Hostname: page-component-758b78586c-wkjwp Total loading time: 0 Render date: 2023-11-29T11:00:38.375Z Has data issue: false Feature Flags: { "corePageComponentGetUserInfoFromSharedSession": true, "coreDisableEcommerce": false, "useRatesEcommerce": true } hasContentIssue false

Assessment of hydration status in a large population

Published online by Cambridge University Press:  24 November 2014

Stephanie Baron
Department of Physiology, European Hospital Georges-Pompidou, AP-HP, Paris Descartes University, Paris, France
Marie Courbebaisse
Department of Physiology, European Hospital Georges-Pompidou, AP-HP, Paris Descartes University, Paris, France Inserm U845, Growth and Signalling Research Center, Paris Descartes University, 156 rue de Vaugirard, 75015Paris, France
Eve M. Lepicard
Institute for European Expertise in Physiology, 21 Rue Leblanc, 75015Paris, France
Gerard Friedlander*
Department of Physiology, European Hospital Georges-Pompidou, AP-HP, Paris Descartes University, Paris, France Inserm U845, Growth and Signalling Research Center, Paris Descartes University, 156 rue de Vaugirard, 75015Paris, France
*Corresponding author: Professor G. Friedlander, fax +331 43 06 04 43, email
Rights & Permissions [Opens in a new window]


Both acute and chronic dehydration can have important implications for human behaviour and health. Young children, non-autonomous individuals and the elderly are at a greater risk of dehydration. Mild hypertonic dehydration could be related to less efficient cognitive and physical performance and has been reported to be associated with frequently occurring pathological conditions, especially nephrolithiasis. The assessment of hydration status in a large sample appears to be of interest for conducting epidemiological and large clinical studies aimed at improving preventive and curative care. Especially in large-population studies, methods that are used have to be accurate, cheap, quick and require no technical expertise. Body weight change is widely used to determine acute hydration changes, but seems to be insufficiently accurate in longitudinal studies. Bioimpedance analysis methods enable the assessment of total body water content, but their use is still under debate. Because plasma osmolality directly reflects intracellular osmolality, it constitutes a good marker to assess acute hydration changes, but not chronic hydration status because it changes constantly. Moreover, venepuncture is considered to be invasive and is not suitable for a large-sample study, especially in children. Urinary markers appear to be good alternatives for assessing hydration status in large populations. Collection of urine samples is non-invasive and cheap. High technical expertise is not required to perform urinary marker measurements and these measurements can be carried out quickly. Thus, methods based on urinary markers are very well suited for field studies. Urine colour is probably the least sensitive marker despite its high specificity. Urine osmolality and especially urine specific gravity could be easily used for determining hydration status in large-sample studies.

Full Papers
Copyright © The Authors 2014 

Both acute and chronic dehydration (body water deficit) can have important implications for human behaviour and health( Reference Adrogue and Madias 1 , Reference Campbell 2 ). Mild dehydration is associated with altered cognitive performance and degraded mood( Reference Adan 3 , Reference Armstrong, Ganio and Casa 4 ) and also with impaired physical performance( Reference Sawka, Burke and Eichner 5 , Reference Murray 6 ). Mild dehydration could thus be associated with less efficient knowledge acquisition, especially during infancy and childhood, and also with less efficient professional activity.

Young children, non-autonomous individuals and the elderly are at a greater risk of dehydration, notably because they do not always have open and easy access to water and because their perception of thirst is neglected( Reference Anonymous 7 ) or altered( Reference Rolls and Phillips 8 Reference Jequier and Constant 10 ). Although nephrolithiasis is the only disorder that has consistently been found to be associated with chronic low daily water intake( Reference Armstrong 11 ), many other frequently occurring pathological conditions, such as constipation, asthma, CVD and chronic kidney diseases, could also be linked to insufficient fluid intake( Reference Popkin, D'Anci and Rosenberg 12 ).

High medical costs, morbidity and mortality can thus result from dehydration, so this condition should be taken into account in the field of public health( Reference Black, Morris and Bryce 13 , Reference Warren, Bacon and Harris 14 ). In the near future, epidemiological and interventional clinical trials will be needed to assess the impact of dehydration in a large sample. The lack of consistency in the evidence concerning hydration status and fluid intake requirements published to date is mainly due to the different methodologies used and also due to the complex and dynamic human fluid–electrolyte regulatory system that defies description as it changes constantly. That is why an attempt should be made to standardise methods for future studies. There is currently no consensus on a ‘gold standard’ for hydration status markers, particularly for mild dehydration. This indicates the need to define the best so-called field method to assess hydration status in a population of supposedly healthy people or patients.

To this end, the choice of accurate, easy-to-perform and non-expensive methods is fundamental. Although methods for assessing hydration status have been defined in previous reviews( Reference Armstrong 15 Reference Shirreffs 17 ), the best methods suitable for assessing hydration status in a large sample have not been precisely discussed to date.

In this review, we focus on the need to study hydration status in a large population and on methods available to assess hypertonic dehydration status. We principally give an overview of methods applicable to a large sample.

Definition and regulation of hydration status

The balance between water outputs and water inputs defines hydration status. Excess loss of water or insufficient intake of water induces a state of dehydration. This dehydration is hypertonic when water loss exceeds electrolyte loss, leading to a higher blood electrolyte concentration and thus to an increased plasma osmolality. To equilibrate osmolality between the intracellular and extracellular compartments, an obligatory increase in plasma osmolality inducing a shift in water from the intracellular to the extracellular compartment occurs. An increase in plasma osmolality thus implies intracellular dehydration( Reference Lang, Busch and Ritter 18 Reference Cheuvront, Kenefick and Charkoudian 20 ).

Water is mainly lost via kidney excretion and sweating. Other routes of loss are the respiratory tract and the faeces. Water excretion via the kidney removes solutes from the blood, and a minimum obligatory urine volume is required to remove the solute load. Obligatory urine volume is defined as the water volume necessary to excrete 24 h urine solutes at the age-related lower limit of maximum urine osmolality. The lower limit of maximum urine osmolality in individuals living in industrialised countries has been estimated to be 830 mOsm/kg minus 3·4 mOsm/kg per year starting from the age of 20 years( Reference Manz and Wentz 16 ). In an 18–25°C environment, a healthy sedentary adult will have moderate water losses ranging from 1·8 to 3·0 litres/d( Reference Grandjean, Reimers and Buyckx 21 ). These water outputs must be counterbalanced by water intake to maintain a neutral hydration balance. Most of the water intake in humans is from pure water (about 61 % of the total daily water intake), while the water that they consume in the form of beverages or water present in foods represents less than 40 % of the total daily water intake( Reference Perrier, Rondeau and Poupin 22 ).

Only a small amount of water (250 ml/d) is produced by metabolism in humans. Because water balance is highly dependent on dietary intakes and nutrient availability, body water balance is highly regulated. A loss of 1 % of body water is usually compensated within 24 h. Both water intake and renal water losses are controlled to achieve water balance, while sweating and expiration of regulatory vapours are not regulated. Minute changes in plasma osmolality are the main factors that stimulate the two main homeostatic mechanisms: release of the antidiuretic hormone arginine vasopressin (AVP) and thirst. AVP is synthesised in the supraoptic and paraventricular nuclei of the hypothalamus and is released from the posterior pituitary( Reference Bourque 23 ) and controls renal water reabsorption. An increase in plasma osmolality immediately triggers the release of AVP, which in turn activates the reabsorption of water from urine by the kidney, the main effective regulator of water loss( Reference Yasui, Marples and Belusa 24 ). The discriminatory power of renal excretion measures for assessing dehydration status is thus always secondary to that of plasma osmolality changes( Reference Shoker 25 ). Following renal water reabsorption, urine osmolality increases, reflecting the concentration capacity of the kidney. Together with the release of AVP, an increase of 1 or 2 % in plasma osmolality elicits thirst and thus water intake.

Dehydration consequences and related disorders

Water is vital to life: when fluid deficit exceeds 8 %, death may occur( Reference Grandjean, Reimers and Buyckx 21 ). Before this extreme state occurs, dehydration is manifested as various signs and symptoms( Reference Popkin, D'Anci and Rosenberg 12 ). Altered cognitive performance( Reference Ritz and Berrut 26 Reference Cian, Barraud and Melin 28 ), degraded mood and headache symptoms( Reference Armstrong, Ganio and Casa 4 ) have been reported to be associated with dehydration in adults and children( Reference Bar-David, Urkin and Kozminsky 29 Reference Fadda, Rapinett and Grathwohl 31 ). A recent review has concluded that being dehydrated by just 2 % impairs performance in tasks that require attention, psychomotor and immediate memory skills, as well as assessment of the subjective state, whereas performance in long-term and working memory tasks and executive functions is better preserved( Reference Adan 3 ). However, as emphasised by Secher & Ritz( Reference Secher and Ritz 32 ), these data have been derived from a small number of children and are not generalisable to older adults (mean age about 60 years) to support a relationship between mild dehydration and cognitive function. Lastly, data are currently lacking in frail elderly and demented individuals.

The role of hydration in physical activity, particularly in athletes, is of considerable interest and is well described in the scientific literature( Reference Sawka, Burke and Eichner 5 , Reference Murray 6 ). Although environmental temperature and heat tolerance of individuals should be taken into account, physical performance is also affected by dehydration. The performance of adults in strength and power exercises is generally less affected when compared with endurance or repeated intense performance( Reference Maughan and Shirreffs 33 ). Rehydration can reverse deficits due to dehydration such as reduced endurance, increased fatigue, altered thermoregulatory capability, reduced motivation and increased perceived effort and can also reduce oxidative stress induced by exercise and dehydration( Reference Paik, Jeong and Jin 34 ). During exercise, children may be at a greater risk of involuntary dehydration than adults. Children may not recognise the need to replace lost fluids, and both children and coaches need to be given specific guidelines regarding fluid intake( Reference Anonymous 7 ).

