Document B8EQRBpab5kQ1Oz5ngjp3kdxw

Published by the American Meciicai Association -- --f v. i ...; , n?H;c;v: LnMJ Ac Asbestos-Assoc&iatied October 8, 1982 Volume 106, Number 11 e,Q'.p c!as rJ&` / Special Issue Archives of Patholcvv ana. Laboratory Medicine Report of the Pnoierrvcorfonie Comm: hoe of the Coilo.no of American PamoiogAu? ami the National institute for Occupational Corah/ and Heaitl John E. Craighead, ivtO, Chairman This p,:otocopy 'Vd AtcViW.0' Services Division c? *s*T'*X*t*'1 w ^'T**TM'*'"'*1 A* Opcode) , T h\i* <-ritlo lii v '-A-. f 1? imAjrVT'i'Jh* Vl `3 Typical ferruginous body within pulmonary macrophage. Note delicate, thin walls of alveoli. Asbosfosis was not evident in this tissue despite ihe presence of asbestos bodies (Prussian blue and eosin). '.`t Asbestos-Associated Diseases Special Issue Archives of Pathology and Laboratory Medicine The Pathology of Asbestos-Associated Diseases of the Lungs and Pleural Cavities: Diagnostic Criteria and Proposed Grading Schema Report of the Pneumoconiosis Committee of the College of American Pathologists and the National Institute for Occupational Safety and Health Chairman: John E. Craighead, MD Department of Pathology University of Vermont College of Medicine Burlington Jerrold L. Abraham, ML University of California San Diego Philip C. Pratt, MD Duke University Durham, NC Andrew Ciiurg, MD University of British Columbia Vancouver Thomas A. Seemayer, MD McGill University Montreal Francis H. Y. Green, MD Appalachian Laboratory for Occupational Safety and Health Morgantown, WVa Val Vallyathan, PhD Appalachian Laboratory for Occupational Safety and Health Morgantown, WVa Jerome Kleinerman, MD Mount Sinai School of Medicine New York Hans Weill, MD Tulanc University New Orleans n \l/3 Vol 106 No. 11 0-iobcr 8, 1982 4 i v7 Chief Editor Kenneth M. Brinkhous, MD University of North Carolina School of Medicine Department of Pathology Chapel Hill, NC Assistant Chief Editor William W. McLendon, MD Chapel Hill, NC Editorial Assistant Jean D. Wright Murray ft. Abell, MD, PhD Tampa, Fla Robert E. Anderson, MD Albuquerque. NM Rex B Conn MD Atlanta * Kenneth M. Earle, MD Washington. DC Thomas J. Gill, MD Pittsburgh EDITORIAL BOARD Robert A. Goyer, MD Research Triangle Park. NC Donald W. King, MD New York - Henry C. Pitot, MD, PhD Madison, Wis . Edvva-d A. Smuckler, MD, PhD ' San Francisco Robert W. Wisster, PhD, MD Chicago PUBLICATION STAFF Jeffrey R. M. Kunz, MD, Managing Editor and Director. Specialty Journal Department Marlene M. Hinsch, Production Manager John R. Sayban, Assistant Production Manager Cheryl Iverson, Editorial Processing Manager Paula Glitman, Assistant Editorial Processing Manager Offices; 535 N Dearborn St Chicago, IL 60610 Ralph W. Ulrich, Reprints and Permissions Copy Editing; Nicole Netlsr (Supervisor). Freder ick Bulzen Production Supervisors; Petty Fngerio (Advertis ing), Tnomas J. Handngan (Illustrations), Marlc-ne M. Hinsch (Edilonal). Elizabeth Petrikenas (Proof reading), JoAnrw Weiskopl (Layout) Proofreading: Verna Cottntan, Patricia Coyne. Winifred O'Pei'ly, Lisetle Sabbach Art and Senior Production Assistants: Karen Branham, Donna Hagler, Shirley Latronica. Denise Minkler. Marilyn Moroney. Terese Nolo, Susan M Routhier, Brenda Tropmski. Bonnie Van eleven, Sandra W<Ilia;ns, Ellen Yoshihara Production Assistants: Diane-Duke. Debra Lucas. Beverly Ramsey. Kathleen Sistek Manuscript Records: Rcnetta Konadu Editorial Assistant: Joan Domkowski Indexing: Phyllis H. O'Donnell (Supervisor), Jane P. Bouterse, Maiy Kay Tinerella Please address communications concerning accepted or published communications to AMA Specialty Journals. 535 N Dearborn St. Chicago. IL 60610. James H. Sammons, MD Executive Vice-President Thomas F. Hannon Deputy Executive Vice-President. Corporate, Publishing and Management Services George D. Lundberg, MD Vice-President. Scienulic Information , John T. Baker Vice-President. Publishing Norman D. Richey Director, Publication Production and Printing Division Lee A. Parent Director, Circulation Fulfillment Division .. Joseph Oennehy General Sales Manager; Director, Advertising Sales and Promotion Services Division Thomas J. Carroll Director. Department of Promotion Raymond Christian Director. Advertising Service Department Advertising and Subscription Information ADVERTISING OFFICES: Eastern: 600 Third Avo, Suite 700. New York, NY 10016 (Manager: Robert C. Corcoran [212-867-6640); Representatives: Phillip R. Allamore, John L. Reeves); Midwest/Farwest: 635 N O.rhorn St. Chicago, IL 00610 (Manager: Thomas J. Carroll [312-751-67521, Representah.eu: Eric K. Larsen, Alice M. Harvey) SUBSCRIPTION RATES--Tire subscription rates lor the ARCHIVES OF PATHOLOGY AND LABORATORY MEDICINE are as follows: lor members of the AMA, SI.50 included in the annual membership dues; tor nonmombers, S30 for one year, S56 for two years in the United Slates and US possessions; all oilier countries, one yer.r, $40; two years, S76 (add S6 surcharge to each one-year subscription, $9 for two years for air delivery to the United Kingdom and Europe only). Special yearly rales to residents and medical students in Ihe Untied Slates and US possessions, one year, $15: two years. S20. Address all subscription communication to American Medical Association, Circulation and Fulfillment Division, 535 N Dearborn St, Chicago, IL C06IO. Phone' (312) 751-0079 CHANGE OF ADDRESS --POSTMASTER, send all address changes to Caryl L. Wer theimer, Director of Fulfillment, 535 N Dear born St, Chicago, IL 60610. Notification of address change must be made at least six weeks in advance, including both old and new addresses, and a mailing label taken from the most recent copy. Include your new xip code number. The ARCHIVES OF PATHOLOGY AND LABO RATORY MEDICINE (ISSN 0003-9905) is published morrlhly by Ihe American Medical Association, 535 N Dearborn St, Chicago, IL 606)0, and is an official publication of tho Association. Copyright 1982 by tho American Medical Association 542 r i CONTENTS 544 Introduction 544 Historic Background 545 Occupational and Nonoccupational Exposure to Asbestos 546 Mineralogy of Asbestos 546 Asbestos Fibers in Tissue . 551 Asbestos-Associated Pleural and Pulmonary Disease . 571 Methods for Pathologic Study 575 Clinical Features of Asbestosis 575 Physiologic Features of Asbestosis (Pulmonary Function) 584 Roentgenographic Features of Asbestosis . 584 Mesothelioma 588 Carcinoma of the Lung 588 Pathogenetic Importance of Different Types of Asbestos 590 Host Factors in Asbestosis 591 Appendix 1: Methods for the Assessment of Lung Fiber Concentrations 592 Appendix 2: Mesothelioma Panels 592 Appendix 3: Pathologic Grading of Asbestosis 594 Appendix 4: Preliminary Evaluation of Grading Schema 595 Acknowledgments 595 References 596 Bibliography Arch Palhol Lab Mod--Vol 106, Ocl 8. 1982 543 j j 1I i j-fXTJ -L l yi The Pathology of Asbestos-Associated Diseases of the Lungs and Pleural Cavities: Diagnostic Criteria and Proposed Grading Schema Report of the Pneumoconiosis Committee of the College of American Pathologists and the National Institute for Occupational Safety and Health John E. Craighead, MD, Chairman Jerrold L. Abraham, MD; Andrew Churg, MD; Francis H. Y. Green, MD; Jerome Kleinerman, MD; Philip C. Pratt, MD; Thomas A. Seemayer, MD; Val Yallyathan, PhD; Hans Weill, MD rphe observations and conclusions X summarized herein represent the work of a committee of North Ameri can pathologists that was organized under the auspices of the National Institute for Occupational Safety and Health (NIOSH) and the College of American Pathologists to evaluate the pathology of asbestos-associated diseases and to establish standards for the grading of asbestosis. This effort was undertaken because of the need to develop a common basis for communication among pathologists, and between pathologists and radiolo gists, clinicians, pulmonary physiolo gists, occupational hygienists, and epidemiologists. During the course of this study, it became clear that the detailed mor phologic features of the asbestosassociated diseases of the pulmonary parenchyma of man have not been described fully in the medical litera- Accepted for publication Feb 11, 1982. From the Department of Pathology, Universi ty of Vermont College of Med'Ane, Rurlington (Dr Craighead); the University './ California, San Diego (Dr Abraham), the University of British Columbia, Vancouver (Dr Churgj; the Appala chian Laboratory fur Occupational Safety and Health, Morgantown, WVa (Drs Green and Vallyathan); Mount Sinai School of Medicine, New York (Dr Kleinerman); Duke University, Dur ham, NO (Dr Pratt), McGill University, Montreal (Dr Seemayer); and Tulane University, New Orleans (Dr Weill). Reprint requests to Department of Pathology', Medical Alumni liuilding, University of Ver mont, Murlington, VT 05-lOfi (Dr Craighead). ture. Thus, considerable attention was given to the histologic features of asbestosis at various stages of what is believed to be its evolutionary course. Because most published descriptions are concerned with disease that results from exposure to an undefined mixture of different types of asbestos, the lesions associated with the various types of asbestos also were studied comparatively, using collections of pathologic material from the United States and abroad. A grading schema that is functional and widely accepted by pathologists would have considerable usefulness in future studies of the clinical and epi demiologic aspects of the disease. Because many pathologists are uncer tain as to the criteria for the diagnosis of asbestosis, a generally accepted description of the lesions would assist in establishing uniformity for pur poses of diagnosis. Although this monograph ad dresses a variety of topics concerned with asbestos and asbestos-associated diseases, the committee was con cerned primarily with the pathologic features of the pulmonary parenchy mal disease associated with exposure to asbestos. Several outstanding books, conference proceedings, and committee reports that examine in detail the mineralogy and public health aspects of asbestos and the asbestos-associated diseases have been published in recent years. We refer the interested reader to them for more detailed presentations (see the "Bibliography"). HISTORIC BACKGROUND The fire-resistant properties of asbestos have been appreciated by man since ancient times. Although the mineral apparently was used in the fabrication of ceramics and tex tiles before the industrial revolution, it was not until the middle of the 19th century that commercial mining began. At that time, asbestos first was used in roofing materials and cement in the United States. During the next several decades, it was incor porated into commercial textile prod ucts in both North America and Europe. By the turn of the century', the industrial applications of asbestos seemed almost limitless and demand escalated. With the expanding recog nition of asbestos as a commercially extractable mineral, mines were opened in many parts of the world. It is unlikely that asbestosis was recognized as a disease before the term pneumonohoniosis was intro duced by Zenker in 1867.' Early in the 20th century, a few case reports from Europe and the United States indi cated a possible association of expo sure to asbestos dust with pulmonary fibrosis; in 1921 the term asbestosis was introduced by Cooke.2 By the 1920s and 19,"0s, the disease was more commonly recognized and series of 544 Arch Pathol Lab Med--Vol 106, Oct 8, 1902 Asbestos-Associated Discaso--Craighead el al cases appeared in the literature. Yet, the industrial use of asbestos increased. As the United States turned to the automobile for trans portation, the use of asbestos in brake linings became commonplace. In the 1930s, the technique of spraying asbestos insulation in construction was developed, and shortly thereafter asbestos began to be used in the man ufacture of insulation and pipe. In the United States, the occurrence of bronchogenic carcinoma in a patient with asbestosis was noted by Lynch and Smith in 1935.3 Their observation was soon confirmed by others, and over time an association was established. Although worldwide concern about the health effects of asbestos mounted, the problem be came more critical during the mobili zation that accompanied World War II. Warfare demanded ships and mod ern weaponry, and the war effort brought countless thousands of per sons into industries that used asbes tos. Although Klemperer and Rabin described tumors known by the desig nation mesothelioma in 1931,* recogni tion of the association of the neoplasm with asbestos awaited the reports of Weiss* and Leichner6 in the 1950s. In 19G0, Wagner and his colleagues' noted the relatively common occur rence of this rare neoplasm in persons exposed to South African cape "blue" asbestos, a type composed of unusual ly long and thin fibers. Subsequent experiments documented the induc tion of morphologically similar pleu ral lesions in animals after the intra pleural introduction of asbestos and of man-made fibers that had similar physical properties.8 Countless re ports since that time have confirmed the etiologic association of asbestos with mesothelioma. Clinical and pathologic evidence accumulated during the past several decades established the role of asbes tos in the pathogenesis of diffuse pul monary fibrosis. Evaluations of popu lation groups by radiologists and epi demiologists documented the common occurrence of pleural plaques in per sons exposed to asbestos. Definition of the role of asbestos in carcinogenesis required the efforts of epidemiolo gists, who established the association of asbestos with bronchogenic carci noma and with neoplasms in other organs. Noteworthy among these studies are the work of Doll in 1955' and the more recent observations of Selikoff and his associates,10 Enter line and Henderson," and investiga tors at the National Cancer Institute12 in this country. OCCUPATIONAL AND NONOCCUPATIONAL EXPOSURE TO ASBESTOS Although asbestos has been mined commercially since the turn of the century, it was not used widely in the United States until after 1930. Approximately 30 million tons of asbestos has been incorporated into construction materials and manufac tured products in this country since then. Each year, US industry con sumes more than 800,000 tons of asbestos. About one million living Americans either work or have worked in industries that manufac ture asbestos products, and countless millions more are exposed to the sev eral thousand asbestos-containing products that are available commer cially. In addition, persons are exposed inadvertently to the break down products of the mineral during the reconstruction and renovation of buildings and because of the deterio ration of manufactured materials that contain asbestos. Because of its heat- and fire-resis tant properties, as well as its high tensile strength and flexibility, asbes tos finds its greatest use in the con struction industry. About one third of the annual US consumption is incor porated into cement and cement prod ucts. Asbestos cement pipe and corrugated sheeting are the major products, among many others. The concentration of asbestos in cement varies considerably, but customarily is about 15% to 20%. Asbestos-con taining products also are used in floor tile, insulation board, plastic, mold ing, and paint, as well as spackle, patching, and taping compounds. In addition, asbestos is used as a filler in the construction of buildings. In the past, asbestos in the form of a suspen sion was applied by spray guns in the construction of steel-girdered build ings to prevent fire damage. This application now is prohibited in the United States because of its unre stricted pollution of the ambient air. Asbestos-containing products are used extensively to insulate electrical and heat-generating equipment and in the plumbing and shipbuilding industries. These products contain asbestos in amounts that range from 10% to almost 100%. Because the material is relatively uncontained and is manipulated during processing and use, breakdown products create a severe hazard. Thus, contamination of the environment often is subtle and the overall importance of the problem with regard to health is difficult to assess. Almost two million persons are employed in automotive sales, service, and repair, of whom some 900,000 are believed to be exposed frequently to asbestos from automobile brakes and clutch repair. Additional thousands, are exposed elsewhere in the trans portation industries. Asbestos is used in yarns and fab rics for protective clothing, safety curtains, and fire blankets. Asbestosreinforced plastics are used-in second ary industries too numerous to men tion. Contemporary standards adopted by the US Occupational Safety and Health Administration are based on the demonstration of fibers 5 gm or longer in ambient air using light microscopy. At present, no more than an average of 2 fibers per cubic centi meter of air during an eight-hour work period is permissible in the occu pational setting. Criteria based on concentrations of "long" fibers are a practical but relatively crude means for assessing environmental pollution, because the "short" (ie, < 5 j;m) asbestos fibers are not counted. Although long fibers are thought to play a major role in the pathogenesis of asbestosis and mesothelioma, the importance of short fibers in carcino genesis and fibrogenesis in the lungs is uncertain. Nonoccupational exposure to asbes tos is particularly insidious. It occurs in persons who reside near asbestos mines and mills, construction and Arch Pathol Lab Med--Vol 106, Ocl 8, 1982 Asbestos-Associated Disease--Craighead et al 545 i -- i j 'S 4 1 A : ! i 3 i { ( ! i v.i j r - V.