reviews influencing restoration offluid and electrolyte in · rehydration in manafterexercise...

BrJ Spors Med 1997;31:175-182

Reviews

Factors influencing the restoration of fluid andelectrolyte balance after exercise in the heat

R J Maughan, J B Leiper, S M Shirreffs

SummaryMaintenance of fluid balance is a majorconcern for all athletes competing in eventsheld in hot climates. This paper reviews recentwork relating to optimisation of fluid replace-ment after sweat loss induced by exercising inthe heat. Data are taken from studies under-taken in our laboratory. Issues investigatedwere drink composition, volume consumed,effects of consuming food with a drink, effectsof alcohol on rehydration effectiveness, volun-tary intake of fluid, and considerations forwomen related to the menstrual cycle. Theresults are presented as a series ofsummaries ofexperiments, followed by a discussion of theimplications. The focus of this review is urineoutput after ingestion of a drink; fluid excretedin urine counteracts rehydration. Also includedare data on the restoration of plasma volumelosses. Ingestion oflarge volumes ofplain waterwill inhibit thirst and will also promote adiuretic response. If effective rehydration is tobe maintained for some hours after fluid inges-tion, drinks should contain moderately highlevels of sodium (perhaps as much as 50-60mmol/l) and possibly also some potassium toreplace losses in the sweat. To surmount ongo-ing obligatory urine losses, the volume con-sumed should be greater than the volume ofsweat lost. Palatability of drinks is important instimulating intake and ensuring adequate vol-ume replacement. Where opportunities allow,the electrolytes required may be ingested assolid food consumed with a drink. There are nospecial concerns for women related to changesin hormone levels associated with the men-strual cycle. Ingestion of carbohydrate-electro-lyte drinks in the post-exercise period restoresexercise capacity more effectively than plainwater. The effects on performance of anuncorrected fluid deficit should persuade allathletes to attempt to remain fully hydrated atall times, and the aim should be to start eachbout ofexercise in a fluid replete state. This willonly be achieved if a volume offluid in excess ofthe sweat loss is ingested together withsufficient electrolytes.

IntroductionSustained hard exercise in a hot environmentpresents a greater challenge to the body'shomoeostatic mechanisms than any othercircumstance. The combination of a high rate

of metabolic heat production and a restrictedcapacity for heat dissipation leads to hyperther-mia, which may progress to heat illness; thiswill inevitably impair exercise performance,and may in extreme cases be fatal.'The 1996 Olympic Games held in Atlanta in

July and August were but one example of amajor sporting event held in unfavourableclimatic conditions; many others are scheduledfor the near future. The combination of hightemperature and high humidity has majorimplications for everyone attending; this in-cludes spectators, officials, and team manage-ment as well as competitors. Performance in alloutdoor endurance events, which we can defineas those lasting longer than a total time ofabout 20-30 minutes, is generally reduced, andthere seems to be no way of avoiding someimpairment of performance. A recent studyunder laboratory conditions2 has shown thatendurance time on a bicycle ergometer at anexercise intensity that could be sustained for 92minutes at a temperature of 11 C was reducedto 83 minutes when the temperature wasincreased to 21 C and to 51 minutes when thetemperature was increased to 30'C; in Atlantaconditions, the reduction would be evengreater. If the athlete is dehydrated beforeexercise begins, the reductions in performanceobserved in the heat are greatly magnified.For those unaccustomed to living in the heat,

the stress will be no different from that experi-enced by those who are accustomed to theheat, but the way in which the athletes dealwith the conditions may be the largest factorinfluencing the impact of the climatic stress ontheir performance. The successful competitorwill have prepared a coping strategy thatincludes acclimatisation, rehydration, and be-havioural and psychological components. Asidefrom behavioural mechanisms to minimise orprevent overheating such as moving indoors toan air conditioned cool environment, the physi-ological mechanisms activated by the acclima-tion process include an earlier onset ofsweatingand an increased sweating rate. Sweating canbe very effective in removing heat from thebody, and substantial amounts of fluid arelikely to be lost in this way by many individuals.The acclimation process results in an increasedsweat loss and therefore increases, rather thandecreases, the need for fluid replacement. Thewater lost must be replaced in order to

University MedicalSchool, Foresterhill,Aberdeen AB25 2ZD,Scotland, UnitedKingdomR J MaughanJ B LeiperS M Shirreffs

Correspondence to:Professor Maughan.