Popkin et al. ( Reference Popkin, D'Anci and Rosenberg 12 ) suggested that the replacement of water with sugar-sweetened beverages, juice and milk is associated with a reduced energy intake. The literature concerning the effect of water intake on energy intake in children is very limited, but a German school intervention study with water has suggested that the effects of water on the overall energy intake of children are comparable to those in adults( Reference Muckelbauer, Libuda and Clausen 35 ).

It has been shown that cardiovascular function is impaired under dehydration conditions. Heart rate increases and blood pressure decreases more rapidly during dehydration. Rehydration improves cardiac function under 2 % body weight dehydration conditions( Reference Charkoudian, Halliwill and Morgan 36 , Reference Aufderheide, Lax and Goldberg 37 ).

Bar-David et al. ( Reference Bar-David, Urkin and Kozminsky 29 ) reported a higher rate of kidney stone formation in a hot environment. Moreover, there is strong evidence that urine dilution in stone formers contributes to the reduction of the average recurrence interval and also the recurrence rate( Reference Borghi, Meschi and Amato 38 ). A high intake of fluids, especially water, is still the most powerful and certainly the most economical means of preventing nephrolithiasis, and it is often not used to advantage by stone formers( Reference Borghi, Meschi and Schianchi 39 ).

Dehydration may also affect kidney function as high fluid intake is associated with a lower risk of chronic kidney diseases( Reference Strippoli, Craig and Rochtchina 40 ) or with a slower decline in kidney function( Reference Clark, Sontrop and Macnab 41 ). Interestingly, in the last study, this association was found to persist even after adjustment for age, sex, baseline estimated glomerular filtration rate, use of medications for hypertension (including diuretics), proteinuria, diabetes and CVD.

Less strong evidence links good hydration status to a reduced incidence of constipation, exercise asthma and hyperglycaemia in individuals with diabetic ketoacidosis. Good hydration status is associated with a reduction in the risk of urinary tract infections, hypertension, fatal CHD and venous thromboembolism, but this needs to be confirmed by clinical trials. For other conditions such as bladder and colon cancers, evidence for a preventive effect of maintaining good hydration status is not consistent( Reference Popkin, D'Anci and Rosenberg 12 ). It has recently been reported that fluid intake of more than 2000 ml/d might be a protective factor in secondary stroke prevention( Reference Mucke, Grotemeyer and Stahlhut 42 ).

In a recent review, Armstrong( Reference Armstrong 11 ) has reported that urolithiasis is the only disorder that has consistently been associated with chronic low daily water intake, whereas evidence suggests that in conditions such as obesity and type 2 diabetes, increased water intake may reduce energy intake in some individuals.

The elderly (aged >65 years) are particularly at an increased risk of dehydration because they exhibit a decrease in thirst sensation and at the same time have an impaired kidney capacity to concentrate urine( Reference Rolls and Phillips 8 Reference Jequier and Constant 10 ). However, in a recent German population-based observational study, the median total water intake was found to decrease with an increase in age in only males( Reference Manz, Johner and Wentz 43 ). Obligatory urine volume was found to increase in both sexes due to a decreased concentration capacity of the kidney. The latter was balanced by a decrease in non-renal water losses, leaving the free water reserve and therefore hydration status almost unchanged, showing that total water intake requirements do not change with age, although ageing affects several parameters of water metabolism. Reduced sweat loss with increasing age appears to be primarily responsible for this observation( Reference Manz, Johner and Wentz 43 ). We must keep in mind that this study was conducted under conditions of free water access in the elderly, which is not the case in those who are institutionalised. Children are dependent on adults for access to water, and a larger surface area:volume ratio makes them more susceptible to changes in skin temperatures, linked to ambient temperature shifts( Reference Falk, Bar-Or and MacDougall 44 , Reference Smolander, Bar-Or and Korhonen 45 ).

Determination of daily fluid intake requirements

The scientific and medical communities have made recommendations regarding daily water intake to fulfil water requirements in infants, children and adults of both sexes. These recommendations are not based on minimal intake as a lot of factors can lead to an increased water output and a negative water balance. Indeed, environmental temperature, altitude, humidity level, physical activity and diet can affect water requirements( 46 , 47 ). Calculations of the recommended water intake made by the European Food Safety Authority (EFSA) are based on the ideal urine osmolality of 500 mOsm/kg to provide a safe margin of a ‘free water reserve’( 46 ). The majority of experts recommend a daily fluid intake of more than 2 litres/d in stone formers to maintain a diuretic state of at least 2 litres/d to optimise urine dilution( Reference Meschi, Nouvenne and Borghi 48 ).

With regard to the impact of seasonal variations on hydration status, it has been reported that dehydration secondary to a heat wave is potentially very harmful, particularly in some susceptible subpopulations, such as the elderly( Reference Kettaneh, Fardet and Mario 49 , Reference Yang, Liu and Ou 50 ), and in fragile patients, i.e. patients on antipsychotics( Reference Fitzgerald, Middleton and Cooper 51 ) and patients suffering from cystic fibrosis( Reference Desmazes-Dufeu, Hubert and Burgel 52 ). However, it has been shown that hydration status assessed by central venous pressure in patients with heat stroke could be normal, indicating that rapid intravenous rehydration should be avoided to prevent overload problems( Reference Seraj, Channa and al Harthi 53 ). Moreover, it is now well known that athletes who train in hot weather are hypohydrated while drinking ad libitum during practice because water intake is not sufficient to replace their sweat loss( Reference Arnaoutis, Kavouras and Kotsis 54 , Reference Shirreffs, Aragon-Vargas and Chamorro 55 ). This is also the case in manual workers working under extremely hot conditions( Reference Biggs, Paterson and Maunder 56 , Reference Bates, Miller and Joubert 57 ). A single study has been conducted in 547 children living in the Mediterranean region to evaluate the effects of seasonal changes in the climate on urine specific gravity and blood pressure. Surprisingly, seasonal changes in Mediterranean climate did not lead to changes in the hydration status of the children, suggesting that the decrease in blood pressure observed during summer should not be attributed to the hydration status( Reference Polat, Akil and Yuksel 58 ).

Studies carried out in diverse populations have shown that daily water intake requirements are not met in children( Reference Decher, Casa and Yeargin 59 , Reference Kaushik, Mullee and Bryant 60 ). Stookey( Reference Stookey 61 ) showed that adults are also sensitive to dehydration. In this study, based on plasma tonicity measured, 60 % of the 14 855 American community-dwelling adults (aged 20–90 years) giving blood for the Third National Health and Nutrition Examination Survey were found to have hypertonic plasma. Besides problems related to disease and the fact that elderly people have a reduced thirst sensation that affects water balance regulation, this study demonstrated that the recommended daily water intake is not met in adults, just as in children.

In a study carried out in eighty-four subjects aged 81–86 years, the mean fluid intake from drinks was found to be 950 ml for women and 1330 ml for men( Reference Ferry, Hininger-Favier and Sidobre 62 ). Only 45 % of the women and 35 % of the men drank at least 1000 ml of drinks a day, with 1000 ml of daily fluid intake from only drinks being the French recommended dietary intake for the general population. Actually, the scientific basis for water intake recommendations for the elderly is scarce, and the recommendations made by different nutrition societies are not consistent.

The high recurrence rate in stone formers strongly suggests that their daily fluid intake is insufficient: up to 85 % of all stone patients could be at a lower risk of stone recurrence with elementary reorientation of their lifestyle and dietary habits, the most important being higher fluid intake( Reference Straub and Hautmann 63 ). Patients’ compliance with this very simple preventive measure could be improved by autoevaluation of their hydration status.

Need to define the best method to assess hydration status in a large population

Dehydration may have a potential economical and sociological impact in terms of cognitive and physical performance. Mild dehydration could thus be related to less efficient knowledge acquisition, especially during infancy and childhood, altered professional activity and more frequent work stoppages. Dehydration could also have a high impact in the field of public health as it appears to be a risk factor for highly prevalent pathological conditions, such as nephrolithiasis, which is a disease affecting about 5–10 % of the population in industrialised countries worldwide with high clinical and economical costs( Reference Bartoletti, Cai and Mondaini 64 ). Lastly, ageing( Reference Ferrucci, Giallauria and Guralnik 65 ) and obesity epidemic in developed countries( Reference Forrester 66 ) highlight the need to study the potential impact of hydration status and of water intake on morbidity and mortality in the elderly and on energy intake.

Due to all these reasons, assessing hydration status in a large sample appears to be of great interest for conducting epidemiological and large clinical studies aimed at improving preventive medicine and also medical supervision in patients, especially stone formers, and in elderly institutionalised people. This may have some important implications for those responsible for forward planning in health care facilities.

‘Field’ methods for assessing hydration status: advantages and disadvantages

A ‘field method’ should be able to be performed in a large sample in everyday-life conditions, while remaining reliable enough to give access to scientifically useful data. It should consequently be ideally non-invasive, acceptable for the majority of people, cheap, easy to perform (not time consuming, not requiring high technical expertise and with a very few pre-analytical requirements), reproducible, sensitive enough and without a large inter-individual variability within a given population. In the following sections, we focus on the methods available for a large-sample study. A summary of the methods available is given in Table 1.

Table 1 Advantages and disadvantages of methods available for assessing hydration status

* Data from Cheuvront et al. ( Reference Cheuvront, Ely and Kenefick 69 ).

