used caffeinated energy drinks, 3 investigations used a caffeinated sport
Nutrients 2019, 11, 440 5 of 15
drink, 1 investigation employed a 20% carbohydrate solution and 1 investigation employed caffeinated
chewing gum.
Most investigations administered caffeine 30–60 min prior to testing, with the exception of the
studies conducted by Andrade-Souza et al. (2015) where the consumption of caffeine was carried
out 3 h after a practice session, 4 h after its effects were evaluated in a simulated match [25]. Also,
Guttierres et al. (2013) used a protocol that included the ingestion of caffeine 1 h before the test and
every 15 min during the protocol [26]. Finally, Ranchordas et al. (2018) employed caffeine 5 min before
the tests because they used caffeinated gums [17]. In summary, different studies examined the effect of
caffeine on soccer physical performance by using a variety of times of ingestion prior to the testing
(5 min–60 min).
3.3. Outcome Measures
Tables 1–3 include information about author/s and year of publication; the sample investigated,
with details of sport level, sex and the number of participants; the study design cites the control
group if the study included one; the supplementation protocol that specifies the type of caffeine used,
the dose and the time that it was administered; the parameters analyzed or main effects either on
sport performance (n = 12; Table 1) and muscle damage (n = 5; Table 2) and finally results or main
conclusions. Additionally, some studies also presented data on the effects of caffeine on perceived
exertion and heart rate (n = 6; Table 3).
Nutrients 2019, 11, 440 6 of 15
Table 1. Summary of studies included in the systematic review that investigated the effect of caffeine ingestion as compared to a placebo on soccer-specific abilities.
Author/s Population Intervention Outcomes Analyzed Main Conclusion
Ellis M. et al.
2018 [21]
15 male elite youth players
(16 ± 1 years)
1, 2 or 3 mg/kg of caffeine capsules
60 min before the start
20-m sprint
Arrowhead agility
CMJ
Yo-Yo IR1
↑ 20-m sprint
↑ Arrowhead agility
↑ CMJ
↑ Yo-Yo IR1
Apostolidis A. et al.
2018 [19]
20 well-trained male players
High (n = 11) and low (n = 9) responders
(21.5 ± 4 years)
6 mg/kg of caffeine capsules
60 min before the start
CMJ
Reaction time
Time to fatigue
↑ CMJ
† Reaction time
↑ Time to fatigue
Guerra MA Jr. et al.
2018 [18]
12 male professional players
(23 ± 5 years)
5 mg/kg of caffeine + 20% carbohydrate solution
60 min before the start
CMJ at 1, 3 and 5 min after the
conditioning stimulus ↑ CMJ
Ranchordas et al.,
2018 [17]
10 male university-standard players
(19 ± 1 years)
200 mg (≈2.7 g/kg) of caffeinated gum
5 min before the start
20-m sprint
CMJ
Yo-Yo IR1
† 20-m sprint
↑ CMJ
↑ Yo-Yo IR1
Andrade Souza, V. et al.
2015 [25]
11 male amateur players
(25.4 ± 2.3 years)
6 mg/kg of caffeine capsules
3 h after the LIST
30-m Repeated-Sprint test
CMJ
LSPT
† 30-m Repeated-sprint test
† CMJ
† LSPT
Jordan, J.et al.
2014 [16]
17 male elite young players
(14.1 ± 0.5 years)
6 mg/kg of caffeine capsules
60 min before the start
Sprint time
Reaction time
† Sprint time
↑ Reaction time on
non-dominant leg
Lara, B. et al.
2014 [27]
18 female semi-professional players
(21 ± 2 years)
3 mg/kg of caffeinated energy drinks
60 min before the start
Height and power of jump
Average speed of running
Total distance covered
Number of sprints
↑ Height and power of jump
↑ Average speed of running
↑ Total distance covered
↑Number of sprints
Astorino, T. et al.
2012 [20]
15 female collegiate players
(19.5 ± 1.1 years)
255 mL (≈1.3 mg/kg) of caffeinated energy drinks (Redbull)
60 min before the start Sprint time † Sprint time
Del Coso, J. et al.
2012 [16]
19 male semi-professional players
(21 ± 2 years)
3 mg/kg of caffeine in energy drink
60 min before the start
Maximum height jump
Maximum running speed
Distance covered
Caffeine concentration in urine
↑Maximum height jump
↑Maximum running speed
↑Distance covered↑Caffeine concentrations in
urine
Gant, N. et al.
2010 [28]
15 male first team level players
(21.3 ± 3 years)
160 mg/L (≈3.7 mg/kg) of caffeinated sport drinks
60 min before the start and every 15 min during the test
Sprint times
Jump power
Test of passes
Blood lactate
Post-exercise caffeine in urine
↑ Sprint times
↑ Jump power
† Test of passes
† Blood lactate
↑ Post-exercise caffeine in urine
Foskett et al.
2009 [15]
12 male professional players
(23.8 ± 4.5 years)
6 mg/kg of caffeine capsules
60 min before the start
LSPT
CMJ
↑ LSPT
↑CMJ
Guttierres, A. P. et al.
2009 [29]
18 male junior players
(16.1 ± 0.7 years)
250 mg/L (≈7.2 mg/kg) of caffeinated sport drinks
20 min before and every 15 min during the test
Jump height
Illinois agility test
↑ Jump height
† Illinois agility test
↑: statistically significant increase; † change with no statistical significance; ↓: statistically significant decrease. CMJ: countermovement jump; LIST: Loughborough Intermittent Shuttle Test;
LSPT: Loughborough Soccer Passing Test; Yo-Yo IR1: Yo-Yo intermittent recovery test level-1
Nutrients 2019, 11, 440 7 of 15
Table 2. Summary of studies included in the systematic review that investigated the effect of caffeine ingestion as compared to a placebo on serum markers of
muscle damage.
Author/s Population Intervention Outcomes Analyzed Main Conclusion
Guttierres, A. P. et al.
2013 [26]
20 male young players
(16.1 ± 0.7 years)
7.2 mg/kg of caffeinated sport drinks
20 min before and every 15 min during
the test
Blood glucose
Blood lactate
Plasma caffeine
Free fatty acids
Urine caffeine
↑ Blood glucose
↑ Blood lactate
↑ Plasma caffeine
† Free fatty acids
† Urine caffeine
Machado, M. et al.
2010 [30]
15 male players
(18.4 ± 0.8 years)
4.5mg/kg of caffeine capsules
Immediately before the test
CK
LDH
ALT
AST
basophils, eosinophils, neutrophils,
monocyte lymphocytes
† CK
† LDH
† ALT
† AST
† basophils, eosinophils, neutrophils,
monocyte lymphocytes
Machado, M. et al.
2009 [31]
20 male players
(18.8 ± 1 years)
4.5 mg/kg of caffeine capsules
Immediately before the test
Basic hemogram
CK
LDH
ALT
AST
AP
γ-GT
† Basic hemogram
† CK
† LDH
† ALT
† AST
† AP
† γ-GT
Machado, M. et al.
2009 [32]
15 male professional players
(19 ± 1 years)
5.5 mg/kg of caffeine capsules
Immediately before the test
CK
LDH
ALT
AST
† CK
† LDH
† ALT
† AST
Bassini-Cameron, A. et al.
2007 [33]
22 male professional players
(26.0 ± 1.6 years)
5 mg/kg of caffeine capsules
60 min before the start
CK
LDH
ALT
AST
↑ CK
† LDH
↑ALT
† AST
↑: statistically significant increase; † change with no statistical significance; ↓: statistically significant decrease. CK: creatine kinase; LDH: lactate dehydrogenase; ALT: alanine
aminotransferase; AST: aspartate aminotransferase; AP: alkaline phosphorylase; γ-GT: γ-glutamyl transferase.
Nutrients 2019, 11, 440 8 of 15
Table 3. Summary of studies included in the systematic review that investigated the effect of caffeine ingestion as compared to a placebo on perceived fatigue and
heart rate.
Author/s Population Intervention Outcomes Analyzed Main Conclusion
Andrade Souza, V. et al.
2015 [25]
11 male amateur players
(25.4 ± 2.3 years)
6 mg/kg of caffeine capsules
3 h after the LIST Perceived effort † Perceived effort
Jordan, J.et al.
2014 [16]
17 male elite young players
(14.1 ± 0.5 years)
6 mg/kg of caffeine capsules
60 min before the start Heart rate † Heart rate
Lara, B. et al.
2014 [27]
18 female semi-professional players
(21 ± 2 years)
3 mg/kg of caffeinated energy drinks
60 min before the start Heart rate † Heart rate
Guttierres, A. P. et al.
2013 [26]
20 male young players
(16.1 ± 0.7 years)
7.2 mg/kg of caffeinated sport drinks
20 min before and every 15 min during the test Perceived effort † Perceived effort
Astorino, T. et al.
2012 [20]
15 female collegiate players
(19.5 ± 1.1 years)
255 mL (≈1.3 mg/kg) of caffeinated energy
drinks (Redbull) 60 min before the start
Perceived effort
Heart rate
† Perceived effort
† Heart rate
Foskett et al.
2009 [15]
12 male professional players
(23.8 ± 4.5 years)
6 mg/kg of caffeine capsules
60 min before the start Heart rate † Heart rate
↑: statistically significant increase; † change with no statistical significance; ↓: statistically significant decrease.
Nutrients 2019, 11, 440 9 of 15
4. Discussion
The purpose of this systematic review was to summarize all scientific evidence for the effect of
acute caffeine ingestion on variables related to soccer physical performance. Due to the differences
of the effects studied among the investigations included in the analysis, the following variables have
been clustered for a more comprehensive scrutiny.
4.1. Impact on Sports Performance
A total of 12 investigations carried out research protocols that studied the effects of caffeine on
one or more variables related to soccer-specific abilities. Overall, these investigations showed an
improvement in soccer-related skills with the pre-exercise ingestion of caffeine (Table 1). Specifically,
Foskett et al., [15], with 12 first division football players (age: 23.8 ± 4.5 years), observed that
the consumption of 6 mg/kg of caffeine before exercise increased passing accuracy and accrued
significantly less penalty time during two validated tests to assess soccer skill performance (intermittent
shuttle-running protocol and Loughborough Soccer Passing Test; LSPT). In addition, this investigation
also found that caffeine improved the functional power of the leg measured by a vertical jump. In the
study conducted by Jordan et al., 17 soccer players from the elite youth category (age: 14.1 ± 0.5 years)
performed an agility test (reactive agility test) validated for football [34]. These authors indicated,
based on the results of their investigation, the intake of 6 mg/kg of caffeine 60 min before the test
significantly improved the reaction time of the players in their non-dominant leg [16]. In another study
conducted with 15 elite young players (age: 16 ± 1 years) that were administered low doses of caffeine
(1, 2 and 3 mg/kg), Ellis et al., [21] observed that improvements in physical performance depended on
the dose and the type of task. Specifically, they concluded that 3 mg/kg of caffeine seems to be the
optimal dose to obtain positive effects on soccer-specific tests (20 m sprint, arrowhead agility and CMJ).
However, the authors also suggested that even higher doses of caffeine might be required to improve
endurance performance, as measured by the Yo-Yo intermittent recovery test level 1 (Yo-Yo IR1).
In this line, Apostolidis et al., [19] showed that 6 mg/kg of caffeine ingested 60 min previous to a
battery of tests improved aerobic endurance (time to fatigue) and neuromuscular performance (CMJ)
in 20 well-trained soccer players (age: 21.5 ± 4 years). Since these authors did not find any change in
substrate oxidation with caffeine, measured by indirect calorimetry during the testing, they commented
that performance improvements could only be attributed to positive effects on the central nervous
system and/or neuromuscular function, although the precise mechanism of caffeine ergogenicity was
not indicated in this investigation. Finally, Guerra et al., [18] investigated the addition of caffeine
(5 mg/kg) to a post-activation potentiation protocol that included plyometrics and sled towing. These
authors found that, in a group of 12 male professional soccer players (age: 23 ± 5 years), caffeine
augmented the effects of the post-activation potentiation, as measured by CMJ. These investigations,
taken together, suggest that caffeine might be effective to improve performance in players’ abilities
and soccer-specific skills (jumps, sprint, agility, aerobic endurance, accuracy of passes and ball control).
Caffeinated energy drinks are considered as one of the most common ways to provide caffeine
before exercise [14], and the effect of this type of beverages have been also investigated in soccer players.
Del Coso et al., [16] chose19 semi-professional players (age: 21 ± 2 years) in order to determine if
the caffeine, provided via a commercially-available energy drink (3 mg/kg), improved performance
during several soccer-specific tests (single and repeated jump tests and repeated sprint ability test)
and during a simulated soccer match. For this investigation, players ingested either an energy drink
without sugar but with caffeine (i.e., sugar-free Redbull), or a sugar-free soda (Pepsi diet without
caffeine) 60 min prior to testing. The results showed that the consumption of the caffeinated energy
drink increased the ability to jump, to repeat sprints, and it affected positively total running distance
and the running distance at >13 km/h covered during the simulated game. In another similar study
carried out with 18 semi-professional women soccer players (age: 21 ± 2 years), Lara et al., [34]
demonstrated that the consumption of an energy drink containing 3 mg/kg of caffeine improved
jump height, the ability to perform sprints, the total running distance and the distance covered at high
Nutrients 2019, 11, 440 10 of 15
running intensity (i.e., >18 km/h). These two investigations together suggest that caffeine in the form
of an energy drink, can improve the physical demands associated with high performance in soccer,
