NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Women's and Children's Health (UK). Diarrhoea and Vomiting Caused by Gastroenteritis: Diagnosis, Assessment and Management in Children Younger than 5 Years. London: RCOG Press; 2009 Apr. (NICE Clinical Guidelines, No. 84.)

Cover of Diarrhoea and Vomiting Caused by Gastroenteritis

Diarrhoea and Vomiting Caused by Gastroenteritis: Diagnosis, Assessment and Management in Children Younger than 5 Years.

Show details

4Assessing dehydration and shock

4.1. Clinical assessment


Dehydration is the primary serious complication of gastroenteritis. To manage gastroenteritis safely and effectively it is necessary to be able to recognise the presence of dehydration based on clinical assessment. Those with dehydration require fluid administration to replace the fluid and electrolyte deficit. Many have suggested criteria for determining the degree of dehydration in order to try to accurately estimate the deficit replacement to be administered. Some children with gastroenteritis may develop hypovolaemic shock, requiring specific emergency treatment. It is critically important that the manifestations of shock are immediately recognised. In this chapter, these matters are addressed in detail.

Recognising the child with a lesser degree of dehydration is not as easy as identifying the presence of advanced dehydration or hypovolaemic shock. In practice, when assessing a child with gastroenteritis, the clinician should first consider whether there are risk factors for dehydration, i.e., how likely it is that the child is or may soon become dehydrated. Second, the clinician should decide whether there are symptoms and signs present that indicate the presence of dehydration. The GDG therefore considered both of these matters as important elements in the assessment for dehydration.

4.1.1. Risk factors for dehydration

Clinical question

What factors are associated with an increased risk of dehydration?

In total, 40 papers of potential importance were retrieved after the literature search, of which 35 were excluded after being assessed. Thus four case–control studies describing risk factors for the development of dehydration and published in five papers have been included.69–73 Since all the studies were conducted in developing countries, efforts have been made to describe clinical risk factors relevant to the UK paediatric population.

Evidence overview

The first paper was a case–control study from India69 with a study sample of 379 infants with acute gastroenteritis of less than 24 hours’ duration. Cases were defined as infants with moderate or severe gastroenteritis (n = 243), while controls had no or mild dehydration (n = 136). Various factors were evaluated for the risk of dehydration: aetiology, feeding practices, management of diarrhoea, hygiene practices, history of measles and clinical features on admission. Univariate analysis identified various risk factors associated with increased risk. However, after controlling for confounding variables during multivariate analysis, only two factors were found to be significantly associated with an increased risk of dehydration: withdrawal of breastfeeding during diarrhoea (OR 6.8; 95% CI 3.8 to 12.2; P < 0.001) and not giving oral rehydration salt (ORS) solution during diarrhoea (OR 2.1; 95% CI 1.2 to 3.6; P = 0.006). Age, severity of symptoms and nutritional status were identified as major confounding variables. There was a significant risk of dehydration if the child was younger than 12 months (OR 2.7; 95% CI 1.5 to 5.0; P = 0.001), had increased frequency of stool > 8 per day (OR 4.1; 95% CI 2.4 to 7.0; P < 0.00001), had increased frequency of vomiting > 2 per day (OR 2.4; 95% CI 1.4 to 4.0; P = 0.001) or was severely malnourished with weight for age < 60th centile according to the Indian Academy of Paediatrics (IAP) classification (OR 3.1; 95% CI 1.6 to 5.9; P = 0.001). [EL = 2+]

Children younger than 5 years with acute gastroenteritis (duration not specified) and with either severe or moderate dehydration (n = 387 cases) or mild or no dehydration (n = 387 controls) and admitted in a hospital were described in another case–control study from India.70 The authors investigated risk factors for dehydration in terms of demographic factors, nutritional status, hygiene practices, clinical features on admission, history of measles and management of diarrhoea. Multivariate analysis showed age younger than 12 months (OR 1.5; 95% CI 1.02 to 2.3; P = 0.038) and Muslim religion (OR 1.64; 95% CI 1.01 to 2.7; P = 0.048) to be associated with risk of dehydration but the lower values of the confidence intervals were close to the null value. Severe undernutrition (weight for age < 60th centile on the IAP classification) was significantly associated with dehydration (OR 1.6; 95% CI 1.3 to 1.9; P < 0.001). Clinical features on admission significantly associated with dehydration included increased stool frequency > 8 per day (OR 8.8; 95% CI 5.9 to 13.0; P < 0.001) and vomiting frequency > 2 per day (OR 2.6; 95% CI 1.7 to 3.8; P < 0.001). History of measles in the past 6 months (OR 2.9; 95% CI 1.5 to 5.6; P = 0.001), withdrawal of breastfeeding during diarrhoea (OR 3.6; 95% CI 2.1 to 6.2; P < 0.001), withdrawal of fluids during diarrhoea (OR 1.6; 95% CI 1.1 to 2.4; P < 0.001) and not giving ORS solution or ‘home available fluids’ during diarrhoea (OR 1.98; 95% CI 1.3 to 2.9; P < 0.001) were all significantly associated with increased risk of dehydration. [EL = 2+]

Results from a third case–control study from Brazil were published in two articles.71,72 Cases included children younger than 2 years admitted with diarrhoea of less than 8 days’ duration with moderate or severe dehydration (n = 192), while controls were children matched to cases by neighbourhood and age who experienced non-dehydrating diarrhoea in the week preceding the study (n = 192). Cases and controls were compared using logistic regression analysis of matched studies. The authors looked at a wide range of prognostic factors including biological, anthropometric and dietary variables, morbidity and clinical symptoms.

