Document 37BZLn0B3NroaGdRd7YpB6rEO

Black Spots Concentrate Oncogenic Asbestos Fibers in the Parietal Pleura Thoracoscopic and Mineralogic Study C. BOUTIN, P. DUMORTIER, F. REY, J, R. VIALLAT, and P. DE VUYST Pneumology Department, Concneption Hospital, and Pulmonology Unit, Paoli-Calmettes Institute, Marseilles, France; and Chest Department, Erasmus Hospital, Brussels, Belgium Epidemiologic and pathologic data demonstrate that malignant mesothelioma occurs preferentially after exposure to long amphibole asbestos fibers. However, mineralogic studies have rarely detected such fibers in the parietal pleura. We hypothesized that the distribution of asbestos fibers in the pleura was heterogeneous and that they might concentrate in certain areas, as does coal dust in patients showing anthracotic "black spots" of the parietal pleura during thoracoscopy. We collected thoracoscopic bi opsy samples from these black spots and from normal areas of the parietal pleura and lung from 14 subjects (eight with and six without asbestos exposure). Asbestos content was determined by trans mission electron microscopy. In exposed subjects, mean fiber concentrations were 12.4 9.8 x 10s fibers/g of dry tissue in lung, 4.1 1.9 in black spots, and 0.5 0.2 in normal pleura. In unexposed patients, concentrations were 0, 03 0.1, and 0, respectively. Amphiboles outnumbered chrysotile in all samples. A total of 22.5% of fibers were > 5 pm in length in black spots. A histologic similarity of these black spots with milky spots is suggested by conventional and electron microscopy. We con clude that the distribution of asbestos fibers is heterogeneous in the parietal pleura. Indeed, the fibers concentrate in black spots, where they can reach high concentrations..These findings could explain why the parietal pleura is the target organ for mesothelioma and plaques. Boutin C, Dumortier P, Rey F, Vlallat JR, De Vuyst P. Black spots concentrate oncogenic asbestos fibers in the parietal pleura; thoracoscopic and mineralogic study. am i respir cru care meo ips;iss:444-9. Asbestos Fibers that reach the respiratory bronchioles and alveoli are subject to different fates. Chrysotile fibers can undergo frag mentation, lixiviation, and progressive clearance, whereas am phibole Fibers may remain unchanged for decades (1,2). If con centrations are sufficiently high in the lung, asbestosis (3) and eventually bronchial carcinoma (4) may develop. In patients with these conditions, asbestos bodies, mostly formed on amphiboles, axe usually found in lung sections and bronchoalveolar lavage fluid (5). Fibers may also migrate towards the periphery of the lung (6), especially the lower lobes (7), into mediastinal lymph nodes (8, 9) and the pleura. Benign pleural plaques and malignant mesothelioma are the main manifestations of occupational or environmental exposure to asbestos fibers (10). Plaques usually develop on the parietal pleura or on the diaphragm. Malignant pleura] mesothelioma also originates from the parietal pleural (11). Up to now, how ever, only few studies have detected significant amounts of as bestos fibers in the parietal pleura. Human lung tissue analyses have demonstrated that the risk for mesothelioma is related to the concentration of long (> 8 pm) amphibole fibers (12, 13). Animal experiments using in tracavitary injections of asbestos confirmed that the most carci- (ftecetved in original form March 23, 1995 and in revised form May 25, 1995) Supported by INSERM Research Grant No. 92 1002. Correspondence and requests for reprints should be addressed to C. Boutin, Hopital de la Conception, 147 bd. Bailie, 13385 Marseille cedex 5, France. Am | Respir Crit Care Med Vol 153. pp 444-449, 1996 nogenic fibers were those measuring > 8 pm in length and <0.25 pm in diameter (14,15). In sharp contrast with these data is the fact that all previous mineralogic studies have shown that short chrysotile fibers are the most common type of asbestos present in the parietal pleura (2,16-19). Amphiboles fibers meeting Stan ton's dimensional criteria