-.-.T/ building demolition sites, and asbes tos dumps. Members of the household of asbestos workers may be exposed to contaminated work clothes. Avocational exposure occurs in those under taking do-it-yourself home repairs and auto body work. Obviously, it is difficult to document the severity of these types of exposure. MINERALOGY OF ASBESTOS The term asbestos refers to a family of naturally occurring, flexible, fibrous hydrous silicate minerals that are relatively indestructible and heatresistant, and have a high lengthto-breadth ratio (aspect ratio). Al though fibrous minerals are ubiqui tous in the earth's crust, only a few types have commercial importance: (l)ch rysotile, derived from serpentine rock, and (2) crocidolite, amosite, and anlhophyllite, which are classified as amphiboles (Fig 1). Fibers of both the serpentine and amphibole types consist of subunit fibrils. Individual serpentine fibers are made up of fibrils that have a layered, silicate structure, formed into scroll-like or concentric cylindric tubes. Presumably because of the high surface concentrations of magnesium hydroxide, these fibers exhibit a strong positive surface charge. The basic subunit of the amphibole is a silicon dioxide tetrahedron arranged in parallel chains and linked laterally by various cations. Amphiboles, in contrast to the serpentines, possess a slightly negative charge. Naturally occurring fibers of both types vary considerably in length, as well as in overall diameter. Fragmentation into shorter fibers and subunit fibrils occurs both with industrial processing and in tissue after inhalation (Figs 2 and 3). Belts of serpentine rock are located in nearly every mountain chain in the world. Chrysotile is found cryptically in rocks and soils of several types, and in potable water in many parts of the United States. It is mined in Califor nia and Vermont. About 8% of the earth's crust con sists of amphibole rock minerals. Because of the complexity of the structure and chemistry of the amphi boles, a wide variety of geologic strata Fig 1.--Mineralogic classification and chemical composition of common commercial types of asbestos (from New England Journal of Medicine, 1982;306:1446-1465). and soils contain fibrous and nonfibrous minerals somewhat akin to commercial amphiboles. The mineral ogy and nomenclature of these sub stances is complex. Amphiboles often are found with commercial deposits of chrysotile; eg, Canadian chrysotile, which is used widely in the United States today, . contains variable amounts of tremolite. The different types of asbestos seem to vary with regard to their importance in the causation of dis ease. Crocidolite is generally consid ered to be the most "dangerous" asbestos because of its strong associa tion with mesothelioma. It now is banned from importation into the United Kingdom, but it is still used to a limited extent in the United States. Amosite is used less widely in this country today. More than ?5% of the asbestos used in the United States is chrysotile, which comes from either domestic or Canadian sources. How ever, it often is impossible to deter mine the type of asbestos a patient has been exposed to, because the vari ous types of asbestos commonly are used interchangeably and as mix tures. ASBESTOS FIBERS IN TISSUE Two forms of asbe.stos fibers char acteristically arc found in the lungs. One, the uncoated, "bare" fiber, is identical to the inhaled particle unless altered by physical and chemical events in the tissue. Because of their narrow diameter, even relatively long fibers (> 5 nm) of inhalable size are detected only with difficulty by light microscopy; shorter fibers (which often are present in large numbers) are demonstrable only by electron microscopy. Polarized light microsco py is of limited use in detecting asbes tos in lung tissue because these rela tively thin fibers are only weakly birefringent. Similarly, the fibers cannot be identified positively in tissue by phase optics microscopy. The second form of fiber, the asbes tos body, is the hallmark of asbestos exposure (Color Fig 1). It consists of a fiber of variable length that is coated with proteins and iron compounds. The core of the fiber is transpar ent, colorless, and usually straight and unbranched; however, curved, branched bodies sometimes are seen. Their appearance has been described 546 Arch Pathol Lab Med --Vol IOC, Oct 8. 1982 Asbestos-Associated Disease--Craighead et al r-.. '. Fig 2. --Scanning electron micrographs of Canadian chrysotile (top, X7.500; bottom, X 18,000). Note curled features as well as variable lengths and diameters of fibers. Top, Fibrillary makeup of individual fibers is evident. In tissue, fibers often break down into individual fibrils and thus may not be evident by light microscopy. All specimens in Figs 2 and 3 were prepared under auspices of Union Internationale Contre Cancer and consist of commercial products. Properties of minerals extracted from individual geologic deposits of asbestos differ, which may affect pathogenicity. '^ Fig 3. -- Scanning electron micrographs of South African crocidolite (top, X7.500; bottom, X IS,000). Note rodlike features of rigid fibers of crocidolite and their variable lengths and diameters. Fibular substructure of fibers is apparent. See Fig 2. ' 548 Arch Pathol Lab Mod--Vol 10G, Oct 8, 1902 Asbestos-Associated Disease--Craighead el at Color Fig 1.--Typical asbestos bodies ir. lung exhib'ting grade 3 asbestosis. Note characteristic hollow cores of many fibers. Phagocytic rnultinucleate cells are unusually prominent in this tissue (hematoxylin-eosin). Arch Pathol Lab Med--Vol 106, Oct 8, 1982 Asbestos-Associated Disease--Craighead et al 549 =8 fO pm 550 Arch Pathol Lab Med --Vol 106, Oct 8, 1982 58a Fig 4.--Configurations of Prussian bluestained asbestos bodies in lung with asbestosis. Bodies were identified in photomicro graphs of lung and cut out to prepare mon tage. Asbestos bodies consist of an asbestos fiber core and surface deposits of protein- and iron-containing compounds. as "dumbbell," "drumstick," "beadednecklace," and "sausage," among oth er descriptors (Fig 4). Typical asbes tos bodies are from 10 to more than 300 gm long, although the usual length is about 30 to 50 pm. However, much smaller bodies of various shapes often are seen free in the tissue or in the cytoplasm of macrophages. The width is less than 3 pm, whereas the core can vary from 0.1 to 1.0 pm. A histologic stain for iron often is help ful in demonstrating asbestos bodies in tissue and should be used when the bodies are not readily demonstrated in lesions believed to be due to asbes tos exposure. The ratio of uncoated fibers to asbestos bodies in the lungs is high, often ranging from 5:1 to 10,000:1." The term ferruginous body has been suggested as a general descriptor for a fiber in tissue with an iron-protein coat.M Ferruginous bodies derived from a variety of inorganic and organ ic fibers have been identified in human lungs. Structures formed around carbon particles have black cores by light microscopy, and bodies formed on other silicates, eg, talc and mica, often have yellow cores. As most ferruginous bodies with colorless, transparent cores in human lungs consist only of asbestos, we prefer to use the more specific term asbestos body.,ilt When birefringent material is present in,the lungs in large amounts, exposure to common silicates like mica, talc, and kaolinite is likely. In workers exposed to talc, the inhaled dust usually contains asbestos be cause small amounts of lno fibrous Fig 5. -- Energy-dispersive x-ray spectrometry (microprobe) spectra of the four commercially important types of asbestos. Individual ele mental component of fiber are displayed at various loci along horizontal axis of cathode ray image; height of peak indicates relative amount of each element present. Upper left shows chrysolite; upper right, crocidolitc; low er left, anthophyllite; and lower right, amosite. Asbestos-Associated Disease--Craighead el al Fig 6. -- Electron microscopic diffraction pat terns of chrysotile (top) and crocidolite (bot tom). Chrysolite pattern is distinctive. Crocidolitc pattern is not specific, but is consistent with amphibole mineral. It is possible to identi fy mineral type using x-ray spectrometry and electron diffraction. mineral customarily are present in commercial deposits of talc. Evidence of exposure to silica dust, eg, quartz, often is found in conjunction with asbestosis. Pulmonary disease conse quent to a mixture of dusts, including asbestos, is an important consider ation in some occupational groups, and this should be rellccted in the diagnosis by the pathologist. The mirieralogic identification of either a specific fiber or a nonfibrous mineral in tissues is a ditricult and time-consuming task that requires specialized equipment and highly trained personnel. Usually, it is neces sary either to digest tissues in alkali or to incinerate them before analy sis. Light microscopic and ultrastructural methods have been developed for the identification and quantifica tion of fibers and asbestos bodies in tissue (appendix 1). These techniques may be useful for documenting expo sure in persons with no history of exposure, and for identifying the spe cific mineral or minerals involved. However, it is difficult to relate the fiber content of tissue to the extent and severity of parenchymal disease of the lung. The physical characteristics of par ticles in tissue vary, and in part reflect changes that occur after inha lation. For example, chrysotile fibers fragment into subunits and are leached of certain mineral constitu ents with the passage of time.17 Ultrastructural examination of individual fibers by either scanning or transmis sion electron microscopy is not a use ful technique for the identification of fibers because the individual fiber types have few specific structural fea tures. Energy-dispersive x-ray spec trometry (electron probe) and electron diffraction transmission mi croscopy permit the chemical and crystallographic analysis of a fiber in tissue (Figs 5 and 6). Because of the specialized nature of these time-con suming techniques, they are of limited diagnostic usefulness. ASBESTOS-ASSOCIATED PLEURAL AND PULMONARY DISEASE Long-term exposure to asbestos may result in disease that is restricted to either the pleura or the pulmonary parenchyma, although lesions of vary ing severity commonly are present in both anatomic locations. Pleural Plaques Plaques of the parietal and dia phragmatic pleura often are found in persons exposed to asbestos.1* In recent years, these lesions have been considered one of the pathologic and radiologic hallmarks of exposure. Plaques may prove to be the only evidence. The interval between the time of initial exposure and the devel opment of plaques is not clearly defined, but ten to 20 years usually elapse before they become evident clinically (Pig 7). As might be ex pected, the prevalence in groups of workers expose;! to asbestos increases with age. In some series of cases, more than 507o of the study group exhibited radiologic evidence of plaques 30 to 40 years after exposure1'''71 (Fig 8). All commercial types of asbestos induce plaques. In one study, the prev alence differed 13-fold between work ers in two regions mining the same vein of chrysotile in Quebec. Plaques are also found in members of the families of asbestos workers and in persons who reside in communities near asbestos mines and mills. They have been reported in persons exposed to mica, industrial talc (which con tains asbestos fibers), fibrous diatomaceous earth, fibrous zeolite, and other silicates. Plaques commonly occur among members of the general population in certain areas of Fin land,71 various places in central Europe, and in a few localities in Turkey.27 Because evidence of occupa tional exposure in these areas is usu ally lacking, it has been suggested that tiie lesions are due to the inhala tion of fiber and minerals that are found in soils and geologic outcrops. The plaques on the parietal pleura typically are located on the postero lateral aspect of the lower part of the thorax and on the dome of the thorac ic surface of the diaphragm (Figs 9 and 10). Customarily, they are not found adjacent to the apices of the lung or in the coslophrenic angles. Plaques also are situated at scattered sites on the peritoneal surfaces of the abdominal cavity, but these are rela tively rare lesions. Thin plaques usually are smooth and grayish white, whereas the thick er lesions are ivory-white and either have a smooth surface or are coarsely nodular. The size and shape vary: whereas plaques on the surface of the diaphragm typically are round and disklike, those located over the inter costal spaces are elongate and have an irregular configuration (Figs 11 through 13). Plaques customarily do not form on the visceral pleura, although ibis surface often is either diffusely or focaily thickened by fibrous tissue. This latter lesion ranges in severity from a simple loss of translueency to a shell of white, fibrous tissue that encases the whole lung. When honeycombing of the lung parenchyma exists, the surface may Arcli Palhol Lnb Med--Vol 106, Oct 8. 1982 Asbestos-Associated Disease--Craighead et al 551 % ol 60 m Asbestos Effusion Pulmonary Fibrosis Pleural Plaques Pleural Calcification 10 0 20-29 30-39 40-49 Years Since First Exposure Fig 7.