Association of polycyclic aromatic hydrocarbon metabolite concentration with sleep quality in workers

Abstract

BackgroundSleep is a crucial physiological activity for the human body, and research has shown that air pollution can affect sleep quality. However, the association between polycyclic aromatic hydrocarbons (PAHs) exposure, neurotoxic compounds in air pollutants, and sleep quality remains uncertain. ObjectiveTo evaluate the association of PAHs exposure with sleep quality, and to provide evidence for improving sleep quality. MethodsThis study used a cross-sectional design. We selected 632 workers from a coking plant of a large state-owned enterprise as the exposure group, and 477 workers from the energy and power plant of the same enterprise as the control group. All workers worked in three shifts. A questionnaire survey was conducted to collect basic information including gender, years of service, age, educational level, smoking, alcohol consumption, consumption of fried foods, cooking frequency, types of cooking fuels. Worker's post-shift morning midstream urine was sampled to determine the concentrations of eight PAHs metabolites (OH-PAHs) using gas chromatography-tandem mass spectrometry (GC-MS). Worker's sleep quality was assessed using Pittsburgh Sleep Quality Index (PSQI). A higher PSQI score indicated a lower sleep quality. Associations of urinary OH-PAHs levels with sleep quality in the workers were analyzed using linear regression, Bayesian kernel-machine regression (BKMR), and quantile g-computation. ResultsThe median (P25, P75) concentration of total OH-PAHs in the exposure group [88.84 (46.27, 151.96) μg·L−1] was higher than that in the control group [54.33 (24.86, 97.97) μg·L−1]. Additionally, the PSQI score (\begin{document}$ \overline{x}\pm {s} $\end{document}) in the exposure group (5.16±3.84) was higher than that in the control group (4.60±3.17). The multiple linear regression revealed that an increase in the sum of the concentrations of eight OH-PAHs after natural logarithmic transformation (lnΣ8OH-PAHs) was associated with an increase of 0.3646 (95%CI: 0.1337, 0.5955) in PSQI score, and an increase in lnΣlow-ring OH-PAHs was associated with an increase of 0.2954 (95%CI: 0.0941, 0.4967) in PSQI score. The BKMR analysis demonstrated that PSQI score was gradually increased as the increasing of lnΣ8OH-PAHs concentration. The quantile g-computation analysis indicated that a quantile increase in lnΣ8OH-PAHs concentration was associated with an increase of 0.4062% (95%CI: 0.1176%, 0.6949%) in PSQI score. ConclusionCompared to the controls, the coking workers show a higher concentration of urinary OH-PAHs and report worse sleep quality. The concentration of OH-PAHs is significantly negatively associated with sleep quality.

Keywords

• polycyclic aromatic hydrocarbon
• hydroxyl metabolite
• sleep quality
• bayesian kernel machine regression
• quantile g-computation