Plain language summary pending.
Rice-based oral rehydration appears to be effective in reducing stool output in people with cholera. This effect was not apparent in infants and children with non-cholera diarrhoea.
Oral rehydration therapy is used to treat dehydration caused by diarrhoea. However the rehydration solution does not reduce stool loss or length of illness. A solution able to do this may lessen the use of ineffective diarrhoea treatments as well as improve morbidity and mortality related to diarrhoea.
The objective of this review was to assess the effects of rice-based oral rehydration salts solution compared with glucose-based oral rehydration salts solution on reduction of stool output and duration of diarrhoea in patients with acute watery diarrhoea.
We searched the Cochrane Infectious Diseases Group trials register, the Cochrane Controlled Trials Register, Medline, Embase, Lilacs and the reference lists of relevant articles. We also contacted researchers in the field.
Randomized trials comparing standard World Health Organization oral rehydration solution with an experimental oral rehydration salts solution in which glucose (20 grams per litre) was replaced by 50-80 grams per litre of rice powder, with the electrolytes remaining unchanged.
Data were extracted independently by a statistician and a clinician.
Twenty-two trials were included. Concealment of allocation was adequate in 15 of these trials. Irrespective of age, people with cholera who were given rice oral rehydration salts solution had substantially lower rates of stool loss than those given oral rehydration salts solution in the first 24 hours. Mean stool outputs in the first 24 hours were lower by 67 millilitres/kg of body weight (weighted mean difference -67.40, 95% confidence interval -94.26 to -41.53) in children, and by 51 millilitres/kg of body weight (weighted mean difference -51.07, 95% confidence interval -65.87 to -36.27) in adults. The rate of stool loss in infants and children with acute non-cholera diarrhoea was reduced by only four millilitres/kg of body weight (weighted mean difference -4.29, 95% confidence interval -9.36 to 0.78).