Document RpV2DeGB8pK4VvkgNmV6zJdLX

Unitad Stata* Environmantal Protaction Agancv Qtfic* o* Wiw RafluHtiona and Standard* Criteria and Standard* Diviaion Waahtngton 0C 20460 EPA`4J0/MU-022* Octooar 1900' C, 2- t------ Ambient Water Quality Criteria for Asbestos - _fr_o_p__wr>r o, r' US Enriroranaraaf Frooctc* A**wn * UC 111930 WM*. 12*30 3ivt. 4.ycr>ua Svx*. V-'A 40ic] - PLAINTIFF'S EXHIBIT CON-52 V ifr AMBIENT WATER QUALITY CRITERIA FOR AS8EST0S Prepared By U.S. ENVIRONMENTAL PROTECTION AGENCY Office of Water Regulations and Standards Criteria and Standards Olvlslon Washington, O.C. Office of Research and Development Environmental Criteria and Assessment Office Cincinnati, Ohio Carcinogen Assessment Group Washington, O.C. Environmental Research Laboratories Corvalis, Oregon Duluth, Minnesota Gulf Breeze, Florida Narragansett, Rhode Island i PROP' 'V OF / o." W 5 4*' REGIONAL. ulBF/*rY ROOM 6"'J FOB. SEATTLE, VYA 9`3174 DISCLAIMER This report has been reviewed by the Environmental Criteria and Assessment Office* U.S. Environmental Protection Agency, and approved for publication. Mention of trade nms or commercial products docs not constitute endorsement or recoonendatlon for use. AVAILABILITY NOTICE This docunent is available to the public through the National Technical Information Service, (NTIS), Springfield, Virginia 22161. ii ACKNOWLEDGEMENTS Aquatic Life Toxicology William A. Brungs, ERL-Narragansett U.S. Envirormental Protection Agency John H. Gentile, ERL-Nar'raoansett ll.S. Environmental Protection Agency Matnnalian Toxicology and Human Health Effects: William Nicholson (author) Mt. Sinai School of Medicine Steven Bayard, CAG U.S. Environmental Protection Agency Debdas Mukerjee (doc. mgr.) ECAO-Cin U.S. Environmental Protection Agency Roy E. Albert, CAG* U.S. Environmental Protection Agency Bonnie Smith (doc. mgr.) ECAO-Cin U.S. Environmental Protection Agency Robert Bruce, ECAO-RTP U.S. Environmental Protection Agency Gary Chapman, ERL-Corvallis U.S. Environmental Protection Agency W. Clark Cooper Robert Carton, OTS U.S. Environmental Protection Agency Patrick Durkin Syracuse Research Corp. Michael Flaherty U.S. Environmental Protection Agency Thomas J.Haley National Center for Toxicological Res. Alfred Garvin University of Cincinnati Phillip M. Cook, ERL-Duluth U.S. Environmental Protection Agency Si Quk Lee, ECAO-Cin U.S. Environmental Protection Agency Richard Lemen National Inst, for Occupational Safety and Health Steven D. Lutkenhoff, ECAO-Cin U.S. Environmental Protection Agency Gary S. Logsdon, MERL-Cin U.S. Environmental Protection Agency James Millette, HERL-CIn U.S. Environmental Protection Agency Charles Poreli, OTS U.S. Environmental Protection Agency James Rowe, OTS U.S. Environmental Protection Agency Jerry F.Stara, ECAO-Cin U.S. Environmental Protection Agency Technical Support Services Staff: D.J. Reisman, M.A. Garlough, B.L. Zwayer, P.A. Daunt, K.S. Edwards, T.A. Scandura A.T. Pressley, C.A.,Cooper, M.M. Denessen. Clerical Staff: C.A. Haynes, S.J. Faehr, L.A. Wade, D. Jones, B.J. Bordicks, 8.J. Quesnell, C. Russom, B. Gardiner. CAG Participating members: Elizabeth L. Anderson, Larry Anderson, Ralph Arnicar, Steven Bayard, David L. Bayliss, Chao W. Chen, John R. Fowle III, Bernard Haierman, Charalingayya Hiremath, Chang S. Lao, Robert McGaughy, Jeffrey Rosenblatt, Charm V. Singh, and Todd W. Thorslund. 1v TABU OF CONTENTS Criteria Sunmary Introduction Aquatic Life Toxicology Effects Sunmary Criteria References Mammalian Toxicology and Human Health Effects Introduction Exposure Analytical Techniques Ingestion from Water Ingestion from Food Exposure from Drugs Inhalation Pharmacokinetics Absorption and Distribution Excretion Effects Acute, Subacute, and Chronic Toxicity Teratogenicity Mutagenicity Carcinogenicity-Animal Oata Carcinogenicity-Human Data Synergism and/or Antagonism Fiber Size Considerations Criterion Formulation Existing Standards and Guidelines Current Levels of Exposure Special Groups at Risk Basis and Derivation of Criteria References Appendix I Appendix II Appendix III A-l B-l 8-1 B-l B-l B-2 c-i c-l C-l C-l C-13 C-19 C-20 C-ZO C-28 C-28 C-3Z C-32 C-32 C-39 C-38 C-40 C-60 C-90 C-94 C-97 C-97 C-98 C-99 C-100 C-U5 C-139 C-l 40 C-141 lv CRITERIA CRITERIA DOCUMENT ASBESTOS Aquatic Life No freshwater organisms have been tested with any asbestlform mineral and no statement can be made concerning acute or chronic toxicity No saltwater organisms have been tested with any asbestlform mineral and no statement can be made concerning acute or chronic toxicity. Human Health For the maximum protection of human health from the potential carcinogenic effects of exposure to asbestos through Ingestion of water and contaminated aquatic organisms, the ambient water concentration should be tero. The estimated levels which would result In Increased lifetime cancer risks of 10*, 10", and 10*7 are 300,000 flbers/1, 30,000 fibers/1, and 3,000 flbers/1, respectively. Estimates for consumption of aquatic organisms only, excluding the consumption of water cannot be made. vi INTRODUCTION Asbestos Is a broad term applied to numerous fibrous mineral silicates composed of silicon, oxygen, hydrogen, and metal cations such as sodium, magnesium, calcium, or Iron. There are two major groups of asbestos, serpentlne (chrysotlle) and amphlbole. Chrysotlle Is the major type of asbestos used In the manufacture of asbestos products. These products Include asbestos cement pipe, flooring products, paper products (e.g., padding), friction materials (e.g., brake linings and clutch facings), roofing pro. ducts, and coating and patching compounds. In 1975, the total consumption of asbestos In the U.S. was 550,900 metric tons. Of the 243,527 metric tons of asbestos discharged to the environment, 98.3 percent was discharged to land, 1.5 percent to air, and 0.2 percent to >ater. Solid waste disposal by consumers was the single largest contrlbu. tlon to total discharges. Although no process water is used In dry mining of asbestos ore, there Is the potential for runoff from asbestos waste-tail ings, wetmining, and Iron ore mining. Mining operations can also contribute substantially to asbestos concentrations In water via air and solid waste contamination. In addition to mining and Industrial discharges of asbestos, asbestos fibers, which are believed to be the result of rock outcroppings, are found In rivers and streaais. The chemical composition of different asbestos fibers varies widely and typical formulas are presented In Table 1 (U.S. ERA, 1976); It should be noted that the values obtained from actual chemical analysis of the various fibers also may differ slightly from the typical formulas. Although chryso tlle Is considered to be a distinct mineral, the five amphlbole minerals are ach varieties of other minerals (Zoltai and Stout, 1976). These minerals differ from each other both chemically and physically with the exception that TABLE 1 Typical Formulas for Asbestos Fibers 1. Serpentines 2. Amphlboles Chrysotlle Amoslte Crocldollte Anthophyl 11 te Tremollte Actlnollte M93$l20s(OH)4 (M9.Fe)7Sl8022(0H)2 Ha2(Hg,Fe)sSl8022(0H)2 (Hg,Fe)7S1a022(0H)2 Ca2Mg5Sl8022(0H)2 C*2(Mg,Fe)5Sl8022(0H)2 ey all contain silicon and all form flbtrs when crushed. Good quality asbestos will fona fibers with higher ratios of length to width than poorer grades. The basic crystal form of the amphlbole minerals Is less complicated than for chrysotlle. The basic structure consists of a double silica chain (S140u) that Is paired back -to>back with a layer of hydrated cations between the chains (Spell and Lelneweber, 1969). Some typical physical properties of three different mineral forms are presented In Table 2 (Gaze, 1965). Asbestos minerals, despite a relatively high fusion temperature, are completely decomposed at temperatures of 1,000*C. Both the dehydroxylatlon temperature and decomposition temperature Increase with Increased MgO con- nt among the various amphlbole species (Spell and Lelneweber, 1969). The solubility product constants for various chrysotlle fibers range from 1.0 x 10"11 to 3 x 10"^. Most materials have a negative surface charge in aqueous systems. However, since chrysotlle has a positive () charge. It will attract, or be attracted to, most dispersed materials. The highly reactive surface of asbestos causes many surface reactions which are intermediate between simple absorption and a true chemical reaction. The absorption of various materials on the surface of chrysotlle supports the premise that the polar surface of chrysotlle has a greater affinity for polar molecules (e.g., HjO.NHj) than for nonpolar molecules (Spell andLelneweber, 1969). Of all the asbestos minerals, chrysotlle Is the most susceptible to acid attack. It Is almost completely destroyed within 1 hour In 1 N HC1 at */-< 95*C. Amphlbole fibers are much more resistant to mineral acids (Llndell, 972). A-3 TABLE 2 Typical Physic*! Properties of Chrysotlle {White Asbestos), Crocldollte (Blue Asbestos), end Amoslte* Units Chrysotlle (white asbestos) "r;i'.7 Crocldollte Amoslte (blue asbestos) Approximate diameter of smallest fibers Specific gravity micron - Average tensile strength Modulus of elasticity Ib/1nch2 lb/1nch2 `Source: Gaze, 1965 0.01 2.55 3.5 x 10$ 23.5 x 10 0.08 3.37 0.1 3.45 5 x 10$ 1.75 x 10 27.0 x 10 23.5 x 10 A-4 The resistance of the asbestos fibers to attack by reagents other than, acid Is excellent up to temperatures of approximately 100`C with rapid deterioration observed at higher temperatures. Chrysotile Is completely decomposed In concentrated KOH at 200*C. In general, organic acids have a tendency to react slowly with chrysotlle (Spell and Lelneweber, 1969). REFERENCES Gaze, R. 1965. The physical and molecular structure of asbestos. Ann. N.r. Acad. Scl. 132: 23. Llndell, K.V. 1972. Biological effects of asbestos. Int. Agency Res. Can cer, Lyon, France. Spell, S. and J.P. lelneweber. 1969. Asbestos minerals In modern technol ogy. Environ. Res. 2: 166. U.S. EPA. 1976. Asbestos: A review of selected literature through 1973 relating to environmental exposure and health effects. EPA-560/2-76-001. U.S. Environ. Prot. Agency, Washington, O.C. Zoltal, T. and J.H. Stout. 1976. Comnents on asbestlform and fibrous min eral fragments relative to Reserve Mining Co. taconlte deposits. Prepared for Minnesota Pollut. Control Agency. *3 7 auatic Life Toxicology EFFECTS No appropriate data on the effects of asbestos on aquatic organisms are available at this time. Therefore, no freshwater or saltwater criterion can be derived for asbestos. However, microscopic inorganic particles, analyzed by transmission electron microscopy, have been detected In fish tissues (Satterman and Cook, 1980). Tissue samples obtained from a river with known chrysotlle asbestos contamination and lake trout, brook trout, and channel catfish exposed to Lake Superior water contaminated with amphibole fibers have been found to contain mineral fibers identical to those In the water. Muscle tissue concentrations are about one-twelfth of the average water con centrations (by volume) but liver and kidney fiber concentrations are 500 times greater than muscle tissue concentrations. jummary The only available data for asbestos and freshwater organisms results from field studies in which chrysotlle and amphibole fibers have been found in tissues of fish collected from freshwater with known concentrations of these mineral fibers. No data are available for saltwater organisms. / CRITERIA No freshwater organisms have been tested with any asbestlfonn mineral, and no statement can be made concerning acute or chronic toxicity. No saltwater organisms have been tested with any asbestlform mineral, and no statement can be made concerning acute or chronic toxicity. 8-1 REFERENCES Batteman, A. R. and P. M. Cook. 1980. A method for the determination of mineral fibers In fish tissues. Paper to be presented at the 13th Ann. Meeting of the Minnesota Chapter of the American Fisheries Society. ( 8-2 L ASBESTOS Mammal 1 an Toxicology and Human Health Effects INTRODUCTION Estimating a risk factor for Ingestion of asbestos presents significant difficulties. Although gastrointestinal cancer has been linked to occupa tional exposures In several groups of workers, no definitive data exist on the effects of direct Ingestion of asbestos, either in animals or humans, further, only limited Information exists on air exposure levels for those human studies showing excess risk of gastrointestinal cancer and peritoneal mesothelioma. Nevertheless, the-most valuable data on risk are those from human Inhalation exposures, and these will form the primary basis for a projected criterion. This document Is not an exhaustive review of all asbestos literature nor e all Important papers mentioned herein. However, the papers selected are deemed relevant for estimating dose-response relationships. EXPOSURE Analytical Techniques For the purposes of this document asbestos Is defined to be ehrysotlle, crocldollte, fibrous cummlngtonlte-grunerlte Including amoslte, fibrous tremollte, fibrous actlnollte, and fibrous anthophylllte. The flbrosity of the above minerals Is ascertained on a microscopic level with fiber defined to be particles with an aspect ratio of 3 to 1 or greater. This definition will apply to fibers of all sizes. Because of the Impossibility of relating fibers In any water system to bulk mineral deposits from whence they came, the mineral nature of fibers will generally be determined, when necessary, by electron beam Instrumentation (morphology, selected area electron dlf>act1on, and electron microprobe analysis). C-l The analytical techniques for the measurement of asbestos minerals In air or water samples collected In occupational or general environmental cir cumstances are time-consuming, and the results are often highly variable. No single method Is suitable for all monitoring circumstances. Techniques appropriate for monitoring workplace exposures are unreliable when used to evaluate the much lower environmental concentrations of asbestos, such as those found In water, largely because of the presence of ouantltles of other Inorganic and organic material. Electron microscopic methods used for envi ronmental monitoring are difficult to perform and costly. Reproducible re sults can be obtained In experienced laboratories If standardized techniques are utilized, careful Quality control Is maintained, and periodic interla boratory comparison of results Is made. With careful analysis of water, Interlaboratory precision can achieve relative standard deviations of 30 to 65 percent (Anderson and Long, 1580; Chopra, 1978), but without standardiza tion Intralaboratory variability can be as great as a factor of ten, and In terlaboratory variability can exceed two orders of magnitude (Brown, et al. 1976). Environmental-Water: Considerable effort has taken place In recent years to standardize techniques for the auantltatlon of mineral fibers In water. All work to date has utilized .electron microscopy. The presence of numerous diatom spicules and other nonasbestos fibers In water and the great difficulty of uniquely Identifying mineral species or classes by optical microscopy would appear to preclude the use of optical microscopy for even the quantitation of large asbestos fibers In water. With electron micro scopy, however, relatively few experimental problems remain, and reproduci ble results can be obtained by experienced laboratories. The disadvantage of this method Is the cost and time of analysis and the limited availability of laboratories for the analysis of samples. C-2 The U.S. EPA has proposed in interim method for the analysis of asbestos ..i water (Anderson and Long, 1980). Prom a 1-liter sample, 50 to 500 ml is. filtered through 0.1 micron polycarbonate (Nuclepore) filter. A portion of the filter is placed on an electron microscope grid and dissolved by the Jaffe-Vick method and scanned by transmission electron microscopy at 10,000 to 20,000 magnification. Prior to dissolution, the flat polycarbonate fil ters are coated with carbon which serves to enmesh the collected material and to reduce losses during dissolution of the filter material by chloro form. Twenty grid squares or 100 fibers are counted. The Identification of fiber type is by morphology for chrysotlle and by selected area electron diffraction for amphlboles. No attempt is made to determine the amphlbole mineral species. If necessary, this can be done using. energy-dispersive X-ray analysis of each fiber. All Individual fibers (length greater than ^ *ee times width), irrespective of length are counted in the grid squares .anned. The fibers in large clumps, though, are not counted Individually. Por surveillance of large numbers of water systems, the procedures serve to Identify those with significant quantities of asbestos present. Por water systems with high concentrations of suspended solids, the collected material and filter can be ashed in an activated oxygen furnace, the remaining mate rial resuspended, ultrasonlfied, and refiltered. The sensitivity of procedure this is such as to be able to detect about 250,000 fibers/liter (f/1) or less in most drinking water systems without the need for the ashing and resuspension step. Most municipal water systems contain less than 1 mg/1 of suspended solids, and thus 200 ml of water can- 2 be filtered through a 10 cm filter for analysis. The counting of 20 grid squares as prescribed above, scans 1.3 x 10"^ cm^ of filter. In this ( -ea typical background counts are less than two fibers. Thus, eight fibers C-3 counted would establish detectable level In a given water sample. with 200 ml of water sampled, this corresponds to 250,000 f/1. In water systems having less suspended solids the lower limit of detection Is proportionally lower. With systems containing more suspended material, similar detection limits can be achieved following the ashing procedure. A previously used technique of condensation-washing of cellulose acetate Will pore filter pieces on carbon-coated grids using acetone can result in significant losses unless extreme care Is taken. Carbon coating of the Mil11pore filter Is Ineffective In enmeshing the fibers because many of tham are trapped deep within the Interstices of the membrane filter. Condensa tion of acetone on the grid can result In the formation of pools of solvent on the filter which wash away fibers. Losses as great as 80 percent have been reported using this technique (Chatfleld, et al. 1978; Beaman and file, 1976; Chopra, 1978). Eighteen analytical laboratories participated In an American Society for Testing and Materials (ASTM) Task 6rcup study of the measurement of amphlbole and chrysotlle fibers In water. Table 1 lists the data on the Interla boratory precision that has been obtained by this group In the analysis of both chrysotlle and amphlbole fibers. The Task Group concluded: The transmission electron microscope 1$ the best basic Instru ment for the analysis, particularly when It is equipped with $elected area electron diffraction and energy-dispersive spectroscopy capabilities. The mean fiber concentrations by different groups aoree within a factor of two. The Interlaboratory reproducibility of 50 percent can be expected In relatively clean water samples unless the concentration Is low. In samples with high concentra tions of Interfering solids, the precision will no,t be as good. When applied on a broad scale there are variable and significant losses associated with the condensation-washing of samples contain ing amphlbole. The losses are low and less variable when condensa tion-washing is used to prepare samples containing chrysotlle (Chopra, 1978). C4 TABLE 1 Inter laboratory Precision Obtained In the Analysis of Water Samples for Chrysotllc and Amphlbole Minerals* Sample Type Number of Laboratories Reporting Mean Fiber Concentration (10* fibers of all slzes/1) Relative Standard Deviation of Analysis (X) ysotlle ysotlle ysotlle vnrysotlle Chrysotlle Chrysotlle Amphlbole Amphlbole Amphlbole 10 9 11 9 9 3 11 4 14 877 119 59 31 28 25 139 95 36 Source: Anderson and Long, 1980 (see also Chopra, 1978) 35 43 41 65 32 35 50 52 66 Environmental--Air; As with water, the analysis of ambient air samples by optical techniques introduces significant difficulties. First, the quan tity of asbestos In ambient air 1$ only a small fraction of the total aero-, sol. This aerosol contains large quantities of organic and mineral material of various origins. Including many fibers other than asbestos. Therefore, enumeration of fibers collected In ambient air may have little relevance to the asbestos material present. In one Instance, a comparlsonof 25 ambient air samples collected in buildings, soaw of which were contaminated with asbestos, showed no correspondence between concentrations of fibers longer than 5 um, as determined using optical microscopic techniques, and the total mass of asbestos present, quantitated by electron microscopic methods (Nich olson, et al. 1975). Here, using the National Institute for Occupational Safety and Health (NIOSH) technique, no fiber concentrations measured ex ceeded 0.03 f/ml, and contributions to the measured filter concentration from other than asbestos fibers were felt to be significant. A review (Ouggan and Culley, 1978) of the results of the analysis of six side-by-side am bient air samples by nine laboratories also highlighted the difficulty of using optical microscopy at low asbestos concentrations. They found that Intralaboratory variability could exceed a factor of 10 and the results be tween laboratories could differ by a factor of 100. The possibility exists that optical techniques using petrographic, polarized light microscopes or dispersion staining techniques could produce better results. This has not been Investigated, however. A variety of techniques, each of which utilizes electron microscopy, have been developed for the analysis of asbestos In the ambient air. At the present time, there Is less agreement on an ideal method for air analysis than for water analysis. Two general electron microscopic techniques are -ellized for the analysis. One involves the collection of asbestos on cell ulose acetate (Mllllpore) or polycarbonate filters (Nuclepore) (Samudra, et al. 1978) and its subseouent transfer to electron microscope grids. For samples collected on cellulose acetate filters, the filter and collected material are ashed, the ash suspended In water, and the suspension filtered through a polycarbonate filter. Such filters are then processed using technioues similar to those used for water and previously discussed (see Water section). Although not well studied, the use of flat-surfaced polycar bonate filters In field situations may lead to losses of particles prior to sample preparation for analysis. Direct transfer techniques have other limitations. Ambient aerosols are made up of agglomerates of particles with asbestos fibers attached to a ( *1ety of other material. Chrysotile asbestos, for example, with a posi- /e surface charge, readily adheres to any of the large number of nega tively charged particles, such as clays, in the ambient air. without,dis persal, these agglomerations can result In the asbestos being obscured when viewed by an electron microscope. Further, agglomeration can occur on the filter during the long collection times required to quantitate low concen trations. In many cases, these agglomerates, which usually are of respir able size, contribute the most to the mass of the sample. Also, they may occur so infrequently that a statistically reliable measure of their quan tity is difficult to obtain. To obviate these difficulties, techniques have been developed in which collected material and filter are ashed in a lowtemperature, activated