Gestational triclosan exposure and infant birth weight: A systematic review and meta-analysis.

ABSTRACT

BACKGROUND: Exposure to triclosan, an antimicrobial chemical used in some personal care and cleaning products, has been associated with reduced birth weight in some, but not all epidemiological studies. OBJECTIVES: We conducted a systematic review and meta-analysis to characterize the relation of gestational triclosan exposure with infant birth weight and identify sources of heterogeneity between studies. METHODS: We identified original studies measuring urinary triclosan concentrations during pregnancy and reporting their association with infant birth weight, gestational age (GA) adjusted birth weight (g), or GA-standardized birth weight z-scores. Using a random effects model, we estimated differences in these outcomes per 10-fold increase in triclosan concentrations and considered triclosan levels and infant sex as sources of heterogeneity. Using Navigation Guide Methods, we evaluated risk of bias within individual studies and across the body of evidence. RESULTS: Among thirteen studies, median triclosan concentrations varied by almost 2-orders of magnitude (0.6-29 ng/mL), with higher concentrations in North American and some European studies compared to Asian ones. Associations between triclosan and birth weight (ß-20 g; 95% CI-65, 26; n = 6) were stronger than those for GA-adjusted birth weight (ß-12 g; 95% CI-29, 5; n = 9). Triclosan was not associated with GA-standardized birth weight z-scores (ß-0.04; 95% CI-0.16, 0.07; n = 5). The association between triclosan and GA-adjusted birth weight was stronger in studies with median triclosan values ≥10 ng/mL compared to studies with median values < 10 ng/mL (ß-27 g; 95% CI-61, 7; n = 4 vs. ß6g; 95% CI-20, 31; n = 5). With a limited number of studies, we observed suggestive evidence that inverse associations were more apparent in studies with ≥ 2 prospective triclosan measures compared to those with one measure. DISCUSSION: Available evidence, with "low" risk of bias, provides limited evidence that triclosan exposure and reduces infant birth weight. We observed stronger inverse associations between triclosan concentrations and birth weight in populations with higher triclosan exposure.