Background: Recent studies have shown that probiotics, most commonly Lactobacillus GG, may be useful in treating acute gastroenteritis. However, beneficial effects appear to be limited to a modest decrease in the duration of diarrhea. No studies have evaluated this therapy in moderate to severe dehydrating diarrhea in a metabolic facility. Methods: Male children less than 2 years of age were admitted to a metabolic unit of the Department of Pediatrics at the Federal University of Bahia, Brazil, with moderate dehydration and were randomized in a double-blind, placebo-controlled fashion. Oral rehydration solution (ORS) was administered per protocol and either placebo or Lactobacillus GG was given in combination with the ORS. Output of urine, stool, and vomitus was recorded along with stool weight, nude body weight, and standard laboratory assessments for hydration. Results: There was no significant reduction in diarrhea duration and stool output in the Lactobacillus GG group. However, Kaplan-Meier survival analysis demonstrated that, even in moderate to severe diarrhea, resolution of the illness occurred so rapidly, that statistically significant benefits of probiotic therapy could not be demonstrated. Conclusion: Our data implies that colonization must occur before benefits of probiotics can be realized. Probiotics are, therefore, likely to be of limited benefit in treating diarrheal illnesses of short duration such as viral enteritis. The beneficial effects of probiotics may be limited to prophylactic usage in high-risk populations.