such as sprints, rapid accelerations/decelerations and constant changes of direction [35,36].
Thanks to the collaboration of 18 junior soccer players (age: 16.1 ± 0.7 years), Guttierres et al., [29]
evaluated whether the physical performance of these players increased with the consumption of
a caffeinated sports drink (250 mg/L ≈ 7.2 mg/kg) compared to a commercial carbonated drink.
They concluded that the caffeine-based sport drink significantly increased jump height and improved
the power in the lower limbs, another determinant factor in soccer physical performance [37]. However,
positive effects were not demonstrated in the “Illinois Agility Test”, a validated routine to assess agility
in team sports players [38]. This lack of positive effects was possibly due to the fact that players’ agility
is a complex process that depends on the coordination of factors such as decision making and speed in
the changes of direction, both aspects of which are trained and constantly improved in training [39].
Gant et al., [28] used a caffeine-containing carbonated drink in 15 professional soccer players (age: 21.3
± 3 years) and were able to evidence that the addition of caffeine to a carbonated drink, in a dose of
3.7 mg/kg (60 min before starting and every 15 min during the test), improved sprint performance
and the vertical jump in soccer players. Andrade-Souza et al., [25] proposed a very interesting study
where the aim was to investigate the effect of a carbohydrate-based drink, with (6 mg/kg) and without
caffeine, and they compared these trials to the isolated ingestion of caffeine. The study was carried
out with 11 college football players (age: 25.4 ± 2.3 years) with the ingestion of the drinks after the
morning training session to see the effects of the 20% carbohydrate solutions on the afternoon training
session of the same day. The main finding of Andrade-Souza’s study was that none of the drinks was
able to increase performance. This fact could be related to reduction of the glycogen levels from the
morning to the afternoon training sessions, a factor that was either not considered or measured in
the study. According to Jacobs [40], a reduction in the content of muscle glycogen below a critical
threshold can affect anaerobic strength and performance. In the Andrade-Souza’s study, the recovery
time between the two practice sessions was very short (4 h) and the nutritional strategies chosen were
not optimal to replenish muscle glycogen [25]. Thus, it is likely that the reduction of glycogen stores
might have precluded the ergogenic effect of caffeine in this experimental design.
Chewing gum provides an alternative mode of caffeine administration that is more rapidly
absorbed (via the buccal mucosa) than capsules and drinks (i.e., 5 min vs. 45 min, respectively) and
less likely to cause gastrointestinal distress [41]. Along these lines, Ranchordas et al., reported that
caffeinated chewing gum, containing 200 mg of caffeine, can enhance aerobic capacity (Yo-Yo IR1) by
2% and increase CMJ performance by 2.2% in 10 male university soccer players (age: 19 ± 1 years) [17].
Therefore, chewing gum could be beneficial for soccer players where the time between ingestion and
performance is short, (e.g., for substitutes that would come on when called upon by the coach and
for players who cannot tolerate caffeinated beverages or capsules because of gastrointestinal distress
before kick-off [17]).
4.2. Impact on Muscle Damage
A total of 5 investigations studied the effect of caffeine on the levels of muscle damage after
a soccer practice (Table 2). In soccer, muscle damage is a very important physiological variable
because all the high-intensity exercises produced in this sport (sprints, accelerations/decelerations,
changes of directions and even tackles) may be associated with myofibril damage [42]. In this way,
Machado et al. carried out three studies on this specific topic: one with 20 healthy soccer players
(age: 18.8 ± 1 years) [31]; another one with 15 male soccer players (age: 19 ± 1 years) [32] and the last
one with 15 male soccer athletes (age: 18.4± 0.8 years) [30]. The main goal of these three investigations
was to determine whether consumption of caffeine in single and acute doses (4.5–5.5 mg/kg) negatively
affected blood markers typically used to assess the level of muscle damage. The authors concluded
that these markers (creatine kinase, lactate dehydrogenase, aspartate aminotransferase and alanine
aminotransferase) increased with exercise, but they did not find that this increase was exacerbated
Nutrients 2019, 11, 440 11 of 15
with the consumption of caffeine. On the other hand, a study conducted by Bassini-Cameron et al., [33]
which measured hematological variables, muscle proteins and liver enzymes in 22 professional soccer
players (age: 26.0 ± 1.6 years), concluded that 5 mg/kg of caffeine ingested 60 min before the start of a
game increased the risk of muscle damage in players because there was an increase in the white blood
cell count. However, the serum concentration of white blood cells is not a definitive marker of muscle
damage level during exercise, as other factors can increase the count of leukocytes during exercise.
In summary, although most of studies found an absence of effect of caffeine on exercise-induced muscle
damage in soccer players, further research is necessary to confirm this notion [43].
Guttierres et al., [26], in an experiment with 20 youth soccer players (age: 16.1 ± 0.7 years),
observed the effect of caffeine (contained in a sport drinks) on free fatty acids mobilization.
Participants consumed the beverage 20 min before a soccer match and every 15 min during the
game with a total caffeine consumption of 7.2 mg/kg. The authors certified that the caffeine did
not increase the mobilization of free fatty acids. Anyway, the caffeinated beverage was also rich
in carbohydrates, increasing blood glucose concentration, promoting more insulin and therefore
inhibiting the mobilization of fatty acids. Graham, et al., [44] showed that caffeine with glucose
increased insulin secretion versus glucose consumption alone. In the Guttierres study, blood
glucose concentrations were higher possibly due to an increase in sympathetic nervous system
activity [26], increasing adrenaline and noradrenaline and glycogenolysis [45,46]. Moreover, blood
lactate concentrations increased with the ingestion of caffeine, likely indicating, in an indirect manner,
that players achieved a greater intensity in the trial with caffeine. These outcomes could suggest that
soccer players exercised at a higher percentage of their maximum heart rate (caffeine: 80.6 % versus
control: 74.7% of maximum heart rate) withthe ingestion of caffeine.
4.3. Impact on the Perception of Fatigue
Astorino et al., [20] concluded that the consumption of a serving portion of an energy drink
(Redbull), with 1.3 mg/kg of caffeine, did not alter the perception of fatigue or heart rate in 15
semi-professional soccer players (age: 19.5 ± 1.1 years; Table 3). An important limitation of this study
was that the amount of caffeine consumed was very low (80 mg) and thus, the effects of caffeine
would have been limited due to dosage. On the contrary, Guttierres et al., [26] and Lara et al., [28]
found that caffeine tended to cause an increase in exercise heart rate with a concomitant reduction in
the perception of fatigue. While caffeine has been proven to reduce perception of fatigue in exercise
protocols that used a fixed exercise intensity [47], these investigations [20,26,28] used exercise protocols
more applicable to soccer, where exercise intensity can be freely chosen, as happens during soccer
play. Under these specific conditions, caffeine served to increase exercise intensity (as indicated by
higher running distances and higher average heart rate) while perception of fatigue was unaffected or
tended to be reduced. Thus, it might be speculated that caffeine might have the capacity to enhance
soccer physical performance without producing higher values of fatigue, which can be understood as
a positive property of this stimulant.
4.4. Caffeine Dose and Inter-Individual Responses to Caffeine Administration
The dose of caffeine administered in the experiments ranged from 1.3 to ~7.2 mg/kg and thus,
the ergogenic effects of caffeine on soccer players must be attributed to this range of dosage. However,
it is still possible that dose-response effects exist or even that a caffeine dose threshold is necessary
to obtain benefits from caffeine in soccer, as recently suggested by Chia et al., for ball sports [6].
Based on previous investigations with other different forms of exercise [48–50], or soccer [20,21], it can
be suggested that doses below 2 mg of caffeine per kg of body mass might not be effective to increase
soccer physical performance. All the experiments included in this systematic review reported their
findings as a group mean comparing caffeine vs. placebo trials. Nevertheless, recent investigations
have shown that not all individuals experience enhanced physical performance after the ingestion
of moderate doses of caffeine [51–54]. These studies have identified the presence of athletes who
Nutrients 2019, 11, 440 12 of 15
obtain minimal ergogenic effects or only slight ergolytic effects after acute administration of caffeine,
and such participants have been catalogued as “non-responders to caffeine” [55]. To date, there is
still no clear explanation for the lack of ergogenic effects after the acute administration of caffeine in
some individuals, although factors such as training status, habitual daily caffeine intake, tolerance to
caffeine, and genotype variation have been proposed as possible modifying factors for the ergogenicity
of caffeine [55]. Whilst this systematic review suggests that the ingestion of 3–6 mg/kg of caffeine is
ergogenic for soccer players, it might not be optimal for everyone. The inter-individual variability in
the ergogenic response to acute caffeine ingestion suggests that caffeine should be recommended in a
customized manner. The development of more precise and individualized guidelines would seem
necessary for soccer players.
4.5. Strengths, Limitations and Future Lines of Research
The current systematic review presents some limitations related to the different research protocols
and performance tests used in the investigations included. Although we selected investigations
in which caffeine was compared to an identical situation without caffeine administration, in some
investigations, caffeine was co-ingested with other ingredients (e.g., carbohydrates). It is still possible
that some of these ingredients produced a synergistic or antagonistic effect on performance. In addition,
the dose of caffeine and posology were not uniform among investigations, which could influence some
of the outcomes of the research included in the review. In addition, in the investigations included in
the analysis there were different competitive levels and age categories while the low number of articles
impeded us from knowing if the effect of caffeine on soccer physical performance depends on level or
players’ age. Despite these limitations, this review suggests a positive effect of caffeine in increasing
soccer players’ physical performance with no or little effect on the levels of muscle damage, perceived
effort, or exercising heart rate. Because soccer is a complex sport in which the variables investigated
in this systematic review represent only a small proportion of the factors necessary for succeeding,
further investigations are necessary to determine the effects of caffeine on more complex and ecological
soccer-specific tests, especially involving decision-making situations. In the same way, studies should
be undertaken into whether the effect of caffeine is different according to the competitive level or
soccer player’s age.
5. Conclusions
In summary, acute caffeine intake of a moderate dose of caffeine before exercise has the capacity
to improve several soccer-related abilities and skills such as vertical jump height, repeated sprint
ability, running distances during a game and passing accuracy. Likewise, so far, it has been shown
that a single and acute dose of caffeine does not have a negative impact on the increase of variables
related to muscle damage during official matches. However, more studies are needed to assess whether
chronic caffeine consumption could alter muscle damage markers. Moreover, caffeine supplementation
does not cause changes in either the perception of effort or heart rate during regular high-intensity
intermittent soccer exercises.
Despite this investigation suggesting several benefits of caffeine in soccer, the use of this stimulant
should only be recommended after a careful evaluation of the drawbacks typically associated with the
use of caffeine [56]. With this aim, the minimal dose with a positive impact would be recommended
(i.e., 3 mg/kg), while it can be consumed in either powder (capsules) or liquid (energy drink or sport
drink) forms. Caffeine should only be recommended to athletes who are willing to use ergogenic aids