oxygen furnace. The resulting residue Is dispersed by physical means, either through the application of ultrasonic energy or I 'ndlng, and is enmeshed In a nitrocellulose or collodion film for mounting electron microscope grids or is reflltered through a polycarbonate fil- ter. Such "rub-out" methods also Involve losses and, as with washing tech( nlaues, require skilled development of the process. A significant disadvan tage of this procedure Is that the Initial physical state of t'hi asbestos Is altered prior to enumeration. Therefore, Information on the fiber size dis tribution 1$ not available. Only mass concentrations can be determined. (Nicholson, 1971a; Nicholson and Pundsack, 1973). To date, there has been less Interlaboratory agreement In the analysis of air samples than for water sample analysis. In one Interlaboratory com parison of samples collected near a road surfaced with serpent!nlte rock and analyzed for the mass of chrysotlle asbestos, Intralaboratory differences exceeded two orders of magnitude, and Interlaboratory differences for labor atories using different analysis techniques exceeded four orders of magni tude. Fiber counts were similarly variable (U.S. EPA, 1977). On the other hand, relatively good agreement (average relative standard deviation of 25 ( percent) was achieved by three laboratories In the analysis for amphlboles of 12 samples collected In Silver Bay, Minnesota (U.S. EPA, 1975). Analysis of amphlboles In air around Lake Superior by the U.S. EPA and the State of Minnesota has been done using a cellulose ester filter for col lection. The filter is shipped to the laboratory where it is ashed in a low temperature oxygen-activated furnace. The residue Is resuspended and fil tered through a polycarbonate filter. Good recovery and low losses are claimed by the Investigators (Cook, 1978). Occupational: In occupational circumstances, the current method of ouantltating asbestos air concentrations Is to enumerate all fibers longer than 5 urn collected on a specified area of filter, utilizing phase-contrast light microscopy at 400X magnification [National Institute for Occupational Safety and Health (NIQSH), 1972]. Such Instrumentation does not allow < identification of the fibers according to mineral type nor is it even C-8 flcient to establish If they are organic or mineral In origin. In general, when the principle fiber In an aerosol Is known to be asbestds; this presents no problem. However, In some occupational circumstances, as with the use of Insulation materials, fibers of various origins are present In the same material, and this can result In overestimates of the actual asbestos concentrations. The adoption of a 5 urn cutoff for the length of fibers enumerated was Imposed by the limitations of light microscopy. It has long been known that fibers longer than 5 wm and visible by phase contrast microscopy represent only a small fraction of the total number of asbestos fibers In the air (Lynch, et al. 1970). This would present no problem were fiber size distri butions similar In different circumstances. However, such is not the case. It has been shown, using electron microscopy, that when chrysotile asbestos -ncentratlons In different exposure circumstances are enumerated, the frac-ion greater than 5 u may vary by 10-fold (from 0.4 percent of the total number of fibers present to approximately S.O percent). When amphlbole varieties of asbestos are also considered, the fraction counted can vary more than 100-fold (Nicholson, et al. 1972). Thus, we do not have an accu rate yardstick for the quantitation of asbestos air concentration In the workplace. This does not present serious problems when monitoring for stan dard compliance but complicates comparisons of health effects between vari ous Industrial processes such as mining, manufacturing, and end-product use. It also complicates extrapolations of dose-response relationships determined In occupational circumstances to lower concentrations of asbestos measured in the general environment by other techniques. Nevertheless, when assess ing exposure In a defined asbestos aerosol, the precision of optical methods C-9 can be good. NIOSH (1976) has estimated that a coefficient of variation of about 20 percent can be achieved In the assessment of asbestos concentri. t r0 tlons greater than 0,1 f/ml. Although fiber counts have been utilized for the assessment of occupa tional asbestos exposure since 1966, In prior years other methods, usually Involving total particle counts (fibrous and nonflbrous), were utilized. Same attempts have been made to relate these earlier counts to present day fiber concentrations (Lynch and Ayer, 1966). However, these have been found to depend strongly on the particular asbestos use process, and no universal conversion factor Is available that would relate total particle concentra tions In a given circumstance with asbestos fiber counts. It Is unfortunate that earlier data have limited relevance, since the disease experience that we are seeing today Is the result of exposures that took place 20, 30, or more years previously when work conditions may have been considerably dif ferent from those currently existing. Thus, dose-response relationships are tenuous and can only be approximate, based upon current data. Intercomparison of Techniques: All data, scant as they are, that relate asbestos disease to exposure are derived from studies of workers exposed In occupational environments. In these studies, concentrations of fibers long er than 5 um were determined using optical microscopy or were estimated from optical microscopic measurements of total particulate matter. On the other hand, all current low-level environmental assessments utilize electron microscopic techniques which are not comparable to those used In the work place since optical techniques do not provide data on the number of fibers less than 5 um In length. To extrapolate dose-response data obtained in studies of working groups to environmental exposures, it Is necessary to establish the relationship between optical fiber counts and mass or total fiber number determined by electron microscopy. C-10 Recent studies have ittempted to relate optical fiber counts (fibers > S urn) and TEM counts (all EM-countable fibers). An Interlaboratory comparison of optical versus EM counts of chrysotlle fibers suggested an average relationshlp between optical counts and TEM counts of ItlOQQ. (Winer and Cossette, 1979). The samples studied Included air samples from six plants (one asbestos cement, one brake lining, two treating mills., and two textile plants). Lower ratios are expected for amphlbole fibers. An analysis by the U.S. ERA (Personal communication, J. Hlllette) relating optical fiber counts of fibers longer than 5 um to total fiber counts by transmission electron microscopy gave a ratio of 400 for six samples of asbestos celling Insulation material (which, however, may contain fibers other than asbestos and were not actual air samples). Other data by Wallingford (1978) suggest a ratio as low as IS for EM count to optical counts. Some data exist that relate optical fiber counts (longer than S um) to che total mass of asbestos as determined by electron microscopic techniques or by other weight determinations of collected airborne asbestos fibers. These are listed In Table 2 and provide crude estimates of a conversion factor relating fiber concentrations (f/ml) to airborne asbestos mass (ug/ffl^). The proposed standards for asbestos In Great Britain by the British Occupational Hygiene Society (B0H5) stated that a `respirable* mass of 0.12 mg asbestos/m* was equivalent to 2 f/ml (80HS, 1968). It was not stated how this relationship was determined. However, if it were from mag nesium determinations In an aerosol, the weight determination would likely be high because of the presence of other nonflbrous, magnesiumcontaining compounds In the aerosol. Such was the case In the work of Lynch, et al. (1970), and their values for the conversion factor are undoubtedly overesti mates. The data of Rohl, et al. (1976) are likely to be underestimates C-ll aA ll fib e r counts used phase-contrast m icroscopy and enumerated fib e rs longer than 5 C o n v e rs io n fa c to r may be low due to losses In C.N. p ro ce ssin g . C o n v e rs io n fa c to r may be h ig h because o f o v e re s tim a te o f asbestos mass on th e b a s is o f t o t a l m a^iesium . ? oW 'S u mm 4 O o 4eu1 * u VI VI e* in SCM <MM <e < lO o' si* c WD U> e ( <U9 i*9W *Ve*I 2 5g u w <2 eg (A ^ veot 9 M oS s (A e V) mesui 4ui W^<>w*J*9mCOmi* HV>I mi ^--^ X WtOfxlAO ,w^&9 c**o dew- W eo8 U<U9l w *o* ^V>I 8 w ou UaN91-*W" w ^bJS ^*.$V) e&4J <sQ u J--S x C tt W < Ol m rvk 0mmm s 9) *X .s &VuaI o&b e +* Umkrmn. e C aQdm.*i us>i C-12 because of possible losses In the determination of mass by electron micro scopy. No data exist on the procedures used to determine the mass of chrysotile in the data presented by Davis, et al. (1978). The range of 5 to 150 for the conversion factor relating mass concentra tion to optical fiber concentration Is great, and any average value derived from it has a large uncertainty. However, for the purpose of extrapolating to low mass concentrations from fiber count, the geometric mean, 30 ug/m^/f/ml, of the above range of conversion factors will be used. The accuracy of this value is felt to be no more than a factor of 5 and this uncertainty severely limits any extrapolation in which it is used. In the case of amosite, the data of Davis, et al. (1978) suggest that a conversion factor of 18 is appropriate. However, since this data yielded lower chryso- tlle values than all other chrysotlle estimates. It may also be low for moslte. Ingestion from Water Asbestos is commonly found In domestic water supplies. Samples from 365 cities have been collected and analyzed by electron microscopy by the U.S. EPA. Of these, 45 percent had detectable levels of asbestos, usually of the chrysotlle variety (Millette, 1979). Table 3 lists the distribution of the concentrations of these samples. 2t* t. . , " Earlier, asbestos had been reported in a variety of Canadian water sup plies (Cunningham and Pontefract, 1971). These waters were found to contain from 2.0 to 172.7 x 10 flbers/1. (In this subsection fibers will denote' all EM-countable fibers, irrespective of length). Two U.S. river systems were also reported to contain chrysotlle at average levels of from 0.3 to 1.5 ug/1 (Nicholson and Pundsack, 1973). Other reports include that of Kay '973) who found from 0.1 to 4 x 10 f/1 In various Canadian drinking water sources. C -13 TABLE 3 Distribution of Reported Asbestos Concentrations In Drinking Hater from 36S Cities luJ^LSlites* Puerto Rico, and the District`of Columbia Asbestos Concentration (106 flbers/1) Nurter of Cities Percentage of Samples Below detectable limits^ Not statistically significant Less than l , 1-10 Greater than 10 Total 110 90 90 34 _41 365 30.1 24.6 24.6 9.3 11.2 99.8 M111ette, 1979 bFor these analyses average detectable limits were 5 x 105 fibers/1. However, significant variations occurred in some instances due to the presence of nonasbestos fibers. tv ` C-14 During 1973, Urge amounts of asbestos-like fibers of amphibole mineral* were found In the waters of Lake Superior, the source of drinking water for Duluth, Minnesota, and other cities (Cook, et al. 1974, 1976; Nicholson, 1974}. Fiber concentrations during normal lake conditions ranged from 20 x 10 to 75 x 10 f/1 and from about 5 to 30 ug/1 In terms of mass (Nicholson, 1974). During storm conditions amphibole fiber concentrations as high as 600 x 10 f/1 were observed (Cook, et al. 1976). Filtration plants now used In Duluth maintain fiber concentrations below 0.1 x 10 f/1 (Mil lette, 1979). leu'-.- . ^ - Certain U.S. water systems currently have high levels of asbestos as a result of serpentine or amphibole deposits In their watersheds. These include Everett, Washington, with concentrations of chrysotlle above 10* '; Seattle, with from 1 to 10 x 10 f/1; and San Francisco, with chryso> concentrations about 10* f/1 In some systems (Mlllette, 1979; Cooper, al. 1978). ' v- Under certain conditions, asbestos -cement (A/C) pipe may also contribute asbestos to municipal water supplies. Asbestos fiber concentrations In A/C pipe distribution systems were found to be as high as 38 x 10 chrysotlle and 4 x 10 amphibole flbers/1 In one Florida city; 17 x 10 in another Florida town; and 47 x 10 f/1 in a Kentucky A/C pipe system. Water at the end of a little-used A/C pipe line In Massachusetts contained as much as 480 x 10 chrysotlle f/1 (Mlllette, 1976). Many of the A/C pipe systems in Connecticut have been sampled and analyzed (Craun, et al. 1977). The majority of samples taken after transit through A/C pipe showed concentra tions under 1 x 10 f/1, and only one sample was over 10 x 10 f/1. While there are an estimated 200,000 miles of A/C pipe now in use in the *d States, it Is apparent that not all A/C pipe sheds fiber?. If the water Is nonaggresslve the pipe dots not erode end contribute fibers to the water (Hallenbeck, et el. 1978). A study (Buelow, et el. 1980) of 10 A/C pipe systems showed thet fibers were edded to the water by the A/C pipes of the 5 systems with eggresslve water (Aggressiveness Index <10.0) and little effect was seen In the nonsggresslve systems. In two systems the pipe was eroded to a depth of 0.3 cm. In one case In a period of only 5 years. In this system fiber counts as high as 550 x 10 f/1 were measured In the distribution network versus 800,000 at the well source. In a third system high concentrations at a dead end sample were attributed to debris from tapping and drilling of pipes In the network. Sampling of representative water utilities throughout the United States has Indicated that over half of the samples had water which was moderately aggressive and 16.5 percent had very aggressive water (Table *) (Millette, et al. 1979b). Water supplies In both the very aggressive and moderately aggressive categories are potentially capable of eroding asbestos-cement pipe (l.e., 68.5 percent of U.S. water systems) although the very aggressive waters could be expected to result, in the contribution of much higher fiber concentrations. Host data on asbestos In water are expressed In terms of fiber concen trations, enumerating fibers of all sizes using appropriate electron micro scope techniques. Some estimates exist (Mlllette, 1979) relating chrysotlle fiber concentrations to mass concentrations. Because the number-to-mass relationship Is highly dependent on average fiber length and diameter, knowledge of the source of the fibers In the water Is Important In determin ing a conversion factor. Some average conversion factors are listed in Table 5. C -16 TABLE 4 Representative Average Water Utility Aggressiveness Indices* ( Highly aggressive** Moderately aggressive^ Nonaggresslved 16.5 percent 52.0 percent 31.5 percent M111ette, et al. 1979b ^Highly aggressive: pH * 1ogio(AH)<10.0 ^Moderately aggressive: pH log (AH) * 10.0 - 12.0 ^Honaggresslve: pH log (AH)>12.0 where A - total alkalinity In mg/1, CaC03 H calcium hardness as mg/1, CaC03 C-17 TABLE 5 Relationship of Total Fiber Counts by Electron Microscopy and Mass of Chrysotile Asbestos In Water* Fiber Source Natural erosion of serpentine rock (shorter fibrils) A/C pipe (longer fibers) Contributions from conwercial dump site runoff and untreated discharge (more fiber bundles) Source: Mlllette, 1979 Average Mass In ug of 10* Fibers of All Lengths 0.002 0.01 0.05 c-xe Similar information on the relationship of fiber count and mass has been olished by Kay (1973), whose data suggest that 10 fibers corresponds^tp from 2 x 10** to 2 x 10"^ ug In water systems. Oata on asbestos concentrations from erosion of fibers from A/C cooling tower panels Indicate that the mass of 10 fibers Is from 0.01 to 0.2 ug (Lewis,. 1977). Based on the aforementioned data, it Is concluded that the majority (approximately 95 percent) of water consumers in the United States are ex posed to asbestos fiber concentrations of less than 10 f/1. In a few areas people are exposed to concentrations between 1 and 10 million f/1 with Intermittent exposures over 100 million f/1. There is at least one area where continuous exposure is over 100 million f/1. Persons using asbestoscement pipe in areas where the water is nonaggresslve or is treated to pre vent corrosion are generally not additionally exposed. In areas of aggresive water, however, the consumer may be exposed to added asbestos fiber centratlons of from fewer than 1 million to over 100 million fibers per liter, depending on factors such as length of pipe, flow rate, and mineral content of the water. The mass concentrations of chrysotlle asbestos In the water of cities with less than 10 f/1 are likely to be less than 0.01 ug/1, corresponding to a daily intake of less than 0.02 vg. However, In areas with significant contamination, whether from natural sources, man's activities, or erosion from A/C pipes, the intake of asbestos from water sources can exceed 2 wg/day. Ingestion from Pood There are scant data on the contribution of food products to population asbestos exposure. Cunningham and Pontefract (1971) showed that various heers and wines could contain quantities of asbestos fibers similar to those C-19 found In water systems (106 to 107 f/1). The source of this contamlnatlon could be from neturel water sources or from the erosion of asbestos fibers from filters used to purify the product. Asbestos filters are cur rently used for the purification of beverages and a variety of other food products, but little data exist on possible fiber contamination from such sources. Contamination of drinking Mater by fibrous glass and other synthe tic fibers used In cartridge filters has been measured at concentrations In excess of 109 f/1 (Cook, ct al. 1978). Exposure from Drugs Erosion of chrysotlle from asbestos filters, used to purify parenteral drugs, has been documented (Nicholson, et al. 1972). Contamination levels up to 1 ug/dose Mere noted In approximately one-third of drugs tested, indi cating that filter erosion can be significant. Because of these findings, the use of asbestos filters for drug purification, without subsequent cleanup, has been prohibited by the Food and Orug Administration (41 FR 16933). Inhalation general Population Exposures: Asbestos of the chrysotlle variety has been found to be a ubiquitous contaminant of ambient urban air. A study of 187 quarterly composite samples collected In 48 U.S. cities from 1969 to 1970 showed chrysotlle asbestos to be present In virtually all metropolitan areas (Nicholson, 1971a; Nicholson and Pundsack, 1973). Table 6 lists the distribution of values obtained in that study. Each represents an average of from five to seven 24-hour samples and thus averages over possible peak concentrations which could occur periodically or randomly. A second set of ambient air analyses Is also shown for comparison (U.S. EPA, 1974). These studies utilized different analytical techniques but the resuits agree well. In both studies, 98.5 percent of the 24-hour samples had chrysotlle asbestos C-20 TABLE 6 Distribution of 24-Hour Chrysotlle Asbestos Concentrations In the Ambient Air of U.S. Cities* Asbestos Concentration (ng/ntf) less than Electron Microscopic Analysis Mount Sinai School of Medicine Nuirber of samples Percentage of samples Battelle Memorial Institute "1 Nuafcer of samples Percentage of samples 1.0 61 2.0 119 32.6 63.6 S.O 164 87.7 10.0 176 94.2 20.0 50.0 184 18S 98.5 99.0 100.0 187 100.0 Source: Nicholson, 1974; U.S. EPA, 1974 27 21.3 60 47.2 102 80.1 124 97.6 125 98.5 127 100.0 127 100.0 C-21 concentrations of less than 20 ng/m3. Of the three samples greater than 20 ng/r3 analyzed by the Mount Sinai School of. Medicine, one was In a cityM F having a major shipyard and another in a city that had four brake manufac turing facilities. Thus, these samples may Include a contribution from a specific source In addition to that of the general ambient air. Similar data with the same range of mass concentrations have recently been reported from France, providing evidence of the presence of chrysotlle in the ambient air of Paris (Sebastien, et al. 1976). In a study of the ambient air of Mew fork City, In which samples were taken during daytime working hours, values higher than, those mentioned .above were obtained (Nicholson, et al. 1971). These were 6- to 8-hour samples collected between 8:00 A.M. and 5:00 P.N., and they reflect what could be intermittently higher concentrations from construction activities or automo bile usage during those hours compared to nighttime periods for example. Table 7 records the chrysotlle content of 22 samples collected in the 5 boroughs of New York. It should be noted that the samples analyzed in all of the studies discussed above were taken during a period when fireproofing hlghrise buildings by spraying asbestos-containing materials was permitted. The practice was especially common In New York City. While no sampling sta tion was known to be located adjacent to an active construction site, unusu ally high levels could nevertheless have resulted from the procedure. To determine If construction activities could Indeed be a significant source of chrysotlle fiber in the ambient air, 6- to 8-hour daytime sampling was conducted In lower Manhattan In 1969 near sites where extensive spraying of asbestos-containing fireproofing material was taking place. Table 8 shows the results of this sampling and demonstrates that spray fireproofing C -22 TABLE 7 Chrysotile Content of Ambient Air In Mew York City by Borough (6- to 8-Hour Daytime Samples)* Sampling Locations Number of Samples Manhattan Brooklyn Bronx Oueens Staten Island 7 3 4 4 4 Source: Nicholson, et al. 1971 Asbestos air level in , 10*9 g/n3 (ng/m3) -- , Range Average 8-65 6-39 2-25 3-18 5-14 30 19 12 9 8 TABLE 8 Chrysotlle Air Levels Hear Spray Fireproofing Sites In New York City (6- to 8-Hour Daytime Samples)* Sampling Locations (distance from site) 1/8 - 1/4 mile 1/4 - 1/2 mile 1/2 - 1 mile Number of Samples 11 6 5 Asbestos air level 10-9 g/m3 (ng/3) Range Average 9 - 375 8 - 54 3.5 - 36 60 25 18 The above concentrations reflect both downwind and upwind sampling locations. Source: Nicholson, at al. 1971 C-24 :ontrlbute significantly to asbestos air pollution. In some Instances, sotlle asbestos levels approximately 100 times the concentrations typlcally found In ambient air were observed. Asbestos contamination has also been documented by analysis of samples collected within buildings. In a study of 116 samples collected In or near 19 buildings (primarily office) In 5 U.S. cities, average chrysotlle air concentrations ranged from 2.5 ng/m3 to 200 ng/m^, with Individual mea surements from 0 to 800 ng/m3 (Nicholson, et al. 1975). For the outside air, the variation for the average concentration at a given site extended from 0 to 48 ng/m^. 