Single and joint associations of exposure to polycyclic aromatic hydrocarbons with blood coagulation function during pregnancy: A cross-sectional study

Abstract

Association linking polycyclic aromatic hydrocarbons (PAHs) to blood coagulation function during pregnancy remains absent. Hence, we conducted a cross-sectional study including 679 late pregnant women (27.2 ± 5.1 years old) drawn from Zunyi birth cohort, Southwest China. During late pregnancy, ten urinary PAHs metabolites and four clinical blood coagulation parameters were measured, including activated partial thromboplastin time (APTT), prothrombin time (PT), thrombin time (TT), and fibrinogen (FIB). Multiple linear regression, Restricted cubic spline (RCS) regression, Bayesian kernel machine regression (BKMR), and quantile g-computation (Q-g) regression were used to investigate their single, nonlinear, and mixed associations. Each 2.7-fold increment in 2-hydroxyfluorene (2-OHFlu), 9-hydroxyfluorene (9-OHFlu), 1-hydroxyphenanthrene (1-OHPhe), 2-hydroxyphenanthrene (2-OHPhe), and 3-hydroxyphenanthrene (3-OHPhe) were associated with 0.287 s, 0.190 s, 0.487 s, and 0.396 s shorter APTT, respectively; each 2.7-fold increment in 2-OHPhe was associated with a 0.047 s longer PT; each 2.7-fold increment in 9-hydroxyphenanthrene (9-OHPhe) and 1-hydroxypyrene (1-OHPyr) were associated with 0.087 s and 0.031 s shorter TT, respectively; and each 2.7-fold increment in 1-hydroxynaphthalene (1-OHNap) was associated with 0.032 g/L higher FIB level. The nonlinear association of 2-OHPhe with APTT and 1-OHNap with FIB were also observed. Furthermore, the shortened APTT and TT associated with PAHs mixture were indicated by BKMR and Q-g model. BKMR also revealed a nonlinear association of 2-OHPhe with PT and an interaction effect of 2-OHPhe and 3-OHPhe on APTT. Our results indicate that urinary PAHs was associated with shortened coagulation time and increased FIB. Therefore, more attention should be paid for late pregnant women to prevent PAHs-associated risk of thrombosis. Future perspective studies to confirm our findings and explore the underlying biological mechanism are warranted.

Keywords: Bayesian kernel machine regression; Coagulation; Joint association; Polycyclic aromatic hydrocarbons; Pregnancy; Quantile g-computation.