Document 3NyjZQe6ZrmxLoJwxY102mn3y

Pathology of Human Asbestosis: A Critical Review Victor L. Roggli, M.D. Department of Pathology. Durham Veterans Administration and Duke University Medical Centers. Durham. North Carolina Although health hazards related to exposure to asbestos dust have been recognized since ancient times, it was not until after the widespread usage of asbestos dunng the Industrial Revolution that asbestos-related pulmo nary disease was identified and defined in modem terms.1 Dr. H. Mon tague Murray is generally credited with the first description of asbestos-re lated pulmonary disease, which was reported to a British parliamentary committee in 1906.' The patient was a 33-year-old man who had worked for 14 years in the carding room of an asbestos textile plant. In 1914. the German pathologist T. Fahr described the characteristic appearance of pul monary interstitial fibrosis in a 35-year-old asbestos worker. Fahr called at tention to crystals in the lung tissue and attributed the pulmonary fibrosis to asbestos exposure.13 Dr. W.E. Cooke provided the first detailed de scription of asbestosis in the general medical literature in 1924 and subse quently coined the term asbestosis.4 Since then, numerous cases of asbes tosis have been published in the medical literature in workers exposed to asbestos dust through mining and milling, manufacture of asbestos prod ucts. or utilization of asbestos-containing products. * Asbestosis has subsequently been defined by a number of investigators in slightly different ways.3-' but none of these definitions has gained wide acceptance or been uniformly applied by pathologists. The most compre hensive description of asbestosis and perhaps most widely cited definition are found in the 1982 report3 of the Pneumoconiosis Committee of the College of Amencan Pathologists and the National Institute for Occupa tional Safety and Health. This group defined the minimum features that permit the morphologic diagnosis of asbestosis as the "demonstration of discrete foci of fibrosis in the walls of respiratory bronchioles associated with accumulations of asbestos bodies/' In addition, a scheme for the pathologic grading of asbestosis was proposed, and the intraobserver and interobserver variability for pathologists using this scheme was assessed. Although previous protocols for the histologic grading of asbestosis had been published.9 19 they have not been applied systematically in largescale studies.3 Adv Pathol 2.31 -oO. 1989 ' 1989 Year Book Medical P'solishers, Inc. 0889-3969-89 U2-0J1-UOO-S04 00 31 83 t>0 V.L. Roggll tos fibers on tissue response in guinea pigs, in Waiton WH (ed): Inhaled Pani cles /V. Oxford. Pergamon Press. 1977. pp 455-474. 80. Davis JMG. Beckett ST. Boiton. RD. et al: Mass and number of fibres in the pathogenesis of asbestos-related lung disease in rats. Br J Cancer 1978; 37:673-688. 81. Crapo JD. Barry BE. Brody AR. O'Neil JJ; Morphological, morphometric, and X-ray microanalytical studies on lung tissue of rats exposed to chrysotile asbes tos in inhalation chambers, in Wagner JC (ed): Biofogicaf Effects of Mineral Fibres. Lyon. France. IARC Scientific Publications. 1980. pp 273-283. no. 30. vol 1. 82. Lee KP. Barras CE. Griffith FD, et al: Comparative pulmonary responses to inhaled inorganic fibers with asbestos and fiberglass. Environ Res 1981: 24:167-191. 83. Gross P: Is short-fibered asbestos dust a biological hazard? Arch Environ Health 1974: 29:115-117. 112 32 V.L. Roggii Though comprehensive in its scope, the aforementioned aescnption of asbestosis5 leaves a number of questions unanswered. How many asbes tos bodies must be observed before the diagnosis of asbestosis can be made9 This point is not directly addressed, although the description of "as bestos bodies" presumably means at least two. Other investigators have suggested that one asbestos body is sufficient when found in the presence of diffuse interstitial fibrosis.11 whereas others have stated that "clusters of three" asbestos bodies are needed.12 Also left unspecified are the number of tissue sections that should be searched for asbestos bodies, the magnifi cation to be used in the search, or whether hematoxylin and eosin (H&E'i or iron-stained sections should be examined. In addition, the preferred staining method (H&E or Masson's trichrome) for assessing fibrosis is not identified. It is the purpose of this study to describe in detail the pathologic features of asbestosis. to address the questions outlined above, and to ex amine the role of quantification of tissue asbestos burden in our under standing of human asbestosis. Asbestos and Asbestos Bodies Asbestos, from the Greek word meaning "unquenchable." refers to a group of fibrous silicates with the qualities of thermal and chemical resis tance. flexibility, and high tensile strength. As a result of these qualities, asbestos is an extremely useful material for many industrial applications, and thus it has been incorporated into more than 3.000 commercial pro ducts. l3 The two major types of asbestos are serpentine and amphiboie forms Table 1). Chrysotiie is the sole fibrous member of the serpentine family of minerals, and it accounts for some 90% to 95% of asbestos used commercially. The fibers tend to be curly rather than straight, a feature that denves from mismatching of the sheet silicate and brucite layers in the crystal structure.14 Chrysotiie has a very unusual structure, whicn consists TABLE 1. Types of Asbestos___________ Serpentine Chrysotiie Commercial amphiboles Amosite Crocidolite Noncommercial amphiboles Actinoiite Anthophyilite Tremolite 84 Pathology of Human Asbesiosis: A Cnticai Review 59 dwellers, urban dwellers and patients with pulmonary neoplasms. South Med J 1976:401-404. ' 60. Bhagavan BS. Koss LG: Secular trends in prevalence and concentration of pulmonary asbestos bodies: 1940 to 1972. Arch Pathol 1976; 100:539 541. 61. Roggli VL. Greenberg SD. Seitzman LH. et al: Pulmonary fibrosis, carcinoma, and ferruginous body counts in amosite asbestos workers: A study of six cases. Am J Clin Pathol 1980; 73:496-503. 62. Kobayashi H. Watanabe H. Zhang WM. et al: A quantitative and histological study on pulmonary effects of asbestos exposure in general autopsied lungs. Acta Pathol Jpn 1986; 36:1781-1791. 63. Pooley FD. Oldham PD. Urn C-H. et al: The detection of asbestos in tissues, in Shapiro HA (ed): Pneumoconiosis: Proceedings of the Internationa!' Confer ence. Johannesburg. 1969. Capetown. Oxford Univ Press. 1970. pp 108-116. 64. Danger AM. Ashley R. Baden V. et al: Identification of asbestos in human tis sues. J Occup Med 1973; 15:287-295. 65. Abraham JL: Recent advances in pneumoconiosis: The pathologists' role in etiologic diagnosis, in The Lung. LAP monograph 19. Baltimore. Williams & Wilkins Co. 1978, pp96-137. 66. Roggli VL. Shelburne JD: New concepts in the diagnosis of mineral pneumo conioses. Semin Respir Med 1982; 4:128-138. 67. Vallyathan V. Green FHY: The roie of analytical techniques in the diagnosis of asbestos-associated disease. CRC Cnt Rev Clin Lab Sci 1984; 22:1-42. 68. Whitwell F. Scott J. Grimshaw M: Relationship between occupations and as bestos fibre content of the lungs in patients with pleural mesothelioma, lung cancer, and other diseases. Thorax 1977: 32:377-386. 69. Ashcroft T. Heppleston AG: The optical and electron microscopic determina tion of pulmonary asbestos fibre concentration and its relation to the human pathological reaction. J Clin Pathol 1973; 26:224-234. 70. Wamock ML. Kuwahara TJ. Wolery G: The relation of asbestos burden 'o as- bestosis and lung cancer. Pathol Annu 1983; 18(part 21:109- 145. 