Content uploaded by Asker Jeukendrup
Author content
All content in this area was uploaded by Asker Jeukendrup on Jan 31, 2018
Content may be subject to copyright.
Optimal Composition of Fluid-Replacement
Beverages
LindsayB.Baker
1and Asker E. Jeukendrup*1,2
ABSTRACT
The objective of this article is to provide a review of the fundamental aspects of body fluid balance
and the physiological consequences of water imbalances, as well as discuss considerations for
the optimal composition of a fluid replacement beverage across a broad range of applications.
Early pioneering research involving fluid replacement in persons suffering from diarrheal disease
and in military, occupational, and athlete populations incurring exercise- and/or heat-induced
sweat losses has provided much of the insight regarding basic principles on beverage palatability,
voluntary fluid intake, fluid absorption, and fluid retention. We review this work and also discuss
more recent advances in the understanding of fluid replacement as it applies to various popu-
lations (military, athletes, occupational, men, women, children, and older adults) and situations
(pathophysiological factors, spaceflight, bed rest, long plane flights, heat stress, altitude/cold
exposure, and recreational exercise). We discuss how beverage carbohydrate and electrolytes
impact fluid replacement. We also discuss nutrients and compounds that are often included in
fluid-replacement beverages to augment physiological functions unrelated to hydration, such as
the provision of energy. The optimal composition of a fluid-replacement beverage depends upon
the source of the fluid loss, whether from sweat, urine, respiration, or diarrhea/vomiting. It is
also apparent that the optimal fluid-replacement beverage is one that is customized accord-
ing to specific physiological needs, environmental conditions, desired benefits, and individual
characteristics and taste preferences. C
2014 American Physiological Society. Compr Physiol
4:575-620, 2014.
Introduction
Body water is a large constituent of the human body and
is important for optimal physiological function and health. In
general, humans drink adequately to maintain body water bal-
ance, that is, fluid replacement to offset losses, under normal
resting, nonstressful conditions. However, situations often
arise in which perturbations in fluid balance occur as a result
of large fluid losses from the body. For example, heat exposure
in an occupational setting or exercise for competitive sport or
recreation can induce significant increases in thermal sweat
loss. Other conditions, such as cold and/or altitude exposure
may increase body water loss through respiration and/or urina-
tion. Upon exposure to microgravity, either due to spaceflight
or bed rest, plasma volume is significantly contracted. In addi-
tion, individuals suffering from diarrheal disease can become
severely dehydrated from fecal water loss. Fluid replacement
in all of these cases is important to prevent or at least min-
imize the detrimental effects of body water deficits and the
only way to replace fluid loss is by drinking. One form of fluid
replacement is plain drinking water. However, oftentimes the
optimal fluid replacement beverage includes carbohydrate,
electrolytes, and/or other ingredients to improve palatability,
stimulate thirst, speed intestinal fluid absorption, and pro-
mote fluid retention. Optimal composition may depend upon
the source of the fluid loss, whether from sweat, urine, res-
piration, or diarrhea/vomiting with additional consideration
for the population and environmental conditions. While the
main objective of a fluid replacement beverage is to replace
water losses, the beverage may also serve as a vehicle for the
provision of nutrients or compounds to augment physiological
function.
The purpose of this article is to review the basic scientific
principles of fluid replacement and provide insight regard-
ing the optimal composition of a fluid replacement beverage
across a broad range of applications. Specifically, this arti-
cle reviews: (i) the early pioneering research involving fluid
replacement in humans; (ii) the basic physiology of body fluid
balance, including its composition, distribution, homeostatic
regulation, and physiological effects of a deficit; (iii) sources
and rates of water loss from the body, both at rest and as a
result various physiological and environmental stressors; (iv)
the physiological processes involved in fluid replacement,
including thirst and voluntary fluid intake, intestinal fluid
absorption, and fluid retention; (v) common ingredients of
a fluid replacement beverage and their potential benefits; and
*Correspondence to asker.jeukendrup@pepsico.com
1Gatorade Sports Science Institute, Barrington, Illinois
2School of Sport and Exercise Sciences, University of Birmingham,
Edgbaston, Birmingham, United Kingdom
Published online, April 2014 (comprehensivephysiology.com)
DOI: 10.1002/cphy.c130014
Copyright CAmerican Physiological Society.
Volume 4, April 2014 575
Optimal Composition of Fluid-Replacement Beverages Comprehensive Physiology
(vi) beverage composition considerations for specific appli-
cations and populations.
Historical Roots of Fluid-Replacement
Beverages
Pioneering work for military needs
Some of the earliest reports of the effect of hydration on work
capacity are from military combats. Military personnel are
often exposed to extreme environmental conditions, such as
hot and dry desert conditions or hot and humid tropical and
subtropical regions (3). In these hot weather operations the
battles could be decided by the availability of water to combat
troops. Several authors have commented on the impact of heat
illness on military campaigns and how wars have been won
or lost by blocking access to water supplies (230). Although
exact numbers were difficult to track, hypohydration-induced
heat stroke was believed (according to intelligence reports)
to account for a very high number of deaths in several wars,
including 11 per 10,000 troops per week in the desert area
of the Persian Gulf Command in the Middle Eastern Theater
in 1943 (510), 378 per 10,000 troops in the month of May
during the Vietnam war (482), and 20,000 Egyptian soldier
deaths during the 1967 Six-Day War with Israel (230). These
causality numbers are striking and perhaps preventable. There
are also likely incidences of heat illness not severe enough to
cause death, but influence combat effectiveness and impact on
the success of a military campaign (230, 323). While several
factors are involved in the etiology of heat illness, including
environmental conditions and level of physical activity (which
are difficult to control since these factors are dictated by the
tactical situation), fluid replacement is also an important factor
in body temperature regulation, and risk for heat illness (3).
The 1940s marked a time when the military devoted a
significant amount of time and energy to better understand
the fluid replacement needs of soldiers operating in desert
environments as well as how to assist soldiers when trapped
with limited water. In 1944, Pitts et al. (375) were the first to
experimentally compare the progressive changes in physio-
logical function when men marched on a treadmill in the heat
with fluid replacement to match sweat loss, ad libitum fluid
replacement, or no fluid replacement (see Fig. 1). The main
findings from their work were that subjects only voluntar-
ily drank about two-thirds of sweat loss, drinking ad libitum
resulted in lower heart rates and rectal temperatures com-
pared to drinking nothing, and drinking water, a 0.2% saline,
or ∼3% glucose solution in a volume to match sweat losses
resulted in even lower rectal temperatures and heart rates than
ad libitum drinking. The main conclusion by the authors was
that the best performance is achieved by replacing water loss
in a volume to match sweat (375).
A few years later Adolph and associates (3) published
a comprehensive report on man’s physiological responses to
marching in the desert. With U.S. warfare beginning in the
African deserts in 1941, the need for scientific investigation
arose. The work of Adolph and associates began in 1942 to
address the following basic questions from military leaders:
what are man’s water needs in the desert and what happens if
they are not met (2)? Adolph and associates’ 1947 report was
the first to address these questions. Many important insights
were gained as a result of this pioneering work conducted
103.0 Key
No water
Zone of
impending
exhaustion
Zone of
comfort
Water adlib
Water equals sweat
102.5
102.0
Rectal temperature °F
101.5
101.0
100.5
100.0
Rest
Rest
Rest
Rest
Rest
Rest
Rest
0.5 1.0
Hours on the march up a grade of 2.5% at 3.5 m.p.h.
2.0 3.0 4.0 5.0 6.0
Figure 1 Effect of no water vs. ad libitum water intake vs. water intake to match sweat
losses on rectal temperature during marching in the heat (37.8◦C, 35%-45% relative humid-
ity) in 1 male subject. Reprinted (with permission) from Pitts et al. (375).
576 Volume 4, April 2014
Comprehensive Physiology Optimal Composition of Fluid-Replacement Beverages
at the Armored Medical Research Laboratory in the desert
area of Freda, California and the Desert Laboratory Unit at
the University of Rochester. Here, they confirmed the find-
ings of incomplete water replacement when drinking ad libi-
tum [termed “voluntary hypohydration” by Rothstein et al.
(400)], that hypohydration results in early onset of fatigue,
and that fluid replacement improved thermal and cardiovascu-
lar function and delayed the onset of fatigue during prolonged
marching in the desert (62, 399). In addition, in 1947, it was
found that drinking a sodium-containing fluid improved ad
libitum fluid intake and reduced renal water excretion (400).
Fluid intake was also improved by cooling and flavoring the
water. They also characterized the physiological and psy-
chological effects of a water shortage, mostly due to inade-
quate circulation, with exhaustion from hypohydration occur-
ring at 5-6% body weight deficit (62, 399). Their work also
helped dispel the widespread belief among the military that
man could be trained to require less fluid intake during heat
exposure (3).
Occupational settings
Observations on fluid loss and replacement in occupational
settings also contributed significantly to the early develop-
ment of the field. Extensive literature (primarily from the
1920s and 1930s) described the sweating rates, sweat com-
position, and dietary intake of workers, as well as the phys-
iological consequences of labor, in extreme environmental
conditions. Of greatest importance was the observation that
profuse sweating (>1 L/h, but as high as ≥2.5 L/h) by the
workers in extreme heat resulted in loss of water and sodium
chloride which was associated with mild to severe, debilitat-
ing muscle cramps (“heat cramps”) (332, 479, 480). One of
the earliest reports of heat cramps in the occupational setting
was in 1878, from the gold mines in Virginia City, Nevada,
where miners worked in environmental temperatures as high
as 48 to 54◦C (276). Many other reports of heat cramping
among miners, stokers, iron or steel workers, and agricultural
workers followed (102, 136, 151, 332, 370, 480). Importantly,
included in many of these reports was the observation that the
ingestion or intravenous administration of sodium chloride
solution was a highly successful therapeutic agent in resolv-
ing heat cramps (151, 479, 480). Moreover, the importance
of daily sodium chloride intake to replace that lost in sweat
was well-recognized; as workers were often provided with
salted drinking water, saline, fresh cow’s milk, barley water,
or salted beer or encouraged to liberally salt their food (313,
479, 480).
