You are here:
MANAGING EXPOSURES TO NEUROTOXIC AIR POLLUTANTS.
Citation:
BOYES, W. K. AND P. J. BUSHNELL. MANAGING EXPOSURES TO NEUROTOXIC AIR POLLUTANTS. ENVIRONMENTAL MANAGEMENT. Springer-Verlag, New York, NY, , 33, (2006).
Impact/Purpose:
To develop a biologically-based dose-response model to describe the neurotoxic effects of exposure to volatile organic compounds
Description:
Researchers at EPA's National Health and Environmental Effects Research Laboratory are developing a biologically-based dose-response model to describe the neurotoxic effects of exposure to volatile organic compounds (VOCs). The model is being developed to improve risk assessments and management of VOCs, many which are listed as Hazardous Air Pollutants under the Clean Air Act, and used in large volumes in industry and commerce. Among the most common are toluene, used in gasoline, paints, glues and printing; perchloroethylene, used in dry cleaning; and trichloroethylene (TCE), used in metal cleaning and degreasing.
Acute, short-term exposure to volatile organic substances can produce neurological impairments. The level of dysfunction increases with larger doses, and symptoms may include confusion, dizziness, slowed reactions, poor coordination, and sensory, motor and cognitive impairment. The effects of chronic exposure are more uncertain, but if the levels are sufficient, also may cause impairments indicative of central nervous system dysfunction.
The use of biologically-based dose-response models is currently considered the state of the art in integrating toxicological and biological information into risk assessments. These models quantify key processes involved in the toxic response at the level of cells, tissues and whole animals to better predict human risks. An important step in modeling is to determine the dose in the target tissue that is associated with toxicity (see figure). Once a toxic dose is known, the model can be used to estimate toxicity of a hazardous air pollutant under various exposure conditions. These quantitative predictions could replace or refine uncertainty factors now used in risk assessment to account for shortcomings in the available data.
The model of acute VOC neurotoxicity being developed at EPA combines estimates of tissue dose with behavioral and neurophysiological assessments. The dosimetry component is called a physiologically-based pharmacokinetic (PBPK) model, which incorporates physiological and biochemical parameters to describe the absorption, distribution, metabolism and elimination of the compound. A key finding is that blood or brain concentration of the neurotoxicant at the time of assessment can be used as a 'dose metric' to assess the potential risks of a variety of acute exposure conditions.
The model can be applied to address several risk assessment and management scenarios. In one application, researchers contributed to development of Acute Exposure Guideline Level (AEGL) values, which provide emergency planners and responders with important information needed to respond to emergency conditions. AEGL values are typically set for exposure durations of 10 and 30 minutes, and 1, 4 and 8 hours. The model was used to determine air TCE concentrations leading to the same target tissue dose for these exposure durations. Because all the resulting exposure conditions lead to the same internal dose, the AEGL standards at each duration should be equally protective.
Biologically-based dose-response models can also facilitate meta-analysis, referring to the analysis of results from combined individual experiments, in order to increase statistical power and reliability via the enlarged sample size. One difficulty in meta-analysis is that the exposure parameters of individual studies often vary greatly, but information about the dose metric can help resolve this dilemma. For example, the data from several experiments measuring the behavioral effects of acute toluene exposure were combined using the conditions from each study to estimate blood toluene concentrations at the time of assessment. Without conversion to a common dose metric, the data appeared widely disparate. When combined, however, the data revealed a systematic dose-response relationship between blood toluene concentration and behavioral impairment across the multiple individual studies.
A third example of the application of such models concerns risk management, which combines risk assessment with other considerations for regulatory or exposure control purposes. One factor being considered increasingly is the costs of emission controls in comparison to the benefits of the health problems that could be avoided if exposures were reduced. A serious impediment to these cost-benefit analyses is the fact that the monetary value of avoided health problems is very difficult to quantify. For example, how much money does a slowed reaction time from acute toluene exposure cost? Not having answers to such questions might mean that some potential health benefits are not considered. For the case of acute toluene exposure, one approach to address this problem is to compare the effects of toluene to those of ethanol (alcohol), for which there is considerable information regarding the monetary costs of overexposure. A function was created describing the blood concentrations of toluene and ethanol causing equivalent behavioral impairments. Thus, any exposure to toluene could be expressed as an equivalent dose of ethanol. This function may be useful for estimating the monetary consequences of exposures yielding any given blood toluene concentration. This general approach may improve the ability to estimate the monetary benefits of controlling exposures for a number of health outcomes that were not otherwise quantifiable.
If the biologically-based dose-response model for the acute effects of VOCs proves to be useful, the principles it illustrates may be applicable to other risk assessment issues. For example, the acute and long-term effects of exposure to other chemical classes, like pesticides and endocrine-disrupting chemicals, may be amenable to this approach. One principle is already apparent: building such a model has proven difficult and has shown the value of starting with a system in which the toxicology is relatively well understood. Application to systems that are less well understood will be a challenge for the future.
William K. Boyes and Philip J. Bushnell of the EPA's National Health and Environmental Effects Research Laboratory in Research Triangle Park NC contributed this month's column.
For more information on the research discussed in this column, contact Deborah Janes, Public Information Officer, U.S. Environmental Protection Agency (B205-01), Office of Research and Development, Research Triangle Park, NC 27711; phone: 1-919-541-4577; e-mail: janes.deborah@epa.gov. Disclaimer: Although this text was reviewed by EPA staff and approved for publication, it does not necessarily reflect official EPA policy.
Figure