relative to that in the extracellular fluid, and this
is often quoted to suggest that sweat losses will
result in the need for potassium supplementa-
tion, but the sweat concentration is low relative
to the intracellular potassium concentration
(Table 17.1). The potassium loss in sweat (about
4–8 mmol · l–1, or 0.15–0.3 g · l–1) is small relative to
the typical daily intake of about 3.2 g for men
and 2.4 g for women (Gregory et al. 1990). In spite
of the relatively high concentration of potassium
in sweat, the normal response to exercise is for
the plasma potassium concentration to increase
due to efflux of potassium from the intracellular
space, primarily from muscles, liver and red
blood cells (Maughan 1994).
There is generally little change in the plasma
magnesium concentration during exercise, but a
slight fall may occur during prolonged exercise,
and this has been attributed to the loss of magne-
sium in sweat. Some support for the idea that
losses in sweat may be responsible comes from
the observation of a larger fall in the serum mag-
nesium concentration during exercise in the heat
than at neutral temperatures (Beller et al. 1972),
but a redistribution of the body’s labile magne-
sium seems to be a more likely explanation for
any fall in plasma magnesium concentration
during exercise (Maughan 1994). Although the
concentration of magnesium in sweat is high
relative to that in the plasma (Table 17.1), the
plasma content represents only a small fraction
of the whole body store; Costill and Miller (1980)
estimated that only about 1% of the body stores
of these electrolytes was lost when individuals
were dehydrated by about 6% of body mass.
Magnesium loss in sweat is considered by
some athletes and coaches to be a potentially
serious problem and to be a contributing factor
to exercise-induced muscle cramp, resulting in
suggestions that magnesium salts should be
included in the formulation of drinks intended
for consumption during exercise, but there is
little evidence to substantiate this belief. Addi-
tion of magnesium to intravenous fluids admin-
istered to athletes with cramp after a triathlon
was shown not to be effective in relieving
the cramp (O’Toole et al. 1993). The causes of
exercise-induced muscle cramp are not well
understood, and descriptive studies measuring
changes in blood or plasma electrolyte concen-
trations or sweat losses of electrolytes are
unlikely to provide any answers.
The sweating response to exercise is influ-
enced by the hydration status of the individual,
and sweat rates and thus thermoregulatory
capacity, will fall if a fluid deficit is incurred
(Sawka 1988). Less sweat is secreted for any
given increase in core temperature. For reason-
ably well hydrated individuals, however,
drinking during exercise seems to have little
(Cageet al. 1970) or no (Davis & Yousef 1987)
effect on sweating rate and to have no effect on
sweat composition, even when plain water or
electrolyte-containing solutions are consumed.
Senay and Christensen (1965), however, ob-
served that fluid ingestion in dehydrated sub-
jects exposed for prolonged periods to hot (43°C)
dry (<40% rh) conditions stimulated a prompt
sweating response and increased skin blood
flow, suggesting that fluid ingestion may restore
thermoregulatory capacity in dehydrated indi-
viduals. It is clear that most of the benefits in
terms of physiological responses and perfor-
mance capacity that accrue from a period of
acclimatization are lost if an individual becomes
dehydrated (Sawka 1988).Gastrointestinal function and
availability of ingested fluids
The available evidence suggests that most ath-
letes do not ingest sufficient fluid to replace
losses (Murray 1996). In some situations, oppor-
tunities for replacement are limited by the rules
of sport, with drinks being available only during
scheduled breaks in play, but even when there is
unlimited access to fluids, intake is generally less
than loss.
The first physical barrier to the availability of
ingested fluids is the rate of gastric emptying,
which controls the rate at which fluids are deliv-
ered to the small intestine and the extent to
which they are influenced by the gastric secre-
tions. The rate of emptying is determined by