Accepted for publication29 May 1997

175

on June 19, 2020 by guest. Protected by copyright.

http://bjsm.bm

j.com/

Br J S

ports Med: first published as 10.1136/bjsm

.31.3.175 on 1 Septem

ber 1997. Dow

nloaded from

http://bjsm.bmj.com/

Maughan, Leiper, Shirreffs

maintain body water balance and allow the bestpossible athletic performance to be achieved.In addition, it is well established that dehydra-tion removes the thermoregulatory advantageand improved exercise tolerance that resultfrom acclimatisation.'Although the physician and physiologist dis-

approve, it seems to be beneficial, at least interms of performance, for athletes in weightcategory sports, including wrestlers, boxers,and weightlifters, to perform while in adehydrated state; the disadvantage arising fromcompeting with a body water deficit has beenfound by trial, error, and the experience ofcoach and athlete to be outweighed by theadvantages of competing in a weight categorybelow the athlete's natural weight.4 The advan-tage probably arises from biomechanical fac-tors, including limb leverage advantages. How-ever, the optimisation of rehydration aftersweat loss induced to make weight at the timeof the weigh-in may be of particular im-portance to performers in these sports. Thereare also many other athletes, including thosewho have to compete in more than one roundof a competition on the same day or on succes-sive days, for whom recovery and rehydrationbetween events is of crucial importance.The difficulty of achieving effective rehydra-

tion after thermal dehydration has long beenrecognised. Ingestion of water causes a promptdiuresis even when the individual is still hypo-hydrated, effectively preventing a return toeuhydration.5 More recent studies have con-firmed that plain water is not the best solutionto be consumed after exercise to replace thewater lost as sweat.67 These studies have high-lighted the need for replacement of electrolytesas well as water, but have generally beennon-systematic, with a number of variableschanging simultaneously, leading to some diffi-culties in the interpretation of the results and inthe formulation of recommendations for ath-letes. These reports have, however, also re-peated earlier observations that voluntary fluidconsumption generally stops short of theamount necessary to replace losses, leading toan involuntary dehydration. Levels of hypo-hydration equivalent to about 1% ofbody massare sufficient to affect thermoregulatory re-sponses to exercise; deficits of 2-3% of totalbody water, although sufficient to impair exer-cise performance, are not effective in initiatingthe drive to drink, and substantial fluid deficits(up to 5-10% of total body water) are well tol-erated at rest by the healthy individual.3 Ifphysical work is to be sustained, however, theneed for adequate hydration becomes critical atthese levels of hypohydration.

This review will describe some of our recentwork concerning issues which must be consid-ered if optimisation of fluid replacement aftersweating induced by exercising in the heat is tobe achieved. The aim of this series of investiga-tions has been to complete a systematic investi-gation of some of the factors that influence res-toration of fluid balance after exercise-induceddehydration. The factors that govern fluidreplacement during exercise have been exten-sively investigated and are the subject of a

number of comprehensive reviews8'-"; theseissues will not be discussed further here. Wehave focused on replacement after exercise inpreparation for the next bout of exercise, be ittraining or competition. Our concerns dealwith the composition of the fluid consumed,with particular regard to its electrolyte content,the volume of fluid consumed, the effects ofconsuming food together with a drink, theeffects of consuming alcohol on rehydrationeffectiveness, voluntary intake of fluid afterexercise, and any special considerations forfemale athletes in view of possible menstrualcycle effects on the rehydration effectiveness ofbeverages. Finally, there has been an initialinvestigation of the possible implications of therestoration of fluid and electrolyte balance onthe ability to perform an exhausting exercisetask. These results are presented as a series ofsummaries of individual experiments, followedby a short discussion of the implications for theathlete in training or competition.

General outline of studiesIn all of these individual experiments, amoderate level of dehydration was induced byintermittent exercise in a hot (about 340C)humid (60-80% relative humidity) environ-ment. Subjects exercised until body mass haddecreased by about 2% of the initial value; fora 70 kg subject, this corresponded to a sweatloss of about 1.4 litres. Beginning 30 minutesafter the end of this exercise period, drinks ofvarying volume or composition were providedover a fixed time (60 minutes in most of thestudies), and thereafter no further intake wasallowed until the end of the study period.Measurements of urine volume and composi-tion and of a variety of plasma variables (elec-trolyte concentrations, hormone levels, osmo-lality, and blood and plasma volume) weremade throughout the study.

Electrolyte content of drinksA study was undertaken to examine the effectof the sodium content of drinks on therehydration process after exercise-induced de-hydration equivalent to 1.9% of body mass ofsix fasted but euhydrated men."2 After dehydra-tion they consumed drinks with sodiumconcentrations of 2, 26, 52, and 100 mmol/lover a 30 minute period beginning 30 minutesafter the end of exercise; the volume consumedwas 1.5 times their mass loss by dehydrationwhich amounted to about 2 litres in all trials.The entire volume of urine produced over the5.5 hours after the end of the drinking periodwas collected and measured (no other food ordrink was consumed after the rehydrationperiod). The volume of urine produced wasinfluenced by the quantity of sodium con-sumed, such that it was greatest when the 2mmol/l drink was consumed and least when the100 mmol/l drink was consumed (fig 1),although the differences did not reach a level ofsignificance between the 2 and 26 and betweenthe 52 and 100 mmol/l drinks. Assuming asweat sodium concentration of 50 mmol/l it ispossible to calculate the whole body sodiumbalance at the end of the study period and

176


http://bjsm.bm

j.com/

Br J S


.31.3.175 on 1 Septem

ber 1997. Dow

nloaded from


Rehydration in man after exercise

1400 - u- mmoI/IE -A- 100 mmol/l1200

o 1000-ax,CD

800

> 600-

E3 200

0 _0 0.5 1.5 3.5 5.5

Time after fluid ingestion (hours)

