(See Spear et al., p. 565)
A field of growing interest in environmental health is that of lead speciation, the study of the various chemical forms of lead. Different lead compounds have different solubilities, which in turn affect bioavailability. Descriptions of lead exposure such as concentrations in air or settled dust are telling quantities, but can only relate part of the story. A variety of biological and geochemical influences will determine how much lead ultimately reaches the blood and becomes available to exert toxic effects. Previous studies have identified the major lead-bearing constituents associated with lead smelting procedures (1,2) and examined the role of speciation and particle size in solubility (3). In the current research article, Spear et al. investigate the geochemical factors of lead speciation and particle size, and their potential role in the bioavailability of lead from bulk and airborne smelter dusts.
The authors present results from a primary lead smelter. Dust samples were obtained from a variety of locations representing different stages of processing. Quantities of bulk dust were taken directly from surface areas within the smelter. Airborne dust was sampled with an apparatus that simulates intake by the human respiratory system. Samples were subjected to X-ray diffraction to identify the leaded compounds associated with the smelter dust. A chemical extraction method was applied to examine the relative solubilities of the dusts. The important characteristics of this study are therefore its abilities to comment on lead speciation and particle size over a range of locations in a lead smelter and to infer how these factors may influence human health.
From X-ray diffraction results, the smelter dust mineralogy is reported to be heavily influenced by lead sulfide. Dusts sampled in the processing areas of the plant reveal proportions of sulfates and complex oxides. The abundance of lead sulfide throughout the smelter is reflected in chemical extraction results from the bulk dust samples, which generally display low solubilities. Bulk dusts from furnace locations tend to be more soluble than those that accumulate near ore storage areas. This locational trend is reinforced by airborne dust results, which are probably more indicative of direct worker exposure. The air sampler allowed dust to also be examined by particle size. Fine particles of leaded dust are more commonly observed in furnace areas than in storage or sinter locations. Regardless of mineralogy, finer particles are reported to be more soluble. Coupled with the intrinsic solubility of the lead sulfates and oxides found near furnace processes, these findings have important implications for the relative health hazards associated with smelter furnace areas.
Lead speciation, particle size, and bioavailability are all factors that may influence human health and that warrant continued investigation.
The Spear et al. article offers further evidence that a complete knowledge of workplace exposure to lead requires more information than simply the concentration in inhaled or ingested media. This is also true of studies of environmental lead. Significantly, environmentally exposed populations do not normally benefit from the regular blood lead monitoring associated with employment in the lead industries. Geochemical issues such as those examined in this article have complicated efforts to model indices of environmental lead exposure, absorption, and toxicity (4). Lead speciation, particle size, and bioavailability are all factors that may influence human health and that warrant investigation in the future.
David E.B. Fleming
Department of Physics and Astronomy, McMaster University
Hamilton, Ontario, Canada
References and Notes
1. Foster RL, Lott PF. X-ray diffractometry examination of air filters for compounds emitted by lead smelting operations. Environ Sci Technol 14:1240-1244 (1980).
2. Clevenger TE, Saiwan C, Koirtyohann SR. Lead speciation of particles on air filters collected in the vicinity of a lead smelter. Environ Sci Technol 25:1128-1133 (1991).
3. Ruby MV, Davis A, Kempton JH, Drexler JW, Bergstrom PD. Lead bioavailability: dissolution kinetics under simulated gastric conditions. Environ Sci Technol 26:1242-1248 (1992).
4. Renner R. When is lead a health risk? Environ Sci Technol 29:256A-261A (1995).
(See Pleil et al., p. 573)
In this issue of EHP, Pleil et al. report on a careful comparison of breath and blood concentrations of volunteers exposed in a chamber to trichloroethylene (TCE) in air. Although studies of this nature were carried out a few decades ago, recent advances in sampling and analytical techniques (some made by Pleil himself) have allowed the authors to follow the blood and breath decay curves for a longer period and with greater precision than before. The results are interesting and a bit puzzling. Before we explore these results, however, it is desirable to place this study in some perspective.