71. Wagner JC. Moncnef CB. Coles R. et al: Correlation between fibre content of the lungs and disease in naval dockyard workers. Br J Ind Med 1986: 43:391-395. 72. Gvlseth B. Churg A. Davis JMG. et al: Analysis of asbestos fibers and asbestos bodies in tissue samples from human lung: An international interlaboratory trial. Scand J Work Environ Health 1985; 11:107-110. 73. Churg A: Asbestos fiber content of the lungs in patients with and without as bestos airways disease. Am Rev Respir Dis 1983; 127:470-473. 74. Churg A: An inflanon procedure for open lung biopsies. Am J Surg Pathol 1983: 7:69-71 75. Becklake MR: Asbestosis criteria. Arch Pathol Lab Med 1984: 108:93. 76. Dodson RF. Williams MG. O'Sullivan MF. et al: A comparison of the ferrugi nous body and uncoated fiber content in the lungs of former asbestos workers Am Rev Respir Dis 1985; 132:143-147. 77. Wamock ML. Wolery G: Asbestos bodies or fibers and the diagnosis of asbes tosis. Environ Res 1987; 44:29 - 44. 78. Holden J. Churg A: Asbestos bodies and the diagnosis of asbestosis in chrysotile workers. Environ Res 1986: 39:232-236. 79. Wnght GW. Kuschner M: The influence of varying lengths of giass and asbes- 111 Pathology of Human Asbestosis. A Critical Review 33 oi a magnesium silicate forming a sheet that is rolled up like a scroll. Con sequently. each individual fibril has a central capillary that can be readily visualized at high magnification in a transmission electron microscope. Chrysotile fibers are composed of numerous individual fibrils, and the fi bers have a tendency to split longitudinally into smaller fibnilar bundles or individual fibnls. The ampniboie group includes commercially valuable forms, crocidolite and amosite. and the noncommercial forms, tremolite. anthophyilite. and acnnolite.13 The latter are pnmanly important as con taminants of other minerals, such as talc or chrysotile asbestos. All of the amphr 'ie fibers tend to be straight and do not break apart into individual fibrils ai readily as does chrysotile. The amphiboles are distinguished from one another based on differences in chemical composition. 14~16 Asbestos bodies are the histologic hallmark of prior exposure to asbes tos. They were first described as "pigmented crystals" in the iung by Marchana ;n 1906.11 and early observers apparently confused these goiden brown, segmented structures with fungi.3 Cooke4 and Gioyne:S were among the first to recognize that these curious bodies had asbestos fibers at their core. They were initially termed asbestosis bodies, which was subsequently changed to asbestos bodies when it was discovered that they occurred in the lungs of some workers who did not have asbestosis.1 Expenmentai animal studies showed that a number of fibrous dusts (fibrous aluminum silicate, silicon carbide whiskers, cosmetic talc, and fibrous glass), wnen administered by intratracheal instillation into the lungs of hamsters, resulted in the formation of structures indistinguishable from as bestos bodies. These findings led Gross et al.19 to propose the noncommital term "ferruginous body" when the precise nature of the fibrous core was unknown. Churg and Warnock20 2i then employed energy dispersive spectrometry and electron diffraction to show that ferruginous bodies iso lated from human lungs and having a thin, translucent fibrous core were virtually always true asbestos bodies. The structure and development of the asbestos body has been described in detail by Suzuki and Churg-" and more recently by Churg and Warnock.-3 and Morgan and Holmes.-'4 Asbestos bodies form when an asbes tos fiber is inhaled and deposited in the distal regions of the lung paren chyma. Here the alveolar macrophages attempt to phagocytose the fiber (but fail because of its size) and in the process cover the fiber with a layer of ferroprotein material. The iron component gives the structure its charactenstic goiden brown appearance and also gives a strong reaction with the Prussian blue stain. The peculiar segmentation is thought to be due to frag mentation of the smooth, sheath-like coating, and it is thought that further "weathering" and dissolution of the coating eventually occurs.25 2b This sequence of events has been supported by scanning electronjnicroscopic observations of asbestos bodies isolated from human tissues.2' Hence for mation of asbestos bodies is a dynamic process, with fibers becoming coated, the coating material undergoing segmentation and dissolution, and the fiber eventually becoming uncoated and thus available to initiate the process all over again. Asbestos bodies are generally 2 to 5 |xm in diame- 58 V.L. Roggii 35. Sebastien P. Gaudichet A. Bignon J. et al: Zeolite bodies in human iungs from Turkey. Lab Invest 1981; 44:420- 425. 36. Hiilerdal G: Asbestos exposure and upper lobe involvement. Am J Roentgenol 1982; 139:1163-1166. 37. Hiilerdal G: Pleural plaques: Occurrence, exposure to asbestos, and clinical importance. Acta Unw Upsaliensis 1980: 363:1-227. 38. Wain SL. Roggii VL. Foster WL: Parietal pleural plaques, asbestos bodies, and neoplasia: A clinical, pathologic, and roentgenograp'nic study of 25 consecu tive cases. Chest 1984: 86:707-713. 39. Murphy RL. Becklake MR. Brooks SM. et al: The diagnosis of nonmalignant diseases related to asbestos. Am Rev Respir Dis 1986: 134:363-368. 40. Franzblau A. Lilis R: The diagnosis of non-malignant diseases related to asbes tos. Am Rev Respir Dis 1987; 136:790-791. 41. Pratt PC: Emphysema and chronic airways disease, in Dail DH. Hammar SP feds): Pulmonary Pathology, New York. Springer-Verlag. 1988. pp 651-669. 42. vorwaid AJ. Durkan TM. Pratt PC: Expenmental studies of asbestosis. Arch Ind Hyg Occup Med 1951: 3:1-43. 43. Wright JL. Churg A: Morphology of small airway lesions in patients with asbes tos exposure. Hum Pathol 1984: 15:68-74. 44. Kuhn C. Kuo T-T: Cytoplasmic hyalin in asbestosis: A reaction of injured alve olar epithelium. Arch Pathol 1973; 95:190-194. 45. Wamock ML. Press M. Churg A: Further observations on cytoplasmic hyaline in the lung. Hum Pathol 1980: 11:59-66. 46. Koss MN. Johnson FB. Hochhoizer L: Pulmonarv blue '"`dies. Hum Pathoi 1981: 12:258-266. ' 47. Brody AR. Hill LH: Interstitial accumulation of inhaled cnrysotile asbestos fi bers and consequent formation of microcalcifications. Am J Pathol 1982; 109:107-114. 48. Ndimbie OK. Williams CR. Lee MW: Dendriform pulmonarv ossification. Arch Pathol Lab Med 1987: 111:1062-1064. ' 49. Hiilerdal G. Heckscher T: Asbestos exposure and aspergillus infection. ur j Respir Dis 1982; 63:420-424. 50. Roggii VL. Johnston WW. Kaminsky DB: Asbestos bodies in fine needle aspi rates of the lung. Acta Cytol 1984; 28:493-498. 51. Kagan E: Current perspectives in asbestosis. Ann Allergy 1985: 54:464-474. 52. Amman K. Aisner J: Asbestos-Related Malignancy. Orlando. Grune & Strat ton. 1987. 53. Ackerman LV. Elliott GV. Alanis M: Localized organizing pneumonia: Its re semblance to carcinoma: A review of its clinical, roentgenographic and patho logic features. Am J Roentgenol Radium Ther Nucl Med 1954; 71:988-996. 54. Mintzer RA. Cugell DW: The association of asbestos-induced pleural disease and rounded atelectasis. Chest 1982; 81:457-460. 55. Churg A. Wnght JL: Small-airway lesions in patients exposed to nonasbestos mineral dusts. Hum Pathol 1983; 14:688-693. 56. Bignon J. Goni J. Bonnaud G. et al: Incidence of pulmonary ferruginous bod ies in France. Environ Res 1970: 3:430-442. 57. Smith MJ. Naylor B: A method of extracting ferruginous bodies from sputum and pulmonary tissues. Am J Clin Pathol 1972: 58:250-254. 