Figure 1 Cumulative urine output after rehydration overtime after exercise-induced dehydration followed byingestion of a fixed volume ofglucose-electrolyte drinks withdiffering sodium content. Significant treatment effects wereseen after the rehydration period. Redrawn from Maughanand Leiper. "

150 r

50

0

-50

-100

-1000

D 9 mmrIl/lL- ,11 | luill I

O 26 mmol/I0 52 mmol/lA\ 100 mmol/l

/i

_ Iw

-750 -500 -250

Water balance (ml)

Figure 2 Relation between whole body water balance andsodium balance at 5.5 hours after the end ofexercise-induced dehydration followed by ingestion of a fixedvolume offluid with different sodium concentration. Asweat sodium concentration of 50 mmolll was assumed.Redrawn from Maughan and Leiper. 2

relate this to water balance, taking the pre-exercise values as the zero point (fig 2); thisclearly shows that there is a strong relationshipbetween the sodium content of the ingestedfluid and its ability to restore water balance.

Blood samples were collected before and 30minutes after the dehydration period (immedi-ately before the test drink was consumed) andthen at intervals until 5.5 hours after the end ofthe rehydration period. Plasma volume changes,calculated according to the formula of Dill andCostill," decreased with dehydration in alltrials and increased after rehydration in alltrials. Restoration of plasma volume occurredless rapidly with the 2 mmol/l beverage, in that1.5 hours after the end of the fluid ingestionperiod, the increase in plasma volume was6.8% in this trial, with increases of 12.4 and12.0% with the 52 and 100 mmol/l drinksrespectively. There was no difference in thechange in plasma volume between trials 5.5hours after the end of the rehydration period,but there was a tendency for it to be related tothe quantity of sodium consumed, with thesmallest increase with the 2 mmol/l beverage

and the greatest increase with the 100 mmol/lbeverage.Sodium has an important function in assist-

ing effective rehydration, largely as a conse-quence of its role as the major ion of the extra-cellular fluid. If sufficient sodium and water areingested, some of the sodium remains in thevascular space, with the result that plasmaosmolality and sodium concentration do notdecline as may occur if plain water is ingested.As a result, the plasma levels ofvasopressin andaldosterone are maintained, and an inappropri-ate diuresis, inappropriate because the body isstill in net negative fluid balance, is prevented.Where ad libitum fluid intake is possible,maintaining the plasma osmolality and the cir-culating sodium concentration also plays a rolein stimulating the drive to drink, and thus helpsto ensure that an adequate volume is con-sumed.

Potassium is the major ion of the intracellu-lar fluid, and it has been postulated that theinclusion of potassium in drinks consumedafter sweat loss may aid rehydration byenhancing the retention of water in theintracellular space.'4 In a study designed toinvestigate this, eight male volunteers weredehydrated by 2.1% of body mass by intermit-tent cycle ergometer exercise in the heat.'5Subjects ingested a glucose beverage (90mmol/l), a sodium containing beverage (NaCl60 mmol/l), a potassium containing beverage(KCl 25 mmol/l), and a beverage consisting ofthe addition of all three over a 30 minuteperiod beginning 45 minutes after the end ofexercise. The drink volume consumed wasequivalent to the volume of sweat lost andamounted to about 1.6 litres in all trials; noother food or drink was consumed during thestudy. The urine produced and excreted fromthe end of the rehydration period for the nextsix hours was again collected.A smaller volume of urine was excreted after

rehydration when the electrolyte containingbeverages were ingested compared with theelectrolyte free beverage (fig 3). An estimatedplasma volume decrease of 4.4% was observed

0

E

CD0cJCDCD

.0

z

-1000

-2000*e,rip !C

c& &

1 2 4 6

Time after rehydration(hours)

Figure 3 Changes in whole body fluid balance over timeafter exercise-induced dehydration followed by ingestion ofdrinks containing no electrolytes or containing sodium,potassium, or both sodium and potassium. Cumulativeurine output was higher after ingestion of the electrolyte freedrink than after ingestion of any of the other drinks,causing a rapid return to a negative fluid balance. Based

on data from Maughan et al. 5

EEa,0cCo

.0E

0C,)

177

lo00


http://bjsm.bm

j.com/

Br J S


.31.3.175 on 1 Septem

ber 1997. Dow

nloaded from



with dehydration over all trials. After drinkingestion, plasma volume increased in all trials,

00 t Obut the rate of recovery was slower when thei, 00 +KC1 beverage was consumed. However, by the

end of the study period, six hours after the endXOO_ of the rehydration period, the increase was not

