ent carbohydrates, either alone or in combina-
tion, are effective. Glucose, sucrose, maltose and
glucose oligomers are all likely to promote
improved performance: addition of small
amounts of fructose to drinks containing other
carbohydrates seems to be acceptable, but high
concentrations of fructose alone are best avoided
because of a risk of gastrointestinal distress.
Fructose is poorly absorbed, and an osmotic
diarrhoea may occur after large doses. Some
sodium should probably be present, with the
optimum concentration somewhere between 10
and 60 mmol · l–1, but there is also a strong argu-
ment that, in events of short duration, this may
not be necessary. Adding sodium will have
several consequences, the most important of
which are a stimulation of water absorption and
the maintenance of plasma volume. In events of
longer duration, replacement of sweat sodium
losses and maintenance of plasma sodium con-
centration and osmolality become important
considerations. Sodium chloride, in high concen-
trations, may have a negative impact on taste,
and home-made sports drinks generally score
badly in this respect. There is little evidence to
suggest that small variations in the concentration
of these components of ingested fluids will sig-
nificantly alter their efficacy. There is not at
present any evidence to support the addition of
other components (potassium, magnesium,
other minerals or vitamins) to drinks intended to
promote or maintain hydration status.
In most situations, the volume of fluid
ingested is insufficient to match the sweat loss,
and some degree of dehydration is incurred
(Sawka & Pandolf 1990), and this suggests an
important role for palatability and other factors
that encourage consumption. It also indicates the
need for an education programme to make ath-
letes, coaches and officials aware of the need
for an adequate fluid intake: a conscious effort is
needed to avoid dehydration. Noakes et al. (1993)
reported that the voluntary fluid intake of
athletes in endurance running events seldom
exceeds about 0.5 l · h–1, even though the sweat
losses are generally substantially higher than
this. Even in relatively cool conditions and in
236 nutrition and exercise
sports that are less physically demanding than
marathon running, sweat rates of more than 1 l ·
h–1are not uncommon (Rehrer & Burke 1996).Pre-exercise hydration
Because of the need to minimize the impact of
sweat loss and volume depletion on exercise per-
formance, it is important to ensure that exercise
begins with the individual fully hydrated. On the
basis that a further increase in the body water
content may be beneficial, there have been many
attempts to induce overhydration prior to the
commencement of exercise, but these attempts
have usually been thwarted by the prompt
diuretic response that ensues when the body
water content is increased. Because this is largely
a response to the dilution of blood sodium con-
centration and plasma osmolality, attempts
have been made to overcome this. Some degree
of temporary hyperhydration can be induced if
drinks with sodium concentrations of 100 mmol ·
l–1or more are ingested, but this seems unlikely
to be beneficial for performance carried out in
the heat, as a high plasma osmolality will ensue
with negative implications for thermoregulatory
capacity (Fortney et al. 1984).
An alternative strategy which has recently
been the subject of interest has attempted to
induce an expansion of the blood volume prior to
exercise by the addition of glycerol to ingested
fluids. Glycerol in high concentrations has little
metabolic effect, but exerts an osmotic action
with the result that some of the water ingested
with the glycerol will be retained rather than
being lost in the urine, although there must be
some concern that the elevated osmolality of the
extracellular space will result in some degree of
intracellular dehydration. The implications of
this are at present unknown (Waldegger & Lang
1998), but it might again be expected that the
raised plasma osmolality will have negative
consequences for thermoregulatory capacity. The
available evidence at the present time seems to
indicate that this is not the case, but the results
of studies investigating the effects on exercise
performance of glycerol feeding before or during