58. Rosen P. Melamed M. Savino A: The "ferruginous body'' content of iung tis sue: A quantitative study of eighty-six patients. Acta Cytol 1972; 16:207-211 59. Breedin PH. Buss DH: Ferruginous (asbestos) bodies in the lungs of rurai 110 Asbestos Consumption Asbestos Body Composition Amosite & Crocidohte 96% Commercial Amphiboles Amosite and Crocidohte C hrysotite Non commercial Amohtboles Percent of Total Composition of Asbestos Consumption Asbestos Body Cores FIG 1. The industrial consumption of asbestos by fiber type compared with the composi tion of the cores of asbestos bodies by Tiber type. (From Roggli VL. Brody AR: im aging techniques for application to iung toxicology, in Gardner DE. Crapo JD. Massaro DJ teds): Toxicology of the Lung. New York. Raven Press. 1988. pp 117-145. Used by permission.) ter. 23 although by scanning electron microscopy 1 have observed bodies that were only about 0.5 yum in diameter. Their length ranges from 10 to over 250 yirrr3 and averages about 35 fim.28 Fiber length is an important variable in determining the formation of asbestos bodies: Morgan and Holmes29 have shown that fibers iess than 20 yim in length rarely become coated, whereas virtually all fibers 80 p.m or greater in length are coatea. The vast majority of asbestos bodies isolated from human lungs nave an amphibole asbestos core: commercial amphiboies. amosite or crocidohte. form the cores of most bodies isolated from the lungs of asbestos work ers10 and of men in the general population.30 whereas noncommercial am phiboies. tremolite or anthophyllite. account for most of the bodies isolated from women in the general population.30 This finding is possibly related to contamination of commercial talcum powder with tremolite and anthophyilite. The predominance of amphibole asbestos body cores is somewhat cu rious. considering that the bulk of asbestos used commercially is chrysonie asbestos (Fig 1). The reason for this anomaly apparently lies in the ready FIG 2. Comparison of the appearance of typical asbestos bodies with vanous types of pseudoasbestos bodies on Nuciepore filter preparations ot lung tissue aigests A, typical asbestos bodies isolated from the lung of an insulator with asoestosis and pleural mesothelioma. Thin translucent cores are visible in several bodies tarrow heads). Magnified 680x. B, pseudoasbestos bodies of tne sheet silicate type, iso lated from the iungs of an insulator with asbestosis and small cell carcinoma ot trie lung, have broad yellow cores larrowsl. True asbestos bodies are aiso present (upper center and lower left). Magnified 520 x. C, pseudoasbestos body isolated from the lungs of a coal-worker has a black carbon core that is coated with seg Pathology of Human Asbestosis: A Cntical Review 57 13. Craighead JE. Mossman BT: Pathogenesis of asbestos-associated diseases. N Engl J Med 1982; 306:1446-1455. 14. Pooiev FD: Asbestos mineralogy, in Antman K. Aisner J teds): Asbestos-Re lated Malignancy. Orlando. Grune & Stratton. 1987. pp 3-27. 15. Churg A: Fiber counting and analysis in the diagnosis of asbestos-reiated dis ease. Hum Pathol 1982: 13:381-392. lb. Roggli VL. Pratt PC. Brody AR: Asbestos content of lung tissue in asbestos- associated diseases: A study of 110 cases. BrJ Ind Med 1986: 43:18-28. 17 Marchand F: Ueber eigentiimliche Pigmentkristalle in den Lungen. Verb Dtsch Ges Pathol 1906: 10:223-228. 13. Gloyne SR: The presence of the asbestos fibre in the iesions of asbestos work ers. Tubercle 1929: 10:404-407. 19 Gross P. de Treville. RTP. Cralley Lj. et al: Pulmonary ferruginous bodies: Development in response to filamentous dusts and a method of isolation and concentration. Arch Pathol 1968: 85:539-546. 20. Churg A. Wamock ML: Analysis of the cores of ferruginous (asbestos) bodies rrom the qeneral population: I. Patients with and without iung cancer. Lab In vest 1977. 37:280-286. 21. Churg A. Wamock ML. Green N: Analysis of the cores of ferruginous (as bestos) bodies from the general population. True asbestos bodies and pseudoasbestos bodies. Lab Invest 1979: 40:31-38. 22. Suzuki Y. Churg J: Structure and development or me asbestos body. Am J Pathol 1969: 55:79-107. 23. Churg AM. Warnock ML: Asbestos and other ferruginous bodies: Their forma tion and clinical significance. Am J Pathol 1981; 102:447-456. 24. Morgan A. Holmes A: The enigmatic asbestos body: Its formation ana signifi cance in asbestos-related disease. Enuiron Res 1985. 38:283-292. 25. Glovne SR: The formation of the asbestosis body in me lung. Tubercle 1931: 12:399-401. 26. Botham SK. Holt PF: Development of asbestos bodies on amosite. chrysotile. and crocidolite fibres in guinea pig lungs. J Pathol 1971: 105:159- 167. 27. Mace ML. McLemore TL. Roggli V. et al: Scanning electron microscopic ex amination of human asbestos bodies. Cancer Lett 1980; 9:95-104. 28. Roggli VL. Pratt PC: Numbers of asbestos-bodies on iron-stained tissue sec tions in relation to asbestos body counts in lung tissue digests. Hum Pathoi 1983: 14:355-361. ' 29. Morgan A. Holmes A: Concentrations and dimensions of coated and uncoated asbestos fibres in the human lung. BrJ Ind Med 1980; 37:25-32. 30. Churg AM. Wamock ML: Analysis of the cores of ferruginous (asbestos) bod ies from the general population: III. Patients with environmental exposure. Lab Invest 1979: 40:622-626. ' 31. Crouch E. Churg A: Ferruginous bodies and the histologic evaluation of dust exposure. Am J Surg Pathol 1980: 8:109-116. 32. Roggli VL: Analytical scanning electron microscopy in the investigation of un usual exposures, in Romig AD. Chambers WF (eds): Microbeam Analysis- 1986. San Francisco. San Francisco Press. 1986, pp 586-588. 33. Dodson RF. O'Sullivan MF, Corn CJ. et al: Ferruginous body formation on a nonasbestos mineral. Arch Pathol Lab Med 1985; 109:849-852. 34. Roggli VL. Shelburne JD: Mineral pneumoconioses, in Dail DH. Hammar SP (eds): Pulmonary Pathology. New York. Springer-Veriag. 1988. Dp 589 617. 109 Pathology of Human Asbestosis. A Critical Review 35 mented terroprotein material. Magnified 800 x. D. pseudoasbestos body with a dark, slightly curved core composed of chromium, isolated from the lungs of a metal polisher. Magnified 670 x. (Part C from Rpggli VL. Mastin JP. Shelburne JD. et al: Inorganic particulates in human lung: Relationship to the inflammatory re sponse. in Lynn WS (ed): Inflammatory Cells and Lung Disease. Boca Raton. Fla. CRC Press 1983. pp 29-62. Used by permission. Part D from Roggli VL: Analyt ical scanning electron miscroscopy in the investigation of unusual exposures, in Romig AD. Chambers WF (eds): Microbeam Analysis --1986. San Francisco. San Francisco Press. 1986. pp 586-588. Used by permission.) 87 56 V.L. Roggli for the pathologic diagnosis of asbestosis. Nonetheless, based on the data summarized in Table 3. it seems unlikely that a patient with clinically sig nificant pulmonary interstitial fibrosis who has fewer than 10 fibers 5 p.m or greater in length per gram of dry lung tissue (1CP fibers, gm wet lung tissue) is suffering from asbestosis. Whereas the fibrogenicity of asbestos fibers 5 p.m or greater in length is well established.42-/9-82 the fibrogenicity of fibers less than 5 p.m in length remains unproved.83 Therefore, no tissue level of fibers in the latter size range should at the present time be pro posed as a critenon for the diagnosis of asbestosis. Acknowledgment The author gratefully acknowledges Susan Embry, James E. Linthicum, and Kenneth Holt for assistance with the illustrations, and Diane Evans for preparation of the manuscnpt. References 1. Castieman Bl: Asbestos: Medicai and Legal Aspects. New York. Harcourt. Brace. Jovanovich. 1984. 