The basic reason for being interested in the breath/blood relationship for some scores or hundreds of volatile organic compounds (VOCs) is that such knowledge would allow us to substitute a noninvasive, nonthreatening breath measurement for the more difficult, generally less precise, and harder to obtain blood sample. Thus, a number of authors have attempted to determine the blood/breath partition coefficient, which is the ratio of the arterial blood concentration to the concentration in exhaled alveolar breath. But there are difficulties in such measurements. We normally measure venous blood, not arterial, and it is not a simple matter to collect alveolar breath. These difficulties in making a direct measurement have forced investigators to estimate partition coefficients by indirect means, using in vitro methods for the most part.
Once the partition coefficient P is known, it is possible using physiologically based pharmacokinetic (PBPK) models such as that of Ramsey and Anderson (1) to calculate the expected blood/breath ratio R at any time. Assuming that local equilibrium is attained instantaneously as the inhaled breath mixes with the alveoli, the governing equation for the relationship between the arterial and venous blood concentrations is
where Cart = concentration in arterial blood, Cven = concentration in venous blood, Cinh = concentration in inhaled air, Qalv = ventilatory volume rate, Qblood = cardiac output, and P = blood/breath partition coefficient.
Because ventilatory flow is often similar to cardiac output at rest (Qalv ~ Qblood), the equation simplifies in such cases to the following approximation:
Setting the exhaled alveolar concentration equal to Cart/P, we find the equation for the blood/breath ratio R during exposure to a nonzero concentration Cinh:
This same equation holds during the decay period, with Cinh = 0, so that the blood/breath ratio is simply equal (approximately) to the partition coefficient plus 1.
At the beginning of the exposure, the venous blood concentration is near zero, so the measured blood/breath ratio starts out at a low value and then increases toward an asymptote. The magnitude of the asymptote is determined by the metabolic rate; for a rate near zero, the asymptote would be nearly P + 1, but for a higher metabolic rate, the asymptote would be considerably smaller. Thus, the complete description of the time-varying blood/breath ratio would be expected to look something like Figure 1, where the ratio increases toward an asymptote during exposure and then makes a discontinuous jump to a constant value of P + 1. For values of the partition coefficient >=10, the blood/breath ratio during the decay period is approximately equal to the partition coefficient. This may be why Pleil et al. tend to use the two terms (partition coefficient and blood/breath ratio) interchangeably, although it would perhaps be better to preserve the differences between the two concepts: one is a fundamental constant, whereas the other varies with time depending on the exposure profile.
Figure 1. The expected behavior over time of the blood/breath ratio of a compound with moderate metabolism and a partition coefficient of 10 is shown for the case of a 2-hr exposure followed by exposure to clean air. In this four-compartment physiologically based pharmacokinetic simulation, the blood/breath ratio rises toward an asymptote that is less than 10 and then jumps discontinuously on cessation of exposure to a constant value of 10 + the ratio of alveolar ventilation to cardiac output (in this case the ratio was set equal to 1, so the final blood/breath ratio is 11).
Partition coefficients have been published for a large number of aromatics, aliphatics, and halogenated compounds (2-4). However, it is a matter of some uneasiness that few studies involving simultaneous measurements of blood and exhaled breath are available to check these values. It makes us even more uneasy when those few studies seem to disagree with the published values. For example, Perbellini et al. (5) found that for persons with occupational exposure to benzene, the measured partition coefficient agreed with the literature value of about 8, but for persons exposed only to environmental levels, the apparent partition coefficient was substantially greater (about 20). This led Travis and Bowers (6) to hypothesize that perhaps a saturable component in the blood was trapping benzene at low levels, possibly as protein adducts. If this sequestered benzene were released during the analytical procedure, it would lead to a higher level in blood than the corresponding breath measurement and thus a higher blood/breath ratio than expected for a given partition coefficient.
