Identifying an appropriate biomarker for the tris (2-chloropropyl) phosphate (TCPP) flame retardant in children

Abstract

Background: Tris(2-chloropropyl) phosphate (TCPP) is an organophosphate flame retardant frequently applied to polyurethane foam. Bis(2-chloropropyl) phosphate (BCPP) is typically measured as the primary urinary biomarker for TCPP, but it is infrequently detected in adult urine and thus often considered unreliable. However, BCPP is frequently detected in urine samples from younger children, suggesting that it could still be effective for monitoring children’s exposure to TCPP. More recently, bis(2-chloro-isopropyl) hydroxy-isopropyl phosphate (BCIPHIPP) was identified as another TCPP metabolite and is considered to be an improved biomarker compared to BCPP due to its high detection in urine from all age groups. Methods: We compiled individual urinary biomarker data from children ages 1-73 months recruited from two U.S. cohorts in our laboratory from September 2014 to January 2018 (n=314). Individual concentrations of BCPP and BCIPHIPP as well as their ratio were examined for changes based on child’s age and demographic characteristics. Results: Children’s ages were significantly associated with the BCIPHIPP:BCPP ratio, with each month increase in age being associated with a 0.11 increase in the ratio (95% CI: 0.08, 0.13; p<0.001). Patterns appeared to be driven by a decrease in urinary BCPP concentrations with age. Increased exposure to TCIPP via breastfeeding may contribute to the higher BCPP observed among the youngest children. Alternatively, metabolism differences during the first few years of life, particularly as they relate to child’s race/ethnicity, may explain observed patterns. Also, children with a parent who graduated from college had a lower ratio (ß = -7.3, 95% CI: -11.4, -3.3; p=0.0004). Conclusion: These results suggest that current analyses utilizing only BCPP as a biomarker could severely underestimate individual exposures to TCPP. Further, this study could have broader implications for biomonitoring as these results demonstrate that the selection of an appropriate exposure biomarker could change based on demographics of the study population.