Document 2ZXz4aqb8VZR81p14yQprYma

Asbestos Content of Lung Tissue, Lymph Nodes, and Pleurai Plaques from Former Shipyard Workers12 PLAINTIFF'S EXHIBIT RONALD F. DODSON, MARION G. WILLIAMS, JR., CAROLYN J. CORN, ALESSANDRO BROLLO, and CLAUDIO BIANCHI Introduction Attention has been given to the amount of asbestos found in lungs of occupation ally exposed workers (1-6) and, to a less er extent, to the concentrations in lungs of the general population (7-11). These analyses have defined the parenchymal SUMMARY Autopsy mamplas tram tight formir ihlpyird worldrs win collactad from lung paranchyma, trachial lymph nodi*, and plaural plaqun. Thi tiaaui from iaeh raapictlvi ana wit pmparad by a modlflid bliaeh dlgutlon Uchnlqua, and thi rotldua was collictid on a 0.2-wn pom polycarbonata or 0.22-qm mlxad eaUulosa istir flltsr. Quantitation of firruglnoua bodloa and uneoatad llboro was dona by tight and transmission ilic. Iron microscopy, raapoctlvoly. Olffinsncas In tlta asbestos burden wars noted for each alia. Fer ruginous bodies ware observed in both parenchyma and nodes but not In plaques. Throe aub|ects burden not cleared by the mucociliary es calator or relocated into the interstitium and removed by the bloodstream and pul monary lymphatics (12). This latter route includes a pathway for relocation that could explain the development of granu lomatous lesions in extrapulmonary sites (lymph nodes, spleen, and liver) such as has been reported after exposure to fine ware found to have more ferruginous bodies par gram dry weight In their lymph nodes than In their lung parenchyma. Likewise, all subjects were found to have more uncoated fibers per nr.m In the nodes than In the parenchyma. Arephlbole and chrysotlla fibers ware noted In the luno and extranulmonarysliea, with chrysotlte being the predominant asbestlfOrm In plaques. The majority at th. undSated fibers In both the nodes and the plaques were < S pm In length. However, some fibers with dimensions conforming to tho "Stanton hypothesis" reschad bath areas. These residual patterns most likely reflect the Impact of clearance on lung burden as opposed to the eventual accumulation and stasis In the extrapulmonary areas. AM REV RESPtR PIS 1WO: 142:S43-S<7 soot and silica dust (13). In studies in volving animals exposed to mixed miner als dusts, McMillan and coworkers (14) v(18), kidney (18, 19), spleen (18-20), found appreciable relocation of titani Aymph nodes (20, 21), and bile duct tuum dioxide and quartz particles from the Imor (22), and in pleural plaques (23). lung parenchyma to the lymph nodes. 'This tissue marker (i.e., FB) is known to Although the lymphatics may be in represent only a fraction of the total fi , volved in only a small fraction of the to ber burden in lung tissue and offers no tal particulate removal, they may never clear insight as to the levels of uncoated theless be very significant in the overall fibers in lung tissue, much less in other pathogenesis o f lung diseases since a slow | tissues. rate of particle movement allows maxi The present study was designed, there mal interactions to occur (15). fore, to expand the limited quantitative Much of the inhaled asbestos is in the data that exist for total asbestos burdens form of thin fibers, and tissue data ob (FB and uncoated fibers) in lung tissue tained by transmission electron micros compared with that in important extra copy indicate that many of these fibers pulmonary sites of relocation--lymph retained in lung tissue are short, uncoat- nodes and pleural plaques. The tissue in i ed fibers 5 pm in length) (1, 16). this study was obtained from occupation These small uncoated asbestos fibers are more easily moved and are thus like ly candidates for relocation from lung pa ally exposed subjects, and the fiber di ameter, length, type, and concentration in each area are