--Clinical occurrence of various asbes tos-associated lesions among industrial work ers chronically exposed to airborne ininerai fibers. Data are based on physical examina tions and roentgenograms obtained systemat ically during 50-year period (from Epler and Gaensler"). Fig 6. -- Chest roentgenogram of 68-year-old ccr.struclion contractor who was hospitalized for treatment of recurrent transitional cell carc.noma of bladder. There were no unusual rcsp.ralory complaints. Calcified plaques on panelal pleural surface of rib cage, pericardi um. and diaphragm were inoi lental finding. v .. -> 552 Arch Pathol Lab Med--Vol 106, Oct 8. 198: Asbestos-Associated Disease--Craighead el ol Fig 9. -- Parietal pleural sur-uces o! thoracic eager, of two Finnish men at autopsy. Lungs and mediastinal structures have been removed. Lett, Note nodularity ot plaques. Right, Note smooth, confluent, sheetlike plaques. Diaphragmatic surface exhibits targe, circumscribed, diskhke plaques. There was no history of occupational exposure to asbestos. I Arch Pathol Lab Mod--Vo! 106, Oct 8, 1982 Asbestos-Associated Disease--Craighead et al 553 '.' &j-: ,-uj jwyif f i- ;* r i %' / :J Fig 10.--Computed lomograms of pleural plaques (arrows). Loft. Calcified paravertebral and peripheral plaques. Right, Calcified mediastinal, paravertebral plaques (from Preger'4). Fig 11.--Parietal pleural surface and parietal pericardium (PP) of diaphragm of American man with history of occupational exposure to asbestos. Disklike configuration and nodularity are typical of these lesions. THT' `` ' ) i-.a. 554 Aren Pathol Lab Med--Vol 10G. Oct a, 1932 .t V Asbestos-Associated Disease--Craighead et ol IF:'.- 1 1 Fig 12.--Roentgenogram of calcified plaques adjacent to ribs in autopsv specimen. Plaques appear to originate on inner surface of ribs and grow onto surface of intercostal muscle (from Preger'5). ' Fig 13.--Nodular plaques on pleural surface of membranous diaphragm of Finnish man with no known occupational exposure to asbestos. I I 3 V r 1 Arch Pathol Lab Mod--Vo! 106, Ocl 8, 1982 Asbestos-Associated Disease--Craighead et al 555 ! i 1 i i \ i} i ( ';#yvi . . Fiy 14. Microscopic lectures of typical pleuial plaque. Note ``basket-weave" arrangement of collagen ' bundles and relative acellularity of tissue. These lesions usually arc not vascularized. 556 Arch Palhol Lab Med--Vol 106, Ocl 8, 1982 Asbeslos-Aasocialed Disoase--Craighead et ol have a coarse nodularity that resem bles a cirrhotic liver. Adhesions between the visceral and parietal pleura are common, particularly at the lung bases. The degree of pleural thickening docs not reflect the severi ty of the pulmonary parenchymal dis ease and it can occur in the absence of fibrosis. Pleural pigmentation is not a feature of asbestosis and its presence suggests exposure to other types of dust. Microscopically, pleural plaques consist of dense strands of hyalinized, relatively acellular, nonvascuiar col lagen lined by a surface of mesothelial cells. The collagen fibers lie parallel to the surface and show a reticulated mesh appearance, the so-called bas ket-weave pattern (Fig 14). The colla gen fibers in noduiar plaques have a whorled arrangement. Occasional plaques exhibit elastic fibers and scat tered, thin-walled blood vessels. Inflammatory cells are not present, but may be found nearby. Calcifica tion is a common but variable feature that has proved useful in radiologic epidemiology. Occasionally, fibers of asbestos can be found in plaques by electron microscopy after either digestion or incineration of the tis sue. Asbestosis Asbestosis is pulmonary fibrosis consequent to the accumulation of air borne asbestos in the lungs. It does not refer to the lesions of the pleura already described. The severity of the disease varies. It depends in large part on the type and duration of exposure, the concentration of dust in the ambi ent air, and the fiber "burden" of the lung. Factors unique to the host (eg, immunologic responsiveness, concom itant disease processes, exposure to other types of dust, and cigarette usej also are believed to be important, although their relative significance in the. evolution of the disease is uncer tain. Asbestosis is a chronic and often progressive disease. Although decades usually elapse before its signs and symptoms become clinically evident, shorter latency periods have been noted in asbestos workers who experi enced exceptionally heavy exposure. dross Features.--The pathologic foa- tures of the advanced stages of pulmo nary asbestosis have been recorded in the classic descriptions of the disease. As expected from physiologic studies and the chest films, the lungs are small and firm. The cut surface is striking because of its dark-brown color. Gray streaks of fibrous tissues outline interlobar and lobular septa and focally interdigitate the lung tis sue. The visceral pleura is thickened by dense collagenous tissue (Fig 15). The occurrence of adhesions varies. The parenchymal fibrosis, which is linear and reticular, is most promi nent in the lower lobes. The bronchi are normal unless the patient has been a heavy smoker or other diseases exist. The tracheobronchial lymph nodes do not have characteristic changes. Commonly, they are en larged and brown-black, but are nei ther firm nor fibrotic. In advanced cases, the parenchyma exhibits honeycombing that is most prominent subpleuraily and in the lower lobes (Color Fig 2). In contrast with emphysema, the airspaces of the honeycomb lung result from tissue fibrosis, with retraction of the residu al structures (Fig 1G). These revised cavities have diameters that range from 1 to 15 mm, and have visibly thickened walls. No evidence indicates that exposure to asbestos contributes to the development of emphysema. Although progressive, massive fibro sis (as in coal workers' pneumoconio sis23 and silicosis24) has been described in persons with asbestosis, it is rare and usually consequent to exposure to a mixture of dusts. The evaluation of fibrosis of the parenchyma on gross examination of the lungs requires a careful inspection of slices of tissue that have been inflated by the intratracheal instilla tion of fixative solution. Because the lesions of advanced asbestosis usually are most prominent in the lower lobes and subpleural tissue, a detailed description of the gross changes is mandatory. In addition, photographic documentation is recommended. The pathologist should attempt to deter mine the relative proportion of a whole lung section that exhibits fibro sis. Although these features are typical of the overt case of asbestosis, the lungs of others with less severe expo sure exhibit only moderate degrees of parenchymal fibrosis that are not clearly evident on gross examination. In the evaluation of these specimens, the histologic assessment of the lungs using the diagnostic criteria to be summarized may be pivotal in estab lishing a diagnosis. Microscopic Features.--The severity of the histologic changes in the lungs in asbestosis varies, presumably in large part because of differences in the intensity and duration of expo sure. At one extreme, the features of typical asbestosis are obvious after a brief examination of a tissue section; at the other, the lesions are subtle, for they may be identified only at scat tered sites in representative histologic sections of the pulmonary tissue. It should be remembered that the lower lobes and subpleural regions of the lung are involved most severely by the disease. Pathologists must critically _ evaluate the lesions of pulmonary fibrosis using the criteria to be cited before implicating asbestos. A history of exposure is supportive but not definitive evidence. Conversely, its absence does not exclude the diagnosis because so many cryptic forms of exposure exist. Correlative evaluation of experi mental and human materials has pro vided insight into the morphogenesis of the disease at an early stage of evolution. Inasmuch as the patholo gist often is obliged to establish a diagnosis in cases with undocumented exposure and mild disease, this sec tion will emphasize the lesions we believe occur early in the pathogenetic sequence of events. Accordingly, an attempt has boon made to develop a construct based on our concept of the disease from its initial stages to the more severe, overt form. Little is known about the develop ment of the early lesion of asbesrosis in man. Thus, the evolution of the process must be inferred from a study of animals exposed to asbestos in inhalation chambers.23 28 Asbestos seems to be deposited first in the respiratory bronchiole and the alveo lar duct. Some observers believe an acute polymorphonuclear leukocyte Arch Pathol Lab Med -Vol 106, Oct 8, 1902 Asbestos-Associated Disease--Craighead ot at 557 \ 1 Fig 15. -- Examples of librosis of visceral pleura (top) and intorlobnr septum (boltoni) in lungs of two English dockworkers with long histor.es of exposure to asbestos. There is pulmonary asbestoses, grade 2B. Note adhesions (arrow and arrowhead) to parietal pleura. In contrast with plaques, winch typically are found on parietal pleura, these lesions often are vascularized and consist of irregular bundles of nonhyalinized fibrous tissue. Involvement of interlobar septum (bottom) is typical of asbestosis (hematoxylm-eosm). 558 Arch Pathol Lab Med--Vol 106, Oct 8, 1902 Asbestos-Associated Disease--Craighead el al A*/^^1** ; *'* * *r; . . .,*>--i- V- -h, . ' '5* ` V,- ; : ' ' m/m../.- Mi V , '.O 1 ' "-*a , .. y. vT~f~. .--*.* /a .. >** ' -.. ,- , k*-' ' **> ' '. ' -i " *v. * ' '" ' ii' , ' 5 V4 -?- 1 'M - V>. - '. ' . ' ; ' '.. . A <8 i ^ , * vi , ' i. i- . ? . v. ' ' A *. \y " . ; 7 `-r.v/a-rfj-r Y ' ! - . : k . '* * M; 1 < % Ml ; * rl1 %. Fig 16. -- Fixed portion of upper lobe of lung of insulation worker. Note diffuse reticular fibrosis and associated honeycombing in subpleural tissue. : . : ^!; response is provoked as an early and transient reaction to dust deposition:'')u (J. Abramowitz, MD, personal communication, June 8, 19S1). Al though this might occur, the macro phage usually is the prominent cell in the distal airways at the time of pathologic examination. In the respiratory bronchioles, the lesion associated with asbestos depo sition consists of a thickened wall attributable to the accumulation of rcticulin and collagen. Either cuboidalization of the epithelial cells or both squamous cell and goblet cell metaplasia can be observed in the mucosa. In typical early asbestosis, only an occasional pulmonary subunit is affected (Figs 17 and IS). Thus, histologic sections obtained from rep resentative sites throughout the lungs must be examined. The criteria that permit the pathol ogist to establish the diagnosis of asbestosis have evolved during a review of many cases of the disease. Presently, the minimal features that permit the diagnosis are the demon stration of discrete foci of fibrosis in the walls of respiratory bronchioles associated u'ith accumulations of asbestos bodies. These morphologic findings, although adequate to estab lish the diagnosis of asbestosis in an early evolutionary stage, have not been shown to result in function'd and radiologic alterations. The. demonstra tion of asbestos bodies in the absence of fibrosis is insufficient evidence to justify the diwjnosis of asbestosis. Con versely, a definitive diagnosis of asbestosis cannot, he made in cases that show characteristic fibrosis in the absence of asbestos bodies, even in a patient with a history of exposure. Because asbestos bodies are unevenly distributed in tissue, an adequate number of samples should be exam ined thoroughly. In the lesions, asbestos bodies are located either in the walls of the bron chioles or in the airspaces, where they are often closely associated with mac rophages. When only a single asbestos body is found in a histologic section, it is necessary to demonstrate addition al bodies (either in deeper sections of the same block or in other samples of tissue) to establish the diagnosis of asbestosis. Initially, first-order respiratory bronchioles are affected by the fibrotic process; in more advanced stages, the second- and third-order branches and alveolar ducts are involved. Con comitantly, the terminal bronchioles Arch Pathol Lab Med--Vol 106, Oct 6, 1982 Asbestos-Associated Disease--Craighead cl al 559 560 Arch Palho! Lab Med Vol 106, Oct 8, 1982 Asbestos-Associated Disease--Craighead el at fibrosis with honeycombing that is particularly prominent in basal portions of lower lobe, and in subpleural parenchyma. Honeycombing is typical of end-stage diffuse pulmonary fibrosis of many different causes, but prominence ol lesions subjacent to pleura and in lower lobes of this case is typical of asbcstosis. Arch Palhol Lab Wed--Vol 106, Ocl 8, 1982 Asbestos-Associatod Disease--Craighead ot el 561 Fig 17. --Lung tissue from 65-year-old woman who had been employer* in textile winding operation for 37 years, "op, Grade 1 lesion. Fibrosis is limited to walls of respiratory bron chioles (hemaloxylin-eosin [HE]). Bottom, Enlargement of boxed area in upper illustra tion shows two asbestos bodies (arrows) (HE). 1rtx - , ._ v' * tK! *T hW, *?> V. . ,, v'.-r * ,'i'y I x '> V >. -.s ' : ^ ^ % Wv V/^-i $ ... : --"vl ' -- V - . .'t ' 1 ' , ` * ` Vi p'9- , .,, ' . .'I... .. . y i '> ' ' ,. ...' - . , .'"= ^ ; "/ --- .-----A / ...... -- * . -. -,. v* \v; , \ L' V v-r! . - V *\ '7 1 f:, - '!' ' ' ' 'A V h ' ; i. v. 5 v -s f; <' ' fj- 7 /<$- ; .--*- >4 :k Tj -' <* ' K. .V \ . \!j , \ . '"7 .. '.V ri*!. S 'J\- ' V"'* '' ' f,V'-. ' Vj ; " v.-, ' 62 Arch Pathol Lab Med--Vol 106, Oct 8, 1982 Asbestos-Associated Disoase--Craighead et at Fig 18.--Lung tissue from 64-year-old man who had been employed as spinner in textile mill for 34 years. Top, Grade 2 lesion. Fibrosis is primarily located around respiratory bronchioles, but there is extension into adjacent alveolar septae (hematoxylin-eosin [HE]). Bottom, Enlargement of boxed area in upper illustration shows several asbestos bodies (arrows). There are fragmented bodies and anthracotic pigment in the fibrotic tissue and pulmonary macrophages (HE). Arch Pathol Lab Mod--Vol 106, Oct 8, 1982 Asbestos-Associated Disease--Craighead et al 563 # 7 * Fig 19.--Lung tissue from 55-year-old foundry worker. Top, Grade 3 lesion. Note fibrosis interconnecting respiratory units. Analysis of tissue gave 6.4 X I0a fibers of anthophyllite per gram of dry tissue (hematoxyHn-eosin [HE]). Bottom, Enlargement of boxed area in upper illustration shows asbestos body (arrow) (HE). # 3f St. a i 5 ! -:<s s' ;V: i C is , V* j 564 Arch Pathol Lab Med--Vol 106. Oct 0, 1982 - -"2 4i-feV *r "/ Asbestos-Associated Disease--Craighead ot al H5; . t/:;~. y .$?? ' ^pT Ji'i . <!. | tff ` -.' .f"' /*"* >r K r ' '0- `v' ' ',;> -; . ^ ,. ;.- :j *};.' . r\ A, ... i e-r. . L /. ^v i * s : i jf 4*i/ Xxi ^ Vv tV' }p L /.Ct * ; ' '5 . *V'4*-'- -- O .. >: / -~- ' *. . * i. '/' ./ x / "t 4 'i % 4'~ I ' ` -A I'i'S. \ '.;_ < r, ' *'' t / . / / V' /,-' > 1 P ..if .- j *. V'.._ . > i ...... :^V y,..^ *'V-4S V/ .a *-'<* <*>i-<c> .4 -'. ^. - ,-; *\ :? .\'v " ' ', ' . &e|,; y' -' '. 1 ,' 'v-rv''* . '! ._ ' J " ... '&,* ' ....... .''" 4 ~\ ' ' ' ` '4*1 '. .:; 1- ' ^ * `'CiT/. T`-*.5'^^-.^N- . -\ ,< ' - v. ;. V-v'-.VA ' Y '% ' i 2`jS * .1 J- V ~!r"`~V i. ='0^~it diiiiai -iTa*. i ^uAaaMaasJ Fig 20. -- Grade 2 lesions in lung of occupationally exposed worker with undocumented period of exposure to asbestos. Arrowheads indicate multinucleate giant cells and asbestos bodies (hematoxylin-eosin). > Arch Pathol Lab Med --Vol 106, Oct 6, 1982 Asbestos-Associated Disease--Craighead et al 565 *" p**ZS^SSS3Z^&^ _ffc X-. V* v:'-v. m^=%" %& <7/** _-r rk, s-;?v -j? <\+x .........., \1- **<>. ''-.'