Survival from diarrheal illness
Another important step in the understanding of the opti-
mal formulation of a fluid replacement beverage came with
research in 1950s and 1960s showing that the cotransport of
glucose and sodium accelerated water absorption in the small
intestine (277). This was groundbreaking in the fight against
diarrheal-induced hypohydration (from cholera) in develop-
ing countries. Survivability from diarrheal disease was likely
if intravenous fluids were given promptly. However, Darrow
(126) suggested that an orally administered electrolyte solu-
tion with glucose could be given to treat patients when intra-
venous fluids were unavailable. This solution proved very
practical and effective and is considered an important medical
advancement in the treatment of illnesses involving diarrhea
and vomiting (277, 278). The mechanism (sodium transport
and glucose transport are coupled so glucose accelerates the
absorption of solute and water) was later discovered by inves-
tigations in the 1950s and 1960s. This was first observed in
the guinea pig small intestine in 1958 (390) and in humans in
the early 1960s (277, 423). The first detailed description of
the mechanism was proposed as the sodium-glucose cotrans-
port process by Bob Crane in 1960 at the Symposium on
Membrane Transport and Metabolism in Prague (120). This
discovery led to a standardized oral replacement solution by
the World Health Organization; but can also be applied to for-
mulation of all fluid replacement beverages as intestinal fluid
absorption is a key process involved in rehydration (discussed
in more detail later).
Sports performance
Another application for fluid-replacement beverages is for
athletes participating in prolonged heavy exercise that induces
sweat loss. This work was pioneered by a series of studies by
Cade et al. in the early 1970s (74,75). In 1971, they reported
the effects of exercise in a hot/humid environment on body
fluid changes in ten players of the University of Florida foot-
ball team during a vigorous 2 h practice session with no fluid
intake (74). They found that body weight, extracellular fluid
volume, and plasma volume decreased by an average of 2.9%,
11%, and 7% respectively. This work was followed by a study
in 1972 (75) to determine whether the fluid deficits affected
performance and whether fluid replacement could prevent the
disturbances in physiology and performance. Cade et al. (75)
found that performance during a standardized walk-run test
(7 mile course) at an ambient temperature of 32 to 34◦Cwas
best when ∼1.0 L of a glucose (3%)-electrolyte (17 mmol/L
sodium, 3.5 mmol/L K, and 12 mmol/L chloride) solution
was consumed, intermediate when the athletes drank 0.1%
saline (1 L), and poorest when they drank nothing. Cade et al.
(75) concluded that the effects of water and carbohydrate
were striking. This early work of Cade et al. eventually led
to the invention of sports drinks. Since the publication of
these seminal studies in football players considerable insight
has been gleaned from research testing the effect of beverage
formulation on the delivery of fluid, electrolytes, and carbo-
hydrate energy to the body during exercise. In the discussion
that follows, water, sodium, and carbohydrate will be the
primary topics covered, but we also discuss other beverage
composition factors that have been hypothesized to impact
physiological function or physical/cognitive performance.
Volume 4, April 2014 577
Optimal Composition of Fluid-Replacement Beverages Comprehensive Physiology
Physiological Review of
Fluid Replacement
Body water
Water functions as a solvent for organic and inorganic mate-
rials and provides a medium for biochemical reactions and
for transportation of solutes throughout the body among var-
ious tissues, supplying nutrients and removing waste. Water
accounts for approximately 60% of total body mass in the
average adult, but ranges from 45% to 75% depending on body
composition, age, and sex (13). The variation due to body
composition is because fat-free mass has a much higher water
content (∼70%-80%) than adipose tissue (∼10%) (496). Total
body water can be divided into two compartments—the intra-
cellular water and the extracellular water. The intracellular
compartment accounts for approximately 55% to 65% of total
body water, while the extracellular compartment accounts
for the remaining 35% to 45% (239, 299). The extracellular
space can be further divided into the interstitial and intravas-
cular fluid (∼7.5% of total body water) compartments (147,
299). Because water is a major component of vascular vol-
ume, [blood volume is about 55%-65% water (plasma) and
35%-45% red blood cells] hydration also plays a critical role
in cardiovascular function and body temperature regulation.
Figure 2 illustrates the body fluid distribution in a 70 kg
human.
Sodium and its conjugate anions (chloride and bicarbon-
ate) comprise the most osmotically active components of
the extracellular fluid. Consequently, sodium balance plays
a key role in governing the size of the extracellular fluid com-
partment and passive water movement according to osmotic
gradients between the intracellular and extracellular water
spaces (299). The most abundant cations in the intracellular
water are potassium and magnesium, while the primary anions
are proteins. The imbalance of sodium and potassium across
the fluid compartments is maintained by the Na-K pump.
Maintenance of this distribution of electrolytes between the
Plasma, 3 L
Gain
Extracellular
fluid, 14 L
Water in fluids, food
Urine, sweat, feces,
respiratory vapor
Loss
Capillary membrane
Cell membrane
Interstitial fluid, 11 L
Intracellular fluid, 28 L
Figure 2 Body fluid compartments that comprise 42 L of total body
water in a 70 kg human, and sources of fluid gain or loss. Reprinted
(with permission) from Armstrong (21).
intracellular and extracellular fluid is critical for cell function
and electrical communication throughout the body.
Daily fluid requirements
Daily fluid requirements are determined by total body water
loss, with the goal of intake approximating daily water losses.
The United States Department of Agriculture’s Adequate
Intake for total water, based on the median total water intake
from the U.S. National Health and Nutrition Examination Sur-
vey data, is 3.7 and 2.7 L/d for young adult (19–30 years) men
and women, respectively (239). The European Food Safety
Authority defines Adequate Intake for total water as 2.5 L/d
for men and 2.0 L/d for women (154). Daily fluid requirements
are higher in individuals who are extremely active, exposed
to environmental stress from heat or altitude, or losing fluid
through vomiting or diarrhea.
When there is a mismatch between fluid intake and fluid
loss this imbalance leads to a body water deficit or surplus.
Throughout this paper, the term “euhydration” refers to main-
tenance of “normal” baseline body water content, while the
terms “hypohydration” and “hyperhydration” refer to body
water deficits and excesses beyond euhydration, respectively.
The term “dehydration” is defined as the process of the
dynamic loss of body water or the transition from euhydration
to hypohydration.
To assess an individual’s acute change in hydration status
their body mass can be compared to baseline or control values.
For example, 3% hypohydration is defined as a water deficit
equal to 3% of body mass. Acute body mass change (e.g., from
before to after exercise) represents approximately 1 mL of
water loss per 1 g of body mass loss. This method of hydration
assessment provides the simplest index in real time (21) and
is referred to frequently throughout this paper. However, it is
important to note that a small portion of body mass loss during
activity occurs due to substrate oxidation, that is, nonwater
mass. Therefore, estimations in water loss based on change
in body mass should be corrected for this nonwater mass loss
during exercise lasting several hours (e.g., >3 h) (414). The
reader is referred to papers that discuss the topic of hydration
assessment in greater detail (21, 311, 414). Euhydration cutoff
values for various biomarkers of hydration status are provided
in Table 1.
Sources and Composition
of Fluid Loss
Body water loss can be categorized as insensible or sensi-
ble. Insensible water loss is evaporative water loss through
respiration and water diffusion through the skin. Under nor-
mal conditions, respiratory water loss is approximately 250
to 350 mL/d (229) but can be higher in dry climates and
when respiration rate is increased, such as during exercise
or at altitude. Loss of water via diffusion through the skin is
approximately 450 mL/d (275). All other water loss, including
urination and thermoregulatory sweating is termed sensible
578 Volume 4, April 2014
Comprehensive Physiology Optimal Composition of Fluid-Replacement Beverages
Table 1 Biomarkers of Hydration Status
Practicality Validity Euhydration cut-off
Total body water Low Acute and chronic <2%
Plasma osmolality Medium Acute and chronic <290 mOsm/kg
Urine specific gravity High Chronic <1.020 g/mL
Urine osmolality High Chronic <700 mOsm/kg
Body weight High Acute and chronic∗<1%
∗Potentially confounded by changes in body composition during very prolonged assessment periods.
Reprinted (with permission) from Sawka et al. (414).
because the person is aware of the loss as it is occurring. Urine
production is the main avenue of body water loss under nor-
mal circumstances. Renal fluid output can vary considerably,
as it is primarily a function of water intake. Obligatory urine
loss (needed to excrete end products, such as creatinine and
urea; and needed to excrete electrolytes to maintain electrolyte
balance) is approximately 500 mL/d (5). However, urine out-
put generally averages 1 to 2 L/d. Other factors that affect
urine output include exercise, heat stress (decrease loss), cold,
and altitude exposure (increase loss). Gastrointestinal (fecal)
water loss in a healthy adult is small, approximately 100 to
200 mL/d (345). Table 2 lists estimated minimum daily water
losses through respiration, urination, the gastrointestinal tract,
and the skin.
