Health · Explainer

Dehydration symptoms and how fluid loss during exercise affects the body

Dehydration symptoms such as thirst, a dry mouth or dark urine are widely used as signs of water loss, yet in older people none of them works well as a test on its own, while losing fluid through sweat during exercise raises body temperature and heart rate in step with how much has been lost (Hooper et al., 2015; Montain and Coyle, 1992). A Cochrane review of 67 tests in older people found that none worked consistently in more than one study (Hooper et al., 2015). During activity, core temperature climbs by about 0.15–0.20°C for every 1% drop in body weight (McDermott et al., 2017).

The article starts with the symptoms. It looks at how well they actually detect water loss and how much fluid people lose in sweat, then turns to what measured deficits do to temperature, the heart and performance, and to the opposite risk of drinking too much. Sources include a Cochrane review, laboratory trials, meta-analyses, a marathon cohort and professional position statements.

Sweating runner seen from behind sitting on a red running track on a hot day with a plain water bottle
A runner sits on a sunlit track holding a water bottle after training Editorial illustration

Dehydration symptoms and what they reflect

Dehydration means the loss of body water, and during exercise most of that water leaves through sweat (McDermott et al., 2017). Other causes include diarrhoea, vomiting, fever, bleeding, frequent urination and simply drinking too little (Hooper et al., 2015). Cheuvront and Kenefick argue that the central distinction is between losing fluid more dilute than the body's own, which pulls water out of cells, and losing fluid of the same concentration (Cheuvront and Kenefick, 2014).

Thirst, urine and other early signs

The position statement of NATA, the US body for athletic trainers, lists thirst and general discomfort as the first signs, at around a deficit of 2% of body mass, followed by flushed skin, tiredness, cramps and apathy (McDermott et al., 2017). Once losses pass 2%, people may feel dizzy or sick, vomit, get headaches or chills, or become short of breath (McDermott et al., 2017). Above 5%, athletes consistently show impaired performance, extreme thirst and headache (McDermott et al., 2017).

Urine darkens as its volume falls (McDermott et al., 2017). The same statement stresses that hydration shifts constantly, so judging it takes several indicators rather than one (McDermott et al., 2017).

Some groups regulate fluid less well. Children and older people are often named, because both can find dehydration harder to notice and correct (Masento et al., 2014). Children may misread thirst through inexperience, and older adults often have reduced kidney filtration, which makes water harder to conserve (Masento et al., 2014).

Severity scales differ by setting. Clinicians call a 1% to 5% body mass deficit mild and anything over 10% severe, whereas the athletic scale treats 2% to 5% as mild to moderate and more than 5% as severe (McDermott et al., 2017).

How well symptoms detect water loss

For older adults the familiar signs perform badly. A Cochrane review analysed 67 tests, using three studies with published accuracy data and raw datasets from 21 more (Hooper et al., 2015). Not one test held up in more than a single study (Hooper et al., 2015).

Heart rate, a dry mouth, feeling thirsty, fluid intake and the colour and volume of urine were all judged not useful as stand-alone checks in this age group (Hooper et al., 2015). Used alone, such tests miss many people who are dehydrated and wrongly flag others who are well hydrated (Hooper et al., 2015).

Exercise in the heat gives a different picture. In 22 healthy men who cycled in a 36°C room for up to 5 hours, a urine colour of 5 or darker flagged a loss of 2% or more of body mass with a sensitivity of 88.9% and a specificity of 84.8%, calculated across 69 urine samples (McKenzie et al., 2015).

The authors call urine colour a valid, cheap tool under those conditions, but they tested it only in a laboratory and only for water lost through exercise in heat (McKenzie et al., 2015). Weight change also tracks fluid loss well in sport, where changes are rapid, though it works less well for older people (Hooper et al., 2015).

In the cycling trial, samples from people who were not dehydrated averaged a colour of 3, against 6 for dehydrated samples (McKenzie et al., 2015). For readings of 5 or darker the positive likelihood ratio was 5.87, so such samples were close to 6 times as likely to come from dehydrated men as from normally hydrated ones (McKenzie et al., 2015).

Age changes the thirst signal itself. Many older exercisers, though not all, feel thirst less sharply when exercise or heat has dried them out (McDermott et al., 2017). Fluid balance is restored eventually, only more slowly (Kenney and Chiu, 2001).