8u11dings in which a loose asbestos fireproofing material was applied to the structural steel surfaces had evidence of sig nificant asbestos contamination. Also, schools In which similar material had been applied have been found to be seriously contaminated. Optical * counts exceeding 2 f/ml In a library and other areas of student use observed during activities which disturbed loose asbestos (Sawyer, 1977; Nicholson, et al. 1978). Ambient air chrysotlle concentrations in schools. In absence of any disturbance of the asbestos ranged up to 2,000 ng/rn^ (Nicholson, et al. 1978; Sebastian, et al. 1976). Finally, analysis of the air of asbestos workers* homes Indicate that chrysotlle concentra tions as high as 5,000 ng/m3 can be encountered (Nicholson, et al. 1978). Figure 1 summarizes the ranges of chrysotlle concentrations In the vari ety of environmental and occupational circumstances discussed above. The concentration ranges are only approximate and In most cases are limited be cause of the limited number of samples taken In given circumstances. Exten sion to higher and lower concentrations would be expected with the avail ability of more data. I I M C-M Although the fate of the asbestos In Inspired air is only approximately known, it appears that eventually more than half the asbestos inhaled will be swallowed (see Effects section). Assuming that an Individual breathes 10 In 24 hours, most ambient air levels of chrysotlle (l to 10 ng/m^) result In exposures to the gastrointestinal tract of froai 0.01 to 0.05 ug/day of asbestos, although, In some circumstances. Inhalation could pro duce gastrointestinal exposures exceeding 0.1 ug/day. These exposures are to be compared with those from water ingestion which lead to dally Intakes of less than 0.02 ug (see Ingestion from Water section). Though the data of Tables 3 and 6 are not related to the same population bases, it would appear that inhalation can give rise to exposures at least equal to that of direct ingestion for most of the population of the United States. ,,* / Only after 1966 has occupational monitoring attempted to quantify asbes- s exposures by fiber counting techniques. Since then, considerable data Ahave accumulated on occupational exposure of workers to asbestos. large compilation of such data Is Included In the 1972 Asbestos Criteria Document (NIOSH, 1972). Levels during the period from 1966 through 1971 were gener ally under lOf (f>5ffl)/ml, although concentrations exceeding 100 f/ml were observed, particularly In two plants producing amoslte insulation materials and in uncontrolled textile mills. Oata on earlier exposures are lacking although some estimates have been made of insulation-workers' exposure (Nicholson, 1976) and factory environments (BOWS, 1968; Newhouse and Berry, 1979). Although average exposures of 10 to 40 f/1 are likely to have pre vailed, peak or localized exposures in excess of 100 f/1 would have been encountered often by some Individuals. for purposes of estimating dose-response relationships, those data that e available for given work environments will be discussed In conjunction with the measured health effects. Absorption md Distribution PHARMACOKINETICS .. . Ingestion: A key Question In the evaluation of cancer risk associated with the Ingestion of asbestos In water is whether microscopic fibers under normal alimentary canal conditions can migrate through the gastrointestinal mucosa. Such movement of fibers could enable their residence In bowel wall or, following hematogenous or ljmiphatlc transport, the peritoneum and other organ tissues. This has been well answered by the work of Carter and Taylor (1980) who demonstrated the presence of amphlbole fibers, characteristic of those In Duluth, Minn, drinking water. In tissue samples of liver, Jejunum, and lung of deceased Ouluth residents. Among 96 tissue specimens of 32 Duluth residents amphlbole fibers were found In 60, with concentrations ranging from 3 x 10 to 16 x 10 fibers of all slzes/gra* of tissue. Amphlbole fibers were found In only 2 of 61 tissue specimens of 21 control subjects deceased In Houston, Texas and St. Paul, Minn. As air sampling gave no evidence of amphlbole air contamination In Ouluth, the authors attribute the highly significant evidence (p <0.001) of tissue contamination to transmucosal uptake of fibers Ingested by drinking amphlbole contaminated Ouluth water. Some studies of tissues of animals that had Ingested fibers report no evidence of fiber transport through the gastrointestinal lining (Gross, et al. 1974). These results, however, have been called Into question on the basis of the Insensitivity of the assay technique used (Cooper and Cooper, 1978). Evidence for such movement Is reported in other studies (Cunningham and Pontefract, 1973). Cunningham, et al. (1977) observed chrysotile fibers In the blood and_t1ssues of rats which previously were fed a diet of one percent chrysotile asbestos for six weeks. Westlake, et al. (1965) identi- chrysotile fibers in the colon mucosa of rats fed chrysotile asbestos, canning electron micrographs have revealed large amoslte asbestos fiber)* penetrating epithelial cells of rat jejunal mucosa tissue (Storeygard and Brqwn, 1977). Kidney cortex tissue of neonate baboon fed chrysotile for nine days was found to contain a statistically significant (p - 0.005) ex cess of chrysotile fibers compared to kidney cortex tissue from an unexposed neonate baboon (PatelJiandllk and Hallenbeck, 1978). Cunningham and Ponte fract (1974) observed passage of chrysotile fibers from the blood across the placenta to the fetus. Ingestion of small particles other than asbestos has also resulted In the subsequent observation of particle accumulation in tissues of animals. Mice that drank water suspensions of 2 urn diameter latex spheres for two months were found to have the latex particles accumulated In macrophages In stlnal Peyer's patches (LtFevre, et al. 1978). Latex particles of 0.22 were reported to migrate from rat stomachs to lymphatics of the mucosa and also to liver and kidney tissues (Sanders and Ashworth, 1960). Much larger particles of silica, opal phytollths from plants, are observed in digested mesenteric lymph node and kidney tissue from sheep which eat cereal chaff and grains (Mottle, 1977). Evidence for the human Intestinal uptake ("persorptlon") of particles as large as 75 um Is provided by the observation of starch granules In blood only minutes after Ingestion (Volkheimer, 1974). Sleep, smoking, and caf feine are reported to Increase the number of starch particles In the blood. Dyed cellulose particles are also Identified In human blood and urine fol lowing ingestion of specially stained plant food (Schreiber 1974). The cel lulose fibers are found in urine several weeks after Ingestion. Langer 74) found asbestos fibers in extrapulmonary organ tissues of asbestos C-29 workers, although fewer then In lung and pleura tissue, and more fibers In kidney than in liver, pancreas, adrenal, or spleen tissue,. i* Human urine sediment examined by transmission electron microscopy may contain amphlbole fibers which originate from Ingestion of drinking water contaminated with these mineral fibers (Cook and Olson, 1979). Ingestion of filtered water results In eventual disappearance of amphlbole fibers from urine. These observations provide direct evidence for the passage of min eral fibers through the human gastrointestinal mucosa under normal alimen tary canal conditions. Measured concentrations of amphlbole fibers elimi nated in urine represent approximately 1 x 10"^ of the number of fibers Ingested with drinking water. To the extent that some fibers are pennanently retained by the body or eliminated by other routes after passage across the gastrointestinal wall, the urine concentrations are an underestl mate of Ingested fiber absorption. Inhalation: Inhalation of asbestos dust Is accompanied by ingestion of many fibers cleared from the respiratory tract by mucociliary action.. The occurrence of peritoneal mesothelioma, excess gastrointestinal tract can cers, and possibly cancers at other nonrespiratory tract sites could result from migration of fibers through the gastrointestinal mucosa. Additionally, fibers may reach organs In the peritoneal cavity by transdlaphragmatlc migration or lymphatic-hematogenous transport. However, this would likely be a very small contribution compared to transmucosal migration following Ingestion, The amount of Inhaled asbestos which is eventually Ingested is important for an assessment of cancer risk based on the excess gastrointes tinal cancer observed for occupational exposures (see Effects section). Whether inspired asbestos fibers will be deposited in the lung depends strongly upon their diameter. Timbrel! (1965) has shown that a fiber, inde- C-30 I it of its length, behaves aerodynamically like i pa. tide having a dia meter three times as great*. Brain and Vo 1 berg (1974) have developed a model" for aerosol deposition In the respiratory tract according to aerodynamic parameters. They indicate that about 50 percent of particles with a mass median diameter of less than 0.1 urn will be deposited on nonclllated pulmo nary surfaces. This fraction falls slowly to 25 percent at 1 u* and to zero at above 10 uffl. Deposition on nasal and pharyngeal surfaces becomes Impor tant at 1 mu and rises rapidly to be the dominant deposition site for parti cles 10 um in diameter or greater. Thus, few fibers with a diameter as large as 2 um are likely to penetrate into the alveolear spaces, although finer fibers, even as long as 200 um, may do so. Once inhaled, a large fraction of the Inhaled dust Is rapidly cleared from the respiratory tract by mucociliary action although some fibers will ( n in the lung and be found there decades after exposure (Poo1ey,1973; l. ..r, 1974). Because of the ubiquitous exposure of individuals to asbes tos, chrysotile fibers can be found In the lungs of most urban dwellers danger, et al. 1971; Gross, et al. 1973). Additionally, larger fibers trapped in the lungs may become coated and form asbestos bodies. These can be readily observed by optical microscopy In tissue sections and in lung smears (Thomson, et al. 1963; langer, et al. 1973). The number of fibers or asbestos bodies found In given circumstances depends strongly upon the nature of the previous exposure of the Individual. The clearance of asbestos from the respiratory tract of rats has been studied directly in a series of experiments (Morgan, et al. 1975; Evans, et al. 1973). Samples were made radioactive by neutron Irradiation, which en abled the mass of asbestos In various tissues to be determined. In a series ^ '-minute exposures with different varieties of asbestos, the deposition C -31 TABLE 25 Expected and Observed Deaths from lung Cancer and Cancer of the Esophagus, Stomach, Colon, and Rectum in Workers Exposed to Amoslte Asbestos (Followed 5 to 35 Years after Employment from 1941 to 1945}* Length of Employment 1 mo 1 mo 2 mo 3-5 mo 6-11 mo 1 yr 2 yrs Total Lung Cancer Expected 1.6 2.5 2.4 4.2 3.2 2.6 6.0 22.5 Source: Seidman, et al. 1979 Observed 4 6 8 9 12 15 39 93 61 Cancer Expected 1.4 2.4 2.6 4.2 3.2 2.5 6.4 22.7 Observed 2 2 3 8 1 5 _7 28 C-78 and clearance In the respiratory tract were followed. At the conclusion of the Inhalation, the distribution in various organ systems was determined. The results are shown in Table 9. As can be seen, rapid clearance from the upper respiratory tract occurs with up to two-thirds of the fibers being swallowed and found In the gastrointestinal tract. Long term respiratory tract clearance or drainage via the lymphatics leads to additional dissemi nation. Other data on the deposition and retention of inhaled asbestos have been reported by Wagner, et al. (1974). Figure 2 shows the dust content of rat lungs following exposures to different asbestos varieties. As can be seen, the chrysotlle content of the lung does not build up as significantly as that of the amphlboles for similar exposure circumstances. This is likely the result of some dissolution of chrysotlle by body fluids. Excretion Host Inhaled or directly Ingested asbestos particles which pass through the gastrointestinal tract are excreted in feces (Cunningham, et al. 1976). As mentioned previously, some fibers are absorbed by the gastrointestinal tract and are eventually eliminated through the urinary tract (Cook and Olson, 1979). EFFECTS Acute. Subacute, and Chronic Toxicity Acute effects are of little consequence in the inhalation exposure of Individuals to high concentrations of asbestos dust. Some temporary breath ing difficulty has been reported by workers In various circumstances, but such discomfort has not limited employment In the Industry. Short-term effects have been described In a recent study by Harllss, et al. (1978) who found airflow abnormalities In 17 of 23 individuals examined C -32 I TABLE 9 Distribution of fiber it the Termination of Exposure (X of Total 0epos1ted)*b Fiber Chrysotile A Chrysotlle 6 site crocidollte Anthophylllte Fluoramphibole Nasal Passages" 9 ' 3 8*2 6+l 8+3 7+2 3+2 aMorgan, et al. 1975 bMean and SO Esophagus 2+1 2+1 21 2l 2+1 1 SI Tract 51+9 54 + 5 57 + 4 51+9 61+8 67 + 5 Lower Respiratory Tract 38 + 8 36M 35 + 5 39 + 5 30 + 8 29 + 4 l c-n Wslght of dust in lungs (mg) FIGURE 2 Me*n Height of Dust In Lungs of Rets In Relation to Dose and Tle Source: Wegner, et el. 1974 C-34 and 8.0 months following a relatively Intense flve^nonth exposure to asbestos. Of the 17, 12 were nonsmokers or current light or ex-light smok ers (less than 10-pack years). The obstructive abnormalities were usually present In measurements both of one minute forced expiratory volume and of closing volume determinations. Although human data on Initial changes art unavailable, Holt, et al. (1964) described early (14-day) local Inflamatory lesions found in the ter minal bronchioles of rats following Inhalation of asbestos fibers. These consisted of multinucleated giant cells, lymphocytes and fibroblasts. Pro gressive fibrosis followed within a few weeks of the first exposure to dust. (These early alterations in animals may be related to the early human find ings above). Oavls, et al. (1978) described similar early lesions In rats <--'slstlng of a proliferation of macrophages and cell debris In the terminal ichloles and alveolae. Jacobs, et al. (1978) fed rats 0.5 mg or 50 mg of chrysotlle dally for 1 week or 14 months and subseguently examined gastrointestinal tract tissue by light and electron microscopy. No effects were noted In esophagus, stomach, or cecum tissue but structural changes In the Ileum were seen, particularly of the villi. Considerable cellular debris was present by light microscopy In the ileum, colon, and rectum tissue. The electron microscopic data con firmed that of light microscopy and Indicated the observed changes were con sistent with a mineral-Induced cytotoxicity. A single oral administration of from 5 to 100 mg/kg of chrysotlle to rats has produced a subsequent Increase In thymidine In the stomach, duo denum, and jejunum (Amacher, et al. 1975). This suggests that an Immediate response of cellular proliferation and OKA synthesis may be stimulated by sot11e Ingestion. The long-term disease entity, asbestosls, resulting from the Inhalation of asbestos fibers is a chronic, progressive pneumoconlosl*. It Is charac terized by fibrosis of the lung parenchyma, usually ridlologlcally evident after 10 years from first exposure, although changes can occur earlier fol lowing more severe exposures. Shortness of breath Is the primary symptom; cough Is less common; and signs such as rales, finger clubbing, and. In later stages of the disease, weight loss appear In a proportion of cases. The disease was first reported 7 decades ago (Murray, 1907) and has occurred freauently among workers occupationally exposed to the fiber in ensuing years. Characteristic X-ray changes are small. Irregular opacities, usually In the lower and middle lung fields, often accompanied by evidence of pleu ral fibrosis or thickening, and/or pleural calcification. Both the visceral and, more conaonly, parietal pleura may be Involved. The mechanism of action and translocation of asbestos fibers to the parietal pleura Is uncertain; both direct migration (Klvlluoto, I960) or transport via lymphatics (Tasklnen, et al. 1973) have been suggested. Currently, 50 to 80 percent of Individuals In occupational groups with exposures beginning more than 20 years earlier have been found to have ab normal X-rays. These Include asbestos Insulation workers (Sellkoff, et al. 1965), miners and millers (Mount Sinai, 1976) and asbestos factory employees (Lewlnsohn, 1972). In many circumstances the disease progresses following cessation of exposure; In a group employed In an asbestos factory for vari ous periods of time between 1941 and 1954, X-ray changes were observed years following exposure In Individuals having exposures as short as one week (Personal connunlcation, I.J. Sellkoff). Restrictive pulmonary dysfunction Is also seen with asbestos exposures and may be accompanied by dlffuslonal defects or airway obstruction (Bader, .. 1961). In the early stages of asbestosls, there is limited correla tion between physiologic parameters, such as lung function tests. Later," X-ray changes and the lung function deficits are more highly correlated, but * still incompletely so. The above chronic effects are common among occupational groups directly exposed to asbestos fibers. They also, however, extend to those employed in other trades working near the application or removal of asbestos. Among workers other than Insulators employed at a shipyard for longer than 15 years, 48 percent were found to have abnormal X-rays (Sellkoff, et al. 1979b). Similar data were obtained .In a study of maintenance personnel in a chemical plant (LI11s and Sellkoff, 1979). Even family contacts (wives, children, etc.) of workers can be affected. Anderson, et al. (1976) have shown that 36 percent of 626 family contacts of workers employed some time ( Jen 1941 and 1954 at an asbestos insulation manufacturing facility had X-.*y abnormalities years later characteristic of asbestos exposure. In addition to disease and disablement during life, asbestosis has ac counted for a large proportion of deaths among workers. The first reports of the disease (Aurlbault, 1906; Murray, 1907) described complete eradica tion of working groups. Much Improvement In dust control has taken place in the industry since the turn of the century, but even recently those exposed in extremely dusty environments, such as textile mills, may have as much as 40 percent of their deaths attributable to this cause (Nicholson, 1976). Groups with lesser exposures for 20 or more years, such as in mining and milling (Mount Sinai, 1976) or Insulation work (Sellkoff, et al. 1979a) may have from 5 percent to 20 percent of their deaths from pneumoconiosis. All varieties of asbestos appear equally capable of producing asbestosis, in I man (Irwig, et al. 1979) and animals (Wagner, et al. 1974). in groups C -37 i exposed at lower concentrations such as the families of workers, there is less Incaoacltatlon, and death from asbestosls has not been reported. * Extra-pulmonary chronic effects reported Include 'asbestos corns* from the penetration of asbestos fibers Into the skin and their incorporation in dermal layers, and Instances of Caplan's syndrome (rheumatoid pneumoconio sis). Mo chronic, nonmalignant gastrointestinal effects are reported. Teratogenicity No data exist on the presence or absence of teratogenic, effects from the Inhalation or Ingestion of asbestos, although transplacental transfer of asbestos has been reported (Pontefract and Cunningham, 1973; Cunningham and Pontefract, 1974) Mutagenicity In a preliminary study chromosomal aberrations were seen In Chinese ham ster cells cultured in a medium containing 0.01 mg/ml of either chrysotlle or crocldollte (Slncock and Seabrlght, 1975). Mo chromosomal aberrations were seen In culture with coarse glass fibers or with control media. A more extensive series of experiments by Slncock (1977), using several chrysotile and crocldollte samples, showed that both positive transformation of morphology and positive genetic responses result from the passive inclusion of asbestos In culture media of CH0-K1 Chinese hamster cells. Very fine fibrous glass produced the same abnormalities, but chemically leached asbestos fibers produced fewer abnormalities than those untreated. The principal results are shown In Table 10. Chamberlain and Tarmy (1977) tested U1CC asbestos samples of chrysotile, amoslte, anthophylllte, and samples of superfine chrysotile on several strains of coll and typhlmurlum bacterial systems in which mutageni city to exogenous materials appears to correlate well with animal carcino- C-33 e ffe c ts a f D iffe re d Treebetels M Otroaot. . (W M (I - C ftlM t* Header C e lls * --iw * O on --w ws oo X ws* |3s*" nOOMw J3s" oo I * s 8 u * M SotSSS zi:l 5|=i k g*i*l "2** a34*is= SSSft $ s 1 vi 5F SSS fin w #* O O* nr i J * its 5$= a==* l.a Jfe a^as & - *, II s. m U0-* as*s jti! ** 2 o* 5* ili ssk- m mm saas i t |si 1:2 Hi {fji fihlli 21=*ia11?1i fcajl C-3* r ( atorles. All available data on these experiments are listed In Table 11. The data are flawed for several reasons. The numbers In each experl, mental group were small, the doses administered limited, and significant Information on experimental procedures lacking. Also, systematic histologi cal examination, which was of most significance, was done on only 53 of over 200 animals. Wagner, et a 1. (1977a) fed groups of 32 rats 100 milligrams per day of chrysotlle or talc In malted milk for 100 days over a 6-month period of time. A small decrease In survival time was observed In the two study groups: 614 and 618 days versus 641' for the controls. Two gastric leiomyo sarcomas were observed, one In each exposure group. Interpretation of the results of this experiment, too. Is difficult because of the small number of wls In experimental groups. ts an outgrowth of concern for the use of asbestos filters In the puri fication of wine products and the possible effects of erosion of asbestos fibers from those filters Into the final product, a study was undertaken In which asbestos filtered material was fed to rats (Glbel, et al. 1976). Twelve malignant tumors developed In experimental animals. Including four kidney tumors. No tumors of this site were found In control groups. This observation of renal cancer takes on significance In light of the finding of an elevated risk of -kidney cancer among asbestos insulation workers (Sellkoff, et al. 1979a) and a high excretion of asbestos fiber In the urine of humans drinking flber-contamlnated water (Cook and Olson, 1979). However, this report provides only limited experimental detail, and the filter material was composed of sulfated cellulose and a condensation resin addition to 52.6 percent chrysotlle asbestos. The presence of other tances confounds the study In relation to asbestos carcinogenicity. C -41 HH l| | ! H i^|M K* l IS M*| *|||UJip MU 'I IK *> l i 01 I m i i ) *i > '* * I e- i-- tr y of [ i f t r l t n l t on U**. .ffcct( of flrol loftstloo of Asbotloti il .tm !U l : :oo l s ?1 *0 * 1 m rt *0 5 0 g # I I 8I 8 < I. 5wi it i! mi 8 0 5k 5i s 1 w !2l 1 I of : K5 i 0 VS k li X Ik* *w0 k*m0 % m ^fc f*c* < k s tf>* s5 x5 r 5 r& Rw C-O 1 Cunninghim, et al. (1977) conducted two limited feeding studies of male .tar rats. One percent chrysotlle asbestos with five percent com oil wee added to rat chow diet and fed to groups of 10 and 40 rats In two separate experiments. In the first study, six of seven surviving animals were found with tumors whereas only one malignancy was observed In eight controls (see Table 11). No gastrointestinal tumors were seen, but two of the treated group tumors were kidney nephroblastomas. In the second larger study, 11 tumors each were observed In treated and control groups of 40 animals. Two of the malignancies In the asbestos-fed group were of the gastrointestinal tract and one of the control group was a nephroblastoma, lessening the sig nificance of the finding of this tumor in the other treated group. With the limited number of animals In this study, the evidence for carcinogenicity of asbestos (by feeding) Is inconclusive. ( Currently, a very large feeding experiment Is being conducted under the pices of the National Institute of Environmental Health Sciences (NIEHS). Results, however, are not anticipated until late 1980. Meanwhile, all previously reported experiments on Ingested asbestos, whether positive or negative, have significant limitations. To extrapolate such data to man for use as a criteria for a standard would not be appropriate. Inhalation: Although lung cancer was suggested as being causally re lated to human asbestos exposure In case reports In 1935 (Lynch and Smith, 1935; Gloyne, 1935), strongly Indicated to be so In 1947 (Merewether, 1947), and unequivocally associated In a cohort study by Doll (1955), no positive animal data of consequence were forthcoming until 1967 when Gross, et al. (1967) showed that lung cancer could be produced by asbestos Inhalation exposure. An early experiment of Nordmann and Sorge (1941) described two ( ng tumors In 10 of 100 mice surviving 240 days following exposure to high C -45 concentrations of chrysotlle. This work, however, was called into question by Smith, et al. (1965) on the basis of the histology of the malignancies lynch, et al. (1957) exposed AC/F, hybrid mice to commercial chrysotlle and observed a higher Incidence of pulmonary adenomas In exposed animals, 45.7 percent (58/127), compared to controls, 36.0 percent (80/222). No malignant tumors were reported, and the Increase of adenomas was not significant at the 0.05 level. The first unequivocal data showing, a relationship between asbestos inhalatlon and malignancy was that of Gross, et al. (1967) who observed carcino mas In rats exposed to a mean concentration of 86 mgft? chrysotlle for 30 hours/week from the age of six weeks. Of 72 rats surviving for 16 months or longer, 19 developed adenocarcinomas, 4 developed squamous cell carcinomas, and 1, a mesothelioma. No malignant tumors were found In 39 control ani Amals. search was made for primaries at other sites which could have metastasized. None were found. These and other data are sunmarized In Table 12. Reeves, et al. (1971) found .2 squamous cell carcinomas In 31 rats sacri> flced after 2 years following exposure to about 48 mg/m^ of croeidollte. No malignant tumors were reported In rabbits, guinea pigs, hamsters, or in animals exposed to similar concentrations of chrysotlle or amoslte. No details of the pathological examinations were given. In a later study (Reeves, et al. 1974), malignant tumors developed in 5 to 14 percent of the rats surviving 18 months. Lung cancer and mesothelioma were produced by exposures to amoslte and chrysotlle and lung cancer by crocldollte Inhalation. Again, significant experimental details were lacking; information on survival times and times of sacrifice would have been useful. 