During exercise and/or exposure to a hot environment
thermoregulatory sweat is the main source of water loss from
the body. Evaporation of sweat secreted onto the skin surface
by eccrine sweat glands is the primary avenue of heat loss
during exercise and/or heat stress. Radiation (heat exchange
between the body and the environment in the form of infrared
rays), conduction (transfer of heat to or from the body through
direct contact with an object), and convection [heat exchange
between the body and surrounding moving air (wind) or body
fluids (blood)] are other potential avenues of heat loss. How-
ever, when ambient temperature is greater than skin temper-
ature, evaporation of sweat is the only means of body heat
loss; which is important to attenuate the increase in body core
temperature. With sweating, heat is transferred from the body
to water (sweat) on the surface of the skin. When this water
gains sufficient heat, it is converted to a gas (water vapor),
thereby removing heat from the body. Evaporation of 1 kg of
sweat from the skin will remove 580 kcal of heat from the
body (508). It is important to note that sweat dripping from
the body is wasted water loss because sweat must evaporate
to allow effective cooling.
Metabolic heat production is directly proportional to exer-
cise intensity. When exercise is performed, a large amount of
heat is produced by the contracting muscles. In fact, less
than 25% of all the energy produced by contracting mus-
cles is used to perform work, with the remaining 75% con-
verted to heat in the muscles. Thus sweating rates increase in
proportion to work intensity. However, heat acclimatization,
higher fitness levels, clothing, and higher ambient temper-
atures also increase an individual’s sweating rate. By con-
trast, wet skin (from high humidity) can reduce sweating rate
(189, 418, 419). Figure 3 illustrates predicted daily water
Table 2 Estimation of Minimum Daily Water Losses and Production∗
Reference Source Loss (mL/d) Production (mL/d)
Hoyt and Honig (229) Respiratory loss 250 to 350
Adolph (5) Urinary loss 500 to 1000
Newburgh et al. (345) Fecal loss 100 to 200
Kuno (275) Insensible loss 450 to 1900
Hoyt and Honig (229) Metabolic production 250 to 350
Total 1300 to 3450 250 to 350
Net loss 1050 to 3100
∗Assuming conditions in which there is minimal water loss from sweating.
Reprinted (with permission) from Institute of Medicine (239).
Volume 4, April 2014 579
Optimal Composition of Fluid-Replacement Beverages Comprehensive Physiology
14
59 68 77 86 95 104
15,070
12,140
10,050
7,950
15 20 25 30 35 40
Energy
expenditure
kJ ⋅d–1
Required daily water intake, liters⋅d–1
Average daytime air temperature,°F
Average daytime air temperature,°C
Minimum daily water requirement
12
10
8
6
4
2
0
Figure 3 Predictions of daily water requirements as a function of daily energy expendi-
ture and air temperature. Reprinted (with permission) from Sawka et al. (416).
requirements as a function of daily energy expenditure and air
temperature.
Because sweating plays a critical role in attenuating
increases in body core temperature, it is apparent that suf-
ficient hydration (via drinking) is needed to maintain sweat-
ing, especially in extreme heat (7). Early observations have
shown that if water supply is adequate, healthy humans can
withstand and even thrive in extremely hot environments (pro-
vided that evaporation of sweat is not impeded by high ambi-
ent humidity or by the wearing of impermeable clothing) (7,
233). For instance, in 1910, healthy European men, observed
during a very hot (≥45◦C) and dry spell of weather, con-
sumed ≥13.6 L of water per day and were able to walk and
perform a considerable amount of physical exercise without
difficulty (233).
In general, women typically exhibit lower sweating rates
than men, primarily due to smaller body mass and lower
metabolic rate achieved during activity (414). However, when
expressed relative to body surface area, mean sweating rates
are similar between sexes in temperate and hot-dry condi-
tions (429). In hot-wet conditions, however, sweating rate
per m2surface area is lower in women than men (30, 429).
The greater suppression of sweat in response to wetted skin
results in less wasted sweat. By having a lower sweating
rate in this condition women are losing less fluid and there-
fore minimizing hypohydration. On the other hand, men drip
more sweat from their bodies (i.e., wasted water since it is
not readily evaporated and does not contribute significantly
to cooling) in humid conditions and become more dehydrated
(30, 417, 429). However at the same level of hypohydration
(5%), women exhibit similar physiological responses (e.g.,
increases in heart rate and body core temperature) to exercise-
heat stress in both hot-dry and hot-wet conditions compared
to men when matched for age, fitness, and percent body
fat (417).
The loss of water due to thermoregulatory sweating is
accompanied by loss of electrolytes. Sodium is the predomi-
nant electrolyte lost in sweat. The total amount of sodium lost
depends on sweating rate and duration as well as sweat sodium
concentration. Average sweat sodium concentration measured
using the “gold standard” whole-body washdown procedure
has been reported to be approximately 40 mmol/L, but ranges
from as low as 15 mmol/L to as much as 90 mmol/L (35).
Even those athletes with low or average sweat sodium con-
centration, can accrue a substantial sodium deficit by virtue
of large sweat losses due to high sweating rates (≥2L/h)
or extended periods of strenuous exercise (two-a-day prac-
tices or ultraendurance events). The wide range is a result
of the myriad of factors that influence sweat sodium concen-
tration, including genetics, diet, heat acclimatization status,
sweating rate, and hydration status (414). Sodium and chlo-
ride are reabsorbed in the duct of the sweat gland, thus sweat
sodium concentration is lower than that of plasma. However,
as sweating rate increases, the sodium secretion rate increases
proportionally more than the rate of sodium reabsorption, thus
sweat sodium concentration increases linearly with increases
in sweating rate (68). Heat acclimatization improves sodium
chloride reabsorption, thus resulting in lower sodium chlo-
ride concentration (>50% reduction) for any given sweat-
ing rate (10). Studies have found that moderate (∼3.5-4 g/d)
to high dietary sodium (∼8-9 g/d) intake results in signifi-
cantly higher sweat sodium concentration compared to that
of low sodium diets (∼1-2 g/d) (11, 23, 221). Increased sweat
sodium chloride concentration can also result from hypohy-
dration (330), but neither sex nor aging seem to have a signif-
icant effect (331). Figure 4 illustrates predicted daily sodium
580 Volume 4, April 2014
Comprehensive Physiology Optimal Composition of Fluid-Replacement Beverages
59 68 77 86 95 104
15,070
12,140
10,050
7,950
15 20 25 30 35 40
Energy
expenditure,
kJ ⋅d–1
Required daily sodium intake, mg ⋅d–1
Average daytime air temperature,°F
Average daytime air temperature,°C
0
2000
4000
6000
8000
10,000
Figure 4 Predictions of daily sodium requirements as a function of daily energy expendi-
ture and air temperature. Reprinted (with permission) from Sawka et al. (416).
requirements as a function of daily energy expenditure and
air temperature. In addition to sodium, several other elec-
trolytes are lost in sweat. These include, but are not limited
to, chloride (∼30 mmol/L), potassium (∼5 mmol/L), calcium
(∼0.5 mmol/L), and magnesium (∼0.1 mmol/L) (152, 414).
There are primarily two different types of hypohydra-
tion, depending on the route of water loss and the amount of
osmolytes (electrolytes) lost in association with the water.
Isoosmotic hypohydration (isotonic hypovolemia) occurs
when fluid loss is iso-osmotic with plasma, that is, loss
of water and osmolytes occurs in equal proportions. This
type of body water deficit is associated with fluid losses
induced by cold, altitude, diuretics, and secretory diarrhea.
For example, fluid (urine) losses induced by administration
of a diuretic, such as furosemide, will cause the intravascular
and interstitial fluid compartments to decrease proportion-
ally. Thus the end result is a state of hypohydration with no
change in plasma osmolality. On the other hand, hypohydra-
tion induced by sweat loss due to exercise and/or heat stress
results in a decrease in the extracellular compartment size
and an increase in plasma osmolality (1, 273). This is because
sweat is hypotonic compared to the plasma. An increase in
plasma osmolality initiates fluid movement from the cellular
compartment into the plasma to maintain osmotic balance.
This results in cellular hypohydration, that is, cell shrinkage
and hypertonicity. This isknown as hyperosmotic hypohydra-
tion and occurs when loss of water is greater than the loss of
osmolytes. Hyperosmotic hypohydration can also occur as a
result of osmotic diarrhea. The effect on blood osmolality has
important implications to cardiovascular and thermoregula-
tory physiology because hyperosmolality increases the tem-
perature threshold for sweating and cutaneous vasodilation
during exercise in the heat (172, 339).
Sources of fluid replacement
Daily fluid requirements can be met from a combination of
drinking water, water in beverages, and water that is in food.
In the U.S. approximately 81% of the water is consumed in
fluids (drinking water and beverages), while foods account
for the remaining 19% of water intake. However, in other
countries with higher intake of fruits and vegetables, such as
Greece and South Korea, water intake from foods is higher
(287, 329).
According to the 1994 to 1996 U.S. Continuing Survey
of Food Intakes by Individuals (aged 20-64 years), drinking
water is the main source of fluid water intake, with coffee,
tea, carbonated drinks, juice, and milk being some of the
other beverages contributing to fluid water intake in the U.S.
(222). Table 3 shows the water content of various foods and
beverages. Other sources of fluid water intake include sports
drinks and oral rehydration solutions. These beverages, which
have specific applications, are discussed in greater detail later
in this article.