The first publication71 reported that, although many factors were associated with an increased risk of dehydration after adjustment for age and socio-economic status, strong association (at P < 0.001) was seen only for the child’s age, birthweight and other anthropometric measures, birth interval and feeding mode. Younger age was significantly associated with an increased risk of dehydration with the risk about seven times higher in the 2–3 month age group compared with those in the 9–11 month age group (adjusted OR 7.1; 95% CI 3.0 to 16.5). Children of low birthweight (<2500 g) were about three times more likely to become dehydrated than other children. Although other growth-related measures (height for age, weight for age, length of age) showed evidence of significant association, these more complex indices were found to be less useful in terms of sensitivity and specificity. The risk of dehydration was also three times higher in children not breastfed compared with those who received only breast milk (adjusted OR 3.3; 95% CI 1.4 to 7.5). [EL = 2+]

In the second publication,72 it was reported that breastfeeding reduced the risk of dehydration when compared with feeding with other types of milk (formula or cow’s milk). After adjustment for age and other confounding variables, it was seen that children only on cow’s or formula milk had a significantly higher risk of developing dehydration compared with children who were exclusively breastfed (adjusted OR 6.0; 95% CI 1.8 to 19.8 for cow’s milk and adjusted OR 6.9; 95% CI 1.4 to 33.3 for formula feeds). There was no difference in the risk of dehydration if children continued with their usual feeds during illness (either breastfeeds or other feeds), but breastfed children who stopped feeding during illness had a statistically significant increase in the risk (adjusted OR 6.4; 95% CI 2.3 to 17.3). It was also observed that the risk of dehydration was greatest during the time period when breastfeeding was stopped, and this higher risk remained statistically significant till after 6 months of full weaning. [EL = 2+]

The fifth paper reported a case–control study conducted in a hospital in Bangladesh73 that included 240 children younger than 2 years with acute gastroenteritis (duration less than 7 days) of which 80 children had severe or moderate dehydration (cases) and 160 children had ‘no signs of dehydration’ (controls). The cases and controls were matched by age. Thirty-eight socio-demographic, clinical or environmental factors were studied for their influence on development of dehydration. In addition to a number of socio-demographic and environmental factors, there was a statistically significant association (at P < 0.05) of the following clinical factors with dehydration: duration of diarrhoea at hospital attendance, stool frequency of more than five per day, ‘vomiting during episode’, receiving oral rehydration therapy (ORT) at home before admission, receiving drugs at home before admission and ‘wasted child’. All the significant factors were then analysed in a step-wise regression model and the results showed two clinical factors to be independently associated with the development of dehydration: vomiting during episode and received ORT at home before admission. Since the information was collected by a pre-tested questionnaire, information on the preparation and method of giving oral fluids could not be collected and the authors attributed the increased risk in children receiving ORT to ineffective preparation and administration of oral fluids. [EL = 2+]

Evidence summary

There were four relevant case–control studies of good quality [all EL = 2+] conducted in countries with similar healthcare settings but different from that of the UK. Despite the different location of research and culture-specific risk factors investigated, these studies showed consistent results for widely applicable risk factors for the development of dehydration in children with gastroenteritis. In terms of demographic factors, younger children (younger than 12 months, with even higher risk for those very young) and those with malnutrition were at a greater risk of dehydration. The studies showed a consistent and strong association of severity of symptoms, i.e. increased frequency of vomiting (>2 episodes per day) and stool production (>5 episodes per day), with a greater risk of dehydration. In terms of management, withdrawal of breastfeeding and other fluids including ORS solution during diarrhoea were strongly associated with risk of dehydration.

GDG translation from evidence to recommendations

The GDG recognised that the clinical studies available were conducted in resource-poor developing countries. In those settings, there would probably be differences from the UK such as a high prevalence of malnutrition. Nevertheless, the GDG considered that the consistency with which these studies identified specific risk factors was likely to be important. Moreover, some of the findings were both intuitively to be expected and consistent with clinical experience in the UK. Thus, frequent or persistent diarrhoea and vomiting were almost certainly important. The risks identified in relation to age and birthweight were consistent with physiological principles and with clinical experience and were also important. The finding in clinical studies that prior administration of ORT reduced the risk seemed intuitively credible. The consistent finding in the studies that continued breastfeeding was associated with a reduced risk was also potentially important.

Recommendations on risk factors for dehydration

Recognise that the following are at increased risk of dehydration:

  • children younger than 1 year, particularly those younger than 6 months
  • infants who were of low birthweight
  • children who have passed more than five diarrhoeal stools in the previous 24 hours
  • children who have vomited more than twice in the previous 24 hours
  • children who have not been offered or have not been able to tolerate supplementary fluids before presentation
  • infants who have stopped breastfeeding during the illness
  • children with signs of malnutrition.

4.1.2. Clinical detection of dehydration and assessment of severity

Clinical questions

What symptoms or signs (individually or in combination) can detect the presence of dehydration? If dehydration is detected, what symptoms or signs (individually or in combination) can indicate its severity?

It is a common practice in textbooks and guidelines to suggest various clinical manifestations as being indicative of dehydration. It has also been suggested in various published guidelines that the severity of dehydration may be estimated on the basis of symptoms and signs. The GDG explored the published literature to examine the available evidence on these important considerations.

Evidence overview

A systematic literature search was undertaken to inform the two questions. Two studies are included for the first question on the accuracy of clinical signs and symptoms in detecting dehydration, while for the second question four published guidelines are included. These guidelines had employed different methods for classifying severity of dehydration.

Clinical detection of dehydration

Two relevant studies were identified, the first a systematic review of diagnostic studies and the second a cohort study comparing digitally measured capillary refill time (CRT) with conventional CRT and overall clinical assessment.