How the body loses water during exercise

One review puts water at 60–80% of what the human body is made of (Masento et al., 2014). At rest, fluid inside cells accounts for roughly 40% of body mass, 15% sits between cells and 5% circulates in blood vessels (McDermott et al., 2017). Total body water normally stays within a narrow band, from 1% above to 3% below the usual level (McDermott et al., 2017). The water stored alongside glycogen is the subject of how much water glycogen holds.

Sweat rates and body mass change

Sweat output varies widely. Among adults during exercise it spans 0.5 to 4.0 L per hour (McDermott et al., 2017). Sweat rates and the salt content of sweat also differ markedly from one person to another (ACSM, 2007).

Weighing offers a simple estimate (ACSM, 2007). Sweat output can be gauged by comparing weight measured before a session with weight measured after it, and the ACSM position stand recommends tailoring fluid plans because individuals differ so much (ACSM, 2007). When people can drink freely according to thirst, they make up roughly two-thirds of the sweat they lose (McDermott et al., 2017). During exercise, thirst may be absent or ignored, a pattern known as voluntary dehydration (McDermott et al., 2017).

Effects on body temperature and the circulation

The clearest measured effects concern heat and the heart. In a classic crossover trial, eight trained cyclists rode for 2 h at 33°C in four sessions, drinking nothing or enough to replace 20%, 48% or 81% of their sweat (Montain and Coyle, 1992). Their body weight fell by 4.2%, 3.4%, 2.3% or 1.1% depending on the session (Montain and Coyle, 1992).

Core temperature and heart rate rose, and the volume of blood pumped per beat fell, in proportion to how much water the cyclists had lost (Montain and Coyle, 1992). The correlations were close to perfect, at r = 0.98 to 0.99, but they were calculated across four session averages rather than across individuals (Montain and Coyle, 1992). The four sessions produced similar sweat rates (Montain and Coyle, 1992).

Summaries of such work put the rise at roughly 0.15–0.20°C in core temperature and 3 to 5 extra heartbeats a minute for each 1% of body mass lost (McDermott et al., 2017). Heat strain adds to cardiovascular stress, lowering stroke volume and raising heart rate and vascular resistance (McDermott et al., 2017). The same statement says a deficit above 2% reduces sweating and delays its onset, although the cyclists above showed no change in sweat rate at losses of up to 4.2% (McDermott et al., 2017; Montain and Coyle, 1992).

Heat illness is part of the picture. NATA describes dehydration as a contributing cause of heat exhaustion and heat stroke during exertion and links a higher risk of heat illness to deficits beyond 3% (McDermott et al., 2017). It also calls dehydration one of many factors behind heat illness (McDermott et al., 2017).

Research on dehydration and exercise performance

Laboratory and field studies have produced apparently conflicting results, and the debate continues (McDermott et al., 2017). How much a deficit matters likely depends on what the task demands, whether endurance, strength, thinking or skill (Cheuvront and Kenefick, 2014). Training and fuelling for performance are covered in sport nutrition.

Endurance exercise

The 2% line appears in older guidance (ACSM, 2007). The ACSM's 2007 position stand, written as guidance on fluid replacement for active people, labels a water shortfall above 2% of body weight as excessive (ACSM, 2007). NATA states that prolonged aerobic performance was consistently reduced once the loss reached 2% (McDermott et al., 2017). Cheuvront and Kenefick likewise support a 2% threshold for endurance, linked to the loss of fluid volume (Cheuvront and Kenefick, 2014).

A meta-analysis of fifteen laboratory studies with 122 participants challenged that rule (Goulet, 2013). When cyclists set their own pace in time-trial tests, dehydration made no measurable difference, while tests at a fixed intensity showed a 1.91% drop in endurance (Goulet, 2013). The author concluded that losses up to 4% are very unlikely to impair performance in time-trial conditions and that the 2% rule came from fixed-intensity tests (Goulet, 2013).

Field observations point the same way, since athletes have performed outstandingly while more than 2% dehydrated (Goulet, 2013). A 2015 consensus panel judged deficits of up to about 3% of body mass tolerable for endurance and muscular power in temperatures between minus 10°C and 20°C (Hew-Butler et al., 2015).