to increase performance and it should be recommended only on an individual basis under careful
supervision, in order to avoid the use of this substance in non-responders or athletes who report
negative side-effects. Experimenting with caffeine while training, before use in any competition,
and avoiding caffeine tolerance may also be further recommendations when using this substance to
increase soccer physical performance.
Nutrients 2019, 11, 440 13 of 15
Author Contributions: J.M.-A.: conceived and designed the investigation, analysed and interpreted the data,
drafted the paper, and approved the final version submitted for publication. J.C.-G.: and J.D.C.: analysed and
interpreted the data, critically reviewed the paper and approved the final version submitted for publication. A.U.,
P.L.-G. and D.F.-L. critically reviewed the paper and approved the final version submitted for publication.
Funding: The authors declare no funding sources.
Conflicts of Interest: The authors declare no conflict of interest.
References
1. FIFA Communications Division. The FIFA Big Count 2006: 230 Million Active in Football. 2007.
Available online: https://www.fifa.com/mm/document/fifafacts/bcoffsurv/bigcount.statspackage_7024.
pdf (accessed on 18 July 2018).
2. Cornelissen, S.; Maennig, W. On the Political Economy of ‘feel-good’ effects at Sport Mega-Events:
Experiences from FIFA Germany 2006 and Prospects for South Africa 2010. Alternation 2010, 17, 96–120.
3. Haycraft, J.A.; Kovalchik, S.; Pyne, D.B.; Robertson, S. Physical characteristics of players withinfrom food, fluids and sup-
plements can be derived from weighed/measured food
records (typically 3–7 day), a multi-pass 24-hour recall or
from food frequency questionnaires.1 There are inherent lim-
itations with all of these methods, with a bias to the under-
reporting of intakes. Extensive education regarding the
purpose and protocols of documenting intakes may assist
with compliance and enhance the accuracy and validity of
self-reported information.
Meanwhile an athlete_s energy requirements depend on the
periodized training and competition cycle, and will vary from
day to day throughout the yearly training plan relative to
changes in training volume and intensity. Factors that increase
energy needs above normal baseline levels include exposure to
cold or heat, fear, stress, high altitude exposure, some physical
injuries, specific drugs or medications (eg, caffeine, nicotine),
increases in fat-free mass and, possibly, the luteal phase of the
menstrual cycle.2 Aside from reductions in training, energy
requirements are lowered by aging, decreases in fat free mass
(FFM), and, possibly, the follicular phase of the menstrual
cycle.3
Energy balance occurs when total Energy Intake (EI)
equals Total Energy Expenditure (TEE), which in turn
consists of the summation of basal metabolic rate (BMR),
the Thermic Effect of Food (TEF) and the Thermic Effect of
Activity (TEA).
TEE ¼ BMRþ TEFþ TEA
TEA ¼ PlannedExerciseExpenditureþ SpontaneousPhysicalActivity
þ Nonthe Australian
football league participation pathways: A systematic review. Sports Med. Open 2017, 3, 46. [CrossRef]
[PubMed]
4. Sporis, G.; Jukic, I.; Ostojic, S.M.; Milanovic, D. Fitness profiling in soccer: Physical and physiologic
characteristics of elite players. J. Strength Cond. Res. 2009, 23, 1947–1953. [CrossRef] [PubMed]
5. Krustrup, P.; Mohr, M.; Steensberg, A.; Bencke, J.; Kjaer, M.; Bangsbo, J. Muscle and blood metabolites during
a soccer game: Implications for sprint performance. Med. Sci. Sports Exerc. 2006, 38, 1165–1174. [CrossRef]
[PubMed]
6. Chia, J.S.; Barrett, L.A.; Chow, J.Y.; Burns, S.F. Effects of caffeine supplementation on performance in ball
games. Sports Med. 2017, 47, 2453–2471. [CrossRef] [PubMed]
7. Holway, F.E.; Spriet, L.L. Sport-specific nutrition: Practical strategies for team sports. J. Sports Sci. 2011, 29,
S115–S125. [CrossRef] [PubMed]
8. Bishop, D. Dietary supplements and team-sport performance. Sports Med. 2010, 40, 995–1017. [CrossRef]
9. Davis, J.K.; Green, J.M. Caffeine and anaerobic performance: Ergogenic value and mechanisms of action.
Sports Med. 2009, 39, 813–832. [CrossRef]
10. Trexler, E.T.; Smith-Ryan, A.E.; Roelofs, E.J.; Hirsch, K.R.; Mock, M.G. Effects of coffee and caffeine anhydrous
on strength and sprint performance. Eur. J. Sport Sci. 2015. [CrossRef]
11. Glaister, M.; Muniz-Pumares, D.; Patterson, S.D.; Foley, P.; McInnes, G. caffeine supplementation and peak
anaerobic power output. Eur. J. Sport. Sci. 2015, 15, 400–406. [CrossRef]
12. Australian Institute of Sport. Australian Institute of Sport. 2014. Available online: https://www.sportaus.
gov.au/ais (accessed on 25 October 2017).
13. Salinero, J.J.; Lara, B.; Del Coso, J. Effects of acute ingestion of caffeine on team sports performance:
A systematic review and meta-analysis. Res. Sports Med. 2018. [CrossRef] [PubMed]
14. Del Coso, J.; Munoz-Fernandez, V.E.; Munoz, G.; Fernandez-Elias, V.E.; Ortega, J.F.; Hamouti, N.;
Barbero, J.C.; Munoz-Guerra, J. Effects of a caffeine-containing energy drink on simulated soccer performance.
PLoS ONE 2012, 7, e31380. [CrossRef] [PubMed]
15. Foskett, A.; Ali, A.; Gant, N. Caffeine enhances cognitive function and skill performance during simulated
soccer activity. Int. J. Sport Nutr. Exerc. Metab. 2009, 19, 410–423. [CrossRef] [PubMed]
16. Jordan, J.B.; Korgaokar, A.; Farley, R.S.; Coons, J.M.; Caputo, J.L. Caffeine supplementation and reactive
agility in elite youth soccer players. Pediatr. Exerc. Sci. 2014, 26, 168–176. [CrossRef]
17. Ranchordas, M.K.; King, G.; Russell, M.; Lynn, A.; Russell, M. Effects of caffeinated gum on a battery of
soccer-specific tests in trained university-standard male soccer players. Int. J. Sport Nutr. Exerc. Metab. 2018.
[CrossRef] [PubMed]
18. Guerra, M.A., Jr.; Caldas, L.C.; De Souza, H.L.; Vitzel, K.F.; Cholewa, J.M.; Duncan, M.J.;
Guimaraes-Ferreira, L. The acute effects of plyometric and sled towing stimuli with and without caffeine
ingestion on vertical jump performance in professional soccer players. J. Int. Soc. Sports Nutr. 2018.
[CrossRef] [PubMed]
19. Apostolidis, A.; Mougios, V.; Smilios, I.; Rodosthenous, J.; Hadjicharalambous, M. Caffeine supplementation:
Ergogenic in both high and low caffeine responders. Int. J. Sports Physiol. Perform. 2018, 14, 1–25. [CrossRef]
https://www.fifa.com/mm/document/fifafacts/bcoffsurv/bigcount.statspackage_7024.pdf
https://www.fifa.com/mm/document/fifafacts/bcoffsurv/bigcount.statspackage_7024.pdf
http://dx.doi.org/10.1186/s40798-017-0109-9
http://www.ncbi.nlm.nih.gov/pubmed/29260420
http://dx.doi.org/10.1519/JSC.0b013e3181b3e141
http://www.ncbi.nlm.nih.gov/pubmed/19704378
http://dx.doi.org/10.1249/01.mss.0000222845.89262.cd
http://www.ncbi.nlm.nih.gov/pubmed/16775559
http://dx.doi.org/10.1007/s40279-017-0763-6
http://www.ncbi.nlm.nih.gov/pubmed/28741186
http://dx.doi.org/10.1080/02640414.2011.605459
http://www.ncbi.nlm.nih.gov/pubmed/21831001
http://dx.doi.org/10.2165/11536870-000000000-00000
http://dx.doi.org/10.2165/11317770-000000000-00000
http://dx.doi.org/10.1186/1550-2783-12-S1-P57
http://dx.doi.org/10.1080/17461391.2014.962619
https://www.sportaus.gov.au/ais
https://www.sportaus.gov.au/ais
http://dx.doi.org/10.1080/15438627.2018.1552146
http://www.ncbi.nlm.nih.gov/pubmed/30518253
http://dx.doi.org/10.1371/journal.pone.0031380
http://www.ncbi.nlm.nih.gov/pubmed/22348079
http://dx.doi.org/10.1123/ijsnem.19.4.410
http://www.ncbi.nlm.nih.gov/pubmed/19827465
http://dx.doi.org/10.1123/pes.2013-0134
http://dx.doi.org/10.1123/ijsnem.2017-0405
http://www.ncbi.nlm.nih.gov/pubmed/29584462
http://dx.doi.org/10.1186/s12970-018-0258-3
http://www.ncbi.nlm.nih.gov/pubmed/30348178
http://dx.doi.org/10.1123/ijspp.2018-0238
Nutrients 2019, 11, 440 14 of 15
20. Astorino, T.A.; Matera, A.J.; Basinger, J.; Evans, M.; Schurman, T.; Marquez, R. Effects of red bull energy
drink on repeated sprint performance in women athletes. Amino Acids 2012, 42, 1803–1808. [CrossRef]
21. Ellis, M.; Noon, M.; Myers, T.; Clarke, N. Low doses of caffeine: Enhancement of physical performance in
elite adolescent male soccer players. Int. J. Sports Physiol. Perform. 2018, 9, 1–21. [CrossRef]
22. Liberati, A.; Altman, D.G.; Tetzlaff, J.; Mulrow, C.; Gotzsche, P.C.; Ioannidis, J.P.; Clarke, M.; Devereaux, P.J.;
Kleijnen, J.; Moher, D. The PRISMA statement for reporting systematic reviews and meta-analyses of studies
that evaluate healthcare interventions: Explanation and elaboration. BMJ 2009, 339, b2700. [CrossRef]
23. Hespel, P.; Maughan, R.J.; Greenhaff, P.L. Dietary supplements for football. J. Sports Sci. 2006, 24, 749–761.
[CrossRef] [PubMed]
24. Maher, C.G.; Sherrington, C.; Herbert, R.D.; Moseley, A.M.; Elkins, M. Reliability of the PEDro scale for
rating quality of randomized controlled trials. Phys. Ther. 2003, 83, 713–721. [PubMed]
25. Andrade-Souza, V.A.; Bertuzzi, R.; de Araujo, G.G.; Bishop, D.; Lima-Silva, A.E. Effects of isolated
or combined carbohydrate and caffeine supplementation between 2 daily training sessions on soccer
performance. Appl. Physiol. Nutr. Metab. 2015, 40, 457–463. [CrossRef] [PubMed]
26. Guttierres, A.P.M.; Alfenas, R.d.C.; Gatti, K.; Lima, J.R.P.; Silva, Â.A.; Natali, A.J.; Marins, J.C.B. Metabolic
effects of a caffeinated sports drink consumed during a soccer match. Motriz: Revista de Educação Física 2013,
19, 688–695.
27. Lara, B.; Gonzalez-Millan, C.; Salinero, J.J.; Abian-Vicen, J.; Areces, F.; Barbero-Alvarez, J.C.; Munoz, V.;
Portillo, L.J.; Gonzalez-Rave, J.M.; Del Coso, J. Caffeine-containing energy drink improves physical
performance in female soccer players. Amino Acids 2014, 46, 1385–1392. [CrossRef] [PubMed]
28. Gant, N.; Ali, A.; Foskett, A. The influence of caffeine and carbohydrate coingestion on simulated soccer
performance. Int. J. Sport Nutr. Exerc. Metab. 2010, 20, 191–197. [CrossRef] [PubMed]
29. Guttierres, A.P.M.; Natali, A.J.; Alfenas, R.D.C.G.; Marins, J.C.B. Ergogenic Effect of a caffeinated sports drink
on performance in soccer specific abilities tests. Revista Brasileira de Medicina do Esporte 2009, 15, 450–454.
[CrossRef]
30. Machado, M.; Koch, A.J.; Willardson, J.M.; dos Santos, F.C.; Curty, V.M.; Pereira, L.N. Caffeine does
not augment markers of muscle damage or leukocytosis following resistance exercise. Int. J. Sports
Physiol. Perform. 2010, 5, 18–26.
31. Machado, M.; Antunes, W.D.; Tamy, A.L.M.; Azevedo, P.G.; Barreto, J.G.; Hackney, A.C. Effect of a single
dose of caffeine supplementation and intermittent-interval exercise on muscle damage markers in soccer
players. J. Exerc. Sci. Fit. 2009, 7, 91–97. [CrossRef]
32. Machado, M.; Breder, A.C.; Ximenes, M.C.; Simões, J.R.; Vigo, J.F.F. Caffeine supplementation and muscle
damage in soccer players. Braz. J. Pharm. 2009, 45, 257–261. [CrossRef]
33. Bassini-Cameron, A.; Sweet, E.; Bottino, A.; Bittar, C.; Veiga, C.; Cameron, L.C. Effect of caffeine
supplementation on haematological and biochemical variables in elite soccer players under physical stress
conditions.Br. J. Sports Med. 2007, 41, 523–530. [CrossRef] [PubMed]
34. Sheppard, J.M.; Young, W.B.; Doyle, T.L.; Sheppard, T.A.; Newton, R.U. An evaluation of a new test of
reactive agility and its relationship to sprint speed and change of direction speed. J. Sci. Med. Sport 2006, 9,
342–349. [CrossRef] [PubMed]
35. Reilly, T.; Williams, A.M.; Nevill, A.; Franks, A. A Multidisciplinary approach to talent identification in
soccer. J. Sports Sci. 2000, 18, 695–702. [CrossRef] [PubMed]
36. Little, T.; Williams, A.G. Specificity of acceleration, maximum speed, and agility in professional soccer
players. J. Strength Cond Res. 2005, 19, 76–78.
37. Bangsbo, J. The Physiology of soccer—With special reference to intense intermittent exercise. Acta Physiol.
Scand. Suppl. 1994, 619, 1–155. [PubMed]
38. Roozen, M. Illinois Agility Test. NSCA’s Perform. Train. J. 2004, 3, 5–6.
39. Serrano, J.; Shahidian, S.; Sampaio, J.; Leite, N. The importance of sports performance factors and training
contents from the perspective of futsal coaches. J. Hum. Kinet. 2013, 38, 151–160. [CrossRef] [PubMed]
40. Jacobs, I. Lactate, muscle glycogen and exercise performance in man. Acta Physiol. Scand. 1981, 495, 1–35.
41. Kamimori, G.H.; Karyekar, C.S.; Otterstetter, R.; Cox, D.S.; Balkin, T.J.; Belenky, G.L.; Eddington, N.D.
The rate of absorption and relative bioavailability of caffeine administered in chewing gum versus capsules
to normal healthy volunteers. Int. J. Pharm. 2002, 234, 159–167. [CrossRef]
http://dx.doi.org/10.1007/s00726-011-0900-8
http://dx.doi.org/10.1123/ijspp.2018-0536
http://dx.doi.org/10.1136/bmj.b2700
http://dx.doi.org/10.1080/02640410500482974
http://www.ncbi.nlm.nih.gov/pubmed/16766503
http://www.ncbi.nlm.nih.gov/pubmed/12882612
http://dx.doi.org/10.1139/apnm-2014-0268
http://www.ncbi.nlm.nih.gov/pubmed/25884315
http://dx.doi.org/10.1007/s00726-014-1709-z
http://www.ncbi.nlm.nih.gov/pubmed/24615239
http://dx.doi.org/10.1123/ijsnem.20.3.191
http://www.ncbi.nlm.nih.gov/pubmed/20601736
http://dx.doi.org/10.1590/S1517-86922009000700010
http://dx.doi.org/10.1016/S1728-869X(09)60011-3
http://dx.doi.org/10.1590/S1984-82502009000200010
http://dx.doi.org/10.1136/bjsm.2007.035147
http://www.ncbi.nlm.nih.gov/pubmed/17473001
