Urinary metabolites of organophosphate flame retardants: temporal variability and correlations with house dust concentrations.

A reduction in the use of polybrominated diphenyl ethers (PBDEs) because of human health concerns may result in an increased use of and human exposure to organophosphate flame retardants (OPFRs). Human exposure and health studies of OPFRs are lacking. We sought to define the degree of temporal variability in urinary OPFR metabolites in order to inform epidemiologic study design, and to explore a potential primary source of exposure by examining the relationship between OPFRs in house dust and their metabolites in urine. Nine repeated urine samples were collected from 7 men over the course of 3 months and analyzed for bis(1,3-dichloro-2-propyl) phosphate (BDCPP) and diphenyl phosphate (DPP), metabolites of the OPFRs tris(1,3-dichloro-2-propyl) phosphate (TDCPP) and triphenyl phosphate (TPP), respectively. Intraclass correlation coefficients (ICCs) were calculated to characterize temporal reliability. Paired house dust and urine samples were collected from 45 men. BDCPP was detected in 91% of urine samples, and DPP in 96%. Urinary BDCPP showed moderate-to-strong temporal reliability (ICC range, 0.55-0.72). ICCs for DPP were lower, but moderately reliable (range, 0.35-0.51). There was a weak [Spearman r (r(S)) = 0.31] but significant (p = 0.03) correlation between urinary BDCPP and TDCPP concentrations in house dust that strengthened when nondetects (r(S) = 0.47) were excluded. There was no correlation between uncorrected DPP and TPP measured in house dust (r(S) < 0.1). Household dust may be an important source of exposure to TDCPP but not TPP. Urinary concentrations of BDCPP and DPP were moderately to highly reliable within individuals over 3 months.