reported. the pleural plaques and lymph nodes were carefully rinsed with water and debrided with precleaned scalpels for removal of all adher ing tissue before the final samples were taken. Subjects were 58 to 82 yr of age (mean, 70 9 yr) at the time of autopsy. Seven had work histories indicating 6 to 46 yr ofexposure in shipyard activities. The group consisted o fj smokers as well as nonsmokers (four smok ers averaging 35 cigarettes per day, one ex smoker, and three nonsmokers). Medical his\ tories for the subjects showed multiple dis eases. Seven of the eight subjects had cardio vascular disease, four had asbestosis, three had cancer, one had a neoplasm, and two had emphysema (table 1). Areas of lung parenchyma were collected from beneath the pleura. The regions were carefully dissected to assure sampling of mul tiple homogeneous areas. Samples of pari etal plaque and tracheal lymph nodes were renchyma either via airway clearance or through interstitial routes to other tissues. However, most information about as bestos in extrapulmonary sites has been derived by light microscopy, which is selectively restricted by resolution limits (0.2 pm) and which identifies only fer ruginous bodies (FB) or longer uncoat ed fibers. Although the numbers of FB reported in extrapulmonary sites have been low, their occurrence has been not ed in tissue from the stomach (17), liver Msthoda Tissue samples were collected at autopsy from eight shipyard workers with histories of oc cupational asbestos exposure. The necropsies were performed at the General Hospital of Monfalcone, which is a small industrial town in northeastern Italy and is the site of large, shipyards. The tracheal lymph nodes, pari-f etal plaques, and lung parenchymawere fxed in prefiltered formalin (0.22-pm filter). To avoid cross-contamination, each tissue was placed in a separate container. Additionally, (Received in originalform October 6, 1989 and in revised form March 5, 1990) ' From the Department of Cell Biology and En vironmental Sciences, The University of Texas Health Center at Tyler, TVler, Texas, and the Labo ratory of Pathological Anatomy, Hospital of Monfaicone, Monfalcone; Italy. 1 Correspondence and requests for reprints should be addressed to Ronald F. Dodson, Ph.D., Chair man and Professor, Department of Cell Biology and Environmental Sciences, The University of Tex as Health Center at Tyler, P. O. Box 2003, TVler, TX 7S7I0. 843 844 OOOSCM, WILLIAMS, COftM. MOLLO. ANO MAMCM TABLE 1 distribution of the chrysotiie and amphi- FORMER SHIPYARD WORKERS bole concentrations in the tissuesjs il- \ ^Jo t 2 3 a Age Sex i/D M se SmOK.ng -- 33 t* :*oay Occupation Shipyard plumper and painter. 30 yr. miner isease Asbestosis: small cell Ca of right lung Plaque Bilateral lustrated by site in figure 1. In seven of ! the eight subjects (table 2), the numbers of FB in the lung parenchymal samples (1,120 to 853,000 FB/gdry) reflected this i past exposure. The total number of un-( M 61 20 : g.cay Shipyard welder, 37 yr Asbestosis; adeno Ca of the right lung; arteriosclerosis Bilateral coated asbestos fibers (chrysotiie plus amphibole) in these seven subjects ranged M 63 40 c.groay Shipyard laborer. 3 yr: Navy: oil mill, crane Laryngeal Ce: arteriosclerosis Bilateral from830,000to 192,000,000 per gram of dry tissue. No FB were found in pleural f operator plaques; however, in three of the subjects/[ M 70 SO cig/day Shipyard worker and Lung librosis; Bilateral the numbers of FB per gram of dry tisJ draftsman, 46 yr. emphysema: sue from the tracheal nodes exceeded the arteriosclerosis numbers per gram in lung (table 2). There 5 M 72 Stopped m Shipyard tracer. 27 yr Arteriosclerosis: 1973 meningioma Bilateral was no linear correlation between FB per gram dry and the chrysotiie or amphi 6 M 7S NO Shipyard welder, 45 yr . Asoestosis; arteriosclerosis Left bole concentrations. In the lung and nodes, the FB mean lengths were 39.2 * 23.5 7 M 78 No Shipwright. 