Short-term variability of bisphenols in spot, morning void and 24-hour urine samples

Due to worldwide regulations on the application of the high production volume industrial chemical bisphenol A (BPA) in various consumer products, alternative bisphenols such as bisphenol F (BPF) and bisphenol S (BPS) are increasingly used. To assess human exposure to these chemicals, biomonitoring of urinary concentrations is frequently used. However, the short-term variability of alternative bisphenols has not been evaluated thoroughly yet, which is essential to achieve a correct estimation of exposure. In this study, we collected all spot urine samples from ten healthy adults for five consecutive days, and an additional 24 h pooled sample. We measured the concentrations of seven bisphenols (BPAF, BPF, BPA, BPB, BPZ, BPS and BPAP) in these samples using gas chromatography coupled to tandem mass spectrometry. BPA, BPF and BPS were frequently found in spot samples (>80%), while bisphenol AP (BPAP) was detected in 43% of spot samples. Calculations of intra-class correlation coefficients (ICCs) showed that reproducibility of these four bisphenols was relatively poor (<0.01-0.200) but improved when concentrations were corrected for urine dilution using creatinine levels (0.128-0.401). Of these four bisphenols, BPF showed the best reproducibility (ICC 0.200-0.439) and BPS the most variability (ICC <0.01-0.128). In general, the within-participant variability of bisphenol levels was the largest contributor to the total variance (47-100%). We compared repeated first morning voids to 24 h pooled urine and found no significantly different concentrations for BPA, BPF, BPS, or BPAP. Levels of BPA and BPF differed significantly depending on the sampling time throughout the day. The findings in this study suggest that collecting multiple samples per participant over a few days, in predefined time windows throughout the day, could result in a more reliable estimation of internal exposure to bisphenols.