Strength and power

One meta-analysis found that dehydration reduced muscle endurance by 8.3%, strength by 5.5% and anaerobic power by 5.8% (Savoie et al., 2015). Anaerobic capacity and vertical jump did not change significantly (Savoie et al., 2015).

The size of the losses did not track the size of the water deficit (Savoie et al., 2015). Dehydrating people through exercise cut performance by a further 5.4% compared with passive methods, so the method of dehydration seems to matter, and the authors attribute the losses to hypohydration or to factors associated with it (Savoie et al., 2015). Trained people tended to lose 3.3% less than untrained people, a difference just short of significance at P = 0.06 (Savoie et al., 2015). In their 2014 review, Cheuvront and Kenefick, two of the meta-analysis authors, saw no clear threshold or mechanism for losses in strength and power (Cheuvront and Kenefick, 2014).

Attention and mood

A 2005 review claimed that mild dehydration already harms short-term memory, alertness and concentration (Ritz and Berrut, 2005). Later work is more cautious. A 2014 review found that studies of mild dehydration had not produced an effect that could be repeated on cognition, although self-reported mood changes such as feeling less alert or more tired were more consistent (Masento et al., 2014).

Pooling thirty-three studies with 413 participants, another meta-analysis found a small but significant overall impairment, with an effect size of −0.21 (Wittbrodt and Millard-Stafford, 2018). The effect was larger when losses exceeded 2% of body mass, and tasks of attention, motor coordination and executive control suffered most (Wittbrodt and Millard-Stafford, 2018).

Heat stress can impair thinking more than dehydration does, which confounds many of these studies (Masento et al., 2014). NATA describes most cognitive impairments as clinically negligible (McDermott et al., 2017).

Measure

Study group

Result

Urine colour 5 or darker

69 samples from 22 men after exercise in heat

88.9% sensitivity for a loss of 2% or more

Core temperature

Summary of laboratory studies

Up about 0.15–0.20°C per 1% lost

Heart rate

Summary of laboratory studies

Up 3 to 5 heartbeats a minute per 1% lost

Endurance in time trials paced by the rider

Subset of a fifteen-study meta-analysis with 122 subjects overall

No measurable change

Muscle strength

Pooled studies

Down 5.5%

Cognition

413 subjects in thirty-three studies

Small impairment, effect size −0.21

Low blood sodium

488 marathon runners

Thirteen percent

Note: the rows come from different designs, mostly small laboratory studies (McKenzie et al., 2015; Goulet, 2013; Wittbrodt and Millard-Stafford, 2018). They show average effects in study groups and are not targets or limits for any individual.

The opposite risk of drinking too much

Drinking well beyond what is lost can lower blood sodium, a condition called exercise-associated hyponatraemia when it develops in the 24 h after activity or during it (Hew-Butler et al., 2015). Most laboratories set the threshold at a sodium level below 135 mmol/L (Hew-Butler et al., 2015). Usually the underlying problem is an excess of body water relative to its normal level (Hew-Butler et al., 2015).

At the Boston Marathon, 488 of 766 enrolled runners provided a blood sample at the finish that could be analysed (Almond et al., 2005). Of these, thirteen percent had hyponatraemia, defined in this study as 135 mmol per litre or less, and 0.6% had critical levels of 120 mmol per litre or less (Almond et al., 2005). Weight gain during the race carried an odds ratio of 4.2, and finishing in over 4 hours rather than under 3:30 an odds ratio of 7.4 (Almond et al., 2005).

Female sex was linked to the condition in the unadjusted analysis but not after adjustment, and the type of fluid drunk showed no link (Almond et al., 2005). A 2015 consensus of 17 experts names sustained drinking beyond losses as the biggest single risk factor, and almost all symptomatic cases occurred in people who gained or kept their weight (Hew-Butler et al., 2015). Weight loss does not rule it out (Hew-Butler et al., 2015).

Symptomatic cases are rare, at 1% against 6% for cases without symptoms in one pooled analysis, yet it has been linked to no fewer than 14 deaths among athletes since 1981 (Hew-Butler et al., 2015). Symptoms range from lightheadedness and nausea to headache, vomiting, confusion and seizures (Hew-Butler et al., 2015). Many overlap with dehydration symptoms, which makes the two conditions hard to tell apart by symptoms alone (McDermott et al., 2017).