The systematic review74 was conducted to review the precision and accuracy of symptoms, signs and basic laboratory tests for evaluating 5% (or worse) dehydration in young children aged 1 month to 5 years. After a systematic literature search of the MEDLINE database, additional searches were conducted on the individual symptoms and signs, the Cochrane Library, and the reference lists of text books and of all included articles. After reviewing hard copies of 110 articles, 26 articles fulfilled the inclusion criteria and underwent a quality assessment, and 13 studies were finally selected for inclusion.

The reference standard used for assessing dehydration was the ‘percentage of volume lost’, calculated as the difference between the rehydration weight (the post-rehydration weight) and the acute weight (the weight at presentation) divided by the rehydration weight. Three of the included studies were based on independent, blind comparison of the test with the reference standard but the participants were enrolled in a non-consecutive manner. The remaining ten studies were based on non-independent comparisons of a test with the reference standard and no selection criteria were defined. Hence overall the quality of the included studies was poor. Meta-analysis of the accuracy results using the random effects model was conducted only if more than two studies evaluated a specific diagnostic test. [EL = 3]

Although the authors reported test accuracy results for detecting 5% dehydration, detailed reviewing of the individual studies included in the review revealed that the results were applicable for the detection of 5% or worse dehydration.


Agreement between parental observation of signs and the signs elicited by trained emergency department nurses was evaluated in a single study. The best level of agreement was reported for sunken anterior fontanelle (k = 0.73) and presence of cool extremities (k = 0.70), while moderate agreement was seen for general appearance, presence of sunken eyes, absence of tears, and presence of dry mouth (k values in the range 0.46–0.57). Three studies reported on agreement among clinicians but wide variation was seen in the results for the various signs. Prolonged CRT had k values ranging from 0.01 to 0.65, while absent tears had values from 0.12 to 0.75. For the rest of the signs, the level of agreement was either slightly better than the chance agreement (k value in the range 0.50–0.60) or worse than the chance agreement (k < 0.50).

Clinical history including symptoms

Three studies were included and all of them evaluated history of low urine output as a test for detecting dehydration. A pooled analysis showed that it did not increase the likelihood of detecting 5% dehydration (+LR 1.3; 95% CI 0.9 to 1.9). However, results from two studies showed that parental reporting of a normal urine output decreased the likelihood of detecting dehydration, although the results were statistically not significant in one study (−LR 0.27; 95% CI 0.14 to 0.51 and −LR 0.16; 95% CI 0.01 to 2.53). One study reported LRs for a number of symptoms, including history of vomiting and diarrhoea (severity), decreased oral intake, and a previous trial of clear liquids, but none of these were found to be helpful in increasing or decreasing the likelihood. This study also suggested that children who had not been previously evaluated by a physician during the illness might be less likely to be dehydrated, but the results were again statistically not significant (−LR 0.09; 95% CI 0.01 to 1.37).


The results of the test characteristics of various signs are given in Table 4.1. Three signs showed evidence of increasing the likelihood of detecting 5% dehydration: prolonged CRT (four studies, +LR 4.1; 95% CI 1.7 to 9.8), abnormal skin turgor (four studies, +LR 2.5; 95% CI 1.5 to 4.2) and abnormal respiratory pattern (four studies, +LR 2.0; 95% CI 1.5 to 2.7). Sunken eyes and dry mucous membranes showed a small increase in the likelihood of dehydration (+LR for both 1.7) and the lower limit of their 95% CI was close to the null value. Results for weak pulse as a predictor for dehydration were variable, with one study showing it to be a fair predictor (+LR 3.1; 95% CI 1.8 to 5.4) while another did not (+LR 7.2; 95% CI 0.4 to 150). The presence of cool extremities as a test for dehydration was examined in two studies and both reported imprecise point estimates for the +LR (95% CI too wide to draw conclusions). The 95% CI for the positive and negative LRs for increased heart rate, sunken fontanelle in young infants, and an overall poor appearance included the null value.

Table 4.1. Summary characteristics of clinical signs used to detect 5% dehydration.

Table 4.1

Summary characteristics of clinical signs used to detect 5% dehydration.

A second prospective cohort study75 aimed to determine whether CRT measured using a digital device (DCRT) could determine the presence of significant dehydration. The study population comprised 83 children (aged 1 month to 5 years) with acute gastroenteritis admitted to an accident and emergency department in Canada. Following admission and enrolment, the degree of dehydration was estimated using a seven-point Likert scale, CRT was clinically assessed in the conventional way by the paediatric medical staff, and DCRT measured using a small digital video camera with customised graphics software. The reference standard (degree of dehydration) was calculated by measuring the difference between the pre- and post-rehydration weight of the child.

Thirteen (16%) children met the WHO definition of dehydration (≥5%), with 12 estimated to have a fluid deficit between 5% and 8% and one with 11% deficit. For these children, there was a strong correlation between the child’s fluid deficit and the DCRT (Pearson’s correlation coefficient 0.75; P < 0.001). The AROC for detecting presence of dehydration ≥ 5% was 0.99 for DCRT and 0.88 for clinical assessment. DCRT showed the best result for predicting dehydration more than 5%, with 100% sensitivity, 91% specificity and a +LR of 11.4 (95% CI 5.4 to 22). Compared with the clinical assessment scale, conventional CRT showed better results for specificity (88% versus 81%) and for +LR (4.5 versus 4.1), but poorer results for sensitivity (54% versus 77%). [EL = 2]

Clinical assessment of the severity of dehydration

Four guidelines had classified degrees of dehydration by using a combination of signs and symptoms. These are summarised in Tables 4.24.4.

Table 4.2. Classification of dehydration severity by Armon et al.

Table 4.2

Classification of dehydration severity by Armon et al.

Table 4.3. Classification of dehydration severity by WHO.

Table 4.3

Classification of dehydration severity by WHO.

Table 4.4. Classification of dehydration severity by ESPGHAN.