Assessment of body water

Body weight change

Determination of body weight change is probably the simplest method for assessing water loss during physical exercise for a short period of time. Total body water (TBW) content corresponds to about 60 % of body weight( Reference Cheuvront, Kenefick and Charkoudian 20 ); thereby, acute changes in body water content can be assessed by determining body weight change. Moreover, this method can be performed easily as it is quick and does not require technical expertise. It is commonly assumed that during physical exercise, body weight loss essentially equals the water loss occurring due to sweating. No other body component is lost at such a rate( Reference Shirreffs 17 ). Harvey et al. ( Reference Harvey, Meir and Brooks 67 ) have recently found that during a match the body weight of nine football players varies in correlation with other indices of hydration status such as urine specific gravity and urine colour. This is a sensitive method that can detect acute changes in hydration status such as moderate fluid losses of between 2 and 3 % of body weight( Reference Armstrong, Costill and Fink 68 ). Moreover, body weight change is commonly used to evaluate the severity of dehydration. Clinical symptoms depend on the severity of dehydration and the tolerance of individuals. Among these symptoms, the more common are an increased heart rate, a lengthening of capillary refill and a decreased systolic blood pressure. Even though no consensual definitions of acute and chronic dehydration exist, these two phenomena are very different. Acute dehydration mainly results from excess water loss due to pathological conditions such as diarrhoea or physical exercise, leading mostly to moderate-to-severe dehydration. On the other hand, chronic dehydration seems to be mainly linked to a lack of water intake as observed in the elderly. This kind of dehydration is often less serious and clinically more difficult to diagnose.

In physiological conditions, intra-individual variations in body weight rarely exceed 1·1 %( Reference Cheuvront, Ely and Kenefick 69 ). Nevertheless, because of the significant inter-individual variations (26·6 %)( Reference Cheuvront, Ely and Kenefick 69 ) due to body composition change (e.g. fat mass, muscular mass, sex and age), a personal precise baseline is absolutely fundamental, but not always available. Moreover, changes in body composition, independently of hydration status changes, make this parameter unusable in studies of long duration. That is why this parameter is mainly used in acute experimental settings, based on sport activity or intense exercising, which greatly differ from free-living conditions from a physiological point of view. In addition, this parameter is relevant for one measurement at a given time point (e.g. after exercising), but cannot reflect hydration status during longer time periods (e.g. 24 h) as food ingestion, fluid intakes, faecal losses and urine production also affect body weight. Due to these reasons, determination of body weight change is not a suitable method for assessing hydration status in free-living condition sample studies.

Isotope dilution methods

The principle of isotope dilution methods is based on the distribution of a tracer substance after oral or intravenous administration( Reference Armstrong 70 ). These methods enable the measurement of TBW content, thanks to a tracer that gets distributed in all body fluid compartments. In this case, the most common tracers used are the stable isotopes of hydrogen and oxygen such as D2O( Reference Westerterp, Wouters and van Marken Lichtenbelt 71 ) and 3H2O( Reference Finn, Plank and Clark 72 ). Briefly, many hours after the administration of a precise quantity of a tracer, the tracer concentration achieved after equilibrium is measured in the plasma and/or urine. Its concentration allows to determine TBW content( Reference Armstrong 70 ). In the same way, tracers that get distributed only in extracellular compartments are used to measure extracellular water (ECW) content. Na, Cl and especially Br( Reference Lukaski 73 ) isotopes are used in this case. The difference between TBW and ECW contents yields intracellular water (ICW) content( Reference Armstrong, Kenefick and Castellani 74 ). Even though these methods are accurate, they cannot be used in large populations because of the significant technical conditions that are required.

Bioelectrical impedance analysis

Another method that can be used for assessing body composition, especially water content, is bioelectrical impedance analysis (BIA). Its principles have been widely described by Kyle et al. ( Reference Kyle, Bosaeus and De Lorenzo 75 , Reference Kyle, Bosaeus and De Lorenzo 76 ). BIA is carried out based on the electrical properties of tissues. Indeed, tissues conduct electrical current differently depending on their water and electrolyte contents. Taking this property into account, equations have been developed to link body resistance to the electrical current of TBW, ECW and ICW. Sex, weight or age is usually taken into account in these equations. The referent isotope dilution method has been used to determine these equations. In the study carried out by Gudivaka et al. ( Reference Gudivaka, Schoeller and Kushner 77 ) the reference values obtained for TBW content were 44·2 (sd 6·3) kg for men (n 14) and 30·6 (sd 3·8) kg for women (n 13); for ECW content they were 15·7 (sd 3·2) kg for men and 12·2 (sd 1·8) kg for women; and for ICW content they were 28·5 (sd 3·7) kg for men and 18·4 (sd 2·5) kg for women.

Several bioimpedance methods have been developed since the 1970s. The first one is the single-frequency BIA (SF-BIA). In this method, a 50 kHz current is passed through the body through the electrodes placed on the hand and the ankle generally. This method enables to measure the sum of ICW and ECW contents, but does not allow determining TBW content. This kind of method has not yet been validated for use in altered hydration conditions( Reference Gudivaka, Schoeller and Kushner 77 ). The multiple-frequency BIA (MF-BIA) was developed in the 1990s to improve sensitivity and accuracy. In this method, electrical currents ranging from 0 to 500 kHz are used to evaluate TBW, ICW and ECW contents( Reference Pialoux, Mischler and Mounier 78 , Reference Piccoli 79 ). Nonetheless, frequencies below 5 kHz and above 200 kHz have poor conductivity reproducibility and their use should be avoided( Reference Hannan, Cowen and Fearon 80 ). Shanholtzer & Patterson( Reference Shanholtzer and Patterson 81 ) found this method to be reproducible with the same technicians taking measurements. Gudivaka et al. ( Reference Gudivaka, Schoeller and Kushner 77 ) evaluated the validity of the MF-BIA in twenty-eight adults and found the maximum standard error estimates for TBW, ECW and ICW contents to be 2·4, 1·4 and 3·5 kg, respectively. Many studies have been realised to compare SF-BIA and MF-BIA. Patel et al. ( Reference Patel, Peterson and Silverman 82 ) demonstrated that better results could be obtained when using the MF-BIA for the determination of ECW content and that the SF-BIA is ideal for the determination of TBW content in critically ill patients. In another study, it was found that changes in ECW and ICW contents in elderly patients could not be detected using the MF-BIA( Reference Olde Rikkert, Deurenberg and Jansen 83 ).

The third method is bioelectrical spectroscopy (BIS). The main difference between BIS and classical BIA (SF-BIA and MF-BIA) is the use of mathematical modelling and mixture equations (e.g. Cole–Cole plots) to determine ECW or ICW content instead of the classical BIA equations. This method has been shown to be accurate and have a low bias in a non-physiological population( Reference Patel, Matthie and Withers 84 ). Nevertheless, the authors do not agree with BIS variability results. Ward et al. ( Reference Ward, Elia and Cornish 85 ) demonstrated a wide biological variation in a control population. Also, there is a debate regarding accuracy results because some authors have demonstrated accuracy improvement( Reference Cox-Reijven and Soeters 86 Reference Earthman, Matthie and Reid 88 ) with mixture equations, while others did not( Reference Ho, Kushner and Schoeller 89 , Reference Baarends, van Marken Lichtenbelt and Wouters 90 ) and have even demonstrated worse accuracy( Reference Gudivaka, Schoeller and Kushner 77 ). Segmental BIA has also been developed, where two additional electrodes are placed on either side of the body to focus on well-defined body segments. Body segmentation is useful because it is less influenced by fat fraction or geometrical boundary conditions. Another method developed using BIA is the bioelectrical impedance vector analysis, which was developed by Piccoli et al. ( Reference Piccoli, Pillon and Dumler 91 Reference Piccoli, Pittoni and Facco 93 ). In this method, results do not depend on equations or modelling and so the variability depends only on analytical errors and biological variations. Few clinical studies have been conducted using this method. Buffa et al. ( Reference Buffa, Mereu and Putzu 94 , Reference Buffa, Saragat and Cabras 95 ) demonstrated promising results, while Cox-Reijven et al. ( Reference Cox-Reijven, van Kreel and Soeters 96 ) demonstrated a low sensitivity, but a high specificity in detecting depletion.

The main advantage of using BIA methods is that they provide a rapid feedback. Moreover, these methods are relatively inexpensive, non-invasive and easy to perform( Reference Kyle, Bosaeus and De Lorenzo 75 ). Nonetheless, such methods exhibit a significant variability. The most important parameter is the choice of the equation or modelling used in the SF-BIA, MF-BIA or even in BIS leading to significant inter-BIA variations. Kyle et al. ( Reference Kyle, Bosaeus and De Lorenzo 76 ) reported more than twenty different equations for determining TBW content using BIA leading to various standard error estimates ranging from 0·88 to 3·8 litres when compared with a reference measure obtained with 2H2O or 18O. With regard to ECW content, Kyle et al. ( Reference Kyle, Bosaeus and De Lorenzo 76 ) reported standard error estimates ranging from 0·98 to 2·2 litres when compared with the results obtained using the isotope dilution method in a review of twenty studies. With regard to ICW content, two studies have reported standard error estimates of 0·9 litres in an elderly population( Reference Dittmar and Reber 97 ) and 1·9 litres in healthy men( Reference De Lorenzo, Candeloro and Andreoli 98 ). The chosen equation must be most relevant for the population studied, depending on ethnic group, age (elderly, adult or child), body shape abnormalities or fat mass distribution. For example, Cox-Reijven et al. ( Reference Cox-Reijven and Soeters 86 ) found BIS to lack sensitivity in an overweight population. Moreover, the reproducibility of BIA measurements depends on many factors. The change in electrode position has been underlined by Sinning & Morgan( Reference Sinning and Morgan 99 ). Roos et al. ( Reference Roos, Westendorp and Frolich 100 ) and O'Brien et al. ( Reference O'Brien, Young and Sawka 101 ) also highlighted the change in electrolyte composition as a significant cause of variations in BIA measurements. Changes in skin and ambient temperatures are also responsible for variations in BIA measurements( Reference Gudivaka, Schoeller and Kushner 102 ). The development of improved BIA methods taking these factors into account is of great interest( Reference Huang, Cheng and Peng 103 ). The placement of electrodes is important and could lead to variations in BIA measurements( Reference Birkemose, Møller and Madsen 104 ). Lastly, standardised protocols that take into account all these parameters are essential for optimising BIA measurements( Reference Armstrong, Kenefick and Castellani 74 ).