http://dx.doi.org/10.1016/j.jsams.2006.05.019
http://www.ncbi.nlm.nih.gov/pubmed/16844413
http://dx.doi.org/10.1080/02640410050120078
http://www.ncbi.nlm.nih.gov/pubmed/11043895
http://www.ncbi.nlm.nih.gov/pubmed/8059610
http://dx.doi.org/10.2478/hukin-2013-0055
http://www.ncbi.nlm.nih.gov/pubmed/24235991
http://dx.doi.org/10.1016/S0378-5173(01)00958-9
Nutrients 2019, 11, 440 15 of 15
42. Thorpe, R.; Sunderland, C. Muscle damage, endocrine, and immune marker response to a soccer match.
J. Strength Cond. Res. 2012, 26, 2783–2790. [CrossRef]
43. Clarkson, P.M.; Hubal, M.J. Exercise-induced muscle damage in humans. Am. J. Phys. Med. Rehabil. 2002, 81,
S52–S69. [CrossRef] [PubMed]
44. Graham, T.E.; Sathasivam, P.; Rowland, M.; Marko, N.; Greer, F.; Battram, D. Caffeine ingestion elevates
plasma insulin response in humans during an oral glucose tolerance test. Can. J. Physiol. Pharmacol. 2001, 79,
559–565. [CrossRef] [PubMed]
45. Davis, J.M.; Zhao, Z.; Stock, H.S.; Mehl, K.A.; Buggy, J.; Hand, G.A. Central nervous system effects of caffeine
and adenosine on fatigue. Am. J. Physiol. Regul. Integr. Comp. Physiol. 2003, 284, R399–R404. [CrossRef]
[PubMed]
46. Yeo, S.E.; Jentjens, R.L.; Wallis, G.A.; Jeukendrup, A.E. Caffeine increases exogenous carbohydrate oxidation
during exercise. J. Appl. Physiol. 2005, 99, 844–850. [CrossRef] [PubMed]
47. Doherty, M.; Smith, P. Effects of caffeine ingestion on rating of perceived exertion during and after exercise:
A Meta-analysis. Scand. J. Med. Sci. Sports 2005, 15, 69–78. [CrossRef] [PubMed]
48. Del Coso, J.; Ramirez, J.A.; Munoz, G.; Portillo, J.; Gonzalez-Millan, C.; Munoz, V.; Barbero-Alvarez, J.C.;
Munoz-Guerra, J. Caffeine-containing energy drink improves physical performance of elite rugby players
during a simulated match. Appl. Physiol. Nutr. Metab. 2013, 38, 368–374. [CrossRef]
49. Astorino, T.A.; Cottrell, T.; Lozano, A.T.; Aburto-Pratt, K.; Duhon, J. Effect of caffeine on RPE and perceptions
of pain, arousal, and pleasure/displeasure during a cycling time trial in endurance trained and active men.
Physiol. Behav. 2012, 106, 211–217. [CrossRef]
50. Turley, K.; Eusse, P.A.; Thomas, M.M.; Townsend, J.R.; Morton, A.B. Effects of different doses of caffeine on
anaerobic exercise in boys. Pediatr. Exerc. Sci. 2015, 27, 50–56. [CrossRef]
51. Doherty, M.; Smith, P.M.; Davison, R.C.; Hughes, M.G. Caffeine is ergogenic after supplementation of oral
creatine monohydrate. Med. Sci. Sports Exerc. 2002, 34, 1785–1792. [CrossRef]
52. Grgic, J.; Mikulic, P. Caffeine ingestion acutely enhances muscular strength and power but not muscular
endurance in resistance-trained men. Eur. J. Sport Sci. 2017, 17, 1029–1036. [CrossRef]
53. Lara, B.; Ruiz-Vicente, D.; Areces, F.; Abian-Vicen, J.; Salinero, J.J.; Gonzalez-Millan, C.; Gallo-Salazar, C.;
Del Coso, J. Acute consumption of a caffeinated energy drink enhances aspects of performance in sprint
swimmers. Br. J. Nutr. 2015, 114, 908–914. [CrossRef] [PubMed]
54. Skinner, T.L.; Jenkins, D.G.; Coombes, J.S.; Taaffe, D.R.; Leveritt, M.D. Dose response of caffeine on 2000-m
rowing performance. Med. Sci. Sports Exerc. 2010, 42, 571–576. [CrossRef] [PubMed]
55. Pickering, C.; Kiely, J. Are the current guidelines on caffeine use in sport optimal for everyone?
Inter-individual variation in caffeine ergogenicity, and a move towards personalised sports nutrition.
Sports Med. 2018, 48, 7–16. [CrossRef] [PubMed]
56. Salinero, J.J.; Lara, B.; Abian-Vicen, J.; Gonzalez-Millán, C.; Areces, F.; Gallo-Salazar, C.; Ruiz-Vicente, D.;
Del Coso, J. The use of energy drinks in sport: Perceived ergogenicity and side effects in male and female
athletes. Br. J. Nutr. 2014, 112, 1494–1502. [CrossRef] [PubMed]
© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).
http://dx.doi.org/10.1519/JSC.0b013e318241e174
http://dx.doi.org/10.1097/00002060-200211001-00007
http://www.ncbi.nlm.nih.gov/pubmed/12409811
http://dx.doi.org/10.1139/y01-026
http://www.ncbi.nlm.nih.gov/pubmed/11478588
http://dx.doi.org/10.1152/ajpregu.00386.2002
http://www.ncbi.nlm.nih.gov/pubmed/12399249
http://dx.doi.org/10.1152/japplphysiol.00170.2005
http://www.ncbi.nlm.nih.gov/pubmed/15831802
http://dx.doi.org/10.1111/j.1600-0838.2005.00445.x
http://www.ncbi.nlm.nih.gov/pubmed/15773860
http://dx.doi.org/10.1139/apnm-2012-0339
http://dx.doi.org/10.1016/j.physbeh.2012.02.006
http://dx.doi.org/10.1123/pes.2014-0032
http://dx.doi.org/10.1097/00005768-200211000-00015
http://dx.doi.org/10.1080/17461391.2017.1330362
http://dx.doi.org/10.1017/S0007114515002573
http://www.ncbi.nlm.nih.gov/pubmed/26279580
http://dx.doi.org/10.1249/MSS.0b013e3181b6668b
http://www.ncbi.nlm.nih.gov/pubmed/19952822
http://dx.doi.org/10.1007/s40279-017-0776-1
http://www.ncbi.nlm.nih.gov/pubmed/28853006
http://dx.doi.org/10.1017/S0007114514002189
http://www.ncbi.nlm.nih.gov/pubmed/25212095
http://creativecommons.org/
http://creativecommons.org/licenses/by/4.0/.
nutrients
Review
Effects of Creatine Supplementation on Athletic
Performance in Soccer Players: A Systematic Review
and Meta-Analysis
Juan Mielgo-Ayuso 1,* , Julio Calleja-Gonzalez 2, Diego Marqués-Jiménez 3,
Alberto Caballero-García 4, Alfredo Córdova 1 and Diego Fernández-Lázaro 5
1 Department of Biochemistry and Physiology, School of Physical Therapy, University of Valladolid,
42004 Soria, Spain; a.cordova@bio.uva.es
2 Laboratory of Human Performance, Department of Physical Education and Sport, Faculty of Education,
University of the Basque Country, 01007 Vitoria, Spain; julio.calleja.gonzalez@gmail.com
3 Academy Department, Deportivo Alavés SAD, 01007Vitoria-Gasteiz, Spain;
diegomarquesjimenez@gmail.com
4 Department of Anatomy and Radiology, Faculty of Physical Therapy, University of Valladolid,
Campus de Soria, 42004 Soria, Spain; albcab@ah.uva.es
5 Department of Cellular Biology, Histology and Pharmacology, Faculty of Physical Therapy,
University of Valladolid, Campus de Soria, 42004 Soria, Spain; diego.fernandez.lazaro@uva.es
* Correspondence: juanfrancisco.mielgo@uva.es; Tel.: +34-975-129-187
Received: 15 February 2019; Accepted: 28 March 2019; Published: 31 March 2019
����������
�������
Abstract: Studies have shown that creatine supplementation increases intramuscular creatine
concentrations, favoring the energy system of phosphagens, which may help explain the observed
improvements in high-intensity exercise performance. However, research on physical performance
in soccer has shown controversial results, in part because the energy system used is not taken into
account. The main aim of this investigation was to perform a systematic review and meta-analysis to
determine the efficacy of creatine supplementation for increasing performance in skills related
to soccer depending upon the type of metabolism used (aerobic, phosphagen, and anaerobic
metabolism). A structured search was carried out following the Preferred Reporting Items for
Systematic Review and Meta-Analyses (PRISMA) guidelines in the Medline/PubMed and Web
of Science, Cochrane Library, and Scopus databases until January 2019. The search included studies
with a double-blind and randomized experimental design in which creatine supplementation
was compared to an identical placebo situation (dose, duration, timing, and drug appearance).
There were no filters applied to the soccer players’ level, gender, or age. A final meta-analysis
was performed using the random effects model and pooled standardized mean differences (SMD)
(Hedges’s g). Nine studies published were included in the meta-analysis. This revealed that creatine
supplementation did not present beneficial effects on aerobic performance tests (SMD, −0.05;
95% confidence interval (CI), −0.37 to 0.28; p = 0.78) and phosphagen metabolism performance
tests (strength, single jump, single sprint, and agility tests: SMD, 0.21; 95% CI, −0.03 to 0.45; p = 0.08).
However, creatine supplementation showed beneficial effects on anaerobic performance tests (SMD,
1.23; 95% CI, 0.55–1.91; pO (Outcome): “soccer-specific skills and relationships with aerobic, phosphagen,
and anaerobic metabolism performance”, and S (study design): “double-blind and randomized design”.
A structured search was conducted in the following databases: PubMed/MEDLINE, web of
science (WOS), Cochrane Library, and Scopus. It included results until 30 January 2019, while no year
restriction was applied to the search strategy. Search terms included a mix of medical subject headings
(MeSH) and free-text words for key concepts related to Cr and soccer performance. Specifically,
we used the following search equation: (“football” [All Fields] OR “soccer” [All Fields]) AND “creatine
supplementation” [All Fields] AND (“physical performance” [All Fields] OR “physical endurance”
[All Fields] OR “physical” [All Fields] OR “endurance” [All Fields] OR “performance” [All Fields]
OR “aerobic” [All Fields] OR “anaerobic” [All Fields]), which returned relevant articles in the field of
applying the snowball strategy. All titles and abstracts from the search were cross-referenced to identify
duplicates and any potential missing studies. Titles and abstracts were screened for a subsequent
full-text review. The search for published studies was independently performed by two different
authors (JMA and JCG) and disagreements were resolved through discussions between them.
2.2. Inclusion and Exclusion Criteria
There were no filters applied to the soccer players’ level, gender, race, or age to increase the power
of the analysis. However, for the articles obtained in the database search, the following inclusion criteria
were applied to select the final studies: (1) in which there was an experimental condition that included
the ingestion of Cr before and/or during exercise which was compared to an identical experimental
condition with the ingestion of a placebo; (2) testing the effects of Cr on soccer-specific tests and/or real
or simulated matches; (3) with a blinded and randomized design; (4) with clear information regarding
the administration of Cr (relative dose of Cr per kilogram of body mass and/or absolute dose of Cr
with information about body mass, timing of Cr intake before the onset of performance measurements,
etc.); (4) on soccer players with previous training backgrounds in this sport; and (5) published in
any language. On the other hand, the following exclusion criteria were applied to the experimental
protocols of the investigation: (1) studies that were not conducted with soccer players; (2) studies that
were performed for clinical purposes or therapeutic use; (3) the absence of a true placebo condition;
and (4) studies carried out using participants with a previous medical condition, illness, or injury.
2.3. Data Extraction
Once the inclusion/exclusion criteria were applied to each study, data on study source (including
authors and year of publication), study design, Cr supplementation (dose and timing), sample size,
characteristics of the participants (level and gender), and final outcomes of the interventions were
extracted independently by two authors using a spreadsheet. Subsequently, disagreements were
resolved through discussion until a consensus was achieved.
Experiments were clustered by the type of test used to assess team sport performance, and groups
of experiments were created which assessed the effect of Cr on aerobic performance (Yo-Yo intermittent
recovery test level 1), phosphagen metabolism performance (strength, jump, sprint, and agility course),
and anaerobic metabolism measures (repeated sprint ability and the Wingate test). Six studies included
measurements of two or more types of performance outcomes (e.g., aerobic and phosphagen abilities)
or even two types of tests for the same performance outcome. In these cases, each test or type of
Nutrients 2019, 11, 757 4 of 17
performance outcome was treated as a single and independent set of data for the meta-analysis and
included in the appropriate performance outcome.
The mean (M), standard deviation (SD), and sample size data were extracted by one author from
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain
the data. When it was impossible, mean and SD were extrapolated from the figures. Any disagreement
was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG).
2.4. Quality Assessment of the Experiments
Methodological quality and risk of bias were assessed by two authors independently (JMA and
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains:
random sequence generation (selection bias), allocation concealment (selection bias), blinding of
participants and personnel (performance bias), blinding of outcome assessment (detection bias),
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias.
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely
to seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1
and Figure 1.
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as
percentages across all included studies.
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
indicates low risk of bias,
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
indicates unknown risk of
bias, and
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
R
an
d
o
m
 s
eq
u
en
ce
 