2. Lee DHK. Selikoff IJ: Historical bacxaround to the asbestos problem. Enuiron Res 1979: 18:300-314. " ' 3. Craighead JE: Eyes for the epidemiologist: The pathologist's roie in shaping our understanding of the asbestos-associated diseases. Am J Clin Pathol 1988: 89:281-287. 4. Cooke WE: Pulmonary asbestosis. Br Med J 1927: 2:1024-1025. 5. Spencer H: The pneumoconioses and other occupational lung diseases, in Spencer H (ed): Pathology of the Lang, ed 4. vol 1. Oxford. Pergamon Press. 1985. pp 413-510. ' 6. McCullough SF. Aresini G. Browne K. et al: Cnteria for the diagnosis of asbes tosis and considerations in the artnoution of lung cancer and mesotnelioma to asbestos exposure. Int Arch Occup Enuiron Health 1982: 49:357-361. 7. Kannerstein M. Churg J: Pathology of Asbestos-Related Diseases. Washington. DC. Armed Forces Institute of Pathoiogy. 1979. 8. Craighead JE. Abraham JL. Churg A. et al: The pathology of asbestos-associ ated diseases of the lungs and pleural cavities: Diagnostic criteria and proposed grading schema. Report of the Pneumoconiosis Committee of the College of Amencan Pathologists and the National Institute for Occupational Safety and Health. Arch Pathol Lab Med 1982: 106:544-596. 9. Report and recommendations of tne working group on asbestos and cancer. Br J Ind Med 1965: 22:165-171. 10. Hinson KFW. Otto H. Webster 1. et al: Criteria for the diagnosis and grading of asbestosis. in Bogovski P (ed): Biological Effects of Asbestos. Lyon, France. World Health Organization, 1973. 11. Churg A: Analysis of asbestos fibers from lung tissue: Research and diagnostic uses. Semin Respir Med 1986: 7:281-288. 12. Wamock ML. Prescott BT. Kuwahara TJ: Correlation of asbestos bodies and fibers in lungs of subjects with and without asbestosis. Scan Electron Microsc 1982: 11:845-857. 108 36 V.L. Roggli fragmentation of chrysotile into shorter fibrils and the fact that asbes tos bodies tend to form only on fibers that are 20 y.m or greater in length. As noted earlier, fibrous dusts other than asbestos can become coated with iron, or ferruginized. so that one must be cautious in identifying asbes tos bodies by light microscopy. Fortunately, most of the nonasbestos fer ruginous bodies, or pseudoasbestos bodies, can be distinguished from true asbestos bodies at the light microscopic level.31 The typical appearances of true asbestos bodies and the more common types of pseudoasbestos bod ies are illustrated in Figure 2. Bodies with broad yellow cores and promi nent knobbed ferroprotein coating on the ends or sides usually have sheet silicate (e.g.. talc) cores.21' 31 A number of metal oxides can assume on occasipn a fibrous form--e.g.. aluminum oxide, chromium oxide.32 iron ox ide.33 and titanium oxide--and these form ferruginous bodies with dark brown to black cores. In addition, coal dust and wood-stove dust may con tain fibrous fragments of coal or burnt wood, and these too can form fer ruginous bodies with black cores.34 Finally, the fibrous zeolite known as erionite can form ferruginous bodies that at the light microscopic level are indistinguishable from asbestos bodies.35 Zeolite bodies have not yet been reported in lung tissues from individuals in North America. Pathologic Features Gross Morphology Asbestosis is characterized by linear interstitial fibrosis that tends to be more severe in the lower lobes. ' These findings are in contrast to silicosis, which is charactenzed by nodular fibrosis that is more severe in the upper lobes.""4 Lungs with asbestosis have a reduced volume (Fig 3). which cor relates with restrictive findings on pulmonary function tests. The weight of the lungs is increased, the consistency is firm, and close inspection of the cut surface demonstrates gray streaks of fibrous tissue (Fig 4). In advanced cases, honeycomb changes may be observed (Fig 5) and are most promi nent subpieurally and in the lower lobes.8 Progressive massive fibrosis has been rarely described in asbestosis. and it is usually the result of exposure to a mixture of dusts. In exceptional cases, fibrosis in asbestosis is more severe in the upper lobes.36 The tracheobronchial tree and hilar lymph nodes do not show any charactenstic changes in asbestosis.8 Although the gross features previously described may be observed in a number of chronic interstitial lung diseases, a useful feature in asbestosis is the frequently associated pleural involvement.8 Diffuse pleural thickening is often present, and adhesions are variable. More characteristic of asbestos exposure is the finding of parietal pieural plaques, which are localized ar eas of ivory-colored pleural thickening with either a smooth or knobby, "candle-wax dripping" surface (Fig 6). These are usually iocated over the domes of the diaphragm or on the posterolateral chest wall, running along 88 Pathology of Human Asbestosis: A Cnticai Reuiew 55 posure to dusts other than asbestos can produce similar lesions and that there is no direct evidence that such lesions progress to alveolar septal fi brosis.43 JD With regard to the first argument, the same is true for diffuse interstitial (i.e.. alveolar septal) fibrosis as for peribronchiolar fibrosis: in ei ther case, the finding of asbestos bodies in histologic sections greatly in creases one s confidence that the fibrosis observed is related to asbestos exposure.11 If one further restricts the definition to cases where the major ity of bronchioles are involved, the chances of overdiagnosing such lesions as being asbestos related is further reduced. With regard to the second ar gument. it is true that progression from peribronchiolar to alveolar septal fibrosis has not been demonstrated experimentally, although the peribron chiolar area appears to be an early site of abnormality in experimental an imal studies.42 It should be noted that peribronchiolar fibrosis is a form of interstitial fibrosis, since the pulmonary interstitium includes not only the interstitium of the alveolar septa, but the subpieural connective tissue, sec ondary lobular septa, and connective tissue enveloping bronchovascular and broncniolovascuiar bundles as well. All of these sites may show in creased fibrous tissue deposition in individuals exposed to asbestos, and it is this wnter's opinion that the preponderance of the evidence does justify inclusion of peribronchiolar fibrosis in the definition of asbestosis. Other investigators have challenged the requirement of finding asbestos bodies in tissue sections before a histologic diagnosis of asbestosis is ren dered. The justification for that opinion has been firstly the observation that chrysotile forms asbestos bodies less readily than^do the amphiboles and many workers are exposed primarily to chrysotile.73 and secondly that some individuals coat fibers to form bodies much less efficiently than oth ers./6 " With regard to the first argument. Holden and Churg'8 have shown that asbestos bodies are readily found in histologic sections of chrysotile miners with asbestosis. and that these asbestos bodies do in fact have chrysotile asbestos cores. With regard to the second argument, there have been a few cases reported of patients with pulmonary fibrosis where the asbestos body content of the tissue was low (less than 100 per gram of wet lung tissue) butjhe uncoated fiber content of the tissue by TEM was considered high.'6 `` Such cases are apparently rare, as no similar cases were observed in our series of 76 patients with asbestosis. Furthermore, the 16 patients in our series with idiopathic pulmonary fibrosis who had relatively low asbestos body counts (see Fig 13) also had low uncoated fi ber counts.10 and it is unlikely that any of these cases were misclassified. The