Page 64 - 15Diarrhoeaandvomiting

Assessing dehydration and shock

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)
70
and admitted in a hospital were described in another case–control study from India. 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 publication reported that, although many factors were associated with an increased
71
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, it was reported that breastfeeding reduced the risk of dehydration
72
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+]
73
The fifth paper reported a case–control study conducted in a hospital in Bangladesh 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

39