Moreover, BIA seems to be insufficiently suitable for large-population studies in free-living conditions( Reference Mazariegos, Pithan and Meyer 105 ). Kyle et al. ( Reference Kyle, Bosaeus and De Lorenzo 76 ) recommended that this method be used in only stable conditions. Furthermore, O'Brien et al. ( Reference O'Brien, Young and Sawka 101 ) underlined the difficulty in distinguishing fluid volume and electrolyte changes during acute hydration changes. Finally, the use of more accurate markers seems to be essential for assessing hydration status in a large-sample study.

Plasma osmolality

Because plasma osmolality reflects intracellular osmolality, it has historically been considered to be a good marker standard for assessing hydration status( Reference Grant and Kubo 106 ), although some limitations have been underlined( Reference Armstrong 15 , Reference Perrier, Rondeau and Poupin 22 ). The most important regulated variable in the central nervous system to control human fluid–electrolyte balance is indeed extracellular osmolality. Consequently, plasma osmolality cannot validly represent chronic hypohydration as the brain is constantly affected by a change in plasma osmolality (i.e. moving it towards a set point or shifting it to exist within an acceptable range). Osmolality is measured using either a freezing-point depression osmometer or, more rarely, a vapour pressure-depression osmometer. Neuroendocrine regulation of plasma osmolality is such that normal values rarely deviate by more than 1–2 % from a basal value of 287 mOsm/kg in healthy, well-hydrated individuals. Intra-individual and inter-individual variations are indeed very low (1·3 and 1·5 %, respectively)( Reference Cheuvront, Ely and Kenefick 69 ). Moreover, its measure is highly reproducible with an analytical CV < 0·4 %( Reference Cheuvront, Ely and Kenefick 69 ). Because of this small deviation window, a cut-off of 290 mOsm/kg is commonly used to define the limit between euhydration and dehydration( Reference Kenefick, Ely and Cheuvront 107 Reference Cheuvront and Sawka 109 ).

When plasma osmolality measurement is not possible, in physiological conditions, it could be replaced with osmolality calculation. Indeed, plasma osmolality depends on plasma solute concentrations. Na is the most abundant electrolyte in ECW and is mainly responsible for plasma osmolality, in association with its matched anions, urea and glucose. The calculated osmolality is defined as follows: osmolality = 2 × [Na+]+[urea]+[glucose]. Apart from pathological conditions, such as hyperglycaemia (diabetes) and the terminal stage of chronic kidney disease with increased uraemia, Na concentration is highly correlated with plasma osmolality, and its measurement could be an alternative for plasma osmolality measurement( Reference Costa, Teixeira and Rama 110 ).

Although methods based on plasma osmolality are not expensive, plasma osmolality may not be the most suitable parameter for field studies because collecting blood samples is considered to be invasive for subjects. Due to this reason, this parameter is not suitable for a large-sample hydration assessment study, especially in children( Reference Oppliger and Bartok 111 ). Moreover, plasma osmolality is more or less relevant, depending on the context. In case of acute changes, especially during physical exercise, plasma osmolality has been described to change, while in a chronic dehydration context, such as in low drinkers or during progressive dehydration, e.g. in case of fluid deprivation, plasma osmolality is preserved, while only urinary indices change because of kidney adaptation( Reference Perrier, Vergne and Klein 112 , Reference Pross, Demazieres and Girard 113 ). Considering all these factors, the use of an indirect marker thereby seems essential for large-sample studies of long duration, and urinary markers appear to be good alternatives for assessing hydration status in such contexts.

Urinary indices

As has been described previously, the kidney is the main regulator of water loss in response to an elevation of plasma osmolality( Reference Jequier and Constant 10 ). During hypertonic dehydration conditions, AVP is secreted, leading to water reabsorption in the collecting duct, without electrolyte reabsorption. This mechanism leads to a decreased urine output with an increased urine concentration. It is essential for scientists to assess hydration status by measuring urine concentration. There are three urinary markers that are widely used for assessing urine concentration: urine osmolality; urine specific gravity; urine colour.

Urine osmolality

Urine osmolality is the concentration of osmotic solutes present in the urine. It is measured, as has been described previously for plasma osmolality, using a freezing-point or vapour pressure-depression osmometer. Urine osmolality depends on two parameters: the quantity of solutes and the volume of water. Regarding the quantity of Na, K and urea are the most abundant solutes in the urine. In physiological conditions, their amounts mainly depend on the diet, with daily osmole elimination in urine being closely related to daily osmole intake.

In a dehydrated healthy individual, a small volume of highly concentrated urine will be produced and will be reflected by an elevated urine osmolality, while in an individual with a high fluid intake, a large amount of urine will be produced, resulting in a low urine osmolality. Thus, urine osmolality reflects the capacity of the kidney to appropriately respond to variations in body water balance. Urine osmolality ranges from 50 to 1400 mOsm/kg. Few rare pathological conditions, such as diabetes insipidus, syndrome of inappropriate AVP secretion and preterminal stage of chronic kidney diseases, could disturb the concentration capacity of the kidney. In these rare cases, urinary indices cannot be used.

Overall, measuring urine osmolality has many advantages (Table 1). First, it is a non-invasive and cheap method that can be performed in large populations in everyday-life conditions. This method permits to detect the trend to dehydration easily because osmolality increases in parallel with hypertonic dehydration( Reference Armstrong, Maresh and Castellani 114 , Reference Oppliger, Magnes and Popowski 115 ). In addition, it is sensitive enough to detect small changes in the hydration status. For 1-unit variation in plasma osmolality, there is a 100-unit variation in urine osmolality showing a larger deviation window( Reference Manz and Wentz 16 ). For instance, Armstrong et al. ( Reference Armstrong, Maresh and Castellani 114 ) showed that urine osmolality reflects dehydration more accurately than blood indices. Moreover, among all the urinary markers, urine osmolality has the best sensitivity (91 %), which is almost equal to that of plasma osmolality (90 %)( Reference Cheuvront, Ely and Kenefick 69 ).

Intra-individual variation in urine osmolality is significant with a 28·3 % variation and even more for inter-individuals with a 57·9 % variation coefficient( Reference Cheuvront, Ely and Kenefick 69 ). Manz & Wentz( Reference Manz and Wentz 16 ) showed that the mean 24 h urine osmolality varies from 360 mOsm/kg in Poland to 860 mOsm/kg in Germany, mainly because of the cultural differences in dietary fluid and osmole intakes.

A large number of studies have shown that urinary osmolality increases in response to dehydration( Reference Armstrong, Soto and Hacker 116 Reference Kovacs, Senden and Brouns 120 ). Nevertheless, defining a cut-off value for euhydrated and dehydrated subjects is difficult. These authors have suggested the use of a population-specific cut-off value that would be equal to the mean maximal value minus 2 sd. In Europe, this cut-off value would be 830 mOsm/kg( Reference Manz and Wentz 16 ). Grant & Kubo( Reference Grant and Kubo 106 ) defined dehydration as a urine osmolality above 1000 mOsm/kg. In 1994, Amstrong et al. ( Reference Armstrong, Maresh and Castellani 114 ) defined dehydration as a urine osmolality exceeding 1052 mOsm/kg. Oppliger et al. ( Reference Oppliger, Magnes and Popowski 115 ) first set a dehydration cut-off value at 700 mOsm/kg for evaluating a hypohydrated group and then decided to set the cut-off value at 800 mOsm/kg to increase the correlation with the results obtained for plasma osmolality. Cleary et al. ( Reference Cleary, Hetzler and Wasson 121 ) used a 700 mOsm/l threshold; Peacock et al. ( Reference Peacock, Stokes and Thompson 122 ) used 900 mOsm/kg as the cut-off value.

To conclude, it is clear that no consensus has been reached regarding the dehydration cut-off value( Reference Armstrong, Pumerantz and Fiala 123 ). However, in line with the results reported by Manz & Wentz( Reference Manz and Wentz 16 , 46 ) and conclusions drawn by the EFSA, an osmolality over 800 mOsm/kg could be a relevant cut-off value to define the limit between a euhydrated and a slightly dehydrated status. This cut-off value is not generalisable because of the great variability between different kinds of populations according to their dietary habits.

Urine specific gravity

Urine specific gravity corresponds to the measure of urine density, defined as the weight of urine compared with that of an equal volume of distilled water. The specific gravity of plain water is equal to 1·000, whereas that of normal urine samples usually ranges from 1·013 to 1·029. To prevent weight loss by dehydration in weight category sport, the National Collegiate Athletic Association( 124 ) has decided that dehydration would be defined by a urine specific gravity value over 1·020–1·025. Armstrong et al. ( Reference Armstrong, Pumerantz and Fiala 123 ) reported that these limits reflect the upper range of a euhydrated state. This cut-off value is in accordance with the results of numerous studies exhibiting a real consensus state regarding urine specific gravity measurements( Reference Cleary, Hetzler and Wasson 121 , Reference Logan-Sprenger and Spriet 125 , Reference Osterberg, Horswill and Baker 126 ). In physiological conditions, intra-individual variation in urine specific gravity is effectively negligible with only a 0·4 % variation coefficient. Inter-individual variation is also very low with a 1·0 % variation coefficient, making the measurement very robust and reliable( Reference Cheuvront, Ely and Kenefick 69 ). Urine specific gravity is measured using a refractometer, which yields results immediately with low technical requirements. Numerous studies have shown that urine osmolality and urine specific gravity are strongly correlated, indicating that the measurements of both these parameters are consistent( Reference Armstrong, Maresh and Castellani 114 , Reference Oppliger, Magnes and Popowski 115 ).