'Available details of the exposures and results are given in Table 13. While C *46 Ti-- ir y mf fx p trta c u ti m U c e ffe c t* ( la t u lit la * f j U n l t i e-47 (a I C-4$ lM rtam t| la M U ttw crcl the relative carcinogenicity of the fiber types was similar, it was noted that the flbrogenlc potential of chrysotfle, which had been substantially reduced In length and possibly altered danger, et al, 1978) by milling, was much less than that of the amphlboles. These results were also discussed in a later paper by Reeves (1976). In an extensive series of experiments, Wagner, et al. (1974) exposed groups of Wlstar SPF rats to the five UICC asbestos samples at concentra tions from 10 to IS mg /a3 for times ranging from 1 day, to 24 months. For all exposure times there were SO adenocarcinomas, 40 squamous-cell carcino mas, and 11 mesotheliomas produced. None appeared prior to 300 days from first exposure. Considerable experimental detail Is provided In the paper. The significant data are presented In Tables 14 and IS. These tumors follow a reasonably good linear relationship for exposure times of three months or greater. The Incidence In the 1-day exposure group, however, is consider ably greater than expected. It was noted that exposure had a limited effect on length of life. Average survival times varied from 669 to 8S7 days for exposed animals versus 754 to 803 days for controls. The development of asbestosls was also documented. The Incidence of lung cancer was found to be greater In animals surviving 600 days. There were 17 lung tumors, 6 In animals with no evidence of asbestosls and 11 in rats with minimal or slight asbestosls. Cancers at extrapulmanary sites were also listed. Seven malig nancies of ovary and 8 of male genitourinary organs were observed In groups of approximately 350 rats. None were observed in groups of 60 male and fe male controls. Incidence of malignancy at other sites was little different from that of controls. If controls are Included from other.experiments in which ovarian and genitourinary tumors were present, the comparative Inci- C-50 TABLE 14 Number of Rats with Lung Tumors or Mesotheliomas After Exposure to Various Forms of Asbestos Through Inhalation* Form of Asbestos No. of Adenocarcinomas Animals Amoslte Anthophylllte Crocldolite Chrysotlle (Canadian) Chrysotlle (Rhodesian) None 146 145 141 137 144 126 Source: Wagner, at al. 1974 5 8 7 11 19 0 Squamous-cell Carcinomas 6 8 9 6 11 0 Mesothelloma 1 2 4 4 0 0 c-si TABLE 15 Numbers of Rats with Lung Tumors or Mesothellomas After Various Lengths of Exposure to Various Forms of Asbestos Through Inhalation* s Length of Exposure No. of No. with Lung No. with Pleural % of Animals Animals Carcinomas Mesotheliomas with Tumors None 126 0 1 day 219 3b 3 months 180 8 6 months 90 7 12 months 129 35 24 months 95 37 *Wagner, et al. 1974 b2 exposed to chrysotlle and 1 to crocldollte C1 exposed to amosite and one to crocidolite 0 2C 1 0 6 2 0.0 2.3 5.0 7.8 31.8 41.0 C-52 Excess deaths due to peritoneal mesothelioma and gastrointestinal cancer (ICO 150-158) eaual approximately 12 percent of the ejected number of deaths for asbestos workers in three different cohorts studied. An average exposure Index of 430 years x fibers >5 um/ml Is calculated for these work ers by multiplying average air fiber concentration estimates by average years of exposure time. Since water measurement for asbestos requires electron microscope analy sis for fibers (asbestos particles with length to width ratios >3.0) of all sizes, the occupational exposure Index must be converted from fibers >5 vm (optical microscope) to fibers of all sizes (electron microscope). A ratio of 200 electron microscope Identifiable fibers to one optical microscope Identifiable fiber Is used for chrysotlle asbestos In workplace air samples. A much smaller ratio Is expected for amphlbole fibers. Assuming a linear dose response, occupational exposure of 5 days/week and 8m3 air Inhaled/workday, and 70 years for Ingestion of drinking water, the criterion Is calculated as follows: (430 f>5 um/ml - years) (5/7) (200 f/f>5 urn) (1/70 years) A BC (lO^l/m3) (8m3/day) (10 *11.1 x 10*1) D EF G 600,000 fibers of all slzes/day A Exposure Index In years x fibers >5 um/ml from Sellkoff, et al. (1976, 1979a) and Newhouse and 8erry (1979) 8 Occupational exposure for 5 days versus 7 days for water exposure. Cj Conversion from optical counts (F >5um) to TEM counts (all fibers) in l fibers/fibers 5vm 0;* 70-year exposure Is assumed for drinking water. E Conversion from ml to m3. dene* in the exposure groups here lacks significance. No data were pro.vlded, however, on the variation of tumor Incidence at extrapulmonary sites with asbestos dosage. Wagner, et al. (1977a) also compared effects of Inhalation of a super fine chrysotlU to a pure, nonflbrous talc. One adenocarcinoma was found In 24 rats exposed to 10.8 mg/m* of chrysotlle for 37.5 hours/week for 12 months. Finally, In a study similar to Wagner's, Davis, et al. (1978) exposed rats to 2.0 or 10.0 g/m* of chrysotlle, crocldollte, and amoslte (equiva lent to from 430 to 1950 f/ml). Adcno- and squamous cell carcinomas were observed in chrysotlle exposures, but not with crocldollte or amoslte (see Table 16). One pleural mesothelioma was observed with crocldollte exposure, and extrapulmonary neoplasms Included a peritoneal mesothelioma. A rela- ively large number of peritoneal connective tissue malignancies were also )b$erved, Including a lelomyofibroma on the wall of the small Intestine. The significance of these tumors Is speculative, however. As discussed In the Pharmacokinetics section. Inhalation exposures . result In concomitant gastrointestinal exposures from the asbestos that Is swallowed after clearance from the bronchial tree. While all Inhalation experiments focused on thoracic tumors, those of Wagner, et al. (1974), Oavls, et al. (1978) and, te a limited extent, Sross, et al. (1967) also Included a search for tumors at extrathoracic sites. A limited number of these were found, but no association can be made with asbestos exposure. One aspect of the Inhalation experiments that Is noteworthy Is the sig nificant number of pulmonary neoplasms that can be produced In the rat by inhalation as compared to other species (Reeves, et al. 1971, 1974). This olnts to the variability of species response to asbestos and the need for C -S3 r TABLE 16 Experimental Inhalation Carcinogenesis In Rats* Exposure Mass (mg/m3) Fiber (f>5u/l) Number of Animals Examined Malignant Tumors Chrysotlle 10 1,950 Chrysotlle 2 390 Anoslte Crocldollte Crocldollte Control 10 10 5 550 860 430 Source: Davis , et al . 1978 40 6 adenocarcinomas 2 squamous-cell carcinomas 42 1 squamous-cell carcinoma 1 peritoneal mesothelioma 43 none 40 none 43 1 pleural mesothelioma 20 none C-54 ' appropriate model before extrapolations to man can be made with conf1 .ence. The absence of significant gastrointestinal malignancy from asbestos exposure in animals, in contrast to that found in humans, may be the result of the use of inappropriate animal models. Intrapleural Administration: Evidence that intrapleural administration of asbestos would result in mesothelioma was forthcoming in 1970 when Oonna (1970) produced mesotheliomas in Sprague-Cawley rats treated with a single dose of 67 mg of chrysotile, amosite, or crocidolite. Reeves, et'al. (1971) produced mesothelial tumors in rats (1 of 3 with crocidolite and 2 of 12 with chrysotile) by intrapleural Injection of 10 mg of asbestos. Two of 13 rabbits Injected with 16 mg of crocidolite developed mesotheliomas. Stanton and Wrench (1972), in a series of experiments, demonstrated that major commercial varieties of asbestos, as well as various other fibers, oduced mesotheliomas In as many as 75 percent of animals into which mateial had been surgically implanted. Extension of these experiments were re ported in 1973 (Stanton, 1973). These results are summarized In Table 17. The authors concluded that the carcinogenicity of asbestos and other fibers is strongly related to their physical size, those fibers of a diameter less than 3 urn being carcinogenic and those of a larger diameter not carcino genic. further, samples treated by grinding in a ball mill to produce shorter length fibers were less likely to produce tumors. While the authors attributed the reduced carcinogenicity to a shorter fiber length, the ques tion has been raised as to the effect of the destruction of crystallinity and perhaps other changes in the fibers occasioned by the extensive ball milling danger, et al. 1978). Another comprehensive set of experiments was conducted by Wagner (Wagr, et al. 1973, 1977b). He, too, has produced mesothelioma from intra- TABLE 17 Dose-response Data Concerning the Effects of Intrapleural Implantation of Asbestos and Other Fibers In Rats* Dose ("9) UICC-SRAS Crocldollte Hand-cobbed Virgin . Crocldollte Special South African crocldollte Partially pulverized crocldollte UICC-SRAS amoslte UICC-SRAS chrysotlle Coarse fibrous glass Glass wool Fine AAA fibrous glass 3urn diameter uncoated coated 1 2 10 20 40 1 20 40 40 40 40 40 40 40 40 40 No. of Rats with Mesotheliomas 2 5 11 12 14 4 10 18 15 8 15 15 1 1 3 5 Source: Stanton and Wrench, 1972 Total no. of rats 25 23 27 25 23 30 24 27 20 25 25 26 24 25 26 28 X of Rats with Tumors 8 22 41 48 61 13 42 67 75 32 60 58 4 4 12 18 056 'aural administration of asbestos to CO Wlstar rats and demonstrated a ong dose-response relationship. Tables 18 and 19 list the results of these experiments. Pylev and Shabad (1973) and Shabad, et al. (1974) reported mesotheliomas In 18 of 48 and In 31 of 67 rats injected with three doses of 20 mg of Rus sian chrysotlle. Other experiments by Smith and Hubert (1974) have produced mesotheliomas In hamsters injected with 10 to 25 mg of chrysotlle, 10 mg of smoslte or anthophylllte, and 1 to 10 mg of crocldollte. Various suggestions have been made that natural oils and waxes contami nating asbestos fibers might be related to their carcinogenicity (Harlngton, 1962; Harlngton and Roe, 1965; Commlns and Slbbs, 1969). This, however, was not borne out In the experiments described above by Wagner, et al. (1973) or Stanton and Wrench (1972). Intratracheal Injection: Intratracheal Injection has been used to study ie combined effect of administration of chrysotlle with benro(a)pyrene In rats or hamsters (see Synergism and/or Antagonism). In rats given three doses of 2 mg chrysotlle (Shabad, et al. 1974) or hamsters given 12 g of chrysotlle (Smith, et al. 1970) no lung tumors were observed. However, the coadministration of benzo(e)pyrene did result In lung tumors. Intraperltoneal Administration: Intraperltoneal Injections of 20 mg of crocldollte or chrysotlle produced three peritoneal mesotheliomas In 13 Charles River CO rats. Twenty mg of amoslte produced no tumors In a group of 11 (Maiton 1 and Annoscla, 1974).- They also Injected 25 mg of crocldollte Into 50 male and 50 female 17-week-old Sprague-Oawley rats and observed 31 mesothellal tumors In males and 34 In females. In an extensive series of experiments, Pott and Friedrichs (1972) and Pott, et al. (1976) produced peritoneal mesotheliomas In mice an(d rats 1n- C -57 TABLE 18 Percentage of Pats Developing Mesotheliomas After Intrapleural Administration of Various Materials* Material SFA chrysotlle (superfine Canadian sample) UICC crocidollte UICC amoslte UICC anthophylllte UICC chrysotlle (Canadian) UICC chrysotlle (Rhodesian) Fine glass fiber (code 100), median diameter, 0.12 * Ceramic fiber, diameter, 0.5-1 61ass powder Coarse glass fiber (code 110), median diameter, 1.8 urn `Wagner, et al. 1977b bwagner, et al. 1973 Percent of Rats with Mesotheliomas 66 61 36 34 30 19 12 103 0 C-S8 TABLE 19 Oose-Response Oita Following Intrapleural Administration of Asbestos to Rats* Material SFA chrysotlle Crocldollte Oose No. of Rats with Total no. (9) Mesothelioma of Rats O.S 1 13 25 44 88 0.5 1 10 23 42 85 12 11 12 12 12 11 12 12 13 11 Source: Wagner, et al. 1973 X of Rats with Tumors 8 27 42 33 62 9 0 25 15 45 C-59 ( Jected with various conmerelal varieties of asbestos and other fibrous mate rial. These results are shown In Table 20. Using experiments with Intrapleural administration, the malignant response was altered by ballmilling fibers for 4 hours. The rate of tumor production was reduced from 55 percent to 32 percent and the time from onset of exposure to first tumor was lengthened from 323 to 400 days following administration of four doses of 25 mg of U1CC Rhodesian chrysotlle. In the ease of the ball allied fiber, 99 percent were reported to be smaller than 3 urn, 93 percent less than 1 u<n, and 60 percent less than 0.3 urn. A strong conclusion which can be drawn from the above experimental data is that large-diameter fibers (greater than 3 vm) are significantly less carcinogenic than finer fibers. The origin of the reduced carcinogenicity of shorter, ball .grilled fibers Is less clear as the relative contributions of shorter fiber length and the significant alteration of the crystal struc ture by Input of physical energy are not, as yet, defined. Further, the extrapolation of data developed on size dependent effects, from Intrapleural or Intraperitoneal administration to Inhalation (where movement of the fibers In airways and subsequently through body tissues Is strongly sizedependent) presents significant difficulties. Finally, since the number of smaller fibers In an exposure circumstance may be 100 times greater than those longer than 5 urn, the reduction of their carcinogenicity must be demonstrated at a level 100 times less before their contribution can be neglected. Carcinogenicity - Human Data The modern history of asbestos disease dates from the turn of the cen tury, when two reports were published documenting uncontrolled conditions In asbestos textile factories. One, the testimony of H. Montague Murray (1907) C-60 f luaors In I M o n i M d /o r ltw r A fte r I r t r i f i r l l M U l u j l d l M f f i l m F U tr t, C r K l M l l i , * r CcriMdua l K e lt* & M * - 31s*i ii 11 tm m +*4 si *aS s s sas N HI a * }* O C-61 * O r *i 8x " aa 55 SS 3 yrw't C-62 r m ln itn l In tiamr r i l a hearing concerning compensation, described severe pulmonary fibrosis uund at autopsy In 1900 In the last survivor of' a group of 10 sorters first employed 14 years previously In a carding room. The second was the descrip tion by Auribault (1906) of deaths during the early years of operation of an asbestos weaving mill established at Conde-sur-Nolreau, France, in 1890. During this period SO men died, including 16 of 17 recruited froai a cotton textile mill previously owned by the factory director. With time, however, the spectrum of diseases associated with asbestos exposure continued to expand. In 1935 two clinical reports were published on lung cancer in asbestos workers who had died with evidence of pulmonary fibrosis (Lynch and Smith, 1935; Gloyne, 1935). While such reports were not sufficient to causally relate asbestos exposure to the lung cancer, the pos sibility was raised. In 1947 it was confirmed by substantial data which ed that 13 percent of a group of Individuals who died with asbestosls In .at Britain also had bronchogenic carcinoma (Merewether, 1947). Mesothe lioma, a rare tumor of the lining of the abdomen or chest, was first de scribed in an asbestos worker in 1953 (Weiss, 1953) subsequently found to be frequently associated with potential asbestos exposure (Wagner, et al. 1960), and unequivocally related to such exposure In 1965 (Newhouse and Thomson, 1965). Gastrointestinal cancer also was found to be In excess among asbestos Insulation workers In the United States (Sellkoff, et al. 1964). Currently, all major commercial asbestos varieties, chrysotile, amoslte, and crocldolite, have been found to produce a significant Incidence of asbestos-related disease among workers occupationally exposed In mining and milling, In manufacturing, and In the use of materials containing the fiber. ( predominant route of exposure has been inhalation, although some asbes- i tos may be swallowed directly or ifter being brought up from the respiratory trect. Not only has asbestos disease been found among Individuals exposed to the'lflber directly as a result of excessive work exposures in decades past, but asbestos-associated cancer has also been identified, albeit less frequently, among those with inhalation exposures of lesser intensity, in cluding those who had worked near the application or removal of asbestos material, those with history of residing in the vicinity of asbestos plants, and those who had lived in the household of an asbestos worker. Water Ingestion: Five studies have considered the relation of asbestos Ingested in drinking water to gastrointestinal cancer. As an outgrowth of the contamination of Lake Superior by fibrous material In the tailings of an Iron ore processing plant, the mortality of the population of Ouluth was compared with that of Minnesota and Hennapln County (Minneapolis) for quin- { quenia to 1969 (Mason, et al. 1974). The relative death rates for digestive cancer, lung cancer, and all neoplasm were elevated from 16 to 49 percent. However, with the exception of colon/rectal cancer, which was highly ele vated, no trends with time or consistency between male and female were clearly discemable. Because of this. Mason, et al. .(1974) concluded that additional followup was necessary to determine if a hazard exists. Levy, et al. (1976) conducted, a similar study with equivalent results. However, the short follow-up from the earliest possible exposure (1956) would make it unlikely that any positive result would be found. Furthermore, while the Reserve plant began production in 1956, current discharge levels did not begin until 1967 when a major plant expansion took place. A'study by Harrington, et al. (1978) reviewed malignancy in the Conmeetleut Tumor Registry from 1935 to 1973 to see if a correlation existed be- , tween the use of asbestos cement (A/C) pipe for public water supply and the i C-64 1 1dencc of gastrointestinal cancer. No association was found between the .