25 yr: chemical industry Asbestosis: emphysema: Bilateral and 3S.9 23.2 pm, respectively (table arteriosclerosis 3), and, though surprisingly similar, were 8 F 82 No Shipyard cleaner for some months: husband shipyard welder Arteriosclerosis Bilateral statistically different (p = 0.0018). Linear correlations between the con centrations or concentration ratios of amphibole to chrysotiie were generally DtUMion of tbonvwionr C* carcinom*. not observed except for a negative corre lation between lung and plaque chryso tiie (r = -0.881; p = 0.002). Still, in all also carefully dissected. Each sample was for each sample, and fibers per gram wet and lived, with one half used for the determina- dry weight were computed. The average de m of the wet weight/dry weight ratio and tection limits were 41,000 fibers/g wet and e other half for digestion in 9.2% sodium 200,000 fibers/g dry. ' '' hypochlorite. The mean weights of the in The following nonparametric statistical tests dividual samples were as follows: parietal were used because the data were not normal plaque, 0.3691 0.028 g wet; parenchyma, ly distributed. Spearman's rank correlation 0.3899 0.018 g wet; lymph nodes, 0.2053 coefficient (25) was used to test for linear rela 0.053 g wet. Aliquots from each digest were tionships between the various fiber and FB filtered through either a 0.22-pm Millipore concentrations, and Wilcoxon's matched pairs mixed cellulose ester filter for FB counts (Mil signed rank test (25) was used to test for sim lipore Corp., Bedford, MA) or a 0.2-prn Nu- ilarities in the amphibole to chrysotiie con clepore polycarbonate filter for fiber counts centration ratios between sites. It was also used (Nuclepore Corp., Pleasanton, CA). This was to test for similarities among the median lengths followed by a further clean-up according to and widths, which had been weighted by the the procedure of Williams and coworkers (24). number of asbestos fibers observed per case. Reagent blanks were prepared using the same reagents and procedures described above. Part of the mixed cellulose ester filter was Remits cleared'(made transparent) using acetone va Asbestosis had been diagnosed histolog por, and the FB were counted at x 200 by light ically in four of the subjects; however, microscopy. FB per gram wet and dry weight r. all eight subjects had a history of occupa were calculated for each digest. The average ,nt lional exposure to asbestos (table 1). The detection limits for FB were 55/g wet and eight subjects, the chrysotiie concentrai, jtions in the plaques exceeded the amphit bole concentrations, whereas in the lungs $ (Subjects 3,5,6, and 7) and nodes (Sub"j jects 4,5,6, and 8), this was true in only! half of the subjects (table 2). i* In seven of eight subjects, the amphi-\ bole to chrysotiie concentration ratios] were greater in the lungs than in the! plaques (p = 0.0173), and in all cases they p were greater in the nodes than in thej , plaques (p =* 0.0117). The physical features of the two types of uncoated asbestiforms in the lungs did vary (p < 0.00005 for length and width). As shown in table 4, 86% of the chryso tiie was < 5 pm in length, whereas 59% of the amphibole was < 5 pm in length, j The mean length of chrysotiie was 2.87 ; 1.07 pm, whereas that for amphibole j was twice that figure (5.82 6.71 pm), j 248/g dry. Carbon extraction replicas were made from the polycarbonate filters, and these were ex amined at x 16,000 to x 100,000, as required, in a JEOL 100CX Ternscan analytical electron microscope equipped with a Tracer North ern TN-2000 energy-dispersive X-ray analyz er. Actual length and width, composition, and .electron diffraction pattern were characterzed for each fiber encountered. All fibers 0.5 Fig.1. Log of