To Eat or Not to Eat? The Guts of PFAS-Related Fish Advisories
Erin L. Pulster, U.S. Geological Survey; Chris McCarthy, Jacobs; Courtney Carignan, Michigan State University; and Jonathan Petali, New Hampshire Department of Environmental Services
Science-based environmental standards and guidance for per- and polyfluoroalkyl substances (PFAS) vary across regulatory jurisdictions from non-existent to conservative and are constantly changing with emerging science. Each state develops its own fish consumption advisories (FCAs), resulting in significantly different advisories across states. Some of the factors contributing to the various FCAs include the use of different toxicity endpoints and the wide range of measured perfluoro-octane sulfonic acid (PFOS) levels (2–200 ppb) in edible fish tissue that trigger state advisories.
At the SETAC North America 43rd Annual Meeting in Pittsburgh, the chairs of the session with the title “To Eat or Not to Eat? The Guts of PFAS Related Fish Advisories,” began by setting the foundation for generating recommendations for a path towards closing data gaps, achieving consistency and communicating risk from fish consumption across North America. Erin Pulster outlined the motivation for the session, highlighting the variation in FCAs and the challenge this presents to regulators and consumers. This was followed by Jonathan Petali, who defined common terminology and processes for establishing FCAs in the United States. The goals for the session and of audience discussion were then provided by Chris McCarthy.
Interest in the session was strong with a steady attendance of approximately 140 in-person participants throughout the morning, in addition to at least 20 virtual participants. Engaging comments and questions followed each platform talk. The session brought together expert chemists, toxicologists, risk assessors and epidemiologists from government, business and academia. Presenters shared their experiences with the aim of improving the understanding of how the evolving science of PFAS impacts fish consumption advisories and exploring the results of their innovative research to better inform risk assessment and management.
Generating Meaningful Fish Tissue Measurements of PFAS
The first two technical presentations focused on issues related to PFAS analysis in aquatic biota samples. Heidi Pickard, Harvard University, highlighted the advantages of using a suite of analytical tools to assess the full extent of PFAS contamination and potential bioaccumulation. Pickard discussed the applicability of using extractable organofluorine (EOF) analysis, target analysis, nontarget analysis, total oxidizable precursor (TOP) assay and Bayesian statistics to better understand the presence of precursors and their bioaccumulation potential. These tools revealed that many precursors below detection in water have a higher bioaccumulation potential than their terminal products, thus suggesting water is not a good proxy for predicting levels in fish. While many of these compounds lack toxicological data to evaluate human health risk, some may degrade to terminal PFAS that are regulated in certain areas. Jacqueline Bangma, U.S. Environmental Protection Agency, then presented analytical challenges and solutions for natural interferants that can mimic PFAS on lower resolution analytical equipment resulting in false positives. Bangma reaffirmed Pickard’s message of utilizing an analytical toolbox, in this case low- and high-resolution mass spectrometry (LRMS, HRMS), to confirm and validate the identification of PFAS and rule out interfering compounds. Understanding which PFAS have had previously detected interfering compounds (e.g., PFPeA, PFBA, PFHxS) will assist in determining whether samples warrant the use of additional tools to eliminate false positives and the overestimation of concentrations. These two presentations stress the importance of using a variety of analytical tools to confirm and validate precursor and terminal PFAS while being aware of interfering compounds and false positives. Applying various tools will help more accurately estimate the amplitude of PFAS contamination, concentration levels, compositional profiles and human health risk assessments.
The subsequent three presentations identified the importance of where and when fish are sampled for determining FCAs. Session chair and presenter, Pulster, USGS, presented PFAS levels and profiles in fish and sediments from Tampa Bay, Florida, and the potential human health implications. This presentation highlighted the importance of being cognizant of adjacent land use patterns, which may help to elucidate potential sources and areas of concern. PFAS levels and compositional profiles varied by site and species. This insight can further help select areas to conduct human health risk assessments from the consumption of contaminated fish. Site- and species-specific considerations are also important as they can both highly influence the fish tissue concentrations that the human health risk estimations are reliant upon.