g
en
er
at
io
n
 (
se
le
ct
io
n
 b
ia
s)
 
A
ll
o
ca
ti
o
n
 c
o
n
ce
al
m
en
t 
(s
el
ec
ti
o
n
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
p
ar
ti
ci
p
an
ts
 
an
d
 p
er
so
n
n
el
 
(p
er
fo
rm
an
ce
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
o
u
tc
o
m
e 
as
se
ss
m
en
t 
(d
et
ec
ti
o
n
 
b
ia
s)
 
In
co
m
p
le
te
 o
u
tc
o
m
e 
d
at
a 
(a
tt
ri
ti
o
n
 b
ia
s)
 
S
el
ec
ti
v
e 
re
p
o
rt
in
g
 
(r
ep
o
rt
in
g
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
indicates high risk of bias.
R
an
do
m
se
qu
en
ce
ge
ne
ra
ti
on
(s
el
ec
ti
on
bi
as
)
A
llo
ca
ti
on
co
nc
ea
lm
en
t
(s
el
ec
ti
on
bi
as
)
Bl
in
di
ng
of
pa
rt
ic
ip
an
ts
an
d
pe
rs
on
ne
l
(p
er
fo
rm
an
ce
bi
as
)
Bl
in
di
ng
of
ou
tc
om
e
as
se
ss
m
en
t
(d
et
ec
ti
on
bi
as
)
In
co
m
pl
et
e
ou
tc
om
e
da
ta
(a
tt
ri
ti
on
bi
as
)
Se
le
ct
iv
e
re
po
rt
in
g
(r
ep
or
ti
ng
bi
as
)
O
th
er
bi
as
Mujika et al., 2000 [17]
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bembenet al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author fromthe tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Bemben et al., 2001 [6]
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authorsindependently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
R
an
d
o
m
 s
eq
u
en
ce
 
g
en
er
at
io
n
 (
se
le
ct
io
n
 b
ia
s)
 
A
ll
o
ca
ti
o
n
 c
o
n
ce
al
m
en
t 
(s
el
ec
ti
o
n
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
p
ar
ti
ci
p
an
ts
 
an
d
 p
er
so
n
n
el
 
(p
er
fo
rm
an
ce
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
o
u
tc
o
m
e 
as
se
ss
m
en
t 
(d
et
ec
ti
o
n
 
b
ia
s)
 
In
co
m
p
le
te
 o
u
tc
o
m
e 
d
at
a 
(a
tt
ri
ti
o
n
 b
ia
s)
 
S
el
ec
ti
v
e 
re
p
o
rt
in
g
 
(r
ep
o
rt
in
g
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performancebias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results).If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Cox et al., 2002 [21]
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
R
an
d
o
m
 s
eq
u
en
ce
 
g
en
er
at
io
n
 (
se
le
ct
io
n
 b
ia
s)
 
A
ll
o
ca
ti
o
n
 c
o
n
ce
al
m
en
t 
(s
el
ec
ti
o
n
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
p
ar
ti
ci
p
an
ts
 
an
d
 p
er
so
n
n
el
 
(p
er
fo
rm
an
ce
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
o
u
tc
o
m
e 
as
se
ss
m
en
t 
(d
et
ec
ti
o
n
 
b
ia
s)
 
In
co
m
p
le
te
 o
u
tc
o
m
e 
d
at
a 
(a
tt
ri
ti
o
n
 b
ia
s)
 
S
el
ec
ti
v
e 
re
p
o
rt
in
g
 
(r
ep
o
rt
in
g
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
R
an
d
o
m
 s
eq
u
en
ce
 
g
en
er
at
io
n
 (
se
le
ct
io
n
 b
ia
s)
 
A
ll
o
ca
ti
o
n
 c
o
n
ce
al
m
en
t 
(s
el
ec
ti
o
n
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
p
ar
ti
ci
p
an
ts
 
an
d
 p
er
so
n
n
el
 
(p
er
fo
rm
an
ce
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
o
u
tc
o
m
e 
as
se
ss
m
en
t 
(d
et
ec
ti
o
n
 
b
ia
s)
 
In
co
m
p
le
te
 o
u
tc
o
m
e 
d
at
a 
(a
tt
ri
ti
o
n
 b
ia
s)
 
S
el
ec
ti
v
e 
re
p
o
rt
in
g
 
(r
ep
o
rt
in
g
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias itemis prohibited.
athletes.15 Nevertheless, there are relationships between body
composition and sports performance that are important to
consider within an athlete_s preparation.
In sports involving strength and power, athletes strive to
gain fat-free mass via a program of muscle hypertrophy at
specified times of the annual macro-cycle. Whereas some
athletes aim to gain absolute size and strength per se, in
other sports, in which the athlete must move their own body
mass or compete within weight divisions, it is important to
optimize power to weight ratios rather than absolute power.16
Thus, some power athletes also desire to achieve low body fat
levels. In sports involving weight divisions (eg. combat
sports, lightweight rowing, weightlifting), competitors typi-
cally target the lowest achievable body weight category,
while maximizing their lean mass within this target.
Other athletes strive to maintain a low body mass and/or
body fat level for separate advantages.17 Distance runners
and cyclists benefit from a low energy cost of movement and
a favorable ratio of weight to surface area for heat dissipa-
tion. Team athletes can increase their speed and agility by
being lean, while athletes in acrobatic sports (eg, diving.
gymnastics, dance) gain biomechanical advantages in being
able to move their bodies within a smaller space. In some of
these sports and others (eg, body building), there is an ele-
ment of aesthetics in determining performance outcomes.
Although there are demonstrated advantages to achieving a
certain body composition, athletes may feel pressure to
strive to achieve unrealistically low targets of weight/body
fat or to reach them in an unrealistic time frame.15 Such
athletes may be susceptible to practicing extreme weight
control behaviors or continuous dieting, exposing them-
selves to chronic periods of low EA and poor nutrient sup-
port in an effort to repeat previous success at a lower weight
or leaner body composition.15,18 Extreme methods of weight
control can be detrimental to health and performance, and
disordered eating patterns have also been observed in these
sport scenarios.15,18
Nevertheless, there are scenarios in which an athlete will
enhance their health and performance by reducing body
weight or body fat as part of a periodized strategy. Ideally,
this occurs within a program that gradually achieves an in-
dividualized ‘‘optimal’’ body composition over the athlete_s
athletic career, and allows weight and body fat to track
within a suitable range within the annual training cycle.18
The program should also include avoiding situations in
which athletes inadvertently gain excessive amounts of body
fat as a result of a sudden energy mismatch when energy
expenditure is abruptly reduced (eg, the off-season or inju-
ry). In addition, athletes are warned against the sudden or
excessive gain in body fat which is part of the culture of
some sports where a high body mass is deemed useful
for performance. Although body mass index is not appro-
priate as a body composition surrogate in athletes, a chronic
interest in gaining weight may put some athletes at risk
for an ‘‘obese’’ body mass index which may increase the risk
of meeting the criteria for metabolic syndrome.19 Sports
dietitians should be aware of sports that promote the at-
tainment of a large body mass and screen for metabolic
risk factors.19
Methodologies for body composition assess-
ment. Techniques used to assess athlete body composition
include dual energy x-ray absorptiometry (DXA), hydro-
densitometry, air displacement plethysmography, skinfold
measurements, and single and multi-frequency bioelectrical
impedance analysis. Although DXA is quick and noninva-
sive, issues around cost, accessibility, and exposure to a
small radiation dose limit its utility, particularly for cer-
tain populations.20 When undertaken according to stan-
dardized protocols, DXA has the lowest standard error of
estimate while skinfold measures have the highest. Air dis-
placement plethysmography (BodPod, Life Measurement,
Inc., Concord, CA) provides an alternative method that is
quick and reliable, but may underestimate body fat by 2%–
3%.20 Skinfold measurement and other anthropometric data
serve as an excellent surrogate measure of adiposity and
muscularity when profiling composition changes in response
to training interventions.20 However, it should be noted
that the standardization of skinfold sites, measurement
techniques, and calipers vary around the world. Despite
some limitations, this technique remains a popular method
of choice due to convenience and cost, with information
being provided in absolute measures and compared with
sequential data from the individual athlete or, in a general
way, with normative data collected in the same way from
athlete populations.20,21
All body composition assessment techniques should
be scrutinized to ensure accuracy and reliability. Testing
should be conducted with the same calibrated equipment,
with a standardized protocol, and by technicians with known
test-retest reliability. Where population–specific prediction
equations are used, they should be cross-validated and reli-
able. Athletes should be educated on the limitations associ-
ated with body composition assessment and should strictly
follow pre-assessment protocols. These instructions which
include maintaining a consistent training volume, fasting
status, and hydration from test to test20 should be enforced to
avoid compromising the accuracy and reliability of body
composition measures.
Body composition should be determined within a sports
program according to a schedule that is appropriate to the
performance of the event, the practicality of undertaking
assessments, and the sensitivity of the athlete. There are
technical errors associated with all body composition tech-
niques that limit the usefulness of measurement for athlete
selection and performance prediction. In lieu of setting ab-
solute body composition goals or applying absolute criteria
to categorize groups of athletes, it is preferred that normative
data are provided in terms of ranges.21 Since body fat con-
tent for an individual athlete will vary over the season and
over the athlete_s career, goals for body composition should
be set in terms of ranges that can be appropriately tracked at
critical times. When conducting such monitoring programs,
NUTRITION AND ATHLETIC PERFORMANCE Medicine & Science in Sports & Exercised 547
SPEC
IA
L
C
O
M
M
U
N
IC
ATIO
N
S
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
it is important that the communication of results with
coaches, training staff, and athletes is undertaken with sen-
sitivity, that limitations in measurement technique are rec-
ognized, and that care is taken to avoid promoting an
unhealthy obsession with body composition.17,18 Sports di-
etitians have important opportunities to work with these
athletes to help promote a healthy body composition, and to
minimize their reliance on rapid-weight loss techniques and
other hazardous practices that may result in performance
decrements, loss of fat-free mass, and chronic health risks.
Many themes should be addressed and include the creation
of a culture and environment that values safe and long-term
approaches to management of body composition; modifica-
tion of rules or practices around selection and qualifica-
tion for weight classes;16,19,22 and programs that identify
disordered eating practices at an early stage for intervention,
and where necessary, removal from play.18
Principles of altering body composition and
weight. Athletes often need assistance in setting appropri-
ate short-term and long-term goals, understanding nutri-
tional practices that can safely and effectively increase
muscle mass or reduce body fat/weight, and integrating
these strategies into an eating plan that achieves other per-
formance nutrition goals. Frequent follow up with these
athletes may have long-term benefitspresented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ngo
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
R
an
d
o
m
 s
eq
u
en
ce
 
g
en
er
at
io
n
 (
se
le
ct
io
n
 b
ia
s)
 
A
ll
o
ca
ti
o
n
 c
o
n
ce
al
m
en
t 
(s
el
ec
ti
o
n
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
p
ar
ti
ci
p
an
ts
 
an
d
 p
er
so
n
n
el
 
(p
er
fo
rm
an
ce
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
o
u
tc
o
m
e 
as
se
ss
m
en
t 
(d
et
ec
ti
o
n
 
b
ia
s)
 
In
co
m
p
le
te
 o
u
tc
o
m
e 
d
at
a 
(a
tt
ri
ti
o
n
 b
ia
s)
 
S
el
ec
ti
v
e 
re
p
o
rt
in
g
 
(r
ep
o
rt
in
g
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Biwer et al., 2003 [15]
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biweret al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary,we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
R
an
d
o
m
 s
eq
u
en
ce
 
g
en
er
at
io
n
 (
se
le
ct
io
n
 b
ia
s)
 
A
ll
o
ca
ti
o
n
 c
o
n
ce
al
m
en
t 
(s
el
ec
ti
o
n
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
p
ar
ti
ci
p
an
ts
 
an
d
 p
er
so
n
n
el
 
(p
er
fo
rm
an
ce
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
o
u
tc
o
m
e 
as
se
ss
m
en
t 
(d
et
ec
ti
o
n
 
b
ia
s)
 