^>' MS . *--rr.r- Ji ' *\/VJL- *I. < JV^*t.*"<.-4*iiY'-^,-'.Z&STV5l?,*N.** <le -'' -^-*w:'J S>V eFr^^ *'>:r- ' '*, ^V'Kl - *''"'* V.>'S;). o- 7% O ?-V* / T^> *- v' ' i fr'*--- - *#? %<; >, -'s: v -v?-* .f;.-<;`-JsX^-''T^X-Kv-- ` '---u ^ / L1'^ ***rv : - V--- - .. > :>*-*'Y*. < ^ >7*/ .* , --/ f*v C W., 3Vy*.'. -> __ ^ . ......... . ^__ ^ - -r^.> f:'.'<S'C4' -*c**x*'~r ~ ' J .<& *,,*3 . -* c^>- r*.T^. *,va, ,&. ... ' 3' " V VV-'v/ +& Y> vr-jr &. * -i-st* ^V-'>=' '*'-'t Fig 21.--Lung parenchyma of case of grade 2 asbestosis revealing extension of fibrotic process to involve adjacent respiratory units. Note fibrosis in walls of both respiratory bronchioles and alveolar ducts (hematoxylin-eosin). Tissue in lower illustration was stained 1o demonstrate elastic fibers. '3^--^1 ^ S,X& i-'isr. * * v,'r' . # *< / . r ^ i * /V- tJ # . - .r ' ^ -*"v'Vr;- .'/--",.i --'*.,- * ::Jrl ) s :yy * h* -O' J't&r f . - r ; . ; -i .'/' ,r' T ;^5*> itffs A* S- > f *_.. : ' - f. / ,, ! ., ' . ' ::-v` ' - \f ? \ t:; ' ;, V. ,.-<. r:;. t - y/ryf'-fl ; " ^'- s - A y'J. V/.Ti-As >.. % f; a* ^ . *..<aS-'V;''74 : >'* Hi* ^C^v' ! ' - \*.'v2 >, * ** %*v * '" , ' ;' - !. V 1 ' '** '* * -*fv V. </.' -V-^ V-: r. * . n . ^ ; -.\\, " f r - -* " A- A'.* /* ; ?*/,/ {,; 'i ` " *1 r' f f \ i * . ; ,/ 'j ` * * " V ui* \k^, *\fy v- .'. :\\\i . ;.>, ',4*-'. : . f A ^ :! ;.; *r. ;. * /-x . v ^ . : r tp.--i-j r` * ^ s. '-.->.* -M'-'.P *: ' 3". ;' * I:'.\ ' s . i y rfiOPs'S' - f ;* ': -;=j to":.A./* <';/ ? . / -.th< ^<. -. - 4/ -' iv. O ? /r t r . .v< / * ,~r !--`', s*'i-' itu \>4 , -/r'-sr ":4f--/.*vis.v ..>/' 566 Arch Pathol Lab Wed --Vol 106, Oct 0, 1982 Asbestos-Associated Disease--Craigliead et at ) 5 ( Fig 22.--Lung tissue from 48-year-old man who worked for 26 years in manufacture of chrysotile-coaled textile products. This is grade 3 lesion. There is peribronchial and diffuse interstitial fibrosis (hematoxylin-eosin). I Arch Pathol Lab Mod--Voi 106, Oct 8, 1982 Asbestos-Associated Disease--Craighead et al 567 *5 1 i F;g 23. -- Upper left and right, Grades 2 to 3 lesions. Fibrosis is centered around respiratory bronchioles and involves adjacent respiratory units (hematoxylm-eostn (HE)). Lower left and lower right, Asbestos bodies (arrows) in boxed areas i of upper left and uppei right illustrations, respectively (HE). i i 568 Arch Pathol Lab Med-*-Vo! tOC, Oct 8, 1902 j Asbestos-Associated Disease--Craighead et at Fig 24.--Lung tissue of 48-year-o!d man who worked 26 years in manufacture of chrysotile-coated textiles. Lesion is grade 3, but elsewhere grade 4 lesions were found (hematoxylin-eosin). exhibit changes. Often the septa adja cent to the respiratory bronchioles are affected so that the fibrosis appears to radiate out from the immediate vicinity of the affected subunit. This process involves variable numbers of acini (Figs 19 through 21). With further progression of the process, increasing amounts of adja cent lung tissue are affected, seeming ly in a centrifugal fashion (Figs 22 through 21). In later stages, the parenchyma is diffusely involved, although the overall structural conti nuity of the lung is preserved (Fig 2-1). As the scarring process progresses, the parenchyma is obliterated by fibrous tissues, and fibrous-walled cysts form, particularly in the subpleural and paraseptal regions of the lower lobes (Fig 2b). At this stage in the evolution of the disease, peribron chial and perivascular fibrosis also is seen. Asbestos bodies often are observed in and around the lymphatic vessels adjacent to these structures. Asbestos bodies usually are obvious in the advanced stages of the disease (Fig 20). However, this is not invari ably the case, because fibers are cleared from the lungs and undergo dissolution and fragmentation with time. Thus, paradoxically, in some cases it may be difficult to demon strate asbestos bodies. In our experi ence, this is a rare occurrence. In the absence of asbestos bodies, t here is no satisfactory means to euaidish the diagnosis in histologic material. Although the demonstration of asbes tos fibers by the electron microscopic study of tissue digestates (appendix 1) provides evidence of exposure, ultrastructural technique cannot be used to establish definitively the etiologic role of asbestos in disease. Multinucleate giant cells are a vari able feature of the lesion; in some cases they are prominent, whereas in others they rarely are seen (Figs 4 and 27). Infiltrates of lymphocytes and plasma cells are found in scattered sites in the lungs of some patients. The importance of this inflammatory process is uncertain, although it could be related to the immunologic reac tions that have been described in some patients with asbestosis. Interstitial cellular infiltrates usually are not prominent in asbestosis. The finding of macrophage accumu lations in air spaces is of no diagnostic usefulness because this feature is commonly associated with the smok ing of cigarettes ami other forms of pulmonary injury. However, in occa sional cases of asbestosis, prominent numbers of macrophages in the air spaces and the histologic picture sug gest the pattern of desquamative interstitial pneumonitis (Fig 28). Cytoplasmic hyaline bodies have 570 Arch Pathol Lab Mod--Vol 106, Oct 8, 1982 Asbestos-Associated Disease--Craighead et al been described in the reactive cuboidal cells that line the air spaces of the lungs in asbestosis." Although this cellular lesion often is seen in asbestosis, it is not specific because similar changes occasionally are found in idiopathic fibrosis and other types of pulmonary disease" (Fig 29). No systematic studies of the tra cheobronchial tree and the pulmonary vasculature have been reported in documented cases of asbestosis. There seems to be no specific lesion attribut able to asbestos in the airways proxi mal to the terminal bronchioles. Scle rotic pulmonary vascular changes are evident in advanced cases of the dis ease. Although this may be secondary to the parenchymal lesion and the related hypoxemia, there is insuffi cient information available to estab lish the pathogenesis of the lesion. Asbestos workers often are exposed to respirable mineral dusts of a nonfibrous type. The pathologist should be alert to the possibility of exposure to a mixture of dusts, even in lungs that exhibit large numbers of asbestos bodies. Modern analytic techniques now make it possible to document the presence of such minerals as silica, talc, feldspar, and other aluminum silicates in lung tissue (Figs 30 and 31). Assay Techniques.--The demonstra tion of asbestos bodies in lung tissue with typical fibrotic lesions is required to establish the diagnosis pathologically. In our experience, these structures are not distributed uniformly in the tissue, even in advanced forms of the disease. Thus, multiple histologic sections must be studied. As previously noted, most of the asbestos fibers in the lung tissue are not coated and are usually rela tively small (ic, < 5 pm long). Uncoated short fibers of chrysotile cannot be visualized by bright-ficld, phase, or polarization light microsco py, and do not stain by the Prussian blue technique. Amphibole fibers are more easily identified in tissue by light microscopic techniques. Methods have been developed for the semiquantitative assessment of asbestos bodies and libers in digestates and incinerated samples of lung tissue. These arc research techniques, subject to considerable variation in inexperienced hands, and their value for the pathologist who is occasionally responsible for the autopsy of a sub ject with asbestosis is doubtful (ap pendix 1). Despite variations introduced by the techniques in use, the burden of pulmonary fibers in members of the general population is many times less than in asbestos workers. The data for persons with modest exposure often are less definitive. The demonstration and measurement of asbestos fibers in lung tissue by these approaches are not a substitute for careful histologic study and the demonstration of asbes tos bodies in association with the typ ical lesions of the disease. METHODS FOR PATHOLOGIC STUDY Postmortem Studies A complete examination of the respiratory tract, including the larynx and thoracic cavity, is critical to the thorough evaluation of the suspected case of asbestosis. When pleural lesions, eg, adhesions and plaques, are widespread or a mesothelioma is present, the thoracic contents should be removed en bloc. Often this is a difficult and time-consuming task, but it w'ill yield detailed information wdth regard to the severity and distri bution of the disease. Because pleural plaques often are found in the absence of significant pulmonary disease and a documented history of exposure to asbestos, exam ination of the parietal pleura is a mandatory routine procedure in good autopsy practice. Plaques often are overlooked and their importance as an indicator of exposure to asbestos is not universally appreciated. Unless other considerations dictate, one lung should be inflated through the major airway with buffered formalin; ideal ly, the perfusion pressure should be regulated to 25 cm H,0. Apparatuses have been fabricated for permanent installation in the morgue, or for rap id assembly for temporary use. After fixation for at least 2-1 hours at a sustained pressure, the lungs should be sliced for gros.-> assessment and blocks of tissue prepared for micro scopic study. We believe that the lungs can be evaluated adequately if tissue blocks are obtained from the apices of each lobe and, in addition, the diaphragmatic portions of the lower lobes. Samples also should be prepared to include (1) the visceral pleura and subpleural tissue, (2) deep parenchyma, and (3) portions of major bronchi, with adjacent paren chyma from each lobe. Examination of the lungs in this manner is neces sary because the lesions of asbestosis are distributed irregularly and tend to be most severe beneath the pleura and in the lower lobes. The blocks should be appropriately labeled so that the distribution of lesions can be assessed subsequently. Obviously, additional blocks of tissue can be selected to evaluate other grossly evident lesions at the discretion of the pathologist. The processing of tissue requires no special techniques. When asbestos bodies are common in lung tissue, they can be detected readily using 5to 6-gm-thick sections stained with hematoxylin-eosin. The standard iron stain helps identify these structures. Without this stain, one may overlook many asbestos bodies, even when using high magnification to screen the tissue. The use of thick sections (eg, 30 gm) has been recommended by several workers to help demonstrate asbestos bodies. Although this technique increases the mass of lung tissue examined, it introduces artifacts and complicates the detailed evaluation of the lung tissue. Thick-section exami nation is an ancillary technique that can be used when asbestos bodies are not identified in tissue sections of standard thickness. Lung.Biopsy The radiologic demonstration of diffuse fibrotic pleural and parenchy mal disease in a person with docu mented occupational exposure *o asbestos customarily is considered ample to establish the diagnosis of asbestosis. Lung biopsy rarely is justi fied in such patients unless the evi dence suggests other complicating, treatable conditions. Accordingly, bi opsy of the lung should be carried out when the nature of the pulmonary lesion is obscure. . Arch Pathol Lab Med--Vol 106, Oct 8, 1982 Asbestos-Associated Disease--Craighead ot at 571 i . L - -. .;/:;V--:_ ,,A, ,j *<?* 'vs * ,->.. ` r'j.^--'^y-X W .'ViV*--- V.;' H . >: r: I ; '--v '-'v ;. '' > W . T ' <"'* ' ; .-. >'4'.." -'--/- -v'-... V/ "^v, ' - ' V. - - ' . 'xir f /;. ,' _ V< - -V ''-'-li'LV `;t <- .'. v;iv, , .-^ '\; '.- -'- . '. '.:';:p - 'l'. . '. X )\ }- . " -^. -v - ' ; ' " ' ; . : : . . -" ':'* ' -- -.; . ' >V' /--. /-'Vi'. ' *,, \ l-.;* ' .> " ^..v. . '. >;.- =? - .*:$ - ... . ^ ' -" A; -. > . SMtri-r . . i. W. \:V 'V. *-r <Vj ''.. - ; 0 t /-' "'.-v->`--- . ' 1 \ '~`'ar.~- -*A- , ' .. V' ' .7 .. V ' 0- -r^ " ' Jl V' X.* ._. - _ _ :.v-sr-:c -`,-V V. . \ ^4-rs J*"' ' - ->'-A / irr^.c.2?-. . . --Y .X, H .-A. . ixy i. A- Fig 20. -- Grade 4 lesions. There is diffuse fibrosis and honeycomb formation. Lining of cyst spaces by bronchiolat epithehuni is variable feature (homaloxylin-cosin). 572 Arch Palliol Lab Mod--Vol 106, Oct 8, 1982 Asbestos-Associated Disoaso--Craighead et al Arch Pathol Lab Mod--Vol 106, Oct 8, 1982 Asbestos-Associated Disease--Craiyhead et at 573 ;: >-? V *^Vr% ' ^ M;- - J .- r- % -' ,,-' * *~J'- v-<: ,!r!, **\- ,Mi- .` *>* e** *.*,;:',, , : / *' '; ;,,;vy6^.,VJ V-. 5,V ** * *- - 'V -"'. \y- ,;- ;*?4^ y " -> "'-' **. f- , J- r T -.- . ' . .- JK_ ( s' 3 |. -' . " `-V* *-;. Y/ **. * ' { -r * ''-* 1 & *3* f -' :.-**. .. - -"y 1* \j- //**' *' ' vn** $ Fig 26.--Grade 3 asbestosis. Occasionally only "heads" of asbestos bodies are evident in histologic material. They are seen as golden spheres (arrowtieads) that approximate pulmonary macrophages in size (hemaloxylin-eosin). r iJ h>_ o*v ; - ' :. -: - - * ..a* ' y " ' "':> ',.'i ' '';>' v'-y f v; // ~ i r- u . n ' - .,"> . v- :v /w ,r . - ' , AV-J - -v-- -r. ' -s . .... f-. . . _ . . . V :f ; .": f {,. ' i' ' >* 0.1 4'cv -y.y vv = ^ P * ; 45,,- '. ->l 'yy ..; V ;d" ' ' #. - '-"A /; ,e-. . f , -'i" j V .;> J J - (` . . ' -A -V - i --.-q . * ^ t .. 4t ttf -1 * yy ',.j ;. . ... ... i-; .: Jii-i. i'J i_zz. -j ' !J-jiT ^.v... . x*.. V4 ,V' 574 Aich Pathol Lab Med--Vol 106, Oct 8, 1982 Asbestos-Associated Disease--Craighead ct at As noted, the lesions of asbestosis often have a spotty distribution when the disease is of either mild or moder ate severity. Adequate sampling of the lung parenchyma is critical and open lung biopsy is strongly recom mended. Transthoracic needle aspira tion and transbronchial biopsy usual ly do not yield sufficient tissue to permit a diagnosis in these cases. Samples of tissue collected by these techniques cannot be used to exclude the diagnosis of asbestosis. A biopsy procedure that yields inflated, well-fixed tissue has been described.1' Areas of relatively nor mal-appearing and fibrotic lung tis sue are identified and sampled to examine tissue at various stages in the evolution of the disease. Severely fibrotic tissue may show honeycomb ing and is of limited use for pathologic interpretation because this lesion is not specific. The tips of the lingula and middle lobe are to be avoided because they often are scarred nonspecifically. A wedge biopsy specimen should be preserved in an inflated state at the time of surgery. The lung is maintained in distention by the anesthesiologist while a wedge is iso lated with two sets of clamps. The surgeon cuts between the clamps and the wedge of tissue is given to the pathologist, who fixes it with the clamps in place to prevent collapse. A technique for either vascular or bron chial perfusion of biopsy specimens has been described." Often, it is useful when the specimen is collapsed. Sputum and Lavage Examination Asbestos bodies usually are demon strable in sputum specimens from persons chronically exposed to the mineral dust and can be present when changes in the chest roentgenogram are not evident.