for carcinogenicity (14) are uncommon or absent. A possible explanation for this paradox could be that fibers are heterogeneously distributed in the parietal pleura. This would explain why random sampling in unselected areas yields poor concentrations. Indeed, when samples were taken from patho logic zones such as pleural plaques (18) or tumors (19), more fibers were found. Several findings led us to speculate that asbestos fibers could accumulate in certain areas of the parietal pleura. In mice, sub cutaneously injected fibers are known to concentrate in the milky spots of the parietal pleura (20). In humans, milky spots are quite invisible in the healthy pleura. However, thoracoscopy sometimes visualizes foci of parietal anthracosis near lymphatic vessels of the parietal pleura (11), which we have called "black spots." In deed, these black spots could correspond to the milky spots de scribed by Kanazawa and colleagues in mice (20) and be rendered visible by trapped coal dust. The present study was carried out to ascertain whether as bestos fibers do concentrate in these black spots. Samples were collected from (i ) black spots, (2) adjacent, macroscopically nor mal pleura, and (3) peripheral lung, in a series of 14 consecutive patients in whom anthracotic spots were detected on parietal pleura during thoracoscopy. The results of mineralogic analysis were compared. HWBUI0008919 Boutin, Dumortier, Rey, el al.: Oncogenic Asbestos Fibers in the Parietal Pleura 445 I METHODS In our department, we perform more than 200 diagnostic video-tho racoscopies a year for pleurisy, spontaneous pneumothorax, and dif fuse interstitial lung fibrosis. Thoracoscopic and sampling techniques have been previously described (21, 22). From January 1993 through February 1994, we collected samples from patients in whom visual in spection of the parietal pleura revealed anthracotic black spots with or without anthracotic lymphatic vessels. In addition to (he anthracotic parietal samples, specimens were taken from an area of macroscopically normal pleura at a distance of 2 cm ("nonanthracotic parietal samples") and from the peripheral lung opposite the black zone ("lung samples"). Lung biopsy samples were prepared by sodium hypochlorite diges tion (5). Because bleach digestion of parietal pleura samples does not provide good preparations, the latter samples were dehydrated, defatted in alcohol and xylene, dried, and ashed in a low-temperature asher. The resulting particles were collected on 0.45-pm-porosity cellulose filters (Millipore*; Millipore Corp., Bedford, MA). After fusion by acetone vapors, pieces of filter were carbon-coated and transferred to transmis sion electron microscopy (TEM) grids. Grids were studied in a Philips EM400T analytical electron microscope equipped with an EDAX PV9900 energy dispersive X-ray analyzer. A magnification of X 22,000 was used to detect all fibers longer or equal to 1 pm, while a X4,600 magnifica tion detected only fibers equal or longer than 5 pm. Theoretical detec tion limits were around 200,000 fibers/g of dry tissue (f/g) for fibers longer than 1 pm and 100,000 f/g for fibers longer than 5 pm. These values were statistically fitted for each sample, depending on the actual concentration of fibers (see Table 3). Fibers were identified on the basis of morphology, chemical composition (EDS spectrum), and, in some cases, electron diffraction pattern. Since data did not show a normal statistical distribution, the Wilcoxon and Mann-Whitney nonparametric tests were used to compare the mean fiber concentration and size values. Because particle size dis tributions are usually log normal, statistical comparison between lengths and widths were also performed on geometric size data. RESULTS Patients Samples were obtained from 14 subjects. The characteristics of these patients are given in Thble 1. Mean age was 60.3 11.2 yr. Eight patients had a history of asbestos exposure: occupa tional in six patients, environmental in one patient who lived in Corsica (23), and both occupational