This randomized, double-blind trial determined whether adding 90 mmol/L of alanine with a reduction in glucose to 90 mmol/L (alanine ORS) improves the efficacy of the standard oral rehydration solution (WHO-ORS). One hundred twenty-nine males aged 3-48 months with weight for length greater than or equal to 70% of NCHS, diarrheal duration less than or equal to 96 h, and clinical signs of mild to moderate dehydration were randomly allocated to either treatment group. During 0-6 h of treatment, ORS was offered at 120 ml/kg for rehydration without food or water. Beyond 6 h, ORS was offered as a volume-to-volume replacement for stool losses and a mixed diet of uniform composition was offered in amounts standardized for body weight. The most frequently isolated pathogens in alanine ORS and WHO-ORS groups were rotavirus (42 and 48%, respectively) and enterotoxigenic Escherichia coli (15 and 12%, respectively). In the 0-6 h period when food was withheld, median urine output in ml/kg (8;5, p less than 0.05) and percentage decrease in total serum solids (9:7%, p = 0.06) was significantly greater in alanine ORS than in WHO-ORS; median ORS intake and stool output were marginally lower in the alanine group but the differences were statistically not significant. Between 0 h and recovery, although the median values for duration of diarrhea (56.5 and 65.0 h), ORS consumption (260 and 323 ml/kg), and stool output (188.4 and 216.3 g/kg) were lower in the alanine ORS group, these differences with the WHO-ORS group were not statistically significant.(ABSTRACT TRUNCATED AT 250 WORDS)