Some 20 years ago, Monster and Houthooper (7) found higher blood/breath ratios than would be expected for subjects exposed to TCE. Since then, advances have occurred allowing shorter sampling times for breath measurements and more accurate analytical techniques for both breath and blood measurements. In the EPA TEAM studies of the early 1980s (8-10), breath was sampled in 20-liter Tedlar bags, requiring about 5 min to take a sample. Successive samples could only be taken on a 20-min cycle. Because half-lives in the first compartment for many VOCs are on the order of 2-3 min, such a system is unsuitable for following uptake and decay curves. Therefore, a new sampling system was developed (11) that allowed approximate 1-min samples to be collected on a 5-min cycle. Even a 5-min cycle time, however, was a little too long for the first few crucial minutes after an exposure in a chamber or a workplace. Therefore, Pleil and Lindstrom (12) developed a single-breath method of sampling, using a 1-liter evacuated canister with a valve operated by the subject after breathing out normally; the vacuum within the canister then sucks some of the remaining alveolar air from the lung. Pleil also developed an analytical method that could quantitate the CO2 in successive samples from the same subject; a CO2 level lower than normal for that subject could be used not only to identify samples that were incompletely alveolar but also to adjust those samples so that they could be used instead of discarded. [The history of these improvements in breath sampling and analytical methods is provided in more detail in a recent article (13)]. Concurrently, the blood analytical methods were eventually greatly improved by employing an isotope dilution method, which detects and quantitates losses in extracting VOCs from the blood and allows all samples to be corrected for such losses. The isotope dilution system was sponsored by the EPA in several unsuccessful early attempts in the mid-1980s but was eventually perfected (14).
In the late 1980s and early 1990s, these methods began to be employed in both field and chamber studies. Wallace et al. (15) found similar results (higher calculated partition coefficients than the literature values) in a chamber study of five subjects exposed at relatively low levels of about 1 ppm. Ashley et al. (16) measured blood levels of a dozen or so prevalent VOCs in 800-odd persons as part of the National Health and Nutrition Examination Survey (NHANES). When these values were compared to the breath concentrations of another set of some 750 persons measured in the EPA TEAM studies (assuming that the two groups were comparable), the apparent partition coefficients were again about twice as high as the literature values for all of the VOCs common to both studies (17).
Now Pleil et al. have found a variation on these results: during the uptake phase in his chamber studies, the blood/breath ratio follows the expected monotonically increasing curve, approaching an asymptote of about 10 toward the end of the exposure period that appears (coincidentally, I believe) to be close to the literature value found for the TCE partition coefficient. However, during the washout period immediately following the chamber exposures, an unusually large increase in the blood/breath ratio is observed (although not for all subjects). Pleil et al. estimate an average blood/breath ratio during the elimination phase of about 29, which is quite close to the value observed by Monster and Houthooper (7). Moreover, instead of staying constant during the decay phase, the blood/breath ratio varies with time. Even more puzzling, it appears to increase for some subjects and decrease for others. Assuming that both the breath and blood measurements are correct (an assumption supported by the good agreement with theory during the uptake phase), we are left with an interesting puzzle: Why does the ratio become so unstable during the decay phase? Because the single-breath sampling method of Pleil et al. has the great advantage of allowing measurements during the first few minutes after exposure ends, a period of extremely rapid changes in the blood concentration, it might be expected that instability would occur due to rapid losses from the blood and little chance for the blood to mix. However, such instability should be damped out after a few minutes, whereas Pleil et al. find it lasts considerably longer. Another puzzle is that the blood/breath ratio in some subjects had nearly opposite behaviors; no simple explanation could possibly account for this. A third puzzle concerns the large jump in the blood/breath ratio from about 10 to 29. Although a discontinuous jump is expected in PBPK models, it is usually more modest. The size of the jump is determined by the metabolic rate and also by the blood flow to the metabolizing tissues. If the liver is the main metabolic site and only about 25% of the blood is shunted to the liver, it would be impossible to make such a large jump at the cessation of exposure.
Possibly these are just anomalous results that will not be repeated by subsequent studies; however, if they are confirmed, there will be interesting things to be learned as we try to understand this behavior. It is to be hoped that such confirmatory studies will not be long in coming. We need such studies either to confirm the literature values of the partition coefficients, thus enabling us to replace blood measurements with breath measurements in many situations, or else to show us the direction we must take to establish partition coefficients that will apply to real persons exposed either to environmental or occupational levels of the many VOCs of interest.