and environmental in one patient. Mesotheliomas were confirmed by the panel of pathol ogists of the French Mesothelioma Register. Thoracoscopic features of parietal pleural anthracosis were observed in all 14 patients. In 12 patients, the lesions consisted Figure 1. Black spots (B) located in an intercostal space. White lym phatic vessels (L) and malignant mesothelioma nodules (M) are visi ble in close contact with black spots. The main nodule is about 3 mm in diameter. of black spots 2 to 4 mm in diameter. In two of these 12 patients, the spots were isolated and, in nine patients, clusters of three to four spots were observed on the lower, posterolateral part of the costo-vertebral gutter (Figure 1). The remaining patient was a former coal miner who had many black spots scattered from the bottom to the top of the pleural cavity. In two patients, an thracotic lesions were located along lymphatic vessels that were very superficially located and clearly distinguishable from in tercostal veins. Both black spots and black lymphatic vessels were observed m intercostal spaces near intercostal veins and arteries and were never found on ribs. In 11 of 14 patients, samples were obtained from the parietal pleura (anthracotic and nonanthracotic) as well as from the lung. In the remaining three patients, lung samples were not available (patients 8,13, and 14). In patient eight, lung sample was taken but pathologic study revealed only thick visceral pleura tissue without any lung structure. In patient 13 (spontaneous pneu mothorax in association with AIDS) and in patient 14 (70 yr old), lung biopsy was not performed to avoid further complications. Mean sample weight was 4.4 2.9 mg of dry tissue for lung samples, 5.5 3.6 mg for anthracotic parietal samples, and 3.1 2.6 mg for nonanthracotic parietal samples. 3 TABLE 1 PATIENT CHARACTERISTICS Patient Age No. Sex (yr) Occupation Asbestos Exposure 1 M SB Insulation worker 2 M S5 Tunnel digger 3 M 65 Insulation worker A M 63 Shipyard worker S F 65 Seamstress M 61 Mechanic 7 M 68 Metallurgy worker 8 M 59 Railroad worker 9 M 45 Hairdresser 10 M 65 Printer 11 M AS Freight handler 12 M S3 Coal miner 13 M 40 Restaurant operator 14 F 70 Hotel operator + f + + 4 4* + +t - - - - * Environmental and occupational exposure. f Environmental exposure in Corsica. * AIDS-related. Diagnosis Asbestos pleural plaques interstitial lung fibrosis Spontaneous pneumothorax Pleural mesothelioma Pleural mesothelioma Pleural mesothelioma Pleural adenocarcinoma Pleural plaques Lung cancer Lung cancer infectious pleural effusion Pleural adenocarcinoma Spontaneous pneumothorax^ Idiopathic pleural effusion ' HWBUI0008920 AMERICAN JOURNAL Of RESPIRATORY AND CRITICAL CARE MEDICINE VOL 153 1^6 **14 Figure 2. Concentration of asbestos fibers > 1 lira (A) and 2= 5 pm (8) as a function of sample types, (white bars: normal pleura; grey bars: pleural black spots; black bars: lung). Results are ranked in or der of decreasing fiber concentration'in lung biopsy. Patients 1 to 8 were previously exposed to asbestos (E1-E8). Patients 9 to 14 have no known previous exposure (NE9-NE14). Empty square-sample not available. Fiber Distribution The concentration of fibers observed in different samples is shown in Figure 2. Fibers 1 pm or longer (Figure 2A) were present in seven of II lung samples, in 12 of 14 anthracotic parietal sam ples, and in six of 14 nonanthracotic parietal samples. Extreme values in lung samples ranged from 0 to 70.9 x 10* f/g of dry tissue, with a mean of 7.9 6.3 x 10s f/g. In anthra cotic parietal samples, fiber concentration ranged from 0 to 15.7 x 10* f/g, with a mean of 2.5 3.7 x 10s f/g. The two mean values were not statistically different. In six of 11 cases, fiber concentration was higher in anthracotic parietal samples than in lung samples. In nonanthracotic parietal samples, mean fiber concentration was 0.3 0.13 x 10s f/g of dry tissue. This is significantly lower than in lung or anthracotic parietal sam ples (p < 0.01). Asbestos fibers were detected in normal pleural tissue