Page 69 - 15Diarrhoeaandvomiting

Diarrhoea and vomiting caused by gastroenteritis in children under 5 years

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).
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.
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

44