Chris Salice, Towson University, presented the variability that can result when estimating PFAS concentrations in fish tissue from limited data sets. Salice pointed out the shortcomings of making risk management decisions based on the results of screening level assessments with surface water ecological effects thresholds. This practice does not consider bioaccumulation in fish and can underestimate potential human health risks and risk to piscivorous wildlife. He also noted the importance of spatial and temporal variations in PFAS concentrations. Temporal variations (e.g., seasonal, post-storm or rain events, etc.) could result in large disparities in concentrations and bioaccumulation factors for the same system. The findings suggest limited small sampling events can yield misleading results. As such, when conducting site evaluations, it is not recommended to solely rely on surface water thresholds as proxies for ecological thresholds or endpoints. Alternatively, it is recommended to (1) conduct routine sampling of biota, (2) incorporate spatial and temporal sampling schemes, and (3) consider sources and environmental influences (e.g., precipitation, drought, hydrodynamics).
Tom Danielson, Maine Department of Environmental Protection, presented a study evaluating the bioaccumulation potential of stocked brook trout in PFAS-contaminated ponds and concentration differences between skin-on versus skinless fillets. Results demonstrated skin-on fillets typically had more PFOS than skinless fillets. Additionally, this study showed that fish can accumulate PFOS quickly in waterbodies with high concentrations of PFAS, resulting in PFOS concentrations 10 times greater than the state’s fish tissue action level after one week.
Improving Risk Assessment Equations
The numeric heart of FCAs is a risk assessment equation comprising three terms used to estimate a tissue concentration of a chemical, in this case PFAS, that triggers an advisory. These three variables are a toxicity effect value (i.e., reference dose, RfD), body weight, and fish consumption rate, all of which are reviewed in a recent document published by the Interstate Technology and Regulatory Council. The toxicity values for multiple PFAS are constantly evolving, which contributes to variation of PFAS FCAs and other regulatory guidance among states and between states and federal agencies.
In the final two presentations, Megan Romano, Dartmouth College, and Evelyn Reátegui-Zirena, GSI Environmental Inc., presented different assumption factors that influence the derivation of FCAs. Romano discussed the implications of fish tissue consumption patterns and the importance of integrating surveys into study designs to determine region-specific fish consumption rates more accurately for inclusion in assessment calculations. Reátegui-Zirena reiterated the importance of region-specific surveys to determine site-specific consumption rates. Reátegui-Zirena presented data from the literature stressing the importance of fish preparation and cooking methods on changes in PFAS concentrations. Although there is limited evidence, the available studies demonstrated the significant variability in increases and/or decreases in PFAS concentrations between marine and freshwater species, different cooking methods, and between long- and short-chain PFAS. Therefore, Reátegui-Zirena suggests that adjustment factors related to preparation and cooking methods may need to be incorporated in risk assessments due to the potential changes in PFAS concentrations.
Throughout the session, there were active discussions regarding the complexity of PFAS FCAs that require the involvement of regulators, academic researchers and industry professionals. Presentations highlighted challenges with aspects of study designs (e.g., analytical chemistry, sample planning or risk assessment) underscoring the necessity for interdisciplinary collaborative efforts to develop robust and meaningful FCAs for the protection of human health. The list of PFAS analytes continues to grow with technological advancements. However, the lack of analytical standards and toxicity data for newly identified PFAS make it difficult to quantify and assess risk. Nevertheless, knowledge of where (e.g., potential PFAS sources) and when (seasonality) to sample can help determine target analytes. In addition, field staff need to be aware of the relevant species, tissues and sample sizes needed to provide useful data for generating FCAs. It is also critical to document current limitations of human health risk assessment for describing FCAs, such as the uncertainty related to the growing list of novel and precursor PFAS that can be detected in fish tissues.
Some key recommendation for improving FCAs in the future include:
- Improving risk communication around uncertainty and evolving PFAS science
- Harmonizing relevant PFAS toxicity thresholds appropriate for risk assessment
- Utilizing a suite of analytical tools to assess total fluorine, target and non-target analytes
- Incorporating spatiotemporal considerations into sampling designs
- Implementing regional surveys to update fish consumption data
- Incorporating potential preparation and cooking method risks
- Increasing PFAS testing of recreationally and commercially important species
This was a successful forum that covered the key areas of epidemiology, analytical challenges, confounding factors, study designs and policy used in developing PFAS-related fish advisories. Attendees gained familiarity with ongoing and new PFAS research, awareness of the interdisciplinary nature needed to develop FCAs, and highlighted critical data gaps needed to inform resource managers and risk assessors. Results of the research associated with many of these presentations have either recently been published or are currently in review emphasizing the importance and demand for this research.