only if fiber concentration in anthracotic lesions exceeded 1 X 10s f/g of dry tissue. Visual comparison of TEM photographs from a given patient showed similar particle burden in lung and anthracotic parietal samples and practically no particles in nonanthracotic parietal samples. The weight of tissue per surface unit of filter was roughly the same for the three samples (Figure 3). In addition to coal and amphibole fibers, black spots also contained rutile and nonfibrous particles such as silica, felspar, talc, kaolinite, mica, and metallic oxides (Fe, Al, Ti). Fiber Concentration and Previous Asbestos Exposure In asbestos-exposed subjects, mean fiber concentrations were 12.4 9.8 X 10* f/g of dry tissue in lung, 4.1 1.9 in black spots, and 0.5 0.2 in normal pleura. In unexposed patients, concentrations were 0, 0.3 0.1, and 0, respectively (Table 2). Known or probable asbestos exposure was consistently cor related with fiber concentration in the lung > 1 x 10* f/g (Fig ure 2A). However, no fibers were observed in the lung of patient seven, who had both chrysotile and tremolite fibers in anthra cotic and nonanthracotic parietal samples. The possibility of in adequate sampling of lung tissue cannot be excluded. In the three patients with mesothelioma, asbestos burden was greater in anthracotic samples than in lung samples. This differ ence was not significant, however, probably because of the small number of patients studied. Types of Asbestos Fibers Thble 3 summarizes the types of asbestos found in the different samples. Chrysotile fibers were found in small amounts in only one lung sample (patient 5) and in two anthracotic and four nonanthracotic pleural samples from exposed patients. Amphibole fibers accounted for 99.9% of fibers recovered in lung samples, 95.5% of fibers in black spots, and 61.3% of fibers in nonanthracotic samples. Crocidolite and amosite were the most common in subjects with a history of occupational ex posure to asbestos. Tlemolite, which is characteristic of environ mental exposure in Corsica (two patients in this series), was the second most common type. Fiber Dimensions Table 4 shows the mean dimensions and length distribution of the fibers obtained from the different samples. There was no sig nificant difference in length between the fibers from lung and anthracotic pleural samples. Conversely, there was a significant difference in diameter, pleural fibers being broader than pulmo nary ones (p < 0.001). Table 2 and Figure 2B show the concentrations of fibers 5 pm or longer; significant amounts were found only in lung and an thracotic pleural samples, except in patient 7, who also had fibers in nonanthracotic pleural samples. Ail fibers 3= 5 pm were amphiboles (crocidolite, amosite, and tremolite) and were associated with previous asbestos exposure, except in patient 14 (in whom amphibole concentration was at the limit of detection). As shown in Thble 2, the mean concentration of long fibers was 1.7 x 10* f/g in lung samples and 0.7 x 10* f/g in anthracotic pleural sam ples, the difference being nonsignificant. The proportion of long fibers (Table 4) was the same in lung and anthracotic pleural sam ples (21.7 and 22.5%, respectively). Microscopy of Black Spots Light microscopic examination of black spots revealed the pres ence of anthracosis associated with a concentration of lympho cytes, plasmocytes, and macrophages. Polynuclear neutrophils and eosinophils were scarce. Whenever it could be visualized, the surface of the mesothelial layer was normal. Congestion of capillaries was sometimes observed, and lymphatic vessels were noted in two patients. In one patient, a giant-cell granuloma was observed. Scanning electron microscopy (Figure 4) confirmed the pres ence of lymphocytes, red cells, and, most importantly, active mac rophages. These findings are in agreement with those reported in milky spots in animals by Kanazawa (24). Pores and crevices HWBUI0008921 Boutin, Dumortier, Bey, el oL: Oncogenic Asbestos Fibers in the Parietal Pleura 447 Figure 3. Comparison