It is important to note that water gain also occurs through
metabolic water production. Water is a byproduct of oxida-
tive phosphorylation and its rate of formation is primarily
dependent upon energy expenditure and to a lesser extent
on the type of substrate oxidized. This has been estimated
to be approximately 250 to 350 mL/d for sedentary per-
sons and up to 500 to 600 mL/d for the physically active
(229).
Regulation of fluid balance
Body water is regulated by various physiological mecha-
nisms to minimize fluctuations in body water content and
Volume 4, April 2014 581
Optimal Composition of Fluid-Replacement Beverages Comprehensive Physiology
Table 3 The Water Content Range for Selected Foods
Percentage Food item
100% Water
90%-99% Fat-free milk, cantaloupe, strawberries,
watermelon, lettuce, cabbage, celery, spinach,
pickles, squash (cooked)
80%-89% Fruit juice, yogurt, apples, grapes, oranges, carrots,
broccoli (cooked), pears, pineapple
70%-79% Bananas, avocados, cottage cheese, ricotta cheese,
potato (baked), corn (cooked), shrimp
60%-69% Pasta, legumes, salmon, ice cream, chicken breast
50%-59% Ground beef, hot dogs, feta cheese, tenderloin
steak (cooked)
40%-49% Pizza
30%-39% Cheddar cheese, bagels, bread
20%-29% Pepperoni sausage, cake, biscuits
10%-19% Butter, margarine, raisins
1%-9% Walnuts, peanuts (dry roasted), chocolate chip
cookies, crackers, cereals, pretzels, taco shells,
peanut butter
0% Oils, sugars
Reprinted (with permission) from Popkin et al. (377).
distribution. Specifically, plasma osmolality and vascular
volume are the hydration-related variables tightly regulated
(239) by reflex adjustments mediated by nerves and hor-
mones. In response to a perturbation in these variables the
nervous and endocrine systems modify fluid distribution (fluid
loss from the capillaries and the rate at which compliant veins
(e.g., skin and splanchnic circulation) fill with blood), fluid
losses (urine flow rate and sweat losses), and fluid intake
(thirst). The most rapid adjustments are initiated by neural
reflexes that control vascular resistance (fluid distribution).
If the perturbation to fluid balance is prolonged (>20 min),
vascular volume/pressure and plasma osmolality are main-
tained by circulating hormones that modify renal salt and
water excretion (299, 404).
Baroreceptors are the pressure sensing reflexes in the heart
and vascular system that function to minimize perturbations
to blood pressure. Both neural and humoral effectors can act
as efferent arms of the baroreflexes. During hypohydration
venous return is decreased due to a reduction in plasma vol-
ume. Thus, hypohydration results in a decrease in central
blood volume or central venous pressure and left ventricular
stroke volume. These vascular changes (decrease in trans-
mural pressure) result in unloading of baroreceptors, which
increases sympathetic nervous activity to vasoconstrict com-
pliant beds to protect central venous pressure and mean arte-
rial pressure. Cardiopulmonary (or low pressure) barorecep-
tors located in the chambers of the heart and the pulmonary
artery and veins sense the decrease in central venous pressure
with hypohydration and cause increased peripheral vascular
resistance (especially in skin and muscle). Arterial (or high
pressure) baroreceptors located in the carotid artery and arch
of the aorta sense the reduction in arterial pressure (and/or
aortic pulse pressure) with dehydration and elicit an increase
in heart rate and vascular resistance of the splanchnic circula-
tion (299, 404). The role of baroreflexes in maintaining blood
volume is depicted in Figure 5.
The kidneys are crucial in regulating water balance.
Changes in body fluids trigger humoral responses that act
on the kidneys to modulate urinary water losses, with the goal
of maintaining a tight range of plasma osmolality between
∼280 and 295 mOsm/kg (159). The main role of the kid-
neys is to filter plasma to remove metabolic wastes from the
body. This process begins with glomerular filtration, which is
the movement of water and solute from the circulation into
the renal tubule at the glomerular capillary. The functional
unit of the kidney, the nephron, consists of an extensive sys-
tem of tubules which interact with the systemic circulation
(peritubular capillaries) to allow movement of water/solute
between the circulatory and renal tubule compartments. In
the nephron, water and solutes (such as sodium chloride) are
reabsorbed back into circulation. Thus, the excretion of water
via urination is the net effect of glomerular filtration versus
renal tubular reabsorption.
Arginine vasopressin (or antidiuretic hormone) plays an
important role in the maintenance of vascular volume and does
so by increasing the renal reabsorption of water. Sweating-
induced dehydration causes an increase in plasma osmolality,
which leads to the movement of fluid from cells into the
plasma to maintain osmotic balance. This results in cellular
hypohydration or cell shrinkage. Shrinkage of osmoreceptor
cells in the hypothalamus and heart stimulate the synthesis of
vasopressin in the supraoptic and paraventricular nuclei of the
hypothalamus. Vasopressin increases renal water reabsorption
by binding to V2 receptors on the basolateral membrane of
cells of the distal tubule and collecting duct of the kidneys,
stimulating aquaporin transport of water across these cells
(from lumen to blood/basolateral side). Hyperosmolality also
stimulates thirst which would increase fluid intake and con-
tribute to the maintenance of vascular volume (299, 404).
The renin-angiotensin-aldosterone system also plays
an important role in the maintenance of vascular volume
and exerts its effects primarily on the renal system to
conserve sodium and water. Renin is released in response
to increased sympathetic nervous system activity, decreased
blood pressure, decreased blood sodium concentration, or
decreased renal blood flow. In response to these stimuli the
juxtaglomerular apparatus cells of the kidneys release the
enzyme renin into the blood. Renin then converts circulating
Angiotensinogen to Angiotensin I. When Angiotensin I
travels through the pulmonary circulation Angiotensin Con-
verting Enzyme cleaves Angiotensin I to form Angiotensin
II. Angiotensin II can affect vascular volume by directly
constricting smooth muscle of arterioles (especially the
splanchnic and renal circulation). In addition, Angiotensin II
stimulates aldosterone synthesis and release from the adrenal
582 Volume 4, April 2014
Comprehensive Physiology Optimal Composition of Fluid-Replacement Beverages
Blood
volume
(~500 mL) (>1000 mL)
Blood
volume
Central venous pressure
Arterial blood pressure
Arterial
baroreflex
ADH or
vasopressin
Sympathetic
nervous
activity
Renal Na+ retention
Water retention
Vasoconstriction
+
+
+
+
+
+
+
+
+
?
++
+
+
+
+
+
+
Cardiopulmonary
baroreflex
Sympathetic
nervous activity
Vasoconstriction
Capillary pressure
Water reabsorption
Glomerular
filtration rate
Urine
volume
Renin ?
Renal
β-receptors
Renal
blood flow
Central venous pressure
Plasma renin
activity
Angiotensin ll
Aldosterone
Figure 5 Two standard schemes for the role of baroreflexes and the renin-angiotensin-
aldosterone axis in maintaining blood volume. Reprinted (with permission) from Rowell (404).
cortex, which in turn stimulates renal sodium chloride
reabsorption. Angiotensin II also regulates body water by
acting on the central nervous system to stimulate thirst and
water and sodium ingestion (299, 404).
Some hormone-mediated differences in the renal han-
dling of water and electrolytes exist between sexes and
within women across different phases of the menstrual cycle.
For example, in response to a water load, women have a
higher rate of water turnover, particularly during the luteal
phase of the menstrual cycle (101). Estrogens and proges-
terone increase renal water and electrolyte retention (463,
465, 466). Furthermore, body core temperature is elevated by
up to 0.6◦C during the luteal phase. Despite these hormonal
effects, there is no evidence that phase of the menstrual cycle
significantly impacts exercise performance, perceived exer-
tion, or risk for heat illness during exercise (301) or renal
water and electrolyte retention with fluid replacement after
exercise (308).
More detailed summaries of body water and electrolytes,
including factors governing water volume, distribution (147,
299), and movement between compartments (468, 469) can
be found in other reviews.
Physiology of fluid replacement
Replacement of fluid losses is a process involving several
steps that starts with ingestion and culminates in fluid being
absorbed into the blood stream and distributed to the intra- and
extracellular fluid spaces of the body. In this section we review
these processes and discuss the physiological, and in some
cases psychological, factors involved in fluid replacement.
Thirst and voluntary fluid intake
The first step in fluid replacement is the act of drinking water
or other fluids. As discussed previously, the renal system
effects fluid balance by limiting water loss through urina-
tion; however, this system has no ability to restore lost water.
Therefore, the replacement of a fluid following a deficit is
ultimately dependent upon fluid intake. As simple as this
sounds thirst and fluid intake are actually very complex and
Volume 4, April 2014 583
Optimal Composition of Fluid-Replacement Beverages Comprehensive Physiology
dictated by multiple mechanisms (6). The consumption of
fluids occurs in response to physiological, psychological, and
environmental stimuli.
Physiological (or regulatory) thirst is stimulated indepen-
dently by cellular and extracellular hypohydration. Increases
in cellular tonicity (cellular hypohydration) are sensed by
osmoreceptors in the central nervous system, while decreases
in extracellular fluid volume (extracellular hypohydration)
are sensed by the cardiopulmonary (low pressure) barorecep-
tors. The sensory information from the osmoreceptors and
barorecptors regarding plasma tonicity and extracellular fluid
volume, respectively, feed into higher brain regions to stimu-
late thirst. The areas of the brain involved in the consciousness
of thirst include the anterior cingulate region and specific sites
in the middle temporal gyrus and periaqueductal gray area
(134).