In
co
m
p
le
te
 o
u
tc
o
m
e 
d
at
a 
(a
tt
ri
ti
o
n
 b
ia
s)
 
S
el
ec
ti
v
e 
re
p
o
rt
in
g
 
(r
ep
o
rt
in
g
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Ostojic 2004 [18]
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently(JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcomedata (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Claudino et al., 2014 [20]
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(sel
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
R
an
d
o
m
 s
eq
u
en
ce
 
g
en
er
at
io
n
 (
se
le
ct
io
n
 b
ia
s)
 
A
ll
o
ca
ti
o
n
 c
o
n
ce
al
m
en
t 
(s
el
ec
ti
o
n
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
p
ar
ti
ci
p
an
ts
 
an
d
 p
er
so
n
n
el
 
(p
er
fo
rm
an
ce
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
o
u
tc
o
m
e 
as
se
ss
m
en
t 
(d
et
ec
ti
o
n
 
b
ia
s)
 
In
co
m
p
le
te
 o
u
tc
o
m
e 
d
at
a 
(a
tt
ri
ti
o
n
 b
ia
s)
 
S
el
ecti
v
e 
re
p
o
rt
in
g
 
(r
ep
o
rt
in
g
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Williams et al., 2014 [16]
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
R
an
d
o
m
 s
eq
u
en
ce
 
g
en
er
at
io
n
 (
se
le
ct
io
n
 b
ia
s)
 
A
ll
o
ca
ti
o
n
 c
o
n
ce
al
m
en
t 
(s
el
ec
ti
o
n
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
p
ar
ti
ci
p
an
ts
 
an
d
 p
er
so
n
n
el
 
(p
er
fo
rm
an
ce
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
o
u
tc
o
m
e 
as
se
ss
m
en
t 
(d
et
ec
ti
o
n
 
b
ia
s)
 
In
co
m
p
le
te
 o
u
tc
o
m
e 
d
at
a 
(a
tt
ri
ti
o
n
 b
ia
s)
 
S
el
ec
ti
v
e 
re
p
o
rt
in
g
 
(r
ep
o
rt
in
g
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campilloet al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreementincluding shepherding
the athlete through short-term goals and reducing reliance on
extreme techniques and fad diets/behaviors.
There is ample evidence in weight sensitive and weight-
making sports that athletes frequently undertake rapid weight
loss strategies to gain a competitive advantage.20,23,24 How-
ever, the resultant hypohydration (body water deficit), loss of
glycogen stores and lean mass, and other outcomes of path-
ological behaviors (eg, purging, excessive training, starving)
can impair health and performance.18 Nevertheless, respon-
sible use of short-term, rapid weight loss techniques, when
indicated, is preferred over extreme and extended energy re-
striction and suboptimal nutrition support.17 When actual loss
of body weight is required, it should be programmed to occur
in the base phase of training or well out from competition to
minimize loss of performance,25 and should be achieved with
techniques that maximize loss of body fat while preserving
muscle mass and other health goals. Such strategies include
achieving a slight energy deficit to achieve a slow rather than
rapid rate of loss and increasing dietary protein intake. In this
regard, the provision of a higher protein intake (2.3 vs 1 g/kg/d)
in a shorter-term (2 w), energy-restricted diet in athletes was
found to retain muscle mass while losing weight and body fat.26
Furthermore, fat-free mass and performance may be better
preserved in athletes who minimize weekly weight loss to
G1% per week. 25
An individualized diet and training prescription for
weight/fat loss should be based on assessment of goals,
present training and nutrition practices, past experiences,
and trial and error. Nevertheless, for most athletes, the
practical approach of decreasing energy intake by ~250–
500 kcal/d from their periodized energy needs, while either
maintaining or slightly increasing energy expenditure, can
achieve progress towards short-term body composition goals
over approximately 3–6 weeks. In some situations, addi-
tional moderate aerobic training and close monitoring can be
useful.27 These strategies can be implemented to help aug-
ment the diet-induced energy deficits without negatively
impacting recovery from sport-specific training. Arranging
the timing and content of meals to support training nutrition
goals and recovery may reduce fatigue during frequent
training sessions and may help optimize body composition
over time.18 Overall barriers to body composition manage-
ment include limited access to healthy food options, limited
skills or opportunity for food preparation, lack of daily
routine, and exposure to catering featuring unlimited portion
sizes and energy-dense foods. Such factors, particularly
found in association with the travel and communal living
experiences in the athlete lifestyle, can promote poor dietary
quality that thwarts progress and may lead to the pursuit of
quick fixes, acute dieting, and extreme weight loss practices.
EAL Question #1 (Table 1) examined the effect of neg-
ative energy balance on sport performance, finding only
fair support for an impairment of physical capacity due to a
hypoenergetic diet in the currently examined scenarios.
However, few studies have investigated the overlay of
factors commonly seen in practice, including the interaction
of poor dietary quality, low carbohydrate availability, ex-
cessive training, and acute dehydration on chronic energy
restriction. The challenge of detecting small but important
changes in sports performance is noted in all areas of sports
nutrition.28 EAL Question #2 summarizes the literature on
optimal timing, energy, and macronutrient characteristics of
a program supporting a gain in fat-free mass when in en-
ergy deficit (Table 1). Again the literature is limited in
quantity and range to allow definitive recommendations to
be made, although there is support for the benefits of in-
creased protein intake.
Macronutrient Requirements for Sport
Energy pathways and training adaptations. Guide-
lines for the timing and amount of intake of macronutrients in
the athlete_s diet should be underpinned by a fundamental
understanding of how training-nutrient interactions affect en-
ergy systems, substrate availability and training adaptations.
Exercise is fueled by an integrated series of energy systems
which include non-oxidative (phosphagen and glycolytic) and
aerobic (fat and carbohydrate oxidation) pathways, using
substrates that are both endogenous and exogenous in origin.
Adenosine triphosphate and phosphocreatine (phosphagen
system) provide a rapidly available energy source for muscu-
lar contraction, but not at sufficient levels to provide a con-
tinuous supply of energy for longer than ~10 seconds. The
anaerobic glycolytic pathway rapidly metabolizes glucose and
muscle glycogen through the glycolytic cascade and is the
primary pathway supporting high-intensity exercise last-
ing 10–180 seconds. Since neither the phosphagen nor the
http://www.acsm-msse.org548 Official Journal of the American College of Sports Medicine
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
glycolytic pathway can sustain energy demands to allow
muscles to contract at a very high rate for longer lasting
events, oxidative pathways provide the primary fuels for
events lasting longer than ~2 minutes. The major substrates
include muscle and liver glycogen, intramuscular lipid, adi-
pose tissue triglycerides, and amino acids frommuscle, blood,
liver and the gut. As oxygen becomes more available to the
working muscle, the body uses more of the aerobic (oxida-
tive) pathways and less of the anaerobic (phosphagen and
glycolytic) pathways. The greater dependence upon aerobic
pathways does not occur abruptly, nor is one pathway ever
relied on exclusively. The intensity, duration, frequency, type
of training, sex, and training level of the individual, as well as
prior nutrient intake and substrate availability, determine the
relative contribution of energy pathways and when crossover
between pathways occurs. For a more complete understand-
ing of fuel systems for exercise, the reader is directed to
specific texts.29
An athlete_s skeletal muscle has a remarkable plasticity to
respond quickly to mechanical loading and nutrient avail-
ability resulting in condition-specific metabolic and functional
adaptations.30 These adaptations influence performance nutri-
tion recommendations with the overarching goals that energy
systems should be trained to provide the most economical
support for the fuel demands of an event while other strategies
should achieve appropriate substrate availability during the
event itself. Adaptations that enhance metabolic flexibility
include increases in transport molecules that carry nutrients
across membranes or to the site of their utilization within the
muscle cell, increases in enzymes that activate or regulate
metabolic pathways, enhancement of the ability to tolerate the
side-products of metabolism and an increase in the size of
muscle fuel stores.3 While some muscle substrates (eg, body
fat) are present in relatively large quantities, others may need
to be manipulated according to specific needs (eg, carbohy-
drate supplementation to replace muscle glycogen stores).
Carbohydrate. Carbohydrate has rightfully received a
great deal of attention in sports nutrition due to a number of
special features of its role in the performance of, and adap-
tation to training. First, the size of body carbohydrate stores
is relatively limited and can be acutely manipulated on a
daily basis by dietary intake or even a single session of ex-
ercise.3 Second, carbohydrate provides a key fuel for the
brain and central nervous system and a versatile substrate for
muscular work where it can support exercise over a large
range of intensities due to its utilization by both anaerobic
and oxidative pathways. Even when working at the highest
intensities that can be supported by oxidative phosphoryla-
tion, carbohydratewas resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
R
an
d
o
m
 s
eq
u
en
ce
 
g
en
er
at
io
n
 (
se
le
ct
io
n
 b
ia
s)
 
A
ll
o
ca
ti
o
n
 c
o
n
ce
al
m
en
t 
(s
el
ec
ti
o
n
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
p
ar
ti
ci
p
an
ts
 
an
d
 p
er
so
n
n
el
 
(p
er
fo
rm
an
ce
 b
ia
s)
 
B
li
n
d
in
g
 o
f 
o
u
tc
o
m
e 
as
se
ss
m
en
t 
(d
et
ec
ti
o
n
 
b
ia
s)
 
In
co
m
p
le
te
 o
u
tc
o
m
e 
d
at
a 
(a
tt
ri
ti
o
n
 b
ia
s)
 
S
el
ec
ti
v
e 
re
p
o
rt
in
g
 
(r
ep
o
rt
in
g
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Ramírez-Campillo et al., 2015 [19]
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochrane Collaboration Guidelines [24]. The items on the list were divided into different domains: 
random sequence generation (selection bias), allocation concealment (selection bias), blinding of 
participants and personnel (performance bias), blinding of outcome assessment (detection bias), 
incomplete outcome data (attrition bias), selective reporting (reporting bias), and other types of bias. 
They were characterized as “low” if criteria for a low risk of bias were met (plausible bias unlikely to 
seriously alter the results) or “high” if criteria for a high risk of bias were met (plausible bias that 
seriously weakens confidence in the results). If the risk of bias was unknown, it was considered 
“unclear” (plausible bias that raises some doubts about the results). Full details are given in Table 1 
and Figure 1. 
Table 1. Risk of bias graph: review of authors’ judgements about each risk of bias item presented as 
percentages across all included studies. indicates low risk of bias, indicates unknown risk 
of bias, and indicates high risk of bias. 
 
Ra
nd
om
 se
qu
en
ce
 
ge
ne
ra
tio
n 
(s
el
ec
tio
n 
bi
as
) 
A
llo
ca
tio
n 
co
nc
ea
lm
en
t 
(s
el
ec
tio
n 
bi
as
) 
Bl
in
di
ng
 o
f p
ar
tic
ip
an
ts 
an
d 
pe
rs
on
ne
l 
(p
er
fo
rm
an
ce
 b
ia
s)
 
Bl
in
di
ng
 o
f o
ut
co
m
e 
as
se
ss
m
en
t (
de
te
ct
io
n 
bi
as
) 
In
co
m
pl
et
e 
ou
tc
om
e 
da
ta
 
(a
ttr
iti
on
 b
ia
s)
 
Se
le
ct
iv
e 
re
po
rti
ng
 
(re
po
rti
ng
 b
ia
s)
 
O
th
er
 b
ia
s 
Mujika et al. 2000 [17] 
 
Bemben et al. 2001 [6] 
 
Cox et al. 2002 [21] 
 
Biwer et al. 2003 [15] 
 
Ostojic 2004 [18] 
 
Claudino et al. 2014 [20] 
 
Williams et al. 2014 [16] 
 
Ramírez-Campillo et al. 2015 
[19] 
Yáñez-Silva et al. 2017 [7] 
 
 
 
