Assessing Exposure to Per- and Polyfluoroalkyl Substances in the Indoor and Ambient Environment Utilizing Silicone Wristbands

Abstract

Per- and polyfluoroalkyl substances (PFASs) are a large class of synthetic compounds produced since the 1940s for their oil and water repellant properties. The strong carbon-fluorine bonds that comprise the backbone of many PFASs contributes to their unique repellant properties but also makes many PFASs recalcitrant to degradation. Due to their widespread use and recalcitrance, humans are ubiquitously exposed to PFASs both indoors and outdoors. Exposure primarily occurs via consumption of contaminated drinking water and food; however, individuals can also be exposed dermally and via inhalation, especially indoors. Furthermore, it is known that some consumer products such as furniture/textiles, carpets and paint have been treated with PFASs, which can potentially off-gas into indoor air. What is currently unclear is the extent to which such sources of PFASs exposure contribute to overall exposure, particularly in indoor residential environments. This knowledge is important to risk assessment and management, especially because humans spend most of their time in some type of indoor environment. Furthermore, some of the most frequently detected PFASs in the indoor environment are polyfluoroalkyl precursors which can potentially break down into the perfluoroalkyl acids such as PFOA and PFOS, which have been associated with adverse health effects. However, these precursors are understudied compared to the perfluoroalkyl acids and the available exposure data are from bulk samples (e.g. air or dust) which do not provide information on behaviors or actions taken on a personal level that contribute to exposure.The overarching goals of this dissertation were to 1) characterize exposure to both volatile and non-volatile PFASs utilizing silicone wristbands, 2) determine whether indoor air, indoor dust, or silicone wristbands best predicts blood PFASs levels in the general population and 3) use silicone wristbands to better understand ambient exposures to PFASs in a firefighter cohort, especially focusing on potential occupational-specific exposures. As mentioned previously, personal-level data on exposure to volatile PFASs is lacking and the goal of this dissertation was to create a method to extract and analyze volatile PFASs from silicone wristbands to address this data gap both for a general population and an occupationally exposed population (i.e. firefighters). In Chapter 2, we developed and validated an analytical method for measuring volatile PFASs in silicone wristbands via gas chromatography high resolution mass spectrometry (GC-HRMS). We applied this method, along with our previously published method for quantifying non-volatile PFASs via liquid chromatography tandem mass spectrometry (LC-MS/MS), to a set of wristbands collected from a cohort of individuals living in a PFAS-impacted community in Michigan. Paired samples of silicone wristbands and blood serum from 87 individuals were analyzed for PFASs. Variations in PFASs levels measured in silicone wristbands were found to vary based on demographics and behaviors such as cosmetics use, and the time spent inside the home. Correlations were also evaluated between the most frequently detected PFASs in silicone wristband and blood serum. A strong and statistically significant correlation was observed between MeFOSE in the wristbands and N-MeFOSAA in blood serum, suggesting that exposure to this PFAS was primarily via inhalation and/or dermal exposure. Taken together, these results suggest that wristbands can provide individual level data on exposure to some polyfluoroalkyl precursors present indoors that reflect serum levels of their suspected biotransformation products. In Chapter 3, a subset of the previous population agreed to indoor air and dust sampling in their homes. This study focused on 46 individuals living in 32 different homes which had paired samples of silicone wristbands, blood serum, indoor air and indoor dust data. Indoor air was analyzed for 14 PFASs via GC-HRMS and 31 PFAS via LC-MS/MS. Indoor dust was analyzed for 13 PFASs via GC-HRMS and 43 PFASs via LC-MS/MS. Survey data was utilized to investigate associations between PFASs levels in indoor air and dust with housing characteristics or personal behaviors such as the type of flooring present or cleaning frequency. This chapter also assessed correlations among PFASs levels measured in all matrices to gain a better understanding of the relative importance of inhalation and dust exposures to overall PFASs exposure and to determine which matrix best predicts internal dose. In the case of MeFOSE, results suggest that silicone wristbands and indoor air samples can equally predict blood PFASs levels of suspected biotransformation products, and wristbands may even be a better predictor than indoor air samples but a larger study with more data is needed to confirm these findings. However, silicone wristbands were not able to capture all PFASs exposures in the indoor environment. PFDA and PFUnA in the dust were observed to have a small but statistically significant impact on blood PFASs levels. These latter long chain PFCAs were not frequently detected in the silicone wristbands and suggest that wristbands may not support exposure assessments for these longer chain non-volatile PFAAs. In Chapter 4, the method developed in Chapter 2 was applied to a set of wristbands collected from a cohort of firefighters recruited from Durham, North Carolina to assess firefighters’ ambient (non-dietary) exposure to PFASs. Twenty-nine firefighters were recruited into the study and provided a blood sample at study onset. Twenty-two of the 29 firefighters returned and provided a second blood sample approximately 4 months later. Silicone wristbands were collected from a total of 23 firefighters over the 4-to-5-month study period. PFASs were detected in all blood samples and two short-chain PFAS (TFA and PFPrA) were detected frequently in blood plasma. Significant associations were observed among blood PFASs levels based on demographics and occupational history (e.g. age and years spent as a firefighter). Occupational activities (e.g. reporting to a fire call) were assessed to determine if they were associated with wristband PFASs levels. The time the wristband was at a fire call was found to significantly predicted wristband MeFBSE concentrations. MeFBSE has been identified as a PFAS in firefighter turnout gear, and it is possible this association reflects exposure to PFAS in gear; however, it’s also possible that the exposure stemmed from soot and smoke associated with each fire. Lastly, MeFOSE levels in silicone wristbands were significantly and positively correlated with plasma N-MeFOSAA concentrations, replicating results found in Chapter 2. Data from this chapter suggests that silicone wristbands can provide some relevant information regarding ambient PFASs exposure among firefighters. Overall, this dissertation provides important information regarding personal level, non-dietary, exposures to both volatile and non-volatile PFASs and examined how behaviors affect PFASs exposures utilizing a simple to use silicone wristband as a personal passive sampler. The capability to measure exposure on a personal level is important and allows us to more accurately understand exposure sources and trends compared to bulk measurements (i.e. air and dust).