asbestos concentration by sit* (fibers/grim dry). NO none detected. pm in length or greater were counted. The definition of a fiber for these counting proce dures included those with > 3:1 aspect ratios and parallel sides for a majority oftheir lengths. Forty 200-mesh grid squares were examined NO LUNO CHRY LUNO AMPH plaque CHRY NO PLAQUE AMPH ft) NOOC CHRY NOOC AMPH AMESTO* CONTENT OF LUNO TIIWI FROM FORMER SHIPYARD WORKERS 843 TABLE 2 FORMER SHIPYARD WORKERS* Subject No. Tissue Location FB/g /lot FB/g Dry CHRY/g Wet CHRY/g AMPH/g Dry f Wet.., AMPH/g Dry ASB/g Dry 1 PL \o ND 2.900 9.900 430 1.500 11,400 LU 4u :oo 953.000 370 2.000 34.000 190.000 192.000 NO `'900 SB 600 ND ND *30.000 1.700.000 1.700,000 2 PL NO 2.200 5.500 500 1.300 6.800 LU ' 200 7 080 650 3.800 780 4.600 8.400 NO t >() 3.420 5.100 15.000 21.000 62.000 77.000 3 PL NO NO 4.800 16.000 850 2.900 18.900 LU 0 238 i 300 90 500 60 330 830 NO 2.100 11 100 ND NO 550 2.900 2.900 4 PL ' NO ND S.500 17.000 soo 1.600 18.600 LU 0.051 0 269 25 140 64 340 480 NO 4 160 17200 3.000 13.000 2.200 9.000 22.000 5 PL NO ND 1.400 4.000 480 1.400 5.400 LU 0.265 3.030 480 5.500 64 720 6.220 NO 1.030 4.370 2.900 12.000 1.200 5.200 17.200 6 PL ND NO 2.100 5.200 250 630 5.830 LU 1.390 9.340 350 2.400 140 930 3.330 NO NO NO 4.200 18.000 1.900 7.700 25,700 7 PL ND ND 1.300 3.900 52 160 4.060 LU 3.300 22.300 41Q 2.800 160 1,100 3.900 NO 2.050 10.200 1.300 6.600 6,700 33.000 39.600 8 PL NO ND 6.200 21.000 NO NO 21.000 LU 0.232 1.120 NO NO 490 2 400 2.400 NO 0.147 0.610 1.700 5.500 1.200 4.100 9.600 Definition of ootmvietione- F8 - ferruginous bodies: CHRY - cHrysotif* fibers. AMPH smbfubofe ''S.n AS9 * torsi as* bsstos. NO - nob* a,tided. PL - plaque. LU lung: NO - node Oala a 1.000 TABLE 3 FERRUGINOUS BODY LENGTH* Plaque^ Lung, n 591 Node, n 241 Mean _ -- 39.2 23.5 35.9 23.2 Median -- 34 30 . Minimum -- 8 9 * All dimensions are in micrometers, t None detected. Maximum _ 119 112 TABLE 4 UNCOATED CHRYSOTILE AND AMPHIBOLE FIBERS* Average Length Avenge Width Length % > 5 um *0 , Chrysotile fibers Plaque Lung Node Amphiboie fibers Plaque Lung Node 1.34 1.50 (Q.85) 2.87 1.07 (1.51) 1.27 os 1.07 (0.94) 1.98 * 3.13 (1.05) 5.82 6.71 (3.39) 2.22 it 3.38 (1.SS) 0.09 * 0.1S (0.06) 0.07 0.06 (0.05) 0.06 0.06 (0.06) 0.15 0.07 (0.14) 0.19 0.21 (0.14) 0.21 0.12 (0.18) 31 14 0 0.0 10.0 41.0 6.0 00 40 00 8.0 20.0 2.5 * All dimension* in micrometer*; numbers in parentheses are geometric meins. The mean amphiboie width was 0.19 0.21 pm, which was more than 2.5 times the chrysotile width of 0.07 r 0.06 pm; figures 2 and 3 illustrate the length and width distributions of all uncoated chrysotile and amphiboie fibers found at each of the three sites. The process involved in relocating as bestos fibers to these extrapulmonary sites results in deposition of a typically shorter uncoated liber, tor example (ta ble 4), no uncoated chrysotile fibers > 5 pm were found in the nodes, and only 3.1?7o of the chrysotile found in the plaques was > 5 pm in length, whereas 14T of the chrysotile in the lung was > 5 pm in length. Ten percent of the amphi boie fibers in the plaques and 6^0 in the nodes were > 5 pm in length compared with 41 To in the lung > 5 pm in length. The lengths of both types of asbestos k'were significantly shorter (p < 0.00005) in both plaques and nodes compared with those found in the lung parenchyma. Node amphiboie fibers were significantly! longer (p < 0.00005) than plaque amphi- -4 bole fibers. Some longer amphiboie fibers (> 10 pm in length) were found in both the nodes (2.5To) and the plaques (8To). The accumulation of nncnated as bestos fibers in nodes ancfplaques resultedln theTiighest concentrations.occurringtrafside the lung in all of the subjecFs Tta5le~~2)/In six subjects, this roccurrecTlh the nodes, whereas in two, the plaques contained the highest num ber