The stimulation of thirst is delayed compared to other
physiological indicators of hypohydration. For example, the
threshold for the release of vasopressin and renal water con-
serving mechanisms occurs at a lower plasma osmolality than
does thirst. The osmotic threshold for vasopressin secretion is
approximately 286 mOsm/kg whereas the average threshold
for thirst is 295 mOsm/kg (394). The stimulation of physi-
ological thirst is also modified by aging and sex hormones
(discussed elsewhere in this article).
Thirst and fluid intake are influenced by a complex interac-
tion of numerous other factors. Afferent signals arising from
the oropharyngeal region influence thirst. For instance, the
sensation of a dry mouth drives thirst and initiates drinking.
However, dry mouth is a perceived sensation of thirst rather
than a physiological signal of hypohydration. Oropharyngeal
factors can also provide signals to terminate drinking. The
act of swallowing produces a sudden inhibition of osmoti-
cally stimulated thirst. Also, a process called oropharyngeal
metering acts to limit the overall rate of fluid ingestion (163).
Gastrointestinal factors such as gastric distension can also
provide sensory input to terminate drinking.
Other nonregulatory factors include cultural/social fac-
tors/preferences, beverage availability or proximity, and
organoleptic properties of beverage (temperature, flavor,
sweetness, saltiness, texture, and aftertaste) (365, 395) and
are discussed in more detail later. Figure 6 illustrates the mul-
tiple factors involved in thirst and voluntary fluid intake.
In general, humans drink adequately to offset losses
on a day-to-day basis. Body hydration status may fluctu-
ate throughout the day, but any fluid deficit is typically
replaced with fluid intake during meals and ad libitum drink-
ing between meals. Intakes of food and water are directly
related (15) and food has a water-retaining effect (108). How-
ever, ad libitum drinking during and/or after periods of fluid
loss (e.g., in workers or athletes losing sweat while physically
active or exposed to heat stress) usually results in incomplete
fluid replacement, a concept known as “voluntary hypohy-
dration” (213, 400). For example, in 1947 men in the desert
only replaced about 50% of losses when active, but when
eating food and/or at rest, the men drank enough to restore
body water balance (3). More recent studies have confirmed
this phenomenon (306). When underdrinking occurs over a
prolonged period of time (e.g., ∼75 min of running in Ref.
364, but will vary depending on sweating rate) body weight
deficits of ≥∼2% during or after exercise and/or heat-stress
have been reported (213, 364, 400). Thus, the optimal fluid
Attenuates
Beverage factors
Voluntary
fluid intake
Extracellular dehydration
(hypovolemia)
Leads to/stimulates
Physiological
thirst
Perceived
thirst Oropharyngeal
factors
Oral metering
(swallowing)
Dry mouth
Availability
Aging
Sex hormones
(estradiol)
CNS
Angiotensin ll
Renin-angiotensin system
Supine posture
Environment
(cold, altitude,
water immersion)
Organoleptic properties
(beverage temperature, flavor,
texture, composition)
Cellular dehydration
(hyperosmolality)
Gastrointestinal factors
(gastric distension)
Social factors Cultural factors
Figure 6 Diagram illustrating the multiple stimulating and attenuating factors involved in thirst and voluntary fluid
intake.
584 Volume 4, April 2014
Comprehensive Physiology Optimal Composition of Fluid-Replacement Beverages
replacement beverage would be one that is palatable enough
so as not to impede drinking and perhaps even promote drink-
ing when individuals would otherwise wish to avoid it (e.g.,
extreme, prolonged loss of appetite and desire to drink due to
sickness, altitude, or cold stress).
Gastric emptying and intestinal absorption
The second major step in fluid replacement is the process of
absorption into the bloodstream. This process involves both
the emptying of fluid from the stomach and transport of water
across the intestinal epithelium. Very little absorption of water
and solutes takes place in the stomach, thus the ingested fluid
must be emptied from the stomach and delivered to the lumen
of the small intestine before it can enter the circulation. The
gastric emptying rate of liquids is faster than that of solids
and solutions with lower energy density empty faster than
those of higher energy density (236). Gastric emptying of
liquids is regulated by the interaction of gastric volume and
feedback inhibition related to the nutrient content of the small
intestine (234,235, 333). Gastric emptying can be maintained
at a high rate (15 to more than 20 mL/min) by maintaining a
high gastric volume with ingestion of either water or a dilute
carbohydrate solution.
Absorption of water and solutes occurs primarily in
the proximal small intestine (duodenum and jejunum). The
structure of the epithelium in the small intestine allows water
and solute to pass into the bloodstream by both paracellular
(via tight junctions between cells) and transcelluar (across
the epithelial cell membrane) processes. Absorption of solute
from the intestinal lumen occurs by diffusion along electro-
chemical gradients and by specific transport mechanisms in
the brush border membrane of intestinal epithelial cells. Water
absorption follows solute absorption; that is, water uptake is
a passive process, dependent on an osmotic gradient which is
created by absorption of solutes (333, 436). Figure 7 illustrates
the proposed mechanisms of water and solute absorption.
Several factors can impact the rate at which gastric emp-
tying and fluid absorption occurs, including various types of
stressors. For example, hypohydration ≥3% is associated with
impaired gastric emptying (344, 381, 491) and exercise inten-
sity >∼70% to 75% slows gastric emptying and intestinal
water absorption (112, 168). Beverage composition, particu-
larly energy (carbohydrate) content and osmolality, also has
a significant influence on the rate at which these processes
occur. In general, the higher the energy density and osmo-
lality of the ingested beverage the slower the rate of gastric
emptying and intestinal fluid absorption, respectively (334,
436). When both energy provision and fluid replacement are
important (e.g., exercise in a warm environment), the optimal
beverage would be formulated to deliver nutrients to the body
without impeding gastric emptying and intestinal absorption
of water. However, when sweat losses are low (e.g., exercise
in very cool weather) or high rates of carbohydrate delivery
Fructose
Fructose
2 Na+
Na+
Na+
2 Na+
K+
K+
AT P
H2O
H2O
H2O
H2O
H2O
H2O, Na+, K+, Cl–
H2O, Na+, K+, Cl–
H2O
(passive diffusion)
GLUT5
LUMEN
BLOOD
GLUT2
AQP
SGLT1 DRTS Paracellular
transport
Sucrose
Solvent drag
???
Fructose
Glucose
Glucose
(galactose)
Glucose, galactose, fructose
Glucose, galactose, fructose
(Galactose)
glucose
Figure 7 Proposed mechanisms of water and solute absorption. Active sodium and glucose
cotransport by the SGLT1 transporter in the enterocyte brush border membrane and extrusion
of sodium and glucose across the basolateral membrane stimulate water and solute absorp-
tion through both transcellular and paracellular routes. The roles of GLUT5, GLUT2, DRTS,
Na+-K+ATPase, and aquaporins (AQP) are also depicted. Following activation of SGLT1,
solutes such as monosaccharides, minerals, and perhaps even small peptides and oligosac-
charides can be absorbed via the paracellular pathway. Reprinted (with permission) from Murray
and Shi (333).
Volume 4, April 2014 585
Optimal Composition of Fluid-Replacement Beverages Comprehensive Physiology
Table 4 The Relative Roles that Physiological Processes Play in Whole-Body Fluid Balance, during Different Life Scenarios
Scenario Renal regulation of
fluid-electrolyte balance Thirst and drinking
behavior Sweat gland secretion
of hypotonic fluid Comments
Sedentary daily activities (16 h) Normal Normal Negligible Normal hormonal
and CNS regulation
Brief, intense exercise (<5 min) Negligible Negligible Minor Volume of fluid loss
is small
Prolonged, strenuous exercise
(5-30 min) Minor Minor to moderate Minor to moderate Volume of fluid loss
is minor when
compared to TBW
Prolonged endurance exercise
(0.5-5 h) at moderate intensity Minor to moderate Minor to large Moderate to large Marge water
turnover due to
sweating and
drinking
Continuous or intermittent
exercise, or labor at low
intensity (5-24 h)
Minor to large Minor to large Large Fluid and electrolyte
losses may exceed
daily dietary intake
Consecutive days of activities,
labor, or exercise (1-180 d) Normal Normal Varied, depending on
labor and exercise Adequate dietary
fluid and electrolyte
consumption is
essential
Reprinted (with permission) from Armstrong (21).
is necessary (e.g., >2 h of exercise) more concentrated bev-
erages may be acceptable. These topics are discussed in more
detail later. Also, the reader is referred to other reviews that
discuss the idea of potential trade-offs between carbohydrate
and fluid ingestion during exercise (115, 119).
Fluid distribution and retention
Finally, once the ingested fluid enters the bloodstream, a
sufficient volume of the fluid needs to be retained to rehy-
drate and offset losses incurred through sweating, respiration,
and/or urination. Thus, the optimal fluid-replacement bever-
age would be formulated to facilitate renal water reabsorption
(described in detail above) as opposed to stimulating diure-
sis. In addition, the beverage should promote retention of
fluid in the compartment (intracellular or extracellular) that
has incurred the fluid deficit. For example, in the case of
significant hypovolemia, the aim would be to promote fluid
retention in the vascular space. The relative roles that various
physiological processes (renal regulation, thirst and drinking
behavior, and sweat secretion) play in determining whole-
body fluid balance during different life scenarios is described
in Table 4.