*e-adjusted, sex-specific incidence data for stomach, colon, and recteV cancer and the use of A/C pipe. While some water supplies reported A/C pipe that was 45 years old in 1975, the majority (66 percent) of the population studied received water through A/C pipes that were only 25 years old. While the majority (56 percent) of A/C pipe systems In Connecticut have water which Is considered aggressive under the AWWA Standard for A/C transmission and pressure pipe, fiber counts done on over 100 A/C pipe systems In Con necticut showed 98 percent to be under 10* f/1 (J. Mlllette, personal com munication). A report published for the University of California analyzed the 19691971 cancer Incidence from 721 census tracts of the five Bay Area Counties along with the ehrysotlle asbestos fiber concentrations In the drinking ( .er (Cooper, et al. 1978). For the census tracts the ehrysotlle asbestos ber counts ranged from below detectable limits to 36 x 10* fibers per liter. The University of California Investigators grouped the census tracts on a gradient of low-to-hlgh asbestos counts and found significant dose-response gradients for the Incidence of several cancers. Statistically significant positive trends were noted for white male lung and stomach cancer and white female gal.1 bladder, esophageal, and peritoneal cancer. The census tracts were cross-classified using both asbestos count and tract socioeconomic status indicators of medium family Income and medium school years completed. The positive dose-response effect between cancer Incidence of certain sites and asbestos counts appeared to be Independent of the effect of socio-economic status. The fact that the significant results are I t restricted to one body site is not surprising considering the knowledge C -65 I that asbestos fibers are probably transported throughout the body. For ex ample, one study using rats has found that Ingested fibers are' deposited In the lung. (Cunningham, et al. 1977). An extension of this study (Cooper, et al. 1979) using six years of data showed a statistically significant as* soclatlon between asbestos levels In the San Francisco Bay area drinking water and cancers of the digestive tract. A study by Wlgle (1977) compared the cancer mortality In two areas of Canada with probable high concentrations of asbestos In drinking water with an area presumably having low concentrations. Only one published asbestos concentration Is provided. Five values are listed from a personal comnunlcatlon with no details given on the sampling and analytical methods uti lized. Ko data are provided to substantiate the assumed exposures of all of the "probably low exposure" group and five of the seven "possible high expo- ( sure" municipalities. The mortality experience was compared with that ex pected from Quebec rates, although for some sites It Is known that the rural counties have lower cancer rates than Quebec, the rates of which are domi nated by the urban center, Montreal. For example, the lung cancer rate of the rural counties near the asbestos mines Is only two-thirds that of Quebec (McDonald, et al. 1971). Elevated rates for cancer of the stomach, colon, and rectum were seen among "high exposure" males (46 observed vs. 38.4 ex pected), "possible high exposure" females (103 vs. 91.3) and "probably low exposure" females (311 vs. 270.3). The rates for the other two male and one female groups were about 5 percent less than expected. In addition to the absense of sampling data on exposure, the small number of deaths observed seriously limits the study. For example, this document estimates that a 10*5 risk of death from asbestos Ingestion may occur from exposures to ( 400,000 f/1. If there were no population migration Into or from the highly C-66 osed areas and everyone deceased In those municipalities were exposed for . lifetime to the concentrations Indicated, the above criteria level would predict about ten excess deaths among the approximately 1,000 that occurred over the observation time of this study. Insulation Application and Removal: A large study by Sellkoff, et al. (1979a) best demonstrates the full spectrum of disease from asbestos expo sure. They studied the mortality experience of 17,800 asbestos insulation workers from January 1, 1967 through December 31, 1976. These workers were exposed primarily to chrysotlle prior to 1940, and to a mixture of chrysotile and amoslte subsequently. No crocldollte is known to have been used In U.S. Insulation material (Sellkoff, et al. 1970). In this group, 2,271 deaths have occurred, and their analysis provides important Insights Into the nature of asbestos disease. Table 21 lists the expected and observed ths by cause, and Includes data on tumors less frequently found. Lung .wrs are cornnon and account for about 20 percent of the deaths; 8 percent are from mesothelioma of the pleura or peritoneum. Additionally, though, cancer of the gastrointestinal tract is significantly elevated; so, too, are cancer of the larynx, pharynx, and buccal cavity, and renal tumors. Other tumors are also increased, but not to a statistically significant degree for an Individual site. Comparing the deaths from cancer and asbestosls in this group with those expected In the general population, more than 40 percent of the deaths among Insulators can be attributed to their occupational exposure to asbestos fiber. Table 21 lists the observed deaths as categorized on death certificates and as determined after a review of all autopsy and medical records (BE). The use of deaths characterized by the best available medical evidence for k analysis Is appropriate when one considers diseases that are virtually C -67 * l TABLE 21 Deaths Among 17,800 Asbestos Insulation Workers in the United States and Canada January 1, 1967 - January 1, 1977*.b Number of Hen: 17,800 Man-Years of Observation: 166,853 Underlying Cause of Death Expected Observed (BE) (DC) Ratio o/e (BE) (DC) Total deaths, all causes Total cancer, all sites Cancer of lung Pleural mesothelioma Peritoneal mesothelioma Mesothelioma, n.o.s. Cancer of esophagus Cancer of stomach Cancer of colon-rectum Cancer of larynx Cancer of pharynx, buccal Cancer of kidney Deaths of less cormon malignant neoplasms Pancreas Liver, biliary passages Bladder Testes Prostate Leukemia Lymphoma Skin Brain All other cancer NonInfectious pulmonary diseases total Asbestosls All other causes 1,658.9 319.7 105.6 c c c 7.1 14.2 38.1 4.7 10.1 8.1 17.5 7.2 9.1 1.9 20.4 13.1 20.1 6.6 10.4 25.5 59.0 c 1,280.2 2,271 2,271 1.37 1.37 995 922 3.11 2.B8 486 429 4.60 4.06 63 25 112 24 0 55 18 18 2.53 2.53 22 18 1.54 1.26 59 58 1.55 1.52 11 9 2.34 1.91 21 16 2.08 1.59 19 18 2.36 2.23 23 49 1.32 2.81 5 19 0.70 2.65 9 7 0.99 0.77 2 1-- .. 30 28 1.47 1.37 15 15 1.15 1.15 19 16 0.95 0.80 12 8 1.82 1.22 14 17 1.35 1.63 55 92 2.16 3.61 212 188 3.59 3.19 168 ... 78 -- -- 1,064 1,161 0.83 0.91 Sel1koff, et al. 1979a ^Expected deaths are based upon white male age specific mortality data of the U.S. National Center for Health Statistics for 1967-1975 and extrapola tlon to 1976. cRates are not available, but these have been rare causes of death in the general population. BE: Best evidence. Number of deaths categorized after review of best available Information (autoosy, surgical, clinical) DC: Number of deaths as recorded from death certificate Information only. i ( tent in the general population (asbestosls and mesothelioma). Since meso.nelloma Is not a cannon cause of death In other than asbestos-exposed Indi viduals, Its misdiagnosis on the death certificates of general population has little significance. However, as It Is a major cause of death of asbes tos-exposed workers. Its proper diagnosis Is necessary In order to evaluate the extent of occupational disease. Moreover, were It not to be properly j characterized one would conclude that cancers of the liver and pancreas were elevated from asbestos exposure. Thus, one would have to consider excesses \ at these sites (as misdiagnosed on death certificates) rather than mesothej Homa In evaluating abdominal cancers. Otherwise, the use of best evidence | rather than death certificate Information Is a minor factor In the evalua- | tlon of gastrointestinal cancer. For example, among cancers of the esopha* ^ gus, stomach, colon, and recttea In 2,271 consecutive deaths In insulation ( Vers, 112 were listed at these sites on death certificates. Best evi ct Indicated that 118 occurred. This difference would have little effect upon the calculation of gastrointestinal cancer. On the other hand, peri toneal mesothelioma per $e was specified In only 24 deaths where best evi dence Indicated 112 occurred from this disease. The difference was largely made up from overdiagnosed cancer of the pancreas (26 cases), cancer of the liver (14 cases), and from 55 mesotheliomas unspecified as to site. The large number of deaths allows an analysis to be made of the onset of effects as related to time from first-.exposure. Figure 3 depicts the excess asbestos-related lung cancers and mesotheliomas according to time from onset of exposure. It Is seen that an Important rise In bronchogenic carcinoma occurs only after 26 years and mesothelioma and asbestosls after 30 years. This long-lapsed period Is seen In Individuals exposed continuously to rela, ly high concentrations of asbestos. At lower exposures, longer periods C-69 FIGURE 3 The Excess, Asbestos-related Mortality Rates for Lung Cancer and Mesothelioma According to Time from Onset of Asbestos Disease Source: Sellkoff, et al, 1979a S* mesothelioma lung cancer ( C-70 ( n exposure onset to tumor development would be expected end, thus, studi that do not provide adequate follow-up can be misleading. Among other groups of Insulation workers, high rates of cancer, particu larly bronchogenic carcinoma and pleural or peritoneal mesothelioma, have been reported by Sellkoff (1976). In this study 632 New York and New Jersey Insulation workers, 20 or more years from onset of exposure were observed from January 1, 1943 through December 31, 1974. Of these, fewer than 300 Individuals were Included In the larger study of 17,800 Insulation workers. With a much longer observation period, even more severe effects were seen. Similarly, a study by Elmes and Simpson (1971, 1977) In the United Kingdom portrays a more severe mortality experience, particularly for lung cancer over a period of time, 1940-1975. Some data on exposure of U.S. Insulation workers exist. These have been ( lewed by Nicholson (1976) and are summarized In Table 22. Estimates of t average exposures were made on the basis of current measurements by four laboratories of fiber concentrations during work activities thought to be typical of those of past years and information on product composition and usage. Time-weighted average concentrations of 10 to 15 f > 5 m/ml and 15 to 20 f > 5 uffl/ml were suggested for conmerclal construction and marine work, respectively. It was noted that, while these average concentrations were not extraordinary, peak concentrations could often be very high and exceed 100 f/ml. At Lyon, In 1972, Cooper and Mledema (1973) reported, peak concentrations may be high for brief periods, while time-weighted averages are often deceptively low." To the extent possible these high exposures were taken Into account and the time-weighted average exposure was largely due to peak exposures. This averaging and the extrapolation to eari 'r years Introduce uncertainties In the estimate. However, the above C -71 TABLE 22 Summary of Average Asbestos Air Concentrations during Insulation Work Research Group Average Fiber Concentration f/ml Light and Heavy Construction Marine Work Average concentrations of fibers longer than 5 u* evaluated by membrane filter techniques and phase-contrast Microscopy Nicholson (1971b) Balzer and Cooper (1968) Cooper and Balzer (1968) Ferris, et al. (1971) Harries (1971a,b) 6.3 2.7 6.6 2.9 8.9 Average concentrations of all visible fibers counted with a konlneter and brlght-fleld microscopy Murphy, et al. (1971) Fleischer, et al. (1946 ) 8.0 30-40 Estimates of past exposure based on current membrane-filter data Nicholson (1976) 10-15 C-72 ( .weighted avenge exposures ire felt to be accurate within a factor of . This is suggested by the good agreement among the exposure estimates and measurements of four different laboratories and by the good agreement of insulator dose-response data with that of other groups. Factory Employment: An early study of workers from an asbestos products factory (Mancusco and Coulter, 1963} showed a significant excess In total mortality, with important contributions to excess death rates from asbesto- < sis, cancer of the lung, bronchus, and trachea, and neoplasms of the diges tive organs and peritoneum. In this latter group of deaths, an Important factor was peritoneal mesothelioma. While in excess. Increases in cancer of the esophagus, stomach, colon, and rectum did not have statistical signifi cance. There was a consistent Increase in the mortality rate with increas ing length of employment in the asbestos industry for all causes of death ( especially for malignant neoplasms and asbestosis. Additional studies of factory employees (Enterline, et al. 1972; Hender son and Enterline, 1979) focused upon a group of retirees from several plants of a major asbestos products manufacturing company. It shows a simi lar pattern of mortality. Table 23 lists standard mortality rates (SMRs) by cause in two time periods. The usual asbestos cancers and asbestosis are seen as significant causes of death. Here, too, a correlation was found be tween total dust exposure and excess mortality for both malignant and nonma11 grant disease. Table 24 lists the data for lung cancer and shows a lin ear relationship with exposure. These authors (Enterline and Henderson, 1973} suggested earlier that crocidolite may have a higher carcinogenic potential (for lung) than amosite or chrysotile. The later analysis (Henderson and Enterline, 1979) shows I at individuals in the textile departments of the company (chrysotile only) C-73 mgmm. mm TABLE 23 Observed Deaths and SMRs for Selected Causes of Death by Period of Follow-wo for 1,075 Males Retiring from a U.S. Asbestos Company from 19*1-67 and Followed through 1973* Cause of Oeath 1941-1973 Observed SMR Oeaths 1941-1969 Observed SMR Oeaths 1970-1973 Observed sm Oeaths Al1 causes Cancer (140-205 )b Digestive (150-159) Respiratory (162-163) All other cancers Stroke (330-334) Heart disease (400-443) Respiratory disease (470427) Pneumoconiosis and pulmonary fibrosis (523-525) Asbestosls (523.2) All other causes Death certificates not located 781 173 55 63 55 74 321 68 31 19 113 32 120.4 159.0 137.8 270.4 120.6 96.4 106.5 616 138 46 49 43 48 269 173.0 54 '-- 92.5 25 16 96 11 115.8 154.5 136.1 270.7 115.0 76.7 108.4 165 35 9 14 12 26 52 178.2 14 m 6 --3 94.6 17 -- 21 141.6 179.5 147.S 269.2 146.3 183.1 97.7 155.6 .. 62.5 -- aHenderson and Enterline, 1979 b01sease code C-74 TABLE 24 Lung Cancer Mortality Rates According to Dust Exposure* Cuawlatlve Dust Exposure (*ppcft> - years) ga <125 125 - 149 250 - 499 500 - 749 750 197.9 180.0 327.6 450.0 777.8 Henderson and Enterline, 1979 ^Million particles per cubic foot C-7S have a lower lung cancer SMR than those In the pipe department (chrysotlle and crocldollte) for equal dust exposures. However, no conclusions could be drawn from an analysis of the mortality rates of all Individuals exposed, or not exposed, to crocldollte. Since the follow-up of this population began only after the cohort members reached age 65, survivor effects may be of Importance. For example, those Individuals who smoke cigarettes and are thus at higher risk for lung cancer may be preferentially excluded by virtue of death before age 65 because of smoking-associated disease such as myocar dial Infarction. Further, the limited number of mesotheliomas (5 of 781 deaths) found In the latest followup of this group could be due to the high incidence of mesothelioma at age SO to 65, 30 to 45 years from onset of first employment (see Figure 3). Mortality data were correlated with esti mates of previous dust concentrations In terms of millions of particles per cubic meter of air (mppcf). No Information was provided on possible fiber concentrations. A study of the largest factory of the company studied by Enterline, et al. (1972), but not limited to retirees, shows a considerably different mor tality pattern (Nicholson, 1976; Nicholson, et al. 1980b), All 689 mainte nance and production employees on January 1, 1959, who were first employed at least 20 years, earlier were followed through 1976. In this group, 274 deaths occurred, whereas 188.19 were expected. Fourteen pleural and 12 peritoneal mesotheliomas accounted for nearly 10 percent of the deaths, most occurring before age 65. A strong correlation with estimated dust exposure was seen In deaths from asbestosls, but not with the asbestos-related malig nancies. Gastrointestinal cancer was especially high In the lowest of four dust categories (11 observed versus 3.15 expected) and only elevated slight ly In the higher exposure categories. In the highest dust category, the C-76 textile mill, cancer was not dramatically Increased, but 40 percent of the eaths were from asbestosls. Individuals In this department tended to..dje of nonmallgnant disease before reaching the age of greatest risk for cancer. A study by Weill, et al. (1979) of two asbestos cement product facili ties has also been published. Here, the mortality experience of 5,645 em ployees Mas followed for at least 20 years. It shows excess mortality for lung cancer In the highest exposed groups but deficits of death from all causes (as great as 40 percent) in all categories. Of the group 3,354 (68 percent) were employed for less than 2 years. Thus, exposures were limited for the majority of the cohort members. Further, as most of the followup Involved observations prior to 25 years from first exposure (18,117 person years at risk <25 years from Initial exposures versus 5,910 person-years >25 years), there was limited risk from asbestos disease in the group. Of most ronseouence, however, 25 percent of the cohort was untraced and all untraced ire considered alive. This could explain the large mortality deficits in all categories other than lung cancer and Invalidates the study for any use In establishing dose-response relationship. A final significant U.S. factory study Is that of Seidman, et al. (1979) which extends an earlier study (Sellkoff, et al. 1972) and documents the experience of workers exposed only to saoslte asbestos In the production of Insulation materials, primarily for use aboard naval vessels. Overall mor tality shows patterns similar to other heavily exposed groups, with 594 deaths observed versus 368.62 expected. Lung cancer was more than five times the number expected, and 16 deaths from mesothelioma occurred. Of particular Importance In this study Is the finding that Individuals employed for periods less than 6 mon.ths had significant excess of lung cancer (Table 25). SastroIntestinal cancer was also elevated for those with exposures of C-77 >ss than 6 months {is observed versus 10.6 expected), but the difference d not have statistical significance.* Further, there was not an Increasing risk with time of employment as In the case with lung cancer. Some data exist that would Indicate the air concentrations of asbestos to which workers In a factory, which operated In Paterson, New Jersey, from 1941 through 1954, were exposed. Following cessation of operations there, two similar plants were opened elsewhere, using the same equipment and manufacturing the same product with the same materials. As In the Paterson factory, dust control was Inadequate In the newer plants. These continued operation through 1971 In one ease and 1975 In the second. During 1967, 1970, and 1971 asbestos fiber concentrations In the plants were measured by the National Institute for Occupational Safety and Health (NIOSH, 1972), and the results are presented In Table 26. The overall arithmetic average axpo-re was 34.9 f/ml with a range from about 20 to 80. Using 40 f/ml, as an Imate of the fiber count In the Paterson factory, one calculates the average dose received by those employed for less than 6 months to be no more than 120 f/mlmonths, the same dose as would be received by a worker employed 20 years at an exposure of 0.5 f/ml. Of significance, also. Is that the mesothelioma risk Is less than that of Insulators (3 percent versus 7 percent). Since times from onset of exposure to amoslte are comparable for each group, the presence of amoslte In Insulation materials cannot explain the high rate of mesothelioma among insulators. In Great 8r1ta1n, a well-studied factory population (Doll, 1955; Knox, et al. 1968) provides useful Information because of the availability of environmental information. The mortality experience of this group has been recently updated (Peto, et al. 1977). Workers exposed prior to 1933 (before H,ist concentrations were significantly reduced) had a marked excess of lung 079 TA8LE 26 Asbestos Fiber Concentrations In Two Amoslte Insulation Production Facilities*.b Operation ASBESTOS INSULATION PLANT Y 1967 No. Of Mean Samples 1970 No. of Mean Samples Mixing Forming Finishing Inspection and Packing Miscellaneous 107.0 98.9 32.2 13.3 3 12 4 2 27.7 2 24.1 13 16.8 2 13.0 8 21.0 14 Operation ASBESTOS INSULATION PLANT X 1967 No. Of Mean Samples 1970 No. of Mean Samples Mixing Forming Curing Finishing Inspection and Packing Miscellaneous 163.0 33.3 2.5 44.6 16.7 S 18 1 3 7 36.2 25.7 31.0 34.8 17.9 13.8 *NI0SH, 1972 &A11 samples expressed as f > 5 ufli/ml* 3 3 1 4 3 2 1971 Ho. of Mean Samples 46.3 7 25.2 32 15.0 17 11.0 19 2.7 5 1971 NO. Of Mean Samples 74.4 11 50.6 39 14.4 5 39.5 26 22.8 15 16.6 24 I 080 L inanavn m cancer (25 observed versus 4.63 expected). Other cancers were elevated, but so greatly. Of significance, however, Individuals employed after 1933, . even after January, 1951, were found to have an excess risk of lung can--, eer. These data were analyzed by Peto (1978) In relation to measured and . . estimated fiber concentrations. Exposures averaged about 10 f/ml after 1933 and were virtually exclusively chrysotlle. Using a linear dose-response relationship for lung cancer and pleural mesothelioma, he estimated that a 2 f/ml exposure for SO years would cause approximately 10 percent of male asbestos workers to die from asbestos-related disease. It should be noted that data available for analysis were very limited and the estimate was based on extremely small numbers (14 deaths from lung cancer,- 4 from meso thelioma, and 17 from nonmallgnant respiratory disease). Furthermore, few Individuals In the cohort were more than 35 years from onset of exposure and at a period of highest risk from asbestos disease. ^ Another factory population has been extensively studied (Newhouse, 1969; puse, et al. 1972; Kewhouse and Serry, 1976, 1979). Exposures were to chrysotlle, crocldollte, and amoslte. Table 27 lists the mortality experi ence of both men and women according to estimates of fiber exposure (no details are provided as to the method of estimation) (Newhouse and Berry 1979). lung cancer, gastrointestinal cancer, and mesothelioma are signifi cantly elevated In the long-term (>2 years) or severe exposure groups. It has been estimated (Newhouse and Berry, 1976) that as much as 11 percent of this entire group will die of pleural or peritoneal mesothelioma. Among fe male workers, cancer of the breast and cancer of the ovary were significant ly higher (p 0.05). Mining and Milling: Three studies exist showing mortality patterns In the mining and milling of pure chrysotlle asbestos. A series of studies C-81 IJftcf >* * e 00 9i IO ( US* I fc, * e ^ gj m* r p* to ST 00 x 00 m w *mw* s* * cw V 00 *s o *s^i 5 Si UXJ o_C<SJ Jo VI vS|> VI 5 wt ws t L. *io e 1 (eJ tf) m g S o fc _ 9 -S(i W Ml (3*5. o w o*> we- C-82 r i TABU _< . .ontlnued) Y >i * 1 II 5 V UKd 2 ft % tO <n ru 41 5 vu v4>i iaus e 41 e OwS i g S S8 S wu s* 6? .1fi 4U t^T 041 3X C-83 (McDonald and Liddell, 1979; McDonald, et al. 1971, 1980) of 10,939 Male Canadian mine and mill employees show excess mortality, particularly of the respiratory system. Table 28 lists the mortality for those Individuals in the cohort that achieved 20 or more years from-first employment. Standard mortality ratios were calculated from the expected number of deaths In the province of Quebec. The risk of death from lung cancer Increases linearly with dust Index with no evidence of a threshold [relative risk 1 0.0014 (mppcf-years)]. The mortality for esophogeal and stoaaeh cancers shows a strong relationship with dust Index, but that of colonrectal cancer does not. Pleural mesothelioma was a cause of 11 deaths to 1975. The use of Ouebec mortality statistics may underestimate the actual risk as the earliest report by McDonald (McDonald, et al. 1971) stated that lung cancer mortality In the five counties near the asbestos mines was only two-thlrds of the province as a whole, the rates of which would be dominated by the urban center, Montreal. The effect of urban<<ura1 difference on the rates of cancer at other sites Is not known. Additionally, It Is not stated In the publication how the 10 percent of the cohort that was untrscad was treated. All data on exposure are given in terms of millions of particles per cubic foot (mppcf). While earlier work described the difficulties of converting particle counts to f/1 (Gibbs and LaChance, 1974), It 1$ now sug> gested that a conversion factor between 1 and 5 f/m per mppcf may be appro. pH ate (McDonald, et al. 1980). A Soviet study of the health effects of chrysotlle mining and milling 1* that of Kogan, et al. (1972). Overall excess mortality of cancer of the respiratory or digestive tract was seen, particularly in the groups aged $0 years or older (and presumably 30 or more years from first exposure), among these, stomach cancer mortality in male miners Is Increased 2.5 times and C-34 d u ll lip tM r t m 4 H arU H ty l C fc rjn ttiu NUfMt, ItlO-TS S ss icjsajue s s c s o * d Om*Onil#* #(W9 * 5f* 3 1 2 i I a-- a*jj aaasjss a **>: m X 2 =- * S I 8 r 8.H , SSn5*S 5w 3 3 3 If * ft ft aassfts ft 3 2 2 S" 8 M m 9a miO m mm I I 5 - sasz-a g S 2X * J * s H3 8 8 388828 St 8 3 3 8 3 * W2 M * V 1:1 o mm R Ht QaOOi * " RR22"S g 9 *X I? u 81 ff*! 