Soy Formulations In The Treatment Of Diarrhea

Journal of Family Practice, Nov, 1999 by Michael Temporal, Scott M. Strayer

Fayad IM, Hashem M, Hussein A, Zikri MA, Zikri MA, Santosham M. Comparison of soy-based formulas with lactose and with sucrose in the treatment of acute diarrhea in infants. Arch Pediatr Adolesc Med 1999; 153:675-80.

Clinical question Do infants with an episode of acute diarrhea have better outcomes when fed soy-based formula with sucrose or lactose?

Background The World Health Organization recommends that children continue their normal diets during diarrheal episodes, and recent data support beginning feeding with formula immediately after the completion of rehydration therapy. There is controversy, however, regarding the best type of formula. Studies have confirmed soy-based lactose-free formula as effective in decreasing the stool output and duration of diarrhea. However, there have been conflicting studies on whether feeding lactose-containing formula or cow's milk aggravates diarrhea.

Population studied Male infants aged 3 to 18 months who were predominantly formula fed and were brought to an urban university children's hospital in Cairo, Egypt, with diarrhea were included. Infants with symptoms for more than 7 days and those without clinical signs of dehydration were excluded. Other exclusion criteria included visible blood in the stool, evidence of systemic infection, severe malnutrition, and weight below 70% of the median for that age. Boys were chosen for the convenience of being able to separate urine from stool output.

Study design and validity This was a randomized clinical trial. After initial intravenous hydration to stabilize vital signs and after tolerating oral rehydration for 3 to 12 hours, 200 infants were randomized to a soy-based formula for which the carbohydrate source was either sucrose (Nursoy ready-to-feed, Wyeth International) or lactose (Nursoy powder, Wyeth International). Infants received 150 mL/kg of their assigned formula in 8 equal feedings per day until the resolution of diarrhea for 15 hours or until they had received formula for 7 days. The study had adequate power to detect differences in stool output and in the duration of diarrhea, and intention-to-treat analysis was appropriately used. The sucrose and lactose formula populations had comparable baseline characteristics.

Outcomes measured The primary outcomes measured were stool output (first 24 hours and total) and duration of diarrhea in hours. Other outcomes were measured but are of unclear clinical and statistical significance because of the number of variables analyzed.

Results The Kaplan-Meier method was used to compare the duration of diarrhea in the 2 treatment groups. Log-rank testing confirmed that the duration of diarrhea in infants fed soy-based formula with sucrose was significantly shorter, with a mean reduction of 16 hours (23 vs 39 hours; 95% confidence interval [CI], 9 - 24 hours). Stool output during the first 24 hours (103 vs 57 gm/kg; relative risk = 1.8; 95% CI, 1.4 - 2.3) and total stool output (164 vs 69 gm/kg; relative risk = 2.4; 95% CI, 1.7 - 3.3) were also significantly higher in the group fed soy-based formula with lactose.

Recommendations for clinical practice This study provides good evidence that soy-based formulas with sucrose can shorten the duration of acute diarrheal episodes in infant boys by almost 1 day and will lead to decreased stool output. The results may not apply to breast-fed infants or to girls. Future studies on this issue may be more applicable to primary care practice and should determine the outpatient efficacy of sucrose-containing soy-based formulas in reducing clinically significant dehydration and preventing hospitalizations.

Michael Temporal, MD Scott M. Strayer, MD St Louis University Family Practice Residency Belleville, Illinois

Note: The opinions and assertions contained in this review are the private views of the authors and are not to be construed as official or as reflecting the views of the United States Air Force or the armed services at large.