classification of the uncommon cases with interstitial fibrosis, absence of asbestos bodies in histologic sections, and elevated tissue asbestos fiber burden remains problematic, but the existence of cases of pulmonary fi brosis due to asbestos fiber inhalation that do not fulfill histologic criteria for the diagnosis of asbestosis seems plausible. In consideration of the lack of a uniform method for the analysis of tis sue mineral fiber content and the variable results obtained from different laboratories analyzing the same sample, it is not presently possible to rec ommend a specific tissue asbestos fiber content to be used as a criterion 107 Pathology of Human Asbestosis. A Critical Review 37 FIG 3. A, posteroanterior chest roentgenogram from an insulator with asbestosis. showing small lung volumes and bilateral reticuionodular infiltrates most prominent in the lung bases. B, CT scan from the same individual, showing prominent interstitial markings with honeycomb changes and peripheral accentuation. (Courtesy of Dr. Colleen Bergin. Department of Radiology. Duke University Medical Center.) the direction of the ribs.37, 38 The plaques are frequently calcified. They may develop after brief or low level exposures to asbestos and are often found in e absence of parenchymal fibrosis.37 38 These pleural abnor malities thus differ from the parenchymal disease in terms of epidemiology, clinical features, and prognosis.39 and the term "asbestosis" should not be 89 54 V.L. Roggli TABLE 6. Energy Dispersive X-Ray Analysis Data on 1,215 Fibers From 76 Patients With Asbestosis Coated Uncoated Total (%) Commercial Amphiboies 497 622 1.119(92) Noncommercial Amphiboles 4 19 23(1.9) Chrysotile 0 8 8(0.6) Other* 13 52 65(5.5) 'Includes talc, silica, rutile, kaolinite. miscellaneous silicates, fiberglass, iron-nch fi bers. and aiummum-nch fibers. fibrosis resulting from inhalation of asbestos-containing dust. Based on the foregoing analysis, reasonable criteria for the histologic diagnosis of asbes tosis can be established. Asbestosis may be defined as the presence of peribronchiolar fibrosis and asbestos bodies in histologic sections, with or without alveolar septal fibrosis. Since pther experimental exposures can produce some peribronchiolar fibrosis.43, 30 it is recommended that in the absence of alveolar septal fibrosis, a histologic diagnosis of asbestosis should be made only when the majority of the bronchioles show increased amounts of fibrous tissue. Occasionally the assessment of fibrosis can be difficult in uninflated tissue that is collapsed, congested, or consolidated by pneumonia, and the pathologist should be careful not to overinterpret such material as showing alveolar septal fibrosis.'4 In cases where the assess ment of presence and extent of fibrosis is not straightforward on H&Estained sections, it is recommended that Masson's trichrome stains be ob tained to assist in this evaluation. Similarly, the pathologist should be care ful not to interpret ferruginous bodies with black or broad yeilow cores as asbestos bodies, and only structures with the typical morphology that has been described8, 20 21 30,3,1 should be interpreted as asbestos bodies. In cases where asbestosis is suspected and asbestos bodies are not readily identified on H&E-stained sections, it is recommended that iron-stained sections be prepared and examined systematically using a mechanical stage and a magnification of 200x.28 With this approach, an average value of five or more asbestos bodies per cm2 of tissue section area exam ined would be expected in 95% of our cases of asbestosis. and two or more asbestos bodies per cm2 in all of our cases. Of course, asbestos bod ies are not necessarily distributed evenly in histologic sections.28 Therefore, more than one section should be examined in cases where asbestos bodies are sparse. Fibrosis was confined to the peribronchiolar region in 18% of the cases in our series (see Table 2), and some investigators have challenged the in clusion of such cases in the definition of asbestosis on the grounds that ex- 106 38 V.L. Roggii FIG 4. Coronal section of the lower portion of the lung of an insulator with asbestosis. There is pale gray, linear interstitial fibrosis especially prominent in the lower lobe. Note the visceral pleurai thickening laterally and adhesion of the diaphragm to the undersurface of the lung. applied to the pleural lesions.5 39 Others have argued that when one con siders that both pleural and parenchymal fibrosis derive from exposure to asbestos, the need for any distinction disappears.40 One has only to con sider the fact that asbestos exposure can produce asbestosis and malignant mesothelioma to see the fallacy in this argument. Asbestos workers are more often than not cigarette smokers, and the pa thologist must take care to distinguish abnormalities related to smoking from those related to asbestos. For example, severe emphysema of the centrilobular type may be present and can overshadow the fibrosis of as bestosis (Fig 7). Emphysema must be distinguished from honeycomb changes, and in this regard, distribution is a useful guide; emphysema tends to be most severe in the upper lobes, whereas honeycomb changes are most severe in the lower lobes. The cystic spaces of honeycombing are rather uniform, averaging about 0.5 cm in diameter and having visibly thickened, fibrotic walls. The spaces of emphysema are variable in size, ranging from just visible up to several centimeters across and do not have ''walls." Strands (representing remnants of blood vessels) are often seen traversing emphysematous spaces, and grossly visible fibrosis is not an ex pected associated finding in centrilobular emphysema.41 As is the case for most other pathologic changes in the lungs, the lesions of asbestosis can be best assessed by careful inspection of slices prepared from lungs that have been fixed under pressure by prolonged intrabronchial instillation of fixa tive solution.3 41 90 Pathology of Human Asbestosis: A Critical Review ' 53 to the wall of small airways (grade 3 or less) have values well below this ievel. This observation is in agreement with the findings by Churg'3 of rel atively low tissue asbestos burden in chrysotile miners with fibrosis con fined to the walls of small airways. There is also a statistically significant association between histologic score and total (coated plus uncoated) fiber content as assessed by SEM, but not between histologic score and asbes tos body content as measured by light microscopy (Table 5). Finally, in this study, there was no significant association between histologic score and pa tient age. duration of occupational exposure, or pack-years of smoking tsee Table 5). Wamock et al.70 have reported finding large numbers of commercial amphiboles. noncommercial amphiboles. and chrysotile fibers in patients with asbestosis. These authors used analytical TEM and examined all de tectable fibers 0.25 pm or greater in length. For comparison, the data from more than 1.200 fibers isolated from 76 patients with asbestosis and iden tified using analytic SEM are shown in Table 6. These data show that for fibers greater than 5pm in length, the vast majority (92%) are commercial amphiboies (mostly amosite with some crocidolite). Less than 2% are non commercial amphiboles and less than 1% chrysotile. Of interest is the observation that for fibers in the stated size range, nonasbestos mineral fibers34 are more common than chrysotile and noncommercial amphiboles combined. Diagnostic Criteria The histologic diagnosis of asbestosis is important because it provides inde pendent assessment of the presence or absence of pulmonary intersititial TABLE 5. Correlation of Histologic Grade of Asbestosis With Tissue Asbestos Content and Other Parameters* Correlation Coefficient (r) P Uncoated fibers gm i<5pm)SEM Total fibers/gm icoaied and uncoated). SEM Asbestos bodies, gm. LM Smoking history, pk-yr Age Duration of exposure, yr 0.46 0.44 0.26 0.22 0.12 0.06 <0.01 <0.01 NS NS NS NS "SEM = scanning electron microscopy: LM = light microscopy; pk-yr = packs smoked daily no. years smoked. 