Nutrients 2019, 11, x FOR PEER REVIEW 4 of 17 
The mean (M), standard deviation (SD), and sample size data were extracted by one author from 
the tables of all of the included papers (DMJ). Whenever necessary, we contacted the authors to obtain 
the data. When it was impossible, mean and SD were extrapolated from the figures. Any 
disagreement was resolved by consensus (JMA and DMJ) or third-party adjudication (JCG). 
2.4. Quality Assessment of the Experiments 
Methodological quality and risk of bias were assessed by two authors independently (JMA and 
DMJ), and disagreements were resolved by third-party evaluation (JCG), in accordance with the 
Cochraneoffers advantages over fat as a substrate
since it provides a greater yield of adenosine triphosphate per
volume of oxygen that can be delivered to the mitochondria,3
thus improving gross exercise efficiency.31 Third, there is
significant evidence that the performance of prolonged
sustained or intermittent high-intensity exercise is enhanced
by strategies that maintain high carbohydrate availability
(ie, match glycogen stores and blood glucose to the fuel
demands of exercise), while depletion of these stores is as-
sociated with fatigue in the form of reduced work rates,
impaired skill and concentration, and increased perception
of effort. These findings underpin the various performance
nutrition strategies, to be discussed subsequently, that sup-
ply carbohydrate before, during, and in the recovery be-
tween events to enhance carbohydrate availability.
Finally, recent work has identified that in addition to its
role as a muscle substrate, glycogen plays important direct
and indirect roles in regulating the muscle_s adaptation to
training. 32 The amount and localization of glycogen within
the muscle cell alters the physical, metabolic, and hormonal
environment in which the signaling responses to exercise are
exerted. Specifically, starting a bout of endurance exercise
with low muscle glycogen content (e.g. by undertaking a
second training session in the hours after the prior session
has depleted glycogen stores) produces a coordinated up-
regulation of the transcriptional and post-translational re-
sponses to exercise. A number of mechanisms underpin this
outcome including increasing the activity of molecules that
have a glycogen binding domain, increasing free fatty acid
availability, changing osmotic pressure in the muscle cell
and increasing catecholamine concentrations.32 Strategies
that restrict exogenous carbohydrate availability (e.g. exercising
in a fasted state or without carbohydrate intake during the ses-
sion) also promote an extended signaling response, albeit less
robustly than is the case for exercise with low endogenous
carbohydrate stores.33 These strategies enhance the cellular
outcomes of endurance training such as increased maximal
mitochondrial enzyme activities and/or mitochondrial content
and increased rates of lipid oxidation, with the augmentation of
responses likely to be explained by enhanced activation of key
cell signaling kinases (eg, AMPK, p38MAPK), transcription
factors (eg, p53, PPARC) and transcriptional co-activators (eg,
PGC-1>).33 Deliberate integration of such training-dietary
strategies (‘‘train low’’) within the periodized training pro-
gram is becoming a recognized,34 although potentially
misused,33 part of sports nutrition practice.
Individualized recommendations for daily intakes of car-
bohydrate should be made in consideration of the athlete_s
training/competition program and the relative importance of
undertaking it with high or low carbohydrate according to
the priority of promoting the performance of high quality
exercise versus enhancing the training stimulus or adapta-
tion, respectively. Unfortunately, we lack sophisticated in-
formation on the specific substrate requirements of many of
the training sessions undertaken by athletes; therefore we
must rely on guesswork, supported by information on work
requirements of exercise from technologies such as consumer-
based activity and heart rate monitors,35 power meters, and
global positioning systems.
General guidelines for the suggested intake of carbohydrate
to provide high carbohydrate availability for designated
training or competition sessions can be provided according
to the athlete_s body size (a proxy for the size of muscle
stores) and the characteristics of the session (Table 2). The
NUTRITION AND ATHLETIC PERFORMANCE Medicine & Science in Sports & Exercised 549
SPEC
IA
L
C
O
M
M
U
N
IC
ATIO
N
S
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
timing of carbohydrate intake over the day and in relation to
training can also be manipulated to promote or reduce car-
bohydrate availability.36 Strategies to enhance carbohydrate
availability are covered in more detail in relation to compe-
tition eating strategies. Nevertheless, these fueling practices
are also important for supporting the high quality workouts
within the periodized training program. Furthermore, it is
intuitive that they add value in fine-tuning intended event
eating strategies, and for promoting adaptations such as gas-
trointestinal tolerance and enhanced intestinal absorption37
that allow competition strategies to be fully effective. During
other sessions of the training program, it may be less impor-
tant to achieve high carbohydrate availability, or there may be
some value in deliberately exercising with low carbohydrate
availability to enhance the training stimulus or adaptive re-
sponse. Various tactics can be used to permit or promote low
carbohydrate availability including reducing total carbohy-
drate intake or manipulating the timing of training in relation
to carbohydrate intake (eg, training in a fasted state, under-
taking two bouts of exercise in close proximity without op-
portunity for refueling between sessions).38
Specific questions examined via the evidence analysis on
carbohydrate needs for training are summarized in Table 2
and show good evidence that neither the glycemic load
nor glycemic index of carbohydrate-rich meals affects
the metabolic nor performance outcomes of training once
TABLE 2. Summary of guidelines for carbohydrate intake by athletes.36
Situation Carbohydrate Targets Comments on Type and Timing of Carbohydrate Intake
DAILY NEEDS FOR FUEL AND RECOVERY
1. The following targets are intended to provide high carbohydrate availability (ie, to meet the carbohydrate needs of the muscle and central nervous system) for different exercise
loads for scenarios where it is important to exercise with high quality and/or at high intensity. These general recommendations should be fine-tuned with individual consideration
of total energy needs, specific training needs and feedback from training performance.
2. On other occasions, when exercise quality or intensity is less important, it may be less important to achieve these carbohydrate targets or to arrange carbohydrate intake over the
day to optimise availability for specific sessions. In these cases, carbohydrate intake may be chosen to suit energy goals, food preferences, or food availability.
3. In some scenarios, when the focus is on enhancing the training stimulus or adaptive response, low carbohydrate availability may be deliberately achieved by reducing total
carbohydrate intake, or by manipulating carbohydrate intake related to training sessions (eg, training in a fasted state, undertaking a second session of exercise without
adequate opportunity for refuelling after the first session).
Light � Low intensity or skill-based
activities
3–5 g/kg of athlete_s body
weight/d
� Timing of intake of carbohydrate over the day may
be manipulated to promote high carbohydrate availability
for a specific session by consuming carbohydrate before
or during the session, or in recovery from a previous session.
Moderate � Moderate exercise program
(eg, ~1 h per day)
5–7 g/kg/d
� Otherwise, as long as total fuel needs are provided, the pattern of
intake may simply be guided by convenience and individual choice.
High � Endurance program (eg, 1–3 h/d
mod-high-intensity exercise)
6–10 g/kg/d
� Athletes should choose nutrient-rich carbohydrate sources to
allow overall nutrient needs to be met.
Very High � Extreme commitment (eg, 94–5 h/d
mod-high intensity exercise
8–12 g/kg/d
ACUTE FUELLING STRATEGIES – these guidelines promote high carbohydrate availability to promote optimal performance in competition or key training sessions
General fuelling up � Preparation for events G 90 min
exercise
7–12 g/kg per 24 h as for
daily fuel needs
� Athletes may choose carbohydrate-richsources that are low in
fiber/residue and easily consumed to ensure that fuel targets are
met, and to meet goals for gut comfort or lighter ‘‘racing weight’’.Carbohydrate loading � Preparation for events 9 90 min of
sustained/intermittent exercise
36–48 h of 10–12 g/kg body
weight per 24 h
Speedy refuelling � G8 h recovery between 2 fuel
demanding sessions
1–1.2 g/kg/h for first 4 h then
resume daily fuel needs
� There may be benefits in consuming small regular snacks
� Carbohydrate rich foods and drink may help to ensure that
fuel targets are met.
Pre-event fuelling � Before exercise 9 60 min 1–4 g/kg consumed 1–4 h
before exercise
� Timing, amount and type of carbohydrate foods and drinks should
be chosen to suit the practical needs of the event and individual
preferences/experiences.
� Choices high in fat/protein/fiber may need to be avoided to
reduce risk of gastrointestinal issues during the event.
� Low glycemic index choices may provide a more sustained source
of fuel for situations where carbohydrate cannot be consumed
during exercise.
During brief exercise � G45 min Not needed
During sustained high intensity
exercise
� 45–75 min Small amounts including
mouth rinse
� A range of drinks and sports products can provide easily
consumed carbohydrate.
� The frequent contact of carbohydrate with the mouth and
oral cavity can stimulate parts of the brain and central nervous
system to enhance perceptions of well-being and increase
self-chosen work outputs.
During endurance exercise
including ‘‘stop and start’’ sports
� 1–2.5 h 30–60 g/h � Carbohydrate intake provides a source of fuel for the muscle to
supplement endogenous stores.
� Opportunities to consume foods and drinks vary according to the
rules and nature of each sport.
� A range of everyday dietary choices and specialised sports
products ranging in form from liquid to solid may be useful
� The athlete should practice to find a refuelling plan that suits their
individual goals including hydration needs and gut comfort.
During ultra-endurance exercise � 92.5–3 h Up to 90 g/h � As above.
� Higher intakes of carbohydrate are associated with better
performance.
� Products providing multiple transportable carbohydrates
(Glucose:fructose mixtures) achieve high rates of oxidation of
carbohydrate consumed during exercise.
http://www.acsm-msse.org550 Official Journal of the American College of Sports Medicine
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
carbohydrate and energy content of the diet have been taken
into account (Question #11). Furthermore, although there is
sound theory behind the metabolic advantages of exercising
with low carbohydrate availability on training adaptations,
the benefits to performance outcomes are currently unclear
(Table 1, Question #10). This possibly relates to the limita-
tions of the few available studies in which poor periodization
of this tactic within the training program has meant that any
advantages to training adaptations have been counteracted by
the reduction in training intensity and quality associated with
low carbohydrate variability. Therefore, a more sophisticated
approach is needed to integrate this training/nutrient interac-
tion into the larger training program.33 Finally, while there is
support for consuming multiple carbohydrates to facilitate
more rapid absorption, evidence to support the choice of
special blends of carbohydrate to support increased carbo-
hydrate oxidation during training sessions is premature
(Question #9).
Protein. Dietary protein interacts with exercise, provid-
ing both a trigger and a substrate for the synthesis of con-
tractile and metabolic proteins39,40 as well as enhancing
structural changes in non-muscle tissues such as tendons41
and bones.42 Adaptations are thought to occur by stimula-
tion of the activity of the protein synthetic machinery in
response to a rise in leucine concentrations and the provision
of an exogenous source of amino acids for incorporation into
new proteins.43 Studies of the response to resistance training
show upregulation of muscle protein synthesis (MPS) for
at least 24 hours in response to a single session of exercise,
with increased sensitivity to the intake of dietary protein
over this period.44 This contributes to improvements in
skeletal muscle protein accretion observed in prospective
studies that incorporate multiple protein feedings after
exercise and throughout the day. Similar responses occur
following aerobic exercise or other exercise types (eg, in-
termittent sprint activities and concurrent exercise), albeit
with potential differences in the type of proteins that are
synthesized. Recent recommendations have underscored the
importance of well-timed protein intake for all athletes even
if muscle hypertrophy is not the primary training goal, and
there is now good rationale for recommending daily protein
intakes that are well above the RDA39 to maximize meta-
bolic adaptation to training.40
Although classical nitrogen balance work has been useful
for determining protein requirements to prevent deficiency
in sedentary humans in energy balance,45 athletes do not
meet this profile and achieving nitrogen balance is second-
ary to an athlete with the primary goal of adaptation to
training and performance improvement.40 The modern view
for establishing recommendations for protein intake in ath-
letes extends beyond the DRIs. Focus has clearly shifted to
evaluating the benefits of providing enough protein at opti-
mal times to support tissues with rapid turnover and aug-
ment metabolic adaptations initiated by training stimulus.
Future research will further refine recommendations directed
at total daily amounts, timing strategies, quality of protein
intake, and provide new recommendations for protein sup-
plements derived from various protein sources.
Protein needs. Current data suggest that dietary protein
intake necessary to support metabolic adaptation, repair,
remodeling, and for protein turnover generally ranges from
1.2 to 2.0 g/kg/d. Higher intakes may be indicated for short
periods during intensified training or when reducing energy
intake.26,39 Daily protein intake goals should be met with a
meal plan providing a regular spread of moderate amounts
of high-quality protein across the day and following stren-
uous training sessions. These recommendations encompass
most training regimens and allow for flexible adjustments with
periodized training and experience.46,47 Although general
daily ranges are provided, individuals should no longer be
solely categorized as strength or endurance athletes and pro-
vided with static daily protein intake targets. Rather, guide-
lines should be based around optimal adaptation to specific
sessions of training/competition within a periodized program,
underpinned by an appreciation of the larger context of ath-
letic goals, nutrient needs, energy considerations, and food
choices. Requirements can fluctuate based on ‘‘trained’’ status
(experienced athletes requiring less), training (sessions in-
volving higher frequency and intensity, or a new training
stimulus at higher end of protein range), carbohydrate avail-
ability, and most importantly, energy availability.46,48 The
consumption of adequate energy, particularly from carbohy-
drates, to match energy expenditure, is important so that amino
acids are spared for protein synthesis and not oxidized.49 In
cases of energy restriction or sudden inactivity as occurs as a
result of injury, elevated protein intakes as high as 2.0 g/kg/day
or higher26,50 when spread over the day may be advantageous
in preventing fat-free mass loss.39 More detailed reviews of