-° ° different between trials and amounted to 7.5(5. 1)% for the glucose-electrolyte beverage, 9.7(6.2)% for NaCl, 7.8 (5.1)% for the KC1 solu-tion, and 7.9 (2.0)% for the beverage contain-ing all three components (results are means(SD)). Although there were differences in the

w __total amount of electrolyte replaced as well asa o C - differences in the type of electrolytes present in-] __> O + _ the drinks, there was no difference in the frac-tion of ingested fluid retained six hours afterthe drinks that contained electrolytes had been

t- 0o finished. It could be postulated, however, that

11 al ova ° -asthe beverage volume consumed was equiva-C., Ad C! tlent to the volume of sweat lost, subjects were

tXS a: > C dehydrated throughout the entire study, even0IA IA %O (1 0 ,after the drinking period. The volumes of urineUsf R excreted were close to basal levels, and signifi-

cant further reductions in output may not have00 al> Obeen possible when both sodium and potas-S + 8 sium were ingested, over and above the reduc--ta oW o tions already induced when the sodium and

00 _4X - X I,potassium were ingested separately. This high-

0 tnso^ lights the need to ensure that the volume of0n -_ _ fluid consumed is sufficient not only to meet

the fluid deficit incurred by sweating, but alsoto provide for ongoing urine and other losses.

Z o o Effects of the volume of fluid consumedAs obligatory urine losses persist even in thedehydrated state, it is clear that any drink con-

e o N - sumed after exercise-induced or thermal sweat-x N t , ingmustbe consumed in a volume greater than<~ Ethe volume of sweat that has been lost. In order00

t _ °to investigate the influence of drink volume on0 ^ a > o rehydration effectiveness, 12 male volunteers

s-^~0 s performed intermittent exercise in the heat ino g-.>̂ a, order to induce a level of dehydration equiva-oC to a -o 0

X lent to a mean of 2.06% of their initial body0 0 tnr CBy c w o mass."6 Over a 60 minute period beginning 30minutes after the end of exercise, drinkvolumes equivalent to 50, 100,150, and 200%of the sweat loss were consumed. In addition,

N tob in order to investigate the possible interactionS @ E ^ ^ co | - between beverage volume and its sodium con-00 tn

S a _tent, six of the subjects consumed a relatively.t S m om t low sodium concentration drink (23 mmol/l),II E wand six a moderately high sodium concentra-

tion drink (61 mmol/l). For six hours after the00 j E ^end of the drinking period, subjects consumed

no other food or drink, and the entire volumeof urine produced and excreted was collecteda X - ^> > a,.andmeasured. When either the low or highe g a ° £- sodium beverage was consumed, the urine vol-

-t " M a v v, .ume produced was, not surprisingly, related tothe beverage volume consumed, such that thesmallest volumes were produced when 50% of

A:: the loss was consumed and the greatest when200% of the loss was consumed (table 1).As different drink volumes were consumed

in each of the trials, a calculation of fluidbalance status relative to the situation before_

O Ug.g m R the dehydration allows easier comparisons thana simple measure oftheurinevolumeproduced(table 1). With dehydration, individuals move

178


http://bjsm.bm

j.com/

Br J S


.31.3.175 on 1 Septem

ber 1997. Dow

nloaded from



Table 2 Volume of urine producedfollowing drink ingestion. Also shown is the fluidvolume consumed and the quantities ofmajor electolytes ingested. Values are mean (SD) ormedian (range) as appropriate

Meal + water Sports drink

Fluid volume (ml) 2076 (371) 2042 (373)Electrolytes ingested (mmol)Na+ 63 (11) 43 (8)K+ 21 (3) 7 (3)

6 h urine volume (ml) 665 (396 to 1190) 934 (550 to 1403)

into negative fluid balance; by drinking theyreturn towards a positive fluid balance status,and, only if the volume consumed is greaterthan the sweat loss, do they become positivelyhydrated. Urine output moves them towardsand/or into a state of negative fluid balance sta-tus again. As shown in table 1, subjects weresignificantly hypohydrated when they con-sumed a volume equivalent to only half theirsweat loss irrespective of the drink composi-tion. With a drink volume equivalent to that ofthe sweat loss, subjects were also hypohydratedat this time, but significantly less so when thehigher sodium beverage had been consumed,confirming the effectiveness of sodium atpromoting rehydration. When a drink volumedouble the sweat loss was consumed, subjectswere still slightly hypohydrated six hours afterdrink ingestion when it had a low sodium con-centration and indeed they were in a similarcondition to when they drank the same bever-age in a volume of only 150% of their sweatloss. With the high sodium drink, subjects hadretained enough of the fluid to maintain themin a state of hyperhydration six hours afterdrink ingestion when they consumed either150% or 200% of their sweat loss, and theexcess retained would eventually be lost byurine production or by further sweat loss if theindividual commenced exercise or moved to awarm environment. Calculated plasma volumechanges indicated a decrease of about 5.3%with dehydration. At the end of the studyperiod, the general pattern irrespective ofwhich drink had been consumed, was for theincreases in plasma volume to be a direct func-tion of the volume of fluid consumed; in addi-tion, the increase tended to be greater for thoseindividuals who ingested the high sodium drink(table 1).