A single gravity test strip could be used to determine urine specific gravity. The major advantage of this is that patients, especially stone formers, or volunteers can use it themselves. Its benefit has been underlined in old institutionalised people( Reference Wakefield, Mentes and Diggelmann 127 ). Although a German study analysing 340 first morning urine samples demonstrated a reasonably good correlation between refractometry and single test strip results( Reference Siegrist, Hess and Montandon 128 ) and a study analysing 174 urine samples demonstrated refractometry measurement to lack accuracy, refractometry measurement remains the ‘gold standard’ to define urine specific gravity( Reference Rowat, Smith and Graham 129 ).

Urine specific gravity measurement has one main disadvantage: both the number and size of the particles in the solution affect it. Indeed, urine specific gravity can vary when unusual quantities of larger molecules such as glucose, proteins and urea are present in the urine, generating falsely elevated values that suggest highly concentrated urine. This phenomenon also occurs during urine osmolality measurement where glucose and urea also have an osmotic effect.

Lastly, this method is considered to be as accurate as urine osmolality measurement( Reference Armstrong, Maresh and Castellani 114 , Reference Perrier, Demazieres and Girard 130 ), with the same specificity (91 %) and an almost equivalent sensitivity (89 %)( Reference Cheuvront, Ely and Kenefick 69 ). Urine specific gravity measurement could even present the advantage of a low inter-individual variability when compared with urine osmolality measurement. Urine specific gravity could thus be recommended to be used for assessing hydration status in large-population studies.

Urine colour

Urine colour is the third common urinary marker used for assessing hydration status. A urine colour chart has been developed to assess urine concentration in healthy humans( Reference Armstrong, Soto and Hacker 116 ). Briefly, this chart has a standardised colour scale ranging from 1 (pale yellow, corresponding to diluted urine) to 8 (dark brown, corresponding to concentrated urine). The general admitted value for a cut-off definition between euhydration and dehydration is mainly set at 4 units( Reference Cheuvront and Sawka 109 , Reference Cleary, Hetzler and Wasson 121 ).

This method has several advantages: it is cheap and non-invasive; it does not require technical expertise and gives immediate results. Moreover, this method has the best specificity (97 %) among all the methods based on urinary markers and the analytical variation is negligible( Reference Cheuvront, Ely and Kenefick 69 ). Armstrong et al. ( Reference Armstrong, Soto and Hacker 116 ) found that there is a linear relationship between urine colour, specific gravity and osmolality, showing that all these urinary markers are suitable for assessing hydration status. However, the main disadvantage of using urine colour chart is its lack of sensitivity (81 % with a 5·5 cut-off value)( Reference Cheuvront, Ely and Kenefick 69 ). Moreover, it can be affected by dietary factors, illness and medications( Reference Simerville, Maxted and Pahira 131 ), leading to significant intra-individual and inter-individual variability (30·9 and 47·4 %, respectively)( Reference Cheuvront, Ely and Kenefick 69 ).

In conclusion, urine colour seems to be less sensitive, but more specific than urine osmolality or specific gravity to assess hydration status and may not be the most suitable marker for large-sample studies.

Validity of urinary indices for assessing hydration status

Because urine is stored in the bladder before excretion, urine can be collected at different time points. Usually, urine is collected either in the morning before ingesting any food or fluid (fasting morning urine) or over a 24 h time period (24 h urine). Urinary indices of morning urine are not always correlated with those of 24 h urine samples( Reference Armstrong, Pumerantz and Fiala 123 ). Several studies have shown that morning urine is more concentrated than 24 h urine samples( Reference Grandjean, Reimers and Buyckx 21 , Reference Shirreffs and Maughan 132 ). Indeed, during night, there is a lack of fluid intake and accumulation of urine in the bladder. Collection of 24 h urine samples provides concentrated morning urine and diluted urine corresponding to the periods of rehydration during the day. First morning urine assessments give information about water balance at a single time point, while 24 h urine collection reflects the whole-day body water balance( Reference Armstrong, Johnson and McKenzie 133 ). In the absence of excessive extra renal water losses by sweating, hydration status (reflected by urine concentration) mainly depends on water intake so that daily repartition of water intake will greatly influence urine concentration. Thereby, urinary measurements should be interpreted relative to the type of urine collection performed.

If the bladder is properly voided before the water load, urine dilution, as judged from urine osmolality, may be observed as early as 30–60 min after a water load. It should be stressed that most studies addressing the impact of water intake on urine osmolality or specific gravity were performed under very specific conditions of acute dehydration elicited by physical exercise. Under these conditions, ingestion of less than 1·0 litres of hypotonic fluid was found to have only a limited effect on urine osmolality in the hour following ingestion( Reference Armstrong, Soto and Hacker 116 , Reference Popowski, Oppliger and Patrick Lambert 134 ). Kovacs et al. ( Reference Kovacs, Senden and Brouns 120 ) showed that nearly 3 h are required to normalise urine osmolality and colour after an acute 3 % dehydration period. In contrast, ingestion of a large amount of fluid within a short period of time during rehydration was found to induce a rapid increase in urinary output even when the subjects were dehydrated( Reference Kovacs, Senden and Brouns 120 ).

The majority of researchers consider 24 h urine collection as the gold standard for urinary hydration markers in daily life( Reference Armstrong, Johnson and McKenzie 133 ). Nevertheless, collecting 24 h urine samples is a heavy procedure that is difficult to perform in large-sample studies on hydration. Perrier et al. ( Reference Perrier, Demazieres and Girard 130 ) demonstrated that afternoon urine collection could be a good representative of 24 h urine collection and become a suitable alternative for 24 h urine collection. Moreover, the use of urine osmolality:urine creatinine ratio has been discussed for assessing hydration status and seems to be reproducible in individuals aged >5 years( Reference Godevithanage, Kanankearachchi and Dissanayake 135 ). Nevertheless, further studies would be needed to validate this marker and its correlation with other accurate hydration status markers.

To conclude, urinary indices allow to accurately assess hydration status during mild dehydration. Among the methods based on these indices, those based on urine specific gravity and colour are easy to be performed, while those based on urine osmolality require technical expertise. Collection of urine samples is non-invasive and cheap. High technical expertise is not required to perform these two measurements, and these measurements can be carried out quickly. These measurements are thus very well suited for field studies. However, these measurements may be less accurate in some situations such as during rehydration, isotonic dehydration (loss of water and Na at the same concentrations as in the plasma) and hypotonic dehydration (loss of Na). In spite of its specificity, urine colour is certainly the least sensitive urinary marker, but urine specific gravity, with good specificity and sensitivity, could easily be used in a large-sample study.

Saliva parameters

Similar to urine, saliva is another easily accessible fluid. Saliva flow rate is a very important parameter among the salivary parameters. In physiological unstimulated conditions, saliva flow rate has been evaluated to be about 0·46 (sd 0·2) ml/min and 0·32 (sd 0·2) ml/min, respectively( Reference Osailan, Pramanik and Shirlaw 136 , Reference Saeves, Reseland and Kvam 137 ). It has been shown that 24 h dehydration is associated with decreased saliva flow rates in a small sample of healthy young and older adults( Reference Ship and Fischer 138 ). During metabolic rehydration of these subjects, unstimulated saliva flow rate increased, but remained significantly lower than the baseline levels. Saliva osmolality is also an important parameter. Cheuvront et al. ( Reference Cheuvront, Ely and Kenefick 69 ) defined normal values of 71 (sd 15) mOsm/kg in an eighteen-person euhydrated population. A further investigation carried out by Walsh et al. ( Reference Walsh, Laing and Oliver 139 ) showed that during acute mild dehydration (3 % body weight change), saliva flow rate decreases (from about 0·5 to 0·2 ml/min), while saliva osmolality (from 50 to about 100 mOsm/kg) and total protein concentration (from 0·7 up to 1·8 mg/ml) increase. These variations are correlated with body weight change, urine osmolality and plasma osmolality( Reference Walsh, Laing and Oliver 139 ). In the study carried out by Pross et al. ( Reference Pross, Demazieres and Girard 113 ), following 24 h of fluid deprivation, saliva osmolality was found to increase slowly when compared with urine specific gravity and colour modifications, which indicates the lack of sensitivity of this parameter. Moreover, values of salivary parameters returned to baseline levels >1 h after the ingestion of a rehydration solution( Reference Walsh, Laing and Oliver 139 ), suggesting that the measurement of these parameters is more relevant for assessing dehydration status than for assessing hydration status during rehydration period. Moreover, Singh & Peters( Reference Singh and Peters 140 ) found a lack of correlation between saliva osmolality and urine osmolality and specific gravity in multiday events.

In addition, the large inter-individual variability highlighted by Ely et al. ( Reference Ely, Cheuvront and Kenefick 141 ) and Walsh et al. ( Reference Walsh, Laing and Oliver 139 ) and measured by Cheuvront et al. ( Reference Cheuvront, Ely and Kenefick 69 ) (35·8 % inter-individual variation coefficient) does not enable the assessment of hydration status in a large population as a baseline value should be determined for each subject to set his or her own euhydration reference value. Moreover, this baseline seems difficult to assess because of a significant intra-individual variability( Reference Cheuvront, Ely and Kenefick 69 ) (9·5 %), probably linked to the profound effect of oral intake( Reference Ely, Cheuvront and Kenefick 141 ). Saliva osmolality is the least sensitive (81 %) and specific (83 %) hydration status marker( Reference Cheuvront, Ely and Kenefick 69 ).

In conclusion, even though saliva osmolality measurement is non-invasive and cheap, it cannot be used in subjects with progressive dehydration in large-sample studies because of its lack of accuracy, its need for technical expertise, and its significant inter-individual and intra-individual variability.

Recently, tear osmolality has also been evocated as a hydration status marker( Reference Fortes, Diment and Di Felice 142 , Reference Santos, Batista and Guare 143 ), with a good correlation with plasma osmolality, but further studies are required to validate its utilisation.