Effect of a body fluid deficit on human
physiological responses
Survival
High levels of hypohydration (∼10% body mass deficit) alone
do not significantly increase the risk of death. However, when
severe hypohydration is combined with other stressors, such
as illness or environmental/physical stress, body water deficits
can contribute to death (4, 239, 505). Reports from humans in
desert survival situations suggest that hypohydration exhaus-
tion occurs at 5% to 10% body mass deficit, thus it has been
suggested that ∼10% hypohydration could be fatal unless
medical assistance is available for recovery and rehydration
(4). In the summer of 1877, soldiers of the U.S. Cavalry
wandered into the arid “Staked Plains” near Fort Concho,
Texas while scouting for Indians and found themselves with-
out water supply for 3.5 days. In a brief account of the ordeal,
King (270) described the sufferings from deprivation of water
and reported that 4 of the men died by the time the troop man-
aged to get back to the Supply Camp. The longest period of
time that any person has been documented to survive without
water is 18 days (516).
No experimental data on the effects of water restriction
on mortality are available in humans, for obvious reasons.
Survival studies with dogs and cats exposed to heat stress
suggest that death from water restriction occurs with body
mass loss of 10% to 20%, due to a rapid rise in body core
temperature, with death consistently occurring at 41.6 to 43◦C
(4). From the work conducted by Adolph and associates in
the desert, the authors speculated that body mass deficits of
15% to 25% would be fatal in humans (4). The environmental
conditions influence the level of hypohydration that is fatal.
As little as 15% hypohydration could be fatal at ambient
temperatures above ∼30◦C, while more severe hypohydration
would be required at cooler temperatures (4).
Cardiovascular and thermoregulatory systems
A reduction in body water is associated with impaired circula-
tory and thermoregulatory function. Hypohydration causes a
586 Volume 4, April 2014
Comprehensive Physiology Optimal Composition of Fluid-Replacement Beverages
decrease in plasma volume and, therefore, a decrease in stroke
volume and a compensatory increase in heart rate (by 3-5 bpm
per 1% body mass deficit) (195, 325, 411) to maintain a given
cardiac output. Hypohydration delays the onset (i.e., increases
the body core temperature threshold) and decreases the sensi-
tivity (i.e., slope of the relation between sweating rate or skin
blood flow relative to the change in body core temperature)
of the sweating and skin blood flow response to hyperther-
mia (169, 172, 269, 326). Both hypovolemia and hyperosmo-
lality mediate these hypohydration-induced adjustments. For
example, studies (169, 339) found that hypovolemia per se
(via diuretic administration) increased the esophageal tem-
perature threshold for cutaneous vascular conductance by
∼0.4◦C (slope of relation not affected) and reduced the slope
of the relation between sweating rate and esophageal temper-
ature (threshold not affected). Fortney et al. (172) then found
that hyperosmolality per se elevated the esophageal tempera-
ture thresholds for both cutaneous vasodilation and sweating.
Thus, hypohydration results in an increased core temperature
during exercise and an increased risk for heat exhaustion (3,
315, 421) and heat stroke (79, 153, 196, 384). The magnitude
of hyperthermia ranges from approximately 0.1 to 0.2◦C per
1% body mass deficit (325, 420) and likely varies with ambi-
ent conditions and exercise intensity. The impact of various
amounts of fluid intake on cardiovascular function, body core
temperature, and rating of perceived exertion during exercise
are illustrated in Figures 8, 9, and 10, respectively.
Heat acclimatization and endurance training increase the
sweating and skin blood flow responses to exercise in the heat
(509). These responses augment heat dissipation and there-
fore minimize the rise in body core temperature during exer-
cise/heat stress. However, the improvements in thermoregu-
latory function conferred by heat acclimatization and aerobic
training are negated by hypohydration (73, 417).
Medical conditions
Low habitual fluid intake has been reported, mostly in epi-
demiological studies, to be associated with a number of
chronic medical conditions, including urinary tract infections,
blood clots, asthma, urolithiasis (kidney stones), cholelithia-
sis (gallstones), constipation, cardiovascular disease, diabetic
hyperglycemia, and some cancers (239, 300, 377). There is
evidence that increased fluid intake reduces the risk of kidney
stones, particularly in the prevention of recurrence in individ-
uals that have already had stones (239). For example, Borghi
et al. (56) tested the effects of increasing water intake in 199
individuals with idiopathic calcium nephrolithiasis. The treat-
ment group increased fluid intake such that 24-h urine volume
increased from ∼1 to 2.6 L, whereas the control group did not
change fluid intake (24-h urine volume remained at ∼1L).
Upon follow-up 5 years later, recurrent stones had occurred
in 27% of the control group, but only 12% of the treatment
group. Similarly, increasing fluid intake has been shown to
decrease recurrence of infections in women with a history of
urinary tract infections (146, 374). Increasing fluid intake is
Cardiac output (L/min)Heart rate (b/min)Stroke volume (mL)Forearm blood flow (mL/100 mL/min)
20
19
18
17
16
170
160
150
140
130
150
140
130
120
110
100
90
17
16
15
14
13
12
11
10
020406080
Time (min)
100 12 0
No fluid
*†§
*†§
*†§
*
*†
*†
*†
*
*
Small fluid
Large fluid
Moderate fluid
Figure 8 Cardiac output, heart rate, stroke volume, and forearm
blood flow during 120 min of cycling (62%-67% VO2max) in the heat
(33◦C, 50% rh) when no fluid or small (0.58 L), moderate (1.42 L), or
large (2.38 L) volume of fluid was ingested, which induced 4.2%, 3.4%,
2.3%, and 1.1% hypohydration, respectively. Values are means ±SE
for 8 different endurance-trained cyclists. ∗Significantly different from
no fluid, P<0.05. †Significantly different from small volume of fluid,
P<0.05. §Significantly different from moderate fluid volume of fluid,
P<0.05. Reprinted (with permission) from Montain and Coyle (325).
Volume 4, April 2014 587
Optimal Composition of Fluid-Replacement Beverages Comprehensive Physiology
39.0
38.6
38.2
37.8
37.4
37.0
36.6
39.2
38.8
38.4
38.0
37.6
37.2
36.8
02040
Time (min)
Rectal temperature (°C) Esophageal temperature (°C)
60 80 100 120
No fluid
*
*
*
*
*
*† *†
*†
*†
*†
*†
*†
*†§
*†§
Small fluid
Large fluid
Moderate fluid
Figure 9 Esophageal temperature during 120 min of cycling (62%-
67% VO2max) in the heat (33◦C, 50% rh) when no fluid or small
(0.58 L), moderate (1.42 L), or large (2.38 L) volume of fluid was
ingested, which induced 4.2%, 3.4%, 2.3%, and 1.1% hypohydration,
respectively. Values are means ±SE for 7 endurance-trained cyclists.
∗Significantly lower than no fluid, P<0.05. †Significantly lower than
small volume of fluid, P<0.05. §Significantly lower than moderate fluid
volume of fluid, P<0.05. Reprinted (with permission) from Montain
and Coyle (325).
18
Very hard 17
Rating of
perceived exertion
16
Hard 15
14
12
10
20 40 60 80
Time (min)
100 12 0
Fairly light 11
Somewhat hard 13
No fluid
Small fluid
Large fluid
*
*
*†
*†
Moderate fluid
Figure 10 Rating of perceived exertion during 120 min of cycling
(62%-67% VO2max) in the heat (33◦C, 50% rh) when no fluid or small
(0.58 L), moderate (1.42 L), or large (2.38 L) volume of fluid was
ingested, which induced 4.2%, 3.4%, 2.3%, and 1.1% hypohydration,
respectively. Values are means ±SE for 7 endurance-trained cyclists.
∗Significantly lower than no fluid, P<0.05. †Significantly lower than
small volume of fluid, P<0.05. Reprinted (with permission) from Mon-
tain and Coyle (325).
also commonly recommended for the treatment of constipa-
tion. However, the few research studies available on this topic
suggest that increased fluid intake is only effective in dehy-
drated individuals and is not useful in mitigating constipation
in euhydrated individuals (28). The relation between daily
water intake and cancer (bladder and colon) has also been
evaluated. Some case-control studies suggest a link between
higher rates of fluid intake and decreased colon and bladder
cancer risk (239, 320, 428); however, others have failed to
demonstrate a significant association (239, 451).
Acute hypohydration may also be a precipitating factor
in some acute medical conditions. For example, hypohydra-
tion is thought to be a risk factor for delirium in the elderly
(284, 499). Poor hydration status has also been linked to an
increased risk of infection, especially in elderly individuals
(161). Increased mortality is commonly observed during heat
waves. Although unusually hot weather is the primary culprit,
failure to increase water intake would exacerbate the effects
of heat strain (303). Another acute condition thought to ben-
efit from provision of water is headache, including migraine.
Water deprivation and hypohydration have been shown to
increase self-reported headache (443) and trigger and pro-
long migraine (54,55). In individuals with headaches caused
by water deprivation, ingestion of water provided relief from
headache within 30 min to 3 h (55). However, there is insuf-
ficient evidence to suggest that fluid intake is an effective
prophylaxis against headache (457). Finally, one study (420)
has reported that severe levels of hypohydration (5% to 7%)
resulted in numerous premature ventricular contractions in
healthy young adults exercising in the heat for over 2 h. It is
important to recognize that only associations and not direct
causal relations have been demonstrated between hydration
status and most of these acute and chronic medical conditions.