Lance Wallace
U.S. Environmental Protection Agency
Reston, Virginia
References and Notes
1. Ramsey JC, Andersen ME. A physiologically based description of the inhalation pharmacokinetics of styrene in rats and humans. Toxicol Appl Pharmacol 73:159-175 (1984).
2. Sato A, Nakajima T. Partition coefficients of some aromatic hydrocarbons and ketones in water, blood, and oil. Br J Ind Med 36:231-234 (1979).
3. Sato A, Nakajima T. A structure-activity relationship of some chlorinated hydrocarbons. Arch Environ Health 34:69-75 (1979).
4. Gargas ML, Burgess RJ, Voisard DE, Cason GH, Andersen ME. Partition coefficients of low-molecular-weight volatile chemicals in various liquids and tissues. Toxicol Appl Pharmacol 98:87-99 (1989).
5. Perbellini I, Faccini GB, Pasini F, Cazzoli F, Pistoia S, Rosellini R, Valsecchi M, Brugnone F. Environmental and occupational exposure to benzene by analysis of breath and blood. Br J Ind Med 45:345-352 (1988).
6. Travis CC, Bowers JC. Protein binding of benzene under ambient exposure conditions. Toxicol Ind Health 5(6):1017-1024 (1989).
7. Monster AC, Houthooper JM. Estimation of individual uptake of trichloroethylene, 1,1,1-trichloroethane, and tetrachloroethylene from biological parameters. Int Arch Occup Environ Health 42:319-323 (1979).
8. Pellizzari ED, Perritt K, Hartwell TD, Michael LC, Whitmore R, Handy RW, Smith D, Zelon H. Total Exposure Assessment Methodology (TEAM) Study: Elizabeth and Bayonne, New Jersey; Devils Lake, North Dakota; and Greensboro, North Carolina, Vol. II. Washington, DC:U.S. Environmental Protection Agency, 1987.
9. Pellizzari ED, Perritt K, Hartwell TD, Michael LC, Whitmore R, Handy RW, Smith D, Zelon H. Total Exposure Assessment Methodology (TEAM) Study: Selected Communities in Northern and Southern California, Vol. III. Washington, DC:U.S. Environmental Protection Agency, 1987.
10. Wallace LA. The TEAM Study: Summary and Analysis: Vol I. EPA 600/6-87/002a. Washington, DC: U.S. Environmental Protection Agency, 1987.
11. Raymer JH, Thomas KW, Cooper SD, Whitaker DA, Pellizzari ED. A device for sampling human alveolar breath for the measurement of expired volatile organic compounds. J Anal Toxicol 14:337-344 (1990).
12. Pleil JD, Lindstrom AB. Measurement of volatile organic compounds in exhaled breath as collected in evacuated electropolished canisters. J Chromatogr B Biomed Appl 665:271-279 (1995).
13. Wallace L, Buckley T, Pellizzari E, Gordon S. Breath measurements as volatile organic compound biomarkers. Environ Health Perspect 104(suppl 5):861-869 (1996).
14. Ashley DL, Bonin MA, Cardinali FL, McCraw JM, Holler JL, Needham LL, Patterson DG. Determining volatile organic compounds in human blood from a large sample population by using purge and trap gas chromatography-mass spectrometry, Anal Chem 64:1021-1029 (1992).
15. Wallace LA, Nelson WC, Pellizzari ED, Raymer JH. Uptake and decay of volatile organic compounds at environmental concentrations: application of a four-compartment model to a chamber study of five human subjects. J Expo Anal Environ Epidemiol 7:141-163 (1997).
16. Ashley DL, Bonin MA, Carkinali L, McCraw JM, Wooten JV. Blood concentrations of volatile organic compounds in a nonoccupationally exposed US population and in groups with suspected exposure. Clin Chem 40:1401-1404 (1994).
17. Wallace L. Environmental exposure to benzene: an update. Environ Health Perspect 104(suppl 6):1129-1136 (1996).
Last Updated: August 25, 1998