To date, no ideal and consensual method has been developed to assess hydration status, especially in large-sample studies. Body weight change seems difficult to assess in such a context because of the necessity of a baseline value. Because plasma osmolality directly reflects intracellular osmolality, it constitutes a good marker to assess acute hydration changes, but cannot represent chronic hydration status because it changes constantly. Moreover, venepuncture is considered to be invasive for subjects and is not suitable for a large-sample hydration assessment study. Urine concentration reflects renal response to changes in plasma osmolality and is in most cases well correlated with plasma osmolality. Among the urinary markers, urine colour is probably the least sensitive marker. Urine osmolality and especially urine specific gravity could be used easily to assess hydration status in a large-sample study. Although 24 h urine collection is the gold standard to assess urine concentration, it is a demanding procedure that is difficult to use in large-sample studies. First morning urine or afternoon urinary spot samples can be used, with the former being easier to standardise and the latter being more representative of the whole-day water balance. Knowledge about the daily repartition of fluid intake is required to analyse urinary markers.

Understanding the advantages and limitations of using each hydration status marker is a key point to conducting large-sample studies concerning hydration status. These large studies will probably be of great interest in the near future in the field of preventive medicine.


The authors thank Dr Yann St George Chaumet for assistance in conducting literature searches.

S. B. and M. C. made an equal contribution in conducting the literature research and writing the manuscript. E. M. L. read and corrected the manuscript. G. F. gave scientific guidelines, and read and corrected the manuscript.

None of the authors has any conflicts of interest to declare.