Table 4.4

Classification of dehydration severity by ESPGHAN.

Evidence summary

Results from a systematic review [EL = III] suggest that prolonged capillary refill time, abnormal skin turgor and abnormal respiratory pattern are the signs most useful to detect 5% or worse dehydration in a child with gastroenteritis. Sunken eyes and dry mucous membrane were also found to be useful although their predictive value was less than the above three signs. For the other signs and symptoms, either the pooled likelihood ratios were statistically not significant or there was wide variation in the results from individual studies. Results also show that there was generally a poor agreement between clinicians on the presence of these clinical signs. Another study [EL = II] showed that CRT measured using a digital video technique (DCRT) had better accuracy in detecting dehydration of 5% or worse than the conventional clinical CRT and the clinical assessment scale.

Although the published guidelines employed different methods of classifying the severity of dehydration, they all used similar symptoms and signs (individually or in combination) for these classification methods.

GDG translation from evidence to recommendations

Clinical detection of dehydration

A range of symptoms and signs have traditionally been considered useful in the detection of dehydration. The GDG found that many of these did not have evidence with regard to their reliability, particularly in those children with less severe degrees of dehydration.

The GDG considered that the identification of symptoms useful for the detection of dehydration would be important, particularly because they could be employed as part of the remote assessment process. However, the only symptom of possible value identified from the evidence was a report of ‘normal urine output’ and the evidence between studies was inconsistent. The GDG considered that enquiry should be made about this matter, and that some reassurance could be taken if the urine output was said to be normal. The GDG also agreed that carers were acutely aware of any change in the child’s behaviour (irritability, lethargy) and appearance (for example ‘sunken eyes’), and so it seemed appropriate to specifically enquire about these.

With regard to the role of physical signs in the detection of dehydration, the GDG examined the evidence from the systematic review and identified two limitations to the included studies. First, the review appeared to report on children with 5% dehydration or worse and, given the symptoms and signs identified, the GDG strongly suspected that many of the patients might have been considerably more than 5% dehydrated. Therefore, the stronger associations reported for CRT, abnormal skin turgor and abnormal breathing pattern did not mean that these signs would be useful for the detection of less severe dehydration. On the contrary, the GDG considered that those signs suggested the presence of relatively severe dehydration. Second, neither the prevalence of dehydration nor the post-test probabilities of dehydration were presented. One or other of these parameters was needed to interpret the likelihood ratios presented.

Therefore, this study did not provide reliable evidence on the value of symptoms and signs for the detection of lesser degrees of dehydration. However, the GDG agreed that the presence of one or more symptoms or signs evaluated in those studies and conventionally employed in assessment for dehydration would suggest clinically significant dehydration.

The study of CRT using a digital technique (DCRT) showed a relationship between abnormal DCRT and dehydration. However, this remains an experimental technique that is not yet established as a method for routine clinical use.

Clinical assessment of dehydration severity

The GDG recognised that there was a lack of compelling evidence to support efforts to accurately distinguish varying degrees of dehydration on the basis of symptoms and signs. In the absence of such evidence, any system of classification was inevitably arbitrary and subjective and based on the clinician’s judgement and a ‘global assessment’ of the child’s condition.

In the past, it was common to describe three levels of dehydration, referred to as mild (3–5%), moderate (6–9%) and severe (≥10%), with an implication that it was possible to make such distinctions based on the clinical assessment (see Table 4.5). A number of recent guidelines (Tables 4.3 and 4.4) had adopted simpler schemes in which just two degrees of dehydration were to be distinguished – ‘some dehydration’ (or ‘mild to moderate dehydration’), variably defined as 3–8% or 5–10% dehydration, and ‘severe dehydration’, variably defined as ≥9% or >10% dehydration. Even these simpler classifications could be difficult to implement in clinical practice. The GDG considered that it was not possible to accurately distinguish ‘sunken’ and ‘very sunken’ or ‘deeply sunken’ eyes, or between skin pinch retracting ‘slowly’ and ‘very slowly’, or between ‘dry’ and ‘very dry’ mucous membranes. There was also no evidence on the reliability of these various signs either individually or in combination in distinguishing varying degrees of dehydration. In addition, there was no evidence to justify arbitrary categorisation on the basis of specific numbers of clinical symptoms or signs as had been suggested (Table 4.3).

Table 4.5. Classification of dehydration severity by the American Subcommittee on Acute Gastroenteritis.

Table 4.5

Classification of dehydration severity by the American Subcommittee on Acute Gastroenteritis.

The GDG decided to adopt a new and even simpler clinical assessment scheme (Table 4.6) Patients would merely be classified as follows: ‘no clinically detectable dehydration’, ‘clinical dehydration’ and ‘clinical shock’. With this assessment scheme the clinician would have to recognise the presence of clinical dehydration. This simplified scheme does not imply that the degree of dehydration is uniform, but rather acknowledges the difficulties in accurately assessing dehydration severity. The GDG recognised that experienced clinicians could distinguish marked differences in the severity of dehydration. They also considered that clinical signs were likely to be more pronounced and numerous in those with severe dehydration. However, firm recommendations linking clinical symptoms and signs with specific varying levels of dehydration were impossible. The crucial point however, is that the scheme is all that is required to guide fluid management (Chapter 5). In this guideline a standard fluid regimen is recommended for all (non-shocked) children with dehydration, with adjustments being made to the fluid regimen over time based on regular reassessment during the rehydration process.

Table 4.6. Symptoms and signs of clinical dehydration and shock.

Table 4.6

Symptoms and signs of clinical dehydration and shock. Interpret symptoms and signs taking risk factors for dehydration into account. Within the category of ‘clinical dehydration’ there is a spectrum of severity indicated by increasingly (more...)