It has been suggested that rhabdomyolysis, which is a
syndrome observed with novel strenuous exercise and mani-
fested as release of skeletal muscle contents leading to acute
renal insufficiency or failure, is exacerbated by hypohydra-
tion. Specifically, when rhabdomyolysis is accompanied by
hypohydration, there is an increased likelihood or severity of
acute renal failure (66, 422). Thus, the combination of novel
training, heat stress, and fluid restriction is especially danger-
ous and potentially fatal (196).
Physical performance
It is widely accepted that hypohydration negatively affects
cardiovascular and thermoregulatory function, the combined
effect of which can increase perceived effort and impair per-
formance during activity which is greatly dependent upon
these two systems, that is, prolonged aerobic exercise (239,
414, 444). Hypohydration >2% body mass can degrade aero-
bic exercise performance in temperate and warm/hot environ-
ments (81, 92, 239, 414). The greater the level of hypohydra-
tion and heat stress, the greater the degradation in aerobic exer-
cise performance (239, 414). Hypohydration only marginally
degrades aerobic exercise performance in cold environments
588 Volume 4, April 2014
Comprehensive Physiology Optimal Composition of Fluid-Replacement Beverages
(89) and has not been found to degrade muscular strength
(156, 215, 239), anaerobic performance (90, 239), or jumping
ability (93). However, performance during stop-and-go type
activities/sports characterized by repeated bouts of intermit-
tent high-intensity exercise has been shown to be impaired
by hypohydration. Body mass deficits on the order of 2%
to 3% has been associated with decrements in high-intensity
endurance (259) and execution of sport-specific skills (32,
137, 141, 314).
Overall, it appears that exercise performance that is aer-
obic in nature and thus dependent upon the cardiovascu-
lar system is most likely to be degraded by a body water
deficit. Another mechanism by which hypohydration impairs
physical performance may be related to its effect on energy
metabolism. Increased lactate, muscle glycogenolysis, and
carbohydrate oxidation during prolonged exercise have been
associated with as little as 1% to 2% hypohydration (193, 220,
296, 297). In addition, Gonzalez-Alonso and Nielsen (192)
found that blood flow to exercising muscles declines signifi-
cantly with hypohydration (hyperosmotic hypovolemia) dur-
ing exercise/heat stress. This is due to a lowering of perfusion
pressure and systemic blood flow. Detailed discussions on
the biological mechanisms of hypohydration-induced perfor-
mance impairments can be found elsewhere (94).
Brain and cognition
Hypohydration can also have detrimental effects on cognitive
performance (short-term memory, attention, and visual-motor
tracking) and ratings of mental fatigue and mood, especially
when the body water deficit is combined with environmen-
tal heat stress (96, 97, 198, 430). In general, hypohydration
>2% of body mass loss, typically induced by exercise and/or
heat stress, degrades cognitive function. However, milder lev-
els of hypohydration have also been suggested to adversely
affect some aspects of mood and cognitive performance (24,
179, 292). For example, Ganio et al. (179) found that 1.6%
hypohydration degraded visual vigilance and visual working
memory and increased ratings of fatigue and tension/anxiety
in healthy young men. Furthermore, there is some evidence
that cognitive performance is degraded in a dose-dependent
manner with graded levels of hypohydration (198, 292). These
effects have the potential to negatively impact the capacity to
perform work involving mentally demanding or skilled tasks
in everyday life as well as in occupational, military, and sports
settings.
Another negative effect and potential safety hazard of
hypohydration related to the brain is an increased risk of ortho-
static intolerance (postural hypotension) in the heat. Upon
standing, fluid displacement from the thorax to the lower
extremities leads to a transient fall in mean arterial blood pres-
sure, sometimes leading to syncope. In 1947, Adolph et al.
reported that dehydrated subjects felt “dizzy” upon standing
and “may faint if asked to stand still” (4, 62). Recently, Carter
et al. (78) reported that the orthostatic intolerance experi-
enced with hypohydration is likely attributed to a decrease
in cerebral blood flow velocity. When subjects moved from
a seated to a standing position, cerebral blood flow velocity
was significantly reduced by prior heat exposure and further
reduced by 3% hypohydration compared to when subjects
were euhydrated. The hemodynamic response to orthostatic
stress has been found to improve with water ingestion (77,
298, 426). For instance, acute water intake of 500 mL blunted
the increase in heart rate and decrease in stroke volume during
head-up tilt with lower body negative pressure and improved
orthostatic tolerance (time to syncope) in healthy young adults
(426). Also, Ando et al. (16) found that 300 mL fluid inges-
tion can prevent vasovagal reaction in blood donors who are at
a high risk of postdonation syncope. The beneficial effect of
300 to 500 mL water ingestion on orthostatic tolerance occurs
without significant changes in blood volume (77). Instead, the
mechanism of action responsible for these improved hemo-
dynamic responses is likely due to the stimulatory effects
of drinking on the sympathetic nervous system to increase
peripheral resistance (426).
Adverse effects of overconsumption
When someone overdrinks (drinks more fluid than they
have lost through sweating, urination, etc.) he or she can
become hypervolemic (overall increased volume of plasma),
and hypo-osmotic (dilution of plasma contents). Excessive
consumption of fluids (especially sodium-free fluid, such
as water) over a prolonged period of time can lead to the
dilution of sodium concentration in the blood. This rare,
but potentially dangerous condition is called hyponatremia.
The plasma sodium concentration of those with symptoms
of hyponatremia is typically less than 130 mmol/L. Nor-
mal resting plasma sodium concentration ranges from 135 to
145 mmol/L. The reduction in solute concentration in plasma
promotes movement of water from the plasma into cells (327).
Symptoms of mild to moderate hyponatremia may include
headache, nausea, dizziness, and muscle weakness, while
severe hyponatremia (typically plasma sodium concentration
<125 mmol/L) is characterized by pulmonary edema, car-
diorespiratory arrest, cerebral edema, seizures, and/or coma
(31, 327). The rapidity, with which the serum sodium con-
centration declines, as well as the absolute change, impacts
the severity of the symptoms. In a 70-kg man, approximately
5.1 L increase in total body water is required to decrease
serum sodium concentration to 125 mmol/L from an initial
concentration of 140 mmol/L (239).
Hyponatremia can occur in anyone partaking in overzeal-
ous fluid consumption, including accidental water intoxica-
tion or forced drinking in children (18, 264), in social situa-
tions (166), in athletes (127, 225, 350, 456) or military recruits
(180, 355) during prolonged exercise, in occupational situa-
tions (332), or in psychiatric patients (particularly those with
schizophrenia) who have psychogenic polydipsia (129, 239).
Acute water intoxication usually occurs as a result of rapid
consumption of large quantities of fluid that greatly exceed the
kidney’s maximal excretion rate (∼1.0 L per h) (239). Factors
Volume 4, April 2014 589
Optimal Composition of Fluid-Replacement Beverages Comprehensive Physiology
which decrease or impair renal water excretion can put some-
one at a higher risk for hyponatremia. Exercise and/or heat
stress reduce urine volume. Also, nausea reduces urine output
by stimulating vasopressin secretion. Hospitalized patients
who develop hyponatremia often exhibit inappropriate secre-
tion of vasopressin during fluid overload (184).
The development of hyponatremia during exercise is usu-
ally associated with excessive overconsumption of fluid (in
excess of sweat losses) over a period of activity that is greater
than 4 to 6 h, such as during endurance events (327). Exercise-
associated hyponatremia has also been reported in American
football and tennis players (138, 223) and in recreational
exercisers (31, 401, 447). In addition, women and children
are more susceptible to hyponatremia because their smaller
total body water and extracellular fluid volume requires less
overdrinking than men to dilute serum sodium concentration
(239).
Symptomatic exercise-associated hyponatremia tends to
occur more commonly in women than men (12, 224, 414).
The reason is not fully understood, but a variety of psychoso-
cial and/or biological factors may be involved. Studies have
found that women overdrink relative to body weight com-
pared to men (34, 224). Other factors that may increase a
woman’s risk for overdrinking includes lower body mass and
total body water (467) as well as longer marathon race times
(12). A recent prospective study comparing women with and
without a history of hyponatremia suggests that female sex
hormones may also play a role in the development of hypona-
tremia during exercise. Stachenfeld et al. (467) found that,
compared to women with no history, women susceptible to
hyponatremia retained more fluid (in response to a fluid over-
load) and lost more sodium when both estradiol and proges-
terone were elevated. The higher severity of cerebral edema
symptoms (i.e., morbidity and mortality) reported in women
may be related to how the brain handles water and electrolyte
imbalances, as some animal studies have shown impaired
Na-K-ATPase pump activity in the female brain during AVP-
induced hyponatremia (175,176).
Components of a Fluid-Replacement
Beverage
Objectives of a fluid-replacement beverage
The main objective of a fluid-replacement beverage is to
replace water lost from the body. However, depending on
the objective and target population, fluid-replacement bever-
ages are often formulated with other ingredients, especially
for beverages intended to improve physical performance.
Beverage composition can have a substantial impact on
fluid ingestion, gastric emptying, intestinal fluid absorption,
fluid distribution, and fluid retention. Therefore, when fluid
replacement is the main objective, the beverage should be
formulated to optimize these processes. Moreover, the opti-
mal composition of the beverage depends upon the source
of the fluid losses. For example, the beverage should be
formulated differently to replace fluid losses from secretory
diarrhea compared to a beverage intended to replace losses
incurred through thermoregulatory sweating.
A fluid-replacement beverage can also act as a vehicle to
provide various nutrients or compounds to aid physiological
processes unrelated to hydration. Some examples include car-
bohydrate for energy provision, caffeine for stimulation of the
central nervous system, and protein to promote postexercise
muscle recovery. Many of these added ingredients can pro-
mote ergogenic effects to physical and mental performance;
and if properly formulated, can do so without hindering hydra-
tion and in some cases even promote the rehydration process.
