957Household Products with Triclosan and Triclocarban Reduce Infectious Disease Symptoms in Newborns - Preliminary Data from an Ongoing Randomized Trial.

Background. Triclosan and triclocarban (TCs) are broad spectrum microbicides found in commercial household and personal cleaning products (HPCP). Effective against gingivitis, TCs have not been proven to protect against other infectious diseases. The FDA, which is reviewing TCs, has requested further information on their efficacy. Methods. Stanford’s Outcomes Research in Kids (STORK) is a rolling cohort of pregnant women and their babies followed to age 36 months. STORK includes a nested, randomized intervention of HPCP with and without TCs to assess the effect of TCs on reported infection. Mothers report daily infectious diseases symptoms in babies using a weekly survey. We compared cumulative infection rates (reported days sick/ total days followed) in each intervention arm. Urinary triclosan levels were measured in a subset of 20 babies. Analyses were blinded to intervention arm; Chi-square test or quasi-Poisson regression were used as appropriate. Results. A total of 119 babies have been born with 52 (43.7%) and 63 (52.9%) randomized to the TC and non-TC arms, respectively (4 were not randomized). As of April 2014, babies had been followed for a median of 48 weeks (range: 3-135 weeks). Breast feeding rates were similar in both arms (p = 0.56). The cumulative infection rate was different between arms (8.3% and 12.3% in the TC and non-TC arms, respectively; p < 0.001) as was themedian infection rate (i.e., per baby) (4.4% vs 7.7%, p = 0.07). Among sick babies, the proportions of days of cold, cough, ear pulling/tugging, diarrhea and fever did not differ significantly between arms; the proportion of vomiting days was less in the TC arm (7.2 vs 11.1%, p < 0.001). Urinary triclosan was detected in 7 (88%) of 8 breastfed babies in the TC arm (mean: 125 pg/ml). While these levels are lower than those seen in TC-using adults (data not shown), they are significantly higher (p = 0.02) than levels in non-breast fed babies in the TC arm (N = 2), babies in the non-TC arm (N = 7) and babies who were not randomized (N = 3) (19.4, 11.7 and 8.8 pg/ml). Conclusion. In this preliminary analysis, residence in a household randomized to TC-containing HPCP is associated with a lower incidence of infectious disease symptoms in infants. Low levels of TC exposure in infants appear to derive from breastfeeding. Disclosures. All authors: No reported disclosures.