-'5''' The bodies are found in the mucin and in close asso ciation with clusters of alveolar mac rophages (Fig 32). The presence of asbestos bodies in the sputum cannot be. used as a diagnostic criterion of asbestosis. On the other hand, most asbestos workers with restrictive pul monary function and radiologic lesions yield sputum with asbestos bodies. Asbestos bodies in sputum strongly suggest past exposure to asbestos and reflect the presence of a significant asbestos burden in the lungs. Systematic studies of persons exposed to asbestos using the tech nique of bronchopulmonary lavage have not yet been reported. This approach undoubtedly would provide large numbers of pulmonary macro phages for examination, and thus might prove to be a sensitive indicator of exposure. CLINICAL FEATURES OF ASBESTOSIS Asbestosis is clinically indistin guishable from other forms of diffuse pulmonary fibrosis. The clinical pic ture ranges from dyspnea associated with exertion to respiratory failure accompanied by cardiac decompensa tion. In addition to exertional dysp nea, symptoms of early disease in clude a nonproductive cough, fatig ability, and a vague feeling of being "unwell." Obviously, these features are not specific because cardiovascu lar and neuromuscular disorders produce similar symptom complexes. As the fibrotic process progresses, shortness of breath becomes apparent at. lesser levels of physical activity and ultimately occurs at rest. Cough may be troublesome and difficult to con trol, although sputum production is unusual in the absence of coexisting chronic bronchitis. There are no specific symptoms and signs of early asbestosis, but endinspiratory crackles (rales) occasion ally are prominent at the lung bases. Crackles are found in a variety of conditions associated with relative underinflation of lung because they result from the rapid opening of peripheral lung units. They are not unique, nor are they more prevalent in asbestosis than in other conditions in which similar pathologic factors exist. Thus, the clinician should be cautious in attributing specificity to ..':ese sounds or even considering them a sensitive means for detecting early asbestosis. As the disease progresses, lung volume reduction leads to a pat tern of rapid, shallow breathing. Hy poxemia develops when significant ventilation-perfusion discrepancies develop and gas exchange fails. Occa sionally it is clinically manifested as cyanosis, initially on exercise, and in the later stages at rest. As with other forms of advanced respiratory dis eases, clubbing of the digits can be present, but it is neither specific nor common in asbestosis. Effusions of serous fluid occasional ly occur in persons with pleural and parenchymal disease due to asbestos. The prevalence seems to be dosage related. Usually, the fluid accumula tions are relatively small and often they are asymptomatic (Fig 7). PHYSIOLOGIC FEATURES OF ASBESTOSIS (PULMONARY FUNCTION) In considering the physiologic con sequences of exposure to asbestos dust, one must distinguish between alterations encountered early in the course of the respiratory disease (such as might be found in a working popu lation) and the impaired lung function associated with well-established, ad vanced asbestosis (which is accompa nied by the characteristic clinical and roentgenographic findings). In the former case, spirometrically deter mined lung volumes and expiratory flow rates are affected adversely in a dose-related manner, at least in the experience of some investigators. This effect can precede roentgenographic evidence of the disease. Similarly, total lung capacity and exercise venti lation become abnormal in relation to the level of exposure to dust. Thus, vital capacity and forced expiratory volumes and flows reflect cumulative exposure. In some cases, diffusing capacities are a sensitive measure of parenchymal disease. That maximum expiratory flow limitation occurs in the early stages of asbestosis is not surprising in view of the pathologic demonstration of bronchiolar lesions in the initial stages of the disease. However, it should not be assumed that clinically significant functional impairment necessarily accompanies early morphologic lesions. Determina tions of pulmonary diffusing capacity can be used to separate persons with well-established asbestosis from those who do not have roentgenographically detectable alterations. Although the diffusing capacity is abnormal in advanced disease, it is not a particularly sensitive means for Arch Pnlhol Lab Mod--Vol 106, Ocl 8, 1982 Asbesl03-AsB0cialed Disease--Craighead el al 575 0 Si'j* 4 1;: A I ' .i s % : ' i i ; V& mmr^ os- .iT#>V^v Vi |v. ;r -v 4 .5, .5 -: >.. Jf-f &at **-:--: -3 ,?y .-/ ***.. . v*- -;/ -,,t '=/ ..-' a , ., - V'- : i / V.. .i-- -/ y ,? , b' a . :..`:,y. *<u * a?* f /- * *s J > -V, <?*2 " V?S2 *$? "y3 \ Kyr-' ^ ;VrV^* : ,. * -... :.. *"< * * . ,v.y <=Jr ?r , 74 y j-^ vr? .fr' . !<* s. - .. -i. - -. JZ& *k Fig 27.--Grade 3 asbestosis exhibiting prom inent numbers of multinucleate cells, many of which contain asbestos fibers and bodies (arrows). Evidence of phagocytosed fibers is variable feature. Multinucleate cells are partic ularly prominent in case material from Finland, where anthophyllite exposure occurs (hema- toxylin-eosin). . u j j i I i 576 Arch Pathol Lab Med -- Vol 106, Oct 8, 1982 Asbestos-Associated Disease--Craighead et at y y. t s. `'s-" I* m?5rG'*7&'m-tersr 7" VV *- -7-V; **'-?! .'' *7*. v'l *U! * JSfP'. 'W % 4-4 t. -*.A ,; ;7-','4.- 4 7'7^ .ffv 4 ,; V *> tv t J&3*' * V* rT. ".''*" \ ' s v^r>. * v - - - - f .. . v-v. -;. y.; .s7 *V~;,` - &*. '. UP ' * 7.7 * .i *-.{ V t-+~. f ,* v 7 J <-; ^ - - '*- "` i Vi r*7 j- / .^/V^ 7 <* > ^v * ;VK/'V*? eS^-'V: " **..*' *{ *w< * k :-; ;.. : 7 .* 7 .*> V -, . J- ^ :- r,, j pv 1 ' ,7 / -4 7 7V . * ^ t.' . *^V % c * '4'.r^ - -.'-r t '. *,, ... . .-4 'r'Vv y& 1'K :V . : * ' \ r '' ' '' V->> 1. '4A7 '1- -:> ' " * ' ,":S '* . .? J /h%: ' . ' " ,17V- * V vV'\^ - ... . . -* J- '- ** ., . <?! . ' v -$ "*- .'.. * . -. <' f? > ; ' t. .4 ^^ - .-< '. J. ' '/ !7 < ; .;,> ' . "V-i .''. . ^v` t*. -' j - - ) - - ?. ; - - ^v rf vv' * %;'?'* $* i _? . ^ v.-^ * -* ,*T - '-4>i ' * t ' .... .-'-'; . ';. VM.1 ' V* l 'iy? *' -'V V "` I: -4 ^ 4j3eZ ' . . S*'*'.v.f.r.*-*'.*<-^-4- r*. |rt i, - -.. _ .. . . . ' *** **.lj<! ,,*j r J* ` -; - -f o. .-, **.. .. --- :T.. - yd -V* C" 1 &.rvxi`?7..v- -t--ttiVb.'*t'4....#;7k;i-'t->.V^Vsv.J.v*0>.wi-4ei' 1,7 c, , r:-^7 [ v 74, V- 'v *$*&-' : <r ;t;v4 <3 4 C ' .:.. & Arch I'athol Lab Med--Vol IOC, Ocl 8. 1982 1 r~^ ;> 7. I-1 \ cJ ii V, " : ".' .'.7--.' f.f;; .;.; ,j S'^' * w' ^ ' ' '"# ..\7Lf}ji:a Asbestos-Associated Disease--Craighead el al 577 ' V - -Of ?'; 01 &* %> '? :v ..--' '--'*^sA> > ; c.;-- \ ~U L- 578 Arch Pathol Lab Mod --Vol 106, Oct 8, 1982 k*F> * v%l V . , e* V O ' -: ,5 .H - ' ~J Asbestos-Associated Disease--Craighead et al  ':v 'Tj Fig 28.--Prominent accumulations of macrophages in air spaces of lungs of patient exhibiting grade 3 fibrosis. This nonspecific cellular response occasionally resembles tissue reaction observed in desquamative interstitial pneumonitis. Arrows indicate subtle asbestos bodies. Arch Pathol Lab Med--Vol 106. Oct 8. 1982 Asbestos-Associated Disease--Craighead et at 579 /! ' "j 1 - - 580 Arch Paihol Lab Med--Vol 106, Ocl 8, 1902 Aaboslos-Associatc-d Diaear.e --Craighead ot al Fig 20. -- Network of flocculent hyaline material in alveolar epithelial cell in lung of patieni with asbesiosis (upper left). This enlarged cell has large nucleus with prominent nucleolus (hematoxylin-eusin, original magnification X 1,250). Upper right and lower left, Electron micrographs demonstrate dense core of inclusion. Lower right, Meshwork of fibrils comprise it (uranyl acetate-lead citrate) (Irom Kuhn and Kuo3'). Arch Pathol Lab Med --Vol 106, Oct 8, 1902 Asbestos-Associated Disease--Craighead et al 1-^ S> 'v &'/ : v ;-i J- .***''- ^rv--:-** <*V . rKYl'*.*.. l B r * &*.:.+A-*f I Il,^Ji -r.' t '' .-V^. '<r c"> ifcr :' -.7- s , ,> J >6 \ .`T7ff--* N*% TS -:i * 4.7 V v < Jt~* a *^ ** -' / . .c? V: ?>C* / g r v- C <s-,, ?. ' ' - *"<? ''D- b**YYi ' ' ir f ->. ...->' *> V- ^-- T* ^-*5',- ~ . > {*.*-. V* F-. ") U \ " -'3 N % ^ >#:" *; ^ Vv'y/ /.:VV > - :...*' ' \i -.-I ,*' s *,i;* * c:?7<''#'";..iivPf. ! '. 'Y* r `^V **''4/ r;' V^` / .. ..- .......77U*< .j gwi.af -"' ,/=? : , " . . V f .* .> `,'y /*' _ 'is *- ts/ i? d ^ - 7./ 1 & xt +? pk ^ . - . *9 & *-v / ; ... . -: > ' ^ ., wr Il3` '* .." L-';' '' ?* t) f ' 'lf3U" O f- W - i * i? > ,, <^ *T- , '..." >>/. ^.'* J.-Mijii-i.#v' ..jj'-/-yv' ? 3- - -;Xv. ^ -'r. > <, ` - - \ v* *' ." * ...jp H -- ^ - . ,> : . -''3. V. gf,j 0 * t. i .S. , Fig 30.--Lung tissue of patient with advanced asbestosis and evidence of exposure to dusts other than asbestos. Analysis of similar dust maculae in this lung showed crystalline aluminum silicates and anthracotic pigment (hematoxylin-eosin). ` . .. ' *-f <^5>t V* -.P, `"'J -1 -* . "i \t .-...Pd ' ' ... < d A ^ ' (TJT > ' vr Fig 31.--Confluent nodular lesion consistent with silicosis superimposed on lesions of asbestosis. Note deposits of anlhracotic pigment and prominent accumulations of mononuclear cells in and around silicotic nodules (hematoxylin-eosin). I r t' r At * * r. k * \ > v i\ 1 .V y i i l V_- l ..-.J .7.,. '. I , .' -J.- i r\ 'X * i. '' i: ?; < ) Fig 32.--Papanicolaou-stained sputum smears showing asbestos bodies associated with pulmonary macrophages. Arrow indicates uncoated portion of asbestos fiber (X450). These specimens were obtained from worker in amosite asbestos pipe insulation plant (from McLarty et al!0 and from Rogli et al's). ^ .. ^ U.*>* * --X V f* -} Arch Pathol Lab Med--Vol 106, Oct 8. 1982 f\Jor Asbestos-Associated Diseaso--Craighead et al I t ' Fig 33. --Roentgenogram of 68-year-old man who worked for 35 years as insulator. Ho had no history of cigarelfe use. His overall health was considered good until about two years before death, when exertional dyspnea was noted. About five months before death, he was forced to retire because of severe respiratory distress. Radiologic examination disclosed evidence of asbestosis and right pleural effusion. Note linear streaks and small irregular opacities in lungs, and "shaggy" heart border. At autopsy, there was grade 3 asbestosis (from Pathology Annual, 1980:2:77-104). demonstrating a dose relationship in a working population. However, in one study, abnormal difTusing capacity correlated with the severity of the histologic lesion demonstrated in the lung biopsy specimen.37 ROENTGENOGRAPHIC FEATURES OF ASBESTOSIS The earliest roentgenographic evi dence of pulmonary fibrosis due to asbestos inhalation appears in the lov.vr lung zones. Typically, small, lin ear, and irregular op'.cbies are lo cated symmetrically at lie lung bases. The densities usually have a "fine" cnaracter, but at times are somewhat coarse. The middle and upper lung zones become involved as the fibrotic process progresses. Subsequently, the densities increase in number and size, bventually, these opacities may oblit erate the definition of normal adja cent structures, resulting in the so- called shaggy heart and. an indistinct diaphragm (Fig 33). Massive conglom erate and coalesccnt densities are uncommon, blit the changes of the honeycomb lung are evident in late stages of the disease. Diffusely dis tributed small, rounded opacities gen erally are not expected as the result of exposure to "pure" asbestos, but are encountered when there also has been exposure to silica dust. Although roentgenographic features of asbestosis are nonspecific and can bo seen in any form of diffuse pulmonary fibro sis, the presence of diaphragmatic plaques and pleural thickening associ ated with fibrosis in the lower lobes strongly suggests the diagnosis. Often, but not invariably, asbestosis is a progressive disease, cither with or without continued .exposure. Using paired serial chest roentgenograms, evidence of progression was evaluated in workers in the asbestos-building products industry." The appearance of the small, irregular opacities that are believed to indicate pulmonary fibrosis was related to cumulative exposure to dust. Progression of pleu ral thickening and plaques, however, was influenced by the duration of exposure and the elapsed time from initial exposure, but not to cumulative dosage. MESOTHELIOMA Malignant mesotheliomas of the pleura and peritoneum either are exceptionally rare or never occur in persons not exposed to asbestos. It is now apparent that asbestos plays an important role in the pathogenesis of these tumors. Although the relative risk is difficult to determine (as the prevalence in the general population is negligible), mesothelioma accounts for about 10% of all deaths in some occupational groups. In a recent study of insulation workers, 112 peritoneal and 63 pleural mesotheliomas were identified among 2,271 deaths, a prev alence of 8%.39 By comparison, there were 486 deaths (21%) from lung can cer in the same group. Mesothelioma typically develops many years after the initial exposure; in most series, no increase in frequency is observed before 25 to 30 years have passed, and cases with a latency of more than 40 years have been reported."0 The protracted interval between exposure and the development of the tumor is important with regard to our expectations for the future occurrence of mesotheliomas. Because a substan tial number of shipyard and industri al workers were exposed to asbestos during World War II, an increasing incidence can be expected during the remaining decades of this century. A rigidly defined relationship to dosage has not been established for mesothelioma, although most cases are found among members of occupa tional groups that experienced heavy exposure. The neoplasm also occurs in persons with relatively limited expo sure. The tumor is found in persons with asbestosis and pleural plaques, but many cases were associated with neither. Malignant mesothelioma in its fully developed form is a diffuse tumor that 504 Arch Pathol Lab Med--Vol 106. Oct 8, 1982 Asbestos-Associated Disease--Craighead et al K - r*v V Ix ?: s 0 1 fc- /. I, f' . , S/f 't S'f. 1,Vi '4 Fig 34.--Postmortem specimen of mesotheli oma from 64-year-old journeyman who denied exposure to asbestos. Top, Left and right lungs are encased by tumor. Note extension of neoplastic tissue along sopiae and fissures, and into lung parenchyma. Bottom, Mediasti nal structures are encased by tumor. Tnis patient died of cardiae arrest, presumably consequent to external compression of heait. / ,c i Ai * '{<&<&?: ` ^v':'/'V;-^''Vv t'ZL-z&s- Teg 35. -- Histologic features of mesotheliomas. Upper left, Typical epithelial mesothelioma ol tubulopapillnry type (hematoxylm-eosin [HE), X50). Upper right, Tumor shown at upper loll illustrating spaces lined by cuboidal cells, and formation ol papillary structures within spaces (HE, X5C0). Lower loft. Typical sarcomatous mesothelioma (HE, X50). Lower right. Tumor patlern shown at lower left illustrating spindled cells forming storiforrn pattern. This pattern ol growth is common in sarcomatous mesotheliomas (HE, X500). . 58G Arch Pathol Lab Med--Vol 106, Oct 8, 1982 Asbestos-Associated Disease--Craighead ot al t i Arch Pathol Lab Med--Vol 106, Oct B. 1982 Fig 36.--Roentgenograms of 76-year-old fur nace worker with long history of cigarette use. He was hospitalized terminally and died with out treatment. Radiologic studies demon strated immobile right diaphragm and pleural effusion (top). Circumscribed cavitary lesion was found by tomography. Postmortem exarn ination showed pulmonary asbestosis (grade 1). noncalcified plaques in pleura, and welldillerentiated cavitary squamous cell carcino ma. Right pleural cavity contained 300 mL of , iiuid (bottom), i J 3 1i* - 4*i j\ Asbestos-Associated Disease--Craighead et al 587 /viv >!