of uncoated asbestos fibers among the sites. The cho,sotile fibers in the lung tissue l were significantly thinner than those in L the plaques or the nodes (p < 0.00005), ' whereas the node amphiboie fibers were significantly wider than those in the Iungsl ^ or the plaques (p < 0.00005). When the* amphiboie and chrysotile dimensions were compared at each site, the amphi boie fibers were significantly longer and thinner (p < 0.00005), Assessing the di mensions of width in another perspec-l five, 97To of all chrysotile and 80To of all amphiboie fibers from all sites were less than 0.3 pm in diameter. Discussion The lungs of humans are protected by various defense mechanisms that are ac tivated by foreign particles from the en vironment. The deposition of dusts (in cluding asbestos) is met with an efficient clearance system that eliminates inhaled particles (26) via the "escalator" concept of an upward shifting of particulates (27) until they are eventually removed in spu- 846 OOOSOM. WILLIAM*. COflN, MOLLO. AMO SEAMCM ii 3 j i i i LLNC LENGTH .UNO .IDTH L*CUE LENGTH "HOVE VOTH NOOE LENOTH NODE VIOTH Fig 2. Log ot chrysotile asbestos di mansions by sit* (micrometers). !T i o Fig 3. Log of amasite asbestos dimen sions by site (micrometers). LUNO LENOTH LUNO VIOTH HAQUE LENOTH PLAOUE VOTH NODE LENOTH NODE VOTH jum (28). When the combined protective plaques is speculative, the deposited /stems are overwhelmed (26), potentially fibers are characterized by significantly rreversible damage and/or a relocation shorter lengths, supporting some type of 4 of inhaled dusts to more permanent de- selection process. ' position sites in subpleural areas (29,30) Chrvsotile has been noted as the pre- can occur. dormnant asbestifortp in parietal pleura In the specific case of inhaled fibers, (36)TTfhis observation is of particular in selective ciearangeratcs occur according terest since the lower limit of fiber length to fiber length. The shortfibers can be used in tnat study would have preferen <Jr relocated more easily and/or cleared by tially excluded many of tne~~i5orter r TEe mucociliary pathway or other routes cftrysotiUTibers fncludedin the >0.5 pm more efficiently over time than are the limlrsct in the~pfesent~studv. fhc data longer fibers (31-33). of both studies indicatethat chrysotile XTIower" pathway for clearance of can accumulate in the pleural regions. If fibers from lung parenchyma is through Several other studies have indicated that 1 the interstitium, the lymphatics (27, 34, chrysotile makes up a majority of these 35). or(the lung vasculature. These routes short, uncoated fibers(8,16. 37). Anex=- are the least rapid but, as shown in our r\ pjanation of this stems from length be- present study, result in an appreciable!- finga determinant of whicn libers were number of asbestos fibers being depos-'H relocated to the pleural regionsT " ited over time within lymph nodes and "TTgreater number of longeraffiphibole pleural tissues. Although most of the] fibers reached the nodes than the plaques fibers in both nodes and plaques are short#""as is reflected by the observation of FBs (< 5 pm),.their numbers are often high- ' (on amphibole cores) in the former and er per gram than those in lung tissue from not in the latter. Although most uncoat the same person. It is assumed that the ed amphibole fibers in both extrapulmohigher burden in nodes and plaques can nary sites are short, amphibole fibers that nbe explained by the less efficient clear fit the dimensions of the "Stanton hy ance ofthese areas as contrasted with the pothesis" (38,39) do reach both of these impact of the more efficient clearance extrapulmonary sites. Thus, although rel mechanisms operating in the lung paren atively few of these longer amphiboles chymal areas. Although the route from^ are present in the overall burden, these the original site of fiber deposition in the fibers, which are potentially more carci lung to relocation into lymph nodes and nogenic (because of their dimensions), do reach the plaques. No chrysotile fibers fitting Stanton's dimensions were found in the nodes, but a small percentage found in the pleural plaques did conform to these aspect ratios."'"'"