Rates vary with the event. In 161 km ultramarathons, cases without symptoms have ranged from 5% to 51% of those tested after the race (Hew-Butler et al., 2015). NATA puts the share of distance athletes found with it after events at 10% to 20%, while noting that it is uncommon in team sports (McDermott et al., 2017). Marathon researchers call it an important cause of deaths and dangerous illness linked to races (Almond et al., 2005).

What is established about dehydration during exercise and what remains uncertain

Some points are settled. Losing water through sweat raises core temperature and heart rate step by step as the deficit grows (Montain and Coyle, 1992; McDermott et al., 2017). Single symptoms and simple tests are poor guides to dehydration in older people (Hooper et al., 2015).

In older people, no single test identified dehydration consistently in more than one study, while in a laboratory trial with young men urine colour detected a 2% loss with 88.9% sensitivity (Hooper et al., 2015; McKenzie et al., 2015).

Other questions remain open. Whether a 2% loss harms endurance depends on how performance is tested (Goulet, 2013; McDermott et al., 2017). Effects on thinking are small and inconsistent across studies, and the field is described as young (Wittbrodt and Millard-Stafford, 2018; Masento et al., 2014).

Everyday health effects are unproven (Ritz and Berrut, 2005). Even the 2005 review that stressed cognitive harm conceded that wider effects of hydration on daily health and well-being are still unproven, mainly because suitable tools for measuring hydration are lacking (Ritz and Berrut, 2005). Funding varies across this literature: the Cochrane review was unfunded, at least one author of the NATA statement consults for Gatorade, and CrossFit paid the travel, hotel and meals of the hyponatraemia panel (Hooper et al., 2015; McDermott et al., 2017; Hew-Butler et al., 2015). Moderate dehydration rarely threatens life in people who are otherwise well, according to the consensus panel (Hew-Butler et al., 2015).

Frequently asked questions

Is thirst a reliable guide to fluid needs during exercise?

Expert bodies lean different ways. NATA notes that thirst signals can be absent or ignored during exercise, and that drinking to thirst replaces about two-thirds of sweat losses (McDermott et al., 2017). The 2015 hyponatraemia consensus, by contrast, treats drinking beyond losses as the single most important risk factor for low blood sodium (Hew-Butler et al., 2015).

Does urine colour show how hydrated a person is?

Partly, and it depends on the setting. After exercise in heat, a colour of 5 or darker detected a loss of 2% or more with 88.9% sensitivity in 69 samples from 22 men (McKenzie et al., 2015). In older people, by contrast, urine colour was not a useful stand-alone test for dehydration (Hooper et al., 2015).

Are older adults more prone to dehydration?

Their thirst response is weaker. A 2001 review found that independent adults over 65 drink enough daily, but feel less thirsty and drink less under heat, exercise or fluid deprivation (Kenney and Chiu, 2001). They also restore lost fluid more slowly, and reduced kidney filtration makes water conservation less efficient (McDermott et al., 2017; Masento et al., 2014).

How is sweat loss measured in studies?

Sweat output can be gauged by weighing people before and after exercise, since rapid weight change in sport largely reflects fluid (ACSM, 2007; Hooper et al., 2015). Sweat rates and the salt content of sweat differ greatly between people (ACSM, 2007). Among adults during exercise, measured rates range from 0.5 to 4.0 L per hour (McDermott et al., 2017).

Can a person be dehydrated without feeling thirsty?

Yes. During exercise, thirst signals can be absent or ignored, which is called voluntary dehydration (McDermott et al., 2017). In older people, feeling thirsty was not a useful test for dehydration, and the thirst response weakens with age (Hooper et al., 2015). Children, by contrast, rarely show voluntary dehydration in activities lasting 45 minutes or less (Kenney and Chiu, 2001).