of particle burden in samples from the same patient (patient 2) (TEM photographs; original magnification; *3,600), Burden was similar in lung (A) and anthracotic parietal (8) samples. Particles were virtually absent from nonanthracotic parietal samples (Q. resembling Wang stoma (25) and corresponding to pleural-lym phatic communications were also visualized. DISCUSSION Black Spots and Milky Spots . Inhaled coal and black dust that reach the deep lung are engulfed by alveolar macrophages, some of which are drained through lymphatic vessels towards the visceral pleura (6) and mediasti nal lymph nodes (8, 9,18). Thus, anthracosis has been reported to be a good marker of pleural lymphatic vessels (26-28). An thracotic deposition in the parietal pleura is characterized by a pigmentation of lymphatic vessels along the intercostal spaces and by clusters of "black spots" located in the posterior and lower part of the costo-vertebral gutter and on the diaphragm. Jones (26) also reported in coal miners black spots within lymphatic vessels located in the middle of the diaphragm and along the lower ribs in the back of the thoracic cavity. TABLE 2 CONCENTRATIONS OF ASBESTOS FIBERS IN TISSUE SAMPLES FROM PATIENTS WITH AND WITHOUT A HISTORY OF ASBESTOS EXPOSURE (* 10s f/g OF DRY TISSUE)* Tissue Sample Lung Black spots Nonanthracotic Parietal pleura Length (pm) >1 ?S 1 ?5 >1 7= S Exposed 12.37*9.8' 1.67 * 0.75b 4.13* 1.86'c 0.73 4 0.47b 0.52 * 0.2C Unexposed 0.26 0.1c ~ " * Value* are expressed as mean SEM. -- = not detected. Statistical analysis: a-a: N5; b-b: NS; c-C p < 0.01. Several findings suggest that these black spots correspond to the milky spots observed in animals. Scanning electron micros copy has confirmed the similarity between these black spots and the milky spots described by Kanazawa (24). Both structures con tain inflammatory cells (macrophages and lymphocytes). Concentration of Asbestos Fibers in the Pleura and Lung The clear-cut concordance between fiber burden in black spots TABLE 3 CONCENTRATION OF AMPHIBOLE AND CHHYSOT1LE FIBERS (* 105 f/g OF DRY TISSUE) IN VARIOUS TISSUE SAMPLES OF 14 PATIENTS Lung Black Spots Nonanthracotic Pleura No. A C A C A c Exposed 1 70.91 2 7.08 3 3.S9 4 2.61 S 1.33 61 7~ 8 ND Nonexposed 9 0.65 10 -- n12 -- 13 NO 14 ND -- - - 0.063* -- ND _ -- ND ND 15.66 0.6 1.46 8.28 1.83 1.28 2,42 -- 0.4 0.39* -- 0.17* 0.6 -- 0.4 - _ 1.2 - *" -- - - 0.83 - 0.37 - 0.33 1.03 - 0.42 -- 0.3S* -- 0.3* 0.33 ---- - ---- -- - Definition of abbreviations: A amphlboles; C ** chrysolite; -- - fibers not detected; NO - not done. * Concentrations within or under detection limit HWBUI0008922 44B AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 153 1996 TABLE 4 MEAN DIMENSIONS (urn) AND PROPORTIONS OF FIBERS IN LUNG SAMPLES AND PLEURAL BLACK SPOTS Lung Black Spots p Value Mean length* Mean diameter* Maximum length Length 3> 5 pm Length > 8 pm Length > 15 pm 4.45 0.45 (2.9) 0.13 0.01 (0.10) 90 21.7% 11.4% 3.9% 3.82 t 0.22 (3.03) 0.1? 0.01 (0.13) 26 22.5% 10% 2.1% NS p < 0.001 * Values are expressed as mean SEM; geometric mean is given in parentheses. and in the lung is remarkable: in most subjects exposed to as bestos, fiber concentrations in lung and anthracotic pleural sam ples were >1x10* f/g, whereas there were practically no fibers in nonanthracotic pleural samples. In five of 11 patients, includ ing all three mesothelioma patients, the concentration of fibers was even higher in black spots than in lung samples. These find ings confirm that fibers are not only heterogeneously distributed Figure 4. Scanning electron microscopic appearance of a black spot. Macrophages (MA) are clustered along the edge of the lesion. Lym phocytes (LY) and red blood cells (R) can be distinguished (upper panel; original magnification: x252). More centrally, lymphatic crevices (C) and macrophages (MA) can be seen (lower panel; original mag nification: *356). in the parietal pleura (2, 16, 17) but also concentrate in black spots, probably by an active mechanism, whereas few or no fibers can be detected