-" It must further be recognized that the data from this study require a resolution best attained by the transmission electron microscope and counting criteria that in clude short, thin fibers. Furthermore, processing must include fiber collection on filters with optimally sized pores since the diameter of many of these uncoated fibers are such that they could pass through filters with larger pore sizes (40). Although this study was not intended to address issues of long versus short fibers and their potential toxicity, it nev-| ertheless has shown that some longer; fibers (> 5 pm) and large numbers of|<A short (< 5 pm) fibers reach lymph nodes' and plaques. Stringent standards of technique, prep aration, and resolution are required to include these short, uncoated fibers in a quantitative analysis of tissue burden, but their presence and numbers justify this type of study. This is particularly im portant since recent data suggest that! j> short fibers may be important in lungp fibrosis (41), and nothing is known about their impact in extrapulmonary areas. References 1. Dodson RF, Williams MG, O'Sullivan MF. Com CJ, Greenberg SD, Hurst GA. A comparison of the ferruginous body and uncoated fiber content in the lungs of former asbestos workers. Am Rev Respir Dis 1985: 132:143-7. 2. Churg A. Asbestos fibers and pleural plaques in a general autopsy population. Am J Pathol 1982; 109:88-96. ------ 3. Churg A, Wood P. Observations on the distri bution of asbestos fibers in human lungs. Environ Res 1983; 31:374-80. 4. Morgan A, Holmes A. The distribution and characteristics of asbestos fibers in the lungs of Finnish anthophyllite mine-workers. Environ Res 1984; 33:62-75. . 5. Holden J, Churg A. Asbestos bodies and the diagnosis of asbestosis in chrysotile workers. En viron Res 1986; 39:232-6. 6. Whitwell F, Scott J, Grimshaw M. Relation ship between occupations and asbestos-fibre con tent of the lungs in patients with pleural meso thelioma, lung cancer, and other diseases. Thorax 1977; 32:377-86. 7. Dodson RF, Greenberg SD, Williams MG, Corn CJ, O'Sullivan MF, Hurst GA. Asbestos content in lungs of occupationally and nonoccupationally exposed individuals. JAMA 1984; 252:68-71. 8. Churg A, Warnock ML. Asbestos fibers in the genera] population. Am Rev Respir Dis 1980; 122' 669-78. 9. Breedin PH, Buss DH. Ferruginous (asbestos) bodies in the lungs of rural dwellers, urban dwellers and patients with pulmonary neoplasms. South Med J 1976; 69:401-4. 10. Churg A, Warnock ML. Analysis of the cores AMUTOt COKTHST OF LUNO T1MUC FIKHl FOMIM MflPYAIM WOCER$ 847 of ferruginous (asbestos) bodies from the general population. I. Patients with and without lung can cer. Lab invest 1977; 37:280-6. 11. Churg A, Wat nock ML. Analysis of the cores of ferruginous (asbestos) bodies from the general population. 111. Patients with environmental ex posure. Lab Invest I9"9: -0:622-6. 12. Lauweryns JM. Baer JH. Alveolar clearance and the role of the pulmonar. .'v mphaties. Am Rev Respir Dis 19"; Ii5:625-e3. 13. Spencer H. The anatomy of the lung. In: Pathology of the lung. 4th ed. Oxford: Pergamon Press Ltd., 1983; 41'. 14. McMillan CH, Jones AD. Vincent JH. John ston AM, Douglas AN, Cowie H. .Accumulation of mixed mineral dusts in the lungs of rats during chronic inhalation exposure. Environ Res 1989; 48:218-37. 15. Fishman AP. Pulmonary diseases and disord ers. Nesv York: McGraw-Hill, 1980; 638. 16. Churg A. Fiber counting and analysis in the diagnosis of asbestos-related disease. Hum Pathol 1982; 13:381-92. 