Sources

  1. Cheuvront SN, Kenefick RW. Dehydration: physiology, assessment, and performance effects. Comprehensive Physiology. 2014;4(1):257–285. DOI 10.1002/cphy.c130017.
  2. Ritz P, Berrut G. The importance of good hydration for day-to-day health. Nutrition Reviews. 2005;63(6 Pt 2):S6–S13. DOI 10.1111/j.1753-4887.2005.tb00155.x.
  3. Masento NA, Golightly M, Field DT, Butler LT, van Reekum CM. Effects of hydration status on cognitive performance and mood. British Journal of Nutrition. 2014;111(10):1841–1852. DOI 10.1017/S0007114513004455.
  4. Hooper L, Abdelhamid A, Attreed NJ, et al. Clinical symptoms, signs and tests for identification of impending and current water-loss dehydration in older people. Cochrane Database of Systematic Reviews. 2015;(4):CD009647. DOI 10.1002/14651858.CD009647.pub2.
  5. McKenzie AL, Muñoz CX, Armstrong LE. Accuracy of urine color to detect equal to or greater than 2% body mass loss in men. Journal of Athletic Training. 2015;50(12):1306–1309. DOI 10.4085/1062-6050-51.1.03.
  6. McDermott BP, Anderson SA, Armstrong LE, Casa DJ, Cheuvront SN, Cooper L, Kenney WL, O'Connor FG, Roberts WO. National Athletic Trainers' Association position statement: fluid replacement for the physically active. Journal of Athletic Training. 2017;52(9):877–895. DOI 10.4085/1062-6050-52.9.02.
  7. ACSM (American College of Sports Medicine); Sawka MN, Burke LM, Eichner ER, Maughan RJ, Montain SJ, Stachenfeld NS. American College of Sports Medicine position stand. Exercise and fluid replacement. Medicine & Science in Sports & Exercise. 2007;39(2):377–390. DOI 10.1249/mss.0b013e31802ca597.
  8. Montain SJ, Coyle EF. Influence of graded dehydration on hyperthermia and cardiovascular drift during exercise. Journal of Applied Physiology. 1992;73(4):1340–1350. DOI 10.1152/jappl.1992.73.4.1340.
  9. Goulet EDB. Effect of exercise-induced dehydration on endurance performance: evaluating the impact of exercise protocols on outcomes using a meta-analytic procedure. British Journal of Sports Medicine. 2013;47(11):679–686. DOI 10.1136/bjsports-2012-090958.
  10. Savoie FA, Kenefick RW, Ely BR, Cheuvront SN, Goulet EDB. Effect of hypohydration on muscle endurance, strength, anaerobic power and capacity and vertical jumping ability: a meta-analysis. Sports Medicine. 2015;45(8):1207–1227. DOI 10.1007/s40279-015-0349-0.
  11. Wittbrodt MT, Millard-Stafford M. Dehydration impairs cognitive performance: a meta-analysis. Medicine & Science in Sports & Exercise. 2018;50(11):2360–2368. DOI 10.1249/MSS.0000000000001682.
  12. Almond CSD, Shin AY, Fortescue EB, Mannix RC, Wypij D, Binstadt BA, Duncan CN, Olson DP, Salerno AE, Newburger JW, Greenes DS. Hyponatremia among runners in the Boston Marathon. New England Journal of Medicine. 2005;352(15):1550–1556. DOI 10.1056/NEJMoa043901.
  13. Hew-Butler T, Rosner MH, Fowkes-Godek S, Dugas JP, Hoffman MD, Lewis DP, Maughan RJ, Miller KC, Montain SJ, Rehrer NJ, Roberts WO, Rogers IR, Siegel AJ, Stuempfle KJ, Winger JM, Verbalis JG. Statement of the 3rd International Exercise-Associated Hyponatremia Consensus Development Conference, Carlsbad, California, 2015. British Journal of Sports Medicine. 2015;49(22):1432–1446. DOI 10.1136/bjsports-2015-095004.
  14. Kenney WL, Chiu P. Influence of age on thirst and fluid intake. Medicine & Science in Sports & Exercise. 2001;33(9):1524–1532. DOI 10.1097/00005768-200109000-00016.
Mara Kenwick

Mara Kenwick — author

Mara Kenwick is a health science writer with academic training in biomedical science and public health. She completed a BSc in Biomedical Science before an MPH focused on epidemiology, prevention and the interpretation of population-health evidence. Her early...

This article is for general information and is not medical advice. It was reviewed for accuracy by a qualified clinician; decisions about your health should be made with your own doctor.

Related