in normal pleural areas only a few centimeters away. This heterogeneity of fiber distribution in the parietal pleura could explain inconsistencies in previous reports in which tissue sampling was random. Dodson and associates (18) documented higher concentrations of fibers in the pleura than in the lung in five of eight cases, Kohyama and Suzuki (19) in six of 13, and Sebastien and coworkers (29) in only eight of 29. Indeed, Sebastien and coworkers stated that eight previously exposed patients had no fibers in the parietal pleura. Both Dodson and associates (18) and Sebastien and coworkers (29) noted that fiber concen tration was higher in costal and diaphragmatic plaques. Distribution of Potentially Carcinogenic Fibers (> 5 pm) According to the criteria proposed by Stanton and colleagues (14) and Pott (15) from animal experiments and confirmed in mesotheiial cell cultures (30), asbestos fibers > 5 pm are the most carcinogenic. In our study, fibers > 5 pm were detected in the parietal pleura of subjects with a history of asbestos exposure, where in some cases their concentration exceeded 1 x 10s f/g of dry tissue. These potentially carcinogenic fibers were observed almost exclusively in black spots (22.5 % of fibers were 5 pm or longer and 10% were 8 pm or longer). The diameter of fibers recovered in black spots was greater than 0.1 pm (mean diameter: 0. 19 0.01 pm). Our observation is thus contrary to the conclu sion by Lippman (31) that only fibers with diameters 4 O.I pm reach the pleural space. The proportion of long pleural fibers has been variable in previous reports. Le Bouffant (16) reported that 7% of amosite fibers and 5% of chrysotilc fibers in the parietal pleura were 5 pm or longer. Sebastien and coworkers (2) stated that 24% of araphibole fibers and 14% of chrysotile fibers in the parietal pleura exceeded 4 pm. Dodson and associates (18) found that 10% of amphibole fibers and 3.1% of chrysotile fibers were longer than 5 pm. Asbestos Type of Pleural Fibers In our study, the proportion of amphibole fibers was 99% in the lung, 95% in pleural black spots, and 61% in nonanthracotic pleura. This finding contrasts with those of previous studies that had underlined the preponderance of short chrysotile fibers in the pleura (2, 16-19, 29). There is no clear explanation for the fact that no chrysotile fibers were found in black spots, even when there was a confirmed history of exposure to chrysotile. An explanation would be that chrysotile clearance is stimulated by large (60 to 70 pm in di ameter), intensely active macrophages observed in black spots (Figure 4). From a technical point of view, short and thin chryso tile fibers could be less easily detected among a "background" of particles in anthracotic samples. Migration of Fibers toward the Parietal Pleura How do pathogenic, carcinogenic, or fibrosing fibers reach the parietal pleura? The answer to this question may involve several potentially interrelated pathways. Few studies have focused on the passage of fibers into the phys iologic pleural fluid. In rats, we (32) observed that after in tratracheal instillation crocidolite and chrysotile fibers pass into the pleural space and diaphragm, but recovered fibers were short, 1. e., 0.44 to 1.32 pm for chrysotile and 0.47 to 2.2 pm for crocido lite (only 5% of crocidolite fibers were longer than 5 pm). These observations are in accordance with those of Pott and colleagues (33) and Le Bouffant and coworkers (8). In the present study, the correspondence between the num- HWBUI0008923  Boutin, Dumortier, Rey, et al.: Oncogenic Asbestos Fibers in the Parietal Pleura 449 ber, type, and dimensions of fibers in the lung and anthracotic pleura is consistent with direct migration of fibers from the alveoli into the pleural cavity through the visceral pleura and a fjreabsorption by the parietal pleura through Wang stoma and/or milky spots. According to Hillerdal (34), connections between pulmonary and pleural lymphatics could also account for this migration. Another possible pathway for rapid