1 Adrogue, HJ & Madias, NE (2000) Hypernatremia. N Engl J Med 342, 14931499.Google Scholar
2 Campbell, N (2011) Dehydration: why is it still a problem? Nurs Times 107, 1215.Google Scholar
3 Adan, A (2012) Cognitive performance and dehydration. J Am Coll Nutr 31, 7178.Google Scholar
4 Armstrong, LE, Ganio, MS, Casa, DJ, et al. (2012) Mild dehydration affects mood in healthy young women. J Nutr 142, 382388.Google Scholar
5 Sawka, MN, Burke, LM, Eichner, ER, et al. (2007) American College of Sports Medicine position stand. Exercise and fluid replacement. Med Sci Sports Exerc 39, 377390.Google Scholar
6 Murray, B (2007) Hydration and physical performance. J Am Coll Nutr 26, 542S548S.Google Scholar
7 Anonymous, (2000) Climatic heat stress and the exercising child and adolescent. American Academy of Pediatrics. Committee on Sports Medicine and Fitness. Pediatrics 106, 158159.Google Scholar
8 Rolls, BJ & Phillips, PA (1990) Aging and disturbances of thirst and fluid balance. Nutr Rev 48, 137144.Google Scholar
9 Kenney, WL & Chiu, P (2001) Influence of age on thirst and fluid intake. Med Sci Sports Exerc 33, 15241532.Google Scholar
10 Jequier, E & Constant, F (2010) Water as an essential nutrient: the physiological basis of hydration. Eur J Clin Nutr 64, 115123.Google Scholar
11 Armstrong, LE (2012) Challenges of linking chronic dehydration and fluid consumption to health outcomes. Nutr Rev 70, Suppl. 2, S121S127.Google Scholar
12 Popkin, BM, D'Anci, KE & Rosenberg, IH (2010) Water, hydration, and health. Nutr Rev 68, 439458.Google Scholar
13 Black, RE, Morris, SS & Bryce, J (2003) Where and why are 10 million children dying every year? Lancet 361, 22262234.Google Scholar
14 Warren, JL, Bacon, WE, Harris, T, et al. (1994) The burden and outcomes associated with dehydration among US elderly, 1991. Am J Public Health 84, 12651269.Google Scholar
15 Armstrong, LE (2007) Assessing hydration status: the elusive gold standard. J Am Coll Nutr 26, 575S584S.Google Scholar
16 Manz, F & Wentz, A (2003) 24-h hydration status: parameters, epidemiology and recommendations. Eur J Clin Nutr 57, Suppl. 2, S10S18.Google Scholar
17 Shirreffs, SM (2003) Markers of hydration status. Eur J Clin Nutr 57, Suppl. 2, S6S9.Google Scholar
18 Lang, F, Busch, GL, Ritter, M, et al. (1998) Functional significance of cell volume regulatory mechanisms. Physiol Rev 78, 247306.Google Scholar
19 Koeppen, B & Stanton, B (2000) Renal Physiology, 3rd ed. St Louis, MO: Mosby.Google Scholar
20 Cheuvront, SN, Kenefick, RW, Charkoudian, N, et al. (2013) Physiologic basis for understanding quantitative dehydration assessment. Am J Clin Nutr 97, 455462.Google Scholar
21 Grandjean, AC, Reimers, KJ & Buyckx, ME (2003) Hydration: issues for the 21st century. Nutr Rev 61, 261271.Google Scholar
22 Perrier, E, Rondeau, P, Poupin, M, et al. (2013) Relation between urinary hydration biomarkers and total fluid intake in healthy adults. Eur J Clin Nutr 9, 939943.Google Scholar
23 Bourque, CW (2008) Central mechanisms of osmosensation and systemic osmoregulation. Nat Rev Neurosci 9, 519531.Google Scholar
24 Yasui, M, Marples, D, Belusa, R, et al. (1996) Development of urinary concentrating capacity: role of aquaporin-2. Am J Physiol 271, F461F468.Google Scholar
25 Shoker, AS (1994) Application of the clearance concept to hyponatremic and hypernatremic disorders: a phenomenological analysis. Clin Chem 40, 12201227.Google Scholar
26 Ritz, P & Berrut, G (2005) The importance of good hydration for day-to-day health. Nutr Rev 63, S6S13.Google Scholar
27 Gopinathan, PM, Pichan, G & Sharma, VM (1988) Role of dehydration in heat stress-induced variations in mental performance. Arch Environ Health 43, 1517.Google Scholar
28 Cian, C, Barraud, PA, Melin, B, et al. (2001) Effects of fluid ingestion on cognitive function after heat stress or exercise-induced dehydration. Int J Psychophysiol 42, 243251.Google Scholar
29 Bar-David, Y, Urkin, J & Kozminsky, E (2005) The effect of voluntary dehydration on cognitive functions of elementary school children. Acta Paediatr 94, 16671673.Google Scholar
30 Edmonds, CJ & Burford, D (2009) Should children drink more water? The effects of drinking water on cognition in children. Appetite 52, 776779.Google Scholar
31 Fadda, R, Rapinett, G, Grathwohl, D, et al. (2012) Effects of drinking supplementary water at school on cognitive performance in children. Appetite 59, 730737.Google Scholar
32 Secher, M & Ritz, P (2012) Hydration and cognitive performance. J Nutr Health Aging 16, 325329.Google Scholar
33 Maughan, RJ & Shirreffs, SM (2010) Development of hydration strategies to optimize performance for athletes in high-intensity sports and in sports with repeated intense efforts. Scand J Med Sci Sports 20, Suppl. 2, 5969.Google Scholar
34 Paik, IY, Jeong, MH, Jin, HE, et al. (2009) Fluid replacement following dehydration reduces oxidative stress during recovery. Biochem Biophys Res Commun 383, 103107.Google Scholar
35 Muckelbauer, R, Libuda, L, Clausen, K, et al. (2009) Promotion and provision of drinking water in schools for overweight prevention: randomized, controlled cluster trial. Pediatrics 123, e661e667.Google Scholar
36 Charkoudian, N, Halliwill, JR, Morgan, BJ, et al. (2003) Influences of hydration on post-exercise cardiovascular control in humans. J Physiol 552, 635644.Google Scholar
37 Aufderheide, S, Lax, D & Goldberg, SJ (1995) Gender differences in dehydration-induced mitral valve prolapse. Am Heart J 129, 8386.Google Scholar
38 Borghi, L, Meschi, T, Amato, F, et al. (1996) Urinary volume, water and recurrences in idiopathic calcium nephrolithiasis: a 5-year randomized prospective study. J Urol 155, 839843.Google Scholar
39 Borghi, L, Meschi, T, Schianchi, T, et al. (1999) Urine volume: stone risk factor and preventive measure. Nephron 81, Suppl. 1, 3137.Google Scholar
40 Strippoli, GF, Craig, JC, Rochtchina, E, et al. (2011) Fluid and nutrient intake and risk of chronic kidney disease. Nephrology (Carlton) 16, 326334.Google Scholar
41 Clark, WF, Sontrop, JM, Macnab, JJ, et al. (2011) Urine volume and change in estimated GFR in a community-based cohort study. Clin J Am Soc Nephrol 6, 26342641.Google Scholar
42 Mucke, S, Grotemeyer, KH, Stahlhut, L, et al. (2012) The influence of fluid intake on stroke recurrence – a prospective study. J Neurol Sci 315, 8285.Google Scholar
43 Manz, F, Johner, SA, Wentz, A, et al. (2012) Water balance throughout the adult life span in a German population. Br J Nutr 107, 16731681.Google Scholar
44 Falk, B, Bar-Or, O & MacDougall, JD (1992) Thermoregulatory responses of pre-, mid-, and late-pubertal boys to exercise in dry heat. Med Sci Sports Exerc 24, 688694.Google Scholar
45 Smolander, J, Bar-Or, O, Korhonen, O, et al. (1992) Thermoregulation during rest and exercise in the cold in pre- and early pubescent boys and in young men. J Appl Physiol 72, 15891594.Google Scholar
46 Panel on Dietetic Products, Nutrition, and Allergies (NDA) (2010) Scientific opinion on dietary reference values for water. EFSA J 8, 14591507.Google Scholar
47 IMNA (2006) Dietary References Intakes: The Essential Guide to Nutrient Requirements. Washington, DC: Institute of Medicine of National Academies.Google Scholar
48 Meschi, T, Nouvenne, A & Borghi, L (2011) Lifestyle recommendations to reduce the risk of kidney stones. Urol Clin North Am 38, 313320.Google Scholar
49 Kettaneh, A, Fardet, L, Mario, N, et al. (2010) The 2003 heat wave in France: hydratation status changes in older inpatients. Eur J Epidemiol 25, 517524.Google Scholar
50 Yang, J, Liu, HZ, Ou, CQ, et al. (2013) Impact of heat wave in 2005 on mortality in Guangzhou, China. Biomed Environ Sci 26, 647654.Google Scholar
51 Fitzgerald, B, Middleton, JK & Cooper, SA (1997) Adverse effects of summer amongst people with learning disabilities: neuroleptic malignant syndrome. J Intellect Disabil Res 41, 273277.Google Scholar
52 Desmazes-Dufeu, N, Hubert, D, Burgel, PR, et al. (2005) Severe dehydration and August 2003 heat wave in a cohort of adults with cystic fibrosis. Presse Med 34, 647648.Google Scholar
53 Seraj, MA, Channa, AB, al Harthi, SS, et al. (1991) Are heat stroke patients fluid depleted? Importance of monitoring central venous pressure as a simple guideline for fluid therapy. Resuscitation 21, 3339.Google Scholar
54 Arnaoutis, G, Kavouras, SA, Kotsis, YP, et al. (2013) Ad libitum fluid intake does not prevent dehydration in suboptimally hydrated young soccer players during a training session of a summer camp. Int J Sport Nutr Exerc Metab 23, 245251.Google Scholar
55 Shirreffs, SM, Aragon-Vargas, LF, Chamorro, M, et al. (2005) The sweating response of elite professional soccer players to training in the heat. Int J Sports Med 26, 9095.Google Scholar
56 Biggs, C, Paterson, M & Maunder, E (2011) Hydration status of South African forestry workers harvesting trees in autumn and winter. Ann Occup Hyg 55, 615.Google Scholar
57 Bates, GP, Miller, VS & Joubert, DM (2010) Hydration status of expatriate manual workers during summer in the middle East. Ann Occup Hyg 54, 137143.Google Scholar
58 Polat, M, Akil, I, Yuksel, H, et al. (2006) The effect of seasonal changes on blood pressure and urine specific gravity in children living in Mediterranean climate. Med Sci Monit 12, CR186CR190.Google Scholar
59 Decher, NR, Casa, DJ, Yeargin, SW, et al. (2008) Hydration status, knowledge, and behavior in youths at summer sports camps. Int J Sports Physiol Perform 3, 262278.Google Scholar
60 Kaushik, A, Mullee, MA, Bryant, TN, et al. (2007) A study of the association between children's access to drinking water in primary schools and their fluid intake: can water be ‘cool’ in school? Child Care Health Dev 33, 409415.Google Scholar
61 Stookey, JD (2005) High prevalence of plasma hypertonicity among community-dwelling older adults: results from NHANES III. J Am Diet Assoc 105, 12311239.Google Scholar
62 Ferry, M, Hininger-Favier, I, Sidobre, B, et al. (2001) Food and fluid intake of the SENECA population residing in Romans, France. J Nutr Health Aging 5, 235237.Google Scholar
63 Straub, M & Hautmann, RE (2005) Developments in stone prevention. Curr Opin Urol 15, 119126.Google Scholar
64 Bartoletti, R, Cai, T, Mondaini, N, et al. (2007) Prevalence, incidence estimation, risk factors and characterization of chronic prostatitis/chronic pelvic pain syndrome in urological hospital outpatients in Italy: results of a multicenter case–control observational study. J Urol 178, 24112415 (discussion 2415).Google Scholar
65 Ferrucci, L, Giallauria, F & Guralnik, JM (2008) Epidemiology of aging. Radiol Clin North Am 46, 643652.Google Scholar
66 Forrester, T (2013) Epidemiologic transitions: migration and development of obesity and cardiometabolic disease in the developing world. Nestle Nutr Inst Workshop Ser 71, 147156.Google Scholar
67 Harvey, G, Meir, R, Brooks, L, et al. (2008) The use of body mass changes as a practical measure of dehydration in team sports. J Sci Med Sport 11, 600603.Google Scholar
68 Armstrong, LE, Costill, DL & Fink, WJ (1985) Influence of diuretic-induced dehydration on competitive running performance. Med Sci Sports Exerc 17, 456461.Google Scholar
69 Cheuvront, SN, Ely, BR, Kenefick, RW, et al. (2010) Biological variation and diagnostic accuracy of dehydration assessment markers. Am J Clin Nutr 92, 565573.Google Scholar
70 Armstrong, LE (2005) Hydration assessment techniques. Nutr Rev 63, S40S54.Google Scholar
71 Westerterp, KR, Wouters, L & van Marken Lichtenbelt, WD (1995) The Maastricht protocol for the measurement of body composition and energy expenditure with labeled water. Obes Res 3, Suppl. 1, 4957.Google Scholar
72 Finn, PJ, Plank, LD, Clark, MA, et al. (1996) Progressive cellular dehydration and proteolysis in critically ill patients. Lancet 347, 654656.Google Scholar
73 Lukaski, HC (1987) Methods for the assessment of human body composition: traditional and new. Am J Clin Nutr 46, 537556.Google Scholar
74 Armstrong, LE, Kenefick, RW, Castellani, JW, et al. (1997) Bioimpedance spectroscopy technique: intra-, extracellular, and total body water. Med Sci Sports Exerc 29, 16571663.Google Scholar
75 Kyle, UG, Bosaeus, I, De Lorenzo, AD, et al. (2004) Bioelectrical impedance analysis – part I: review of principles and methods. Clin Nutr 23, 12261243.Google Scholar
76 Kyle, UG, Bosaeus, I, De Lorenzo, AD, et al. (2004) Bioelectrical impedance analysis – part II: utilization in clinical practice. Clin Nutr 23, 14301453.Google Scholar
77 Gudivaka, R, Schoeller, DA, Kushner, RF, et al. (1999) Single- and multifrequency models for bioelectrical impedance analysis of body water compartments. J Appl Physiol 87, 10871096.Google Scholar