Effects of solid food and fluidconsumptionIn many situations it is not uncommon to con-sume solid food between exercise bouts, andindeed in most situations it probably should beencouraged. We undertook a further study inwhich eight volunteers (five men, threewomen) dehydrated by 2.1% of their bodymass by exercising in the heat and then, over a60 minute period starting 30 minutes after theend of exercise, consumed either a solid mealplus virtually electrolyte free flavoured water or

Table 3 Volume of drink consumed andfraction offluid retained six hours after ingestionofdrink of various alcoholic strengths (0, 1, 2, or 4%). Values are mean (SD)

0% 1% 2% 4%

Drinkvolume (ml) 2178 (191) 2240 (147) 2275 (154) 2155 (125)Fraction of fluid retained(%) 59.3 (15.7) 53.1 (11.0) 50.0 (16.2) 40.7 (13.7)

a commercially available sports drink"; thevolume of fluid contained within the meal pluswater was the same as the volume of sportsdrink consumed. For six hours after the end ofthe eating and/or drinking period, the entirevolume of urine produced and excreted wascollected. The volume of urine produced afterfood plus fluid ingestion was significantly lessthan that when the drink alone was consumed(table 2). It was calculated that plasma volumedecreased by 5.4% with dehydration over alltrials. The plasma volume increased after rehy-dration in all trials, and there was no differencein the increase between the food plus fluid trial(11.7 (2.0)%) and drink only trial (1 3.2(4.2)%) (means (SD)). The quantity of waterconsumed with both these rehydration meth-ods was the same, but the meal had a greaterelectrolyte content (table 2), and it seems mostlikely that the greater efficacy of the meal pluswater treatment in restoring whole body waterbalance was a consequence of the greater totalcation content.

Alcohol consumptionMany people advise against the consumptionof alcohol and caffeine containing beverageswhen fluid replacement is desired because theyhave been shown, at least in some circum-stances, to act as diuretics. However, it is obvi-ous that many people enjoy consuming thesetypes of beverages in many circumstances. Wetherefore undertook a study to investigate theeffect ofconsuming alcohol after exercise in theheat sufficient to induce dehydration equiva-lent to 2.01 (0.06)% of body mass.'8 19 Over a60 minute period beginning 30 minutes afterthe end of exercise, subjects consumed avolume equivalent to 150% of their mass loss ofdrinks containing 0, 1, 2, or 4% alcohol; thedrink composition was identical in all otherrespects. The volume of urine produced andexcreted for the six hours after drink ingestionwas measured and was found to be related tothe quantity of alcohol consumed, with a largervolume the more alcohol that was consumed(table 3). However, despite the definite ten-dency for the urinary output to increase withincreasing alcohol, only with the 4% beveragedid the increased value approach significance.The calculated decrease in plasma volume withdehydration was about 7.6% across all trials.With rehydration the plasma volume increased,but the rate of increase seemed to be related tothe quantity of alcohol consumed; at the end ofthe study period, the increase in plasma volumerelative to the dehydrated value was 8.1 (3.2)%with 0% alcohol, 7.4 (2.7)% with 1%, 6.0(3.4)% with 2%, and 5.3 (3.4)% with 4% (allresults means (SD)).

Voluntary beverage consumptionWe have already demonstrated and discussedthe need for drinking more than the sweat vol-ume lost to have a chance of rehydrating effec-tively. However, in practice it is left to the dis-cretion of an individual how much theyconsume and what they choose to drink. In astudy to examine the effect of palatability,together with the solute content ofbeverages in

179


http://bjsm.bm

j.com/

Br J S


.31.3.175 on 1 Septem

ber 1997. Dow

nloaded from



Table 4 Volume ofdrink consumed and netfluid balancefour hours after rehydration. Values are mean (SD). For netfluid balance: 0 = euhydrated, + = hyperhydrated, - = hypohydrated

Glucose 90 mmol/l + Sports drink (sodium Orange/lemonadesodium 60 mmol/1 + 24 mmol/l + (sodium 2 mmolll+potassium 24 mmol/l Aerated water potassium 4 mmol/l) potassium 24 mmolll)

Drink volumne (ml) 1796 (758) 1750 (560) 2492 (661) 2488 (233)4 h net fluid balance (ml) -123 (615) -474 (333) -135 (358) 45 (238)

promoting rehydration after sweat loss, eightmen exercised in the heat to lose 2.1% of theirmass.20 Over a two hour period after exercise,subjects were allowed to drink ad libitum; thedrinks they received, each on a separateoccasion, were a glucose-electrolyte beverage,aerated water, a commercial sports drink, andan orange juice/lemonade mixture. Subjectsdrank a greater volume of the sports drink andorange juice/lemonade mixture, the taste ofwhich was perceived as being more pleasant(table 4). As different volumes of drinks wereconsumed, a calculation of fluid balance statusrelative to the situation before the dehydrationallows easiest comparisons; with dehydration,individuals move into a negative fluid balancestatus, and by drinking they move towards apositive fluid balance status, and, only if thevolume consumed is greater than the sweatloss, do they become positively hydrated, andurine production and excretion moves themtowards and/or into the negative fluid balancestatus again. By consuming a greater volume,individuals give themselves a better chance ofeffectively rehydrating, and therefore the palat-ability of drinks, together with their composi-tion, should be considered when free choice isgiven to individuals, as indeed occurs in mostsituations.