105 Pathology of Human Asbestosis: A Cntical Reutew 39 FIG 5. Coronal section of the left lung of an insulator with asbestosis and cavitary squamous cell carcinoma of the right lower lobe. Note the honeycomb cnanges :p. the medial portion of the lower lobe. Histopathology There is a range of microscopic changes in asbestosis depending upon the seventy of the process. These have been extensively illustrated in a recent monograph,3 and only the main features need be summarized here. The two microscopic features that are the sine qua non for the histologic diag nosis of asbestosis are pulmonary interstitial fibrosis and asbestos bodiesT Experimental animal studies42 and studies of autopsy lung tissue from as bestos workers8 suggest that the earliest changes of asbestosis involve in creased collagen deposition in the interstitium surrounding respiratory bronchioles, although this idea has been challenged by some investiga tors.43 With more extensive disease, there is extension both proximaily and 91 52 V L. Roggli greater in length as assessed by SEM is shown in Figure 14. These data are based on the 36 autopsied cases of asbesiosis from the author's con sultation files for which tissue was available for analysis of asbestos con tent. There is a statistically significant (P <.01) relationship between histo logic score and uncoated fiber content, although there is a wide range of scatter of the data points. The degree of correlation would likely improve with more extensive histologic and mineralogic sampling of the lungs and expression of the data as total lung burden rather than fiber concentra tion. Furthermore, the intercept for the regression line is approximately 100.000 fibers per gram of wet lung (or one million fibers per gram of dry lung), which coincides with the lower limit of the range of values shown in Table 3. In addition, it can be seen from Figure 14 that very few patients with alveolar septal fibrosis (grade 4 or higherl have uncoated fiber counts less than 100.000 per gram, although some patients with fibrosis confined Correlation between uncoated fiber count by scanning electron microscopy and histologic assessment of the seventy of asbestosis for 36 autopsied cases, using grading scheme of College of American Pathologists and National Institute for Oc cupational Safety and Health.8 The correlation coefficient (r) for the linear regres sion line is 0.46 (P <0.01). SEM = scanning electron microscopy. 104 40 V.L. Roggli FIG 6. A, Gross appearance of nemidiaphragm with parietal pleural plaque shows irregu lar. 10-cm plaque with both smooth and nodular areas, the latter resembing can dle-wax drippings. B, photomicrograph of typical plaque shows bundles of acellu lar. hyalinized collagen arranged in a "basket-weave " pattern. A focus of chronic inflammation is present at the interface of piaque and adjacent pleura. H&E. mag nified 68x. (Part A from Wain SL. Roggli VL. Foster WL: Parietal pleural plaques, asbestos bodies, and neoplasia: A clinical, pathologic, and roentgenographic study of 25 consecutive cases. Chest 1984; 86:707-713. Used by permission.) distally to involve the terminal bronchioles and alveolar ducts in the fibrotic process. Ultimately, there is radial extension to alveolar septa surrounding these structures (Fig 8). The fibrosis is usually most severe in the subpleural regions and in alveoli in closest proximity to the bronchioles. Scamng and distortion of bronchioles occasionally results in the lining of adjacent alveoli by cuboidal bronchiolar epithelium, a process sometimes referred to 92 -3 < I3 103 -3 < <* .0 *> L_UJ .-*3- 5< V> H</5 Pathology of Human Asbestosis: A Critical Review 51 CO XXX X ? CO CM TT -3i -O a* XXXXX ^ O -- CM 00 05 CM --' -3 + 0 XX U </> X 5S 3 V 3/) c>i) O'Ji a cm x X X < <. rr X X X cm =t -- 2 "2 ^^ -- = c " *f\ 11 2 I *3* (/) --. - --y E io --lc art a ^^rr V alues indicated represent the m edian counts derived from the data presented in the reference that is cited Asbestosis grade is as defined in each original s m iu v l ust tw o studies em ployed phase contrast light m icroscopy whereas the study by W ar nock el al used transmission electron m uioscopy and the study of W agner el al used both (phase contrast results from Figure 2 of the latter s t u d y /1 listed fiist. and EM results from Figure 1, listed below). W agner et al grading scheme of 0 to 4 has been m odified to 0 to 3 simply for purposes of tabulation Pathology of Human Asbestosis: A Cntical Review 41 FIG 7. Coronal section of lung from an insulator and cigarette-smoker showing moderate asbestosis and severe centrilobular emphysema. Note the visceral pieurai thickening enveloping the lung and extending into the interlobar fissure. as pulmonary adenomatosis. Secondary lobular septa may be of markedly increased thickness due to collagen deposition, and there is orten diffuse fibrotic thickening of the visceral pleura. In the most advanced cases, large zones of lung consist of alveoli with fibrotic walls, and there may be hon eycomb change. The latter is characterized by irregular, cystlike structures 1 to 15 mm in diameter, which are lined by cuboidal to low columnar ep ithelium and have fibrotic walls. Pools of mucus often accumulate in these spaces. A background of chronic inflammatory cells, consisting of lympho cytes and plasma cells, is often scattered within the fibrotic intersnnum. As bestos bodies may be found lying free within alveolar spaces or embedded within the fibrotic pulmonary interstitium (see Fig 8). The detection of as bestos bodies in histologic sections can often be enhanced by iron stains (e.g., Prussian blue), particularly when there are few bodies present. Fur thermore, connective tissue stains (e.g., Masson's trichrome) may facilitate the assessment of the extent of pulmonary interstitial fibrosis. The fibrotic process tends to be patchy, so that many sections may need to be searched to find the diagnostic features.3 Other histologic abnormalities are seen less commonly in patients with 93 50 V.L. Roggii croscopy. The median uncoated fiber count exceeds 1 million fibers per gram of dried lung tissue in all five studies. For companson. the median count in my laboratory for uncoated fibers 5 urn or greater in length from individuals with normal lungs and no known occupational asbestos expo sure is approximately 0.034 x 10 fibers/gm. The asbestos body content of the lung in 76 patients with histologically confirmed asbestosis is shown in Figure 13 and is compared with the re sults from 16 patients with idiopathic pulmonary fibrosis (1PF) and with 64 nonexposed controls. The results are expressed on a logarithmic scale as asbestos bodies per gram of wet lung tissue (which can be approximately converted to bodies per gram of dry lung tissue by multiplying by a factor of 10).13 The median count for the patients with asbestosis is 37.800 as bestos bodies per gram of wet lung tissue (AB/gm). whereas the median values for the patients with IPF is 16 AB/gm and for the controls is 0.4 AB/gm. There is substantial overlap between the IPF and "control'' groups, although a few IPF patients with low-level asbestos