3 a jl4d s 8-- --2 8O8a8O8~8~8J 8JI 3 3 o' 8 88 -- -I II I SM <r pn K K g* ~ R nown ne i III a i % a1 * u5 1 C-85 that of female workers by 3.6 times. The corresponding Increases for femalt and male mill workers are 4.3 and 19.9 times expected. Additionally, Intestlnal cancer is elevated among the 50* year group 4.3 times for male miners, 6.9 times for female miners and 14.3 for women mill employees. Unfortunately, data on the number of deaths are not provided. Mo cases of mesothelioma are reported. Anthophylllte mining has also been found to produce a high risk of bronchogenlc carcinoma (Meuman, et al. 1974). In a study of, miners exposed to fibers of cummingtonite-grunerlte ore series (In which amoslte Is formed), Slllam, et al. (1976) reported excess malignant respiratory disease (10 observed versus 2.7 expected) at an average air concentration of 0.25 f/ml. No cohort mortality studies exist for the mining or milling of crocidollte or amoslte. In the above studies of chrysotlle mining and milling, mesothelioma was present to much less a degree than In the following three Instances: a fac ( tory using chrysotlle exclusively, (4 percent of 20* year employees) (Peto, 1978); the largest U.S, chrysotlle using facility (10 percent) (Nicholson, et al. 1979); or Insulation work using chrysotlle and amoslte (7 percent) (Sellkoff, et al. 1979a). It appears that as the fibers are manipulated through milling, processing, and use, their carcinogenic potential In creases. Whether this is related to a reduction In fiber size or other fac tors Is yet to be definitively established, particularly in view of animal data which Indicates a reduction In carcinogenic potential following ball milling (see Animal Inhalation section). 8eeause of Its relevance to Ingestion, a summary of the available data on gastrointestinal cancer and peritoneal mesothelioma is given in Table 29. i C-86 G*itroltcstinl CMccr In O ccupillM ilIt C>pai(4 A U ntoi U orttri if]>|J2* s a Ja - a O S nU Mft H ft ft f2t *** < St * RR a O A NO 2 ** S T 2S n R5 2 sas s a a* -24.t o * ft ft i! g a s - | si ^ z 8 5Sfi i* w*2S8 55S5 asai 555 ii1 NQ ^ r? a || 111 -o-iV8stgi K^*! s*m i*ia |3A aJi s4 s2. a 5ft Ja c* C ?* ~ ;M V a1ec "?*V --?^* ^ V 3s a t = 5 ts S a S| S i :* 1 SiIb* ?vi> i52M i*W; ii 1 a I] ! I sa k C-B7 (2 jf! i i! * SI 1 Sm 85 ii !| u1- I-a. a i i (1 K U-- mm m I2 .1 ) 3 * *a M Sm = tk-Eiisli IS i ii 52 H it if C-88 f Indirect Occupational Asbestos Exposure: In 1968 It was pointed out by ries (1968) that shipyard workers other than insulators were at risk from isoestos disease. Among Oevonport Dockyard employees, five cases of meso>. thelioma were found among men who had not been 'asbestos workers" but had followed other trades In the yard. These men presumably had been inadver tently exposed to asbestos merely by working In the same shipyard areas where asbestos had been used. Continuing to follow this group, Harries later documented 55 cases of mesothelioma In this shipyard alone, only 2 of which occurred In asbestos workers (Harries, 1976), and 1 of which occurred in a man who had previously sprayed asbestos. A study of the distribution of all verified cases of mesothelioma found In Scotland between the years of 1950 and 1967 Is also revealing. Of 89 cases available for study, 55 were In shipyard employees, dockers, or naval personnel. Of the 55, again only 1 was an asbestos Insulation worker (McEwen, et al. 1977). A study by Edge (1976) of men who had worked in a shipyard In Barrow, ...gland, attempted to establish a risk of low-level asbestos exposure on a population basis. He selected 235 shipyard workers with pleural plagues but no parenchymal fibrosis on X-ray, and followed their mortality experience from 1970 through 1973. Seventy died, 17 of mesothelioma and 13 from lung cancer, 2.6 times greater than expected. However, the relevance of these data have been called .into question by the possibility of bias In the selection of the 235 cases (Edge 1979). The previously mentioned radiological evidence (see Indirect Occupa tional Asbestos Exposure section) that asbestos concentrations in general shipyard work (Selikoff, et al. 1979a) or maintenance activities in a chemi cal factory (LI 11s and Selikoff, 1979) are sufficient to produce fibrosis points to the existence of a widespread carcinogenic problem from indirect .bestos exposures. C-89 Environmental Asbestos Disease: Wagner, et al, (1960) reviewed 47 cases of mesothelioma found in the Northwest Cape Province, South Africa in the * previous 5 years. Of this number, roughly half the cases were In people who had worked with asbestos. Virtually all the rest were in individuals who had, decades before, simply lived or worked in an area of asbestos mining (one living along a roadway in which asbestos fibers were shipped). This germinal observation demonstrated that asbestos exposure of limited inten sity, often intermittent, could cause mesothelioma. The hazard was further pointed out by the findings of Newhouse and Thomson (1965), who showed that mesothelioma could occur among people whose potential asbestos exposure con sisted of their having resided near an asbestos factory or In households of asbestos workers. Twenty of 76 cases from the files of the London Hospital (1917 to 1964) were the result of such exposure; 31 were occupational in origin, and asbestos exposure was not identified for 25. Both pleural and peritoneal mesotheliomas have been found to occur from environmental asbestos exposure. For example. In the neighborhood and fam ily cases documented by Heben and Plstawka (1967), two of three family con tacts and two of eight neighborhood mesotheliomas were peritoneal. In gen eral, a greater percentage of environmental mesotheliomas compared to occu pational are pleural in origin. This, however, may be the result of a greater propensity for peritoneal mesotheliomas to be misdiagnosed. In occupational circumstances, 40 percent of pleural mesotheliomas were cor rectly classified on death certificates versus only 21 percent of peritoneal mesotheliomas (Selikoff, et al. 1979a). Synergism and/or Antagonism Asbestos exposure and cigarette smoking have been found to act synergistically to produce dramatic increases in lung cancer over that from expo- C-90 e to either agent alone. In a prospective study by Hwmond, et al. i1979) of 17,800 Insulation workers, smoking histories were solicited from all Individuals during 1966 prior to observation. Of 12,051 workers who passed the 20-year point since entering the trade before or during the 10-year observation period, January l, 1967 to December 31, 1976, 891 re ported they had never smoked, 488 had smoked only a pipe and/or cigars, and 6,841 gave a history of cigarette smoking. No Information was available from the remaining 3,831. Using data of the American Cancer Society (ACS) on age- and calendar year-specific cancer rates among smokers and nonsmokers In a prospective study of more than one million people In the United States, It was possible to make smoking-specific comparisons of the mortality exper ience of insulation workers with nonasbestos exposed Individuals In the gen eral population. Those Insulation workers who claimed never to have smoked .garettes were found to have an Increased risk of death from lung cancer -ompared with nonsmokers In the general population, although there were relatively few deaths, 8 observed versus 1.3 expected. However, among those with a history of cigarette smoking, the risk was also increased and Its effect was large, 268 deaths being recorded versus 4.7 expected. Among non cigarette smokers in the general population Table 30 lists the death rates and mortality ratios of smoking and nonsmoking asbestos workers compared to the ACS control population. Asbestos exposure appears to multiply the risk of death of lung cancer by four to six times, Irrespective of smoking hab its. When that risk Is already high, as In cigarette smokers, the result Is catastrophic. An earlier study by Sellkoff, et al. (1968) indicated that the risk of death from lung cancer in cigarette-smoking asbestos workers was 92 times that among Individuals who were neither exposed to the fiber nor .moked cigarettes. C-91 I TABLE 30 Age-Standardized lung Cancer Death Rates* For Cigarette Smoking and/or Occupational Exposure to Asbestos Dust Compared with No Smoking and No Occupational Exposure to Asbestos Dust*3 Group Exposure to Asbestos History Cigarette Smoking Oeath Rate Mortality Difference Mortality Ratio Control6 Asbestos workers Control Asbestos workers No ves NO res No 11,3 0.0 No 58.4 47.1 ves 122.6 111.3 Yes 601.6 *590.3 1.00 5.17 10.85 53.24 *Rate per 100,000 man-years standardized for age on the distribution of the man years of all the asbestos workers. Number of lung cancer deaths based on death certificate information. bHammond, et al. 1979. cThe central population Is a group of 73,763 white, male workers exposed on the job to dust, fumes, vapors, chemicals, or radiation. C-92 Cancers of the larynx, pharynx and buccal cavity, and of the esophagus i insulation workers are' also associated with cigarette smoking (Hanmond, et al. 1979). Among SO deaths due to tumors of these sites, none were among nonsmkers and 3 were among Individuals who smoked only pipes or c1gar.s. Mesothelioma of the pleura or peritoneum and cancer of the stomach, colon, and rectum, however, were unrelated to smoking habits. It Is worth noting that In these studies by Sellkoff and Hanmond over 200 excess deaths occurred from peritoneal mesothelioma and gastrointestinal cancer (excluding esophagus) In 2,271 deaths of Insulation workers. Here smokingrelated lung cancer not a factor, abdominal cancer deaths would dominate the mortality experience of this group of asbestos workers. Other studies have substantiated the synergistic effect of cigarette smoking. Berry, et al. (1972) obtained retrospective smoking histories on a group of asbestos workers and analyzed their mortality according to smoking tablts over a 10-year period of time. The results Indicated that the com bined effect of cigarette smoking and asbestos exposure on the development of lung cancer Is multiplicative rather than additive. Although synergistic effects have been documented for bronchogenic car cinoma, only cigarette smoking has been investigated in the etiology of abdominal cancers. The possibility exists, of course, that these tumors too could have a multiple factor etiology and that other contaminants, ingested with asbestos, may potentiate tumor development. Additionally, some nonmallgnant asbestos effects are related synergistlcally to cigarette smoking. Among a group of factory employees It was found by Weiss (1971) that evidence of fibrosis, as manifest on X-rays, was In creased among Individuals who smoked cigarettes compared to nonsmokers. Deaths due to asbestosls appear also to be Increased In cigarette smokers compared to nonsmokers (Hanmond, et al. 1979). C-93 { l In animal experiments, exposure to benzo(a)pyrene (BP) and asbestos >a act synerglstlcally. Pylev and Shabad (1973) reported, that Intratrachea Injections of 6 mg of chrysotlle onto which was absorbed 0.144' mg of 8P (from a benzene suspension) and 2 mg of chrysotlle coadministered with 5 mg BP produced malignant tumors In 29 percent and $4 percent of rats, respec tively. Administration of 6 mg of chrysolite or 5 mg BP yielded no tumors. Miller, et al. (1965) found Intratracheal Injection of chrysolite with BP to Increase tumor yield over that of BP alone while amoslte appeared to have little such effect. No data exist on antagonistic or prophylatlc compounds In relation to animal or human disease. In vitro experiments by Schnltzer, et al. (1971) have shown that hemolysis of red cells can be Inhibited by coating the fibers with ionic polymers such as carboxymethylcellulose. Fiber Size Considerations Experimental systems, particularly those used by Stanton and Wrench (1972) and Pott, et al. (1976), Indicate a significantly reduced carcino genicity of fibers as the length Is reduced or the diameter Increased. On the other hand, human data suggest an Important role for small fibers. From analyses of tissue samples from 29 mesothelioma cases, Sebastlen, et al. (1979) found that larger fibers, often amphlboles, tend to be found In the lung parenchyma. In contrast. In the pleura, the fibers were finer and shorter and generally chrysotlle. The mean length In' pleura was 2.3 um and that of the lung 4.9 um. In 20 pleural samples of 29. autopsy cases in which asbestos fibers were found, chrysotlle was Identified as the only fiber in 8 and only a trace (<1 percent) of amphlboles was found in 2 others. In con trast, significant percentages of amphlbole fibers (>18 percent) were found 'in 26 of 29 lung parenchyma samples from the same cases. C-94 I 1 In an examination of the mortality of workers In different types of asbestos Industries, significant differences occur that may be related 'to fiber size. In amoslte and chrysotlle mining, few mesotheliomas are seen, whereas, in manufacturing and end product use, large percentages of deaths occur from this tumor. For example, chrysotlle mining and milling, while related to a significantly Increased risk of death from lung cancer and asbestosls (McDonald and Liddell, 1979; Nicholson, et al. 1979), has not been associated with an extraordinary mesothelioma risk. Similarly, amoslte mining and milling does not appear to significantly Increase the risk of mesothelioma, while crocldoilte mining and milling does (Webster, 1970). On the other hand, the manufacture of amoslte products is associated with a significant risk of death from mesothelioma, 3.S percent of the deaths of individuals 20 or more years from first employment being from this cause 'Seidman, et al. 1979). Further, Insulators who were exposed to chrysotlle nd amoslte, but never to crocldoilte (Sellkoff, et al. 1970) have 9 percent of their deaths, 20-plus years from onset of exposure, from mesothelioma (Sellkoff, et al. 1979a). As neither amoslte nor crocldoilte can account for this extraordinary risk, chrysotlle must contribute significantly. This Is also borne out by observations of the mortality of workers In a chrysotlle using factory. 4.3 percent of long-term deaths were from mesothelioma In a facility using 5,000-6,000 tons of chrysotlle, approximately 50 tons of amoslte, and less than 4 tons of crocidolite annually (except for 3 years when 375 tons of amoslte were used annually) (Robinson, et al. 1979). Much of these differences In risk may be accounted for by the differ ences in fiber size distributions In the three work environments rather than y fiber type. The greatest percentage of longer and thicker fibers would ;cur in the work environment of miners and millers. As the asbestos is C-95 used In manufacturing processes, it Is broken ipert as it is incorporated finished products. Ourlng application or removal of insulation products, is further manipulated and the fiber reduced In length and diameter, tthese`smaller fibers can readily be carried to the periphery of the lung, penetrate the visceral pleura and lodge in the visceral or parietal pleura, they may be of greater Importance in the etiology of mesothelioma, even though longer fibers, once there, are more carcinogenic. In the case of crocldolite, fine fiber aefosols are produced even. In mining and, thus, all uses of that fiber are associated with mesothelioma. { C-96 CRITERION FORMULATION Istlng Guidelines and Standards The current Occupational Safety and Health Administration (OSHA) stan dard for an 8-hour time-weighted average (TWA) occupational exposure to asbestos Is 2 fibers longer than S microns In length per milliliter of air (2 f/ml or 2,000,000 f/m*). Peak exposures of up to 10 f/ml art permitted for no more than 10 minutes (29 CFR 1910.001). This standard has been in effect since July 1, 1976, when It replaced an earlier one of 5 f/m1 (TWA), In Great Britain, too, a value of 2 f/ml is the accepted level, below which no controls are required (BOHS, 1968); the British standard. In fact, served as a guide for the OSHA standard (NIOSH, 1972). The British standard was developed specifically to prevent asbestosis among working populations; data were felt to be lacking that would allow a termination of a standard for cancer (BOHS, 1968). Unfortunately, among ccupatlonal groups, cancer Is the primary cause of excess death among work ers (see Carcinogenicity section). Three-fourths or more of asbestosrelated deaths are from malignancy. This fact has led OSHA to propose a lower TWA standard of 0.5 f/ml (500,000 f/m3) (29 CFR 1910.001). The National Institute for Occupational Safety and Health (NIOSH), in their cri teria document for the hearings on a new standard, have proposed a value of 0.1 f/ml (NIOSH, 1976). In the discussion of the NIOSH proposal, It was stated that the value was selected on the basis of the sensitivity of ana lytical techniques using optical microscopy and that 0.1 f/ml may not neces sarily protect against cancer. Recognition that no Information exists that would define a threshold for asbestos carcinogenesis was also contained In the preamble to the OSHA proposal. The existing standard In Great Britain has also been called Into question by Peto (1978), who estimates that asbes- C-97 tos disease may cause the death of 10 percent of .workers exposed at 2 f/ml for a working lifetime. A fiber concentration limit of 1.0 Y/ml has recently been published In Great Britain (Advisory Comnlttee on Asbestos, 1979). The existing Federal standard for asbestos emissions Into the environ ment prohibits "visible emissions" (40 FR 4S291). No numerical value was specified because of difficulty In monitoring ambient air asbestos concen trations In the ambient air or In stack emissions. (Time-consuming and expensive electron microscopy Is often required.) Same local government agencies, however, may have numerical standards (New York, 27 ng/m3 for example). No standards for asbestos In foods or beverages exist even though the use of filtration of such products through asbestos filters has been a com mon practice In past years. Asbestos filtration, however, is prohibited or limited for human drugs (41 FR 16933). Current Levels of Exposure As detailed In the Exposure section, asbestos Is a ubiquitous contami nant of our air and water. Air concentrations over 24 hours In metropolitan areas usually are less than S ng/m3 but can range up to 20 ng/m3. Val ues up to 50 ng/m3 are found during daytime hours In locations where con struction activities and traffic can be contributing sources. A significant fraction of the fibers Inhaled can be brought up from the respiratory tract and swallowed. This leads to an Ingestion exposure from air sources of up to 0.1 u9/day, although most of the population exposure 1$ from 0.01 to 0.05 ug/day. Water concentrations of asbestos are usually less than 10 fibers of all sizes per liter although significantly higher values (10 f/1) have been found in circumstances where water systems have been in contact with C-98 f tiform minerals or where contamination of the water supply exists. .r mass concentrations corresponding to fiber concentrations are usually less than 0.01 ug/1 but could exceed l g/l. Thus, direct water ingestion usually leads to exposures of less than 0.02 ug/day. Clearly, point source pollution can cause both air and water concentra tions to exceed the above values. Such instances are discussed In the Exposure section. Special Groups at ftlsk Special groups at risk may Include neonates and children; however, no data exist on the relative sensitivity to asbestos of Infants and children undergoing rapid growth. Concern exists because fibers deposited In the tissues of the young may have an extremely long residence time during which malignant changes could occur. In addition, risk could be influenced by cferent1a1 absorption rates which have not been fully studied at this time. Individuals on kidney dialysis machines may also be at greater risk as fluids, potentially contaminated with asbestos fibers can enter the blood stream directly or, in selected instances, the peritoneal cavity (peritoneal dialysis). Although no synergistic effects have been Identified in the etiology of asbestos-related gastrointestinal cancer, they cannot be ruled out. Thus, people exposed to other carcinogens. Initiators, or promotors could be at increased risk. An increased risk Is also associated with increased exposure to asbestos in water in municipalities such as San Francisco or Seattle .where asbestos occurs naturally in water, in cities where there Is an interaction between S's* s ' ' aggressive water and asbestos-cement pipe, or in cities whose water may be contaminated as a result of asbestos operations. Also, the use of asbestos y C-99 cement products for the collection of water, such as In cisterns In the Virgin Islands or in roof run-offs In tropical areas, Increases exposure. Basis and Derivation of Criterion As previously discussed, no definitive studies, either animal or human, exist that would establish risk levels from Ingestion of asbestos fibers. Those studies published provide both positive and negative data, but all have methodological limitations. In the case of the human studies, these Include observations on only recently exposed Individuals, small study groups, low exposures, population mobility, uncertainty over the effect of confounding variables, and Inappropriate control populations. Animal stud ies have usually been conducted with very small numbers of animals, have lacked proper pathology, used limited doses of asbestos, and poorly defined the materials Ingested. On the other hand, human studies of workers exposed to airborne asbestos uneaulvocally demonstrate an excess risk of gastrointestinal cancer In vir. tually all groups surveyed. A route of exposure to the gastrointestinal tract from such exposures Is also clear from the fibers cleared from the lung and bronchial tract and subsequently swallowed. Using Information on airborne exposures to workers, It is possible to estimate an approximate exposure level to. the gastrointestinal tract from estimates of airborne asbestos concentrations. This, however, involves the use of data having, in some cases, significant uncertainties and, thus, the criterion level on asbestos In water that will produce a specified risk cannot be established with high precision. Experimental uncertainties exist as to the air concentrations In fibers longer than 5 um/ml to which workers were exposed in past years, the conver sion of these >5 u<n fiber concentrations to concentrations of fibers of all C-lOO *es In air, ind on the size distributions of water and airborne asbestos s. Information is also lacking on the Importance of fiber size in the production of human cancer. On the one hand, longer fibers are more car*. t cinogenic in experimental systems although quantitative data are limited* On the other hand, smaller fibers appear to more readily cross body barriers to reach sites of Importance for human carcinogenesis-. The relative Impor' ;J tance of these two factors cannot be accurately estimated. v . / A substantial body of data exists which shows Increased incidence of cancer of the esophagus, stomach, colon, and rectum or peritoneal mesothe lioma in humans exposed to asbestos occupationally, for several of these groups, data exist on the approximate airborne fiber concentrations to which individuals were exposed (see Effects section). These human data will serve as the primary basis for a standard of asbestos In water. Experimental data 'see Pharmacokinetics section) indicate that a major fraction of the asbes- is deposited In the lungs is subsequently swallowed. In this section, the dose to the gastrointestinal tract of four occupational groups will be cal culated from knowledge of the air concentrations to which the workers were exposed and the assumption that all the asbestos Inhaled subsequently passed through the gastrointestinal tract and provided the exposure that led to the observed Increase In abdominal cancer. The assumption that all inhaled as bestos is Ingested is an overestimate but not a significant one. No account has been taken of the material that a worker may swallow directly, and this quantity could be Important. The extent to which these factors are offset ting cannot be estimated. Uncertainties exist in the extrapolation of ani mal data on clearance to man and In the effect of the aerosol size distribu tion on the fraction swallowed. These uncertainties, however, are felt to be unimportant in comparison to our inability to estimate the quantity of C-101 asbestos that might be directly swallowed. If the amount of asbestos directly Ingested Is less than that which remains In the lungs or is cleared from the body by other than Ingestion, the estimated criteria level will be less stringent. If the directly Ingested asbestos Is of more Importance, the criteria level will be more stringent. Table 31 lists the percentage of death from excess gastrointestinal cancer and peritoneal mesothelioma In four groups of asbestos workers. Calcu. 