dissemination of short fibers is the bloodstream (35). The glomerular vascular struc ture of milky spots may facilitate the uptake of circulating fibers. Indeed, after subcutaneous or intravenous injection of asbestos fibers in mice, Kanazawa and associates (20) demonstrated that fibers concentrated only in milky spots. Fibers can also migrate via the lymphatic pathway (28, 34). It is responsible for translocation of inhaled particles such as silica or coal into lymphoid tissue, particularly in regional lymph nodes (8, 9,18). It is not possible, however, to deduce which of these routes is preferential on the basis of current evidence. Though our results suggest a direct transpleural migration, our study was not designed to assess such a type of translocation. This study demonstrates that, whatever route they followed, long amphibole fibers concentrate in the same structures of the parietal pleura that trap other particles such as coal dust. The accumulation of potentially carcinogenic fibers in these black spots ^nake them possible starting points for asbestos-induced lesions of the parietal pleura. This mechanism is consistent with the clinical observation that early-stage mesothelioma is observed on the parietal and diaphragmatic pleura rather than on the visceral pleura and with the usual location of pleural plaques. Acknowledgment The writers thank Drs. P. Cau and A. Henric for their colla boration. References 1. Churg, A., and L. DePaoIi. 1988. Clearance of chrysotile asbestos from human lung. Exp. Lung Res. 14:567-574. 2. Sebastian, P., X. Janson, A. Gaudichec, A. Hirsch. and J. Bignon. 1980. Asbestos retention in human respiratory tissues: comparative mea surements in lung parenchyma and in parietal pleura. In J. C. Wagner, editor. Biological Effects of Mineral Fibers. IARC, Scientific Publi cations, Lyon, France. 237-246. 3. Churg, A., J. L. Wright, L. DePaoIi, and B. Wiggs. 1989. Mineralogic correlates of fibrosis in chrysotile miners and millers .Am. Rev. Respir. Dts. 139:891-896. 4. Anttila, S., A. Karjalainen, O. Taikina-aho, P. Kyyronen, and H. Vainio. 1993. Lung cancer in the lower lobe is associated with pulmo nary asbestos fiber count and fiber size. Environ. Health Perspect. 101:166-170. 5. De Vuyst, P., P. Dumortier, E. Moulin, N. Yourassowsky, P. Roomans, P. De Francquen, and J. C. Yemault. 1988. Asbestos bodies in bronchoalveolar lavage reflect lung asbestos body concentration. Eur. Respir. J. 1:362-367. 6. Gibbs, A. R., M. Stephens, D. M. Griffiths, B. J.N. Blight, and F. D. Pooley. 1991. Fibre distribution in the lungs and pleura of subjects with asbestos related diffuse pleural fibrosis. Br. J. Ind. Med. 48: 762-770. 7. Teschler, H., N. Konietzko, B. Schoenfeld, C. Ramin, T. Schraps, and U. Costabel. 1993. Distribution of asbestos bodies in the human lung as determined by bronchoalveolar lavage. Am. Rev. Respir. Dis. 147:1211-1215. 8. Le Bouffant, L-, S. Bruyere, H. Daniel, and G. Tichoux. 1979. Etude expirimentale de devenir des fibres d'amiante dans l'gppaieil respiratoire. Rev. Fr. Mai. Resp. 7:707-716. 9. Roggli, V. L., andT. L. Benning. 1990. Asbestos bodies in pulmonary hilar lymph nodes. Mod. Pathol. 3:513-517. 10. Craighead, J. E., and B. T. Mossman. 1982. The pathogenesis of asbestos-associated diseases. N. Engl. J. Med. 306:1446-1455. 11. Boutin, C., J. R. Viallat, and Y. Aelony. 1991. Practical Thoracoscopy. Springer-Verlag, Heidelberg, Germany. 12. McDonald, J. C., B. Armstrong, B. Case, D. Doell, W. T. E. McCaughey, A. D. McDonald, and P. Sebastien. 1989. Mesothelioma and asbestos fiber type. Evidence from lung tissue analyses. Cancer 63:1544-1547. 13. Wagner, J. C., M. T. Newhouse, B. Corrin, C. E. R. Rossiter, and D. M. Griffiths. 1988. Correlation between fibre content of the lung and disease iit east London asbestos factory workers. Br. J. Ind. Med. 45:305-308. 14. Stanton, M. F., M. Cayard, A. Tegeris, E. Miller, M. May, E. Mor gan, and A. Smith. 1981. Relation of particle dimension to carcinoge nicity in amphibole asbestoses and other fibrous minerals. J. Natl. Cancer Inst. 67:965-975. 