Menstrual cycle effects: concerns forfemale athletesAlmost all previous investigations into rehydra-tion after exercise, as with so much humanphysiology, have been concerned exclusivelywith men. This is in part a consequence of thepossibility that the steroid hormones, which aresubject to cyclical variation in women, mayhave an influence on fluid balance, and thewish to avoid this confounding factor. Wetherefore undertook a specific study to estab-lish whether or not there is an acute effect ofthese hormones on fluid balance in the fewhours after exercise-induced sweat loss.21 Fivewomen, each with a regular menstrual cycle,exercised in the heat to dehydrate themselvesby 1.8% of their body mass. They did this atthree different stages of their menstrual cycle(two days before, five and 19 days after theonset of menses), and, over a 60 minute periodbeginning 30 minutes after the end of exercise,they consumed the same quantity of the samebeverage on each occasion; the volume con-

Table 5 Volume of urine produced after drink ingestion at various stages of the menstrualcycle. Also shown is the volume ofdrink consumed. Values are mean (SD) or median(range) as appropriate

2 days before the 5 days after the 19 days after theonset of menses onset of menses onset ofmenses

Drink volume (ml) 1662 (195) 1550 (172) 1392 (241)6 h urine volume (ml) 714 (469 to 750) 476 (433 to 639) 534 (195 to 852)

sumed was 150% of the mass loss and thedrink, a commercially available sports drink,was the same in all trials. For six hours after theend of the rehydration period, the entirevolume of urine excreted was collected andmeasured. There was no difference in thevolume of urine produced (table 5), and hencein the volume of the ingested fluid that wasretained at the different stages of the menstrualcycle. These results suggest that the acutereplacement ofvolume losses incurred by sweatloss due to exercising in the heat are notaffected by the normal regular menstrual cycle.Therefore women seem not to be disadvan-taged when rapid and complete restoration ofexercise-induced sweat loss is required.

Performance effectsThe studies described above have used restora-tion of fluid balance as an index of recoveryafter exercise-induced dehydration, but manyother factors are involved in the restoration ofexercise capacity. To investigate some of thefactors that influence recovery of performancecapacity, a further experiment was undertaken;the study design was similar to that used previ-ously, but subjects performed a standardisedexercise test four hours after the end of therehydration period. Rehydration was achievedby administration of plain water, a flavouredplacebo, or a drink containing carbohydrate(68 g/l) and electrolytes (25 mmol/l sodium, 4mmol/l potassium) in a volume equal to 1.5times the sweat loss of about 2% of body mass.The performance test consisted of an incre-mental cycling exercise test continued to thepoint of exhaustion.The glucose-electrolyte test drink had an

electrolyte content similar to that of most com-mercial sports drinks, and there was a tendency(P = 0.073) for the cumulative urine outputover the four hour recovery period to be less inthis trial (1162 (816-1429) ml; median(range)) than in the placebo (1470 (1228-1667) ml) or water (1395 (954-1555) ml)trials. This is consistent with the results of theprevious trials, and confirms that this low levelof electrolytes will give a small increase in thevolume of fluid retained over this time scale.

All subjects recorded their best exercise per-formance, measured as the time to fatigue, onthe carbohydrate-electrolyte trial, with a mean(SD) exercise time of 18.00 (2.43) minutes,compared with 16.42 (2.15) minutes afterconsuming the plain water and 17.00 (1.76)minutes in the placebo trial. These results sug-gest that there is a benefit from the addition ofelectrolytes and glucose to drinks ingested afterexercise, but cannot distinguish between thevarious possible mechanisms that might beinvolved. It is, however, worth noting that there

180


http://bjsm.bm

j.com/

Br J S


.31.3.175 on 1 Septem

ber 1997. Dow

nloaded from



were no differences in the blood glucoseconcentration at any time between trials inspite of the ingestion of a substantial amount ofcarbohydrate on one of the trials. There isclearly a need for more systematic studies toresolve the outstanding questions.

Discussion and conclusionsRapid and complete restoration of fluidbalance after exercise is an important part ofthe recovery process, and becomes even moreimportant in hot humid conditions. The choiceof drink to be consumed after exercise may andindeed should be different depending on theindividual and their particular circumstances.Replacement of substrate in addition to waterand electrolyte losses may be of concern in thepost-exercise period in preparation for afurther bout of exercise. In terms of sustaininglife, substrate (muscle and liver glycogen)depletion is unlikely to have an adverse effect inan otherwise healthy individual, but waterdepletion, if not reversed, may have seriousconsequences. Moderate levels of dehydration(2-3% reductions in body mass) will impairperformance, and when an Olympic medal is atstake, it might be wise to assume that even verylow levels of dehydration may have a negativeimpact on performance; it is certainly the casethat hypohydration of only 1% of body masshas a measurable effect on the thermoregula-tory response to exercise.3 Severe dehydration(losses of more than about 6-7% ofbody mass)can result in a life threatening situation, andthis scenario is rendered more likely when theambient temperature is high. ' Such extremes ofwater depletion should not occur in any athletein the case of sports competition if a consciouseffort is made to ensure adequate fluid replace-ment. However, even if the health concerns areignored, the effects on performance of a fluiddecrement should be enough to persuade allindividuals to attempt to remain fully hydratedat all times, and particularly to ensure that theybegin each bout of exercise in a water repletestate.Team doctors, coaches, and athletes must be