**" J` ' tends to surround one and sometimes both lungs and mediastinal structures (Fig 34). In the abdomen, it encases the bowel, often leading to obstruc tion. Less commonly, pleural mesothe lioma is a discrete mass that invades lung. In early cases, it may be difficult to distinguish the tumor grossly from pleural plaques (eg, at the time of exploratory thoracotomy). For this reason, the pathologist is advised to obtain multiple samples of the lesion when biopsy is performed for the pur poses of diagnosis. Although plaques are found in persons with mesothelio ma, no evidence suggests that meso theliomas arc derived from plaques. Asbestos bodies are customarily not found in the tumor tissue. Mesothelioma may penetrate the diaphragm and invade viscera. Metastases once were considered to be an indicator that a tumor was not a mesothelioma. However, metastases, particularly within lung and local lymph nodes, are not rare. A charac teristic feature of mesothelioma is the tendency to grow along needle tracts and through surgical incisions. The microscopic patterns of meso thelioma are numerous and not clear ly defined, particularly in the case of the sarcomatous tumors (Fig 35). Mesotheliomas are grouped generally into epithelial, sarcomatous, and mixed types. Detailed descriptions of the gross and microscopic features of malignant mesotheliomas arc beyond the scope of this monograph and are published elsewhere.'0 Assistance in the diagnosis of mesothelioma is pro vided by panels of pathologists in both the United States and Canada (appen dix 2). The most crucial step in the diagno sis of malignant mesothelioma is to distinguish these tumors from prima ry bronchogenic adenocarcinomas, and from metastatic carcinoma and sarcoma. The autopsy finding of a typical gross appearance and the absence of another primary neoplasm are useful, hut care should he taken to rule out the occasional adenocarcino ma of lung that spreads over the pleura. Staining for mucins often is helpful in dilferentialing an adeno carcinoma from a mesothelioma. The carcinomas often contain 1'AS-posi- tive, diastase-resistant mucin within cytoplasmic droplets, whereas epithe lial cells in mesotheliomas may con tain glycogen that is removed by pre treatment with diastase. The epitheli al cells of the mesothelioma often contain droplets of hyaluronic acid that do not react with the Schiff reagent. This acid polysaccharide stains with either colloidal iron or Alcian blue; it can be removed by digestion with testicular hyaluroni-' dase. The demonstration of hyaluron ic acid by electrophoresis of tumor tissue also may be valuable.'1 Electron microscopy probably is of little help in establishing the diagno sis, because the ultrastructural fea tures of mesothelioma are similar to those of many carcinomas and sarco mas. Specifically, the epithelial types show the features of secretory cells, including the formation of lumina, whereas the sarcomatous forms are similar ultrastructurally to fibrosar comas and malignant fibrous histio cytomas. In recent immunoeytochemical studies, material resembling carcinoembryonic antigen was demon strated consistently in both bronchial adenocarcinomas and bronchioloal veolar carcinomas, but not in meso theliomas.'2 On the other hand, inter mediate fibers made up of keratin usually were found in the cells of mesotheliomas.'2 These new histochemical approaches may prove help ful to the pathologist. CARCINOMA OF THE LUNG The association of lung cancer with exposure to asbestos has been estab lished conclusively by epidemiologic studies conducted in a number of dif ferent population groups. Asbestosassociated lung cancer usually has a latency period in excess of 20 years, but generally occurs at an earlier age than does lung cancer in nonexposed persons'1 (Fig 3f>). All histologic types are represented, although in some series adenocarcinomas predomi nate." The latter neoplasms tend to he located in areas of fibrosis and thus are more common in t lie lower lobes'' "' (Fig 37). The relationship of the number of asbestos bodies and uncoated fibers in the lungs with the pathogenesis of the neoplasm has not been resolved. The greater the cumulative exposure to asbestos, the higher the risk of lung cancer developing."'" All types of as bestos seem to be carcinogenic; how ever, the relative carcinogenicity of each type of asbestos is not defined and remains a matter of controversy. Asbestos workers who smoke have an incidence of lung cancer higher than that of nonexposed smokers and of those who work with asbestos but do not smoke.50 In one study of insula tion workers, nonsmokers had an approximately fivefold increase in the risk of lung cancer compared with a nonsmoking population.51 This in crease in prevalence of neoplasms in nonsmoking asbestos workers has not been found in most studies. On the other hand, the risk of lung cancer was increased greater than 50-fold in asbestos workers who smoked, com pared with nonsmokers who lacked exposure to asbestos.50'51 These obser vations cannot be accounted for on the basis of an additive effect of two independently acting carcinogens, which suggests that asbestos and cig arette smoke act synergistieally.52 PATHOGENETIC IMPORTANCE OF DIFFERENT TYPES OF ASBESTOS Asbestos is not one, but a number of different, commercially available, fi brous minerals that have different physical and chemical properties. Fiber length and diameter, as well as the fiber aspect ratio (ie, ratio of length to diameter) are critical with regard to transport of particles in the respiratory tract. The diameter of the fiber is a major determinant of the depth of pcnetralion into the lung.51 All types of asbestos can produce pulmonary fibrosis, and no evidence suggests that the morphologic fea tures of the lesions differ. This state ment must he qualified, however, because much of the case material available for study represents severe, far-advanced disease in which subtle dilferences in the pathogenicity of the dust and the morphologic aspects of the lesion would be difficult to dis cern. In addition, documentation of exposure to only one type of mineral 508 Arch Pathol tab Med--Vol 106, Oct 8, 1982 Asbestos-Associated Disease -- Craicjhead et ul ' Fig 37.--Top, Chest roentgenogram of 44 : year-old man whose chest film had been , normal three years earlier. Solitary right lower ' lobe lesion is seen, with slight increase in | linear, reticular marking in both lower lung { fields. No pleural changes are evident. Patient | had been exposed to asbestos for 17 years -; while working as boiler tender in Navy. He had ..} smoked two to three packs of cigarettes per | day for 30 years. Ventilation perfusion scan ; was normal, except for evidence of chronic obstructive pulmonary disease. Patient under- f went right lower iobectomy. Bottom, Right !I ; lower-lobe lesion proved to be 'adenocarci.noma, presumable arising in scar. Asbestotic *' lesions of varying severity were located clse- - where in specimen (see Figs 21 and 28). This II L I finding emphasizes importance ot thorough ; evaluation ot pulmonary tissue in persons J exposed to asbestos (hematoxylin-oosin). *<; ? ; ^: . v Arch Pathol Lab Wed--Vol 106, Oct 8. 1982 fc- M Asbestos-Associated Disease Craighead et al 589 Fig 38.--Recommended sites for preparation of tissue sections in left lung. Right lung is sampled in similar manner, although specimens are also prepared from middle lobe. can rarely be established with cer tainty'. From a practical point of view, it must be emphasized that many commercial asbestos products contain mixtures of more than one type of fiber. Because exposure may occur over a worker's lifetime, and because industrial processes change and are often poorly documented, it is rare when only one type of fiber is involved in (he disease. The physical characteristics of the fiber seem to be of paramount impor tance in mesothelioma. When either naturally occurring or man-made fibers are introduced experimentally into the pleural cavity of rats, those fibers longer than S pm and of a diameter less than 1.5 pm produce mesotheliomas with a high degree of efficiency.5 Because the chemical com position and origin of the material does not seem to be critically impor tant, the aspect ratio of the asbestos particles must be a factor that deter mines the pathobiologic properties of a mineral species. A possible confir mation of this notion has come from recent reports of mesothelioma that developed in persons who lived in areas of Turkey where the soil and building materials contain large amounts of a zeolite, fiber, crionitc.22 These fibers are similar in size and shape t.o many types of amphibole asbestos, but they arc dissimilar structurally and chemically. The importance of different types of asbestos in the pathogenesis of meso thelioma in man has been explored in numerous epidemiologic studies. Anthophyllite failed to produce the tumor, even though pleural plaques occur commonly in populations ex posed to this mineral.'1 The apparent benignity of anthophyilite is believed to reflect the relatively broad diame ter of its fiber. Mesotheliomas were strongly associated with crocidolite. However, the physical characteristics of this mineral type (and, thus, the prevalence of mesothelioma) differ from one mine to another. Crocidolite from the Transvaal region of South Africa is considerably less likely to induce the tumor than are the fibers extracted in the Northwest Cape region.1 The importance of chrysotile in the etiology of mesothelioma is less clear, and the question remains a sub ject of controversy. HOST FACTORS IN ASBESTOSIS Although persons may differ in their susceptibility to the pathologic effects of asbestos, the possibility is not well documented. A genetic pre disposition to asbcslosis has been sug gested. In two studies, HLA-B27 was demonstrated more commonly in per sons with asbestosis than in members of the genera] population.55-54 Subse quent investigations failed to confirm this observation.51 Additional detailed studies using carefully selected groups of patients and controls will be required before possible association between a specific histocompatibility antigen and disease can be estab lished. Immunologic aberrations have been documented in blood sera of workers exposed to asbestos, especially those with roentgenographic evidence of disease. Considerable clinical evidence suggest that disturbances of immunoregulation may be operative in asbes tosis. For example, impaired expres sion of cell-mediated immunity has been demonstrated.586" Persons with asbestosis also exhibit altered humor al immune responsiveness, as exem plified by the occurrence of non organ-specific autoantibodies (rheu matoid and antinuclear factors) in relatively low concentrations in the blood serum.61'2 Increased concentra tions of scrum secretory immunoglob ulin are frequently noted. The patho genetic and clinical significance of these immunologic alterations have not been defined. Because similar aberrations are observed in other forms of interstitial lung disease, the finding lacks diagnostic significance. 590 Arch Palhol Lab Mod --Vol 106, Oct 8, 1982 Asbestos-Associated Diseaso--Craighead el al Appendix 1: Methods for the Assessment of Lung Fiber Concentrations Techniques have been developed to measure the relative numbers of uncoated and coated asbestos bodies in lung tissue. Most of the studies thus far reported have been performed on .autopsy specimens, although biopsy material also can be used. Although these techniques seem simple, labora tories differ in the number of fibers demonstrated in tissues from persons with seemingly comparable degrees of exposure. Experience and critical con trol of variables are required to obtain meaningful data. Unfortunately, as of this writing only a few laboratories in North America accept specimens of tissue for routine quantitative study. (Interested parties should write to the authors for further information.) The size (particularly length) and the relative number of coated and uncoated particles in the lung influ ence the sensitivity of the method. Because some asbestos fibers either break down into smaller subunits in tissue over time or are inhaled as small particles, the burden of dust in the lung may not bo reflected accu rately in light microscopic counts of fibers longer than 5 pm. In addition, the various types and sizes of fibers are probably cleared from the lung at different rates. These considerations provide some basis for the observation that particle counts by light microsco py do not correlate directly with the severity of the pulmonary parenchy mal disease. Both formalin-fixed and fresh lung tissue can be used in the techniques to be described. Contamination of the fixative and glassware with asbestos theoretically is a problem because many municipal water supplies con tain fibers in concentrations of 1 to 5 X 10VL. This may prove to be an important consideration in some labo ratories. Examination of histologic sections from patients wdth asbestosis has demonstrated the uneven distribution of asbestos bodies in the lungs. This finding has been confirmed by quanti tative studies.13-63 In addition to seem ingly random variation within a given area of parenchyma, short fibers tend to concentrate under the pleura. An uneven distribution of fibers seems to be greater in scarred lungs. These factors should be borne in mind when selecting tissue for analysis; for exam ple, variations in fiber distribution may make the use of small biopsy specimens unreliable. . Most published studies used one of the following two techniques. In one, fixed lung tissue is dissolved in sodi um hypochlorite (Chlorox). The resulting slurry is washed to remove organic debris and the material col lected on a membrane filter. The filter then is mounted on a slide and exam ined in a light microscope to demon strate fibers and asbestos bodies.'''1*6 Alternatively, pieces of the filter can be placed on electron microscope grids and the libers transferred to the grid by dissolving the filter in an appropri ate solvent. This method has the advantages of (1) gentle chemical digestion, (2) maximal concentration of the. sample (a feature of great importance in persons whose exposure has been minimal), and (3) the preser vation of a permanent sample for light and electron microscopy. These preparations are excellent for count ing asbestos bodies by light microsco py and uncoated fibers by electron microscopy, but are not suitable for counting uncoated fibers by light microscopy. Ashcroft and Heppleston described an alternate method.61 Tissue is dis solved in concentrated potassium hydroxide solution and an appropriate volume of the digestate is placed in a standard counting chamber to be examined by phase-contrast micros copy. Generally, this method neither produces as much concentration of the sample as the technique' just de scribed, nor is the counting chamber sample permanent. However, it does permit counting of long (> 5 pm) uncoated fibers by light microscopy. Determination of the number of asbestos bodies in lung tissue digestates provides a quick, although some what crude, estimate of the pulmo nary burden of fibers. When many bodies are found by this method, the result usually can be considered indic ative of substantial exposure to asbes tos (although the actual values found in patients with occupational expo sure vary from laboratory to laborato ry). When small numbers of bodies are identified, occupational exposure is less certain. As discussed in the section entitled "Asbestos Fib'v.;. in Tissue," refined analytical studies are required to document the types of minerals present in tissue. The num bers and types of asbestos fibers and bodies found in patients with various asbestos-related diseases are detailed in several reports.6*'''' Arch Pathol Lab Mod -- Vol 106. Oct 8, 1982 Asbestos-Associated Disease--Craighead et al 591 Appendix 2: Mesothelioma Panels Panels of experienced pathologists were established by the Union Inter nationale Contre Cancer (UICC) Working Group on Asbestos and Can cer in 1964 to provide a referral mech anism for the diagnosis of this rela tively rare tumor. The function of these panels was recently evaluated.70 In 1972, the US panel reviewed 168 cases. Of the lesions, 70% were con sidered either probable or definite mesotheliomas and 16% were thought to be possible mesotheliomas. In the remainder, the diagnosis of mesothe lioma was rejected. Members of the panel customarily concurred; in only 10 % of the cases was there substan tial disagreement. The present US panel organized under the auspices of the UICC is chaired by Charles Carrington, MD (Department of Pathology, Stanford University, Stanford, CA 94305). The Canadian panel is chaired by W. T. E. McCaughcy, MD (Clinical Studies Unit Bldg, 60 Ruskin Ave, Ottawa, Ontario, Canada KlY 4M9). Appendix 3: Pathologic Grading of Asbestosis The classification and grading of morphologic lesions has potential use in epidemiologic investigations and correlative studies of various aspects of disease. A useful system should be functional and acceptable to the pathologist who is expected to apply it consistently and in a reproducible fashion. In addition, the interobserver variability must be sufficiently low to permit the application of the system to studies that involve large numbers of cases and, thus, different patholo gists. In 1965, the Working Group on Asbestos and Cancer of the UICC rec ommended a classification schema for asbestosis that was based on both the assessment, of the severity of fibrosis and the extent of the changes in the lung as a whole.71 Shortly thereafter, Hinson et ah" proposed a grading pro tocol based on gross and microscopic criteria that could be applied to pul monary fibrosis in general, and to asbestosis specifically. To the best of our knowledge, this grading system lias not been applied systematically in large-scale studies. Grading protocols for the radiologic assessment of the pneumoconioses are at an advanced stage of development. Groups of radiologists working large ly under the auspices of the UICC and the International Labor Organization (ILO) proposed protocols that were generally accepted and have been applied in numerous epidemiologic studies. Moreover, it has been possible to train many radiologists in many parts of the world in the application of the system. The UICC and ILO grading systems were developed with silicosis and coal workers' pneumoco niosis in mind, and were not as appli cable to asbestosis as might be desired. Because of this problem, the standard UICC and ILO systems were amalgamated into the UICC-ILO 1971 system (updated in 19S0), which now is used universally for all pneumoco nioses, including asbestosis. To date, no attempt has been made to correlate the radiologic severity of disease with studies of pathologic material. A universally accepta1 :c grading system for pathologic evaluation of fihrotic iungs would have relevance to the clinical and radiologic investiga tion of asbestosis. In addition, it could provide a common language for com munication among pathologists and with other groups in the health fields. The histologic grading system pro posed here, is a further modification and extension of the protocol of Hin son et a!,7' although it is based exclu sively on microscopic criteria. It pre supposes systematic study of the lung tissue at the time of autopsy and requires the representative sampling of the lungs described earlier (see "Methods for Pathologic Study) (Fig 38). Although perfusion fixation would be highly desirable, the schema is applicable to lung tissue obtained under less-than-ideal conditions of fixation (ie, without perfusion). It also can be used in the evaluation of lung biopsy material, accepting the limita tions of the sample. Under this cir cumstance, evaluation of the patho logic material might appropriately be carried out in conjunction with radio logic studies to assess the distribution of disease. GRADING OF ASBESTOSIS The grading system is intended only for the semiquantitative estimation of the degree of asbestosis, and not as an aid to diagnosis. Therefore, the diag nosis of asbestosis should be estab lished before grading is attempted. Because librosis of the peribronchio lar and interstitial tissues is the key lesion in asbestosis, this grading sys tem is based solely on the evaluation 592 Arch Pathol Lab Med--Vol 106, Oct 8, 1982 Asbestos-Associated Disease--CraiQhend ct al 'Description of a hypothetical case. fRUL indicates right upper lobe; RML, right middle lobe; RLL, right lower lobe; LUL, left upper lobe; and LLL, left lower lobe. I Second slide missing. Grao'fe is total score divided by the number of lobes examined. Table 2. --Asbestosis Grading Schema" Score Magnitude of increase Score Magnitude of increase Grades From Products 12 3 4 2X 1.5X 1.3X Straight Grades 14- 24- 34- 44- 2X 1.5X 1.3X 6 1.5X 541.25X 8. 9 1.5X 641.2X 12 1.3X 741.1X * Quantitative comparison of incremental intervals between histologic grades, determined either as a product (severity X extent) or as an evaluation of sequential cnanges in disease severity. of fibrosis. Numbers of asbestos bodies, macrophages, and inflamma tory cells within the tissues and changes in the epithelium are not considered. Each histologic slide is evaluated from two perspectives. Initially, the severity of the lesion either in or surrounding the bronchioles is as sessed and the numeric score re corded. Then, the proportion of the bronchioles in the section that have any degree of involvement is evaluat ed (not just the proportion involved to the most severe degree). This grade, which represents extent of disease, is recorded as a letter score. To establish an overall grade for a tissue section, the distribution letter score is converted to a number and multiplied by the lesion score. To characterize the disease of a patient, the products of the scores of all the slides are added and the sum divided by the number of slides examined. This average indicates the degree of disease. Table 1 gives data from a hypothetic case to demonstrate the suggested format for recording data. ASBESTOS1S GRADING SCHEMA Severity Lesions associated with individual respiratory bronchioles are evaluated. The grade is based on the most severe lesion in the slide, not a visual average of the changes found in the various individual respiratory units, as fol lows. Grade 0: No fibrosis is associated with bronchioles. Grade 1: Fibrosis involves wall of at least one respiratory bronchiole with or without extension into the septa of the immediately adjacent layer of alveoli; no fibrosis is present in more distant alveoli. Grade 2: Fibrosis appears as in grade 1, plus involvement of alveolar ducts or two or more layers of adja cent alveoli; there still must be a zone of nonfibrotic alveolar septa between adjacent bronchioles. Grade 3: Fibrosis appears as in grade 2, but with coalescence of fibrotic change such that all alveoli between at least two adjacent bron chioles have thickened, fibrotic septa; some alveoli may be obliterated com pletely. Grade 4: Fibrosis appears as in grade 3, but with formation of new spaces of a size larger than alveoli, ranging up to as much as 1 cm; this lesion has been termed honeycombing. Spaces may or may not be lined by epithelium. Extent The proportion of the respiratory bronchioles involved by the disease process is assessed. The grades per tain to the relative numbers of bron chioles in the slide involved by any degree of fibrosis--not just the num bers involved to the maximum degree as recorded under severity--as fol lows. Grade A (1) Only occasional bron chioles are involved--most show no lesion. Grade B (2): More than occasional involvement is seen, but less than half of all bronchioles are involved. ' Grade C (3): More than half of all bronchioles are involved. Multiplication of the scores for severity and extent requires explana tion because it might be argued that a product score is inappropriate. For example, a product score of 4 obtained on a tissue with grade -1 severity with an extent score of A has implications with regard to disease that differ from a tissue with grade .2 severity and grade B extent. Although theoret ically the validity of this argument cannot be refuted, in the evaluation of our case material the severity and extent of the lesions were not found to vary independently, hut seemed to be interdependent. Thus, severe lesions are usually found when disease is widespread in (he tissue. The second concept, which argues for the multiplication of the severity and extent scores, relates to the range of possible scores that result (TaKe 2). Thus, the products of grades (i X 1, 1X2, 2X3, etc) produce a series of eight possible scores (1, 2, 3, 4, 6, 8, 9, and 12). As shown in Table 2, the magnitude of increment between products remains relatively constant throughout the range. This is not the case when a numeric grading system (1+ through 7+) is used (Table 2). Arch Pathol Lab Med--Vol 106, Oct 0, 1982 Asbeslof.-Associaled Disease--Craighead et al 593 Appendix 4: Preliminary Evaluation of We evaluated the proposed grading system using pathologic material in the files of the Armed Forces Insti tute of Pathology, Washington, DC. In this study, we were interested only in determining the applicability of the system to a series of cases of asbestosis of varying degrees of severity and assessing preliminarily both the in terobserver and intraobserver varia tions in scoring. Two tissue sections from each of 20 representative cases were chosen by John Rust, MD, of the Pulmonary Disease Section, Armed Forces Institute of Pathology. Nine pathologists (ourselves and Dr Rust) then reviewed the material micro scopically. All 40 slides were exam ined once by each participant, and 20 slides were evaluated twice, once in the morning and again in the after noon. The results of an analysis of intraobserver variability in the evalu ation of the 20 histologic sections are given in Table 3. As can be seen, observers scored slides differently on the two occasions, but the average algebraic difference between the morning and afternoon readings for the group as a whole was --0.3. Disre garding temporal considerations, the variability was 0.47, which is 4% of the total possible range of 0 to 12. When the mean of all differences in the scoring of individual slides is eval uated, the average difference was only Grading Schema 1.3 (ie, 10.8% of the total possible range). This latter analysis magnified differences maximally. We concluded that the degree of intraobserver vari ability was acceptable. The average scores of the nine pathologists on all 40 slides for morn ing and afternoon readings are given in Table 4. As might be expected, differences between observers were found, but the SD of the mean score (6.5) for all readings was 0.8. This SD is 12% of the mean value, or 7% of the total range. The coefficient of correlation be tween the scores of every possible pair of pathologists on individual slides was evaluated. The average of all coef ficients was .69 (P < .0001). Parametric statistical methods were used in these evaluations. It could be argued that this approach is inappropriate as the `'scores" are not a true measure of disease. According ly, a nonparametric analysis was also carried out.7' In this evaluation, the data on the cases were arranged in order of increasing severity as deter mined by each pathologist, and the agreement in ranking by the observ ers was analyzed. When the coeffi cient of concordance was determined, the overall concordance among the pathologists as a group was +.71 (P = .001). Additional analyses were per formed to address objections to a grading system based on the multipli cation of the extent and severity scores. The measures of severity and extent were evaluated individually by determining the agreement between all possible pairs of pathologists. With both measures, it was found that the scores of the pairs of observations were identical for more than half of the slides reviewed. In nine of ten slides, the maximum discrepancy was only one grade. The preliminary evaluation was carried out by pathologists who have a special interest in pulmonary disease. The members of the group had worked together to evaluate pathologic mate rial from almost 200 cases of asbestosis. It would be unrealistic to suggest that they are unbiased subjects for a test of the proposed grading system. Nonetheless, it is evident that agree ment among members of the group with regard to evaluating both severi ty and.extent of the lesions was excel lent. At this juncture, further evalua tions of the grading system by pathol ogists with different backgrounds seems appropriate. In addition, studies should be undertaken to deter mine whether the pathologic and radiologic grades correlate. Table 3 -- Evaluation of Asbestosis Grading Schema' Pathologist 1 2 3 4 t 6 7 6 9 Average Mean Algebraic Difference 0.5 -0.7 -0.3 0.1 -0.6 -0.1 -0.3 -0.7 -0.7 -0.3T Mean Random Ditference 0.5 0.7 0.3 0.1 0.6 0.1 0.3 0.7 0.7 0.47f Mean Absolute Difference 1.6 1.0 1.6 1.1 1.8 1.3 1.0 1.3 1.3 1 3 Intraobsetver variation in the evaluation of disease grade for 20 histologic sections examined by nine pathologists in the morning and afternoon of the same day. tAfternoon grades were sligtitiy lower than the morning grades. Score is 4% ol total range (0 to 12). 53coro n to 8'*j ol lotal range (0 to 12). Table 4.