February 25, 2020

PFAS Hazard Characterization: Inhalation Exposure

Travis R. Kline

Perhaps the single biggest challenge confronting us in our understanding and management of per- and polyfluoroalkyl substances (PFAS) is the overall scarcity of defensible toxicological information. Such toxicological information is the basis for understanding the potential for chemical health impacts, which in turn defines which chemicals are at issue for environmental, occupational, and consumer product–related exposures. These toxicological implications also define the level of sensitivity required in their analytical assessment and testing. Health protection is normally the starting point for assessing remedial goals or the level of inherent risk we are willing to accept, either from the contaminants in situ or following remediation. After considering health protection, other factors such as technical achievability and cost can then be considered. Although toxicological testing of PFAS continues, very few state or federally promulgated PFAS toxicological criteria are available and these criteria are only available for a tiny fraction of the more than 4,000 PFAS that have been produced and used in global commerce.

When addressing emerging contaminants, state and federal environmental programs tend to focus on drinking water exposures; however, ambient air inhalation–based PFAS hazards have historically been equivalent to drinking water exposures especially for communities adjacent to production facilities. Significantly, there are no state or federally promulgated toxicity criteria to support inhalation exposure assessment: All of the available PFAS toxicological criteria are ingestion-based (e.g., U.S. Environmental Protection Agency (EPA) oral reference doses (RfDo) or Agency for Toxic Substances and Disease Registry (ATSDR) minimal risk levels (MRLs)). Based on the voluntary phaseout of widely used commercial PFAS compounds (such as perfluorooctanoic acid (PFOA) and perlfuorooctanesulfonic acid (PFOS)) in the United States and implementation of control technologies, plant emissions have been significantly reduced and PFAS blood serum levels are declining. However, U.S. worker populations at production facilities were associated with blood serum levels hundreds of times greater than the general population and these exposures were driven by inhalation of fugitive and stack emissions, not drinking water. Still, inhalation exposures persist, whether through stack and fugitive emissions or through disassociation from consumer products in the home or at work, with subsequent particulate inhalation. Compounding the potential for inhalation exposure, some PFAS compounds exhibit physicochemical properties classifying them as volatile organic compounds with implications for exposure pathways to indoor air from groundwater sources (i.e., vapor intrusion).

To date, hazard characterizations of PFAS compounds have been limited to ingestion-based assessments (e.g., incidental ingestion of soil, drinking water, and dietary exposures). Recognizing that inhalation is a significant component affecting worker and residential exposures, omission of this pathway represents a data gap and undermines effective risk and site management decision-making.

Assessing Route-to-Route Extrapolation for Inhalation Exposure

EPA, in important risk assessment guidance documents such as the Vapor Intrusion Guidance, has employed a common convention whereby a toxicologist may develop provisional toxicity criteria referred to as route-to-route extrapolation. Via this process, a toxicologist can start with an ingestion-based criterion and adjust for certain pathway and exposure parameter values (e.g., inhalation rate, body weight) to develop a provisional inhalation criterion, in concert with the basis for EPA’s inhalation-based reference concentration (RfC). Development of this type of provisional value provides site managers with a basic understanding of the potential implications for inhalation exposure and a basis for an assessment of relative hazard contribution in comparison to other complete exposure pathways. Use of this, admittedly provisional, value in quantitative expression of hazard can help support decisions about where to commit available resources to affect the greatest good for a particular site or condition (e.g., fish ingestion, drinking water, or in-home ambient air exposures).

In order to assess whether development of provisional toxicological values for PFAS using a route-to-route extrapolation is appropriate, one must first determine whether there is any condition or phenomenon that could contraindicate such an approach. The following facts are relevant to this assessment:

  • Regardless of administration route, very similar PFAS distribution percentages are evident in the same target organ systems (e.g., blood, liver, and extravascular tissues). With the same target systems subject to impact in the body, we do not expect substantive differences in elicited toxic response based on ingestion versus inhalation exposure.
  • PFAS are readily absorbed via the gastrointestinal system, with efficiencies for a compound like PFOA in the 95 to 99 percent range. No definitive fractional absorption studies are available for the inhalation route of exposure, but PFAS are recognized as readily absorbed following inhalation administration. Based on our current understanding, there is no indicated need for a quantitative adjustment predicated on absorption efficiency.
  • There is low evidence of substantive portal-of-entry effects, specific to the respiratory system. There is some evidence of low-grade lung irritation; however, this effect was elicited in both inhalation and ingestion studies and so may be discounted as a contraindication.
  • PFAS compounds are very poorly metabolized in the human body and are slow to be eliminated. Subsequently, PFAS compounds have a long half-life (t1/2) in humans, with residency times for compounds like PFOS spanning the six- to nine-year time frame. The human body does a good job of detoxifying most toxicants we contact over the span of a routine day. In some instances, metabolization in the body can yield metabolites with heightened toxicological implications. No contraindication is evident based on a very low potential for metabolism and the unlikely buildup of toxic metabolites in different target systems based on differing routes of administration (i.e., ingestion versus inhalation).