The GDG was aware of the crucial importance of identifying those children with hypovolaemic shock. They would require specific emergency management with administration of IV fluid boluses (Section 5.4) and so it was essential that signs of shock should be recognised without delay. Many patients with hypovolaemic shock were likely to have obvious and pronounced signs of dehydration in addition to the specific clinical manifestations of shock. However, this might not always be the case. For example, a small infant with gastroenteritis might experience sudden severe fluid loss at the onset of gastroenteritis sufficient to cause hypovolaemic shock before any signs of dehydration (for example, dry mucous membranes or reduced skin turgor) were present. Hence it was appropriate to distinguish the symptoms and signs of shock from those of dehydration. Inevitably, there was some overlap, in that both dehydration and shock might be associated with a change in conscious state. In dehydration, lethargy or irritability might commonly occur, while in shock there might be a more profound depression of consciousness. Likewise, dehydration would often cause an increased heart rate but in shock this might be much more pronounced. The diagnosis of shock would be based on the clinician’s global assessment, taking account of each of the relevant symptoms and signs. With severe shock the manifestations would be unequivocal. In lesser degrees of shock, for example as the symptoms and signs first appeared, there might be some difficulty in distinguishing it from severe dehydration. The GDG concluded that when there was uncertainty the safe approach would be to treat as though shock was present (Section 5.4).

The GDG identified several ‘red flag’ signs in dehydration whose presence should alert the clinician to a risk of progression to shock (see Table 4.6). These were altered responsiveness (for example, irritable, lethargic), sunken eyes, tachycardia, tachypnoea, and reduced skin turgor. Children with such red flag signs require especially careful consideration and close monitoring. The GDG considered that monitoring to follow the ‘illness trajectory’ was critically important particularly in these ill children. Thus tachycardia (a red flag sign) would be of even greater concern if it worsened over time, pointing to a serious risk of clinical deterioration and shock.

The GDG recognised that this recommended clinical assessment scheme was novel and would be unfamiliar to clinicians. However, it had the great advantage of simplicity, would be easy to implement, and would provide the clinical information necessary for appropriate fluid management. As discussed later in Chapter 5, those with dehydration will usually be treated with oral fluid rehydration, those with red flag symptoms and/or evidence of deterioration will require careful management, probably in a hospital setting, while those with suspected or definite shock will require emergency IVT in hospital. In the community setting, it will be necessary for the healthcare professional to decide whether monitoring the response to rehydration therapy can be carried out safely in the home setting and if so under what level of supervision (general practitioner, community children’s nurse, etc.). Where there are concerns about a parent’s ability to monitor their child’s condition and to provide appropriate care, referral to hospital might be required.

The GDG considered that recognition of the symptoms and signs of dehydration and shock needs considerable expertise. Clinicians therefore require training and experience in order to ensure competence in assessing children with gastroenteritis. This should be at an appropriate level to allow the individual to work safely and effectively in their specific clinical role.

Recommendation on clinical detection of dehydration and assessment of severity

During remote or face-to-face assessment ask whether the child:

  • appears unwell
  • has altered responsiveness, for example is irritable or lethargic
  • has decreased urine output
  • has pale or mottled skin
  • has cold extremities.

Use Table 4.6 to detect clinical dehydration and shock.

Research recommendation

In children with gastroenteritis, what is the predictive value of clinical symptoms and signs in assessing the severity of dehydration, using post-rehydration weight gain as the reference standard, in primary and secondary care settings?

Why this is important

Evidence from a systematic review* suggests that some symptoms and signs (for example, prolonged capillary refill time, abnormal skin turgor and abnormal respiratory pattern) are associated with dehydration, measured using the accepted ‘gold standard’ of the difference between pre-hydration and post-hydration weight. However, 10 of the 13 included studies were not blinded and had ill-defined selection criteria. Moreover, all these studies were conducted in secondary care where children with more severe dehydration are managed.

Most children with gastroenteritis can and should be managed in the community* but there is a lack of evidence to help primary care healthcare professionals correctly identify children with more severe dehydration. Symptoms and signs that researchers may wish to investigate include overall appearance, irritability/lethargy, urine output, sunken eyes, absence of tears, changes in skin colour or warmth of extremities, dry mucous membranes, depressed fontanelle, heart rate, respiratory rate and effort, character of peripheral pulses, capillary refill time, skin turgor and blood pressure.


Steiner MJ, DeWalt DA, Byerley JS. Is this child dehydrated? JAMA: the Journal of the American Medical Association 2004;291(22):2746–54.


Hay AD, Heron J, Ness A; the ALSPAC study team. The prevalence of symptoms and consultations in pre-school children in the Avon Longitudinal Study of Parents and Children (ALSPAC): a prospective cohort study. Family Practice 2005;22(4):367–74.

Clinical question

What symptoms and/or signs suggest the presence of hypernatraemic dehydration?

Hypernatraemic dehydration may be defined as dehydration associated with a plasma sodium concentration greater than 150 mmol/l. Some textbooks suggest that the presenting symptoms and signs associated with this condition differ from those in dehydration without hypernatraemia. It is said that these patients may have ‘doughy’ skin, and tachypnoea, and that many of the signs normally associated with dehydration (reduced skin turgor, dryness of the mucous membranes, skin mottling, sunken eyes, altered vital signs) may not occur. The evidence for these reported differences was sought.