--A sbestosis Grading Test' Pathologist 1 2 3 4 5 6 7 a 9 Average Slide A (AM) 7.4 7.1 8.0 6.6 6.8 5.4 5.6 6.8 6.1 6.6 Slide A (pm) 7.8 6.4 7.7 6.7 5.9 5.4 5.3 6.1 5.5 6.3 Slide B (AM) 7.9 7.1 7.0 7.5 7.0 5.6 4.4 6.3 5.7 65 Average 7.7 6.9 7.6 6.9 6.5 5.5 5.1 6.4 5.8 6.5 0.8t , * Mean of the grades determined on 20 histologic sections by nine pathologists in the morning and afternoon of the same day. The SD is 12% of the mean. 594 Arch Pathol Lab Med-Vol 106, Ocl 8, 1982 Asbestos-Associated Oiceaso--Craighead of al During its deliberations the Committee con sulted many scientists and pathologists in this country and abroad, including the following. Mike Attfield, MS, Appalachian Laboratory for Occupational Respiratory Disease, NIOSH, Mor-gantown, WVa Professor Jean Bignon, Laboratoire de Recherches sur les Affections Respiratoires et L'environnement, Cretcil, France Charles Carrington, MD, College of Medicine, Stanford University, Palo Alto, Calif Brian Corrin, MD, Bromptom Hospital, London John M. Dement, PhD, Appalachian Laboratory for Occupational Respiratory Diseases, NIOSH, Morgantown Edward A. Gaensler, MD, School of Medicine, Boston University S. Donald Greenberg, MD, College of Medicine, Baylor University, Houston Russel! A. Harley, Jr, MD, School of Medicine, University of South Carolina, Charleston Paul Kotin, MD, Manville Corp, Denver Marvin Kuschner, MD, College of Medicine, State University of New York at Stony Brook James A. Merchant, MD, Appalachian Laborato ry fur Occupational Respiratory Diseases, NIOSH, Morgantown L. 0. Mourman, MD, University Hospital, Turku, Finland Roger Seal, MD, IJandough Hospital, Penarth, Glamorgan, United Kingdom Irving J. Selikoff, MD, Environmental Science Laboratory, Mt Sinai School of Medicine, New York Professor P. Sebasticn, Direction des Affaires Sanitaires et Sodales, Paris Geoffrey Taylor, MD, Wausau Medical Center, Wausau, Wise J. H. Tucker, Appalachian Laboratory for Occu pational Respiratory Disease, NIOSH, Morgan town . Frank Whitwell, MD, Broadgreen Hospital, Liv erpool, England Ralph Yodaiken, MD, NIOSH, Rockville, Md J. C. Wagner, MD, of the Medical Research Council, Pneumoconiosis Unit, Llandough Hospi tal, Penarth, Glamorgan, graciously provided case material for the study of the committee, including the material shown in Figs 15 and 31, and spent many hours discussing his views on the pathologic aspects of the disease with the chair man. We wish to acknowledge the interest and help of Peter C. Elmos. MD, director of the unit, and the members of his staff. Lisdotte Hochholcer, MD. director of the pul monary disease section, Armed Forces Institute of Pathology, and her associate John Rust, MD, made available selected material from the collec tion of the institute for the purpose of evaluating our grading system. COL Elgin Cowart, director of the institute, provided support and encourage ment. Jean R. Lemieux, MD, Directeur des Services Modicaux, and Louis-Gilles Cloutier, MD, of the Commission de la Santo et de la Security du Travail du Quebec, made available for study representative histologic sections of asbestosis from the pruvince ol (Quebec. Much of litis case material had been previously e ..mined and clas sified by the late Roland Guy MD, consultant pathologist to the commission for several decades. We sincerely acknowledge the privilege granted us to review this valued material. Yrjo Collan, MD, professor of pathology and director of pathology at the University of Kuo pio, Finland, graciously provided pathologic material from Finland for study, including the material shown in Fig 27. Jack Abramowitz, MD, of the Institute of Occupational Health, Johannesburg. Soutn Afri ca, made available to the committee material from his country for study. ' The collection of pulmonary diagnostic.materi al assembled by the late Averrill A. Liebow, MD, was used in this study. The Department of Pathology, University of California (San Diego), gave us access to the collection. The illustrations shown in Figs 2 and 3 were supplied by Craig WoodwoYth, University of Ver mont, Burlington. The illustrations shown in Fig 9 were supplied by L. O. Meurman, MD, University Hospital, Turku, Finland. . . The material shown in Figs 18. 22 and 24 was supplied by Russel! A. Harley, MD, University of South Carolina, Charleston. The material shown in Pig 19 was supplied by D. Groth, MD, NIOSH, Cincinnati. The material shown in the color figure on the inside back cover was supplied by Sergo Masse, MD, University of Sherbrooke, Quebec. ' 1. Zenker FA: Iron lung: Sclerosis pulmonuni. Dtsrh Arch Klin Med lt>G7;2:11C-120. 2. Cooke WE: Fibrosis of the lungs due to the inhalation of asbestos dust. Br Med J 1924;2:147- 150. 3. Lynch J.M, Smith \VA: Pulmonary asbesto- sis: Carcinoma of the lung in asbesto-silicosis. Am J Cancer 1935;21:56-G1. 4. Klemperer P, Rabin CB: Primary neoplasms of the pleura Arch Pathol Lab Med 1931;11:3S5- 412. 5. Weiss A: Pleurakrebs bei Lunge.nasbestose, in vivo murpholugisch Gesirheri. Medizinische 1953;3:93-100. C. Leiehner F: Primary mesothelial-cell tumor of the peritoneum in asbestosis. ArrA Geieerbc- pathul Geicerbehtju 1951;13:582-390. 7. Wagner JC, Sieggs CA, Marchand P: Diffuse pleura! mesothelioma and asbestos exposure in Northwestern Cape Province. Br J Jml Med 19GO;17:2GO- 27(1. 8. Stanton MF, Layard M, Tegcris A, et al: Carcinogenicity of fibrous glas^: Pleural re sponse in the rat in relation to fiber dimension. Natl Cancer In*! Monoyr 1977;58:587-(i03. 9. Dull R: Mortality from lung cancer in asbes tos workers. Br J Ind Med 1955;12:81-85. 10. SelikoR !J. ('burg A, llammund EC: Asbes tos exposure and neoplasia. JAMA 19G4;188:22- 2G. 11. KnterUne PE, Henderson V: T; (f asbes tos and respiratory cancer in the a-l.-aos indus try. Arch L'm'iron Health 1973,27.513-317. 12. Blot WK, Harrington JM, Toledo A, et al: Lung, cancer after employ, merit in shipyards after World War II. .V Haul J Med 1978;299:G20-G21. 13. Churg A, Warnock Ml,: Asbestos fibers in the general population. Am Roc Respir His 198t);122.Gfi9-G77. M. Gross P, Cralley LG. doTreville RFP: 'As bestos huilies. Their nonspecilicity. A/a hid !hj;t Assoc. J 19G7;28 511-552. ' References 15. Gaensler SA, Addington WW: Asbestos or ferruginous bodies. N Emjl J Med 1909; 280:4S8492. 1G. Churg AM, Warnock ML: Asbestos and other ferruginous bodies: Their formation and clinical significance. Am J Pathol 19S1;102:4-1745G. 17. Jaurand MC, Sebasticn P, Bignon J, et al: Leaehings of chrysotile asbestos in human lungs. Environ Res 1977;14:245-254. 18. Roberts GH: The pathology of parietal pleural plaques. J Clin Fedho! 1971,24:348-353. 19. Weiss W, Levin R, Goodman L: Pleural plaques and cigarette smoking in asbestos work ers. J Occup Med 1981;25:427-430. 20. Hurries PG, Mackenzie FAF, Sheers G, et al: Radiological survey of men exposed to asbes tos in naval dockyards. Br J hid Med 1972;29:271279. 21. Mcurman L: Astastos bodies and pleural plaques in a Finnish scries of autopsy cases. Acta Pathol Microbiol Scum/ 19GG, suppl 181, pp i107. 22. Baris YI, Artvinli M, Sahin AA: Environ mental mesothelioma in Turkey. Ann NY Acad Sci 1979:350:423-132. 23. Kleinerman J: Pathology standards for coal workers' pneumoconiosis. Arch Pathol Lab Mid 1979;103:374-429. 21. Spencer H: Puihohfjy of the Luuij. Phila delphia, WB Saundur.-v Co, \ul 1, p 592. 25. Yorwald AJ, Durban T.M, I'ratt PC: Hxporimental studies of asbestosis Arch ind Ihjy Occup Med 1951;3:1-43. 20. Wagner JC: Asbestosis in experimental animals Br J hid Mid 19G5;2fJ:l-12. 27. Wagner JG: The sequelae of exposure to asbestos dust. Ann NY Acad Sci 19G5;132G91G95. 28. Wagner JC, I terry G, Skidmore JW, et al: The ellYrts of the inhalation of asbestos in rats. Br J Cancer PJ7l;29.252-2fi9. 29. Bignon J, Atassi K, Jaurand MC, et al: tude cytologique et biochimique du liquide du lavage broneho-alveolairt- (LBA) clans la fibrose ptilmonaire idiopnthique et l'asbestse. Rev Fr Mul Respir 197S;G:353-35S. ' 30. Hunninghake GW. Kawanunii 0, Ferrans VJ, et al: Characterization of the inflammatory and immune elTects on cells in the lung paren chyma of patients with inlerstiLial lung disease. .4m Rev Respir Dis 1981 ;123:107-412. 31. Kuhn C, Kuo TT. Cytoplasmic hyaline in asbestosis. Arch Bathol Lab Med 1973,95:190 194. 32. Warnock ML, Press M, Churg A: Further observations on cytoplasmic hyaline in the lung. Hum Pathol 1980,1]:59-GG. 33. Gaensler EA, Carrington Cl>: Open biopsy for chronic diffuse infiltrative lung disease: Clin ical, roentgenographic and physiological correla tions in 502 patients. A>tu Thoruc Snry 1950:30:411-420. 31. Brody AR, Craighead JE: Preparation of. human lung biopsy specimens by perfusion fixa tion. Am Rev Respir Ois 1975,112:045-019. 35. Farley ML, Greenberg SD, Shuford EH, et al: Ferruginous bodies in sputa of former asbes tos workers. Acta Cntnl 1977:21:095-700. 30. McLarty JW, Greenberg SD, Hurst GA, et al: The clinical significance of ferruginous bodies in sputa. J Occup Med 195();22.92-9li. 37. Kpler G R, Samet JM, Gaensler HA: Diffus ing capacity in the diagnosis of asbestosis, abstracted. Chest 1978;7 1.550. 58. Jones RN, Diem JK, Glindineyer H, et al: Progression of asbestos radiographic abnormali ties: Relationships to estimates of dust exposure and annual decline in lung function, abstracted, in Biological I\(hcts of Mineral Fibres. Lyon, France, World Health Organization, 1980. 39. Selikolf 1J, Hammond EG, Seidmait H: Mortality experience of insiilal ion workers in the United Stales anti Canada, 1943-11*70. Ana .VI* Arch Pathol Lab Med--Vol 106, Oct 8, 1982 Asbestos-Associated Disease--Craighead el al 595 Acad Sci 1979;330:91-116. 40. Kannerstein M. ChnigJ, McOaughry WTK: Asbestos and mesothelioma: A review. Pathol An nu 1978:13:81-11^1. '11. Waxier B, Eiscnslein R, Baitifora H: Elec trophoresis of tissue glycosaminogivcan* as an aid in the diagnosis of mesotheliomas. Cunccr 1079;14:221-227. 42. Corson JM, Pinkus GS Mesothelioma: Pro file of keratin proteins and carcinoemhryomc antigen-immunoperoxidaso study of 20 eases and comparison with pulmonary adenocarcino ma. Am J Pathol l9S2;vKK):l>0()-000. 13. Martischnig KM. Nowall I)J, Barnsley \VC, ct al: Unsuspected exposure to asht-slos and bronchogenic carcinoma. Hr Med } 1977:3.71U. 4-1. Whitwell F. Newhouse ML, Bennett OB; A study of the histological edl types of Ion*; cancer in workers suffering from ashestosis in the United Kingdom. Br J hid Med 1974,31:898-403. 45. Kannerstein M, Churg J: Pathology' of car cinoma of the lung associated with asbestos exposure. Cancer 3972:30:11-21. 40. Sluis-Cremer GK: The relationship be tween asbestosis and bronchial cancer. Chest 1980;78:380-381. 47. Weill M, Hughes J, Waggenspack G: Influ ence of dose and fiber type on respiratory malig nancy risk in asbestos cement manufacturing. Am Rev Respir Dis 1979;! 3<>:815-354. 4S. McDonald JC, Lidded FDK, Gibbs CW. et al: Dust exposure and mortality in chrysmile mining, 1910-1975. Hr J led Med 1980;37:11-24. 49. Dement JM, Harris RL, Symons MJ, n al: Estimates of dose-response for respiratory can cer among chrysotile asbestos textile workers. Inhaled Part, in press. 50. Selikolt IJ, Hammond EG, Churg J: Asbes tos exposure, smoking and neoplasia. .MAM J9CS;201:100-112. 51. Hammond EC, Selikoff IJ, Seidman H: Asbesitu exposure, cigarette smoking and death rales. A?.m XY Acad Sci 1979:350:173-490. 52. Saraeci R: Asbestos and lung cancer An analysis el the epidemiological evidence on the asbestos smoking interaction. Ird J Cancer 1977:3*323-331. 53. Timbri-H V, Pomey F, Wagner JC: Charac teristics of respirable asbestos fibres, in Shapira HA (edj: Pneumoenniosis; Proceedings of the international Conference, John n ncsburp, 1969. Capetown, South Africa, Oxford University Press, 1970, pp 120-125. 54. Meurnian LO, Kiviluoto R, Hakama M: Mortality and morbidity among the working population of anthophyllitc asbestos miners in Finland: Br J hid Mid 1974;31:105-112. 55. Merchant JA, Klouda PT, Soutar CA, et al: The HL-A system in asbestos workers. Br Med J 1975:1:189-192. . 56. Matej M. Lange A, Sniolik R: HLA anti gens in asbestosis. Arch Immunol Ther Exp 1977;25:489-491. 57. Evans CC, Lcwisohn ME, Evans JM: Fre quency of HLA antigens in asbestos workers with and without fibrosis. Br Med J 1970;1;603- 605. 58. Kagan E, Solomon A, Cochrane JC, et al: immunological studies of patients with asbesto sis: I. Studies of cell mediated immunity. Clin Exp Immunol 1977;25:251-267. 59. Kagan E, Solomon A, Cochrane JC, et al: Immunological studies of patients with asbesto sis: II. Studies of circulating lymphoid Cell num bers and humoral immunity. Chn Exp Immunol 1977:28:268-275. 00. Kang KY, Sera Y, Okochi T, et al: Lympho cytes in asbestosis. \r Enpl J Med 1974;291:735- 745. ` Gl. Turner-Warwiek M, Parkes WR: Circulat ing rheumatoid and antinuclear factors in asbes tos workers. Hr Mud J 1970.3.492-193. G2. Stansfiekl D, Edge JR: Circulating rheu matoid factor and antinuclear antibodies in ship yard asbestos workers w ith pleural plaques. Br J Din Chest 1974,68:166-170. 63. Scbastien P, Fonriimare A, Gibnon J, et al: Topographic distribution of asbestos liber in human lung in relation to occupational and non- occupational exposure. Inhaled Part 1P77;4:435- 444. 64. Gross P, de'freville RTF, Crallcy 1-J, ct al: Pulmonary ferruginous bodies: Development in response to filanieniary dusts and a method of isolation and concentration. Arch Pathol Lah Med 1965:85:539-516. 65. Bignon J. Goni J, Gonnaud G, et ah Inci dence of pulmonary ferruginous bodies in France. Environ Res 1970;3:430-442. 66. Smith MJ, Naylor R A method for extract ing bodies from sputum and pulmonary tissue. Am J Clin Pathol 1972;58:250-254. 67. Ashcroft T, Heppleston G: The optical and electron microscope determination cf pulmonary asbestos fibre concentration and its relation to the human pathologic reaction. J Clin Pathol 1973;26:224-234. 68. Whitwell F, Scott J. Grimshaw M: Rela tionship between occupations and asbestos-fibre content of the lungs in patients with pleural mesothelioma, lung cancer, and other diseases. Thorax 1977;32:377-386. 69. Churg A, Wurnuck ML: Asbestos fibers in the general population. Am Rev Respir Dis 1980:122:669-678. 70. Kannerstein M. Churg J, McCaughey WTE: Functions of mesothelioma panel. Ann XY Acad Sci 1979;330:433-439. 71. Report and recommendations of the work ing group on asbestos and cancer. Br J hid Med 1965;22:165-171. 72. Hinson KFW, Otto H, Webster I, et al: Criteria for the diagnosis and grading of asbesto sis, in Bogovski F, led); Be>lo:/icai E/fects of Asbestos. Lyon, France, World Health Organiza tion, 1973. 73. Gibbons GD: Xon-ftarn metric Statistical Inference. Nevr York, McGraw-Hill Book Co, 1971. ' 74. Epler GR, Gnonsier EA: Prevalence of asbestos pleural elfusion in a working popula tion. JAMA 19S2;24 7:017-622. 75. Preger I, (ed): Asbestos-Related Disease. New York, Crime & Stratton Inc. 1978. 76. Roggli YL. Greenberg SD, MeLarty JW, et al: Comparison of sputum and lung asbestos body counts in former asbestos workers. Am Rev Respir Dis 19SO;122.J41-946. Bibliography Bccklake MR: Asbestos-related diseases of the lung ami-other organs: Their epidemiology and implications for clinical practice, in Murray JF led;: Lxnaj Disease: Stale of the Aii. Mew York, American Lung Association, 1977. Clifton RA:.\sbestos. US Dept of Interior pub lication TN'930, C57 5.72: 07 79-2572. Pittsburgh, Bureau of Mines, 1979. Levine RV: Asbestosis: An Information Resource, US Dept of Health, Education, and Welfare publication (NTH) 78-1981. Beihesda, Mil. National Institutes of Health, 1978. Selikoff J, Hammond EC leds): Health hazards of asbestos exposure. Auk XY Acad Sci 1979, vol 330. SelikofT IJ, Lee DHK: Asbestos <0 Disease. New York, Academic Press Inc, 1978. Simpson W (ed): Asbestos: Final Reports of the Advisory Committee oh Asbestos. London, 1980, 2 vol. Workplace Exposure to Asbestos, US Dept of Health and Human Services publication (NTOSH) 81-103. Cincinnati, National Institute of Occupational Safety and Health, 1980. 596 Arch Pnlhol Lab Mod--Vol 106, Ocl 8, 1082 Asbeslos-Associated Disoose--Craighead ct al * -. iVt-V.^-" .' '*_*,, "**- .V> -V. }.'vs - 'Vijv' ?V' _ -4*4 '5* ?&>-/ -4 ., ' -* - r V-. V-':V44'^ r- : r .-,V "-*'*. . * ' A .'< 'V* tv ' :' .~V v. . '' *Jf . p ': ! K fC ' ' '' , --':' ' -*. ' ' ' *- -. \ s. , v . ' Unfixed lung of Canadian chrysotile worker. Upper left. Note diffuse interstitial fibrosis and brownish discoloration attributable to hemo siderin accumulations in macrophages. Upper right, Note diffuse parenchymal fibrosis and honey omh lesions that are particularly strik ing adjacent to visceral pleura. Bottom, Note fibrous thickening of visceral pleura. There is dense parenchymal fibrosis and honeycomb ing, which is particularly prominent in lower lobe. Accumulations ot anthracotic pigment account for gray-black discoloration of fibrotic parenchyma.