78 Pialoux, V, Mischler, I, Mounier, R, et al. (2004) Effect of equilibrated hydration changes on total body water estimates by bioelectrical impedance analysis. Br J Nutr 91, 153159.Google Scholar
79 Piccoli, A (2010) Bioelectric impedance measurement for fluid status assessment. Contrib Nephrol 164, 143152.Google Scholar
80 Hannan, WJ, Cowen, SJ, Fearon, KC, et al. (1994) Evaluation of multi-frequency bio-impedance analysis for the assessment of extracellular and total body water in surgical patients. Clin Sci (Lond) 86, 479485.Google Scholar
81 Shanholtzer, BA & Patterson, SM (2003) Use of bioelectrical impedance in hydration status assessment: reliability of a new tool in psychophysiology research. Int J Psychophysiol 49, 217226.Google Scholar
82 Patel, RV, Peterson, EL, Silverman, N, et al. (1996) Estimation of total body and extracellular water in post-coronary artery bypass graft surgical patients using single and multiple frequency bioimpedance. Crit Care Med 24, 18241828.Google Scholar
83 Olde Rikkert, MG, Deurenberg, P, Jansen, RW, et al. (1997) Validation of multi-frequency bioelectrical impedance analysis in detecting changes in fluid balance of geriatric patients. J Am Geriatr Soc 45, 13451351.Google Scholar
84 Patel, RV, Matthie, JR, Withers, PO, et al. (1994) Estimation of total body and extracellular water using single- and multiple-frequency bioimpedance. Ann Pharmacother 28, 565569.Google Scholar
85 Ward, LC, Elia, M & Cornish, BH (1998) Potential errors in the application of mixture theory to multifrequency bioelectrical impedance analysis. Physiol Meas 19, 5360.Google Scholar
86 Cox-Reijven, PL & Soeters, PB (2000) Validation of bio-impedance spectroscopy: effects of degree of obesity and ways of calculating volumes from measured resistance values. Int J Obes Relat Metab Disord 24, 271280.Google Scholar
87 Hannan, WJ, Cowen, SJ, Plester, C, et al. (1998) Proximal and distal measurements of extracellular and total body water by multi-frequency bio-impedance analysis in surgical patients. Appl Radiat Isot 49, 621622.Google Scholar
88 Earthman, CP, Matthie, JR, Reid, PM, et al. (2000) A comparison of bioimpedance methods for detection of body cell mass change in HIV infection. J Appl Physiol 88, 944956.Google Scholar
89 Ho, LT, Kushner, RF, Schoeller, DA, et al. (1994) Bioimpedance analysis of total body water in hemodialysis patients. Kidney Int 46, 14381442.Google Scholar
90 Baarends, EM, van Marken Lichtenbelt, WD, Wouters, EF, et al. (1998) Body-water compartments measured by bio-electrical impedance spectroscopy in patients with chronic obstructive pulmonary disease. Clin Nutr 17, 1522.Google Scholar
91 Piccoli, A, Pillon, L & Dumler, F (2002) Impedance vector distribution by sex, race, body mass index, and age in the United States: standard reference intervals as bivariate Z scores. Nutrition 18, 153167.Google Scholar
92 Piccoli, A, Piazza, P, Noventa, D, et al. (1996) A new method for monitoring hydration at high altitude by bioimpedance analysis. Med Sci Sports Exerc 28, 15171522.Google Scholar
93 Piccoli, A, Pittoni, G, Facco, E, et al. (2000) Relationship between central venous pressure and bioimpedance vector analysis in critically ill patients. Crit Care Med 28, 132137.Google Scholar
94 Buffa, R, Mereu, RM, Putzu, PF, et al. (2010) Bioelectrical impedance vector analysis detects low body cell mass and dehydration in patients with Alzheimer's disease. J Nutr Health Aging 14, 823827.Google Scholar
95 Buffa, R, Saragat, B, Cabras, S, et al. (2013) Accuracy of specific BIVA for the assessment of body composition in the United States population. PLOS ONE 8, e58533.Google Scholar
96 Cox-Reijven, PL, van Kreel, B & Soeters, PB (2003) Bioelectrical impedance measurements in patients with gastrointestinal disease: validation of the spectrum approach and a comparison of different methods for screening for nutritional depletion. Am J Clin Nutr 78, 11111119.Google Scholar
97 Dittmar, M & Reber, H (2002) Evaluation of different methods for assessing intracellular fluid in healthy older people: a cross-validation study. J Am Geriatr Soc 50, 104110.Google Scholar
98 De Lorenzo, A, Candeloro, N, Andreoli, A, et al. (1995) Determination of intracellular water by multifrequency bioelectrical impedance. Ann Nutr Metab 39, 177184.Google Scholar
99 Sinning, WE & Morgan, AL (1993) The effects of body position on bioimpedance spectroscopy. Basic Life Sci 60, 377380.Google Scholar
100 Roos, AN, Westendorp, RG, Frolich, M, et al. (1992) Tetrapolar body impedance is influenced by body posture and plasma sodium concentration. Eur J Clin Nutr 46, 5360.Google Scholar
101 O'Brien, C, Young, AJ & Sawka, MN (2002) Bioelectrical impedance to estimate changes in hydration status. Int J Sports Med 23, 361366.Google Scholar
102 Gudivaka, R, Schoeller, D & Kushner, RF (1996) Effect of skin temperature on multifrequency bioelectrical impedance analysis. J Appl Physiol (1985) 81, 838845.Google Scholar
103 Huang, JJ, Cheng, KS & Peng, CJ (2000) Temperature-compensated bioimpedance system for estimating body composition. IEEE Eng Med Biol Mag 19, 6673.Google Scholar
104 Birkemose, M, Møller, AJ, Madsen, ML, et al. (2013) Electrode placement in bioimpedance spectroscopy: evaluation of alternative positioning of electrodes when measuring relative dehydration in athletes. Conf Proc IEEE Eng Med Biol Soc 2013, 30283031.Google Scholar
105 Mazariegos, M, Pithan, C, Meyer, A, et al. (1998) Bioelectrical impedance spectroscopy (BIS) in young children with acute and semi-acute hydration disorders: potentials and limitations. Appl Radiat Isot 49, 611614.Google Scholar
106 Grant, MM & Kubo, WM (1975) Assessing a patient's hydration status. Am J Nurs 75, 13071311.Google Scholar
107 Kenefick, RW, Ely, BR, Cheuvront, SN, et al. (2009) Prior heat stress: effect on subsequent 15-min time trial performance in the heat. Med Sci Sports Exerc 41, 13111316.Google Scholar
108 Gaebelein, CJ & Senay, LC Jr (1980) Influence of exercise type, hydration, and heat on plasma volume shifts in men. J Appl Physiol 49, 119123.Google Scholar
109 Cheuvront, SN & Sawka, MN (2005) Hydration assessments of athletes. Sports Sci Exchange 18, 16.Google Scholar
110 Costa, RJ, Teixeira, A, Rama, L, et al. (2013) Water and sodium intake habits and status of ultra-endurance runners during a multi-stage ultra-marathon conducted in a hot ambient environment: an observational field based study. Nutr J 12, 13.Google Scholar
111 Oppliger, RA & Bartok, C (2002) Hydration testing of athletes. Sports Med 32, 959971.Google Scholar
112 Perrier, E, Vergne, S, Klein, A, et al. (2013) Hydration biomarkers in free-living adults with different levels of habitual fluid consumption. Br J Nutr 109, 16781687.Google Scholar
113 Pross, N, Demazieres, A, Girard, N, et al. (2013) Influence of progressive fluid restriction on mood and physiological markers of dehydration in women. Br J Nutr 109, 313321.Google Scholar
114 Armstrong, LE, Maresh, CM, Castellani, JW, et al. (1994) Urinary indices of hydration status. Int J Sport Nutr 4, 265279.Google Scholar
115 Oppliger, RA, Magnes, SA, Popowski, LA, et al. (2005) Accuracy of urine specific gravity and osmolality as indicators of hydration status. Int J Sport Nutr Exerc Metab 15, 236251.Google Scholar
116 Armstrong, LE, Soto, JA, Hacker, FT, et al. (1998) Urinary indices during dehydration, exercise, and rehydration. Int J Sport Nutr 8, 345355.Google Scholar
117 Hackney, AC, Coyne, JT, Pozos, R, et al. (1995) Validity of urine–blood hydrational measures to assess total body water changes during mountaineering in the sub-Arctic. Arctic Med Res 54, 6977.Google Scholar
118 Melin, B, Jimenez, C, Savourey, G, et al. (1997) Effects of hydration state on hormonal and renal responses during moderate exercise in the heat. Eur J Appl Physiol Occup Physiol 76, 320327.Google Scholar
119 Shirreffs, SM, Watson, P & Maughan, RJ (2007) Milk as an effective post-exercise rehydration drink. Br J Nutr 98, 173180.Google Scholar
120 Kovacs, EM, Senden, JM & Brouns, F (1999) Urine color, osmolality and specific electrical conductance are not accurate measures of hydration status during postexercise rehydration. J Sports Med Phys Fitness 39, 4753.Google Scholar
121 Cleary, MA, Hetzler, RK, Wasson, D, et al. (2012) Hydration behaviors before and after an educational and prescribed hydration intervention in adolescent athletes. J Athl Train 47, 273281.Google Scholar
122 Peacock, OJ, Stokes, K & Thompson, D (2011) Initial hydration status, fluid balance, and psychological affect during recreational exercise in adults. J Sports Sci 29, 897904.Google Scholar
123 Armstrong, LE, Pumerantz, AC, Fiala, KA, et al. (2010) Human hydration indices: acute and longitudinal reference values. Int J Sport Nutr Exerc Metab 20, 145153.Google Scholar
124 NCAA (National Collegiate Athletics Association) (2003) 2003 NCAA Wrestling Rules and Interpretations. Indianapolis, IN: NCAA.Google Scholar
125 Logan-Sprenger, HM & Spriet, LL (2013) The acute effects of fluid intake on urine specific gravity and fluid retention in a mildly dehydrated state. J Strength Cond Res 27, 10021008.Google Scholar
126 Osterberg, KL, Horswill, CA & Baker, LB (2009) Pregame urine specific gravity and fluid intake by National Basketball Association players during competition. J Athl Train 44, 5357.Google Scholar
127 Wakefield, B, Mentes, J, Diggelmann, L, et al. (2002) Monitoring hydration status in elderly veterans. West J Nurs Res 24, 132142.Google Scholar
128 Siegrist, D, Hess, B, Montandon, M, et al. (1993) Urinary specific gravity – comparative measurements using reagent strips and refractometer in 340 morning urine samples. Schweiz Rundsch Med Prax 82, 112116.Google Scholar
129 Rowat, A, Smith, L, Graham, C, et al. (2011) A pilot study to assess if urine specific gravity and urine colour charts are useful indicators of dehydration in acute stroke patients. J Adv Nurs 67, 19761983.Google Scholar
130 Perrier, E, Demazieres, A, Girard, N, et al. (2013) Circadian variation and responsiveness of hydration biomarkers to changes in daily water intake. Eur J Appl Physiol 113, 21432151.Google Scholar
131 Simerville, JA, Maxted, WC & Pahira, JJ (2005) Urinalysis: a comprehensive review. Am Fam Physician 71, 11531162.Google Scholar
132 Shirreffs, SM & Maughan, RJ (1998) Urine osmolality and conductivity as indices of hydration status in athletes in the heat. Med Sci Sports Exerc 30, 15981602.Google Scholar
133 Armstrong, LE, Johnson, EC, McKenzie, AL, et al. (2013) Interpreting common hydration biomarkers on the basis of solute and water excretion. Eur J Clin Nutr 67, 249253.Google Scholar
134 Popowski, LA, Oppliger, RA, Patrick Lambert, G, et al. (2001) Blood and urinary measures of hydration status during progressive acute dehydration. Med Sci Sports Exerc 33, 747753.Google Scholar
135 Godevithanage, S, Kanankearachchi, PP, Dissanayake, MP, et al. (2010) Spot urine osmolality/creatinine ratio in healthy humans. Kidney Blood Press Res 33, 291296.Google Scholar
136 Osailan, SM, Pramanik, R, Shirlaw, P, et al. (2012) Clinical assessment of oral dryness: development of a scoring system related to salivary flow and mucosal wetness. Oral Surg Oral Med Oral Pathol Oral Radiol 114, 597603.Google Scholar
137 Saeves, R, Reseland, JE, Kvam, BM, et al. (2012) Saliva in Prader–Willi syndrome: quantitative and qualitative characteristics. Arch Oral Biol 57, 13351341.Google Scholar
138 Ship, JA & Fischer, DJ (1997) The relationship between dehydration and parotid salivary gland function in young and older healthy adults. J Gerontol A Biol Sci Med Sci 52, M310M319.Google Scholar
139 Walsh, NP, Laing, SJ, Oliver, SJ, et al. (2004) Saliva parameters as potential indices of hydration status during acute dehydration. Med Sci Sports Exerc 36, 15351542.Google Scholar
140 Singh, NR & Peters, EM (2013) Markers of hydration status in a 3-day trail running event. Clin J Sport Med 23, 354364.Google Scholar
141 Ely, BR, Cheuvront, SN, Kenefick, RW, et al. (2011) Limitations of salivary osmolality as a marker of hydration status. Med Sci Sports Exerc 43, 10801084.Google Scholar
142 Fortes, MB, Diment, BC, Di Felice, U, et al. (2011) Tear fluid osmolarity as a potential marker of hydration status. Med Sci Sports Exerc 43, 15901597.Google Scholar
143 Santos, MT, Batista, R, Guare, RO, et al. (2011) Salivary osmolality and hydration status in children with cerebral palsy. J Oral Pathol Med 40, 582586.Google Scholar
Figure 0

Table 1 Advantages and disadvantages of methods available for assessing hydration status