17. Telischi M, Rubenstone AI. Pulmonary asbestosis. Arch Pathol 1961; 72:116-25. 18. Langer AM. Inorganic particles in human tis sues and their association with neoplastic disease. Environ Health Perspect 1974; 9:229-33. 19. Auerbach O, Conston AS, Garfinke! L. Parks VR, Kaslow HD, Hammond EC. Presence of as bestos bodies in organs other than the lung. Chest 1980: 77:133-7. 20. Godwin MC, Jagatic JJ. Asbestos and meso theliomas. Environ Res 1970; 3:391-416. 21. Keal EE. Asbestosis and abdominal neoplasms. Lancet 1960; 2:1211-6. 22. Szendroi M, Nemeth L, Vajta G. Asbestos bod ies in a bile duct cancer after occupational exposure. Environ Res 1983; 30:270-80. 23. Rosen P. Gordon P, Savino A. Melamed M. Ferruginous bodies in benign fibrous plural plaques. Am J Clin Pathol 1973: 60:608-17. 24. Williams MG, Dodson RF, Com C, Hurst GA. A procedure for the isolation of amosite asbestos and ferruginous bodies from lung tissue and spu tum. J Toxicol Environ Health 1982; 10:627-38. 25. Hollander M, Wolfe DA. Nonparametric statistical methods. New York: John Wiley and Sons, 1973. 26. Gross P, Detreville RT. The lung as an embat tled domain against inanimate pollutants. Am Rev Respir Dis 1972; 106:684-91. 27. Lippmann M, Yeates DB, Albert RE. Deposi tion, retention, and clearance of inhaled particles. Br J Ind Med 1980; 37:337-62. 28. Dodson RF, Williams MG, Corn CJ, Idell S, McLarty JW. Usefulness ofcombined light and elec tron microscopy evaluation of sputum samples for asbestos to determine past occupational exposure. Mod Pathol 1989; 2:320-2. 29. Holt PF. Translocation of inhaled dust to the pleura. Environ Res 1983; 31:212-20. 30. Morgan A, Evans JC, Holmes A. Deposition and clearance of inhaled fibrous minerals in the rat. Studies using radioactive tracer techniques. In: Walton WH, ed. Inhaled particles. 4th ed. Edin burgh: Pergamon, 1975; 259-74. 31. Roggli VL, Brody AR. Changes in numbers and dimensions of chrysotile asbestos fibers in lungs of rats following short-term exposure. Exp Lung Res 1984; 7:133-47. 32. Brody AR, Hill LH. Interstitial accumulation of inhaled chrysotile asbestos fibers and consequent formation of microcalcifications. Am J Pathol 1982; 109:107-14. 33. Churg A, Wright JL, Gilks B, Depaoli L. Rapid short-term clearance of chrysotile compared with amosite asbestos in the guinea pig. Am Rev Respir Dis 1989: 139:885-90. 34. Leak LV. Lymphatic removal of fluids and par ticles in the mammalian lung. Environ Health Perspect 1980; 35:55-76. 35. Lauweryns JM, Baert JH. The role of the pul monary lymphatics in the defenses of the distal lung: morphological and experimental studies of the transport mechanisms of intratracheally mstillated particles. Ann NY Acad Sci 1974; 2il:244-7j, 36. Sebastien P, Janson X, Gaudichet A. Hirsdi A, Bignon J. Asbestos retention in human respira tory tissues: comparative measurements in lung pa renchyma and in parietal pleura. In: Wagner JC. ed. Biological effects of mineral fibers. Lyon: 1ARC. 1980; 237-46. 37. Warnock ML, Kuwahara TJ. Wolery G. The relation of asbestos burden to asbestosis and lung cancer. Pathol Annu 1983; 18:109-45. 38. Stanton MF. Layard M. Tegeris A. Miller E. May M, Kent E. Carcinogenicity of fibrous glass: pleural response in the rat in relation to fiber dimen sion. J Natl Cancer Inst 1977; 58:587-603. 39. Stanton MF, Layard M, Tegeris E. ei at. Rela tion of particle dimension to carcinogenicity in am- phibole asbestos and other fibrous materials. J Natl Cancer Inst 1981; 67:965. - 40. O'Sullivan MF, Corn CJ, Dodson RF. Com parative efficiency of nuclepore filters of various pore sizes as used in digestion studies of tissue. E nviron Res 1987; 43:97-103. 41. Churg A, Wright JL, Depaoli L, Wiggs B. Minetalogic correlates of fibrosis in chrysotile miners and millers. Am Rev Respir Dis 1989; 139: 891-6. \