factors that influence changing protein needs and their rela-
tionship to changes in protein metabolism and body composi-
tion goals can be found elsewhere.51,52
Protein timing as a trigger for metabolic adap-
tation. Laboratory based studies show that MPS is opti-
mized in response to exercise by the consumptionof high
biological value protein, providing ~10 g essential amino
acids in the early recovery phase (0–2 h after exercise).40,53
This translates to a recommended protein intake of 0.25–
0.3 g/kg body weight or 15–25 g protein across the typical
range of athlete body sizes, although the guidelines may need
to be fine-tuned for athletes at extreme ends of the weight
spectrum.54 Higher doses (ie, 940 g dietary protein) have not
yet been shown to further augment MPS and may only be
prudent for the largest athletes, or during weight loss.54 The
exercise-enhancement of MPS, determined by the timing and
pattern of protein intake, responds to further intake of protein
within the 24-hour period after exercise,55 and may ultimately
translate into chronic muscle protein accretion and functional
change. While protein timing affects MPS rates, the magnitude
of mass and strength changes over time are less clear.56
However, longitudinal training studies currently suggest that
increases in strength and muscle mass are greatest with im-
mediate post-exercise provision of protein.57
NUTRITION AND ATHLETIC PERFORMANCE Medicine & Science in Sports & Exercised 551
SPEC
IA
L
C
O
M
M
U
N
IC
ATIO
N
S
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Whereas traditional protein intake guidelines focused on
total protein intake over the day (g/kg), newer recommenda-
tions now highlight that the muscle adaptation to training can
be maximized by ingesting these targets as 0.3 g/kg body
weight after key exercise sessions and every 3–5 hours over
multiple meals.47,54,58 Table 1, Question #8 summarizes the
weight of the current literature of consuming protein on
protein-specific metabolic responses during recovery.
Optimal protein sources. High-quality dietary pro-
teins are effective for the maintenance, repair, and synthesis
of skeletal muscle proteins.59 Chronic training studies have
shown that the consumption of milk-based protein after re-
sistance exercise is effective in increasing muscle strength
and favorable changes in body composition.57,60,61 In addi-
tion, there are reports of increased MPS and protein accre-
tion with whole milk, lean meat, and dietary supplements,
some of which provide the isolated proteins whey, casein,
soy, and egg. To date, dairy proteins seem to be superior to
other tested proteins, largely due to leucine content and the
digestion and absorptive kinetics of branched-chain amino
acids in fluid-based dairy foods.62 However, further studies
are warranted to assess other intact high-quality protein
sources (eg, egg, beef, pork, concentrated vegetable protein)
and mixed meals on mTOR stimulation and MPS following
various modes of exercise. When whole food protein sources
are not convenient or available, then portable, third-party
tested dietary supplements with high-quality ingredients
may serve as a practical alternative to help athletes meet
their protein needs. It is important to conduct a thorough
assessment of the athlete_s specific nutrition goals when
considering protein supplements. Recommendations regard-
ing protein supplements should be conservative and primarily
directed at optimizing recovery and adaptation to training
while continuing to focus on strategies to improve or maintain
overall diet quality.
Fat. Fat is a necessary component of a healthy diet, pro-
viding energy, essential elements of cell membranes and
facilitation of the absorption of fat-soluble vitamins. The
Dietary Guidelines for Americans, 2015–20208 and Eating
Well with Canada_s Food Guide63 have made recommenda-
tions that the proportion of energy from saturated fats be
limited to less than 10% and include sources of essential fatty
acids to meet adequate intake (AI) recommendations. Intake
of fat by athletes should be in accordance with public health
guidelines and should be individualized based on training
level and body composition goals.46
Fat, in the form of plasma free fatty acids, intramuscular
triglycerides and adipose tissue provides a fuel substrate that
is both relatively plentiful and increased in availability to the
muscle as a result of endurance training. However, exercise-
induced adaptations do not appear to maximize oxidation
rates since they can be further enhanced by dietary strategies
such as fasting, acute pre-exercise intake of fat and chronic
exposure to high-fat, low-carbohydrate diets.3 Although
there has been historical64 and recently revived65 interest in
chronic adaptation to high-fat low carbohydrate diets, the
present evidence suggests that enhanced rates of fat oxidation
can only match exercise capacity/performance achieved by
diets or strategies promoting high carbohydrate availability at
moderate intensities,64 while the performance of exercise at
the higher intensities is impaired.64,66 This appears to occur
as a result of a down-regulation of carbohydrate metabolism
even when glycogen is available.67 Further research is
warranted both in view of the current discussions65 and the
failure of current studies to include an adequate control diet
that includes contemporary periodized dietary approaches.68
Although specific scenarios may exist where high-fat diets
may offer some benefits or at least the absence of disadvan-
tages for performance, in general they appear to reduce rather
than enhance metabolic flexibility by reducing carbohydrate
availability and capacity to use it effectively as an exercise
substrate. Therefore, competitive athletes would be unwise to
sacrifice their ability to undertake high-quality training or
high-intensity efforts during competition that could deter-
mine the outcome.68
Conversely, athletes may choose to excessively restrict
their fat intake in an effort to lose body weight or improve
body composition. Athletes should be discouraged from
chronic implementation of fat intakes below 20% of energy
intake since the reduction in dietary variety often associated
with such restrictions is likely to reduce the intake of a va-
riety of nutrients such as fat-soluble vitamins and essential
fatty acids,9 especially n-3 fatty acids. If such focused re-
strictiveness around fat intake is practiced, it should be
limited to acute scenarios such as the pre-event diet or
carbohydrate-loading where considerations of preferred
macronutrients or gastrointestinal comfort have priority.
Alcohol. Alcohol consumption may be part of a well-
chosen diet and social interactions, but excessive alcohol
consistent with binge drinking patterns is a concerning be-
havior observed among some athletes, particularly in team
sports.69 Misuse of alcohol can interfere with athletic goals in
a variety of ways related to the negative effects of acute intake
of alcohol on the performance of, or recovery from, exercise,
or the chronic effects of binge drinking on health and man-
agement of body composition.70 Besides the calorie load of
alcohol (7 kcal/g), alcohol suppresses lipid oxidation, in-
creases unplanned food consumption and may compromise
the achievement of body composition goals. Research in this
area is fraught with study design concerns that limit direct
translation to athletes.
Available evidence warns against intake of significant
amounts of alcohol in the pre-exercise period and during
training due to the direct negative effects of alcohol on exercise
metabolism, thermoregulation, and skills/concentration.69 The
effects of alcohol on strength and performance may persist for
several hours even after signs and symptoms of intoxication
or hangover are no longer present. In the post-exercise phase,
where cultural patterns in sport often promote alcohol use,
alcohol may interfere with recovery by impairing glycogen
storage, 71 slowing rates of rehydration via its suppressive
effect on anti-diuretic hormone,72 and impairing the MPS
http://www.acsm-msse.org552 Official Journal of the American College of Sports Medicine
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproductionof this article is prohibited.
desired for adaptation and repair.69,73,74 In cold environments,
alcohol consumption increases peripheral vasodilation result-
ing in core temperature dysregulation75 and there are likely
to be other effects on body function such as disturbances in
acid–base balance and cytokine-prostaglandin pathways,
and compromised glucose metabolism and cardiovascular
function.76 Binge drinking may indirectly affect recovery
goals due to inattention to guidelines for recovery. Binge
drinking is also associated with high-risk behaviors leading
to accidents and anti-social behaviors that can be detrimental
to the athlete. In conclusion, athletes are advised to consider
both public health guidelines and team rules regarding use of
alcohol and are encouraged to minimize or avoid alcohol
consumption in the post-exercise period when issues of re-
covery and injury repair are a priority.
Micronutrients. Exercise stresses many of the meta-
bolic pathways in which micronutrients are required, and
training may result in muscle biochemical adaptations that
increase the need for some micronutrients. Athletes who
frequently restrict energy intake, rely on extreme weight-loss
practices, eliminate one ormore food groups from their diet, or
consume poorly chosen diets, may consume sub-optimal
amounts of micronutrients and benefit from micronutrient
supplementation.77 This occurs most frequently in the case of
calcium, vitamin D, iron, and some antioxidants.78–80 Single-
micronutrient supplements are generally only appropriate for
correction of a clinically-defined medical reason [eg, iron
supplements for iron deficiency anemia (IDA)].
Micronutrients of key interest: Iron. Iron deficiency,
with or without anemia, can impair muscle function and limit
work capacity 78,81 leading to compromised training adapta-
tion and athletic performance. Suboptimal iron status often
results from limited iron intake from heme food sources and
inadequate energy intake (approximately 6 mg iron is con-
sumed per ~1,000 kcals).82 Periods of rapid growth, training
at high altitudes, menstrual blood loss, foot-strike hemoly-
sis, blood donation, or injury can negatively impact iron
status.79,81 Some athletes in intense training may also have
increased iron losses in sweat, urine, feces, and from in-
travascular hemolysis.
Regardless of the etiology, a compromised iron status can
negatively impact health, physical and mental performance,
and warrants prompt medical intervention and monitoring.83
Iron requirements for all female athletes may be increased by
up to 70% of the estimated average requirement.84 Athletes
who are at greatest risk, such as distance runners, vegetarian
athletes, or regular blood donors should be screened regu-
larly and aim for an iron intake greater than their RDA
(ie, 918 mg for women and 98 mg for men).81,85
Athletes with iron deficiency anemia (IDA) should seek
clinical follow up, with therapies including oral iron sup-
plementation,86 improvements in diet and a possible reduc-
tion in activities that impact iron loss (eg, blood donation, a
reduction in weight bearing training to lessen erythrocyte he-
molysis).87 The intake of iron supplements in the period im-
mediately after strenuous exercise is contra-indicated since
there is the potential for elevated hepcidin levels to interfere with
iron absorption.88 Reversing IDA can require 3 to 6 months;
therefore, it is advantageous to begin nutrition intervention
before IDA develops.78,81 Athletes who are concerned about
iron status or have iron deficiency without anemia (eg, low
ferritin without IDA) should adopt eating strategies that pro-
mote an increased intake of food sources of well-absorbed
iron (eg, heme iron, non-heme iron + vitamin C foods) as the
first line of defense. Although there is some evidence that iron
supplements can achieve performance improvements in ath-
letes with iron depletion who are not anemic,89 athletes
should be educated that routine, unmonitored supplementa-
tion is not recommended, not considered ergogenic without
clinical evidence of iron depletion, and may cause unwanted
gastrointestinal distress.89
Some athletes may experience a transient decrease in
hemoglobin at the initiation of training due to hemodilu-
tion, known as ‘‘dilutional’’ or ‘‘sports anemia’’, and may
not respond to nutrition intervention. These changes ap-
pear to be a beneficial adaptation to aerobic training and do
not negatively impact performance.79 There is no agree-
ment on the serum ferritin level that corresponds to a
problematic level of iron depletion/deficiency, with var-
ious suggestions ranging from G10 to G35 ng/mL.86 A
thorough clinical evaluation in this scenario is warranted
since ferritin is an acute-phase protein that increases with in-
flammation, but in the absence of inflammation, still serves as
the best early indicator of compromised iron status. Other
markers of iron status and other issues in ironmetabolism (e.g.
the role of hepcidin) are currently being explored.88
Micronutrients of key interest: Vitamin D. Vitamin
D regulates calcium and phosphorus absorption and metabo-
lism, and plays a key role in maintaining bone health. There is
also emerging scientific interest in the biomolecular role of
vitamin D in skeletal muscle90 where its role in mediating
muscle metabolic function91 may have implications for
supporting athletic performance. A growing number of stud-
ies have documented the relationship between vitamin D
status and injury prevention,92 rehabilitation,93 improved
neuromuscular function,94 increased type II muscle fiber
size,94 reduced inflammation,93 decreased risk of stress frac-
ture,92,95 and acute respiratory illness.95
Athletes who live at latitudes 935th parallel or who pri-
marily train and compete indoors are likely at higher risk for
vitamin D insufficiency (25(OH) D = 50-75 nmol/L) and
deficiency (25(OH) D G50 nmol/L). Other factors and life-
style habits such as dark complexion, high body fat content,
undertaking of training in the early morning and evening
when UVB levels are low, and aggressive blocking of UVB
exposure (clothing, equipment, and screening/blocking lo-
tions) increase the risk for insufficiency and deficiency.93
Since athletes tend to consume little vitamin D from the
diet93 and dietary interventions alone have not been shown
to be a reliable means to resolve insufficient status,96 supple-
mentation above the current RDA and/or responsible UVB
exposure may be required to maintain sufficient vitamin D
NUTRITION AND ATHLETIC PERFORMANCE Medicine & Science in Sports & Exercised 553
SPEC
IA
L
C
O
M
M
U
N
IC
ATIO
N
S
Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
status. A recent study of NCAA Division 1 swimmers and
divers reported that athletes who started at 130 nmol/L and
received daily doses of 4,000 IU of vitamin D (100 Kg)
were able to maintain sufficient status over 6 months (mean
change +2.5 nmol/L), while athletes receiving placebo ex-
perienced a mean loss of 50 nmol/L.97 Unfortunately, de-
termining vitamin D requirements for optimal health and
performance is a complex process. Vitamin D blood levels
from 80 nmol/L and up to 100 nmol/L93 to 125 nmol/L94
have been recognized as prudent goals for optimal training
induced adaptation. Although proper assessment and cor-
rection of deficiency is likely vital to athlete well-being and
athletic success, current data do not support vitamin D as an
ergogenic aid for athletes. Empirical data are still needed to
elucidate the direct role of vitamin D in musculoskeletal
health and function to help refine recommendations for ath-
letes. Until then, athletes with a history of stress fracture, bone
or joint injury, signs of over training, muscle pain or weak-
ness, and a lifestyle involving low exposure to UVB may re-
quire 25(OH)D assessment98 to determine if an individualized
vitamin D supplementation protocol is required.
Micronutrients