Evidence overview

Only one study was found that reported signs and symptoms associated with hypernatraemic dehydration. A prospective comparative study was conducted in South Africa78 to determine the incidence of hypernatraemia in children with diarrhoea and to define its distinguishing symptoms and signs. Serum sodium levels were determined in all children admitted with diarrhoea at the hospital over the course of 1 year (n = 3889). In total, 147 (3.8%) were found to be hypernatraemic (serum sodium > 150 mmol/l). A group of 50 consecutive children with an initial serum sodium < 150 mmol/l formed the control group. No inclusion and exclusion criteria were reported. The study participants underwent a full clinical examination and the degree of dehydration was categorised as ‘not dehydrated’, ‘5% dehydrated’ or ‘10% dehydrated’. The percentage dehydration was calculated from the difference between the weight on admission and after rehydration.

A significantly greater proportion of those with hypernatraemia were younger than 6 months (P < 0.01) compared with the control group. There were no differences regarding gender or nutritional status. Symptoms of central nervous system dysfunction were more common in the hypernatraemic group compared with the non-hypernatraemic children (38% versus 4%; P < 0.001). The authors also reported the numbers of children presenting with various central nervous system symptoms for the two groups: 32 versus 2 were drowsy but rousable; 15 versus 0 were jittery, hypertonic or hyperreflexic; 9 versus 0 children were in coma or had convulsions. When clinical estimation of dehydration was compared with the actual degree of dehydration (based on weight change), dehydration was underestimated in 72.5% of the hypernatraemic group compared with 36% of the non-hypernatraemic group (P < 0.001). The authors reported that in the hypernatraemic group dehydration was often grossly underestimated. [EL = 2]

Evidence summary

Evidence from a single prospective study indicated that hypernatraemia was more common in young infants (<6 months) with diarrhoea. Children with hypernatraemic dehydration had an increased frequency of symptoms of central nervous system dysfunction. Using clinical assessment, the severity of dehydration was more often underestimated in hypernatraemic dehydration than in children with dehydration associated with a normal plasma sodium concentration.

GDG translation from evidence to recommendation

The GDG noted that there was a lack of evidence on this topic, No evidence was found for the often described phenomenon of ‘doughy skin’, and so it was concluded that this finding could not be relied on to clinically identify patients with hypernatraemic dehydration. The GDG also noted that in some publications from North America the term ‘doughy skin’ was used with a different meaning – seemingly being synonymous with ‘reduced skin turgor’, a sign of dehydration more generally. Therefore GDG consensus was that the term ‘doughy’ was not helpful and hence it has not been used in this guideline. On the other hand, it was the experience of GDG members that hypernatraemic dehydration is associated with neurological signs such as an altered level of consciousness, ‘jitteriness’ or muscle hypertonicity, and the presence of these signs should prompt laboratory investigation.

Recommendation on assessment of hypernatraemic dehydration

Suspect hypernatraemic dehydration if there are any of the following:

  • jittery movements
  • increased muscle tone
  • convulsions
  • drowsiness or coma.

4.2. Laboratory investigations for assessing dehydration


There are potential biochemical complications associated with gastroenteritis that can only be identified through blood testing. For example, electrolyte disturbances such as hypernatraemia and hyponatraemia may occur. Hypoglycaemia may occur in some cases. Children with gastroenteritis may become acidotic. Dehydration may lead to pre-renal failure or even acute renal failure. To suggest that all children with gastroenteritis should undergo blood testing would be inappropriate, and yet the clinician must consider the possibility that in some cases clinically important biochemical abnormalities may occur that could require specific treatment. Evidence was therefore sought on the incidence of such biochemical disturbances in children presenting with gastroenteritis, and their accuracy in detecting severity of dehydration. An attempt was made to determine whether the incidence of such complications was increased in specific and clinically identifiable categories of patient.

Clinical questions

How common are biochemical abnormalities in children with gastroenteritis and dehydration?

How accurate are laboratory tests in detecting varying degrees of dehydration?

Evidence overview

After the primary screening, 40 papers were retrieved for reviewing. Most of the retrieved studies had been published in the 1980s and 1990s, used a non-comparative study design and did not give adequate data to calculate the incidences. In the end, five studies were included to provide data on the incidence of biochemical disturbances in children with acute gastroenteritis. For the second part of the question relating to diagnostic accuracy, two studies were included – a systematic review of diagnostic studies and a prospective diagnostic study.

Incidence of biochemical abnormalities

There were three prospective cross-sectional studies from the UK,53,57,58 one from Turkey79 and one retrospective case series from the USA.80 All three studies from the UK have already been included previously under Section 3.2.1.

The first UK study53 included 1148 children younger than 16 years admitted to a sub-regional infectious disease hospital with a diagnosis of gastroenteritis over a 1 year period. Of the admitted children, 55% (635/1148) were younger than 1 year while 5% were over 5 years of age. Admissions were predominantly from socially disadvantaged families (62% from social classes IV and V). At the time of admission, 8.8% of children (101/1148) were clinically dehydrated, with 1% assessed to have greater than 5% dehydration. The group of dehydrated children (n = 101) showed a higher incidence of biochemical disturbances compared with those who were not dehydrated (n = 1047): hypernatraemia (sodium levels > 145 mmol/l) 10.9% versus 0.6%, uraemia (urea > 7 mmol/l) 30% versus 5.3% and low bicarbonate levels (<21 mmol/l) 72% versus 55%. The difference in the incidence of biochemical abnormalities between the two groups was statistically significant (P < 0.001) for all the three parameters. [EL = 3]

In the second UK study,57 447 children younger than 2 years and admitted to a hospital with gastroenteritis were recruited over a 1 year period. Seventy-four percent of the children were younger than 1 year and two-thirds of under-1-year-olds were younger than 6 months. The overall incidence of moderate to severe dehydration (assessed clinically) was 14%. Hypernatraemia (sodium levels ≥ 150 mmol/l) occurred in 0.8% of cases, 8% had raised urea concentration (>6 mmol/l), and 3% had bicarbonate concentration ≤ 15 mmol/l. However, it was not specified whether biochemical abnormalities were found only in children with moderate to severe dehydration. [EL = 3]