1 ations of these percentages were made using expected numbers of death, rather than the observed, because the latter Is often significantly inflated by Including other asbestos-related deaths (asbestosls, lung cancer, and pleural mesothelioma). Table 32 lists the fiber concentration estimates (see Carcinogenicity section) and an exposure Index for each cohort (years of exposure x fiber concentration). This Index will be used to calculate the number and mass of asbestos fibers ingested during a working lifetime. As the observed mortal. Ity Is, to a large extent, after 20 years from first exposure, the Inter mixing of time and exposure does not present significant problems. The average length of exposure for the insulation workers in the first group was calculated from data on employment time at entry Into the cohort In 1967. A working lifetime of <0 years was used for the smaller group of New York and New Jersey Insulators, virtually all of whom were deceased or retired. The estimate of the person-weighted exposure Index for the amosite factory Is simply the average employment time multiplied by 40 f/ml. Oat* from Table 33 were used to estimate a person-weighted exposure- index far the Newhouse and Berry group. [Person-weighted exposure Index i No. at risk x exposure x time) m 180.) 2(No. at risk) Percentage of Excess Gastrointestinal Cancers and Peritoneal Mesotheliomas in Four Groups of Asbestos Workers ( +<d4 *4 4 co p44 4 44 oO k- --X4 X u 44 S SL& --4 n 4 Pi CSJ PI 4 Q X44 u O VI s r*4 CSJ CSJ 4 X Ph CSJ vs *4 *3 k cm & U 41$ KUO Ut W 0. 4 X4 01 o I CSJ PO4 XO 00 m CSJ CSJ P4i o e o CSJ e 41 44 O*4 CS4J X tJfc M k S| SL| 4 0 s X S CO 40 X *n x5 LUX o PH X CO X XX n Pi O* fi X 44 4 pT CSJ M VI 41 41 ^ UX K4 UJ H- <- e oo w k <4* 41 w* -3 <4 Q e *-- O 41 44 X *m +* kO 41 V ="f k 41 u s m <u 3 a 3 c k o 4) VI & X Ui N CSJ as P-4 <NJ X CV S 6 * C3 Xo <n 3 PI X o u X Pi X CNJ 4* o w J_L <t4 lO ue 1 V c cO f- 44 44 O 5 2, 3k Vi x eu J_L 4441 *pm VI xo VS C b <4 X TO we pm % e 41 o f 44 J4 o 54 3b V) X eu 4 *** mt X 23 Om X p-4 o o 'w* o VI u K 41 CD X 00 o XX a sj u 44 e 51 e a E |4J k * O VI 44 Q uI Ml 5 U. *o *4 eu s o k 5i u o S*3. IT <-- 44 p-- 44 VI >i o g w Q sJ 44 k U X*3 <4 u c U w * Pi 4-4 XX w* ! d "5ci V x ph kn -m -4 CSJ ^ Oo u w P 44 p 0mm o 3U vl 3 3k a> u w P- 4* OK 44 VI >* o o k? o. >i m 44 k W<4 X u ^e U- 4 VI V ^- T3 ^ e *4 44 o VI VI k > k ex : u x C-103 aSellkoff, et a l. 1979* bSel1koff, 1976' cseldman, et a l. 1979 dNenhouse and Berry, 1979 ^Henderson and Enterline, 1979 ^McDonald, et a l. 1980 m ( TABLE 32 Exposure Indices for Asbestos Worker Groups Exposed Group Air Fiber Concentration (f/l) U.S. Insulators Sellkoff, et al. (1979a) 15 (Table 22) NY/NJ Insulators Sellkoff (1976) IS (Table 22) Amoslte factory workers Seldman, et al. (1979) 40 (Table 26) British factory workers Newhouse and Berry (1979) 10-30 Factory retirees Henderson and Enterline (1979) See note a Chrysotlle miners and millers See note a Person-weighted Exposure Index Average Exposure (years x f/ml) Time (yrs.) 34 510 40 600 1.9 76 See Table 33 180 740 S8S aThe cumulettve exposure Index In ffm1 x years was calculated by multiply ing a person-weighted exposure Index In mppef x years by 3. t t i t ( 0104 1 Exposure Estimates for Workers In a British Factory* Exposure Group No. at Risk Exposure (f/el) Tine of Exposure Severe <2 years >2 years Low to Moderate <2 years >2 years 711 1,333 503 933 Source: Newhouse and Berry, 1979 30 30 10 10 20 2 20 2 The fiber exposures for the studies.of MeOonald, et al. (1980) and Hen derson and Enterline (1979) were calculated from the estimate of McDonald,, et al. (1980) that 1 mppcf Is tool valent to 3 f/ml. While no data support this suggestion, It appears reasonable and was also used for the factory exposure circumstances. The majority of s*mpl analyzed for the EPA to date were characterized by a concentration of all electron microscopic visible fibers per liter of water (see Exposure section). Further, techniques for the determination of fiber concentrations (as opposed to mass concentrations) have been pub lished as Interim EPA procedures (Anderson and Long, 1980). Thus, a criter ion for the concentration of fibers of all sizes In water corresponding to a 10" risk will be calculated directly from the concentrations of fibers greater than 5 m measured In the occupational circumstances that produced disease. Unfortunately, the data currently available relating air concen trations of fibers longer than 5 um counted by optical microscopy, to those determined by electron microscopy, are extremely limited. These Include those by Wallingford (1978), 1:15; Mlllette (personal conmunlcation), 1:400; and Winer and Gossett (1979), 1:1,000 and are only for chrysotlle asbestos. Using the geometric mean of 200 for this factor from all available data, a total fiber concentration corresponding to a 10"" risk can be calculated from the data of Tables 31 and 32. The scant data on the relation between electron and optical microscopic counts Is uncertain. The variability between these three measurements Is likely the result of losses during the preparation of specimens for electron microscopy. Thus, the value by Wal lingford appears unduly low and Is In disagreement with electron microscopic size distributions showing 1 to 3 percent of fibers in chrysotlle asbestos 'aerosols to be longer than 5 m. With these considerations, the uncertainty in the value 200 may be estimated to be a factor of 3. C-L06 In making the calculation, ont tacitly assumes the same fiber size dfsoutlon In water as In occupational air samples. Some data show that water fiber size distributions vary grdatly (Mlllette, et al. 1979a,b), axxj, occupational air distributions have been- shown to be so variable that thgs fraction of fibers longer than 5 um can range over a factor of 10'depending on sampling circumstances (Nicholson, et al. 1972). -Although sizing of airborne and waterborne fibers has not been done using the same methods, qualitatively, water appears to have fiber distributions with more smaller fibers than In occupational.air samples. Experimental studies, previously discussed, have shown that once In place, longer fibers are more carcinogenic than shorter ones. However, shorter fibers appear to more easily cross organ barriers and migrate throughout the body, and may, thus, be of greater Importance for some asbestos malignancies, particularly meso thelioma (Sebastian, et al. 1979). The extent to which the assumption of ie same fiber size distribution In water as In air will likely yield a con servative criteria (from the point of view of health) cannot be estimated. A detailed calculation of the 70-year lifetime risk from the Injection of 10 fibers of asbestos per day Is given In Appendix I. Oata of the oc cupational risk of both gastrointestinal cancer and peritoneal mesothelioma were used (Table 31). Account was taken of the fact that occupational expo sures took place over a 5-day work week and that the Ingestion exposure may encompass a lifespan of 70 years. It was assumed that a worker breathes at the rate of 1 m^/hr during work exposure for the purpose of calculating total asbestos Intake per day. Using a linear dose-response relationship and a specified risk of 10"5, the calculated 70-year daily intake result ing from these calculations .are given in Table 34. It Is not correct to simply average intake levels (rather than risks) as a single study showing C-107 ( TABLE 34 The Calculated Risk of Death over a 70-Year Lifetime from Gastrointestinal Cancer and Peritoneal Mesothelioma from Ingestion of 106 flday of Asbestos Exposure Group Estimated Risk 10 f/day Sellkoff, at al. (1979a) Sellkoff (1976) Newhouse and Berry (1979) Henderson and Enterline (1979) McDonald, et al. (1980) . Average 1.1 x 10-5 1.7 x 10-5 3.1 x 10-5 1.9 x 10- 9.5 x 10-7 1.24 x 10"5 This average corresponds,to a daily intake of. 300,000 fibers for a 10*5 lifetime risk. t t V i C-10S "i* ( /, low risks could yield in Intake level of unlimited magnitude. The In fusion of such a level In any averaging process would clearly not be appro priate. The data from Seldman, et al. (1979) were not used because it was *1 exclusively from amoslte exposures. While exposure over the last 10 or 2<j years of life may not have been of great Importance In the generation of asbestos related cancers, those ingested during the first 10 years may be much more Important than fibers Ingested later, and no consideration was taken of this possibility In establishing criteria levels. Further, the occupational exposures from which the criteria were developed utilized exposures through the lifetime of the populations. Assuming that two liters of water are Ingested per day, this would correspond to a concentration of *00,000 fibers of all sizes/liter of water. It Is remarkable that three long exposure groups had similar exposure Hees. This would suggest that these estimates are Indeed reasonable, e exposure Index for the study of Newhouse and 8erry may be low, and this would produce a higher risk estimate. On the other hand, as previously dis cussed, the mortality data of Henderson and Enterline (1979) and McDonald, et al. (1980) may underestimate effects producing lower risk estimates. A criterion for a mass concentration of asbestos can also be calculated using the conversion value of 30 g/m^/f/ml derived from the data of Table 2 for predominantly chrysotlle exposures. A value of 150 ug/m^/f/ml for amoslte appears more appropriate, based on the finding of Davis, et al. (1978) that amoslte has approximately a three time greater conversion factor than chrysotlle. A detailed calculation Is given In Appendix II and the results summarized in Table 35. Assuming that 2 liters of water are ingest ed per day, a risk of 10"^ would be produced from Ingesting water contain g 0.05 ug/liter. As mentioned in the "Exposure" section, the variability C-X09 ( TABLE 35 The Calculated Risk of Oeath over a 70-Year lifetime from Gastrointestinal Cancer and Peritoneal Mesothelioma from Ingestion of 1 ug/day of Asbestos Exposure Group Estimated Risk ug asoestos Sellkoff, et al. (1979a) Sellkoff (1976) Seldman, et al. (1979) Newhouse and Berry (1979) Henderson and Enterline (1979) 7.3 x 10-5 1.1 x 10-A 5.5 x 10-5 2.1 x 10-4 1.3 x 10-5 McDonald, et al. (1980) 6.4 x 10-6 Average* 8.6 x 10-5 This corresponds to a dally Intake of 0.12 ug for a 10*5 lifetime risk. (h t t i c-uo ( e data used to convert optical fiber counts to mass (factor 0 In Appenc, .1) leads to a large uncertainty (a factor of 5) In the above estimate. -4 Considering chrysotlle and depending on the source of the asbestos In : water (see Table 5), 0.05 g/Hter corresponds to from 10 to 25 x 10 1 fibers of all lengths per day. Such estimates are considerably higher than those derived previously and are most likely a reflection of the differences In the sizes of the fibers found In water, as compared to those found In air. Because of these uncertainties, high priority should be given to ob taining accurate size and mass distribution of typical fibers found In dif ferent circumstances (air and water) which would allow appropriate conver sions to be made between fiber concentrations In air and water. Although positive animal experiments had various experimental limita tions, risk estimates were calculated from their data using a modified one model as previously discussed In the Methodology document. The data are anted In Table 36. Considering the large number of experimental uncer tainties, these values provide reasonable support for the concentration derived from human exposure data. This document was concerned with the estimation of that concentration of asbestos In water which will produce a lifetime risk of 1 In 100,000 In a population exposed continuously. The risk estimate was made using a linear extrapolation from existing human data and would appear to constitute a con servative extrapolation. However, In the case of asbestos, the risk factor of 1/100,000 Is not conservative. If we were concerned with Intermittent or localized contamination Incidents of some carcinogen that once identified, could be abated, such a value would have utility. With asbestos, however, we are concerned with a ubiquitous contaminant In the environment to which *ge populations are continuously exposed for decades, further, the estid value has a high degree of uncertainty associated with it, based upon the data from which It was derived. c-Ul Risk Estimates from Animal Experiments* Effect Estimated 10"* Dosage (g/l) 4/42 Kidney carcinomas 0/49 control 12/42 Malignancies 2/49 control 3.2 1.1 Source: Glbel, et al. 1976 i Unde** the Consent Decree in NRDC v, Train, criteria are to state "recoin:d maximum permissible concentrations (Including where appropriate,- ** zero) consistent with the protection of aquatic organisms, human health, and i recreational activities." Asbestos Is suspected of being a human carclno- * gen. Because there is no recognized safe concentration for a human carcino gen, the recommended concentration of asbestos In water for maximum protec tion of human health Is zero. Because attaining a zero concentration level may be Infeasible In some cases and In order to assist the Agency and states In the possible future development of water quality regulations, the concentrations of asbestos corresponding to several Incremental lifetime cancer risk levels have been estimated. A cancer risk level provides an estimate of the additional Inci dence of cancer that may be expected in an exposed population. A risk of 10^ for example, indicates a probability of 1 additional ease of cancer / very 100,000 people exposed, a risk of 10*6 Indicates 1 additional case of cancer for every million people exposed, and so forth. In the federal Register notice of availability of draft ambient water quality criteria, EPA stated that It Is considering setting criteria at an Interim target risk level of 10*, 10^, or 10"^ as shown In the fol lowing table. Exposure Assumption 2 liters of drinking water Risk Levels and Corresponding Criteria (1) 10*f. I :*1 *5 3,000 f/1* 30,000 f/1 300,000 f/1 Consumption of fish and shellfish only No Criterion *f fibers C-113 (1) Calculated by applying a relative risk epidemiological model as described In the Methodology Document to the human epidemiological data presented in Appendix III. Since the extrapolation model Is i linear at low doses, the additional lifetime risk is directly proportions 1 to the water concentration. Therefore, water concentra tions corresponding to other risk levels can be derived by multiplying or dividing one of the risk levels and corresponding water concentrations shown In the table by factors such as 10, 100, 1,000, and so forth. Concentration levels were derived assuming a lifetime exposure to various amounts of asbestos occurring from the consumption of drinking water only. Although total exposure information for asbestos is discussed and an estimate of the contributions from other sources of exposure can be made, this data will not be factored into ambient water Quality criteria formula tion until additional analysis can be made. The criteria presented, there fore, assume an Incremental risk from ambient water exposure only. REFERENCES Aavisory Committee on Asbestos. 1979. Vols. I and II: Final report of the,. Advisory Committee. Health and Safety Coimlsslon. Her Majesty's Stationery, Office, London. Amacher, O.E., et al. 1975. The dose-dependent effects of Ingested chrysotlle on ONA synthesis In the gastrointestinal tract, liver, and pancreas of the rat. Environ. Res. 10: 208. Anderson, C.H. and J.M. Long. 1980. Interim method for asbestos In water. EPA 600/4-80-005. U.S. Environ. Prot. Agency Res. Lab., Athens, Georgia. Anderson, H.A., et al. 1976. Household -contact asbestos neoplastic Hsk. . N.v. Acad. Scl. 271: 311. Auribault, M. 1906. Bull, del-Inspect, du travail. Bader, M.E., et al. 1961. Pulmonary function In asbestosls of the lung on alveolar-caplllary block syndrome. Am. Jour. Med. 30: 235. Balzar, J.L. and W.C. Cooper. 1968. The work environment of Insulating workers. Am. Ind. Hyg. Assoc. 29:-222. Beaman, D.R. and O.M. File. 1976. Quantitative determination of asbestos' fiber concentrations. Ann. Chem. 48: 1. c-115 Berry, G., et *1. 1972. Combined effect of asbestos exposure'and smoking on mortality from lung cancer In factory workers. Lancet. 2:' 476. 8onser, G.M, and 0.8. Clayson. 1967. Feeding of blue asbestos to rats. Br. empire Cancer Campaign for Research. 45: 242. Brain, J.0. and P.A. Volberg. 1974. Models of lung retention based on ICRP task group report. Arch. Environ. Health. 28: 1. British Occupational Hygiene Society. 1968. Hygiene standard for chrysotlle asbestos dust. Ann. Occup. Hyg. 11: 47. Brown, A.L., et al. 1976. The reliability of measures of amphlbole fiber concentration In water. Environ. Res. 12: 150. 8ue1ow, R.W., et al. 1980. The behavior of asbestos-cement pipe under var ious water quality conditions: A progress report. Res. Tech. Feb: 91. Carter, R.E. and W.6. Taylor. 1980. Identification of a particular amphibole asbestos fiber In tissues of persons exposed to a high oral Intake of the mineral. Environ. Res. 21: 85. Chamberlain, M. and E.M. Tarmy. 1977. Asbestos and glass fibres in bacter1al mutation tests. Mutat. Res. 43: 159. t Lk Chatfleld, E.J., et al. 1978. Preparation of water samples for asbestos fiber counting by electron microscopy. U.5. Environ. Prot. Agency. EPA 600/4-78*4)11. C-116 ( ora, K.S. 1978. Inter!aboratory measurements of amphlbole and chryso,e fiber concentration In water. Jour. Test. Eval. 6: 241. ia Commins, 8.T. and G.W. Gibbs. 1969. Contaminating organic material In as** bestos. 8r. Jour. Cancer. 23: 3S8. Cook, P.M. 1978. Amphlbole Fiber Concentration Determination for Series of Communication Air Samples: Use of X-ray Diffractions to Supplement Electron Microscope Analysis. In: P.A. Russell and A.E. Hutchins (eds.), Proc. Symp. Electron Microscopy and X-ray Applications to Environ, and Occup. Health Analyses. Ann Arbor Science Publishers, Ann Arbor, Michigan, p. 107. Cook, P.M. and G.F. Olson. 1979. Ingested mineral fibers: Elimination In ( an urine. Science. 204: 195. Cook, P.M., et al. 1974. Asbestlform amphlbole minerals: Detection and measurement of high concentrations In municipal water supplies. Science. 185: 853. Cook, P.M., et al. 1976. X-ray Diffraction and Electron Beam Analysis of Asbestlform Minerals In Lake Superior Waters. In: Proc. Int. Conf. Environ. Sensing Assessment. 34: 1. Cook, P.M., et al. 1978. Evaluation of cartridge filters for removal of. small fibers from drinking water. Jour. Am. Water Works Assoc. 70: 459. C-117 Coooer, R.c. and W.C. Cooper. 1978. Public health aspects of asbestos fibers in drinking water. Jour. Am. Water Works Assoc. 72: 33. Cooper, R.c., et al. 1978. Asbestos in domestic water supplies for five California counties. Progress report for period April 25, 1977 to June 30, 1978, under Contract No. R804366-02. U.S. 'Environ. Prot. Agency. Cooper, R.C., et al. 1979. Asbestos In domestic water supplies for five California counties. Part II, EHS Publ. No. 79-1, School of Public Health, ttniv. Calif. Berkeley, p. 247. Cooper, W.C. and J.L. Balzer. 1968. Evaluation and Control of Asbestos Exposures in the Insulating Trade. In: Holstein and Anspach (eds.), International Konferenz uber die Biologlschen Wirkungen des Asbestos, Dres den, Germany, April 22-24. Berlin, Oeutsche lent. Arbeltsmed. p. 151. Cooper, W.C. and J. Hi edema. 1973. Asbestosis in the Manufacture of Insu lating Materials. In: P. Bogovski, et al. (eds.). Biological Effects of Asbestos. IARC World Health Org., Lyon, France, p. 175. Craun, G.F., et al. 1977. Exposure to Asbestos Fibers in Water Distribu tion Systems. In: Proc. 97th Ann. Conf. Am. Water Works Assoc., Anaheim, California, May 8-13. Cunningham, H.M. and R.D. Pontefract. 1971. Asbestos fibers in beverages and drinking water. Nature. 232: 332. c-113 ( gingham, H,H. and R.O. Pontefract. 1973. Asbestos fibers In beverages, inking water and tissues: Their passage through the Intestinal wall and movement through the body. Jour. Assoc. Off, Anal. Chem. 56: 976. , Cunningham, H.M. and R.O. Pontefract. 1974. Placental transfer of asbes tos. Nature. 249: 177. Cunningham, H.M., et al. 1976. Quantitative relationship of fecal asbestos to asbestos exposure. Jour. Toxicol. Environ. Health. 1: 377. Cunningham, H.M., et al. 1977. Chronic effects of Ingested asbestos in rats. Arch. Environ. Contam. Toxicol. 6: 507. ( 'Is, J.M.G., et al. 1978. Mass and number.of fibers In the pathogenesis asbestos-related lung disease In rats. 8r. Jour. Can. 37: 673. Doll, R. 1955. Mortality from lung cancer in asbestos workers. Br. Jour. Ind. Med. 12: 81. Donna, A. 1970. Tumori sperlmentall da amlanto di crisotilo, crocldolite e amosite In ratto Sprague-Dawley. Med. Lavoro. 61: 1. Duggan, M.J. and E.W. Culley. 1978. The counting of small numbers of as bestos fibers on membrane filters: A comparison of results from some commer cial laboratories. Ann. Occup. Hyg, 21: 85. Edge, J.R. 1975. Asbestos-related disease In Barrow and Furness. Environ. Res. 11: 244. Edge, J.R, 1979. The Incidence of carcinoma of the bronchus In shipyard workers with pleural plaques. Ann. N.Y. Acad. Sci. 330: 289. Elmes, P.C. and M.J.C. Simpson. 1971. Insulation workers In Belfast. III. Mortality (1940-56). 8r. Jour. Ind. Med. 28: 226. Elmes, P.C. and M.J.C. Simpson. 1977. Insulation workers In Belfast. A further study of mortality due to asbestos exposure (1940-75). Br. Jour. Ind. Med. 34: 174. Enterllne, P.E. and V. Henderson. 1973. Type of asbestos and respiratory cancer In the asbestos Industry. Arch. Environ. Health. 27: 312. Enterline, P., et al. 1972. Mortality In relation to occupational exposure In the asbestos Industry. Jour. Occup. Med. 14: 897. Evans, J.C., et al. 1973. Studies on the deposition of inhaled fibrous material In the respiratory tract of the rat and Its subsequent clearance using radioactive tracer techniques. I. UICC crocidollte asbestos. Environ. Res. 6: 180. Ferris, 8.6., Jr., et al. 1971. Prevalence of chronic respiratory disease. Arch. Environ. Health. 23: 220. C-120 ^ ischer, W.E., et *1. 1946. A health survey of pipe coverage operetions construction of navel vessels. Jour. Ind. Hyg. Toxicol. 28s 9. -U 1 Gibbs, G.W. and M. LaChance. 1974. Oust-flber relationships in the Quebed chrysotlle Industry. Arch. Environ. Health. 28: 69. Glbel, W., et al. 1976. Tlerexperimentelle untersuchungen uber elne kanzerogene wirkung von asbestflltermaterlal nach oraler aufnatme. Arch. Geschwulstforsch. 46: 437. Gill am, J.D., et al. 1976. Mortality patterns among hard rock gold miners exposed to an asbestlform mineral. Ann. N.v. Acad. Scl. 271: 336. yne, S.R. 1935. Two cases of squamous carcinoma of the lung occurring asbestosls. Tubercle. 