exposure have slightly elevated values. It should be noted that for 95% of the cases of asbestosis. the asbestos body content is 1.700 AB/gm or greater. This finding is useful because, at this tissue asbestos body concentration, it has been shown that several asbestos bodies should be observed on most 2 x 2 cm histologic sections when the sections are stained for iron and system atically examined.28 Thus the finding of asbestos bodies in histologic sec tions is a reasonable histopathologic discriminator between asbestosis and IPF. A few studies have investigated the relationship between tissue asbestos burden and the fibrotic response in humans, and these are summarized in Table 4. The study by Whitwell et al.DB shows a progressive increase in median total coated and uncoated fiber count from patients with mild (1 -) to severe (3 + ) fibrosis. Ashcroft and Heppleston09 showed similar progres sion in severity of fibrosis with increasing uncoated fiber count from no fi brosis to moderate !2^) fibrosis, but no further increase in fiber count from moderate to severe disease. The latter authors concluded that additional factors other than tissue fiber burden must be involved in the progression from moderate to severe disease.09 Wamock et al.,0 graded the severity of fibrosis on a scale of 0 to 3+ based on visual inspection of the cut surface of inflation fixed specimens, with 12- defined as microscopic fibrosis only (Table 4). Their data show no apparent correlation between the severity of fibrosis and total fiber content for_all fibers 0.25 ym or greater in iength as assessed by TEM. Wagner et al.'* graded the severity of fibrosis micro scopically on a scale of 0 to 4. which for the sake of convenience has been tabulated as 0 to 3 in Table 4 with their grade 1 fibrosis listed under l/2+. Their data show a progressive increase in optically visible and electron microscopically enumerated fibers with increasing seventy of asbestosis. The relationship between histologic asbestosis score using the method recently proposed by the Pneumoconiosis Committee of the College of Amencan Pathologists8 and the tissue content of uncoated fibers 5 ym or 102 42 V.L. Roggh FIG 8. A, low-power photomicrograpn ot a bronchiole from an insulator with asbestosis. There is peribronchiolar fibrosis with distortion of the bronchiole. The fibrosis extended to involve adjacent alveoli. H&E. magnified 52 x. B, higher magnification of area in box A shows asbestos bodies embedded within fibrotic penbronchiolar connective tissue (arrowheads). H&E. magnified 250 x. C, elsewhere, clusters ot typical asbestos bodies are present embedded within fibrotic interstitium. H&E. magnified 325 x 94 Pathology of Human Asbestosis: A Cntical Renew 49 rones examine the same sample'2 and considering the different techniques employed in each of the studies referred to in Table 3. For example. Ash croft and Heppleston09 used phase contrast light microscopy (PCLM) at a magnification of 400 x and counted all visible fibers, enumerating coated and uncoated fibers separately. Whitwell et al.b8 used PCLM and counted all fibers greater than or equal to 6 |x in length, counting coated and un coated fibers together. Wamock et al.' used transmission electron micros copy (TEM) and counted all fibers that exceeded 0.25 pm in length and had an aspect ratio (length to width) of 3 or greater. Wagner et al.'' used the PCLM method of Ashcroft and Heppleston09 as well as TEM. The present author's results were obtained by scanning electron microscopy (SEM) at a magnification of l.OOOx. counting all visible fibers with a length greater than or equal to 5 pm.10 The study by Wamock et ai.' and the study by Roggli also counted asbestos bodies by conventional light mi- (16) FIG 13. Histogram of asbestos body counts in 76 patients with asbestosis. 16 patients with idiopathic pulmonary fibrosis, and 64 nonexposed "controls." Honzontal bars indi cate median values. Note logarithmic scale. 101 Pathology of Human Asbestosis: M Cntical Review 43 asbestosis (Table 2). Although increased numbers of alveolar macrophages are often seen within alveolar spaces in asbestosis. in some cases, the alve oli are packed with sheets of macrophages in a pattern similar to desqua mative interstitial pneumonitis (Fig 9). Occasionally, foreign body type gi ant cells may be seen within the alveoli or. less commonly, within the fibrotic interstitium. Hyperplastic alveolar type 11 cells often line the fibrosed alveolar septa in asbestosis. and in some cases (roughly 7% of all cases with histologically documented asbestosis in the author's series), these type 11 cells contain irregular, waxy-appearing, deeply eosinophilic material (Fig 10). This cytoplasmic hyalin has the same tinctorial and ultrastructurai characteristics as the alcoholic hyalin observed in hepatocytes.44 However, this abnormality is not specific to asbestos and probably represents a pecu liar. nonspecific reaction to injury.45 These uncommon histologic abnor malities are in the author's experience observed in the more advanced stages of asbestosis. Rarely observed in patients with asbestosis are the so-called pulmonary blue bodies. These basophilic, laminated concretions are present in alveo lar spaces and consist primarily of calcium carbonate.46 They are not visu alized with polarizing microscopy in H&E-stained tissue sections, but are bnghtly birefringent in filter preparations of tissue digests (Fig 11) or in un stained histologic sections. Their mechanism of formation is unknown, although asbestos can cause abnormal accumulations of calcium salts in the lung parenchyma in experimental animal models of asbestosis.4' Pul monary blue bodies are not specific for asbestosis and may also be ob- TABLE 2. Histologic Findings in 100 Cases of Asbestosis Histologic Feature Always present Asbestos bodies Peribronchiolar nbrosis Often present Alveolar seDtal fibrosis Occasionally present Honeycomb changes Foreign body giant cells Pulmonary adenomatosis Cytoplasmic hyalin Desquamative interstitial pneumonitislike areas Rarely present Osseous metaplasia (dendriform pulmonary ossification i Pulmonary biue bodies Percent 100 100 82 15 15 10 / 6 2 1 95 48 V L. Roggli some asbestos can be found in the lungs of virtually all adults in industrial ized nations. 3b-02 Therefore, the mere demonstration of the presence of asbestos in tissue digests is of no particular value, and quantitative studies are required to understand the relationship between tissue fiber burdens and various disease processes. A variety of techniques have been de: scribed for the isolation and quantification of tissue asbestos content,63-'1' and it is beyond the scope of this review to consider these here. Rather, the intent is to review the information that has been obtained regarding the asbestos content of lung tissue in patients with asbestosis. Relatively few studies have been published examining the asbestos con tent of lung tissue in series of patients with asbestosis.68-'1 The data from these studies and the author's own senes are compared in Table 3. With the exception of the unusually high median count for asbestos bodies in the study by Ashcroft and Heppleston. and the high mean count for un coated fibers by electron microscopy in the study by Wagner et al. the val ues are roughly similar among the reported senes. This is rather remark able considering the wide range of values obtained when different labora- TABLE 3. Asbestos Content of Lung Tissue in Reported Series of Patients with Asbestosis4' Source Whitwell et al.ort No. of Cases 23 Method^ PCLM Asbestos Bodies/gm Dried Lung Ashcroft and Heppleston09 Wamock et al. 22 PCLM 22 TEM Wagner et al.'1 100 PCLM 12.2 (0.49-192) 0.123 (0.001 7.38) 170 TEM Roggii (present study) 76 SEM? 0.378 (0.006-16) Uncoated Fibers, gm Dried Lung (1.0-70) 32 (1.3-493) 5.08 (1.6-121) 1.5 (0.001-31.6) 372 (<1.0-10.000) 3.3s (0.18-125) `Values reported are the median counts for millions {10) ot asbestos bodies or uncoated fibers per gram ot dned lung tissue, with ranges indicated in parentheses, except for the study of Wagner et ai. 1 where only the mean value could be obtained from the data pre sented. hPCLM = phase contrast light microscopy: TE.M = transmission electron microscopy; SEM = scanning electron microscopy. tin these two studies, asbestos bodies were counted by conventional light microscopy Walues multiplied by a factor of 10 (approximate ratio of wet to dry lung .veigntl for pur poses of companson. 