Another UK study58 included 215 children admitted to four paediatric units in south Wales with gastroenteritis over a 1 year period. The age of the study population ranged from 2 weeks to 9 years and 61% of children were younger than 1 year. The primary aim of the study was to describe the clinical characteristics, incidence of complications, and management (pre-admission and hospital) of the patients. The authors did not specify the total number of cases with clinical dehydration, but overall only 7% were judged to be severely dehydrated. At the time of admission, blood testing was carried out in 35% of children (76/215) on clinical grounds. The incidence of hypernatraemia among all the children (sodium levels > 145 mmol/l) was 0.9%, while 7.9% each had hyponatraemia (sodium < 135 mmol/l) and raised urea concentration (>6 mmol/l). About 6% of children had acidosis with bicarbonate levels < 15 mmol/l. [EL = 3]

The study from Turkey79 aimed to investigate the relationship between blood glucose and serum electrolytes since it was hypothesised that changes in blood glucose levels during diarrhoea complicate the course of the illness, especially when it is associated with electrolyte abnormalities. The study population included 119 children (age range 2 months to 15 years) with gastroenteritis and moderate to severe dehydration (according to WHO criteria) admitted to a tertiary children’s hospital over a 3 month period. In order to reduce age-dependant variability of laboratory findings, the study population was further divided into two groups: younger than 2 years and more than 2 years of age. More than half of the study population had body weight/age ratio less than the 10th percentile. Blood samples were drawn at the time of admission in all children. Hypernatraemia (sodium levels > 150 mmol/l) was present in 7.6% of all cases and hyponatraemia (sodium levels < 130 mmol/l) in 3.4%, while 48% of children had bicarbonate levels < 15 mmol/l. Potassium levels < 3 mmol/l were noted in 4.2% of children. Hyperglycaemia (blood glucose levels > 140 mg/dl) was observed in 10.9% of cases while hypoglycaemia (threshold value not defined) was noted in only one child. The mean sodium levels were significantly higher in the hyperglycaemic group of children compared with the rest of children, but there was no difference between the two groups regarding serum bicarbonate levels. Similarly, mean sodium levels were noted to be higher in children younger than 2 years with bicarbonate levels < 15 mmol/l compared with those with higher bicarbonate levels (>15 mmol/l). A positive correlation was found between blood glucose and serum sodium levels in children younger than 2 years with bicarbonate levels < 15 mmol/l (r = 0.35; P < 0.05), and this correlation became stronger when the analysis was limited to children with bicarbonate levels < 10 mmol/l (r = 0.73; P < 0.05). No relationship was observed between blood glucose and serum sodium levels in the older age group. However, the authors did not give detailed information about the correlation data. [EL = 3]

A retrospective case series from the USA80 aimed to estimate the prevalence of hypoglycaemia among children with dehydration due to acute gastroenteritis who presented to an urban hospital emergency department. For this study, dehydration was considered to be present in children who received an IV fluid bolus. Hypoglycaemia was defined as serum glucose concentration < 60 mg/dl (3.3 mmol/l). Medical records of 196 children (younger than 5 years) admitted over a 1 year period were reviewed and the mean age of the study sample was 23 months (SD 14 months). Overall, 9.4% of children (18/192) were found to be hypoglycaemic but only one child had serum glucose levels < 40 mg/dl (2.2 mmol/l). On comparing the characteristics of the hypoglycaemic group of children (n = 18) with the non-hypoglycaemic group (n = 178), the mean duration of vomiting (± SD) was found to be significantly longer in hypoglycaemic children (3.3 ± 1.7 days versus 2.4 ± 2.6 days; P < 0.05). Of those children with hypoglycaemia and dehydration, 94% had bicarbonate levels < 18 mEq/l and 19% had blood urea nitrogen (BUN) levels > 18 mg/dl, while in the group of children having normal glucose levels and dehydration, 92% had bicarbonate levels < 18 mEq/l and 29% had BUN levels > 18 mg/dl. The difference between the two groups was not statistically significant for these two parameters. [EL = 3]

The incidences of various biochemical disturbances as identified in the above five studies are shown in Table 4.7. It is important to note that the investigators arbitrarily employed varying definitions for biochemical abnormality, and the clinical importance of these disturbances should be taken into account when considering the results from these studies.

Table 4.7. Incidence of biochemical disturbances in children with gastroenteritis.

Table 4.7

Incidence of biochemical disturbances in children with gastroenteritis.

Accuracy of laboratory tests in detecting dehydration

Two studies were included that evaluated the diagnostic accuracy of laboratory investigations for assessing dehydration – a systematic review and a prospective cohort study. The methodology of the systematic review74 and the results on the accuracy of signs and symptoms are described in detail under Section 4.1. In this section, only the findings relevant to accuracy of laboratory tests are given.

In the systematic review,74 six studies were identified that evaluated the ability of laboratory tests to assess dehydration. Five studies evaluated BUN levels or BUN/serum creatinine ratio as a test for dehydration but they used different thresholds to define an increased level. With a cutoff value of 8, 18 and 27 mg/dl for a high BUN level, the + LRs ranged from 1.4 to 2.9, while a single study found urea levels > 40 mg/dl to significantly increase the likelihood of at least 5% dehydration (+LR 46; 95% CI 2.9 to 733). However, this study had a small sample population and the confidence limits of the likelihood ratio were wide. Acidosis was evaluated in four studies but these studies also used different cut-off values. Two studies defined acidosis as base deficit > 7 mEq/l and they reported +LR of 1.4 and 1.8, and −LR of 0.4 and 0.7, respectively. The other two studies used serum bicarbonate levels < 15 and < 17 mEq/l as indicative of acidosis. Both the studies reported that bicarbonate levels lower than the cut-off values were not helpful in increasing the likelihood of dehydration (+LR of 1.5 and 3.5, respectively), but higher levels were found to be useful in decreasing the likelihood of dehydration (−LR of 0.18 and 0.22, respectively). One study evaluated elevated serum uric acid levels (>600 mmol/l) and increased anion gap (>20 mmol/l) as tests for dehydration but their likelihood ratios contained the null value. [EL = 3]