17: 5. Gross, P., et al. 1967. Experimental asbestosls: The development of lung cancer In rats with pulmonary deposits of chrysotlle asbestos dust. Arch. Environ. Health. 15: 343. Gross, P.t et al, 1973. Lymphatic transport of fibrous dust from the lungs. Jour. Occup. Med. 15: 186. Gross, P., et al. 1974. Ingested mineral fibers. Oo they penetrate tissue^ or cause cancer? Arch. Environ. Health, 29: 341. i C-121 I ( Hallenbeck, W.H., et al. 1978. Is chrysotlle asbestos released from asbes tos -cement pipe Into drinking water? Jour. Am. Water Works Assoc. 70: 97. Hammond, E.C., et al. 1979. Asbestos exposure, -cigarette smoking, and death rates. Ann. N.Y. Acad. Scl. 330: 473. Harlngton, J.S. 1962. Occurrence of oils containing 3,4-benzpyrene and related substances In asbestos. Nature. 193: 43. Harlngton, J.S. and F.J.C. Roe. 1965. Studies of carcinogenesis of asbes tos fibers and their natural oils. Ann. N.Y. Acad. Scl. 132: 439. Harllss, K.W., et al. 1978. The acute effects of chrysotlle asbestos expo( sure on lung function. Environ. Res. 16: 360. Harries, P.6. 1968. Asbestos hazards In naval dockyards. Ann. Occup. Hyg. 11: 135. Harries, P.G. 1971a. A comparison of mass and fibre concentrations of asbestos dust In shipyard Insulation processess. Ann. Occup. Hyg. 14: 234. Harries, P.6. 1971b. Asbestos dust concentrations In ship repairing: A practical approach to Improving asbestos hygiene In naval dockyards. Ann. of Occup. Hyg. 14: 241. f i% C-122 ies, P.6. 1976. Experience with asbestos disease and Its control In w at Britain's naval dockyards. Environ. Res. 11: 261. 1 a Harrington, J.M., et al. 1978. An investigation of the use of asbestos1 cement pipe for public water supply and the Incidence of gastrointestinal cancer In Connecticut, 1935-1973. An. Jour. Epidemiol. 107: 96. Henderson, V.I. and P.E. Enterline. 1979. Asbestos exposure: Factors asso ciated with excess cancer and respiratory disease mortality. Ann. N.Y. Acad. Scl. 330: 117. Holt, P.F., et al. 1964. The early effects of chrysotl.le asbestos dust on the rat lung. Jour. Pathol. Bacterlol. 87: 15. Jig, L.M., et al. 1979. Risk of asbestosls In crocldollte and amosite mines in South Africa. Ann. N.Y. Acad. Scl. (In press) Jacobs, R., et al. 1978. Light and electron microscope studies of the rat digestive tract following prolonged and short-term Ingestion of chrysotile asbestos. Br. Jour. Exp. Pathol. 59: 443. Kay, 6. 1973. Ontario Intensifies search for asbestos In drinking water. Water Pollut. Control. 9: 33. Klviluoto, R. 1960. Pleural calcification as a roentgenologic sign of nonoceuoatlonal endemic anthophyllite-asbestosls. Acta Radiol. Suppl. 194: 1. C-123 Klelnfeld, M., et al. 1967. Mortality experience In a group of asbestos workers. Arch. Environ. Health. IS: 177. . Knox, J.F., et al. 1968. Mortality from lung cancer and other causes among workers In an asbestos textile factory. Br. Jour. Ind. Med. 25: 293. Kogan, F.M., et al. 1972. The cancer mortality rate among workers of as bestos Industry of the Urals. 61g. 1 Sanlt, 37: 29. Langer, A.M. 1974. Inorganic particles In human tissues and their associa tion with neoplastic disease. Environ. Health Perspect. 9: 229. . Longer, A.M,, et al. 1971. Chrysotlle asbestos In the lungs of persons In New York City. Arch. Environ. Health. 22: 348. Langer, A.M., et al. 1973. Identification of asbestos In human tissues. Jour. Qccup. Med. 15: 287. Langer, A.M., et al. 1978. Variation of some physical, chemical, and bio logical properties of chrysotlle asbestos subjected to prolonged milling. Jour. Toxicol. Environ. Health. 4: 173. LeFevre, M., et al. 1978. Accumulation of 2 wm latex particles In mouse Peyer's patches during chronic latex feeding. Experlentla. 34: 120. ( t i Levy, B.S., et al. 1976. Investigating possible effects of asbestos In city water: Surveillance of gastrointestinal cancer Incidence In Duluth., Minn. Am. Jour. Epidemiol. 103: 362. C-U4 { ;ohn, H.C. 1972. The medical surveillance of asbestos workers. R. Sv Health Jour. 92: 69. i i Lewis, 8.6. 1977. Asbestos In cooling-tower waters. Argonne Natl. Lab., Argonne, Illinois. ANl/ES-63. Lleben, J.P. and H. Plstawke. 1967. Mesothelioma and asbestos exposure. Arch. Environ. Health. 14: 559. LIlls, R. and I.J. Sellkoff. 1979. Asbestos disease In maintenance workers of the chemical Industry. Ann. N.v. Acad. Scl. 330: 127. Lynch, J.R, and H.E. Ayer. 1966. Measurement of dust exposures In the ^ tos textile Industry. Jour. Am. Ind. Hyg. Assoc. 27: 431. Lynch, J.R., et al. 1970. The Interrelationships of selected asbestos exposure Indices. Jour. Am. Ind. Hyg. Assoc. 31: 598. Lynch, K.M. and W.A. Smith. 1935. Pulmonary asbestosls. III. Carcinoma of lung in asbestos-silicosis. Am. Jour. Cancer, 24: 56. Lynch, K.M., et al. 1957. Pulmonary tumors In mice exposed to asbestos dust. Arch. Ind. Health. 15: 207. Maltonl, C. and C. Annoscla. 1974. Mesotheliomas In Rats Following the Intraperltoneal Injection of Crocldollte. In: W. Davis and C. Maltonl ( ), Advances In Tumour Prevention, Detection, and Characterization. Vo 1. 1 .racterizatlon of Human Tumours. Excerpta Medica, Amsterdam, p. 115. C-i2S Mancuso, T.F. and E.J. Coulter. 1963. Methodology In Industrie heilth studied. The cohort approach, with special reference to., w asbestos com pany. Arch. Environ. Health. 6: 210. Mason, T.J., et al. 1974. Asbestos-1 Ike fibers In Ouluth water supply. Relation to cancer mortality. Jour. Am. Med. Assoc. 228: 1019. McDonald, J.C. and O.K. Liddell. 1979. Mortality in Canadian miners and millers exposed to ehrysotlle. Ann. N.Y. Acad. Scl. 330: 1. McOonald, J.C., et al. 1971. Mortality In the chrysotile asbestos mines and mills of Quebec. Arch. Environ. Health. 22: 677. McOonald, J.C., et al. 1980. Oust exposure and mortality In chrysotile mining, 1910-75. 8r. Jour. Ind. Med. 37: 11. McEwen, J., et al. 1977. Mesothelioma in Scotland. Jour. Br. Med. 4: 575. Merewether, E.R.A. 1947. Annual report of the chief Inspector of factories for the year 1947. Her Majesty's Stationery Office, London. Meurman, L.O., et al. 1974. Mortality and morbidity among the working population of anthophyllite asbestos miners in Finland. Br, Jour. Ind. Med. 31: 105. * 1i Mirller, L., et al. 1965. Tests for effect of asbestos on benzol a )pyrene carcinogenesis in the respiratory tract. Ann. N.Y. Acad. Sci. 132: 489. c-126 ( Mette, J.R. 1976. Analyzing for asbestos In drinking water. Mews of /iron. Res., Cincinnati, Ohio. Jan. 16. U J Millette, J.R. 1979. Exposure to asbestos from drinking water In the United States. EPA 600/1-79-150. Environmental Health Effects Research Report, Off. Res. Oev. Millette, J.R., et al. 1979a. Aggressive water and water utilities from representative areas of the United States. Jour. Am. Water Works Assoc. (Submltted) Millette, J.R., et al. 1979b. Concentration and size of asbestos in water supplies. Paper presented at NIEHS Workshop on the Biological Effects of Inerel Fibers and Particulates, Washington, O.C., June 7-8. (Submitted to .viron. Health Perspect.) Morgan, A., et al. 1975. Studies on the deposition of Inhaled fibrous material in the respiratory tract of the rat and its subsequent clearance using radioactive tracer techniques. II. Deposition of the UICC standard reference samples of asbestos. Environ. Res. 10: 196. Mount Sinai. 1976. Rapport final comite d'eltlde selr la selubrite' daus 1'indestive de 1'amlante. Montreal, Canada. Annex. Murphy, R.L.H., et al. 1971. Effects of low concentrations of asbestos. New England Jour. Med. 285: 1271. C-127 Myrriy, H.M, 1907. Report of the departmental comlttee on compensation for industrial disease. Her Majesty's Stationery Office, London. National Institute for Occupational Safety and Health. 1972. Criteria for a recommended standard: Occupational exposure to asbestos. HSM 72-10267. U.S. 'owernment Printing Office, Washington, D.C. National Institute for Occupational Safety and Health. 1976.. Re-examina tion and update on the health effects of occupational exposure to asbestos. U.S. Government Printing Office, Washington, O.C. Newhouse, M.L. 1969. A study of the mortality of workers In an asbestos factory. Br. Jour. Ind. Med. 26: 294. Newhouse, M.L. and 6. Berry. 1976. Predictions of mortality from mesothellal tumours In asbestos factory workers. Br. Jour. Ind. Med. 33: 147. Newhouse, M.L. and G. Berry. 1979. Patterns of mortality in asbestos fac tory workers In London. Ann. N.v. Acad. Scl. 330: 53. Newhouse, M.L. and H. Thomson. 1965. Mesothelioma of pleura and peritoneum following exposure to asbestos In the London area. Br. Jour. Ind. Med. 22: 261. Newhoese, M.L., et al. 1972. A study of the mortality of female asbestos k i workers. Br. Jour. Ind. Med. 29: 134. t C-128 io1son, W.J. 1971a. Measurement of asbestos In ambient air. Final .port under Contract CPA 70-92. National Air Pollution Control Adnlnlstra- tlon. i i Nicholson, W.J. 1971b. Insulation hygiene progress reports 3: 2. Mount Sinai School of Medicine of the City University of New York. Nicholson, W.J. 1974. Analysis of amphlbole asbestlforts fibers In Munici pal water supplies. Environ. Health Res. 9: 165. Nicholson, W.J. 1976. Case study 1: Asbestos - the TLY approach. Ann. N.Y. Acad. Scl. 271: 152. :ho1son, W.J. and F.L. Pundsack. 1973. Asbestos In the Environment. In: Bogovskl, et al. (eds.). Biological Effects of Asbestos. IARC, Lyon, France. IARC Scl. Publ. No. 8. p. 126. Nicholson, W.J., et al. 1971. Asbestos Air Pollution In New York City. In: H.M, England and W.T. Barry (eds.), Proc. Second Clean Air Cong. Aca demic Press, New York. p. 136. Nicholson, W.J., et al. 1972. Asbestos contamination of parenteral drugs. Science. 177: 171. Nicholson, W.J., et al. 1975. Asbestos contamination of the air In public buildings. Final report under Contract No. 68^)2-1346. U.S. Environ. Prot. ency. C-129 A Nicholson, W.J., et al. 1978. Control of sprayed asbestos surfaces in school buildings: A feasibility study under NIEHS Contract -No. 1 ES-7-2113. Natl. Inst. Environ. Health Safety. Nicholson, W.J., et al. 1979. Long term mortality experience of chryso* tile miners and millers In Thetford Mines, Quebec. Ann. N.v. Acad. Scl. 330: 11. Nicholson, W.J., et al. 1980a. Environmental Asbestos Concentrations In the United States. In: Proceedings of Conference on Asbestos and Mineral Fibers. IARC, Lyon, France. Nicholson, W.O., et al. 1980b. Mortality experience of asbestos factory workers: Effect of differing Intensities of asbestos exposure. Environ. Res. (In press) Nordmann, M. and A, Sorge. 1941/ Lungenkrebs durch asbeststaub 1m tlerversuch. Z. Krebsforsch. 51: 168. Nottle, M.C. 1977. Plant particles In kidneys and mesenteric lymph nodes of sheep. Aust. Vet. Jour. S3: 405. Patel4tand1Ik, K. and W. Hallenbeck. 1978. Fate of Ingested chrytotila asbestos fiber in the newborn baboon. EPA 600/1-78-069. Report prepared for U.S. Environ. Prot. Agency. t t Peto, J. 1978. The hygiene standard for asbestos. Lancet. 8062: <84. C-130 , J., et al. 1977. A mortality study among workers in an English as- ,,jtos factory. 8r. Jour. Ind. Med. 34s 169. * Pontefract, R.O. and H.M. Cunningham. 1973. j i Penetration of asbestos through the digestive tract of rets. Nature. 243; 3S2. Pooley, F.D. 1973. Mesothelioma In Relation to Exposure. In: P. Bogovskl, et al. (eds.). Biological Effects of Asbestos. IARC, Lyon, France. IARC Scl. Publ. No. 6. p. 222. Pott, F. and K.H. Friedrichs. 1972. Tumortn der ratte nach 1 .p.-Injektlon faserformlger staube. Naturalssenschaften. 59: 318. t, F., et al. 1976. Ergebnlsse aus tlerversuchen zur kanzerogenen <ung faserformlger staube und Ihre deutung 1m hlnbllck auf die tumorentstehung belm menschen. Zbl. Bakt. Hyg., I Abt. Orlg. B. 162: 467. Pylev, L.N. and L.M. Shabad. 1973. Some Results of Experimental Studies In Asbestos Carcinogenesis, ^n: P. Bogovskl, et al. (eds.). Biological Effects of Asbestos. IARC, Lyon, France. IARC. Scl. Publ. No. 8. p. 99. Reeves, A.L. 1976. The carcinogenic effect of Inhaled asbestos fibers. Ann. Clin. Lab. Scl. 6: 459. Reeves, A.L., et al. 1971. Experimental asbestos carcinogenesis. Environ. Res. 4: 496. C-Ul ( ( ' I Reeves, A.L., et il. 1974. Inhalation carcinogenesis from various forms of ! asbestos. Environ. Res. 8: 178. *| Robinson, C., et al. 1979. Mortality Patterns, 1940-75, Among Workers Employed In an Asbestos Textile Friction and Packing Products Manufacturing Facility. In: R. Lemen and J.R. Dement (eds.), Dusts and Disease. Pathotox Publishers, Park Forest, Illinois, p. 131. Rohl, A.N., et al. 1976. Asbestos exposure during brake lining maintenance and repair. Environ. Res. 12: 110. Samudra, A., et al. 1978. Electron microscope measurements of airborne asbestos concentrations. A provisional methodology manual. Prepared under Contract No. 68-02-2251. EPA 600/2-77-178. U.S. Environ. Prot. Agency. Sanders, E. and C.T. Ashworth. 1960. A study of particulate Intestinal absorption and hepatocellular uptake. Exp. Cell Res. 22: 137. Sawyer, R.N. 1977. Asbestos exposure In a va1e building: Analysis and resolution. Environ. Res. 13: 146. Schnltzer, R.J., et al. 1971. Interaction of mineral fiber surfaces with cells In vitro. Ann. N.Y. Acad. Scl. 172: 747. t Scirelber, G. 1974. Ingested dyed cellulose In the blood and urine of man. i. Areh. Environ. Health. 29: 39. c-132 I sf ` len, F., et *1. 1976. Les pollutions atmospherlgues urbanles par 1 ste. Rev. Franc. Mai. Resp. 4: 51.- Sebastien, P., et al. 1979. Translocation of Asbestos Fibers Through Respiratory Tract and Gastrointestinal Tract According to Fiber Type and Si2e. In,: R. lemen and J.R. Oement (eds.). Dusts and Disease. Pathotox Publishers, Park Forest, Illinois, p. 65. Seldman, H., et al. 1979. Short-term asbestos work exposure and long-term observation. Ann. N.v. Acad. Sci. 330: 61. Selikoff, I.J. 1976. Lung cancer and mesothelioma during prospective sur veillance of 1,249 asbestos insulation workers, 1963-1974. Ann. N.v. Acad. Y 271: 448. Selikoff, I.J., et al. 1964. Asbestos exposure and neoplasia. Jour. Am. Med. Assoc. 188: 22. Selikoff, I.J., et al. 1965. The occurrence of asbestosis among insulation workers in the United States. Ann. N.Y. Acad. Sci. 132: 139. Selikoff, I.J., et al. 1968. Asbestos exposure, smoking and neoplasia. Jour. Am. Med. Assoc. 204: 106. Selikoff, I.J., et al. 1970. Mortality Experience of Asbestos Insulation Workers, 1943-68. In: M.A. Shapiro (ed.), Proc. Int. Conf. Pneumoconiosis, Jc`"'inesburg. Oxford Univ. Press, Capetown, p. 180. C-133 Sellkoff, I.J., et al. 1972. Carcinogenicity of amoslte asbestos. Arch. Environ^ Health. 25: 183. Sellkoff, I.J., et al. 1979a. Mortality experience-of Insulation workers In the United States and Canada, 1943-1977. Ann. N.Y. Acad. Sel. 330: 91. Sellkoff, I.J., et al. 1979b. Asbestos disease In United States shipyards. Ann. N.Y. Acad. Scl. 330: 295. Shabad, L.M., et al. 1974. Experimental studies on asbestos carcinogeni city. Jour. Natl. Cancer Inst. 52: 1175. Slncock, A.M. 1977. In vitro Chromosomal Effects of Asbestos and Other Materials. In: Origins of Human Cancer. Cold Spring Harbour, 1976. Slncock, A. and M. Seabright. 1975. Induction of chromosome changes in Chinese hamster cells by exposure to asbestos fibers. Nature. 257: 56. Smith, W.E. 1973. Asbestos, talc, and nitrates In relation to gastric can cer. Am. Ind. Hyg. Assoc. Jour. 33: 227. Smith, W.E. and 0.0. Hubert. 1974. The Intrapleural Route as a Means for Estimating Carcinogenicity. _In: E. Karbe and J.F. Park (eds.),-Experimental Lung Cancer. Springer-Verlag, 8erl1n. p. 92. a \l A k C-134 Sf W.E., et a 1. 1965. Tests for carcinogenicity of asbestos. Ann. n, .cad. Sel. 132: 456. *? Smith, W.E., et al. 1970. An Experimental Model for Study of Cocarcino- 1 genesis In the Respiratory Tract. In,: P. Netteshelm, et al. (eds.), Morpho logy of Experimental Respiratory Carcinogenesis. U.S. Atomic Energy Conm., Oak Ridge, Tennessee, p. 299. Stanton, M.F. 1973. Some Etiological Considerations of Fibre Carcino genesis. In: p* Sogovskl, .et al. (eds.), 81olog1cal Effects of Asbestos. IARC, Lyon, France. IARC Sel. Pub1. No. 8. p. 289. Stanton, M.F. and C. Wrench. 1972. Mechanisms of mesothelioma induction 'sbestos and fibrous glass. Jour. Natl. Cancer Inst. 48: 797. Storeygard, A.R. and A.L. Brown. 1977. Penetration of the small Intestinal mucosa by asbestos fibers. Mayo Clin. Proc. 52: 809. Taskinen, E., et al. 1973. A current hypothesis of the lymphatic transport of Inspired dust to the parietal pleura. Chest. 64: 193. Thomson, J.G., et al. 1963. Asbestos as a modern urban hazard. S.A. Med. Jour. 37: 77. Timbrell, V. 1965. The Inhalation of fibrous dusts. Ann. N.v. Acad. Scl. 132: 255. C-13S U.S. EPA. 1974, A preliminary report on asbestos In the Duluth, Minnesota, area. Off. of Tech. Analysis. January. i U.S. EPA. 1976. Quarterly report of the Environmental Research Laboratory. Duluth, October-December, 1976, p. S. U.S. EPA. 1977. Comparison of ambient asbestos levels determined by vari ous laboratories. Environ. Monitor, Support Lab. September. Volkhelmer, G. 1974. Passage of particles through the wall of the gastro intestinal tract. Environ. Health Perspect. 9: 215. Wagner, J.C., et al. 1960. Olffuse pleural mesothelioma and asbestos expo sure in the Northwestern Cape Province. Br. Jour. Ind. Med. 17: 260. Wagner, J.C., et al. 1973. Mesothellomata In rats after Inoculation with asbestos and other materials. Br. Jour. Cancer. 28: 173. Wagner, J.C., et al. 1974. The effects of Inhalation of asbestos in rats. 8r. Jour. Cancer 29: 252. Waoner, J.C., et al. 1977a. Animal Experiments with Talc. In: W.C. Walton (ed.). Inhaled Particles and Vapors, IV. Pergamon Press, New York. (In press) f%c > t C-136 f ner, J.C., et al. 1977b. Studies of the Carcinogenic Effect of fibre oiass of Different Olameters Following Intrapleuril Inoculation In Experi;* mental Animals. In: Natl. Inst. Occup. Safety and Health Symp. Occup* Exposure to Fibrous Glass. (Jnlv. of Maryland, 1977. (In press) Wallingford, K.M. 1978. Chrysotlle asbestos In Industry. Paper presented to Natl. Inst. Occup. Safety and Health meeting, Washington Baltimore sec tion of the Am. Ind. Hyg. Assoc., February. Webster, I. 1970. Asbestos Exposure In South Africa. In: H.A. Shapiro (ed.). Pneumoconiosis Proc. of Int. Conf., Johannesburg. Oxford Unlv. Press, Capetown, p. 209. ( 1, H., et al. 1979. Influence of dose and fiber type on respiratory Ignancy risk In asbestos cement manufacturing. Am. (lev. Resp. 01s. 120: 345. Weiss, A. 1953. Pleurakrebs bel lungenasbestose, In, vivo morphologlsch geslchert. Medlzlnlsche. 2: 93. Weiss, W. 1971. Cigarette smoking, asbestosls and pulmonary fibrosis. Am. Rev. Resp. 01s. 104: 223. Westlake, G.E., et al. 1965. Penetration of colonic mucosa by asbestos particles. An electron microscopic study In rats fed asbestos dust. Lab. Invest. 14: 2029. C-137 Wlgle, D.R. 1977. Cancer mortality In relation to asbestos In municipal water supplies. Arch. Environ. Health. 31: 185 Winer, A.A. and M. Cossette. 1979. The effect of aspect ratio on fiber counts: A preliminary study, Ann. N.v. Acad. Scl. 330: 661. { Appendix I Sample calculation of risk per 10 flbers/day Ingested using a linear dose-response relationship. 16.8 x 10"2 x 1 1 600(f/ml) yrs 8 x 10 ml/day 8C 1 70 yeers xl 10 200 S 0F A * Percentage of excess 61 cancer and peritoneal mesotheli oma In study group. 8 Exposure Index. C - Exposure took pi ace .for 8 hours and the worker was as sumed to breathe 1 3/hr (10 ml/hr). 0 - Conversion from optical counts (fibers >5 vm) to electron microscopic counts (all fibers). E - 70-year exposure to water is assumed. F - Exposure was concentrated In 5 days rather than 7 days/ week. 6 - Calculation Is for a 70-year risk per 10 flbers/day. C-139 Appendix II Sample calculation of risk per ug of asbestos Ingested ustng a lin ear dose-response relationship. 16.8 x 10-2 x x 1 x ^jj033_fLl2i x 70 years x 600(f/1) yrs 8 m3/dayug/m3 A B CO EF A Percentage of excess 61 cancer and peritoneal mesotheli oma In study group. B Exposure Index. C a Exposure took place for 8 hours and the worker was as sumed to breathe 1 m3/hr. 0 - Conversion of 30 wg/fl3 per 1 f/ml of chrysotlle (Table E a 70-year exposure to water Is assumed. F a Exposure was concentrated In 5 days rather than 7 days/ week. I Appendix III Sunnary and Conclusions Regarding the Carcinogenicity of Asbestos* Asbestos is a collective mlneraloglcal term referring to naturally oc curring minerals which have crystalled In the form of masses of long fibers which can be easily separated. This term also comaonly refers to certain mineral occurrences In which fibrous silicate mineral ean be extracted and used commercially for insulation, textiles, brake linings, asbestos cement, construction products, etc. Chrysotlle, the fibrous form of serpentine, provides over 95 percent of the approximately 900,000 tons of asbestos con sumed each year in the United States. The remaining asbestos used consists of the fibrous amphlbole minerals crocldollte, amoslte (fibrous grunerlte), anthophylllte. fine dusts produced from the mining, milling, manufac\ *g, and use of these asbestos minerals contain discreet microscopic, eiw.^ated mineral particles of "fibers" which when Inhaled by man are known to cause bronchogenic carcinoma and pleural and peritoneal mesothelioma. Asbestos particles and other inorganic fibers Introduced into the pleu ra, peritoneum, and trachea of rodents have Induced malignant tumors in numerous studies reported In the literature. Limited, and contradictory data exist for the carcinogenicity of asbestos administered to animals by Inges tion. One study In which asbestos filter material was fed to rats (Sibel, et al. 1976) reports 12 malignant tumors In 42 exposed animals versus only 2 liver-cell carcinomas In 49 control animals. Electron microscope analysis "This summary has been prepared and approved by the Carcinogens Assessment Group of EPA on June 23, 1979. C-141 of animal tissues for asbestos Indicates that Ingested fibers can accumulate at many sites following hematogenous or lymphatic transport of Ingested fibers which pass through the gastrointestinal mucosa. The strongest evidence for the carcinogenicity of Ingested asbestos Is provided by epidemiology of populations occupationally exposed to high con centrations of airborne asbestos dust. Inhalation exposure to asbestos dust Is accompanied by Ingestion exposure because high percentage of Inhaled fibers are removed from the respiratory tract by mucociliary clearance and swallowed. Peritoneal mesothelioma, often In great excess since It Is very rarely observed In the absence of asbestos exposure, and modest excesses of stomach, esophagus,, colon-rectum, and kidney cancer have been observed asso ciated with occupational exposure. The Influence of long-term chrysotlle fiber contamination of San Fran cisco Bay area water supplies on cancer Incidence has recently been studied by the University of California under an EPA grant. Significant dose response gradients for the Incidence of several cancers. Including white male lung and stomach and white female esophageal and peritoneal cancer, were noted Independent of the effect of socioeconomic status. Other water supply studies are of limited value due to factors such as very low exposure and Insufficient time elapsed since Initial exposure of the population. Observation In human urine of mineral fibers previously Ingested with drink ing water has established that Ingested asbestos can pass through the human gastrointestinal mucosa and migrate to various tissues. Asbestos is a known carcinogen when inhaled. The demonstrated ability of asbestos to induce malignant tumors In different animal .tissues, the i passage of ingested fibers through the human gastrointestinal mucosa, and the? extensive human epidemiological evidence for excess peritoneal, gastrc- C-142