aware of the factors that influence rehydrationand of the electrolyte content of the availabledrinks. The amount of electrolytes lost in sweatis highly variable between individuals, and,although the optimum drink may be achievedby matching electrolyte loss with equal quanti-ties from the drink, this is virtually impossiblein a practical situation. Sweat composition var-ies considerably between individuals, but alsovaries with time during exercise and will befurther influenced by the state of acclimation.22However, a moderate excess of salt intakewould appear to be beneficial as far ashydration status is concerned without any det-rimental effects on health provided that fluidintake is in excess of sweat loss and that renalfunction is not impaired. Concerns over thepossible adverse effects of a high salt intakehave led some athletes to restrict dietary saltintake.23 For the athletes with large sweatlosses, sodium loss will be correspondinglyhigh; loss of 5 litres of sweat with a sodiumcontent of 50 mmol/l requires ingestion of

almost 15 g of sodium chloride to restore bal-ance. This amount of sweat can easily be lost in2-3 hours of hard training or match practice inhot humid conditions. Although the diet willmake a major contribution to replacement,normal daily salt intake from food is only about6-8 g for the UK population, about half ofwhom rarely or never add salt to food at thetable.24 There is clearly a high risk of salt deficitwhen losses are high, unless a conscious effortis made to increase intake.

It is clear from the results described above, aswell as those of many other studies, thatrehydration after exercise can only be achievedif sweat electrolyte losses as well as water arereplaced. The oral rehydration solution recom-mended by the World Health Organisation forthe treatment of acute diarrhoea has a sodiumcontent of 60-90 mmol/l, reflecting the highsodium losses that may occur in some types ofdiarrhoea."5 In contrast, the sodium content ofmost of the major commercial sports drinks isin the range 10-25 mmol/1,22 and is even lowerin some products that are marketed as sportsdrinks. Most commonly consumed soft drinkscontain virtually no sodium and these drinksare therefore unsuitable when the need forrehydration is crucial. The problem with a highsodium concentration in beverages is that itmay exert a negative effect on taste, resulting inreduced consumption. It is equally clear, how-ever, that drinks with a low sodium content areineffective, and they will also reduce the stimu-lus to drink.'6

It can generally be concluded that in order toachieve effective rehydration after exercise inthe heat or heat exposure sufficient to causesweat loss, the rehydration beverage shouldcontain moderately high levels of sodium (atleast 50 mmol/l), plus possibly some potas-sium; a source of substrate is not necessary forrehydration, although a small amount ofcarbohydrate (<2%) may improve the rate ofintestinal uptake of sodium and water. Thevolume of beverage consumed should begreater than the volume ofsweat lost in order tomake provision for the ongoing obligatoryurine losses. Therefore the palatability of thebeverage is of importance; many individualsmay lose substantial amounts of sweat and willtherefore have to consume large amounts offluid to replace them, and this is more likely tobe achieved if the taste is perceived as beingpleasant. Although we have shown that wateralone is adequate for rehydration purposeswhen food that replaces the electrolytes lost insweat is also consumed, there are manysituations where intake of solid food is avoided.This is particularly true in weight categorysports where the interval between the weigh-inand competition is short, but is also the case inevents where only a few hours intervenebetween succeeding rounds of the competition.The above studies were carried out in a sys-

tematic way and did not attempt to replicatereal life situations where many other factorswill intervene. After the exercise in the heat,individuals were removed to a relatively coolindoor environment and undertook no furtherexercise. For athletes living in training camps

181


http://bjsm.bm

j.com/

Br J S


.31.3.175 on 1 Septem

ber 1997. Dow

nloaded from


182 Maughan, Leiper, Shirreffs

or in the village at major championship eventswhere the outside environment is hot for themajority of most days, it is highly likely thatthey will be unable to avoid some exposure to athermally stressful environment. Team supportstaff, officials, and spectators will also be at risk.All must be aware of the need to maintain fluidbalance and should consider rehydration afterpassive heat exposure as well as after exercise.Indicators such as a reduction in urine outputand failure to maintain body mass are morereliable measures of dehydration than the feel-ing of thirst.

SS was supported by a student research award from theGatorade Sports Science Institute for the study of menstrualcycle effects on rehydration. We would like to thank BelindaGooding of Masterfoods, UK, for the provision of Uncle Ben'srice and chili sauce used in the food plus fluid intake study. Wewould also like to thank Bass Brewers Ltd for the provision ofdrinks used in the alcohol study.

1 Sutton JR. Clinical implications of fluid imbalance. In:Gisolfi CV, Lamb DR, eds. Perspectives in exercise science andsports science, Vol 3. Fluid homeostasis during exercise. Carmel:Benchmark Press, 1990: 425-56.

2 Galloway SDR, Shirreffs SM, Leiper JB, Maughan RJ.Exercise in the heat: factors limiting exercise capacity andmethods for improving heat tolerance. Sports, Exercise andInjury 1997;3: 19-24.

