Page 110 - 15Diarrhoeaandvomiting

Nutritional management

groups at any point. However, there was a significant decrease in the duration of diarrhoea in
the wheat peas diet and potato milk diet groups compared with the soy formula group (57 hours
versus 55 hours versus 154 hours; P = 0.005). [EL = 1−]

The second study 132 was conducted in Nigeria and recruited boys aged 6–24 months who had
been hospitalised for acute watery diarrhoea (defined as more than three liquid stools per day,
and for less than 72 hours). Those with gross faecal blood, who had received more than one dose
of antibiotics, who were breastfeeding more than once a day or who were under 2 SD of the
weight-for-length reference data from the US National Center for Health Statistics were excluded.
ORT (according to WHO guidelines) was given as necessary for up to 8 hours before children
started feeding with their assigned diet type. Children were randomly assigned (using separate
block randomisation lists for under and over age 1 year) to either a maize–cow pea diet (maize
flour, cowpea flour, palm oil, sugar) (n = 35) or a soy-based lactose-free formula (n = 34). The
allocation of diets was not masked. Data gathered over 4 days for 69/74 children were presented.
Only partial data were available for nine of these children. The two groups had broadly similar
characteristics at the start of the study although compared with the maize–cow pea diet group
the children receiving the soy formula were more dehydrated at admission (P = 0.08), had lower
weight-for-age z scores (P = 0.08), had lower serum bicarbonate values (P = 0.04) and had a
greater stool output during rehydration (P = 0.01).There were five treatment failures – two were
in the maize–cow pea diet group and three in the soy formula group. The children on the soy
formula consumed more on days 1–6 than the maize–cow pea diet (P < 0.001).The median
duration of diarrhoea was significantly less in the maize–cow pea diet compared with the soy
formula diet (42 hours versus 104 hours; P < 0.001). Mean weight change data were poorly
described and data were presented in graph form only. [EL = 1−]
The third study, 133 conducted in Mexico, recruited male infants aged 5–36 months who had
acute watery diarrhoea (defined as more than three liquid stools per day during the previous
24 hours, over a period under 96 hours) and clinical evidence of dehydration (according to
WHO guidelines). Exclusion criteria were severe systemic infections (for example, pneumonia
or sepsis), breastfeeding, a diarrhoeal episode in the previous 2 weeks or poor nutritional status
(under 2 SD of the weight-for-length reference data from the US National Center for Health
Statistics). ORT (according to WHO guidelines) and subsequent hydration maintenance was given
as necessary prior to children being randomly assigned (permuted block randomisation) to either
a mixed diet of rice, chicken, brown beans, carrots and vegetable oil (n = 44) or soy formula
(n = 45). Lab evaluations were made at 6 and 24 hours and weight and length observations were
made daily. Partial data were available for seven children who remained in the study for less
than 6 days. There were six treatment failures (defined as recurrent dehydration, severe diarrhoea
or prolonged high purging), all of whom were in the soy formula group (P < 0.01) There was
a significant reduction in the median duration of diarrhoea in the mixed diet compared with
the soy group (25 hours versus 67 hours; P < 0.001). Although, at day 1, infants were heavier
in the mixed diet group than the soy group, by day 2 and up to day 6 there was no statistically
significant difference in weights between the two groups. [EL = 1−]

Evidence summary
Seven RCTs were examined but only three of them were of fair quality [EL = 1+]. All the trials
recruited fewer than 100 subjects and they compared a wide range of mixed diets based on rice,
lentils, porridge, soy and/or maize. No single solid diet or composition of solid diet in terms of
carbohydrate, protein or fat was shown to be more effective than another for the post-rehydration,
maintenance phase for children with acute diarrhoea. The diets examined were specific to the
various regions in the world. Although these diets conferred no advantage, they did not have
any harmful effects. However, the porridge diets suggested that a thinner food consistency aids
consumption and acceptability. In the studies with soy formula, the data suggested that the soy
formula may prolong the duration of diarrhoea compared with solid foods.

85