15. Pott. F. 1980. Animal experiments on biological effects of mineral fibres. In I.C. Wagner, editor. Biological Effects of Mineral Fibers. IARC Scientific Publications, Lyon, France. 261-269. 16. Le Bouffant, L. 1980. Physics and chemistry of asbestos dust- In]. C. Wagner, editor. Biological Effects of Mineral Fibers. (ARC Scien tific Publications. Lyon, France. 15-33. 17. Boutin, C.,P.Sebastien, X. Janson, and J. R. Viallat. 1981-Mdtrologie des fibres mindraies dans des biopsies thoracoscopiques pulmonaires et pleurales. Rdsultats prdliminaires. Poumon-Coeur 37:253-257. 18. Dodson, R. F., M. O. Williams, C. J. Com, A. Brollo, and C. Bianchi. 1990. Asbestos content of lung tissue, lymph nodes, and pleural plaques from former shipyard workers. Am. Rev. Respir. Dis. 142:843-847. 19. Kohyama, N., and Y. Suzuki. 1991. Analysis of asbestos fibersin lung parenchyma, pleural plaques, and mesothelioma tissues of North American insulation workers. Ann. N.Y. Acad. Sci. 643:27-52. 20. Kanazawa, K., F. J. C. Roe, and T. Yamamoto. 1979. Milky spots (taches laiteuses) as structures which trap asbestos in mesotheliai layers and their significance in the pathogenesis of mesotheliai neoplasia. Int. J. Cancer 23:858-865. 21. Boutin, C., J. R. Viallat, P. Cargnino, and P. Farisse. 1981. Thora coscopy in malignant pleural effusions. Am. Rev. Respir. Dis. 124: 588-592. 22. Boutin,C., J. R.Viallat, P. Cargnino, andF. Rey. 1982. Thoracoscopic lung biopsy--experimental and clinical preliminary study. Chest 82: 44-48. 23. Viallat, J. R., C. Boutin, J.Steinbauer, A. Gaudichet, andG. Dufour. 1991. Pleura effects of environmental asbestos pollution in Corsica. Ann. N.Y. Acad. Set. 643:438-443. 24. Kanazawa, K. 1985. Exchange through the pleura. Cells and particles. In S. Chretien, J. Bignon, and A. Hirsch, editors. The Pleura in Health and Disease. Marcel Dekker, New York. 195-231. 25. Wang, N. S. 1983. Morphological data of pleura. Normal conditions. In J. Chretien and A. Hirsch, editors. Diseases of the Pleura. Mas son, New York. 10-24. 26. Jones, J. S. P. 1987. Pathology of the Mesothelium. Springer-Veriag, New York. 123-127. 27. Cottier, H.. F. Meister, A. Zimmennann, R. Kraft, A. Burkhardt, P. Gehr, and G. Poretti. 1987. Accumulation of anthracotic particles along lymphatics of the human lung: relevance to "hot spot" forma tion after inhalation of poorly soluble radionuclides. Radial. Envi ron. Biophys, 26:275-282. 28. Taskinen, E., K. Aftlman, and M. Wiikeri. 1973. A current hypothesis of the lymphatic transport of inspired dust to the parietal pleura. Chest 64:193-196. 29. Sebastien, P., X. Janson, G. Bonnaud, G. Riba, R. Masse, and J. Big non. 1979. Translocation of asbestos fibers through respiratory tract and gastrointestinal tract according to fiber type and size. In R. Le nten and J. M. Dement, editors. Dusts and Disease. Pathotox Publi cations, Park Forest South, 1L. 65-85. 30. Jaurand, M. C. 1989. Particulate^state carcinogenesis: a survey of re cent studies on the mechanisms of action of fibres. In J. Bignon, J. Peto, and R. Saracci, editors. Non-Occupational Exposure to Mineral Fibres. IARC Scientific Publications, Lyon, France. 54-73. 31. Lippman, M. 1988. Asbestos exposure indices. Environ. Res. 46:86-106. 32. Rey, F., J. R. Viallat, P. Farisse, and C. Boutin. 1993. Pleural migra tion of asbestos fibers after intratracheal injection in rats. Eur. Respir. Rev. 3:145-147. 33. Pott, F., F. Huth, and K. Spurny. 1980. Tumour induction after Intraperitoneal injection of fibrous dusts. InS.C. Wagner, editor. Bio logical Effects of Mineral Fibres. IARC Scientific Publications, Lyon, France. 337-342. 34. Hillerdal, G. 1980. The pathogenesis of pleural plaques and pulmonary asbestosis: Possibilities and impossibilities. Eur. J. Respir. Dis. 61: 129-138. 35. Monchaux, G., J. Bignon, A. Hirsch, and P. Sebastien. 1982.Translo cation of mineral fibres through the respiratory system after injec tion into the pleural cavity of rats. Ann. Occup. Hyg. 26:309-318. HWBUI0008924