The second diagnostic study from the USA81 evaluated the accuracy of urine specific gravity, urine ketone levels and urine output in detecting dehydration. This study was part of a larger study to compare the safety and efficacy of rapid IVT given over 1 hour with that of infusion over 3 hours. The study population included 75 children aged 3–36 months admitted to the emergency department with moderate dehydration (clinically estimated) and requiring IVT due to failure of ORT (refusal, recurrent emesis or inadequate intake). After admission, urine samples were collected by catheterisation or spontaneous void and, following successful rehydration with IVT, repeat samples were collected. The reference standard for estimating the degree of dehydration was the percentage weight loss calculated by dividing the difference between the initial weight and final rehydrated weight with the rehydrated weight. Two-thirds of the children (50/75) had ≥3% dehydration while 21% had ≥5% dehydration confirmed by the weight-based criterion. No statistically significant correlation was found between urine specific gravity or urine ketone levels with the degree of dehydration. For urine specific gravity, there was no statistically significant increase in the likelihood of either 3% or 5% dehydration at any of the cut-off values (with the 95% confidence limits containing the null value of 1). Similar results were seen for urine ketone levels. Finally, urine output measured after admission and during rehydration therapy did not correlate with the degree of dehydration, and it was not helpful in increasing or decreasing the likelihood of dehydration. [EL = 3]

Evidence summary

There was a lack of good-quality studies relevant to these two questions.

Result from one UK study showed that the incidence of biochemical abnormalities was significantly higher in dehydrated children compared with those who were not dehydrated, while the other two UK studies did not report separately the distribution of abnormalities among dehydrated and non-dehydrated children. Although the three studies employed variable definitions of abnormality for the various biochemical measurements, the incidence of hypernatraemia was reported to be less than 1% in the two latter studies and in the first study in non-dehydrated children. High urea levels (>6 or >7 mmol/l) were found in 5–8% of children while variation was seen regarding the incidence of acidosis. A study from Turkey found a higher incidence of biochemical abnormalities, but more than half of the children in this study were undernourished. This study also reported a positive correlation between serum sodium levels and blood glucose levels in children younger than 2 years with low bicarbonate levels. The last case series from the USA found 9.4% of children admitted to an emergency department having glucose levels < 3.3 mmol/l, and the mean duration of vomiting in these children was significantly longer compared with rest of the children admitted with gastroenteritis.

Evidence from the systematic review suggests that blood levels of bicarbonate > 15 or > 17 mEq/l (or absence of acidosis) is helpful in decreasing the likelihood of dehydration, while lower levels are not helpful in increasing the chances of detecting dehydration. High urea/BUN levels, high serum uric acid and increased anion gap were also found to be unhelpful in detecting dehydration. Results from another study indicate that urine specific gravity, urine ketone levels and low urine output are not useful diagnostic tests in detecting dehydration.

GDG translation from evidence to recommendations

The GDG found that there was a lack of satisfactory evidence with regard to the incidence of clinically important biochemical disturbances in children with gastroenteritis in the UK. One large study did indicate that hypernatraemia, uraemia and acidosis were more commonly found in those with clinical dehydration. The available studies did not, however, make clear the frequency with which such abnormalities occurred in children with varying levels of dehydration. In studies of large populations of children with gastroenteritis in the UK, the incidence of hypernatraemia was 1% or less, and those populations included children with severe dehydration. Increased plasma bicarbonate levels were significantly associated with dehydration but the practical usefulness of bicarbonate estimation to detect dehydration was unclear. Studies on the potential value of other blood and urine investigations for the detection of dehydration also failed to provide evidence in support of their use.

The GDG recognised that children with shock may develop metabolic acidosis and monitoring of the venous blood acid–base status is clinically important in such cases. Studies of acid–base status usually include the pH, pCO2, bicarbonate, base deficit and lactate. When acidosis is detected, the underlying mechanism should be considered. Diarrhoea is often associated with substantial bicarbonate loss and this is a common cause of acidosis. In hypovolaemic shock, tissue blood perfusion is impaired and this can cause lactic acidosis. In those with acidosis due to bicarbonate loss, spontaneous resolution can be anticipated as the diarrhoea resolves. Patients with hypovolaemic shock require IV fluid bolus administration (see Chapter 5). Such fluid therapy would not be indicated in those with acidosis related to bicarbonate loss. To distinguish these two types of acidosis, the GDG suggested that measurement of the anion gap: ([Na+] + [K+]) − ([HCO3] + [Cl]) could be helpful, particularly in those with symptoms or signs of shock. If the anion gap was increased, this would suggest impaired tissue perfusion as the underlying mechanism. If the anion gap was normal, bicarbonate loss was likely to be important.

The GDG considered that routine measurement of plasma glucose was unnecessary in children who would not otherwise require a blood test. However, if there was reason to clinically suspect hypoglycaemia, for example in an infant with unexplained drowsiness, then it should be measured.

Recommendations on laboratory investigations for assessing dehydration

Do not routinely perform blood biochemical testing.

Measure plasma sodium, potassium, urea, creatinine and glucose concentrations if:

Measure venous blood acid–base status and chloride concentration if shock is suspected or confirmed.

Copyright © 2009, National Collaborating Centre for Women’s and Children’s Health.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK []. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

Bookshelf ID: NBK63845


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...