3 Sawka MN, Pandolf KB. Effects of body water loss onphysiological function and exercise performance. In:Gisolfi CV, Lamb DR, eds. Fluid homeostasis during exercise.Carmel: Benchmark Press, 1990:1-38.

4 Horswill CA. Physiology and nutrition for wrestling. In:Lamb DR, Knuttgen HG, Murray R, et al, eds. Perspectivesin exercise science and sports science. Vol 7. Physiology andnutrition for competitive sport. Carmel: Cooper, 1994:131-80.

5 Costill DL, Sparks KE. Rapid fluid replacement followingthermal dehydration. JAppl Physiol 1973;34:299-303.

6 Gonzales-Alonso J, Heaps CL, Coyle EF. Rehydration afterexercise with common beverages and water. Int J SportsMed 1992;13:399-406.

7 Nielsen B, Sjogaard G, Uglevig J, Knudsen B, Dohlmann B.Fluid balance in exercise dehydration and rehydration withdifferent glucose-electrolyte drinks. Eur J Appl Physiol1986;55:318-25.

8 Coyle EF, Coggan AR. Effectiveness of carbohydratefeeding in delaying fatigue during prolonged exercise.Sports Med 1984;1:446-58.

9 Lamb DR, Brodowicz GR. Optimal use of fluids of varying

formulations to minimise exercise-induced disturbances inhomeostasis. Sports Med 1986;3:247-74.

10 Maughan RJ. Carbohydrate-electrolyte solutions duringprolonged exercise. In: Lamb DR, Williams MH, eds. Per-spectives in exercise science and sports science. Volume 4.Ergogenics: the enhancement of sport performance. Carmel:Benchmark Press, 1991:35-85.

11 Murray R. The effects of consuming carbohydrate-electrolyte beverages on gastric emptying and fluid absorp-tion during and following exercise. Sports Med 1987;4:322-51.

12 Maughan RJ, Leiper JB. Sodium intake and post-exerciserehydration in man. EurJ,Appl Physiol 1995;71:311-19.

13 Dill DB, Costill DL. Calculation of percentage changes involumes of blood, plasma, and red cells in dehydration. JAppl Physiol 1974;37:247-48.

14 Nadel ER, Mack GW, Nose H. Influence of fluidreplacement beverages on body fluid homeostasis duringexercise and recovery. In: Gisolfi CV, Lamb DR, eds.Perspectives in exercise science and sports medicine. Volume 3.Fluid homeostasis during exercise. Carmel: Benchmark Press,1990:181-205.

15 Maughan RJ, Owen JH, Shirreffs SM, Leiper JB. Post-exercise rehydration in man: effects of electrolyte additionto ingested fluids. EurJAppl Physiol 1994;69:209-15.

16 Shirreffs SM, Taylor AJ, Leiper JB, Maughan RJ. Post-exercise rehydration in man: effects of volume consumedand sodium content of ingested fluids. Med Sci Sports Exerc1996; 28:1260-71.

17 Maughan RJ, Leiper JB, Shirreffs SM. Restoration of fluidbalance after exercise-induced dehydration: effects of foodand fluid intake. EurJApplPhysiol 1996;73:317-25.

18 Shirreffs SM, Maughan RJ. The effect of alcohol consump-tion on fluid retention following exercise-induced dehydra-tion in man. J Physiol (Lond) 1995;489:33P-34P.

19 Shirreffs SM, Maughan RJ. The effect of alcohol consump-tion on the restoration of blood and plasma volume follow-ing exercise-induced dehydration in man. J Physiol (Lond)1996;491:64P-65P.

20 Maughan RJ, Leiper JB. Post-exercise rehydration in man:effects of voluntary intake of four different beverages. MedSci Sports Exerc 1993;25:12S.

21 Maughan RJ, McArthur M, Shirreffs SM. Influence ofmenstrual status on fluid replacement after exercise-induced dehydration in healthy young women. BrJI SportsMed 1996;30:41-7.

22 Maughan RJ. Fluid and electrolyte loss and replacement inexercise. In: Harries M, Williams C, Stanish WD, MicheliLJ, eds. Oxford textbook ofsports medicine. New York: OxfordUniversity Press, 1994:82-93.

23 Bergeron MF. Heat cramps during tennis: a case report. IntY Sport Nutr 1996;6:62-8.

24 Gregory J, Foster K, Tyler H, Wiseman M. The dietary andnutritional survey ofBritish adults. London: HMSO, 1990.

25 Walker-Smith JA. Recommendations for oral rehydrationsolutions for use in the treatment of diarrhoeal disease inEuropean children. J Pediatr Gastroenterol Nutr 1992;14:113-15.

26 Nose H, Mack GW, Shi X, Nadel ER. Role of osmolalityand plasma volume during rehydration in humans. J ApplPhysiol 1988;65:325-31


http://bjsm.bm

j.com/

Br J S


.31.3.175 on 1 Septem

ber 1997. Dow

nloaded from


reviews influencing restoration offluid and electrolyte in · rehydration in manafterexercise...

Documents