Based on our current understanding of PFAS toxicokinetics, there does not appear to be any contraindication to employing route-to-route extrapolation in the derivation of a provisional inhalation toxicological criterion for individual PFAS, where accepted and promulgated ingestion-based criteria are available

Derivation of Screening Levels for PFOA and PFOS

The most widely studied of these compounds, PFOA and PFOS, are considered as extrapolation examples in Table 1. Table 1 presents the conversion of ingestion-based toxicity criteria from the EPA and ATSDR, a federal public health agency of the U.S. Department of Health and Human Services, to develop target inhalation doses. The EPA RfDo is an estimate, with uncertainty spanning perhaps an order of magnitude, of a daily oral exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime. The ATSDR MRL is that agency’s ingestion-based toxicity criterion. In this case, the PFOA and PFOS MRLs are “intermediate” MRLs (MRLint), protective of sensitive individuals among the human population based on exposures ranging between 14 and 364 days. The EPA- and ATSDR-based target inhalation doses are calculated and presented in Table 1.

Table 1:  Conversion of Ingestion-based Toxicity Criteria 

Table 1: Conversion of Ingestion-based Toxicity Criteria
Compound Adult Body Weight (kg) USEPA RfDo  (mg/kg/d)
USEPA Target Inhalation Dose    (ug/day) ATSDR MRLint  (mg/kg/d) 
ATSDR Target Inhalation Dose (ug/day)
PFOA 70 2.00E-05 1.40E+00 3.00E-06 2.10E-01
PFOS 70 2.00E-05 1.40E+00 2.00E-06 1.40E-01

Based on the conversion in Table 1 from ingestion to inhalation criteria, Table 2 presents the derivation of residential ambient air screening levels (SLs) predicated on EPA’s RfDos and ATSDR’s MRLints. 

Table 2: Derivation of Residential Ambient Air Screening Levels

Table 1: Conversion of Ingestion-based Toxicity Criteria
Compound Adult Inhalation Rate (m3/d)
USEPA Target Inhalation Dose    (ug/day)
USEPA-Based Air SL (Residential) (ug/m3)
ATSDR Target Inhalation Dose (ug/day)
ATSDR-Based Air SL (Residential) (ug/m3)
PFOA 20 1.40E+00 .07 2.10E-01
PFOS 20 1.40E+00 .07 1.40E-01

Consistent with the original derivation of the toxicological criteria, the ambient air SL values are predicated on an adult bodyweight of 70 kg (although current EPA guidance suggests the use of 80 kg) and a daily inhalation rate of 20 m3. It is interesting to note the rather significant differences in the screening criteria, especially given the exposure duration basis (i.e., the EPA RfDos are based on 70 years of exposure, while the ATSDR MRLints are based on a period of less than one year).  These values may be used to screen ambient air empirical data-based exposure point concentrations to help site managers determine whether additional scrutiny is necessary.


In the absence of regulatory agency-derived or promulgated toxicity criteria, the route-to-route extrapolation process can provide site managers with quantitative expressions of human health hazard related to inhalation exposure as a basis for risk management decisions. There is significant uncertainty attendant in any quantitative expression of risk or hazard, but such quantitative point estimates are inherently uncertain and frequently incorporate excess conservatism in the face of this uncertainty. Although there is significant difference in the derived values, based on USEPA and ATSDR toxicity criteria, this range is worthy of consideration. Ideally, quantitative point estimates of risk and hazard should be viewed as a range of values to inform site management, with a lessened degree of reliance on discrete point estimates. This methodology can be employed as new toxicity criteria become available for additional PFAS compounds, although continued assessment of contraindications is critical for these constituents of emerging concern.


    Travis R. Kline

    Travis R. Kline is a Senior Principal Toxicologist with Geosyntec Consultants, Inc. in Washington, D.C.