100 44 V.L. Roggii FIG 9. Low-power photomicrograph of lung of patient with asbestosis showing alveoii packed with sheets of alveolar macrophages {asterisks]. This pattern resembles that seen in desquamative interstitial pneumonitis. H&E. magnified 68 x. served in patients with asbestos exposure who do not meet histologic critena for the diagnosis of asbestosis.46 An additional unusual process that is rarely observed in patients with advanced asbestosis is dendriform pulmo nary ossification.46 This is characterized by branching spicules of bone (often with bone marrow) in association with pulmonary scarring, appar ently resulting from metaplasia of interstitial fibroblasts to osteoblasts. An- FIG 10. Photomicrograph of lung of patient with asbestosis showing hyperplastic aiveoiar type II pneumocytes. many containing cytoplasmic hyalin (arrowheadsi This ma terial has the same tinctorial characteristics as the hyalin found within nepatocytes in patients with alcoholic hepatitis. H&E. magnified 680x. Pathology of Human Asbestosis: /A Cntical Review 47 gist should look for asbestos bodies and penbroncniolar fibrosis in the ad jacent lung, since this process is frequently associated with exposure to as bestos.34 Differential Diagnosis Asbestosis must be distinguished from pulmonary injury secondary to inha lation of other toxic substances and also from other forms of pulmonary interstitial fibrosis. Peribronchiolar fibrosis may be associated with inhaia; tion of other mineral dusts, such as silica, iron oxides, or aluminum oxides.33 and with exposure to cigarette smoke.43 although the percentage of small airways with abnormalities due to these exposures is generally less than that observed with asbestos exposure.43 33 Diffuse interstitial fibrosis can also be seen with exposure to a variety of inorganic particulates.34 Thus, the finding of penbronchiolar or diffuse alveolar septal fibrosis alone is not sufficient for a diagnosis of asbestosis. and the identification of asbestos bodies in histologic sections is necessary to relate the fibrosis to asbestos. This is also true for idiopathic pulmonary fibrosis of the usual or desqua mative type, which has many features that overlap with asbestosis. Indeed, since idiopathic pulmonary fibrosis is a diagnosis of exclusion, asbestosis is one of the conditions that should be ruled out to make that diagnosis. The presence of asbestos bodies and visceral pleural fibrosis are helpful fea tures in this regard, as is the identification of parietal pleural plaques. Individuals who are occupationally exposed to asbestos are often ex posed to other dusts as well. For example, shipyard workers may receive substantial exposures to silica, talc, or welding fumes as well as asbestos. Crystalline silica constitutes one component of the lining of steam boilers in ships and is sometimes still used in sandblasting. Individuals engaged in boiler-scaling or sandblasting or working in the vicinity of these operations often have silicotic nodules in the hilar lymph nodes and occasionally in the lung parenchyma, especially in the upper lobes.34 Shipyard weiders of ten are exposed to asbestos, so that care should be taken to distinguish welder's pneumoconiosis from asbestosis. Welders are exposed to metal oxides (especially iron oxide), and sheet silicates. Iron oxides appear as in terstitial deposits of dark brown to black spherical particles, often with a golden brown rim.34 There is very little collagen deposition in response to these particles. Pseudoasbestos bodies with biack or broad yellow cores may be identified in histologic sections. Among 12 cases of welder's pneu moconiosis in shipyard workers from the author's consultation files, only four cases satisfied histologic criteria for the diagnosis of asbestosis.3 i.e.. peribronchiolar fibrosis and true asbestos bodies. Quantitation of Tissue Asbestos Burden A number of studies have demonstrated that when digestion-concentration techniques are employed to analyze adequate amounts of lung tissue. 99 Pathology of Human Asbestosis: A Critical Review 45 Jt It '4* .. FIG II. A, these mrraalveolar puimonary blue bodies have a somewhat laminated appear ance [arrows) and consist primarily of calcium carbonates. H&E. magnified 400 x. B. Nuciepore filter preparation of lung tissue digested in sodium hypochlorite. from same case illustrated in A. When examined by polarizing microscopy, the "blue bodies-' are brightly birefnngent and the laminations dearly visible (arrowheads). Magnified 325 x. (Courtesy of Dr. Fred Askin. Department of Pathology. University of North Carolina. Chapel Hill. NC.) other peculiar association with asbestosis is Aspergillus pneumonia. Hillerdai and Hecksher reported four cases of this unusual combination.401 and the author has personally observed five additional cases30 (and unpub lished observations). No other opportunistic fungal infections associated with asbestosis have to my knowledge been described. The mechanism may involve the suppression of cell-mediated immunity by asbestos.31 al though the reason for the specific predisposition for Aspergillus species is not known. It is not the purpose of this review to discuss asbestos-related malignan cies. and the interested reader is referred elsewhere for detailed discussion of this subject.8,52 However, the pathologist should be aware of some nonneoplastic pulmonary parenchymal processes that can mimic neoplasia . both clinically and radiographically, since he or she may be called upon to examine such cases by frozen section. The author has observed several | cases of organizing pneumonia in patients with asbestos exposure who un!j derwent thoracotomy for suspected malignancy. Organizing pneumonia is | a well-known mimicker of carcinoma of the lung.38 Whether this process f has an increased incidence in patients with asbestosis or whether these pa- 97 46 V.L. Roggli dents are merely observed more closely for potential malignancy is un known. In addition, a peculiar process known as rounded atelectasis (also known as folded lung syndrome) can occur adjacent to thickened, buckled pleura, and this lesion can radiographically resemble a peripheral coin le sion (Fig 12). Its radiographic features are described elsewhere.34 On his tologic examination, the lung parenchyma may be atelectatic or normal, although the overlying visceral pleura is invariably thickened. The patholo- FIG 12. A, lateral chest roentgenogram shows a peripheral mass posteriorly <arrowheads,1. B, CT scan from the same individual shows distorted bronchovascular bundles coursing into the mass which abuts the pleura. This radiographic appearance is charactenstic ot rounded atelectasis. (Courtesy of Dr. Colleen Bergin. Depanmer.t of Radiology. Duke University Medical Center.) 98