Overview of potential ingredients
In this section, we introduce ingredients commonly found
in fluid-replacement beverages and summarize the general
effects of these ingredients on fluid replacement, provision of
energy, or augmenting physiological function. The relevance
and efficacy of these ingredients to specific populations will
be discussed in detail later.
Carbohydrate
Carbohydrate is included in fluid-replacement beverages in
part because of its impact on the rate of water absorption.
The cotransport of glucose and sodium facilitate the passive
absorption of water across the intestinal mucosa. To improve
water absorption, the minimum glucose concentration needed
in a beverage is 0.9% (185). In addition, carbohydrate intake
is important for maintaining blood glucose concentrations
and high rates of carbohydrate oxidation, especially when
endogenous carbohydrate stores are being depleted, such as
during physical activity (247, 251).
As will be discussed in more detail later, both the type
and amount of carbohydrate impacts the rate of fluid and
solute absorption from a beverage at rest and during exercise.
This is related to the fact that different transport mechanisms
are utilized by different carbohydrate types. Glucose (derived
from a glucose source or hydrolysis of maltose, maltodex-
trin, glucose polymers, or starch) and galactose are trans-
ported across the intestinal mucosa via the energy-dependent
sodium-glucose link transporter (SGLT1), whereas fructose
utilizes GLUT5, and sucrose can either be hydrolyzed into
glucose and fructose or utilize its own disaccharide transporter
SCRT (160, 265, 319). In addition, carbohydrate type greatly
impacts the osmolality of a beverage, with maltodextrin and
starch having lower osmolality than mono- and disaccharides.
Carbohydrate amount and type also impact the sweetness and
thus the palatability of fluid replacement beverages. For exam-
ple, compared to sucrose (table sugar, relative sweetness rat-
ing of 100) solutions that contain crystalline fructose (rating
of 180) or high fructose corn syrup (rating 105-130) have
higher relative sweetness whereas glucose (rating of 50-70)
and maltose (rating 50) taste less sweet. Moreover, complex
590 Volume 4, April 2014
Comprehensive Physiology Optimal Composition of Fluid-Replacement Beverages
carbohydrates such as maltodextrin and starch have very low
sweetness levels (44, 337).
Electrolytes
The presence of sodium in a fluid-replacement beverage
enhances palatability and stimulates the physiological drive
to drink. By contrast, consumption of plain water decreases
plasma osmolality and sodium concentration, which reduce
the drive to drink, oftentimes before body water volume has
been fully restored (351, 478). It is important to note that very
high sodium concentrations (≥50 mmol/L) decrease drink
palatability (507), which could hinder ad libitum fluid intake.
However, the unpalatable salty taste from high concentra-
tions of sodium chloride (the typical source of sodium) can
be decreased by substituting other anions (such as citrate) for
chloride (440). The increase in serum sodium concentration
and osmolality with sodium ingestion stimulates renal water
reabsorption. Furthermore, because sodium is the primary
extracellular osmolyte, ingestion of sodium helps maintain
extracellular fluid volume, including plasma volume.
Potassium is the major cation in the intracellular fluid
compartment. The balance of potassium in the extracellu-
lar versus intracellular space has important effects on neural
transmission, muscle contraction, and vascular tone. Potas-
sium is often included in fluid replacement beverages in rel-
atively small amounts to replace losses due to sweating or
diarrhea. Furthermore, potassium influences the concentra-
tion of sodium in the blood. This is because plasma sodium
concentration is a function of the mass balance of sodium,
potassium, and water (148, 346, 361).
Chloride is usually included as the major anion in most
fluid-replacement beverages. This is due in part to chloride
being the anion lost in the greatest amount in sweat. There
is limited information available about the effect of a fluid-
replacement beverages’ anion content on intestinal water and
solute absorption. One study has investigated the absorption
rate of a test solution with various anion compositions while
holding sodium and glucose concentrations constant. Using
the triple-lumen catheter technique, Fordtran (167) perfused
a solution with 80 mmol/L of sodium and 65 mmol/L of
glucose with either chloride, bicarbonate, or sulfate as the
anion to sodium. The author reported that maximal water and
sodium absorption in the jejunum was attained with chloride,
followed by bicarbonate, and then sulfate.
Protein
Protein is included in postexercise fluid-replacement bever-
ages to promote muscle protein synthesis and thus aid in the
recovery process after physical activity (373). There are also
recent data on the role that protein may play in rehydration
and hydration-related physiological outcomes. For instance,
ingestion of milk has been studied for its efficacy in promot-
ing fluid retention during the postexercise rehydration process
(244,245). Furthermore, the promotion of albumin synthesis
with protein ingestion after repeated days of exercise in the
heat has been shown to enhance plasma volume expansion
and thermoregulatory adaptations in both young adults and
older individuals (199, 357). These studies are discussed
in more detail in the competitive sport and exercise section
below.
Amino acids have been investigated for their effects
on intestinal water and solute absorption, primarily in the
oral rehydration from diarrheal disease literature. Theoreti-
cally, amino acids could improve fluid absorption because
they utilize sodium-coupled transporters independently of the
sodium-hexose cotransporter. Thus, combining amino acids
with glucose in a fluid replacement beverage provides separate
but additive sodium cotransport systems (424). The addition
of glutamine to glucose-based oral rehydration solutions has
been shown to improve sodium chloride absorption in the
piglet jejunum (385, 386). However, because glutamine (and
other nitrogen-containing compounds) negatively affect bev-
erage palatability and stability, it is difficult to formulate oral
rehydration solutions with glutamine (424).
Osmotically active solutes
Glycerol is an osmotically active solute that is reported to
be evenly distributed among all fluid compartments. Scien-
tific evidence supports the efficacy of glycerol for inducing
a temporary hyperhydrated state when ingested with addi-
tional volumes of water. Glycerol promotes hyperhydration
by inducing an osmotic gradient that enhances renal water
reabsorption. Thus, when glycerol (typically 1 g of glycerol
per kg body mass) is ingested with water (typically ∼20 mL
per kg body mass), subjects retain more fluid (due to lower
free water clearance) compared to when they ingest a placebo
with water (341). Although glycerol can be used to retain fluid,
the physiological and performance advantages with hyperhy-
dration are inconsistent and the side effects often outweigh
the potential advantages (341, 414, 492). For a more com-
plete overview of the literature on glycerol hyperhydration
the reader is referred to recent review papers (341, 492).
Compounds that have central effects
Some beverages contain compounds that affect neurotrans-
mitter systems and are consumed for their potential to pre-
vent central fatigue and improve aspects of mood, cognitive
function, or physical performance. Some of these compounds
include caffeine, branched-chain amino acids, and tyrosine.
Caffeine’s central effects are likely mediated by the
blocking of adenosine receptor sites. Adenosine is a
neurotransmitter-inhibitor with sedative-like properties, thus
blocking the action of adenosine results in a stimulatory
effect on the sympathetic nervous system. Caffeine is a rel-
atively nonspecific adenosine antagonist, but seems to have
the highest affinity for the A1 receptor subtype, which is
located in several regions of the brain involved in arousal
(165, 177). Caffeinated beverages are routinely consumed
Volume 4, April 2014 591
Optimal Composition of Fluid-Replacement Beverages Comprehensive Physiology
for their beneficial effects on mood, alertness, cognition, and
physical performance. Also, contrary to popular belief, sev-
eral studies have shown that caffeinated fluids contribute to
hydration in a similar manner to noncaffeinated fluids (26,
133, 202). To determine the effects of caffeinated bever-
ages on hydration status, Grandjean et al. (202) had 19 to
39 year old free-living men (participating in light exercise
only) consume various combinations of caffeinated and non-
caffeinated beverages, including water only; water and caf-
feinated cola; water, caffeinated cola, and instant coffee; or
water and noncaffeinated citrus soft drink; in a counterbal-
anced, crossover manner for one 24-h period. The subjects’
diets were standardized and total fluid intake was controlled
at 35 mL/kg/d (clinical practice guideline) to avoid over- and
underdrinking. There were no significant differences in uri-
nary output or change in body weight among beverage treat-
ments, suggesting that caffeinated beverages do not hinder
hydration.
Through their potential effects on brain serotonin synthe-
sis, branched-chain amino acids (BCAA; including leucine,
isoleucine, and valine) have been proposed as a nutritional
countermeasure to central fatigue (128). Changes in brain
serotonin are known to affect mood, arousal, and sleepiness.
Brain serotonin synthesis increases when the ratio of plasma
free tryptophan-to-BCAA rises, thus ingestion of BCAA has
been hypothesized to reduce the concentration of serotonin in
the brain and mitigate central fatigue.
Tyrosine is a dietary precursor for catecholamine
(dopamine and norepinephrine) synthesis and has been pur-
ported to mitigate some adverse behavioral, cognitive, and
physiological effects of acute stress. It has been hypothesized
that central catecholamine neurons are unable to synthesize
sufficient neurotransmitter (particularly norepinephrine) dur-
ing acutely stressful situations (e.g., sleep deprivation, emo-
tional/mental stress, or exposure to cold, heat, or altitude
in military applications) (519). Administration of tyrosine
(a catecholamine precursor) is thought to enhance the abil-
ity of neurons to release neurotransmitter, potentially pre-
venting the cognitive deficits that typically occur with stress
(219, 521).
<